Food Packaging Materials

Methods and Protocols
in Food Science
Caio Otoni Editor
Food
Packaging
Materials
Current Protocols
METHODS AND PROTOCOLS IN FOOD SCIENCE
Series Editor
Anderson S. Sant’Ana
University of Campinas
Campinas, Brazil
For further volumes:

http://www.springer.com/series/16556
Methods and Protocols in Food Science series is devoted to the publication of research
protocols and methodologies in all fields of food science.
Volumes and chapters will be organized by field and presented in such way that the
readers will be able to reproduce the experiments in a step-by-step style. Each protocol will
be characterized by a brief introductory section, followed by a short aims section, in which
the precise purpose of the protocol will be clarified.
Food Packaging Materials
Current Protocols
Edited by
Caio Otoni
Department of Materials Engineering, Federal University of São Carlos, São Carlos, Brazil
Editor
Caio Otoni
Department of Materials Engineering
Federal University of São Carlos
São Carlos, Brazil
ISSN 2662-950X ISSN 2662-9518 (electronic)

Methods and Protocols in Food Science
ISBN 978-1-0716-3612-1 ISBN 978-1-0716-3613-8 (eBook)
https://doi.org/10.1007/978-1-0716-3613-8
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part
of Springer Nature 2024
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and
retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter
developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to
be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty,
expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been
made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This Humana imprint is published by the registered company Springer Science+Business Media, LLC, part of Springer
Nature.
The registered company address is: 1 New York Plaza, New York, NY 10004, U.S.A.
Printed on acid-free paper

Preface to the Series
Methods and Protocols in Food Science series is devoted to the publication of research
protocols and methodologies in all fields of food science. The series is unique as it includes
protocols developed, validated, and used by food and related scientists as well as theoretical
basis are provided for each protocol. Aspects related to improvements in the protocols,
adaptations, and further developments in the protocols may also be approached.
Methods and Protocols in Food Science series aims to bring the most recent develop-
ments in research protocols in the field as well as very well-established methods. As such, the
series targets undergraduate, graduate, and researchers in the field of food science and
correlated areas. The protocols documented in the series will be highly useful for scientific
inquiries in the field of food sciences, presented in such way that the readers will be able to
reproduce the experiments in a step-by-step style.
Each protocol will be characterized by a brief introductory section, followed by a short
aims section, in which the precise purpose of the protocol is clarified. Then, an in-depth list
of materials and reagents required for employing the protocol is presented, followed by a
comprehensive and step-by-step procedures on how to perform that experiment. The next
section brings the dos and don’ts when carrying out the protocol, followed by the main
pitfalls faced and how to troubleshoot them. Finally, template results will be presented and
their meaning/conclusions addressed.
The Methods and Protocols in Food Science series will fill an important gap, addressing
a common complain of food scientists, regarding the difficulties in repeating experiments
detailed in scientific papers. With this, the series has a potential to become a reference
material in food science laboratories of research centers and universities throughout the
world.
Campinas, Brazil Anderson S. Sant’Ana
v
Preface
Once, I heard that the future of packaging is having “no packaging at all.” This is a ground-
breaking approach that might solve some concerns related to the boundaries of the linear life
cycle of materials: from the cradle, as far as the depletion of critical raw materials from
nature; to the grave, in whatever concerns the environmental fate of such an often consid-
ered “useless piece of matter,” mainly when one deals with single-use packaging that is
persistent and ends up being treated unsuitably once it has played its protective role.
Although revolutionary, this approach may be limited to short-range businesses, mean-
ing that I consider it idealistic in our industrialized and globalized society, in which
packaging actually plays vital roles as far as logistics, marketing, and quality assurance, to
mention a few. The role of packaging gets even more pivotal when dealing with food
systems, known to be prone to a range of spoilage mechanisms. Food Packaging is the
core topic of this contribution, wherein I merge both disciplines in which I have been
trained—namely, Food Engineering and Materials Engineering—to gather what I believe to
be the most relevant nontraditional analytic techniques, detailed by a team of leading experts
with the ultimate goal of supporting contemporary packagers in industry and academia
during their journey toward consistency and innovations in the field.
This text is also devoted to encouraging sustainability, multifunctionality, safety, and
performance through, respectively, linear-to-circular, passive-to-active, threatening-to-
harmless, and disposable-to-durable paradigm shifts. In my opinion, these transitions will
continue to be the main drivers of the packaging revolution that is already ongoing.
Herein, circularity is put forward when protocols are proposed to (i) track potentially
hazardous compounds arising from mechanical recycling, (ii) monitor biodegradation from
a biological recycling angle, and (iii) characterize the performance of packaging materials in
terms of additives and defects, so these can be improved toward extended longevity within
the economic cycle.
The different roles that packaging can play in an active fashion are addressed when
methods are described to (i) predict microbial development in modified atmosphere pack-
aging, evaluate the efficiency in actively preventing the growth of (ii) fungi and (iii) bacteria
or the occurrence of (iv) oxidative reactions, (v) prospect the potential of edible packaging
to carry and deliver probiotics and prebiotics, (vi) outlook the use of phase-change packag-
ing for thermal control, and (vii) assess the interaction between packaging and consumers in
terms of sensory perception and acceptance.
The safe use of packaging in food systems is well characterized by assays of (i) migration
and release of potentially harmful molecules or particles, (ii) manifestations of toxicity in
different biological contexts, and (iii) environmental and food contaminations with micro-
plastics arising from the mismanagement of long-lasting packaging materials.
Finally, packaging performance is comprehensively pictured in terms of barrier against
the permeation of (i) moisture, (ii) gases, and (iii) microorganisms, besides molecular-level
(iv) microstructural investigation of defects and (v) spatiospectral distribution of additives,
both via nondestructive techniques.
vii
viii Preface
While this text almost omits well-established protocols—e.g., those already common-
place in the literature—I do hope it catalyzes ongoing and future endeavors toward next-
generation food packaging, soundly designed based on reliable, comparable, and reproduc-
ible characterizations. Enjoy the pack, wherein every layer unfolds a narrative of possibilities!
Sao Carlos, Brazil Caio Otoni

Contents
Preface to the Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
PART I ENVIRONMENTAL AND TOXICOLOGICAL ASPECTS OF FOOD PACKAGING

MATERIALS
1 Biodegradability of Biodegradable Plastics in Compost, Marine,

and Anaerobic Environments Assessed by Automated Respirometry. . . . . . . . . . . 3
Joseph P. Greene, William Hart-Cooper, Lennard F. Torres, Julia Cunniffe,
Artur Klamczynski, Gregory M. Glenn, and William J. Orts
2 Biodegradability of Polymers by Relatively Low-Cost and Readily Available
Nonautomated Respirometry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Alex S. Babetto, Laı́s T. Possari, Baltus C. Bonse, and Sı́lvia H. P. Bettini
3 Detection and Identification of Microplastics in Food and the Environment . . . 57
Walter R. Waldman, Cristiane Vidal, Mariana A. Dias, Victor Z. Resende,
and Cassiana C. Montagner
4 Identification of Intentionally and Non-intentionally Added Substances
in Recycled Plastic Packaging Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Magdalena Wrona, Davinson Pezo, Robert Paiva, and Sandra A. Cruz
5 Poly- and Perfluorinated Alkyl Substances in Food Packaging Materials. . . . . . . . 99
Rachel C. Scholes, William Hart-Cooper, Gregory M. Glenn,
and William J. Orts
6 Migration of Building Blocks, Additives, and Contaminants
from Food Packaging Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
Victor G. L. Souza, Regiane Ribeiro-Santos, Patricia F. Rodrigues,
Carolina Rodrigues, João R. A. Pires, Ana T. Sanches-Silva,
Isabel Coelhoso, Fátima Poças, and Ana L. Fernando
7 In Vitro Cytotoxicity Testing of Food Packaging . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Arthur B. Ribeiro, Juliana G. F. Silva, Lucas N. F. Trevizan,
Hernane S. Barud, Flávia A. Resende, and Denise C. Tavares
8 In Vitro Genotoxicity/Mutagenicity Testing of Food Packaging . . . . . . . . . . . . . . 149
Flávia A. Resende, Juliana G. F. Silva, Arthur B. Ribeiro,
Lucas N. F. Trevizan, Hernane S. Barud, and Denise C. Tavares
PART II MICROSTRUCTURAL AND BARRIER FEATURES OF FOOD PACKAGING

MATERIALS
9 Microstructural and Defect Analysis of Food Packaging Materials

Through X-Ray Microtomography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Marcos V. Lorevice, Pedro I. C. Claro, Diego M. Nascimento,
and Rubia F. Gouveia
ix
x Contents
10 Mapping the Distribution of Additives Within Polymer Films

Through Near-Infrared Spectroscopy and Hyperspectral Imaging. . . . . . . . . . . . . 183
Jussara V. Roque, Cı́cero C. Pola, Larissa R. Terra, Taı́la V. Oliveira,
Reinaldo F. Te'o filo, Carmen L. Gomes, and Nilda F. F. Soares
11 Water Vapor Permeability of Hydrophilic Films. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
Roberto J. Avena-Bustillos, Noah M. Klausner, and Tara H. McHugh
12 Permeation of Oxygen and Carbon Dioxide Through Food Packaging
Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
Victor G. L. Souza, Carolina Rodrigues, João R. A. Pires, Ana L. Fernando,
Vitor Alves, and Isabel Coelhoso
13 Microbial Permeation Through Food Packaging Materials . . . . . . . . . . . . . . . . . . . 233
Julia V. Ernesto, Patricia Severino, Anna C. Venturini,
Cristiana M. P. Yoshida, Classius F. da Silva, and Patricia S. Lopes
PART III NONTRADITIONAL ROLES PLAYED BY FOOD PACKAGING MATERIALS
14 Do Not “Pack and Pray”: Use Predictive Models to Assess the Microbial
Safety and Shelf-Life of Modified Atmosphere Packaged Foods . . . . . . . . . . . . . . . 245
Arı́cia Possas, Fernando Pérez-Rodrı́guez, and Antonio Valero
15 Antifungal Activity of Edible Films and Coatings for Packaging of Fresh
Horticultural Produce . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Lluı́s Palou and Marı́a B. Pérez-Gago
16 Antibacterial Activity of Active Food Packaging Materials . . . . . . . . . . . . . . . . . . . . 279
Paula J. P. Espitia and Rejane A. Batista
17 Antioxidant Activity Assays for Food Packaging Materials . . . . . . . . . . . . . . . . . . . . 293
Fabiana H. Santos, Danielle C. M. Ferreira, Julia R. V. Matheus,
Ana E. C. Fai, and Franciele M. Pelissari
18 Release of Active Agents from Food Packaging Materials . . . . . . . . . . . . . . . . . . . . 311
Murilo S. Pacheco, Mariana A. de Moraes, Mariana A. da Silva,
and Andréa C. K. Bierhalz
19 Bioactive Properties of Probiotic and Prebiotic Edible Films . . . . . . . . . . . . . . . . . 325
Jackson A. Medeiros, Carolina M. Niro, Mateus K. Salgaço, Kátia Sivieri,
and Henriette M. C. Azeredo
20 Sensory Acceptance Test of Edible Packaging Using Hedonic Scale . . . . . . . . . . . 337
Suzana Maria Della Lucia and Tarcı́sio Lima Filho
21 Consumer Choice Probabilities for Food Packaging. . . . . . . . . . . . . . . . . . . . . . . . . 349
Tarcı́sio Lima Filho, Suzana Maria Della Lucia,
and Valéria Paula Rodrigues Minim
22 Thermal Performance of Food Packaging Containing Phase Change
Materials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365
Bianca C. N. Fernandes and Ana S. Prata
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
Contributors
VITOR ALVES • Linking Landscape, Environment, Agriculture and Food (LEAF), Instituto
Superior de Agronomia, Universidade de Lisboa, Lisbon, Portugal
ROBERTO J. AVENA-BUSTILLOS • Western Regional Research Center, Agriculture Research
Service (ARS), United States Department of Agriculture (USDA), Albany, CA, USA
HENRIETTE M. C. AZEREDO • Embrapa Tropical Agroindustry, Fortaleza, CE, Brazil;
Embrapa Instrumentation, São Carlos, SP, Brazil
ALEX S. BABETTO • Graduate Program in Mechanical Engineering, Centro Universitário
FEI, São Bernardo do Campo, SP, Brazil
HERNANE S. BARUD • University of Araraquara (UNIARA), Araraquara, SP, Brazil
REJANE A. BATISTA • Institute of Technology and Research of Sergipe, Aracaju, SE, Brazil
SÍLVIA H. P. BETTINI • Graduate Program in Materials Science and Engineering
(PPGCEM), Federal University of São Carlos (UFSCar), São Carlos, SP, Brazil
ANDRÉA C. K. BIERHALZ • Center of Technology, Exact Sciences and Education, Federal
University of Santa Catarina (UFSC), Blumenau, SC, Brazil
BALTUS C. BONSE • Graduate Program in Mechanical Engineering, Centro Universitário
FEI, São Bernardo do Campo, SP, Brazil
PEDRO I. C. CLARO • Brazilian Nanotechnology National Laboratory (LNNano), Brazilian
Center of Research in Energy and Materials (CNPEM), Campinas, SP, Brazil
ISABEL COELHOSO • LAQV-REQUIMTE, Department of Chemistry, NOVA School of Science
and Technology, Universidade NOVA de Lisboa, Lisbon, Portugal
SANDRA A. CRUZ • Department of Chemistry, Federal University of São Carlos, São Carlos,
São Paulo, Brazil
JULIA CUNNIFFE • Western Regional Research Center, Agriculture Research Service (ARS),
United States Department of Agriculture (USDA), Albany, CA, USA
CLASSIUS F. DA SILVA • Institute for Environmental, Chemical, and Pharmaceutical Sciences
(ICAQF), Federal University of São Paulo (UNIFESP), Diadema, SP, Brazil
MARIANA A. DA SILVA • Center of Agricultural Sciences, Federal University of São Carlos
(UFSCar), Araras, SP, Brazil
MARIANA A. DE MORAES • Department of Chemical Engineering, Federal University of São
Paulo (UNIFESP), Diadema, SP, Brazil; School of Chemical Engineering, Universidade
Estadual de Campinas (UNICAMP), Campinas, SP, Brazil
SUZANA MARIA DELLA LUCIA • Federal University of Espı́rito Santo (UFES), Alegre, ES,
Brazil
MARIANA A. DIAS • Institute of Chemistry, University of Campinas (UNICAMP),
Campinas, SP, Brazil
JULIA V. ERNESTO • Institute for Environmental, Chemical, and Pharmaceutical Sciences
PAULA J. P. ESPITIA • Nutrition and Dietetics School, Universidad del Atlántico, Puerto
Colombia, Colombia
ANA E. C. FAI • Food and Nutrition Graduate Program (PPGAN), Federal University of
the State of Rio de Janeiro (UNIRIO), Rio de Janeiro, RJ, Brazil; Department of Basic
and Experimental Nutrition, Institute of Nutrition, Rio de Janeiro State University
(UERJ), Rio de Janeiro, RJ, Brazil
xi
xii Contributors
BIANCA C. N. FERNANDES • Department of Food Engineering, School of Food Engineering,

University of Campinas (UNICAMP), Campinas, SP, Brazil
ANA L. FERNANDO • MEtRICs/CubicB, Departamen of Chemistry, NOVA School of Science
and Technology, Universidade NOVA de Lisboa, Lisbon, Portugal
DANIELLE C. M. FERREIRA • Institute of Science and Technology, Federal University of
Jequitinhonha and Mucuri Valleys (UFVJM), Diamantina, MG, Brazil; Department of
Food Technology, Federal University of Viçosa (UFV), Viçosa, MG, Brazil
GREGORY M. GLENN • Western Regional Research Center, Agriculture Research Service
(ARS), United States Department of Agriculture (USDA), Albany, CA, USA
CARMEN L. GOMES • Nanoscale Biological Engineering Laboratory, Iowa State University,
Ames, IA, USA
RUBIA F. GOUVEIA • Brazilian Nanotechnology National Laboratory (LNNano), Brazilian
Center of Research in Energy and Materials (CNPEM), Campinas, SP, Brazil; Center of
Natural and Human Sciences, Federal University of ABC (UFABC), Santo Andre, SP,
Brazil
JOSEPH P. GREENE • Department of Mechanical Engineering, Mechatronic Engineering, and
Manufacturing Technology, California State University, Chico, CA, USA
WILLIAM HART-COOPER • Western Regional Research Center, Agriculture Research Service
ARTUR KLAMCZYNSKI • Western Regional Research Center, Agriculture Research Service
NOAH M. KLAUSNER • Environmental Engineering, College of Agriculture and Life Sciences,
Cornell University, Ithaca, NY, USA
TARCÍSIO LIMA FILHO • Federal University of Espı́rito Santo (UFES), Alegre, ES, Brazil
PATRICIA S. LOPES • Institute for Environmental, Chemical, and Pharmaceutical Sciences
MARCOS V. LOREVICE • Brazilian Nanotechnology National Laboratory (LNNano),
Brazilian Center of Research in Energy and Materials (CNPEM), Campinas, SP, Brazil
JULIA R. V. MATHEUS • Food and Nutrition Graduate Program (PPGAN), Federal
University of the State of Rio de Janeiro (UNIRIO), Rio de Janeiro, RJ, Brazil
TARA H. MCHUGH • Western Regional Research Center, Agriculture Research Service
JACKSON A. MEDEIROS • Postgraduate Program in Food and Nutrition, School of
Pharmaceutical Sciences, São Paulo State University (Unesp), Araraquara, SP, Brazil
VALÉRIA PAULA RODRIGUES MINIM • Federal University of Viçosa (UFV), Viçosa, MG, Brazil
CASSIANA C. MONTAGNER • Institute of Chemistry, University of Campinas (UNICAMP),
DIEGO M. NASCIMENTO • Brazilian Nanotechnology National Laboratory (LNNano),
Brazilian Center of Research in Energy and Materials (CNPEM), Campinas, SP, Brazil
CAROLINA M. NIRO • Postgraduate Program in Biotechnology, Federal University of São
Carlos (UFSCar), São Carlos, SP, Brazil
TAÍLA V. OLIVEIRA • Food Packaging Laboratory, Universidade Federal de Viçosa (UFV),
Vicosa, MG, Brazil
WILLIAM J. ORTS • Western Regional Research Center, Agriculture Research Service (ARS),
United States Department of Agriculture (USDA), Albany, CA, USA
MURILO S. PACHECO • Department of Chemical Engineering, Federal University of São
Paulo (UNIFESP), Diadema, SP, Brazil
Contributors xiii
ROBERT PAIVA • Department of Chemistry, Federal University of São Carlos, São Carlos,
São Paulo, Brazil
LLUÍS PALOU • Centre de Tecnologia Postcollita (CTP), Institut Valencia ` d’Investigacions
Agràries (IVIA), València, Spain
FRANCIELE M. PELISSARI • Institute of Science and Technology, Federal University of
Jequitinhonha and Mucuri Valleys (UFVJM), Diamantina, MG, Brazil
MARÍA B. PÉREZ-GAGO • Centre de Tecnologia Postcollita (CTP), Institut Valencia `
d’Investigacions Agràries (IVIA), València, Spain
FERNANDO PÉREZ-RODRÍGUEZ • Department of Food Science and Technology, UIC Zoonosis y
Enfermedades Emergentes (ENZOEM), CeiA3, Universidad de C'ordoba, Campus
Rabanales, Cordoba, Spain
DAVINSON PEZO • Faculty of Health Sciences, San Jorge University, Villanueva de Gállego,
Spain
JOÃO R. A. PIRES • MEtRICs, CubicB, Departamento de Quı́mica, NOVA School of Science
and Technology, Universidade NOVA de Lisbon, Lisbon, Portugal; Bio4Plas – Biopolı́meros,
Lda, Cantanhede, Portugal
FÁTIMA POÇAS • Centro de Biotecnologia e Quı́mica Fina (CBQF), Universidade Cat'olica
Portuguesa, Lisbon, Portugal
CÍCERO C. POLA • Nanoscale Biological Engineering Laboratory, Iowa State University,
Ames, IA, USA
LAÍS T. POSSARI • Graduate Program in Materials Science and Engineering (PPGCEM),
Federal University of São Carlos (UFSCar), São Carlos, SP, Brazil
ARÍCIA POSSAS • Department of Food Science and Technology, UIC Zoonosis y Enfermedades
Emergentes (ENZOEM), CeiA3, Universidad de C'ordoba, Campus Rabanales, Cordoba,
Spain
ANA S. PRATA • Department of Food Engineering, School of Food Engineering, University of
Campinas (UNICAMP), Campinas, SP, Brazil
FLÁVIA A. RESENDE • University of Araraquara (UNIARA), Araraquara, SP, Brazil
VICTOR Z. RESENDE • Institute of Chemistry, University of Campinas (UNICAMP),
ARTHUR B. RIBEIRO • University of Franca, Franca, SP, Brazil
REGIANE RIBEIRO-SANTOS • Instituto de Quı́mica, Laborat'orio de Bioquı́mica Nutricional
e de Alimentos, Federal University of Rio de Janeiro , Rio de Janeiro, Brazil
CAROLINA RODRIGUES • MEtRICs/CubicB, Departament of Chemistry, NOVA School of
Science and Technology, Universidade NOVA de Lisboa, Lisbon, Portugal
PATRICIA F. RODRIGUES • Centre for Mechanical Engineering Materials and Process,
CEMMPRE, Department of Mechanical Engineering, University of Coimbra, Coimbra,
Portugal
JUSSARA V. ROQUE • Multivariate Chemical Data Analysis Laboratory, Universidade
Federal de Viçosa (UFV), Vicosa, MG, Brazil; Chemistry Institute, Universidade Federal
de Goias, Goiania, GO, Brazil
MATEUS K. SALGAÇO • Postgraduate Program in Food and Nutrition, School of
Pharmaceutical Sciences, São Paulo State University (Unesp), Araraquara, SP, Brazil
ANA T. SANCHES-SILVA • National Institute for Agricultural and Veterinary Research ,
INIAV, Vila do Conde, Portugal; Faculty of Pharmacy, University of Coimbra, Polo III,
Coimbra, Portugal; Center for Study in Animal Science (CECA), ICETA, University of
Porto, Porto, Portugal; Associate Laboratory for Animal and Veterinary Sciences
(AL4AnimalS), University of Lisbon, Lisbon, Portugal
xiv Contributors
FABIANA H. SANTOS • Institute of Science and Technology, Federal University of

Jequitinhonha and Mucuri Valleys (UFVJM), Diamantina, MG, Brazil; Department of
Food Technology, School of Food Engineering, University of Campinas (UNICAMP),
RACHEL C. SCHOLES • Western Regional Research Center, Agriculture Research Service
(ARS), United States Department of Agriculture (USDA), Albany, CA, USA;
Department of Civil Engineering, University of British Columbia, Vancouver, Canada
PATRICIA SEVERINO • Tiradentes University (UNIT), Aracaju, SE, Brazil
JULIANA G. F. SILVA • University of Araraquara (UNIARA), Araraquara, SP, Brazil
KÁTIA SIVIERI • Postgraduate Program in Food and Nutrition, School of Pharmaceutical
Sciences, São Paulo State University (Unesp), Araraquara, SP, Brazil; Postgraduate
Program in Biotechnology and Health Innovation, Anhanguera University (UNIAN),
São Paulo, SP, Brazil
NILDA F. F. SOARES • Food Packaging Laboratory, Universidade Federal de Viçosa (UFV),
Vicosa, MG, Brazil
VICTOR G. L. SOUZA • MEtRICs, CubicB, Departamento de Quı́mica, NOVA School of
Science and Technology, Universidade NOVA de Lisbon, Lisbon, Portugal; International
Iberian Nanotechnology Laboratory (INL), Braga, Portugal
DENISE C. TAVARES • University of Franca, Franca, SP, Brazil
REINALDO F. TEÓFILO • Multivariate Chemical Data Analysis Laboratory, Universidade
Federal de Viçosa (UFV), Vicosa, MG, Brazil
LARISSA R. TERRA • Laboratory for Theoretical and Applied Chemometrics, Institute of
Chemistry, Universidade de Campinas (UNICAMP), Campinas, SP, Brazil
LENNARD F. TORRES • Western Regional Research Center, Agriculture Research Service
LUCAS N. F. TREVIZAN • University of Araraquara (UNIARA), Araraquara, SP, Brazil
ANTONIO VALERO • Department of Food Science and Technology, UIC Zoonosis y
Enfermedades Emergentes (ENZOEM), CeiA3, Universidad de C'ordoba, Campus
Rabanales, Cordoba, Spain
ANNA C. VENTURINI • Institute for Environmental, Chemical, and Pharmaceutical Sciences
CRISTIANE VIDAL • Institute of Chemistry, University of Campinas (UNICAMP),
WALTER R. WALDMAN • Science and Technology Center for Sustainability, Federal University
of São Carlos (UFSCar), Sorocaba, SP, Brazil
MAGDALENA WRONA • Department of Analytical Chemistry, Aragon Institute of
Engineering Research I3A, University of Zaragoza, Zaragoza, Spain
CRISTIANA M. P. YOSHIDA • Institute for Environmental, Chemical, and Pharmaceutical
Sciences (ICAQF), Federal University of São Paulo (UNIFESP), Diadema, SP, Brazil
Part I
Environmental and Toxicological Aspects of Food Packaging

Materials
Chapter 1
Biodegradability of Biodegradable Plastics in Compost,

Marine, and Anaerobic Environments Assessed by
Automated Respirometry
Joseph P. Greene, William Hart-Cooper, Lennard F. Torres, Julia Cunniffe,
Artur Klamczynski, Gregory M. Glenn, and William J. Orts
Abstract
Biodegradability is an increasingly beneficial property of sustainable materials, particularly for single-use
packaging. Biodegradation rates can vary dramatically depending on the conditions, whether aerobic or
anaerobic, aqueous or nonaqueous (e.g., compost). We describe protocols of several standard biodegrada-
tion test methods, spanning marine, compost, and anaerobic environments. Simple methods to analyze
biodegradation rates are also described.
Key words Biodegradability, Biodegradation, Industrial composting, Marine environment, Standard

test methods, ASTM D5338-15, ISO 14855-2, ASTM D6691, ISO 14851, ISO 14852
1 Introduction
1.1 Background Biodegradable plastics are available throughout the world. These
materials provide an opportunity to meet market needs for materi-
als with increased biobased and biodegradable content, while miti-
gating environmental and human health concerns. Food packagers
are adopting more of these materials into their supplies and utiliz-
ing them for everything from cups to coffee pods. One of the
advantages of using biodegradable plastics, especially the ones
made from sustainable resources, is the significant reduction in
carbon emissions and energy requirements during the
manufacturing process. The commercial appeal of biodegradable
plastics hinges on their good processability and mechanical proper-
ties. For example, polylactic acid (PLA) is a biodegradable polymer
derived from renewable resources, such as starch or sugar, through
fermentation. It has good durability and can be processed using
existing manufacturing equipment typically designed and originally
Caio Otoni (ed.), Food Packaging Materials: Current Protocols, Methods and Protocols in Food Science,
https://doi.org/10.1007/978-1-0716-3613-8_1,
© The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer Nature 2024
3
4 Joseph P. Greene et al.
used for petroleum-based plastics. PLA has been widely used for
food packaging and other single-use products, such as injection
molded cups and cutlery. Of course, the most important feature
of biodegradable plastics is the breakdown of these products by
microorganisms, such as bacteria and fungi, in industrial compost-
ing facilities and marine environments. Crucially, biodegradable
plastics can be made with reduced carbon emissions, waste, and
toxic pollution compared to traditional plastics.
1.2 Biobased and Biobased and biodegradable polymers have two different mean-
Biodegradable ings. Biobased products are materials made from some amount of
Definitions biomass, such as plants, trees, animals, and marine materials
[1]. Biobased products have been defined in the 2002 Farm Bill
as commercial or industrial products that are composed in whole,
or in significant part, of biological products, renewable agricultural
materials, or forestry materials. The definition has been expanded
with the 2008 Farm Bill that incorporated biobased intermediate
ingredients or feedstock [2]. The USDA has established minimum
biobased content standards for many product categories. Products
must meet or exceed the minimum biobased content in its category
to be certified as biobased products.
Biodegradable polymers are converted to biomass, CO2, and
water through a thermochemical process in a specified time frame
and in a specified disposal environment. Biodegradable polymers
meet ASTM or ISO standards for biodegradation in a specific
surrounding, for example, industrial compost or marine environ-
ments. Many biobased polymers are biodegradable, but not all
biodegradable polymers are biobased. While some biobased poly-
mers do not biodegrade, some biodegradable or compostable poly-
mers are petroleum-based synthetics. Compostable polymers are
those that meet the ASTM requirements for biodegradation under
industrial composting conditions. Replacing fossil carbon with
renewable carbon can reduce the carbon footprint of the plastic
material based on life cycle assessment (LCA) [3].
1.3 Biodegradation Biodegradation is an important feature of biodegradable plastics.

Mechanism for Two essential components of the biodegradation process are that
Biodegradable and the material must be a food source for the bacteria in the disposal
Compostable Plastics environment and that the biodegradation must take place within a
six-month period of time. Therefore, biodegradation can occur in
an industrial compost environment for biodegradable plastics if
they are used as food source for the bacteria in the compost and
that they are generally consumed within 4–12 weeks at 40–60 °C.
Likewise, biodegradation can occur in the marine environment if
the bacteria in the sea water generally consume a major portion of
the plastic within 4–12 weeks at a minimum of 30 °C.
The biodegradation of plastics into carbon dioxide, methane,
water, and biomass is achieved through interaction with
Biodegradability Assessed by Automated Respirometry 5
microorganisms and enzymes. The bacteria that degrade the poly-

mers are widely distributed in various environments. The common
biodegradation environments may be categorized as either hot or
cool. The hot biodegradation environment includes industrial
composting where the operating temperatures can range from
50 to 65 °C. The cool biodegradation environments include
home composting, soil, anaerobic digestion, landfill, and marine
waters. The mechanisms are similar in these environments, but the
activity rates of the bacteria and enzymes are different. The pro-
cesses involved in biodegrading a macromolecule to an oligomer
and an oligomer to a monomer or small molecule can involve the
following:
1. Wetting of the polymer
2. Abiotic hydrolysis
3. Chain scission
4. Transformation to simple chemicals
5. Conversion of carbon to humus and/or volatile carbon (e.g.,
carbon dioxide, methane)
The two key components of biodegradation are (i) the rate of
biodegradation by microorganisms and (ii) the life span of the
product in the disposal environment [4]. The rate of biodegrada-
tion is influenced by environmental factors, which can have a crucial
effect on the microbial population and activity. Biodegradation
typically occurs through a two-step process: (i) scission of the
main and side chains of the macromolecules through hydrolysis
induced by thermal activation, resulting in a decrease in molecular
weight, followed by (ii) conversion of short polymer chains into a
biogas through respiration by microorganisms. Environmental
parameters such as humidity, temperature, pH, absence/presence
of oxygen, and nutrients can dictate whether microorganisms con-
sider the plastic as a viable food source. Aside from environmental
factors, the biodegradation rate can also be dependent upon the
chemical and physical characteristics of the plastic. Some of these
characteristics include porosity, chemical reactivity, thermal prop-
erty, and morphology. Finally, all these conditions must be consid-
ered when testing the biodegradability of plastics, which must
occur within a relatively short time span.
1.4 Biodegradation Certification is needed for biodegradable plastics to ensure that

Certification they meet the performance specification requirements in the bio-
degradation standards. In the United States, Biodegradable Pro-
ducts Institute (BPI) and the US Composting Council (USCC)
established the Compostable Plastics certification program in the
United States that meet the American Standards for Testing Mate-
rials (ASTM) compostability standards as specified in ASTM
D6400 or ASTM D6868 [5, 6].
Fig. 1 Overview of standard specifications and test methods for determining biodegradability in various
disposal environments
1.5 Biodegradation Biodegradation can occur under a variety of conditions: anaerobic,

Standards aerobic, compost, landfills, and marine environments (Fig. 1). For
example, biodegradable plastic products are degraded by microor-
ganisms in landfill facilities in the absence of oxygen through anaer-
obic digestion. Alternatively, aerobic biodegradation refers to the
conversion of plastic material into carbon dioxide and water with
the consumption of oxygen. To place specific parameters on bio-
degradation, worldwide organizations developed acceptable stan-
dards for biodegradable products. These standards have an
important role in the information infrastructure that guides design
and manufacturing in the biodegradable plastics market.
Biodegradation standards for plastic materials are established in
two necessary categories for biodegradation, one for biodegrada-
tion performance specifications and one for a biodegradation test-
ing method (Fig. 1). Both types of standards are necessary and
sufficient to adequately establish the biodegradation performance
of plastic materials. The performance specification standard assigns
a minimum value to establish biodegradation. Performance specifi-
cation establishes the biodegradation requirement for a plastic
product. ISO 17088 and EN13432 refer to the international and
European specification standards, respectively, for plastics in indus-
trial composting facilities. ASTM has developed specification stan-
dards for both industrial composting (ASTM D6400 and D6868)
and marine (D7081) environments [7]. Currently, there are no
international or European performance specification standards for
plastics in a marine environment. Furthermore, neither one of the
organizations has specification standards for plastic articles in land-
fills and anaerobic digestors, possibly due to extensive variabilities
in those disposal environments.
Table 1
Summary of performance specification requirements
Performance
Disposal specification Level of Environmental
environment standard biodegradationa Rate of biodegradationa effectb
Compost EN 13432 ≤10% original wt. ≥90% carbon converted to CO2 None
after 84 d after 6 months
Compost ISO 17088 ≤10% original wt. ≥90% carbon converted to CO2 None
after 3 months after 6 months
Compost ASTM >90% original ≥90% carbon converted to CO2 None
D6400 wt. disintegration after 180 d
after 12 weeks
Compost ASTM ≤10% original wt. ≥90% carbon converted to CO2 None
D6868 after 12 weeks after 180 d
Marine ASTM ≤30% original wt. ≥70% carbon converted to CO2 None
D7081 after 12 weeks after 180 d
Landfill X X X X
Digestor X X X X
a
At ≥58 °C, 50% humidity for industrial composting facilities. For marine environment, temperature requirement is
30 °C
b
This effect refers to phytotoxicity and/or presence of heavy metals
Despite the different nomenclatures, all performance specifica-

tion standards require the plastic products (e.g., packaging, coat-
ings) and demonstrate three criteria [5–9]. Table 1 summarizes the
specifications for each of the standards.
1. Sufficient disintegration of the plastic products.
2. Specified rate of biodegradation.
3. No adverse effects on the disposal environment.
The standards require the plastic products demonstrate ca. 90%
loss of their original weight after 3–4 months. Furthermore,
ca. 90% of the original carbon content must be converted to CO2
by microorganisms after 6 months. Lastly, the end products must
have low phytotoxicity and contain low levels of heavy metals or
other toxic substances.
The biodegradation testing method accurately simulates the
intended environment and specifies a method for measuring bio-
degradation. Table 2 summarizes the test methods associated with
the specification standards.
The ISO 16929 and 20200 disintegration test methods involve
gravimetric analysis of plastic products in industrial composting
facilities or marine environment. The test duration is generally
Table 2
Summary of test methods for biodegradation of plastic materials
Disposal Specification Disintegration Respirometry Biogas

environment standard test method test method measured
Compost EN 13432 ISO 16929 ISO 14855 CO2
ISO 17088 ISO 20200
Compost ASTM D6400 ISO 16929 ASTM D5338 CO2
ASTM D6868
Marine X X ISO 14851 CO2
ISO 14852
Marine ASTM D7081 ISO 16929 ASTM D6691 CO2
Landfill X X ASTM D5526 CH4, CO2
ASTM D7575
Digestor X X ISO 14853 CH4, CO2
ASTM D5511
12 weeks [10, 11]. The test method for determining the rate of
biodegradation typically involves respirometric testing as specified
by organizations such as ISO or ASTM. Respirometry techniques
are an effective tool to measure the respiration of microorganisms
and are associated with readily biodegradable plastics. Modern
respirometers can automate data collection and are thus considered
simple and effective instruments to measure carbon dioxide during
respiration over a specified length of time. Because numerous
readers may not have access to modern respirometers, they are
invited to refer to Chapter 2 for nonautomated biodegradability
assessments.
Currently, there are several testing methods used for evaluating
biodegradable plastic products depending on the disposal environ-
ment (Table 2). ASTM D5338 and ISO 14855 are widely recognized
by various municipalities and regulatory agencies as the test methods
for biodegradability of products or materials in industrial composting
facilities [12, 13]. These tests involve introducing a material to a
mixed bacterial and fungal inoculum and use respirometry to measure
biodegradation. ISO 14851, 14852, and ASTM D6691 are used for
marine disposal environment, which typically involve similar proce-
dures as their terrestrial counterparts but only in an aqueous medium
and lower temperatures (ca. 30 °C).
While several test methods for measuring biodegradability have
been developed, several issues can limit their applicability (and
ultimately, their reliability) when attempting to predict rates of
biodegradation. These issues originate from uncertainties pertain-
ing to (i) inoculum, (ii) test sample morphology, and (iii) a suitable
mathematical model used to predict product half-lives.
Recommendations for the type of inoculum used for the test

methods are vague, and, given its rich diversity, it is difficult to
obtain a standardized inoculum with constant characteristics for
biodegradability tests. Moreover, the type of inoculum can signifi-
cantly affect the rate of biodegradation [14–17]. Secondly, the
available test methods recommend test samples to be ground and
sieved through a specific mesh size to ensure a homogenous particle
size and high surface area-to-volume ratio. Unfortunately, none of
the published procedures provide guidelines for testing the biode-
gradability of samples with varying morphologies, such as semi-
liquid resins or foamed plastic films, where grinding may not be
applicable. Finally, the test methods do not provide simple mathe-
matical models that would predict ultimate material half-life, a
valuable parameter to end-users when designing and
manufacturing their products.
This chapter describes the US biodegradation standards for
biodegradable plastic food packaging, including starch-based pack-
aging, in common disposal environments, including compost,
marine, and anaerobic digestion. Compost environments include
aerobic conditions within hot aerobic industrial compost environ-
ments and cool aerobic home composting environments. Marine
environments include cold aerobic conditions. Landfill disposal
environments include aerobic and anaerobic conditions. Anaerobic
digestion environments include mesophilic anaerobic conditions.
2 Materials
2.1 Biodegradation . Compost soil

in a Composting . Plastic samples: films, powders, pellets, pieces, or fibers
Environment
. Positive control reference: biodegradable material (e.g., cellu-
lose powder)
. Negative control reference: nonbiodegradable material (e.g.,
polyethylene film)
. Personal computer with software (e.g., Micro-Oxymax Respi-
rometer proprietary software, Columbus Instruments)
. Respirometer (e.g., Micro-Oxymax Respirometer, Columbus
Instruments):
– Tank of compressed air (CO2-free and H2O-saturated)
– Composting vessels (typically 125 mL to 1 L)
– Humidified chamber
– Flexible tubing nonpermeable to CO2
– Stopper equipped with ports for the flexible tubing
2.2 Biodegradation . Aqueous medium (e.g., sea water or surface fresh water as
in a Marine indicated in the method; typically used within 3 days)
Environment . Micronutrient supplementation (N, P, S) as indicated in the
specified test method
. Plastic samples: films, powders, pellets, pieces, and fibers
lose powder)
polyethylene film)
. Respirometer:
– Tank of compressed air (CO2-free and H2O-saturated)
– Composting vessels (typically 125 mL to 1 L)
– Humidity controlled chamber
– Flexible tubing nonpermeable to CO2
– Stopper equipped with ports for the flexible tubing
2.3 Biodegradation . Blank anaerobic digester inoculum

in Anaerobic Digestion . Plastic samples: films, powders, pellets, pieces, and fibers
or Active Landfill
lose powder)
polyethylene film)
. Test vessels (typically 125 mL to 1 L)
. Low pH fluid bath or other temperature control device
. Flexible tubing nonpermeable to CH4, CO2, and O2
. Stoppers equipped with sampling ports
. Graduated cylinder or plastic tube
. Analytical balance (±0.1 mg)
. pH meter
. Gas chromatograph
3 Methods
3.1 ASTM D6400-04. This specification standard establishes the performance require-
Standard Specification ments for biodegradation of compostable plastic materials that are
for Compostable designed to biodegrade into CO2, water, and biomass in an indus-
Plastics trial compost environment at a temperature maintained above 40 °
C. It requires that the product must demonstrate each of the three
3.1.1 Summary characteristics as follows:
1. Disintegration: Sufficient disintegration during composting.

2. Biodegradation rate: Adequate level of inherent
biodegradation.
3. Nontoxic to plants: No adverse impacts on the ability of com-
post to support plant growth.
3.1.2 Procedure Three test procedures for the ASTM D6400-04 standard specify
that three types of tests are performed on the plastic samples:
Disintegration
The first test measures the percentage of disintegration of the
plastic samples while under hot and moist compost conditions.
1. The plastic samples are weighed prior to exposure to test
conditions.
2. The samples are placed in compost soil (see Note 1) with the
use of a sack, bag, or screened container. The composting
conditions needs to be maintained at least 50 °C and 50%
relative humidity.
3. The mass of the plastic sample (see Note 2) is measured after
12 weeks by passing the plastic sample and compost through a
2-mm sieve.
Biodegradation Rate
The second test procedure for D6400-04 standard specifies a bio-

degradation rate, which converts 90% of the carbon in the original
plastic samples under composting conditions at least 50 °C and 50%
moisture for 180 d into CO2 as measured by a CO2 respirometer.
The details of the test procedure are listed in ASTM D53317-11
test method (see Subheading 3.2).
Phytotoxicity
The third test procedure for ASTM D6400-04 standard specifies
the ability of the compost soil at the end of the biodegradation
testing to support plant growth through phytotoxicity testing and
very low regulated heavy metal concentrations.
1. Phytotoxicity testing is achieved through planting of tomato,
cucumber, radish, rye, barley, or cress grass seeds in the tested
compost soil.
2. Plant growth after 10 d indicates positive soil conditions. Plant
biomass tests can reveal quality differences between composts
and can indicate potential plant stress induced by the compost
at the given level used in the test.
3. The level of regulated heavy metals (see Note 3) can be

measured with flame atomic absorption spectrophotometer
using an air-acetylene flame and equipped with a Pb hollow-
cathode lamp.
3.2 ASTM D5338-11. This test method standard measures the degree and rate of biodeg-
Standard Test Method radation of plastic materials under controlled composting condi-
for Determining tions, simulating industrial composting conditions. The plastic test
Aerobic samples are exposed to an inoculum that is derived from industrial
Biodegradation of compost.
Plastic Materials
Under Controlled
Composting
Conditions
3.2.1 Summary
3.2.2 Procedure 1. Sieve the compost through a 1.4-mm screen and analyze it for
moisture content (see Note 4).
2. Place 40 g of compost in the vessel.
3. Cut the plastic samples into small pieces, ca. 0.5 g total weight,
and then place in a vessel with warm and moist compost soil,
making sure plastic sample is in good contact with the compost
(see Note 5).
4. Adjust the moisture content to 58.5% (see Note 6).
5. Connect the composting vessels to the respirometer, equipped
with the CO2 sensor ranging from 0% to 3%.
6. Maintain the test containers at (58 ± 2) °C for 180 d, as
indicated by the test method. The biogas from the container
is measured for CO2 and O2 over the testing period. Analyze
the headspace of the composting vessels for development of
CO2. Sampling period can be 2 h (see Note 7).
7. Run samples in triplicates. Compost baseline controls lacking
test sample are run in triplicate or quadruplicate.
3.3 ASTM D-7081- This specification standard establishes the performance require-
05. Nonfloating ments for biodegradation of plastic materials and products, includ-
Biodegradable Plastic ing packaging, films, and coatings in a marine environment, which
in the Marine includes conditions of aerobic marine waters or anaerobic marine
Environment sediments, or both. It establishes the requirements for biodegrada-
tion of plastic materials that have rates that are similar to known
3.3.1 Summary compostable materials. The standard requires the product must
demonstrate each of the three characteristics as follows:
1. Disintegration: Sufficient disintegration during marine
biodegradation.
2. Biodegradation rate: Adequate level of inherent biodegrada-

tion of the plastic material.
3. Nontoxic to plants: Minimal adverse effect on the marine
environment.
3.3.2 Procedure Three test procedures for the ASTM D-7081-05 standard specify
that three types of tests are performed on the plastic samples.
Disintegration
The first test measures the percentage of disintegration of the
plastic samples while under 30 °C marine conditions.
Biodegradation of biodegradable plastics in marine environ-

ment is based upon two sets of standards, the first for a test method
standard and the second for a performance specification standard.
The marine biodegradation standard covers nonfloating products
made from plastics that are designed to biodegrade in the aerobic
marine environment. It applies to deep sea water, shallow sea water,
and brackish inland waters. Plastic materials must demonstrate
disintegration and inherent biodegradation during marine water
exposure and not exhibit adverse environmental impacts on the
survival of marine organisms while in the marine environment.
Biodegradation Rate
The second test procedure for ASTM D-7081-05 standard specifies
a biodegradation rate, which converts 90% of the carbon in the
original plastic samples under marine conditions at least 30 °C and
50% moisture for 180 d into CO2 as measured by a CO2 respirom-
eter. The details of the test procedure are listed in ASTM D53317-
11 test method (see Subheading 3.4).
Phytotoxicity
The plastic sample also must pass several marine toxicity tests,
including Polytox (microbial oxygen absorption), Microtox
(microbial bioluminescence) test, Fish Acute Toxicity (static con-
ditions) OPPTS 1750.1075, Daphnia Acute Toxicity (static con-
ditions) OPPTS 1750.1010, or Static Algal Toxicity Test OPPTS
1750.5400. The plastic samples must also have less than 25% of
maximum allowable concentrations of regulated heavy metals.
Marine biodegradation standards require that the plastic sam-

ples also pass the ASTM D-6400 standard for biodegradation
under industrial aerobic compost conditions. The ASTM D-6400
standard requires plastic samples to convert 90% of the carbon in
the plastic sample to CO2 after 180 d while at 58 °C.
3.4 ASTM D6691-09. This test method is used to determine the aerobic biodegradation
Standard Test Method rate of plastic materials exposed to sea water or synthesized sea
for Determining water with pre-grown population of at least ten aerobic marine
Aerobic microorganisms of known genera. It consists of preparing a
Biodegradation of uniform inoculum of marine water, exposing the plastic samples
Plastic Materials in the to the marine water, measuring biodegradation with a carbon
Marine Environment by dioxide respirometer or equivalent measurement method, and
a Defined Microbial assessing the percentage of carbon conversion in the plastic to
Consortium or Natural carbon dioxide.
Sea Water Inoculum
3.4.1 Summary
3.4.2 Procedure 1. Filter water (see Note 9) through 0.5-mm screen to remove
sand and other impurities.
2. Add NH4Cl and KH2(PO4) salts as per ASTM 6691D.
3. Cut the plastic samples into small pieces, typically 20 mg to
several grams, depending on the detection limits of the instru-
ment in use.
4. Place the plastic sample with 75 mL of marine stock solution in
125-mL bottles.
5. Provide containers also for the following samples: blank
(marine water only), positive control (e.g., cellulose or starch),
and if needed, negative control (e.g., polyethylene).
6. Run experiments in triplicates. Use test water (e.g., ocean
water) as a baseline in quadruplicates.
7. Connect the chambers to the respirometer and keep at (30 ± 2)
°C with continuous agitation for 180 d.
8. Analyze the headspace for development of CO2. Take head-
space samples in 2-h intervals and replace 50% of gas with
atmospheric air to provide adequate oxygenation (see Note
10).
3.5 ASTM D5511-02. This test method measures the biodegradation rate of plastic mate-
Standard Test Method rials under anaerobic thermophilic conditions in an aqueous envi-
for Determining ronment. The plastic test samples are exposed to an inoculum that
Anaerobic is derived from an aerobic digester or wastewater treatment opera-
Biodegradation of tion. The plastic samples can be in the form of films, powders,
Plastic Materials pellets, or molded pieces and are placed in a vessel with warm
Under High-Solids inoculum with proper anaerobic bacteria.
Anaerobic-Digestion
Conditions
3.5.1 Summary
3.5.2 Procedure 1. Place 1 kg of inoculum derived from properly operating anaer-

obic digester that is made from pretreated household waste.
The inoculum should be derived from a digester operating
under greater than 20% total solids conditions.
2. Add the plastic samples to each test container in quantities up
to 100 g.
3. The test apparatus includes a graduated cylinder or plastic
column [16]. The graduated cylinder or plastic column is
inverted in a low-pH fluid to avoid CO2 loss through the
dissolution in the fluid. The biogas is calculated through a
pressure measurement of the inverted tubes. Through ideal
gas law, the pressure can be converted to grams of biogas.
The concentration of biogas can be converted to concentra-
tions of CO2 and CH4. The conversion of carbon from the
plastic sample to CO2 and CH4 can be determined. This will
result in the carbon biodegradation percentage over 30 d in a
high-solids anaerobic digester.
4. A minimum of 12 test vessels are required for the test.
5. Maintain the test containers at (50 ± 2) °C for 30 d.
6. Provide containers for the following samples: blank, positive
control, and negative control, as described previously. The
positive control must obtain greater than 70% biodegradation
in 30 d.
7. Complete the testing in triplicate.
8. Test the inoculum for pH.
9. Measure the biogas from the container for CH4, CO2, and O2
over the testing period.
3.6 Active Landfill Landfills in the United States are typically built with the EPA
(ASTM D5511-02) guidelines with the use of clay linings and a landfill cap (Criteria
for Solid Waste Disposal Facilities 2013). The most common mate-
rial for landfill caps is made from asphalt or concrete (Remediation
Technologies Screening Matrix and Reference Guide 2013). Land-
fills can operate with the creation of biogas that is composed of
methane, carbon dioxide, and other trace gases. Methane gas can
be vented and burned or can be captured and stored for energy
purposes. The carbon dioxide and other gases must be scrubbed to
provide a clean methane gas without carbon dioxide or other gases.
Some landfills are considered active and provide clean methane gas
for energy consumption. Biodegradable plastics can hold the waste
as trash bags for disposal and provide food source for the aerobic
and anaerobic bacteria that are in the landfill. Standards are needed
to evaluate the biodegradation of biodegradable plastics in landfills.
Biodegradation of plastics in active landfill can use the ASTM
standards in ASTM 5511 conditions to measure biodegradation
under anaerobic conditions.
3.7 Analysis of The total amount of carbon in a plastic sample is generally deter-
Respirometry Results: mined by elemental analysis (see Note 11). In each round of experi-
Carbon Content and ments, a baseline must be established, which represents control
Mineralization Kinetics media (e.g., compost, marine water, wastewater, etc.) lacking a
test material (see Note 12). Subtracting the accumulated CO2 of
the baseline experiments from the test cases, which contain media
and test material, gives the quantity of CO2 that can be attributed
to the mineralization of the test material. In both the baseline and
test material conditions, mineralized gases measured through res-
pirometry can be converted to moles carbon using the ideal gas law.
The observed and theoretical moles carbon can be used to calculate
percent mineralization and % theoretical carbon remaining in the
test sample. Percent carbon mineralization can be converted to %
theoretical carbon remaining and kinetic models can be applied to
obtain rate constants.
The most practical interpretation of biodegradability rates can
be obtained using linear fits which represent average biodegrada-
tion rates (Eq. 1), which represent the average percent degradation
of a test material divided by the duration of the experiment. This
approach is useful when measuring initial rates of degradation,
which, like many other chemical processes, are often linear. How-
ever, biodegradation kinetics often deviate from linearity after the
first ca. 30% of the starting material is consumed. This effect implies
that, typically, linear approximations (Eq. 1) can accurately describe
initial rate kinetics but are less accurate when describing kinetics
curves where higher levels of degradation are observed. In Eq. 1,
Δ% degradation represents the percent of material that was
degraded over a given time span (Δtime). Linear fits can also be
useful in determining induction periods, which describe kinetics
where a period of slow degradation precedes a rapid increase in
mineralization. Rates often vary by orders of magnitude, relative
rates (relative rate constant = krel, Eq. 2) provide an intuitive
ranking of biodegradation kinetics (Fig. 2). In Eq. 2, kfast represents
the most rapidly degraded rate constant, compared to kslow that
represents the slower rate constant:
Δ%degradation
Average rate = ð1Þ
Δtime
kfast
krel = ð2Þ
kslow
Average biodegradation rates for various materials are listed in

Table 3 [18–33].
These materials represent three groups: bioplastics, natural
materials, and synthetic materials. In bioplastics, PLA degrades
rapidly under industrial compost conditions, but degradation
does not appreciably proceed in fresh water, marine water, soil, or
45
Relative rates
40 Cellulose: 1.0 PHA: 0.20% per day
Percentage biodegradation
35 PHA: 1.1 y = 0.2048x + 0.925
HDPE: 0.039 R2 = 0.9946
30
PLA: 0.18
25 Cellulose
Cellulose: 0.19% per day
20 y = 0.1874x + 1.2289 PHA
R2 = 0.9906
15 HDPE
PLA: 0.0074% per day HDPE: 0.0034% per day PLA
10 y = 0.0074x - 0.169 y = 0.0034x - 0.0768
5 R2 = 0.9284 R2 = 0.9284
0
0 50 100 150 200
-5
Days
Fig. 2 Linear fits and average rates of marine biodegradation data (ASTM D6691)
home compositing conditions. Polycaprolactone (PCL) and poly-

hydroxybutyrate (PHB) degrade much quicker than PLA in all
conditions. Polyhydroxyoctanoate (PHO) underwent rapid degra-
dation in fresh water, marine water, and industrial compost,
whereas it barely degraded in soil. Poly(1,4-butylene succinate)
(PBS) degraded similarly in both soil and compost conditions,
presenting a large degradation range based on the condition
temperature.
Of the literature data surveyed, most of the natural materials
degraded in 2 months or less in soil or compost, except for beeswax
which was 3–4 months in oil-contaminated soil. Of these natural
materials, guar gum degraded the most rapidly. On the other hand,
the synthetic materials had a larger range of degradation rates from
months to years. Poly(vinyl alcohol) (PVA) degraded slowly under
wastewater, soil, and compost environments. Carboxymethylcellu-
lose (CMC) in wastewater showed that the higher the degree of
substitution (DS), the longer the degradation rate. Low-density
polyethylene (LDPE) showed similar results to PLA in soil with no
significant degradation.
Pseudo first-order fits have been widely used to describe the
nonlinear kinetics often observed in biodegradation tests. A pseudo
first-order fit (Eq. 3) is applied to mineralization data (Fig. 3),
affording rate constants that can be used to calculate half-lives
(Eq. 4):
½A] = ½A ]0 e - kobs t ð3Þ

lnð2Þ
t 1=2 = ð4Þ
k
In Eq. 3, [A] represents the concentration of a material at a
given time, [A]0 is the starting concentration of the same material,
Table 3
18
Biodegradation rates of bioplastics, natural, and synthetic materials from literature
Temperature Average biodegradation per Residence

Material Condition [°C] day [%] time References
Bioplastics
Poly(lactic acid) (PLA) Fresh water 21 <0.01a >20 yra [18]
Marine water 30 <0.01a >20 yra
Soil 25 0.01 20 yr
Industrial compost 58 1.33 1–2 mo
Joseph P. Greene et al.
Soil 25–37 <0.01a >20 yra [19]

Compost 25–50 <0.01a–0.37 8 mo to
>20 yra
Polycaprolactone (PCL) Fresh water 21 0.25 1 yr [18]
Marine water 30 0.30 10 mo
Soil 25 0.55 5–6 mo
Home compost 28 1.19 2–3 mo
Industrial compost 58 2.67 1 mo
Anaerobic digestion 52 0.71 4–5 mo
Soil 25–37 0.19–0.21 14–15 mo [19]
Compost 25–50 0.30–1.10 2–10 mo
Anaerobic methane 37 0.38 7–8 mo [20]
sludge
Polyhydroxyoctanoate (PHO) Fresh water 21 0.30 10 mo [18]
Marine water 30 0.86 3–4 mo
Soil 25 0.01 20 yr
Industrial compost 58 0.85 3–4 mo
Polyhydroxybutyrate (PHB) Fresh water 21 1.50 2 mo [18]
Marine water 30 1.20 2–3 mo
Soil 25 0.67 4–5 mo
Industrial compost 58 2.56 35 d
Soil 25–37 0.08–0.21 1–3 yr [19]
Compost 25–50 0.22–0.35 8–14 mo
Anaerobic methane 37 2.38 38 d [20]
sludge
Thermoplastic starch (TPS) Fresh water 21 2.00 1–2 mo [18]
Marine water 30 3.00 1 mo
Soil 25 0.55 5–6 mo
Home compost 28 1.22 3–4 mo
Anaerobic digestion 52 0.63 1–2 mo
Poly(butylene adipate-co-terephthalate) Compost 28 0.03 8–9 yr [21]
(PBAT)
Poly-(β-hydroxybutyrate- Anaerobic methane 37 0.69 4–5 mo [20]
co-β-hydroxyvalerate) (PHBV) sludge
Poly(1,4-butylene succinate) (PBS) Soil 25–37 0.03–0.16 1–10 yr [19]
Compost 25–50 0.01–0.21 1–17 yr
Natural materials
Beeswax Oil-contaminated soil 20 0.94 3–4 mo [22]
Wheat gluten Soil 20–22 1.90 47 d [22]
Compost 58 4.55 20 d [24]
Cellulose Marine Water 30 0.19 16 mo This work
(Fig. 2)
Compost 58 1.06 2–3 mo [25]
Sodium alginate edible film Soil 37 2.60 35 d [26]
Guar gum Compost 15–25 7.14 13 d [27]
Zein film Soil (solid waste 30 4.05 22 d [28]
landfill)
Soil (lagoon) 30 4.05 22 d
Alginate film Compost 32 2.57 35 d [29]
Synthetic materials
Poly(vinyl alcohol) (PVA) Marine and 25 0.62 to <0.01a 5 mo to [30–33]
Biodegradability Assessed by Automated Respirometry
Wastewater >20 yra

Soil 15–25 0.30–0.31 9–10 mo [34]
Compost 58 0.49 6 mo [25]
19
(continued)
Table 3
20
(continued)
Temperature Average biodegradation per Residence

Material Condition [°C] day [%] time References
Carboxymethyl cellulose (CMC) [DS 0.44] Wastewater 25 1.34 2–3 mo [30]
CMC [DS 0.75] Wastewater 25 0.54 5–6 mo
Cellulose acetate (CA) [2.85] Soil 15–25 0.25 1 yr [35]
Paraffin wax Oil contaminated soil 20 1.01 3 mo [22]
Joseph P. Greene et al.
Low-density polyethylene (LDPE) Soil 15–25 0.01 20 yr [36]

a
Material showed no biodegradation – set average % per day to “<0.01” and residence time to “>20 yr” as minimum and maximum cutoffs
Marine
120.00 y = 9.99E+01e-2.29E-03x
R2 = 9.97E-01
Half-life: 300 days
100.00
% Cellulose remaining
80.00 Aerobic wastewater

y = 1.01E+02e-307E-03x
R2 = 9.79E-01
60.00 Half-life: 230 days
40.00 Industrial compost

y = 8.44E+01e-403E-02x
R2 = 9.75E-01
20.00
Half-life: 17 days
0.00
-4.000 1.000 6.000 11.000 16.000 21.000
time/days
Fig. 3 Pseudo first-order fits and half-lives of cellulose in marine, aerobic wastewater, and compost
environments
kobs is the observed pseudo first-order rate constant, and

t represents time. In Eq. 4, t1/2 signifies the material’s half-life
under the given test conditions and k is the first-order rate constant
(e.g., kobs from Eq. 1). Residence times, defined as the time needed
to obtain 90% biodegradation, can be estimated using linear fits of
average rates, as shown in Table 3, or pseudo first-order kinetics.
Due to the nonlinearity of typical kinetics, linear fits of initial rates
tend to underestimate material residence times. In the absence of an
induction period, which is typically modeled with additional para-
meters, a residence time is more accurately determined by ca. 3.3
half-lives obtained from a pseudo first-order fit.
Conclusions
Biodegradation rates are a product of many factors, including test
medium (e.g., compost, marine, active landfill), inoculum selection
(e.g., organisms are acclimated or nonacclimated to the test mate-
rial), and temperature. Standard conditions exist for many of these
environments. Rates can be evaluated using kinetics to estimate
material residence times under a given set of conditions. The rigor-
ous evaluation of material biodegradability has the potential to
mainstream the use of nonpersistent materials as part of a more
circular economy that generates waste as a feedstock.
4 Notes
1. Compost needs to be either newly purchased or no more than

2 months old.
2. Sample selection process is of utmost importance for making
proper comparisons. As the contact of compost with plastic
within the respirometer chamber is critical, the plastic samples
need to be uniformly selected for similar surface area. For this
reason, grinding the samples can be performed. However,
some would argue that the results obtained from ground plas-
tic are unrealistic. On one hand, grinding of the plastic enlarges
the specific surface area and provides better contact with the
degradation media thus enhancing the overall degradation rate.
While using films, the samples have to be selected for a uniform
thickness and cut to the same size. This is especially difficult
when plastic blends are analyzed. Foams and fiber mats that
have smaller density than the media tend to float on the surface,
thus only one surface is exposed for the degradation. This is a
challenge that the experimenter must address individually.
3. The level of regulated heavy metals can be measured with flame
atomic absorption spectrophotometer using an air-acetylene
flame and equipped with a Pb hollow-cathode lamp. The com-
post samples must have regulated metals concentrations less
than 50% of the acceptable levels of regulated heavy metals as
prescribed in 40 CFR Part 503.13, that is, lead (75 mg·kg-1),
cadmium (17.5 mg·kg-1), chromium (not specified), copper
(375 mg·kg-1), nickel (105 mg·kg-1), zinc (700 mg·kg-1),
and mercury (4.25 mg·kg-1).
4. The sieving process removes all large particles like stones, glass,
metal, and pieces of wood. It is critical to minimizing experi-
mental variations in the mineralization rate. After the screen-
ing, the compost is thoroughly mixed and is allowed to
equilibrate for moisture.
5. To ensure there is good contact of the compost with the
sample, first place half of the necessary amount of compost in
the chamber. This is followed by the addition of the sample and
finally by the rest of the compost.
6. The moisture of the compost, which usually falls within a
35–40% range, is determined in triplicates that have to agree
within 0.25%. As the wet compost does not mix well with the
sample, the final moisture is adjusted after the sample and
appropriate amount of “dry” compost are placed in the respi-
rometer chamber.
7. With every sampling time, 50% of the headspace gas is replaced

with atmospheric air, providing adequate oxygenation of the
compost.
8. It is important to have a positive control, either cellulose or
starch, to observe total carbon released and the rate of the
reaction. The positive control would show almost an instant
response. Starch, either as film or as powder, degrades within
the first week. Cellulose is used completely within 2–4 weeks.
9. The water is either collected from the open sea area or other
surface water sources, and used within 3 days of collection.
10. The chambers were situated away from the sunlight to mini-
mize the effect of ultraviolet degradation.
11. The amount of carbon can also be determined experimentally
via calorimetry or found in the literature.
12. Biodegradation of bioplastic is calculated on the basis of carbon
dioxide emitted from the samples with respect to CO2 emitted
from the positive control samples.
References
1. Fornasiero P, Graziani M (2006) Renewable the marine environment (withdrawn 2014).
resources and renewable energy: a global chal- ASTM International, West Conshohocken.
lenge, 2nd edn. CRC Press www.astm.org
2. BioPrefer red Program. https://www. 8. International Organization for Standardization
biopreferred.gov/BioPreferred/faces/pages/ (2012) Specifications for compostable plastics
AboutBioPreferred.xhtml. Accessed (ISO 17088:2012) https://www.iso.org
5 Feb 2021 9. European Standards (2000) Requirements for
3. Choi B, Yoo S, Park S (2018) Carbon footprint packaging recoverable through composting
of packaging films made from LDPE, PLA, and and biodegradation. (BS EN 13432:2000)
PLA/PBAT blends in South Korea. Sustain- https://www.en-standard.eu
ability 10:2369. https://doi.org/10.3390/ 10. International Organization for Standardization
su10072369 (2019) Determination of the degree of disinte-
4. Greene JP (2014) Introduction to gration of plastic materials under defined com-
sustainability. In: Sustainable plastics. Wiley, posting conditions in a pilot-scale test. (ISO
Hoboken 16929:2019) https://www.iso.org/
5. ASTM D6400-19 (2019) Standard specifica- 11. International Organization for Standardization
tion for labeling of plastics designed to be aer- (2015) Determination of the degree of disinte-
obically composted in municipal or industrial gration of plastic materials under simulated
facilities. ASTM International, West Consho- composting conditions in a laboratory-scale
hocken. www.astm.org test. (ISO 20200:2015) https://www.iso.org/
6. ASTM D6868-21 (2021) Standard specifica- 12. ASTM D5338-15 (2021) Standard test
tion for labeling of end items that incorporate method for determining aerobic biodegrada-
plastics and polymers as coatings or additives tion of plastic materials under controlled com-
with paper and other substrates designed to be posting conditions, incorporating thermophilic
aerobically composted in municipal or indus- temperatures. ASTM International, West Con-
trial facilities. ASTM International, West Con- shohocken. www.astm.org
shohocken. www.astm.org 13. International Organization for Standardization
7. ASTM D7081-05 (2005) Standard specifica- (2018) Determination of the ultimate aerobic
tion for non-floating biodegradable plastics in biodegradability of plastic materials under
controlled composting conditions — Method 23. Ye P, Reitz L, Horan C, Parnas R (2006) Man-
by analysis of evolved carbon dioxide — Part 2: ufacture and biodegradation of wheat gluten/
Gravimetric. (ISO 14855-2:2018) https:// basalt composite material. J Polym Environ 14:
www.iso.org/ 1–7. https://doi.org/10.1007/s10924-005-
14. Ohtaki A, Akakura N, Nakasaki K (1998) 8701-3
Effects of temperature and inoculum on the 24. Zhang X, Gozukara Y, Sangwan P et al (2010)
degradability of poly-ε-caprolactone during Biodegradation of chemically modified wheat
composting. Polym Degrad Stab 62:279–284. gluten-based natural polymer materials. Polym
https://doi.org/10.1016/S0141-3910(98) Degrad Stab 95:2309–2317. https://doi.org/
00008-1 10.1016/j.polymdegradstab.2010.09.001
15. Mezzanotte V, Bertani R, Innocenti FD, Tosin 25. Lešinský D, Fritz J, Braun R (2005) Biological
M (2005) Influence of inocula on the results of degradation of PVA/CH blends in terrestrial
biodegradation tests. Polym Degrad Stab 87: and aquatic conditions. Bioresour Technol 96:
5 1 – 5 6 . h t t p s : // d o i . o r g / 1 0 . 1 0 1 6 / j . 197–201. https://doi.org/10.1016/j.bio
polymdegradstab.2004.06.009 rtech.2004.05.008
16. Krzan A, Hemjinda S, Miertus S et al (2006) 26. Mahcene Z, Khelil A, Hasni S et al (2020)
Standardization and certification in the area of Development and characterization of sodium
environmentally degradable plastics. Polym alginate based active edible films incorporated
Degrad Stab 91:2819–2833. https://doi.org/ with essential oils of some medicinal plants. Int
10.1016/j.polymdegradstab.2006.04.034 J Biol Macromol 145:124–132. https://doi.
17. Beegle JR, Borole AP (2017) An integrated org/10.1016/j.ijbiomac.2019.12.093
microbial electrolysis-anaerobic digestion pro- 27. Kaith BS, Sharma R, Kalia S (2015) Guar gum
cess combined with pretreatment of wastewater based biodegradable, antibacterial and electri-
solids to improve hydrogen production. Envi- cally conductive hydrogels. Int J Biol Macro-
ron Sci Water Res Technol 3:1073–1085. mol 75:266–275. https://doi.org/10.1016/j.
https://doi.org/10.1039/C7EW00189D ijbiomac.2015.01.046
18. Narancic T, Verstichel S, Reddy Chaganti S 28. Romero-Bastida CA, Flores-Huicochea E,
et al (2018) Biodegradable plastic blends create Martin-Polo MO et al (2004) Compositional
new possibilities for end-of-life management of and moisture content effects on the biodegrad-
plastics but they are not a Panacea for plastic ability of Zein/Ethylcellulose films. J Agric
pollution. Environ Sci Technol 52:10441– Food Chem 52:2230–2235. https://doi.org/
10452. https://doi.org/10.1021/acs.est. 10.1021/jf0350414
8b02963 29. Deepa B, Abraham E, Pothan LA et al (2016)
19. Al Hosni AS, Pittman JK, Robson GD (2019) Biodegradable nanocomposite films based on
Microbial degradation of four biodegradable sodium alginate and cellulose Nanofibrils.
polymers in soil and compost demonstrating Materials 9:50. https://doi.org/10.3390/
polycaprolactone as an ideal compostable plas- ma9010050
tic. Waste Manag 97:105–114. https://doi. 30. Wirick MG (1974) Aerobic biodegradation of
org/10.1016/j.wasman.2019.07.042 Carboxymethylcellulose. J Water Pollut Con-
20. Abou-Zeid D-M, Müller R-J, Deckwer W-D trol Feder 46:512–521
(2001) Degradation of natural and synthetic 31. Rolsky C, Kelkar V (2021) Degradation of
polyesters under anaerobic conditions. J Bio- Polyvinyl Alcohol in US Wastewater treatment
technol 86:113–126. https://doi.org/10. plants and subsequent Nationwide emission
1016/S0168-1656(00)00406-5 estimate. Int J Environ Res Public Health
21. Pinheiro IF, Ferreira FV, Souza DHS et al 18(11):6027. https://doi.org/10.3390/
(2017) Mechanical, rheological and degrada- ijerph18116027
tion properties of PBAT nanocomposites rein- 32. Chiellini E, Corti A, Solaro R (1999) Biodeg-
forced by functionalized cellulose nanocrystals. radation of Poly(Vinyl Alcohol) based blown
Eur Polym J 97:356–365. https://doi.org/10. films under different environmental conditions
1016/j.eurpolymj.2017.10.026 11 part of work herewith reported was pre-
22. Hanstveit AO (1992) Biodegradability of sented at the 5th scientific workshop on biode-
petroleum waxes and beeswax in an adapted gradable polymers and plastics, Stockholm
CO2 evolution test. Chemosphere 25:605– (Se) June 1998. Polym Degrad Stab, 64(2):
620. https://doi.org/10.1016/0045-6535 305–312. https://doi.org/10.1016/S0141-
(92)90291-X 3910(98)00206-7
33. Alonso-López O, López-Ibáñez S, Beiras R 35. Fei Z, Huang S, Yin J et al (2015) Preparation
(2021) Assessment of toxicity and biodegrad- and characterization of bio-based degradable
ability of Poly(Vinyl Alcohol)-based materials plastic films composed of cellulose acetate and
in marine water. Polymers 13(21):3742. starch acetate. J Polym Environ 23:383–391.
https://doi.org/10.3390/polym13213742 https://doi.org/10.1007/s10924-015-
34. Abd El-Mohdy HL, Ghanem S (2008) Biode- 0711-1
gradability, antimicrobial activity and proper- 36. Goheen SM, Wool RP (1991) Degradation of
ties of PVA/PVP hydrogels prepared by polyethylene–starch blends in soil. J Appl
γ-irradiation. J Polym Res 16:1. https://doi. Polym Sci 42:2691–2701. https://doi.org/
org/10.1007/s10965-008-9196-0 10.1002/app.1991.070421007
Chapter 2
Biodegradability of Polymers by Relatively Low-Cost

and Readily Available Nonautomated Respirometry
Alex S. Babetto, Laı́s T. Possari, Baltus C. Bonse, and Sı́lvia H. P. Bettini
Abstract
Humanity is currently consuming natural resources 1.75 times faster than the planet can regenerate in a
year, so the regeneration of these natural resources has become an issue of pressing concern. New drivers
have pointed to actions that minimize future impacts, which include reducing plastic waste in the environ-
ment, recycling, and the use of biodegradable polymers. The definition and determination of biodegrad-
ability of polymers has been a topic of discussion in the scientific community, mainly related to the criteria
used to define biodegradability. Academic studies have shown excellent results on polymer biodegradation
tests with automated respirometry methods; however, these tests are relatively expensive and may not be
readily available. This protocol presents an alternative to automated respirometry (procedure and monitor-
ing), detailing the procedure of a polymer biodegradation test with nonautomated respirometry and
monitored by titrimetry. The protocol is based on the international standards ASTM D5338-15, ASTM
D5988-18, ASTM D6400-19, ISO 14855-1:2012, and ISO 17556:2019. Proper execution of the proto-
col will guarantee the correct performance of respirometric polymer biodegradation tests, providing
reproducible and accurate results.
Key words Biodegradation, Nonautomated respirometry, Titrimetry, Simplified Bartha, Standards
1 Introduction
Biodegradable polymers have shown to be a promising alternative

to conventional nonbiodegradable polymers, which accumulate in
landfills and ecosystems, contributing to serious environmental
problems [1]. Briefly, biodegradable polymers can be described as
those capable of undergoing degradation due to biotic action pro-
moted by microorganisms, such as fungi, algae, and bacteria [2]. In
this process, the polymer molecules are converted into an energy
source for the microorganisms, as well as biomass and simple
molecules, such as water (H2O), carbon dioxide (CO2), and meth-
ane (CH4) [3].
It should be mentioned that not all biodegradable polymers
derive from renewable resources. Biodegradable polymers can be
https://doi.org/10.1007/978-1-0716-3613-8_2,
27
28 Alex S. Babetto et al.
obtained both from renewable (biobased) and fossil (fossil-based)

resources [2]. The origin of the polymer is not directly related to its
biodegradability [4].
The biotic degradation process is complex and can be divided
into three important stages, namely: (i) biodeterioration
(or bioerosion), (ii) biofragmentation, and (iii) assimilation. Biode-
terioration occurs due to the adhesion and growth of microbial
colonies onto the surface of the material. The biofilm formed
penetrates the pores of the substrate changing their size and distri-
bution, resulting in brittleness and cracks. The material starts dis-
integrating into smaller pieces and suffers significant loss in physical
and mechanical properties. The growth of the biofilm may be either
intensified or restricted by environmental conditions (such as pH,
temperature, and moisture) as well as by the roughness and polarity
of the material. In addition, microorganisms secrete enzymes that
may catalyze the scission of specific bonds in the polymer chains,
such as lipases and esterases that catalyze hydrolysis reactions, pro-
moting surface erosion of the substrate [3, 5].
In the biofragmentation stage the biotic action causes polymer
chain scission by degradation reactions catalyzed by the excretion of
enzymes and byproducts (such as organic acids and peroxides),
resulting in progressive reduction in molar mass and release of
monomers and/or oligomers [3, 6]. For this to happen, the bond
that will be attacked should be accessible to the active site of the
enzyme. Therefore, amorphous regions and flexible linear chains
are more susceptible to microbiological attack [7].
The final stage of biodegradation is assimilation, in which
molecules of sufficiently low molar masses are diffused into cells
and metabolized to produce adenosine triphosphate (ATP) and
compose cellular structures (biomass). In this process, small mole-
cules (e.g., H2O, CO2, CH4, and salts) are released, a phenomenon
called mineralization [3]. Equations 1 and 2 represent, respectively,
the chemical conversion involved in the aerobic and anaerobic
biodegradation of any polymer composed only of carbon, hydro-
gen, and oxygen atoms [6]:
Cpolymer source þ O2 → CO2 þ H2 O þ Cbiomass ð1Þ
Cpolymer source → CH4 þ CO2 þ Cbiomass ð2Þ
Biodegradation is a complex process, and its occurrence and
rate are conditioned by several factors, from the chemical and
physical characteristics of the material to the environmental condi-
tions in which the material is used and disposed of. Therefore,
assessing a polymer’s biodegradability is not trivial, and several
experimental methods can be used in the laboratory to monitor
biodegradation at different stages [3, 8].
Some studies employ methods that use enzymes and cell cul-
ture; however, these tests provide situations that strongly deviate
Assessing Polymer Biodegradability via Nonautomated Respirometry 29
from reality. Natural environments, into which plastic waste is often

discarded, have a great diversity of microbiota and other carbon
sources that may be preferable to microorganisms, in addition to
the polymer. Therefore, for more reliable results, most of the
methods standardized by the American Society for Testing and
Materials (ASTM) and by the Organization for Economic Cooper-
ation and Development (OECD) determine the use of simulated or
real environments [8]. The incubation media for assessing biodeg-
radation can be soil, compost, marine waters, activated sludge, or
wastewater effluents [9].
Monitoring the level of disintegration and the percentage mass
loss of the degraded material in different media are frequently used
in scientific studies. These tests provide information on biodeterio-
ration rather than on the biodegradation process as a whole
[3]. Sample disintegration does not necessarily reflect mineraliza-
tion and may result only from its fragmentation into
microplastics—readers are invited to refer to Chap. 3 for more
information on microplastics. This may even render the technique
susceptible to high margins of error due to the difficulty of
recovering the sample fragments as biodeterioration advances.
Other methods are often used to complement this type of analysis,
such as observing surface erosion with microscopic techniques and
changes in physicochemical properties [10, 11].
Other techniques enable the identification of low-molar mass
fragments, changes in the chemical structure of the polymer itself,
and the presence of degradation products resulting from biofrag-
mentation, such as size-exclusion chromatography (SEC), nuclear
magnetic resonance (NMR), and Fourier-transform infrared spec-
trometry (FTIR). In this case, it is possible to access information
about sample biofragmentation [3, 11].
The aforementioned analyses certainly provide important per-
spectives for understanding biodegradation; however, they are not
sufficient to support claims on the biodegradability of a polymer
[1]. The occurrence of mineralization and, therefore, of the biotic
degradation process is only effectively assessed by monitoring O2
consumption and/or CO2 production under aerobic conditions
and, in the case of anaerobic degradation, the release of CH4.
These methods are called respirometric methods [3]. To avoid
erroneous conclusions and guarantee some real progress in the
field of biodegradable polymers, it is essential to carry out respi-
rometry. Note that biodegradation is not an intrinsic characteristic
of the material but depends on the conditions at which it occurs.
Thus, rigorous discussion is needed [1]. In addition to the different
possible media (soil, compost, seawater, etc.), the standards specify
parameters such as presence or absence of oxygen (aerobic or
anaerobic condition), pH range, relative humidity, and incubation
temperature (thermophilic condition, typically 58 °C; or
mesophilic, up to 30 °C). Thermophilic temperatures are normally

used in industrial composting plants or anaerobic digesters
[12, 13].
Respirometric methods are specified by international (ISO,
OECD) or regional (ASTM, CEN, etc.) regulatory agencies cover-
ing a wide range of time intervals, apparatuses, media, and condi-
tions [14]. Some examples are shown in Table 1. More information
on biodegradation standards can be found elsewhere [15, 16].
These standards also define the criteria (time and rate of min-
eralization) to certify a polymer as biodegradable in each culture
medium and environmental conditions. ASTM 6400-19, ISO
17088:2012, and EN 13432:2012 standards, for example, state
that for a polymer to be certified as biodegradable, at least 90% of
its carbon must be converted into CO2 within 6 months of biodeg-
radation, compared to the positive reference polymer or in absolute
terms, when in a simulated industrial composting medium. In
addition, disintegration and ecotoxicity criteria are also established.
In the case of biodegradation in soil, the European standard EN
17033:2018 for mulching films states that mineralization must
reach 90% within 2 years.
In most of the published studies, respirometry is conducted
under simulated industrial composting conditions [17]. Some
polymers are considered biodegradable only under thermophilic
conditions, as is the case with one of the most well-known polymers
in this class–poly(lactic acid), PLA [13]. However, installing indus-
trial composting plants is still not a reality in many locations, and
the waste collection and separation systems are often ineffective.
Annually, 400 Mt of plastic waste are produced worldwide and of
this amount, 58% ends up in landfills and in the environment
[11]. Therefore, the destination of most of the waste is common
soil, where the temperature does not normally exceed 30 °C (meso-
philic condition). In this context, respirometry systems in meso-
philic and aerobic conditions (common soil or domestic compost)
can be considered more realistic, although still scarce in the
literature [17].
Quantification of mineralization over time is often performed
by monitoring the evolution of CO2, using either continuous or
discrete measurements. To this end, different methods can be used:
(i) automated systems with direct analysis of evolved gases by means
of gas chromatography or infrared spectroscopy (direct measure-
ment respirometry, DMR)—readers can refer to Chapter 1 for
details on automated respirometry; (ii) by capturing CO2 in alka-
line solutions (such as Ba(OH)2, KOH, or NaOH) and quantifying
by titration (cumulative measurement respirometry, CMR); and
(iii) CO2 capture can also be carried out in absorption columns
with NaOH pellets and is measured by gravimetric measurement
respirometry (GMR), but the use of this technique is less frequent
in the literature [12].
Table 1
Biodegradability standards and certifications of polymer materials in different types of environments
Simulated
environment Standard Certification organizations
Industrial EM 13432, ASTM D5338 and ISO 14855 TUV € Austria (Austria), DIN CERTCO
composting for packaging; EN 14995 and ISO (Germany), Vinçotte (Belgium),
17088 for polymer materials in general; Biodegradable Products Institute
ASTM 6868 materials containing (BPI—USA), Japan BioPlastics
polymer components Association (JBPA—Japan), Finnish
Solid Waste Association (Finland)
Domestic prEN 17427, AS 5810 (Australia), NF T € Austria (Austria), DIN CERTCO
TUV
composting 51800 (France) (Germany), Vinçotte (Belgium)
Soil ASTM D5988, ISO DIN CERTCO (Germany), Vinçotte
17556, EN 17033 (Belgium)
Seawater OECD 306, ISO € Austria
Vinçotte (Belgium), TUV
16221, ASTM D6691 (Austria)
Landfills ASTM D5526
Anaerobic ISO 15985, ISO 14853, ASTM D5511
Digestion
Aqueous ISO 14851, ISO 14852, and ISO 14853 Japan BioPlastics Association (JBPA—
medium Japan)
Kale et al. [18] assessed the biodegradation of PLA bottles

under composting conditions using CMR and GMR systems. In
the former case, the experiment was developed by the researchers to
comply with ASTM D5338-15 and ISO 14855-1:2012 standards.
The GMR apparatus was purchased, and its construction was based
on the ISO 14855-2:2012 standard. The results showed consider-
able differences between the biodegradation curves obtained with
the different systems, which the authors attributed to the different
compost/sample ratios and sample sizes. In addition, it should be
mentioned that due to the configuration of the GMR device, the
adequate number of samples required, according to the ISO
14855-2 standard, could not be met.
In the study of Cadar et al. [19], two methods were used to
assess the biodegradation of PLA and its copolymers. In both cases,
CO2 was absorbed by alkaline solutions, one method with quanti-
fication by titrimetry and the other by elemental analysis of the
solution, in a partially automated way. The same compost was used
in both analyses, but the solutions were different: 0.125 M BaOH
and 0.05 M NaOH, respectively. The curves obtained showed
similar biodegradation behavior of the positive control and of the
polymers; however, the total absolute values at the end of the
period showed significant differences.
Comparing respirometry results of samples that do not belong

to the same set of tests is hampered by the many variables affecting
the biodegradation process [12]. Some of these cannot be strictly
controlled, such as the characteristics of the culture medium and
microorganisms present in this medium. As shown in the study by
Castro-Aguirre et al. [12], even with the use of the same system
(apparatus) and methodology, the activity of the pure compost
(evolution of CO2) varied with its physicochemical properties
and, consequently, from where and when it was taken.
This can be seen in Table 2, which presents the results of the
biodegradation of microcrystalline cellulose (MCC) in the form of
powder under the conditions set by ISO 14855-1:2012 and ASTM
D5338-15, that is, compounded at a mass proportion of 6:1 in
relation to the sample and temperature of (58 ± 2) °C. In the same
way, procedures in accordance with the standards used in studies
with PLA films are listed in Table 3. In this case, additional variables
are included, such as characteristics of the PLA, film thickness, and
the compost/sample ratio, which in most of the studies diverged
from the value specified by the standards.
Table 2
Biodegradation data of cellulose in an industrial composting environment reported in the literature,
using different respirometry methods
Biodegradation Time
[%] [d] Method References Standard Origin of the compost
72.4–82.5 45 DMR [14] ISO 14855- 3-month-old industrial
1 plant compost
78 85 DMR [20] ISO 14855- 2- to 3-month-old industrial
1 plant compost
83 110 DMR [21] ISO 14855- 2-month-old municipal
1 organic waste compost
76 110 CMR [19] ISO 14855- 3-month-old organic
1 domestic waste compost
83 110 CMR with [19] ISO 14855- 3-month-old organic
automated 1 domestic waste compost
measuring
70 46 CMR [22] ASTM 3-month-old municipal
D5338 organic waste compost
90 100 CMR [23] ASTM Not mentioned
D5338
74 56 CMR [24] ISO 14855- Industrial plant compost
1
94.34 120 CMR [25] ASTM Agriculture waste compost
D5338
Table 3
Biodegradation data of poly(lactic acid) (PLA) in an industrial composting environment reported in the literature, using different respirometric
methods
Biodegradation Time Compost/ Thickness

[%] [d] Method References Standard Origin of the compost sample [mm] PLA
45 75 DMR [26] ISO 14855-1 Commercial compost 15-1 0.2 Ingeo 3001D
adapted
80 90 CMR [27] ISO 14855-1 Municipal solid waste 85-15 0.5 4032D
compost
69–72 110 CMR [19] ISO 14855-1 3-month-old organic 6-1 Not Synthesized and
domestic waste compost mentioned commercial
83–86 110 CMR with [19] ISO 14855-1 3-month-old organic 6-1 Not Synthesized and
automated domestic waste compost mentioned commercial
measuring
94.2 136 DMR [28] ASTM Prepared by authors 7-5 0.45 2003D
D5338
85.75 120 CMR [25] ASTM Organic agriculture waste 6-1 0.3 Commercial
D5338 compost grade
Assessing Polymer Biodegradability via Nonautomated Respirometry
33
Results from laboratory aerobic biodegradation tests using

conventional (nonautomated) respirometry allow one to estimate
the biodegradability of polymer materials using soil or compost as a
culture medium and, thus, simulate the material’s aerobic biodeg-
radation behavior when discarded in a terrestrial environment or
discarded and destined to a composting plant (municipal, indus-
trial, or domestic). For soil biodegradation, the test temperature
may vary between 20 and 28 °C, simulating a condition that can be
easily found worldwide. On the other hand, biodegradation in
compost must be carried out at (58 ± 2) °C, which is the standard
temperature in the composting plants, as the microorganisms in
these environments are thermophilic. The validation of the tests
will depend mainly on the microbial activity of the culture medium.
Therefore, the choice of culture medium is important for the
success of the test, primarily in relation to the choice of the soil if
the option is biodegradation in soil. If the test is conducted in
compost, this concern is less significant, as this substrate is normally
supplied with significant microbial activity. Comparatively, the bio-
degradation tests in compost are significantly more intense than in
soil, even if the selected soil is extremely active (fertile), due to the
amount and thermophilic characteristics of the microorganisms
that compose the compost [29–35].
The Materials and Methods addressed in this chapter are based
on the international standards ASTM D5338-15, ASTM D5988-
18, ASTM D6400-19, ISO 14855-1: 2012, and ISO 17556:2019.
2 Materials
2.1 System for the Glass vessel with internal volume between 2 and 4 L, glass lid with
Development of sealing ring (elastomer), and lock that guarantees the system is
Biodegradation in airtight. Inside this vessel, a 250-mL Erlenmeyer flask, a 100-mL
Large Volumes beaker, and a perforated plate (glass or porcelain) will be fitted. The
dimension of the perforated plate must be such to fit in the con-
tainer and to hold the Erlenmeyer flask and the beaker. The culture
medium is deposited on the bottom of the vessel and the perforated
plate must be held by supports (glass or porcelain) at a minimum
free height (i.e., above the culture medium) of 1 cm. A representa-
tive image of this system is shown in Fig. 1. See the required
amounts in Subheading 3.1 of this protocol.
2.2 System for the Simplified Bartha Respirometer consisting of a 250-mL Erlenmeyer
Assessment of flask and a 125-mL test tube (40 mm in diameter and 100 mm in
Biodegradation in height) connected to one another by a tube soldered on the side of
Small Volumes both constituents, forming, together with two rubber stoppers, an
airtight system. An illustrative image of this system is shown in
Fig. 1. See the required amounts in Subheading 3.2 of this
protocol.
Fig. 1 Flowchart for the assessment of polymer biodegradability via titrimetry in both large- and small-volume
samples
Note (1): this system can be purchased ready-made. Compared to

the previous system, it presents two drawbacks: low amount of
culture medium (50 g) and sample for the test (between 0.4 and
2 mg per gram of culture medium) and increased possibility of
errors in the analysis by titrimetry due to the need to transfer the
analytical solution.
2.3 Biodegradation Darkened temperature-controlled chamber (between 20 and 28 °C

Test Chamber for soil and 58 °C for compost, maintaining the selected tempera-
ture at ±2 °C).
Note (2): The volume of the chamber must be sufficient to hold

all the systems (highlighted in Subheadings 2.1 or 2.2) of the same
test, as it is not recommended to separate the systems of the same
test in different chambers. Test temperature and darkness condi-
tions should be equal.
2.4 Apparatus for See ASTM D5988-18.

Analysis of the Sample
and Culture Medium
2.5 Culture Medium See Subheadings 3.3.1 (soil) and 3.3.2 (compost) of this protocol.
2.6 Positive MCC with average particle size less than 20 μm should be used as a
Reference Polymer positive reference for biodegradation. Thermoplastic starch can also
be used as positive reference.
2.7 Negative Low-density polyethylene (LDPE) can be used as a negative refer-

Reference Polymer ence. Other polyethylenes, such as high-density polyethylene
(HDPE), can also be used as negative reference.
2.8 Test Polymer Characteristics described in Subheading 3.4 of this protocol.
2.9 Materials for 2.9.1 Conical polypropylene funnel (250 mL in volume, top
Determining the diameter of 120 mm, stem diameter of 19 mm, stem length
Residual Moisture of 81 mm, and cone height of 89 mm): three units.
Content and the Field 2.9.2 Ring stand: one unit.
Capacity 2.9.3 Iron ring (diameter of 7 cm) with clamp: one unit.
2.9.4 Narrow-mouth Erlenmeyer flask (250 mL in volume): three
units.
2.9.5 Pharmaceutical grade cotton ball: ca. 100 g.
2.9.6 Glass Petri dish 100/15 mm: three units.
2.9.7 Large rectangular stainless-steel tray (ca. 40 cm long, 30 cm
wide, and 4 cm deep or greater): two units.
2.9.8 Plastic bucket (polyethylene or polypropylene) of at least

5 L, with lid: one unit.
2.9.9 Oven for drying at 105 °C: one unit.
2.9.10 Desiccator (minimum size of 5 L): one unit.
2.9.11 Semi-analytical balance (minimum accuracy: 0.01 g):
one unit.
2.9.12 Beaker (500 mL in volume): one unit.
2.9.13 Beaker (150 mL in volume): one unit.
2.9.14 Common laboratory utensils (spatulas, glass rods, etc.).
2.10 Materials for 2.10.1 Glass Erlenmeyer flask (250 mL in volume): one unit per
Biodegradation system under test.
Monitoring (Titrimetry) 2.10.2 Graduated glass burette (100 mL, for the system
described in Subheading 2.1 of this protocol) or automatic
titrator: two units.
2.10.3 Graduated glass burette (50 mL, for the system described
in Subheading 2.2 of this protocol) or automatic titrator:
two units.
2.10.4 Volumetric glass pipette (100 mL, for the system
described in Subheading 2.1 of this protocol): one unit.
2.10.5 Volumetric glass pipette (10 mL, for the system described
in Subheading 2.2 of this protocol): one unit.
2.10.6 Polypropylene syringe (10 mL): one unit.
2.10.7 Urinary catheter size 6 Fr (outer diameter of 2 mm):
one unit.
2.10.8 Pipette filler bulb with a three-way valve system: two units.
2.10.9 Ring stand: two units.
2.10.10 Burette clamps: four units.
2.10.11 Glass beaker (1 L): one unit.
2.10.12 Stopwatch (analog or digital): one unit.
2.11 Reagents for 2.11.1 CO2-free distilled water.

Biodegradation
Testing and
Monitoring
Note (3): To remove CO2 from distilled water, boil the water for
30 min and place the warm water in a glass vessel (Erlenmeyer flask,
e.g.) equipped with an ascarite valve/filter (CO2 absorber) to cool.
After cooling, close the valve, keeping it closed when not in use.
CO2-free water that is not used should be discarded at the end of
the reagent preparation procedures. Whenever CO2-free water is
needed, it should be prepared on the day of use.
2.11.2 Potassium hydroxide (KOH) or sodium hydroxide

(NaOH) as 0.5 N solution in CO2-free distilled water.
Prepare 1–2 L of solution at a time.
Note (4): Alternatively, a 0.25 N barium hydroxide [Ba(OH)2]

solution may be used; however, this solution may promote forma-
tion of a barium carbonate (BaCO3) film, preventing diffusion of
CO2 in the alkaline solution. Therefore, KOH is more suitable. Yet,
the use of KOH involves the preparation of a 1 N barium chloride
(BaCl2) solution, which is used to precipitate the CO2 from the test
in the form of the stable BaCO3 salt (see reactions in Subheading
3.6 of this protocol).
Note (5): Standardize the alkaline solution (KOH or NaOH or Ba

(OH)2) against a 0.5 N solution of potassium acid phthalate (item
2.11.3 of this protocol) using methyl red (two drops) as an indica-
tor; standardization should be carried out before each CO2 deter-
mination. Note that 0.5 N is the nominal concentration of the
alkaline solution, which must be confirmed or corrected by finding
the correction factor ( fcor), using a standard acid solution (in this
protocol: potassium acid phthalate).
2.11.3 Standard solution of 0.5 N potassium acid phthalate in

CO2-free distilled water. Prepare 1 L of solution at a time.
2.11.4 Methyl red indicator solution.
2.11.5 Hydrochloric acid (HCl) solution at 0.25 N in CO2-free
distilled water. Standardize this solution against a 0.25 N
sodium carbonate solution using methyl red (two drops)
as indicator. Prepare 1–2 L at a time.
Note (6): 0.5 N is the nominal concentration of the acid solution,

which must be confirmed or corrected by finding the correction
factor ( fcor), using a standard alkaline solution (in this protocol:
sodium carbonate).
2.11.6 Sodium carbonate (Na2CO3) 0.25 N standard solution in

CO2-free distilled water. Prepare 1 L at a time.
2.11.7 Barium chloride (BaCl2) solution in CO2-free distilled
water at 1 N. Prepare 200–500 mL at a time.
2.11.8 Phenolphthalein indicator solution.
Note (7): All reagents used must meet purity standards in accor-
dance with ASTM D5988-18 (item 7.1). The reagents must be
kept in airtight packaging and stored away from light and under
controlled temperature (between 20 and 25 °C).
3 Methods
3.1 System for the 3.1.1 The tests should be carried out in triplicate and, therefore,
Assessment of three systems will be required for the control (containing
Biodegradation in only culture medium, also called “blank”), three systems for
Large Volumes the positive reference (culture medium and sample of a
biodegradable polymer, such as MCC), three systems for
each sample to be tested, and three systems for the negative
reference.
Note (8): The ASTM D5988-18 standard does not describe the
use of a system with a negative reference polymer; however, it is
important to prepare a triplicate of this system to compare the
biodegradation of the test polymer both with a positive reference
polymer (e.g., MCC) and with a negative reference polymer (e.g.,
LDPE) and, thus, to determine the relative biodegradability, posi-
tive or negative, of the test polymer. The control has the function of
discounting both the production of CO2 from the organic matter
present in the culture medium and the CO2 introduced in the
system during aeration after each titration.
Note (9): The ASTM D5988-18 standard calls for a triplicate set
called “technical control,” which serves to check the airtightness of
the system and to subtract the CO2 introduced during aeration.
However, from our point of view, this “technical control” is not
necessary, as the CO2 introduced during aeration has already been
subtracted with the triplicate “control” and the airtightness of the
vessels must be tested before the beginning of the experiments, as
during the experiment this test is not possible. If any of the systems
shows suspicious CO2 determination, the experiment must be
restarted, as there may have been a leak in the system.
Note (10): instead of the “technical control” system, prepare

three systems (blank monitoring) equal to the blank for monitoring
the loss of moisture in the biodegradation systems (by determining
the mass of the systems at each titration). In the titration, these
systems must be subjected to the same procedures of the other
systems with two differences: first, these systems should not be
titrated and, thus, the alkaline solution does not need to be
changed and, second, the mass of the system must be determined
at each titration step (monitoring) to check for moisture loss in the
system. If moisture is not within the range determined by this
protocol (Subheading 3.5.4 of this protocol), this should be cor-
rected by adding CO2-free distilled water.
Note (11): this system may present relative difficulty in its con-
struction, but it has two advantages: the system can be used in tests
with relatively larger amount of culture medium (between 100 and
500 g) and test sample (between 0.4 and 2 mg per gram of culture
medium) and eliminates possible errors due to analytical solution
transfers in the titrimetric analysis.
Note (12): For the glass vessel (internal volume between 2 and
4 L) with glass lid and airtight sealing system, the parts that cannot
be made of glass (sealing, e.g.) must be made of material that
neither adsorbs nor absorbs and is impermeable to CO2. The
container must have an opening that allows manipulation within.
Note that desiccator containers are not recommended, as the bio-
degradation process, despite oxygen consumption, may generate
positive pressure inside the system and the desiccator is not
designed to operate with positive pressure. Desiccators may be
used if changes are made (adaptation of an elastomeric sealing
ring and fastening clips on the cover, e.g.) to guarantee airtightness
under conditions of positive pressure.
Note (13): The ASTM D5988-18 standard indicates the use of a

150-mL beaker (to contain 100-mL of alkaline solution) and a
100-mL beaker (to contain 50 mL of distilled water) for each
biodegradation system. However, we recommend replacing the
150-mL beaker with a 250-mL Erlenmeyer flask (Subheading 2.1
of this protocol), because 100 mL of alkaline solution is a relatively
high volume to be contained in a 150-mL beaker that will after-
ward be used as the titration flask. Even with great care, spillage of
solution may occur (due to agitation in the titration procedure),
causing errors in the results.
3.2 System for the Alternatively, the use of a biodegradation system that involves
Assessment of reduced amounts of culture medium and sample may be used.
Biodegradation in The tests should also be carried out in triplicate and, therefore,
Small Volumes the number of systems is the same as described in Subheading 3.1
of this protocol. In this system, the culture medium is deposited on
the bottom of the Erlenmeyer flask and the reagent (alkaline solu-
tion) is deposited in the side tube. This system is described in
Subheading 2.2 of this protocol.
3.3 Culture Medium The procedures for selecting, collecting, and determining soil prop-
erties for use as culture medium can be performed in accordance
3.3.1 Soil
with ASTM D5988-18, item 9.
Note (14): Note that the definition and procedures for determin-
ing the moisture-holding capacity (MHC), also called field capacity
(FC), may introduce errors in soil moisture during the biodegrada-
tion test and adversely affect the tests. If soil moisture content
during the biodegradation test is lower than the amount
recommended by the standards, the microbial activity of the soil

will be reduced, and biodegradation will be compromised. On the
other hand, if the soil moisture during the biodegradation test is
higher than the indicated amount, the microbial activity will also be
reduced and, depending on the amount of excess water, aerobic
biodegradation may change to anaerobic biodegradation (flooded
soil), simulating a swamp. Therefore, the correct determination and
interpretation of FC is important for the proper execution of the
aerobic biodegradation tests.
Note (15): As an alternative to the FC determination methods

presented in ASTM D5988-18, this protocol presents a more
accessible and practical method for determining residual moisture
(RM) and FC, as described next.
3.3.1.1 Preparation of the Soil for RM and FC Testing

After executing items 9.1 and 9.3 of the ASTM D5988-18 stan-
dard, separate approximately 1 kg of soil, divide into approximately
equal parts, deposit in the stainless-steel trays, and allow to dry for
24 h in a temperature-controlled environment (between 20 and
25 °C) and without direct incidence of sunlight. After drying,
perform intense tumble-mixing of the soil in a plastic bucket with
lid. The resulting product is the test soil. If soil remains after the
determinations, save it for use, together with additional soil
obtained from ASTM D5988-18 (items 9.1 and 9.3), in the bio-
degradation tests.
3.3.1.2 Determination of RM
Add ca. 10 g of the test soil (mt) in a previously weighed and dried
Petri dish and store in a drying oven at 105 °C for 24 h. Then,
remove the Petri dish with dry soil from the oven and store in the
desiccator to cool down to room temperature. After cooling, deter-
mine the mass of the set and determine (subtracting the mass from
the Petri dish) the mass of dry soil (md). RM is the original moisture
of the soil (after drying for 24 h at room temperature) and is
expressed in grams of water (which the soil originally has) per
100 g of dry soil, according to Eq. 3. It is important to know
RM, as it is necessary to know how much water the test soil already
has, to be able to adjust the biodegradation test moisture. This
procedure should be performed in triplicate:
RM = ½ðm t - m d Þ:100]=m d ð3Þ
3.3.1.3 Determination of FC
Start the procedure by blocking the exit of the conical part of the
polypropylene funnel (250 mL) with cotton to prevent flow of soil
(use minimum necessary cotton to avoid occupying too much free
space of the conical part of the funnel). Then, fill approximately

70% of the free height left in the conical part of the funnel with the
test soil (Subheading 3.3.1.1), compacting it after every 2 cm of soil
layer, using the “repeating impact” technique, which consists of
releasing the funnel, containing the first layer of soil, in vertical
position against the iron ring (fixed on the ring stand) from a height
of approximately 10 cm between the iron ring and the exit of the
cone (opening blocked by cotton). Repeat this procedure ten times
for every 2 cm layer of added soil (CAUTION: perform the proce-
dure with the funnel always in vertical position and centered with
the iron ring to avoid accidental falls and loss of the procedure).
After filling and packing the soil in the funnel, place the funnel with
the soil in the Erlenmeyer flask and carefully add distilled water at
will, to soak the soil. After soaking the soil, wait for the water to
drain into the Erlenmeyer and, when draining stops (no dripping),
extract a sample of ca. 10 g of the soaked soil (mss) from the central
part of the funnel and add in a previously weighed Petri dish. Place
the Petri dish with the soaked soil in an oven for drying at 105 °C
for 24 h and then place the Petri dish and the dry soil in a desiccator
to cool down to room temperature. After cooling, determine the
mass of the set (subtracting the mass of the Petri dish) to obtain the
mass of dry soil (md). This procedure should be performed in
triplicate from the beginning, that is, a funnel for each determina-
tion. FC is defined as the amount of water that the soil can hold per
100 g dry soil, according to Eq. 4:
FC ðwater=100 g dry soilÞ = 100 . ðmss - md Þ=md ð4Þ

However, FC can also be expressed as the amount of water that
the soil can hold per gram of dry soil (Eq. 5):
FC ðwater=g dry soilÞ = ðmss - md Þ=md ð5Þ
3.3.2 Compost If compost is chosen as culture medium, see ASTM D5338-15

(item 9) for selection, characterization, and care related to the
culture medium.
Note (16): an adaptation related to the ASTM D5338-15 stan-

dard (item 9) must be performed. This adaptation is related to the
average particle size of the compost, which must be less than 2 mm,
requiring grinding of the compost in a rotary knife mill to reduce
particle size.
3.4 Test Specimen The characteristics that the test specimen (TS) must present are
specified by ASTM D5988-18, item 10 (if the culture medium is
soil) or by ASTM D5338-15, items 10 and 11.1, item 11.1.3
excluded (if the culture medium is compost).
Note (17): When the TS consists of pure polymer, or polymer

with a known composition of additives, the amount of carbon can
be determined depending on the chemical structure of the polymer.
As an example, we demonstrate the calculation (Eq. 6) of the
amount of carbon suitable for biodegradation for MCC (normally
used as a positive reference in polymer biodegradation tests).
Cellulose: (C12H20O10)n, molar mass of the repeating unit

324 g∙mol-1.
Hence: Xc = fraction of carbon in the cellulose molecule
X Carb
ðmolar mass of CarbonÞ × ðamount of Carbon in the moleculeÞ
=
ðmolar mass of polymer repeating unitÞ
12:12
X Carb = = 0:444 carbon in cellulose molecule ð6Þ
324
Consequently, the amount of carbon in the MCC TS is the
mass of the TS multiplied by the carbon fraction in the cellulose
molecule (Xcarb). Use the same procedure to calculate the amount
of carbon in the polymer molecule (or other organic material),
either pure or having other constituents (known quantity), which
will be subjected to the biodegradation test.
3.5 Biodegradation Comply with ASTM D5988-18, item 11.1.

Procedure
3.5.1 Number of Systems
per Sample
Note (18): It is important to guarantee the biodegradation sys-

tems are airtight (Subheadings 2.1 or 2.2 of this protocol) before
the start of the test. One should not expect the technical control to
perform this task, because if there is a “leak” during the test, the
test must be discarded. It should also be mentioned that the CO2
present in the aeration air of the systems holding the samples
(positive, negative, and test) is discounted (calculations) by the
system containing only culture medium (blank soil or compost).
3.5.2 Amount of Culture Note (19)–soil: For systems with a large internal volume (Sub-
Medium heading 2.1 of this protocol), comply with ASTM D5988-18, item
11.2. For systems with a small internal volume (Subheading 2.2 of
this protocol), use (50.0 ± 0.1) g.
Note (20)—compost: For systems with a large internal volume

(Subheading 2.1 of this protocol), comply with ASTM D5338-15,
item 11.1.3. For systems with a small internal volume (Subheading

2.2 of this protocol), use (50.0 ± 0.1) g at the same mixing condi-
tions described in item 11.1.3 of ASTM D5338-15. Adjust the
carbon/nitrogen ratio according to item 11.1.3 of ASTM D5338-
15 for the two system sizes (Subheadings 2.1 and 2.2 of this
protocol).
Note (21): The samples (positive, negative, and test) that will be
subjected to the biodegradation test must have similar mass (preci-
sion of 0.1 g), whether for large or small internal volume.
Note (22): The culture medium that will be used in the biodegra-
dation tests of the samples (positive, negative, and test) must be
extracted from the same batch of culture medium that has been
subjected to the tests on determination of properties (Subheading
3.3 of this protocol, item 9 of ASTM D5988-18, and item 9 of
ASTM D5338-15).
3.5.3 Carbon/Nitrogen Perform in accordance with ASTM D5988-18, item 11.3.

Ratio of the Culture
Medium
Note (23): Alternatively, the use of the “as collected” culture

medium is allowed (Subheadings 3.3.1 and 3.3.2 of this protocol)
and, thus, the biodegradation test will provide results closer to the
environmental reality.
Note (24): For compost, see ASTM D5338-15 (items 10.2 and
11.1.3).
3.5.4 Adjusting Moisture 3.5.4.1 If the culture medium is soil, perform in accordance with
Content of the Culture ASTM D5988-18, item 11.4.
Medium
Note (25): The procedures for determining RM and FC described

in this protocol (Subheadings 3.3.1.2 and 3.3.1.3) are equivalent
to those of the ASTM D2980-17 standard and, therefore, adjust
the soil moisture content between 50% and 70% of the determined
FC (or MHC).
3.5.4.2 If the culture medium is compost, adjust the moisture

content of the system so that the total mass of the culture
medium consists of 50 wt% dry solids (compost plus test
sample) and 50 wt% distilled water.
3.5.5 Recording the Record the mass of each biodegradation system (Subheading 2.1 or
Mass of Each 2.2 of this protocol) containing the culture medium with the
Biodegradation System adjusted moisture for monitoring soil moisture loss.
3.5.6 Incubation 3.5.6.1 If the culture medium is soil, perform in accordance with
ASTM D5988-18, item 11.6.
Note (26): The ASTM D5988-18 standard recommends between
0.4 and 2.0 mg of sample per gram of culture medium. We suggest
using the mean value of this interval, that is, 1.2 mg of sample per
gram of culture medium.
3.5.6.2 If the culture medium is compost, prepare a homoge-

neous mixture of dry compost and dry sample using a
compost:sample mass ratio of 6:1, considering the
amount of culture medium for each biodegradation sys-
tem size (Subheading 3.5.2 of this protocol).
3.5.7 Selection of the 3.5.7.1 If the culture medium is soil, select the chamber temper-
Test Temperature ature between 20 and 28 °C, maintaining the selected
temperature at ±2 °C.
3.5.7.2 If the culture medium is compost, select the temperature
of 58 °C, maintaining this temperature at ±2 °C.
3.5.8 Closing of the 3.5.8.1 Definitions

Systems and Initiating At the beginning of the biodegradation (time zero), the alkaline
Biodegradation solution is the last component added to each of the systems before
closing. The systems must be closed one at a time, appreciating a
time interval defined as the aeration time (taer). The aeration time
depends on the titration manipulation time (tman). It is, therefore,
important to define each of the following times:
. tman: time interval related to titration manipulations (opening
the system, removing the alkaline solution, titration, recording
the result, and restoring the total volume in the titration
burette). If the system is that addressed in Subheading 2.2 of
this protocol, consider the time for the triple washing.
. taer: time interval during which the systems must remain open
and exposed to the environment for oxygen renewal within the
systems. This time should range between 15 and 60 min (ASTM
D5988-18).
Note (27): During the first taer, which takes place at time zero of
the biodegradation (onset of biodegradation), no titration manip-
ulation occurs, only aeration and closing sequence of the systems
and the closing sequence at the start of biodegradation should be
observed in each titration procedure throughout the
biodegradation test.
3.5.8.2 Determination of tman and taer

To define all time intervals (i.e., tman and taer), a titrimetric manipu-
lation test is performed using a set with at least three systems
containing only the alkaline solution. The test consists of simulat-
ing the titration procedure that occurs throughout the biodegrada-
tion process. The titrimetric manipulation test aims to determine
the appropriate tman for the titration operation and to allow the
operator to become familiar with good laboratory practices, titra-
tion techniques, and manipulation of the titration instruments.
Having determined tman, then taer can be determined. It is recom-
mended that taer be 5 min longer than tman as long as the result of
the sum of tman plus 5 min lies within the interval between 15 and
60 min.
3.5.8.3 To close and start biodegradation for a system with a large

internal volume (Subheading 2.1 of this protocol), comply with
ASTM D5988-18 item 11.7.
Note (28): It is recommended that the 150-mL beaker (for hold-

ing 100 mL of alkaline solution) be replaced by a 250-mL Erlen-
meyer flask in the large volume system (Subheading 3.1 of this
protocol), as 100 mL of solution is a relatively high volume for a
150-mL beaker that will afterward be used as titration flask. Even
with great care, spillage of solution may occur, due to agitation in
the titration procedure, introducing errors in the titration result.
3.5.8.4 To close and start biodegradation for a small internal

volume system (Subheading 2.2 of this protocol), add 10 mL of
0.5 N KOH solution (item 2.11.2 of this protocol) in the side test
tube of the system (simplified Bartha). For this system, there is no
need to add distilled water. After adding the alkaline solution, seal
the system and store in a dark, temperature-controlled chamber.
Note (29): Regardless of system size (Subheading 2.1 or 2.2 of

this protocol), the alkaline solution should be the last component
incorporated into the system (one system at a time), following a
procedure that consists of (i) identifying and organizing the sys-
tems; (ii) selecting the first system and starting the timer in count-
down mode for a time interval equal to taer; (iii) at the end of taer,
add the alkaline solution and close the system; and (iv) repeat steps
(ii) and (iii) for the next system, and so on up to the last system. Use
the organizational sequence in the titrimetric procedure
(Subheading 3.5.9.3).
3.5.9 CO2 Analysis 3.5.9.1 Caution with BaCO3 Film Formation

Comply with ASTM D5988-18, item 11.9.1.
Note (30): If the alkaline solution of choice is KOH 0.5 N, there

are no concerns about the formation of a BaCO3 film in the
150-mL beaker (system with large internal volume) or in the side
test tube of the simplified Bartha (system with small internal vol-
ume). However, BaCl2 should be used in the titration.
3.5.9.2 Interval Between Titrations

3.5.9.2.1 Soil: When the culture medium is soil, it is recom-
mended to carry out the first titration approximately
12 h after the beginning of the biodegradation test,
because at the beginning of the test the organic material
present in the soil is activated (oxygenated and
hydrated) due to its handling during the preparation
procedures (sieving, homogenization, and moisture
adjustment). When the period between titrations
reaches 3 d, comply with ASTM D5988-18, item
11.9.2.
Note (31): It is important to know the maximum volume of HCl
solution (Vmax) required to neutralize the original alkaline solution
(KOH, or other) (Subheading 3.2 of this protocol), that is, the
alkaline solution that has not been subjected to the biodegradation
test or any other form of contact with CO2. If, after a certain
biodegradation interval (BI), the volume of HCl titrating solution
equals Vmax, this means that no CO2 was produced during this
BI. On the other hand, if this volume equals zero (the turning point
occurs when adding the indicator to the alkaline solution present in
the biodegradation system), this means that during this BI the
amount of CO2 produced has saturated the alkaline solution. If
this happens, the biodegradation test must be discarded and a new
test must be prepared.
Note (32): In the titration procedures, carried out after each BI, it
is recommended that the amount of HCl solution required to
neutralize the alkaline solution that remained after the respective
BI be in the range of 30–70% of Vmax. In the initial titration (first
BI, i.e., 12 h), the volume of HCl solution may not comprise this
range, but it should not equal zero. The titration of the first BI
(12 h) will be the reference for the second BI and this, in turn, will
be the reference for the third BI and so on, so that the volume of
HCl solution required to neutralize the alkaline solution remains in
the range between 30% and 70% of Vmax. If the volume of HCl
titration solution is greater than 70% of Vmax, BI should be
increased, and if less than 30%, BI should be reduced.
3.5.9.2.2 Compost: When the culture medium is compost, it is

also recommended to perform the first titration after
12 h from the start of the biodegradation test due to the
reasons mentioned in step 3.5.9.2.1. However, com-
posts have a significantly higher microbial activity than
soils and, therefore, the 12-h intervals between titra-
tions may be repeated for a longer period when com-
pared to soil. When the period between titrations
reaches 3 d, comply with ASTM D5988-18, item
11.9.2.
Note (33): Regardless of the biodegradation system (large or
small internal volume), titration of the alkaline solution (after an
incubation period) will be against the 0.25 N HCl solution using a
burette.
3.5.9.3 Titrimetric Procedure

3.5.9.3.1 At the end of each incubation period, perform titrime-
try of the alkaline solution that absorbed (pre-existing
and generated) CO2 in the systems (large or small
internal volume).
3.5.9.3.2 For large internal volume systems (Subheading 2.1 of
this protocol) and 0.5 N KOH solution, before titration
add, via burette, 20 mL of 1 N BaCl2 solution in the
250-mL Erlenmeyer flask containing the alkaline solu-
tion to cause precipitation, in the form of stable BaCO3,
of all the CO2 absorbed by the KOH solution, avoiding
displacement of CO2 during titration. Hence, only the
KOH that did not absorb CO2 will be titrated against
the 0.25 N HCl solution.
3.5.9.3.3 For systems with a small internal volume (Subheading
2.2 of this protocol) and 0.5 N KOH solution, before
titration add, via burette, 2 mL of 1 N BaCl2 solution in
the 250-mL Erlenmeyer flask that will receive the alka-
line solution stored in the side test tube of the simplified
Bartha. As previously mentioned, BaCl2 will cause pre-
cipitation, in the form of stable BaCO3, of all the CO2
absorbed by the KOH solution, avoiding displacement
of CO2 during titration and, hence, only the KOH that
has not absorbed CO2 will be titrated against the
0.25 N HCl solution.
3.5.9.3.4 Add phenolphthalein indicator to the titration flasks:
four drops in the flask for the large internal volume
system (Subheading 2.1 of this protocol) and two
drops in the flask for the small internal volume system
(Subheading 2.2 of this protocol).
3.5.9.3.5 For large internal volume systems (Subheading 2.1 of

this protocol), perform the titration directly in the
250-mL Erlenmeyer flask containing the alkaline solu-
tion, to which 20 mL of the 1 N BaCl2 solution and the
phenolphthalein indicator have already been added.
3.5.9.3.6 For small internal volume systems (Subheading 2.2 of
this protocol), transfer (using the urinary catheter
attached to the 10-mL polypropylene syringe) the alka-
line solution (which is in the simplified Bartha side test
tube) to the titration flask (250-mL Erlenmeyer flask),
which should already contain 2 mL of 1 N BaCl2 solu-
tion and phenolphthalein.
Note (34): Regardless of the system used (Subheadings 2.1 or 2.2
of this protocol), titration must follow a standard procedure, which
consists of (i) starting the timer in the countdown mode with the
taer, opening the first system of the organizational sequence (see
Note 28), and removing the alkaline solution to be titrated;
(ii) during the aeration time countdown, perform titration, prepare
the insertion of the new alkaline solution (250-mL Erlenmeyer flask
containing 100 mL of alkaline solution for large volume system or
10-mL polypropylene syringe containing alkaline solution to be
added to the side tube of the small volume system) and prepare
the next system for titration; (iii) after the aeration time, insert the
alkaline solution into the titrated system and close it; and (iv) repeat
items (i) to (iii) for the next system in the organizational sequence
(see Note 28).
Note (35): Regardless of the biodegradation system used (Sub-

headings 2.1 or 2.2 of this protocol), titration must be performed
quickly to prevent CO2 absorption by the alkaline solution from the
environment wherein the procedure will be performed. To guaran-
tee the success of the test, before starting the biodegradation, the
operator should be knowledgeable about the titration procedures
and precautions.
3.5.9.4 Total Incubation Period

Comply with ASTM D5988-18, item 11.10.
3.6 Calculations The calculations for determining the amount of CO2 produced by
the test polymer and the polymers used as positive and negative
reference are based on the stoichiometry of the reactions involved
in the process and on the titration outcome. Firstly, it is necessary to
know the biodegradability potential of the polymer, which means
the maximum amount of CO2 the material can generate. This
amount is defined as the theoretical amount of CO2 (ThCO2)
produced [mg]. ThCO2 is calculated based on the amount of carbon
in the polymer. To simplify this protocol, we describe the calcula-

tions using cellulose as an example. In the biodegradation process:
Cpresent in cellulose þ O2 → CO2 þ H2 O þ Cbiomass ð7Þ
Therefore, one mole of carbon produces one mole of CO2, that
is, 12 g of carbon produces 44 g of CO2. If Y mg of carbon, from
the polymer, is added to the biodegradation system, this amount
Y can produce, at most, the following amount of CO2:
ThCO2 ½mg] = Y . ð44=12Þ ð8Þ
As demonstrated in Subheading 3.4 of this protocol (Eq. 6),
the molar fraction of carbon in the cellulose molecule (Xcarb) is
0.444.
If one adopts, as an example, system 2.2 of this protocol
(simplified Bartha), the amount of culture medium is
(50.0 ± 0.1) g. Adopting the sample/culture medium ratio of
1.2 mg sample per gram culture medium, 60 mg MCC will be
incubated and the amount of carbon (Y) in 60 mg MCC is:
Y = X carb . mass of sample = 0:444 . 60 = 26:64 mg ð9Þ
Hence:
ThCO2 = Y . ð44=12Þ = 26:64 . ð44=12Þ = 97:68 mg ð10Þ
Therefore, 97.68 mg is the maximum amount of CO2 that
60 mg MCC (26.64 mg of carbon) can produce in the biodegrada-
tion system throughout the test. Therefore, 97.68 mg is the bio-
degradability reference of cellulose in the biodegradation test.
Note (36): If the 60 mg MCC produces 97.68 mg CO2, the

biodegradation of MCC will be 100% in relation to the production
of CO2; however, this is a utopian result, because, in addition to
CO2, microorganisms also produce biomass (cell reproduction
process) from carbon. The individual determination of CO2 and
biomass in the same aerobic biodegradation system is complex due
to biomass, which is intermixed with the culture medium, making
quantification difficult. Therefore, this protocol considers the bio-
degradability of MCC (or another positive reference polymer, e.g.,
thermoplastic starch) as the biodegradation reference for the test
polymers, considering only the data obtained with the quantifica-
tion of CO2 in the biodegradability calculation of the test polymer
and positive reference polymer. Hence, the biodegradability of the
test polymer will be relative to the biodegradability of the positive
reference polymer.
To proceed with the example of calculating biodegradation, it

is important to know the reactions involved in the titration process,
which start with absorption, by the alkaline solution (we will adopt
KOH), of the CO2 generated in the biodegradation system:
CO2 þ H2 O → H2 CO3 ð11Þ
H2 CO3 þ 2 KOH → K 2 CO3 ð12Þ
Therefore, one mole of KOH is consumed by half a mole of
CO2.
Because potassium carbonate, just like sodium carbonate, is not
a stable salt, BaCl2 is added before titration to shift the equilibrium
toward the formation of barium carbonate:
BaCl2 þ K 2 CO3 → 2 KCl þ BaCO3 # ð13Þ
So, the KOH that has not been consumed by CO2 in each of
the biodegradation systems (blank, positive reference polymer,
negative reference polymer, and test polymer) is titrated using the
HCl solution. It is at this moment that we highlight the importance
of biodegradation systems containing only culture medium (blank),
since the difference between the titration volume of the HCl solu-
tion of the blank system and the system containing the test sample
(test polymer, positive reference polymer, and negative reference
polymer) is exactly the volume of the HCl solution equivalent to
the KOH that was consumed by the CO2 produced by biodegrada-
tion of the test sample only.
Note (37): Note that in the blank system the culture medium
produces CO2 and the aeration air within the system also contains
CO2, whereas in the test systems CO2 is produced by the culture
medium and by the test sample and is also present in the aeration
medium, that is, the difference is the test sample (Eq. 14):
V bio = V blank - V test sample ð14Þ

where:
Vbio = volume of the HCl solution related to the amount of CO2
produced by the test sample.
Vblank = volume of HCl solution to titrate the blank system.
Vtest sample = volume of the HCl solution to titrate the system
containing the test sample (test polymer, positive reference
polymer, and negative reference polymer).
We already know that one mole of CO2 consumes two moles of
KOH. Regarding the neutralization reaction of KOH and HCl, one
mole of HCl neutralizes one mole of KOH:
KOH þ HCl → KCl þ H2 O ð15Þ
From Eqs. 12, 13, and 15, the stoichiometry involving KOH,
HCl, and CO2 is 1:1:1/2, respectively. As the titration data yield
the volume of the HCl solution and we want to calculate the
amount of CO2 produced, from this point on we are only interested
in the relationship between HCl and CO2. Hence:
1 HCl → § CO2 , i:e:, 36:5 g HCl → 22 g CO2 ð16Þ
We already have the correlation between mass HCl and our
unknown (CO2) and hence we can calculate the mass of CO2 using
normality (N):
N = ðm: = Þ= mol:V ðLÞ → m = N :mol:V ðLÞ == ð17Þ
where:
* = for acids, it is the number of ionizable hydrogens (for HCl, e.g.,
* = 1).
mol = molar mass of the molecule (for HCl, mol = 36.5 amu).
V(L) = volume of solution [L] and, in our case, V[L] = Vbio [L].
Therefore:
m = N . 36:5 . V bio½L] ð18Þ
Note that the HCl solution used is 0.25 N, but 0.25 N is the
nominal concentration (Nnom). To know the final concentration of
the HCl solution, this solution should be standardized against a
standard Na2CO3 solution (Subheading 2.11.5 of this protocol)
and this standardization will generate a correction factor that may
be different from one. Therefore, the final concentration of the
solution of HCl will be:
N f = N nom . f cor ð19Þ
Hence:
m = 36:5 . N nom . f cor . V bio ðLÞ ð20Þ
From stoichiometry:
36:5 g HCl → 22 g CO2 ð21Þ
36:5 . N nom . f cor . V bio ðLÞ → y g CO2 ð22Þ
Hence:
y ½g] = 22 . N nom . f cor . V bio ðLÞ ð23Þ
or:
y ½mg] = 22 . N nom . f cor . V bioðmLÞ ð24Þ
Thus, from Eq. 24, one can calculate the mass of CO2
[mg] produced by a sample in a biodegradation test, whether it is
the test, the positive reference, or the negative reference polymer.
The titration is periodic, and each determination of CO2 must be

accumulated throughout the biodegradation test and the sum at
the end of the six-month test will determine the amount of CO2
(X [mg]) produced by the biodegradation of the different polymers
tested (test, Xt; positive reference, Xp; and negative reference, Xn).
The data can be presented in the form of tables and/or figures
according to Subheading 3.8 of this protocol. Note that the tests
are performed in triplicate and, therefore, all statistical procedures
of reliability (95%) must be performed using the results of the three
tests of each test and blank sample.
3.7 Interpretation of The standards (ASTM D5338-15, ASTM D5988-18, ASTM

the Results D6400-19, ISO 14855-1:2012, and ISO 17556:2019) that sup-
ported this protocol do not consider which amount of carbon from
the tested samples (test, positive, and negative) is used to produce
biomass (Eq. 1) and, therefore, it is not possible to determine the
absolute biodegradability without knowing the amount of biomass
produced. This protocol, therefore, proposes the term “relative
biodegradability” of the test sample (Biort), which consists of
determining the biodegradability of the test sample relative to the
biodegradability of the positive reference sample (e.g., MCC) over
6 months. The mathematical result of Biort should be determined
from the amount of CO2 (Xt [mg]) accumulated during the test by
the sample relative to the amount of CO2 (Xr [mg]) accumulated
during the test by the positive reference. Hence:
Biort = X t =X r ð25Þ
%Biort = ðX t =X r Þ . 100 ð26Þ
In accordance with ASTM D6400-19, the polymer under test
will be considered biodegradable if Biort or %Biort is equal to or
greater than 0.90 or 90%, respectively.
3.8 Report Comply with ASTM D5988-18, item 14.

3.8.1 Soil
3.8.2 Compost Comply with ASTM D5338-15, item 14.
4 Final Considerations
Respirometric polymer biodegradation tests are, in general, time-

consuming and laborious procedures. The use of automatic systems
significantly eases these obstacles (see Chapter 1 for details); how-
ever, access to automated biodegradation respirometric systems is
restricted due to the relatively high cost of the system. On the other
hand, correct execution of polymer biodegradation tests using
hand-operated respirometry reported in this protocol enables

reproducible and accurate results at relatively low costs, using sim-
ple laboratory systems and routines. The accuracy and, mainly, the
reproducibility of the results, associated with the low cost, are
essential ingredients to globalize the biodegradation tests by
means of nonautomated respirometry.
References
1. Zumstein MT, Narayan R, Kohler HPE et al 30:567–573. https://doi.org/10.1179/
(2019) Dos and do nots when assessing the 1743284713Y.0000000503
biodegradation of plastics. Environ Sci Technol 11. Chamas A, Moon H, Zheng J et al (2020)
53:9967–9969. https://doi.org/10.1021/ Degradation rates of plastics in the environ-
acs.est.9b04513 ment. ACS Sustain Chem Eng 8:3494–3511.
2. Emadian SM, Onay TT, Demirel B (2017) https://doi.org/10.1021/acssuschemeng.
Biodegradation of bioplastics in natural envir- 9b06635
onments. Waste Manag 59:526–536. https:// 12. Castro-Aguirre E, Auras R, Selke S et al (2017)
doi.org/10.1016/j.wasman.2016.10.006 Insights on the aerobic biodegradation of poly-
3. Lucas N, Bienaime C, Belloy C et al (2008) mers by analysis of evolved carbon dioxide in
Polymer biodegradation: mechanisms and esti- simulated composting conditions. Polym
mation techniques - a review. Chemosphere Degrad Stab 137:251–271. https://doi.org/
73:429–442. https://doi.org/10.1016/j. 10.1016/j.polymdegradstab.2017.01.017
chemosphere.2008.06.064 13. Kliem S, Kreutzbruck M, Bonten C (2020)
4. Vert M, Doi Y, Hellwich K-H et al (2012) Review on the biological degradation of poly-
Terminology for biorelated polymers and mers in various environments. Materials (Basel)
applications (IUPAC recommendations 13:4586–4604
2012). Pure Appl Chem 84:377–410. 14. Way C, Wu DY, Dean K, Palombo E (2010)
https://doi.org/10.1351/pac-rec-10-12-04 Design considerations for high-temperature
5. Laycock B, Nikolić M, Colwell JM et al (2017) respirometric biodegradation of polymers in
Lifetime prediction of biodegradable polymers. compost. Polym Test 29:147–157. https://
Prog Polym Sci 71:144–189. https://doi.org/ doi.org/10.1016/j.polymertesting.2009.
10.1016/j.progpolymsci.2017.02.004 10.004
6. Bastioli C (ed) (2005) Handbook of biode- 15. Rudnik E (2008) Biodegradability testing of
gradable polymers. Smithers Rapra Press, compostable polymer materials. In: Rudnik E
Shrewsbury (ed) Compostable polymer materials. Elsevier
7. Nikolic MS, Djonlagic J (2001) Synthesis and Science, Amsterdam
characterization of biodegradable poly(butyl- 16. Funabashi M, Ninomiya F, Kunioka M (2009)
ene succinate-co-butylene adipate)s. Polym Biodegradability evaluation of polymers by
Degrad Stab 74:263–270. https://doi.org/ ISO 14855-2. Int J Mol Sci 10:3635–3654.
10.1016/S0141-3910(01)00156-2 https://doi.org/10.3390/ijms10083635
8. Haider TP, Völker C, Kramm J et al (2019) 17. Rudnik E, Briassoulis D (2011) Degradation
Plastics of the future? The impact of biodegrad- behaviour of poly(lactic acid) films and fibres in
able polymers on the environment and on soci- soil under Mediterranean field conditions and
ety. Angew Chemie - Int Ed 58:50–62. laboratory simulations testing. Ind Crop Prod
https://doi.org/10.1002/anie.201805766 33:648–658. https://doi.org/10.1016/j.
9. Bonilla J, Paiano RB, Lourenço RV et al (2020) indcrop.2010.12.031
Biodegradability in aquatic system of thin 18. Kale G, Auras R, Singh SP, Narayan R (2007)
materials based on chitosan, PBAT and Biodegradability of polylactide bottles in real
HDPE polymers: Respirometric and physical- and simulated composting conditions. Polym
chemical analysis. Int J Biol Macromol 164: Test 26:1049–1061. https://doi.org/10.
1399–1412. https://doi.org/10.1016/j. 1016/j.polymertesting.2007.07.006
ijbiomac.2020.07.309 19. Cadar O, Paul M, Roman C et al (2012) Bio-
10. Banerjee A, Chatterjee K, Madras G (2014) degradation behaviour of poly(lactic acid) and
Enzymatic degradation of polymers: a brief (lactic acid-ethylene glycol-malonic or succinic
review. Mater Sci Technol (United Kingdom) acid) copolymers under controlled composting
conditions in a laboratory test system. Polym 27. Luo Y, Lin Z, Guo G Biodegradation assess-
Degrad Stab 97:354–357. https://doi.org/ ment of poly (lactic acid) filled with functiona-
10.1016/j.polymdegradstab.2011.12.006 lized titania nanoparticles (PLA/TiO 2) under
20. Petinakis E, Liu X, Yu L et al (2010) Biodegra- compost conditions. https://doi.org/10.
dation and thermal decomposition of poly(lac- 1186/s11671-019-2891-4
tic acid)-based materials reinforced by 28. Kalita NK, Sarmah A, Bhasney SM et al (2021)
hydrophilic fillers. Polym Degrad Stab 95: Demonstrating an ideal compostable plastic
1704–1707. https://doi.org/10.1016/j. using biodegradability kinetics of poly(lactic
polymdegradstab.2010.05.027 acid) (PLA) based green biocomposite films
21. Weng YX, Wang XL, Wang YZ (2011) Biodeg- under aerobic composting conditions. Environ
radation behavior of PHAs with different Challeng 3:100030. https://doi.org/10.
chemical structures under controlled compost- 1016/j.envc.2021.100030
ing conditions. Polym Test 30:372–380. 29. Tate RL (2021) Soil microbiology, 3rd edn.
https://doi.org/10.1016/j.polymertesting. Wiley, New Jersey
2011.02.001 30. Elsas JD, Trevors JT, Rosado AS, Nannipieri P
22. Calil MR, Gaboardi F, Guedes CGF, Rosa DS (eds) (2019) Modern soil microbiology, 3rd
(2006) Comparison of the biodegradation of edn. CRC Press, Boca Raton
poly(ε-caprolactone), cellulose acetate and 31. ASTM (2018) D5988-18: standard test
their blends by the Sturm test and selected method for determining aerobic biodegrada-
cultured fungi. Polym Test 25:597–604. tion of plastic materials in soil. American Soci-
https://doi.org/10.1016/j.polymertesting. ety for Testing and Materials, West
2006.01.019 Conshohocken
23. Pradhan R, Reddy M, Diebel W et al (2010) 32. ASTM (2015) D5338-15: standard test
Comparative compostability and biodegrada- method for determining aerobic biodegrada-
tion studies of various components of green tion of plastic materials under controlled com-
composites and their blends in simulated aero- posting conditions, incorporating thermophilic
bic composting bioreactor. Polycaprolactone temperatures. American Society for Testing
Int J Plast Technol 45–50. https://doi.org/ and Materials, West Conshohocken
10.1007/s12588-010-0009-z 33. ASTM (2019) D6400-19: standard specifica-
24. Du YL, Cao Y, Lu F et al (2008) Biodegrada- tion for labeling of plastics designed to be aer-
tion behaviors of thermoplastic starch (TPS) obically composted in municipal or industrial
and thermoplastic dialdehyde starch (TPDAS) facilities. American Society for Testing and
under controlled composting conditions. Materials, West Conshohocken
Polym Test 27:924–930. https://doi.org/10. 34. ISO (2019) 17556:2019: plastics - determina-
1016/j.polymertesting.2008.08.002 tion of the ultimate aerobic biodegradability of
25. Leejarkpai T, Suwanmanee U, Rudeekit Y, plastic materials in soil by measuring the oxy-
Mungcharoen T (2011) Biodegradable kinetics gen demand in a respirometer or the amount of
of plastics under controlled composting condi- carbon dioxide evolved. International Organi-
tions. Waste Manag 31:1153–1161. https:// zation for Standardization, Geneve
doi.org/10.1016/j.wasman.2010.12.011 35. ISO (2012) 14855-1:2012: plastics - evalua-
26. Brdlı́k P, Borůvka M, Borůvka B, et al (2021) tion of the ultimate aerobic biodegradability
Biodegradation of poly(lactic acid) biocompo- and disintegration under controlled compost-
sites under controlled composting conditions ing conditions—method by analysis of released
and freshwater biotope. https://doi.org/10. carbon dioxide. International Organization for
3390/polym13040594 Standardization, Geneve
Chapter 3
Detection and Identification of Microplastics in Food

and the Environment
Walter R. Waldman, Cristiane Vidal, Mariana A. Dias,
Victor Z. Resende, and Cassiana C. Montagner
Abstract
Microplastic contamination is a relevant topic in Food and Environmental Sciences. Microplastic analyses in
matrices such as food, soil, sediment, water, and air require a specific method approach according to their
respective aims and scope. This chapter presents a comprehensive discussion about standard practices
currently applied to ensure representative sampling, adequate sample preparation, and unequivocal identi-
fication and characterization of microplastic particles.
Key words Microplastic contamination, Quality assurance, Quality control, Representative

sampling, Density separation, Digestion, Chemical composition, Visual inspection, Microscopy,
Spectroscopy.
1 Introduction
Plastic is an extensively used material in daily life and an obvious

primary source of microplastics called by range size of 1 μm to
5 mm. Almost a hundred years ago, the industry discovered the
advantages of plastics, and the sector was heavily influenced by it
ever since. Transportation and storage of food and beverages are
safer and cheaper using plastics rather than glass or metal. Nowa-
days, research is primarily focused on environmentally friendly
alternatives to replace or minimize plastic use, especially single-
use plastics that have been demonstrated to negatively impact the
environment because of inadequate waste disposal over the years.
Biodegradable materials in composting conditions or in other
environments—soil, sea water, among others (see Chapters 1 and
2 on biodegradability)—have been studied as compelling alterna-
tives. However, conventional plastics are low-cost, and their physi-
cochemical and mechanical properties, especially the low weight
and easy processability, are exceptional for the packaging industry.
https://doi.org/10.1007/978-1-0716-3613-8_3,
57
58 Walter R. Waldman et al.
Over the last years, toxicologists have aimed to understand the

human risk associated with plastic and microplastic exposure, such
as inhalation and ingestion routes of exposure [1]. In general,
polymers are nontoxic for humans, but the additives and plasticizers
can be toxic depending on exposure level. Readers are invited to
check Chapters 7 and 8 on toxicity assessment of food packaging
materials. Chemicals migrating from polymer bulk to the surface
and then to the food were the main concern for food contamina-
tion from the packaging so far [2, 3]—see Chapter 6 on migration
of constituents of food contact materials. In addition, despite plas-
tic resistance, mechanical stimuli such as shearing, folding, and
fatigue can wear surfaces and produce fragments that might go
into the food [4].
Scientific research has consistently shown the presence of
microplastics in different types of food such as salt, honey, mineral
water, etc. [5–10], their leaching from take-out containers [11], or
microplastic generation by simple tasks such as opening food pack-
aging (plastic containers, bags, tapes, caps) using scissors, tearing
with hands, cutting with knives, or twisting manually [4]. Thus,
primary, secondary, or tertiary packaging can be sources of micro-
plastics depending on the contact with the food and the type of
food (dry solid foods, pasty food, beverages, etc.). Finally, regard-
ing in natural foods, contamination with microplastic is minimal
and, if existent, it may be due to environmental contamination (air,
soil, and water) during the cultivation [12–14].
Once disposed in the environment, the packaging can be a
source of microplastics. Secondary microplastics are formed by
the breakdown/degradation of macroplastics or primary micro-
plastics in the environment (weathering exposure). Such small
particles can be transported to long distances, carrying chemical
contaminants or microorganisms. Today, microplastics are consid-
ered ubiquitous contaminants in the environment. Poor solid waste
management causes soil and water contamination, both important
compartments for food production [12, 13].
This chapter introduces an analytical protocol to extract, iso-
late, characterize, and chemically identify microplastics. Many pro-
cedure steps are common for both food and environmental
matrices. However, food and environmental samples are intrinsi-
cally complex (see Note 1). The sample particularities (aqueous,
viscous, solid, oleaginous, powder, etc.) will justify whether a step is
necessary or not. There are no official methods yet, and accurate
quantification is still challenging. Further details on the differences
and similarities in analytical protocols for food and environmental
samples are highlighted in the Notes section.
Figure 1 shows a didactic analytical workflow that synthesizes
the topics discussed herein regarding contamination control, sam-
pling, sample preparation, and characterization.
Microplastics from Food Packaging 59
Fig. 1 Analytical workflow presented in this chapter. SEM: scanning electron microscopy; EDS: energy-
dispersive X-ray spectroscopy; IR: infrared; py-GC-MS: pyrolysis-gas chromatography mass spectrometry;
TED-GC-MS: thermal extraction-desorption gas chromatography mass spectrometry
2 Materials
We consider as a minimal set of resources for assembling a general

microplastic detection laboratory or setup:
• A laminar flow hood, to keep contamination out of the separa-
tion systems.
• A centrifugation system to separate better and faster by density.
• A separating funnel to separate the denser and lighter fractions.
• A vacuum filtration system to accelerate filtration when the filter
might clog by the size of the dispersed phase.
• An analytical balance to weigh the samples accurately (precise to
0.1 mg [15]) and the filters to keep track of the mass balance.
• Specific chemicals and apparatuses depending on the method of
choice, as detailed in the following sections.
3 Methods
3.1 Microplastic First, this analytical protocol will address an important step to
Background guarantee the quality of results that will be generated. To avoid
Contamination microplastic contamination in sampling, sample preparation, and
analysis, the analyst should pay attention to three main topics: air
contamination (i.e., exposure to airborne microplastic contamina-
tion), cross-contamination (i.e., lab material contaminated during
sampling, sample preparation or analysis), and contamination con-

trol (i.e., measures to promote quality assurance and quality
control) [16].
• Reducing air contamination:
– Avoid personal protective equipment (PPE) manufactured of
synthetic fibers (e.g., polyester); use preferably natural fibers
(e.g., cotton). Cellulosic fibers can be digested if needed (see
Note 2).
– Always record the color and material of the clothes used
during lab assays. This practice can facilitate the identification
of possible contamination sources.
– If possible, perform all lab activities in clean rooms with
controlled air circulation and limited staff access.
– Use laminar flow hood or similar during all lab work.
– Maintain scheduled cleaning of the lab, at least once a week.
– Whenever possible, cover all working solutions, samples, fil-
ters, and other related materials with precleaned aluminum
foils. In the case of flasks, cap whenever not in use.
– Provide proper storage to all lab materials.
• Reducing cross-contamination:
– Use preferably lab materials made of glass and metals.
– Wash lab materials with ultrapure or filtered distilled water
before use.
– Record sample information, sampling and sample prepara-
tion order, and adopted protocols to identify possible sources
of cross-contamination.
– All inner surfaces of the laminar flow hood or bench surfaces
must be cleaned with residue-free detergent or ethanol fol-
lowed by ultrapure or filtered distilled water before use.
– Perform proper washing of lab materials between samples as
stated above.
– Filter all working solutions. Purity reagents p.a. must also be
filtered (see filter and conditions in Subheading 3.3.2).
– Filters composed of stainless steel or glass materials can
undergo heat treatment (e.g., 450 °C for 3 h) to cleanse
possible microplastic cross-contamination. On the other
hand, materials such as natural or synthetic fibers (e.g., cellu-
losic or polyamide) may not be thermally treated. In this case,
compressed air can be used for decontamination, and the
removal efficacy can be performed using optical microscopy.
• Performing contamination control:

– Collect field controls during sampling.
– Prepare blanks during sample preparation (e.g., filtration).
– Collect open-air background to identify possible air contami-
nation. Procedures for possible airborne microplastic con-
tamination can be found at Zhang et al. [17] and Chen
et al. [18] (see Note 2).
3.2 Sampling The sampling protocol is dependent on the type of samples. Thus,
several procedures may be found in the literature [19, 20]. However,
the lack of standard procedures leads to relevant differences in
sampling and sample processing steps. Therefore, microplastic
abundance may be expressed in several different ways (e.g., fibers
per cubic meter, microplastics per liter, microplastic mass per sam-
ple mass, pellets per square meter, etc.) (see Note 3). It is vital to
keep the representative sampling in accordance with the analysis,
methodology, and amount of collected material [19, 21]. This
chapter addresses microplastic detection in food (solid or liquid)
and environmental matrices (water and soil/sediment).
3.2.1 Food Matrices Lower-viscous (e.g., bottled water, wine, beer, refreshments, and
milk) and higher-viscous (e.g., honey, syrup, and ketchup) liquid
samples can be collected from different batches, brands, or places.
Viscous samples are usually diluted after sampling and before treat-
ment. In general, a volume of 350–750 mL has been used for wine
sampling [22]. About 500 mL to 2 L of the sample have been used
for microplastics analysis for bottled water [5, 23, 24]. One liter of
warm milk samples can be filtered in a dried precleaned glass
vacuum system [25]. Liquid honey has been diluted in warm
water and passed through a steel sieve [7, 26, 27].
For the sampling of solid matrices in food, defining the sample
size is essential to obtain representative samples since they may not
be homogeneous. Often, samples can be collected from different
batches in several places to enhance representativeness.
Microplastics can contaminate animals destined for human
consumption (e.g., fish, crustaceans, and mussels). Such particles
may come from habitat surroundings (environment sample) or
food processing (food sample). Samples can be stored until analysis
under refrigeration in chemical preservatives (e.g., formaldehyde
and ethanol). Generally, the digestive tract of these animals has
been investigated for microplastics [28]. More information about
biota sampling and sample preparation can be found in Hermsen
et al. [29] and Lusher et al. [28].
3.2.2 Environmental A wide range of equipment (trawl nets, pumps, Niskin bottles, etc.)
Matrices has been used to collect microplastics in aquatic environmental
samples (freshwater, seawater, drinking water, and groundwater
from rivers, lakes, estuaries, and oceans). This sampling may be
performed in the surface or depth water according to the aim of
the study. The examples of equipment and sampling details can be
found in the works of Prata et al. [19] and Silva et al. [30].
Similar to water, soil/sediment sampling requires specific
equipment to collect microplastics in the environment. Surface
sampling from marine sediment, river sediment, soil, and sandy
beaches has been conducted using steel materials (spatula, twee-
zers, or spoon). Conversely, depth samples have been collected
using grabs, augers, or corers. More detailed information can be
found in Möller et al. [31] and in Yang et al. [32]. For detection of
microplastics in atmospheric air (indoor and outdoor), we direct
the reader to Zhang et al. [17] and Chen et al. [18].
3.3 Sample Sample preparation is the process of extracting microplastics from

Preparation their respective matrices. A single step, like a filtration, shall be
enough for clear samples like drinking and bottled water. However,
as the matrix complexity increases, it is necessary to combine differ-
ent sample preparation steps, such as dissolving or digesting the
matrix, using separation methods to isolate the microplastics from
the matrix constituents. The protocol order is matrix-dependent.
In cases where microplastic is trapped in samples (e.g., biota, pro-
cessed food), digestion may be the first step to promote the release
of the microplastic (for more comments on sample disassembling,
see Note 4). Alternatively, in samples from water, salt, sugar, etc.,
this step usually occurs after density separation [28]. Commercial
salt samples, for instance, can be directly dissolved in water and
submitted to sample processing (density separation and digestion,
if necessary). Warm water can be used to facilitate dissolution [8–
10, 33–35]. Such a step has also been applied to commercial sugar
and solid honey [7].
3.3.1 Digestion Several sample processing by digestion can be performed to elimi-

nate or minimize the presence of organic matter. Among the
options, the most used are acid/alkaline reagents (e.g., HNO3,
HCl, NaOH, and KOH) and oxidizing agents (e.g., Fenton’s
reagent and hydrogen peroxide (H2O2) solution). Peroxide oxidiz-
ing digestion has been considered more effective to degrade natural
organic matter than acidic or alkaline due to less damage being
caused to the polymer. Enzymatic digestions are also adopted,
leading to a reduction in polymer degradation compared to acid/
alkaline processing. On the other hand, enzymes are costly and
differences between biological matrices promote irregularly enzy-
matic activities [21].
Solid matrices composed of carbohydrates (e.g., honey and
sugar), inorganic salts (e.g., sea salt), and proteins (e.g., fish,
mussel, shrimp, and bivalve) are usually pretreated using oxidizing

agents as H2O2 or Fenton’s reagent and inorganic acids or bases
[28]. The following procedure is based on the NOAA guidelines
and literature specialized in microplastic extraction for these matri-
ces [15, 28, 36, 37]. If necessary, literature shows that filtration
steps combined with digestion steps would enable correct micro-
plastic extraction [21].
Liquid matrices are mainly beverages, including soft drinks,
beers, wines, and bottled and tap water. Beverage samples usually
have low organic content and particulate matter. Thus, filtration as
pretreatment can be adopted without digestion steps. However,
some matrices with high organic content (e.g., milk or liquid
honey) may require digestion steps to eliminate fats, carbohydrates,
and larger proteins [21].
• H2O2 and Fenton’s Reagent Oxidizing Digestion/Acid or Alka-
line Digestion
CAUTION: Hydrogen peroxide solutions and respective
mixtures are highly reactive. Safety precautions are necessary to
avoid accidents in the laboratory. Please follow and review labo-
ratory safety measures to comply with this reagent use.
– Use a dried precleaned glass flask to determine the mass of
dried solids after the filtration steps. Always use an analytical
balance to the nearest 0.1 mg. Liquid matrices or biological
soft tissues, in some cases, cannot undergo dried solids deter-
mination. Whenever the case, proceed directly to digestion
and use defined metrics in volume, mass, or size to identify
the samples. The correct identification will allow microplastic
counting by measure.
– Aqueous Fe(II) and H2O2 solutions, used separately or
jointly, acid or alkaline solutions can be added to remove
organic matter in sequence. Strong acid or alkaline digestion
might damage the microplastic structure. Thus, possible
destruction of microplastics or incorrect results should be
considered [28] (see Notes 5 and 6).
– Accurate reaction conditions will vary according to the matrix
under analysis or microplastic intended to obtain:
– Usual concentrations:
Oxidizing digestion agents: 15–35 wt% H2O2 solutions;
0.05 M Fe(II) aqueous solution.
Acidic or alkaline digestion agents: 10 wt% KOH solutions,
22.5 M or 65–100 wt% HNO3 solutions, 68 wt% HClO4
solution.
– Usual reaction time: 30 min up to 72 h. Longer reaction
periods of 7 to 10 d may be used to eliminate all biological
soft tissue [28].
– Magnetic agitation: 80–120 rpm. Orbital shakers may

be used.
– Reaction temperatures:
Oxidizing digestion agents: 40–85 °C.
Acidic or alkaline digestion agents: 20–100 °C.
Room temperature and incubators may be used.
– After aqueous Fe (II) solution and/or H2O2 solution, or
acid/alkaline solutions addition, let the mixture stand on
the lab bench at room temperature for some time before
proceeding to the next step.
– Cover the flask with a watch glass or cap, add the magnetic
stir bar, and heat up to the defined temperature.
– If the reaction exhibits excessive gas bubbles at the surface,
remove the beaker from the hotplate and place it in the
laminar flow hood until boiling decreases. In case of overflow,
ultrapure or distilled filtered water can be added to slow
down the reaction.
– Magnetic agitation or shaking is necessary during the
reaction.
– If natural organic matter is still visible at the end of the
reaction, add more digestion agent solution and repeat the
procedure above.
– After oxidizing or acid/alkaline digestion, ultrapure or dis-
tilled filtered water and salts may be added for re-dissolution,
filtration, and density separation steps (Subheading 3.3.2). In
some cases, a manual pick-up can be used in glass Petri dish
substrates.
3.3.2 Density Separation Density separation is a simple step for isolating microplastics from
other sample compounds, generally inorganic materials (e.g., sand)
through salt solutions with known densities. Microplastics are
separated by density from heavier fractions like minerals and undis-
solved impurities, organic matter, and even other microplastics (see
Note 7). High-density salt solutions (1.2–1.8 g cm-3; e.g., NaBr,
NaCl, NaI, ZnBr2, and ZnCl2) are added to the samples to separate
the denser fraction of the media from the microplastics. The sepa-
ration of the two fractions (low- and high-density fractions) can be
performed by gravity or centrifugation systems.
NaCl is beneficial because it is inexpensive, nontoxic, and
highly soluble in water. However, the density of NaCl solution is
about 1.2 g cm-3 (Table 1) which is lower than the densities of
some polymers like poly(ethylene terephthalate) (PET), poly(vinyl
chloride) (PVC), and poly(lactic acid) (PLA). Thus, such polymers
cannot be separated using a NaCl solution, and other solutions
must be used to solve this limitation. NaI, ZnCl2, and ZnBr2
Table 1
Densities and costs of salt solutions used in density separation for
microplastics [19, 20, 38] (see Note 9)
High-density salt solution Density [g cm-3] Cost per 100 g [US$]a

NaBr 1.37 46.10
NaCl 1.2 6.79
NaI 1.6–1.8 88.70
ZnBr2 1.7 51.40
ZnCl2 1.5–1.8 27.40
a
Quotes for the United States dated from March 26, 2021 [39]
solutions have higher densities, expanding the range of polymers

that can be separated. However, these salts have a higher cost, and
they are environmentally unfriendly due to their toxicity [19, 20,
38] (see Note 8). It is worthy of highlighting that the densities on
Table 1 are not all related to saturated solutions because there is a
tradeoff between the density and the viscosity so, if needed, higher
densities can be achieved but at the expenses of longer times for
decantation and filtration.
Density separation presents some limitations, such as the inter-
ference of organic matter in samples, leading to overestimated or
underestimated results in density separation. For instance, an
organic film (biofilm) can adhere to a microplastic surface and
change the particle density.
Microplastic particles have different degradation degrees, mod-
ifying some polymer physical properties as wettability and crystalli-
zation [40]. Additive concentrations and adsorbed substances can
also change the processed polymer density [19].
• Add high-density salt solutions to the samples in a dried
pre-cleaned glass separating funnel. Samples may be previously
digested to reduce or eliminate the organic matter.
CAUTION: Use personal protective equipment when
handling NaBr, NaI, ZnB2, and ZnCl2 salts.
• Water-soluble samples, such as commercial salt, can be directly
dissolved in water.
• The sample may be manually shaken using a separating funnel.
However, the separation step can take several hours (ca. 24 h).
• Alternatively, the sample may be centrifuged using dried pre-
cleaned glass centrifuge tubes. In this case, the centrifugation
step lasts for a few minutes according to the rotation speed
(200–500 rpm).
• Particles with density higher than the salt solution will remain in
the precipitate, whereas particles with lower density will remain
in the supernatant (low-density fraction).
• After separation, the supernatant is filtered in a dried precleaned
glass vacuum system covered with precleaned aluminum foil in a
laminar flow hood (see Notes 10 and 11).
– Filter diameter: 47 mm.
– Filter porosity: 0.2–149 μm.
– Filter: cellulose nitrate, cellulose acetate, glass fiber, alumi-
num oxide, nylon, polycarbonate, or polytetrafluoroethylene.
• The filter is placed and sealed in dried precleaned Petri dishes
and dried at room temperature or in an oven (40–50 °C) for
5–12 h. Afterward, it may be directly submitted to the next step
of the sample preparation.
• The procedure of density separation may be repeated according
to the matrix complexity.
3.4 Characterization Microplastics are usually characterized by morphology (color, size,

shape, and surface texture), origin (primary or secondary), and by
chemical composition (polymer identification). Many techniques
are presented below, and their choice will depend on the objective
of the analysis, deeply influenced by the size of the microplastic. If
the microplastics have been extracted and are isolated, the techni-
ques can be chosen regardless of the original matrix they were
extracted from.
3.4.1 Objective: Sorting Visual inspection can be used to sort suspected microplastics
of Suspected Microplastics [41, 42]. However, this is a subjective technique for environmental
or Morphological/Origin samples. The decision depends on the analyst skills to observe the
Characterization particle features (texture, physical behavior, overall appearance) and
recognize them as plastic (see Note 12). When not instrument-
assisted, it is not suitable for small microplastics. For smaller parti-
cles, a visual inspection must be combined with optical microscopy.
To enhance the detection, microplastics can be stained with dyes
prior to visual inspection [43]. The dying process facilitates locat-
ing suspected microplastics via optical or fluorescence microscopy
(see Note 13).
During sorting with visual inspection, assisted or not with a
microscope, morphology is classified as size and shape. Shape clas-
sification includes pellets, sphere, hemisphere, grain, nurdle, fiber
(singular fiber, fiber bundle), and fragment (foam, film, angular/
sub-angular, rounded/sub-rounded).
Scanning electron microscopy (SEM) is used for surface analy-
sis and to evaluate signs of degradation [44] (see Note 14).
3.4.2 Objective: Polymer To perform microplastic confirmation and polymer identification,

Identification use vibrational spectroscopy (infrared or Raman), or thermal analy-
sis (pyrolysis, -py, or thermal desorption, TED) combined with gas
chromatography coupled to mass spectrometry (py-GC-MS or
TED-GC-MS) [43].
Spectroscopic Techniques
For spectroscopic techniques [45], microplastic identity confirma-
tion is done by spectral signatures related to each polymer by using
libraries or multivariate analysis (chemometric classification mod-
els). Quantification is performed by items (number of particles).
• Fourier-transform infrared (FTIR) spectroscopy
FTIR spectroscopy is the most used technique, and there is
plenty of instrumentation described below [46].
• ATR-FTIR: More suitable for a size range of 0.5–5 mm, which is

not likely to occur in food, but it is likely that food packaging
breaks down to that size range in the environment. No substrate
is required as the microplastic is individually pressed against the
crystal for analysis.
• FTIR microscopy: Usually referred as μFTIR, which is a micro-
scope coupled to the spectrometer. More suitable for microplas-
tics smaller than 0.5 mm until ca. 20 μm. Acquisition mode
(transmittance, reflectance, or ATR) will depend on the particle
size as well. Transmittance is the most used and compatible
substrates are metallic filters (e.g., aluminum oxide) or CaF2
plates. For microplastics displayed on Petri dishes, analysis can
be performed by reflectance or by the ATR objective (lens) as
well, but not in transmittance.
• Imaging FTIR microscopy: The technique is usually reported as
μ-FPA-FTIR. The spatial dimension added to the spectroscopic
analysis allows shape and size analyses, despite the chemical
composition. Instruments with focal plane array detector
(FPA) are faster than point scan FTIR microscopy made by
single element detectors [47].
• Raman spectroscopy
Suitable for size ranges from 10 μm to 5 mm, depending on
instrumentation and particle characteristics [48–50].
Main optimization may include laser wavelength, laser power,
and integration time to achieve proper signal-to-ratio spectra and
to avoid fluorescence or other interference from polymer constitu-
ents. Raman microscopy is suitable for small microplastics down to
ca. 3 μm, covering a size range not possible with FTIR
(<10–20 μm). Raman imaging is also possible, but it is a slow
analysis. Compatible substrates are metallic filters, but not

restricted to them, depending on the microplastic size.
Thermal Degradation-Based Techniques

The techniques py-GC-MS and TED-GC-MS are destructive, as
the samples are thermally degraded. The identity of microplastics is
confirmed through the evaluation of their degradation products.
These techniques are suitable for the detection of additives or
organic contaminants in the microplastics, as well as for the simul-
taneous analysis of different polymers, reaching detection and
quantification levels in the nanogram range. Quantification is per-
formed in mass-based concentration, upon construction of analyti-
cal curves of target microplastics [51–53].
• py-GC/MS
Suitable for microplastics down to 100 μm, if the particle is
previously isolated.
• TED-GC-MS
Suitable for bulk sample analysis and considered promising for
the analysis of nanoplastics. However, for bulk analysis, it is not
possible to directly assess microplastic size. Size can be known
indirectly by analyzing previously size-fractionated samples
(by sieving, e.g.) during sample preparation.
4 Notes
1. Plastic physical properties differ according to the two main

contexts outlined in this chapter: (i) plastics collected in the
environment exposed for a while to the weather will be more
brittle, crystalline, with a lower molar mass or crosslinked,
while (ii) plastics handled during the production process will
be more inert, amorphous, and closer to the raw resin proper-
ties. It might influence the output of some assessments, for
instance, the size of the microplastics after the separation pro-
cess because more brittle materials are also more likely to
fragment during the handling and the digestion step.
2. Cellulosic fibers are a common contamination for monitoring
microfibers, and some environments may have more than 80%
of the fibers as cellulosic ones [54]. In cases like that, an
additional step of selective cellulosic digestion might be tried
to reduce the burden of the characterization step if it is signifi-
cantly time-consuming, like the spectroscopic
identification [55].
3. When comparing your results with the literature, one topic is

always present: the microplastics abundance. Because the need
for quantification always answers a specific demand of knowl-
edge, sometimes the determination of the microplastics con-
centration is made as a function of the area (relevant, for
instance, for terrestrial epigeic species or the aquatic buoyant
positive microplastics) or as a function of the volume (relevant,
for instance, for terrestrial endogenic species or the aquatic
buoyant neutral microplastics). The numerator of the concen-
tration is a matter of dissent, with the number of particles and
weight as the predominant units. The microplastics amount
might be described by number, weight, or volume. Not so
often, volume can also be found describing the quantity of
microplastics. However, as there are no official or standardized
methods yet, quantification accuracy is low as important figures
of merit (recovery, repeatability, and reproducibility) are not
yet established for microplastics. Concentration results are
often estimative and not yet comparable.
4. The density separation step might change substantially accord-
ing to the complexity of the system to be treated. Drinking
products and samples from the aquatic bodies with a low
amount of organic matter can be filtered and then passed
through a digestion to clean the surface from, for instance,
biofilms. More complex samples, such as soil or some multi-
component solid food, might have microplastics heavily
entangled inside microstructures, like soil aggregates or pro-
cessed industrial food preparations. In that case, more aggres-
sive treatments as continued stirring or ultrasound might help
to disassemble organized soil structures to release the micro-
plastics to be separated by density. In the case of complex food,
where microplastics might come from the ingredients or the
process and be heterogeneously spread through the sample,
aggressive homogenization, like stirring, may be necessary. It
comes, obviously, at the expense of information such as size,
shape, and format since microplastic fragmentation can occur.
5. Nitric acid (HNO3) is the most effective digestion agent com-
pared to the others in terms of removing organic content.
However, high temperatures or concentrations could dissolve
or degrade polymers sensible to acid hydrolysis—for example,
polyamides, polyesters such as PET, and polycarbonates [28].
6. Potassium hydroxide (KOH) is proven to promote the dissolu-
tion of animal digestive tracts due to basic hydrolysis of chemi-
cal bonds presented in soft biological tissues, such as larger
proteins and fat [56].
7. The separation of suspected collected microplastics is one of
the main challenges for environmental researchers. Because the
origin and diversity of the plastics are unknown but likely to be

broad regarding type and weathering time, the protocols must
also follow broad criteria to separate the plastics from the media
where they are. Concerning density, just for the most used
plastics, there is a range from 0.9 to 1.4 g∙cm-3, therefore the
need for a general-purpose separation method. Regarding a
controlled environment, like the development or optimization
of new packaging, process, or even a quality control protocol, it
is possible to develop a process to monitor a specific micro-
plastic. A cost-effective way of doing that is fine-tuning the
density separation, using solutions right below and above the
range density of the chosen microplastic to perform the sepa-
ration before the digestion and the identification of
microplastics.
8. Regarding the separation by density, salts like ZnBr2, CaCl2,
sodium polytungstate, lithium metatungstate, among others,
one can optimize the process to have a faster or a better
separation according to the system where the microplastic is
[15, 57]. To choose the best salt for the separation step, one
must consider several features, namely: (i) the microplastic
density under analysis, usually unknown for environmental
samples but likely to be known for packaging analysis. Exam-
ples of especially denser plastics are PTFE (ca. 2.2 g mL-1),
polyvinylidene chloride (PVDC; ca. 1.6 g mL-1), or formula-
tions with inorganic materials like TiO2, glass fibers, or colored
pigments; (ii) the density of the media from which it is needed
to separate the microplastic; (iii) the chemical affinity between
the media where the microplastic is and the salt. One example
worthy of highlighting is the ZnCl2, which forms a hydro
soluble complex between the zinc and humid substances
[58, 59], increasing the density of the liquid media, and dark-
ening and contaminating the aqueous phase; (iv) salt hygro-
scopicity might lead to difficult handling if it is needed to
fabricate several solutions at a time; (v) mixing heat might be
intense, so the preparation must be a sensitive step regarding
safety if the salt has an exothermic dissolution. It is recom-
mended the analyst always test which salt offers the best
benefit-to-cost ratio, checking the separation with positive
controls of a known amount of the microplastic of interest in
the media chosen to quantify the microplastics.
9. High-density salt solutions can be filtered and reused more
than once. A dried precleaned pycnometer may be used to
determine the density of these solutions.
10. Filtration might be a time-consuming step if the system con-
tains small domains, like fine particles or unstable emulsions
and colloids, which may clog the filter pores, demanding vac-
uum filtration or even several filtration repetitions. Some
strategies to prevent clogging are shifting the heterogeneous

equilibrium of inorganic particulates using chelating agents
[6], using oxidants to decompose globules of fatty acid in
milk [60], or heat to decrease milk viscosity [60].
11. Chemical composition of the filter influences polymer charac-
terization. Choose the filter according to its compatibility with
the instrumental analysis performed in the next step (see
Subheading 3.4).
12. When monitoring a well-controlled environment, the com-
plexity of the unknown polymer types, which is the usual
environmental monitoring condition, can be avoided. In a
well-controlled environment with excellent contamination
control, the identification step can be simplified considering
that all microplastics come from specific process equipment or
packaging. In that case, it might be considered that the optical
microscopy or even visual inspection for larger microplastics
could be enough identification if there is enough color
contrast.
13. When the characterization step is optical microscopy, the risk of
confusing plastic fragments with organic matter must be
reduced using dyes with a remarkable interaction with nonpo-
lar structures rather than polar ones. The most used dye is the
Nile red [19, 61], which has a solvatochromic property, shift-
ing the color as a function of the amount of hydrophobicity on
the microplastics surface. Other dyes might also stain organic
particles, misleading the identification.
14. Energy-dispersive X-ray spectroscopy (EDS) [44], usually cou-
pled to microscopy SEM-EDS, is used for elementary analysis,
and it is not a polymer identification method, but it is a tech-
nique that can eliminate if a suspected particle is not a plastic
based on the elements ratio present. In other words, EDS may
not confirm if the particle is a microplastic, but it can confirm if
it is not, for example, if other elements are more abundant than
carbon content.
References
1. Campanale M, Savino L, Uricchio. (2020) A derived from plastic food containers. Compr
detailed review study on potential effects of Rev Food Sci Food Safe 11:453-470
microplastics and additives of concern on 4. Sobhani Z, Lei Y, Tang Y, Wu L, Zhang X,
human health. Int J Environ Res Public Health Naidu R et al (2020) Microplastics generated
17(4):1212 when opening plastic packaging. Sci Rep
2. Schmid P, Welle F (2020) Chemical migration 10(1):4841
from beverage packaging materials—a review. 5. Zuccarello P, Ferrante M, Cristaldi A, Copat C,
Beverages 6(2):37 Grasso A, Sangregorio D et al (2019) Exposure
3. Bang DY, Kyung M, Kim MJ, Jung BY, Cho to microplastics (<10 μm) associated to plastic
MC, Choi SM et al (2012) Human risk assess- bottles mineral water consumption: the first
ment of endocrine-disrupting chemicals quantitative study. Water Res 157:365–371
6. Oßmann BE, Sarau G, Holtmannspötter H, 20. Stock F, Kochleus C, B€ansch-Baltruschat B,

Pischetsrieder M, Christiansen SH, Dicke W Brennholt N, Reifferscheid G (2019) Sampling
(2018) Small-sized microplastics and pigmen- techniques and preparation methods for micro-
ted particles in bottled mineral water. Water plastic analyses in the aquatic environment – a
Res 141:307–316 review. TrAC Trends Anal Chem 113:84–92
7. Liebezeit G, Liebezeit E (2013) Non-pollen 21. Shruti VC, Pérez-Guevara F, Elizalde-
particulates in honey and sugar. Food Addit Martı́nez I, Kutralam-Muniasamy G (2021)
Contam Part A 30(12):2136–2140 Toward a unified framework for investigating
8. Yang D, Shi H, Li L, Li J, Jabeen K, Koland- micro(nano)plastics in packaged beverages
hasamy P (2015) Microplastic pollution in intended for human consumption. Environ
table salts from China. Environ Sci Technol Pollut 268:115811
49(22):13622–13627 22. Prata JC, Paço A, Reis V, da Costa JP, Fer-
9. Iñiguez ME, Conesa JA, Fullana A (2017) nandes AJS, da Costa FM et al (2020) Identifi-
Microplastics in Spanish table salt. Sci Rep cation of microplastics in white wines capped
7(1):8620 with polyethylene stoppers using micro-Raman
10. Karami A, Golieskardi A, Keong Choo C, spectroscopy. Food Chem 331:127323
Larat V, Galloway TS, Salamatinia B (2017) 23. Makhdoumi P, Amin AA, Karimi H,
The presence of microplastics in commercial Pirsaheb M, Kim H, Hossini H (2021) Occur-
salts from different countries. Sci Rep 7(1): rence of microplastic particles in the most pop-
46173 ular Iranian bottled mineral water brands and
11. Du F, Cai H, Zhang Q, Chen Q, Shi H (2020) an assessment of human exposure. J Water Proc
Microplastics in take-out food containers. J Eng 39:101708
Hazard Mater 399:122969 24. Mason SA, Welch VG, Neratko J (2018) Syn-
12. Crossman J, Hurley RR, Futter M, Nizzetto L thetic polymer contamination in bottled water.
(2020) Transfer and transport of microplastics Front Chem 6:407
from biosolids to agricultural soils and the 25. Kutralam-Muniasamy G, Pérez-Guevara F, Eli-
wider environment. Sci Total Environ 724: zalde-Martı́nez I, Shruti VC (2020) Branded
138334 milks – are they immune from microplastics
13. Huang Y (2020) Agricultural plastic mulching contamination? Sci Total Environ 714:136823
as a source of microplastics in the terrestrial 26. Liebezeit G, Liebezeit E (2015) Origin of syn-
environment. Environ Pollut 260:114096 thetic particles in honeys. Pol J Food Nutr Sci
14. Brahney J, Hallerud M, Heim E, 65(2):143–147
Hahnenberger M, Sukumaran S (2020) Plastic 27. Mühlschlegel P, Hauk A, Walter U, Sieber R
rain in protected areas of the United States. (2017) Lack of evidence for microplastic con-
Science 368(6496):1257–1260 tamination in honey. Food Addit Contam Part
15. Masura J, Baker J, Foster G, Arthur C (2015) A 34(11):1982–1989
Laboratory methods for the analysis of micro- 28. Lusher AL, Welden NA, Sobral P, Cole M
plastics in the marine environment. NOAA (2017) Sampling, isolating and identifying
16. Prata JC, Reis V, da Costa JP, Mouneyrac C, microplastics ingested by fish and inverte-
Duarte AC, Rocha-Santos T (2021) Contami- brates. Anal Methods 9(9):1346–1360
nation issues as a challenge in quality control 29. Hermsen E, Mintenig SM, Besseling E, Koel-
and quality assurance in microplastics analytics. mans AA (2018) Quality criteria for the analy-
J Hazard Mater 403:123660 sis of microplastic in biota samples: a critical
17. Zhang Y, Kang S, Allen S, Allen D, Gao T, review. Environ Sci Technol 52(18):
Sillanp€a€a M (2020) Atmospheric microplastics: 10230–10240
a review on current status and perspectives. 30. Silva AB, Bastos AS, Justino CIL, da Costa JP,
Earth-Sci Rev 203:103118 Duarte AC, Rocha-Santos TAP (2018) Micro-
18. Chen G, Fu Z, Yang H, Wang J (2020) An plastics in the environment: challenges in ana-
overview of analytical methods for detecting lytical chemistry - a review. Anal Chim Acta
microplastics in the atmosphere. TrAC Trends 1017:1–19
Anal Chem 130:115981 31. Möller JN, Löder MGJ, Laforsch C (2020)
19. Prata JC, da Costa JP, Duarte AC, Rocha- Finding microplastics in soils: a review of ana-
Santos T (2019 Jan) Methods for sampling lytical methods. Environ Sci Technol 54(4):
and detection of microplastics in water and 2078–2090
sediment: a critical review. TrAC Trends Anal 32. Yang L, Zhang Y, Kang S, Wang Z, Wu C
Chem 110:150–159 (2021) Microplastics in soil: a review on
methods, occurrence, sources, and potential comparison with various physical and chemical
risk. Sci Total Environ 780:146546 approaches. Mar Pollut Bull 148:210–216
33. Fadare OO, Okoffo ED, Olasehinde EF 45. K€appler A, Fischer D, Oberbeckmann S,
(2021) Microparticles and microplastics con- Schernewski G, Labrenz M, Eichhorn K-J
tamination in African table salts. Mar Pollut et al (2016) Analysis of environmental micro-
Bull 164:112006 plastics by vibrational microspectroscopy:
34. Lee H-J, Song N-S, Kim J-S, Kim S-K (2021) FTIR, Raman or both? Anal Bioanal Chem
Variation and uncertainty of microplastics in 408(29):8377–8391
commercial table salts: critical review and vali- 46. Renner G, Schmidt TC, Schram J (2017) Char-
dation. J Hazard Mater 402:123743 acterization and quantification of microplastics
35. Peixoto D, Pinheiro C, Amorim J, Oliva-Teles- by infrared spectroscopy. In: Comprehensive
L, Guilhermino L, Vieira MN (2019) Micro- analytical chemistry [Internet]. Elsevier; [cited
plastic pollution in commercial salt for human 2021 Mar 29], pp 67–118. Available from:
consumption: a review. Estuar Coast Shelf Sci https://linkinghub.elsevier.com/retrieve/pii/
219:161–168 S0166526X16301556
36. Kim J-S, Lee H-J, Kim S-K, Kim H-J (2018) 47. Löder MGJ, Kuczera M, Mintenig S,
Global pattern of microplastics (MPs) in com- Lorenz C, Gerdts G (2015) Focal plane array
mercial food-grade salts: sea salt as an indicator detector-based micro-Fourier-transform infra-
of seawater MP pollution. Environ Sci Technol red imaging for the analysis of microplastics in
52(21):12819–12828 environmental samples. Environ Chem 12(5):
37. Conesa JA, Iñiguez ME (2020) Analysis of 563
microplastics in food samples. In: Rocha- 48. Araujo CF, Nolasco MM, Ribeiro AMP,
Santos T, Costa M, Mouneyrac C (eds) Hand- Ribeiro-Claro PJA (2018) Identification of
book of microplastics in the environment microplastics using Raman spectroscopy: latest
[Internet]. Springer International Publishing developments and future prospects. Water Res
[cited 2021 Mar 29], Cham, p 1–16. Availa- 142:426–440
ble from: http://link.springer.com/10.1007/ 49. von der Esch E, Kohles AJ, Anger PM,
978-3-030-10618-8_5-1 Hoppe R, Niessner R, Elsner M et al (2020)
38. Zarfl C (2019) Promising techniques and open TUM-ParticleTyper: a detection and quantifi-
challenges for microplastic identification and cation tool for automated analysis of (micro-
quantification in environmental matrices. Anal plastic) particles and fibers. George M, editor.
Bioanal Chem 411(17):3743–3756 PLoS One 15(6):e0234766
39. Merck (2021) Sigma-Aldrich [Internet]. 50. Munno K, De Frond H, O’Donnell B, Roch-
Available from: https://www.sigmaaldrich. man CM (2020) Increasing the accessibility for
com/ characterizing microplastics: introducing new
40. Waldman WR, Rillig MC (2020) Microplastic application-based and spectral libraries of plas-
research should embrace the complexity of sec- tic particles (SLoPP and SLoPP-E). Anal Chem
ondary particles. Environ Sci Technol 54(13): 92(3):2443–2451
7751–7753 51. Picó Y, Barceló D (2020) Pyrolysis gas
41. Karlsson TM, K€arrman A, Rotander A, Hassel- chromatography-mass spectrometry in envi-
löv M (2020) Comparison between manta ronmental analysis: focus on organic matter
trawl and in situ pump filtration methods, and and microplastics. TrAC Trends Anal Chem
guidance for visual identification of microplas- 130:115964
tics in surface waters. Environ Sci Pollut Res 52. Peñalver R, Arroyo-Manzanares N, López-
27(5):5559–5571 Garcı́a I, Hernández-Córdoba M (2020) An
42. Lusher AL, Bråte ILN, Munno K, Hurley RR, overview of microplastics characterization by
Welden NA (2020) Is it or isn’t it: the impor- thermal analysis. Chemosphere 242:125170
tance of visual classification in microplastic 53. Yakovenko N, Carvalho A, ter Halle A (2020)
characterization. Appl Spectrosc 74(9): Emerging use thermo-analytical method cou-
1139–1153 pled with mass spectrometry for the quantifica-
43. Renner G, Schmidt TC, Schram J (2018) Ana- tion of micro(nano)plastics in environmental
lytical methodologies for monitoring micro samples. TrAC Trends Anal Chem 131:
(nano)plastics: which are fit for purpose? Curr 115979
Opin Environ Sci Health 1:55–61 54. Suaria G, Achtypi A, Perold V, Lee JR,
44. Gniadek M, Da˛browska A (2019) The marine Pierucci A, Bornman TG et al (2020) Micro-
nano- and microplastics characterisation by fibers in oceanic surface waters: a global char-
SEM-EDX: the potential of the method in acterization. Sci Adv 6(23):eaay8493
55. Olsen LMB, Knutsen H, Mahat S, Wade EJ, environment with iron and zinc ions. Water
Arp HPH (2020) Facilitating microplastic Resour 37(1):65–69
quantification through the introduction of a 59. Güngör EBÖ, Bekbölet M (2010) Zinc release
cellulose dissolution step prior to oxidation: by humic and fulvic acid as influenced by pH,
proof-of-concept and demonstration using complexation and DOC sorption. Geoderma
diverse samples from the Inner Oslofjord, Nor- 159(1–2):131–138
way. Mar Environ Res 161:105080 60. Diaz-Basantes MF, Conesa JA, Fullana A
56. Wu J, Lai M, Zhang Y, Li J, Zhou H, Jiang R (2020) Microplastics in honey, beer, milk and
et al (2020) Microplastics in the digestive tracts refreshments in Ecuador as emerging contami-
of commercial fish from the marine ranching in nants. Sustainability 12(14):5514
East China Sea, China. Case Stud Chem Envi- 61. Erni-Cassola G, Gibson MI, Thompson RC,
ron Eng 2:100066 Christie-Oleza JA (2017) Lost, but found
57. Quinn B, Murphy F, Ewins C (2017) Valida- with Nile Red: a novel method for detecting
tion of density separation for the rapid recovery and quantifying small microplastics (1 mm to
of microplastics from sediment. Anal Methods 20 μm) in environmental samples. Environ Sci
9(9):1491–1498 Technol 51(23):13641–13648
58. Dinu MI (2010) Comparison of complexing
ability of fulvic and humic acids in the aquatic
Chapter 4
Identification of Intentionally and Non-intentionally Added

Substances in Recycled Plastic Packaging Materials
Magdalena Wrona, Davinson Pezo, Robert Paiva, and Sandra A. Cruz
Abstract
Intentionally added substances (IAS) and non-intentionally added substances (NIAS) play an important
role in food contact materials. Even though food packaging is being used to ensure the quality and safety of
food, chemicals may be transferred to food. This aspect is more critical for recycled polymers destined to
contact with food. The presence of contaminants associated with the high shear rates and temperatures
characteristics of the recycling process results in new molecules with low molar mass. These have a greater
migration potential contaminating the food and constitute a risk to consumers. Despite the importance of
NIAS and IAS, there are significant difficulties in their identification and quantification due to the confi-
dential composition of the polymers, the complexity of the chemical structure, and unequivocal confirma-
tion analytes. Therefore, this chapter addresses an overview of the challenge of NIAS and IAS
determination, as well as the most modern analytical methods for determination and quantification in
complex polymeric matrix. Moreover, the usage of analytical techniques has been shown in the context of
direct analysis of recycled polymer surface, the importance of odorous research, and samples from migration
assays (volatile and non-volatile IAS and NIAS). Therefore, techniques such as SERS, ASAP, HS-SPME-
GC-O-MS, DI-GC-MS, SPME-GC-MS, GC-FID, APGC-Q-TOF-MSE, UPLC-Q-TOF-MSE, UPLC-
QqQ-MS, LTQ-Orbitrap, and UPLC-IM-Q-TOF have been discussed, and examples of analysis of real
IAS and NIAS in the complex matrix have been added. Finally, the application of European legislation and
risk assessment have also been discussed.
Key words NIAS, IAS, Recycled packaging, Analytical methods, Migration, Food contamination,
Legislation, Risk assessment, Food safety
1 Introduction
Food contact materials (FCMs) are defined as all materials that

come into contact with food during processing, production, stor-
age, packaging, transportation, preparation, and serving. Different
types of materials, such as polymers, paper, glass, metal, as well as
inks, adhesives, and/or even a combination of them, may be used as
FCMs [1, 2]. In the last few years, studies and regulations on FCMs
have increased significantly due to the presence of diverse chemicals
https://doi.org/10.1007/978-1-0716-3613-8_4,
75
76 Magdalena Wrona et al.
that may present potential adverse health effects to humans or the

environment [1, 3, 4]. This is more critical for food contact articles
based on polymers due to their greater diffusivity, and the potential
migration of contaminants when compared to other materials such
as metal or glass. The migration of a substance from a polymer to
food depends on the mass transfer phenomenon, especially the
diffusion process [5]. Readers can refer to Chap. 6 for further
information on migration of constituents of FCMs. Additionally,
the time, temperature, physicochemical properties of the sub-
stances, and characteristics of the molecular structure of the poly-
mer impact these processes. Consequently, this is a complex system,
and the migration depends on intrinsic and extrinsic factors and
their relationship to the physicochemical properties of each poly-
mer and substance [6, 7].
Although food contact articles, especially food packaging, are
used to ensure the quality and safety of food, chemicals may still
migrate and endanger the health of consumers [8]. The list of
possible chemical migrants is very broad and includes substances
that are named intentionally added substances (IAS). IAS can be
defined as (i) residual compounds from the polymerization process,
such as monomers, catalysts, initiators, and (ii) additives inserted
deliberately to improve the properties or processability of polymers,
for example, plasticizers, antioxidants, flow aids, and others
[4, 9]. Recently, a new class of substances has been highlighted in
the field of food safety: non-intentionally added substances (NIAS).
NIAS may originate from degradation products, additives, and
impurities, as well as the reactions between them. On the other
hand, oligomers have recently been classified as NIAS once they
form due to incomplete reactions during polymerization, according
to the European Union Requirements [10]. IAS and NIAS can also
be classified as volatile, semi-volatile, and non-volatile based on
their molar masses. Therefore, highly sensitive analytical methods
and digital libraries are required for their identification and quanti-
fication [11–13].
Most studies on chemicals present in plastics have focused on
the identification of monomers [14, 15] and additives, especially
antioxidants [16] and those used at higher concentrations as plas-
ticizers [17, 18]. Ibarra and co-workers [17] studied the extraction
of seven compounds, most of them phthalates, from real food
matrices, such as snacks, cookies, and cakes. All compounds were
previously identified in 34 packaged foodstuffs and the quantifica-
tion was performed by gas chromatography coupled with mass
spectrometry (GC-MS). The plasticizer acetyl tributyl citrate was
detected in 94% of the samples, followed by different types of
phthalates, and dibutyl phthalate at levels exceeding the specific
migration limits established by Regulation 10/2011 [10].
Moreover, NIAS formation occurs most frequently due to
degradative processes of polymers, additives, and/or impurities.
IAS and NIAS in Recycled Packaging 77
As a result, new molecules with low molar mass and high diffusion
coefficients are formed, resulting in compounds with greater migra-
tion potential. Several studies have reported the formation
of NIAS, mainly related to poly(ethylene terephthalate) (PET)
[19–21], where the most relevant and well-known compounds
are formaldehyde and acetaldehyde.
Subsequently, an additional aspect when using recycled materi-
als in direct contact with food is the presence of contaminants and
degradation products that can migrate into food. Despite its rele-
vance, this is an issue that is rarely addressed in the literature
[22, 23]. In this industry, the most used recycling process is
mechanical due to its relative low-cost, large-scale, and solvent-
free features, besides being applicable to most thermoplastics.
However, the high shear rates and temperatures employed in the
recycling result in an increase in the degradation process, which
leads to chain scission, chain crosslinking, and branching forma-
tion, as well as the oxidation of polymer molecules. Additionally,
the presence of contaminants, such as lids and labels, printing inks,
pigments, foreign compounds due to misuse by consumers, and
polymer cross-contamination may accelerate the degradation reac-
tions during the recycling process. Moreover, the presence of con-
taminants from previously packaged foods associated with incorrect
disposal has caused a more critical degradation [24, 25].
The packaging sector has always had the highest consumption
in the global plastic market, with the largest amount destined for
single-use applications. Around 36.5% of the world’s plastic pro-
duction in 2019 was attributed to this sector [26]. As a conse-
quence, these products contribute significantly to environmental
pollution and waste generation. The use of recycled plastic for the
same application is one of the strategies to valorize it and close the
loop for these materials. Notwithstanding this fact, the potentially
hazardous compounds (e.g., NIAS, present in recycled packaging)
may contaminate food and pose a risk to the consumers. Therefore,
several works are being carried out to create a database, identify and
develop methodologies for precise identification of NIAS using
analytical techniques. Most studies on recycled polymers and
NIAS are focused on PET [27], low- (LDPE) and high-density
polyethylene (HDPE) [28, 29], expanded polystyrene (EPS) [30],
and polypropylene [31].
As previously described, the greatest difficulty for the imple-
mentation of the use of recycled packaging for food contact is
related to the migration of NIAS and IAS. Health authorities
have clearly defined the criteria for the use of recycled FCM [32–
35]. These recycled plastics must be submitted to a decontamina-
tion technology that is effective in reducing the level of contami-
nants to those allowed by regulatory agencies [36]. The regulations
and directives follow the recommendations of the Codex Alimen-
tarius Commission (CAC) and the Joint FAO/WHO Expert
Committee on Food Additives (JECFA) concerning the safety

assessment of these substances [37, 38]. Although the use of IAS
for the production of FCMs is allowed, these substances must meet
the food safety profile established in the standards and directives of
each country. In general, IAS must be registered and cataloged in a
database, and they must not represent a health risk to the popula-
tion. Another requirement is that the concentration of these sub-
stances in food must not exceed the specific migration limit
established by the legislation [32, 39–42]. Regulatory agencies,
however, encounter difficulties regarding registering NIAS. The
European Commission and the US Food and Drug Administration
recommend that these NIAS should not exceed the analytical
migration tolerance of a maximum of 0.01 mg kg-1 in food
[10, 43, 44]. In addition, if the substance has a potential mutage-
nicity (see Chap. 8 for the determination of genotoxicity and muta-
genicity of food packaging materials), carcinogenicity, or toxicity
(see Chap. 7 for the assessment of the cytotoxicity of food packag-
ing) risk, it should not be used in FCMs.
1.1 Difficulties of Recycled materials have a complex chemical composition, especially

NIAS and IAS when compared to their pristine counterparts. The degradative
Determination processes of a polymer its additives, and impurities results in the
formation of molecules denoted as NIAS. Due to the lack of
information regarding the composition of plastic products by the
manufacture associated with new molecules formed due to the
degradative process, the identification of NIAS via qualitative anal-
ysis is a very challenging task [12]. Establishing the structure of
detected NIAS is also not easy to work on due to the lack of suitable
references and low concentrations of those substances.
Figure 1 shows an example of isomers of nonylphenols (NPs)
and nonylphenol ethoxylated (NPEO). NPs are formed as a result
of hydrolysis of the antioxidant tris(nonylphenyl) phosphite
(TNPP), which is used as a stabilizer for polymeric FCMs. NPs
are endocrine disruptors and xenoestrogens, while NPEOs are
HO OH
2-nonylphenol 4-nonylphenol
2-(4-nonylphenoxy) ethan-1-ol
Fig. 1 The structural formula of the example of nonylphenols and nonylphenol

ethoxylated isomers considered as NIAS and are registered with different CAS
numbers
nonionic surfactants that can be found as potential migrants from

food packaging adhesives. NPEOs degrade into NPs in the envi-
ronment and both families of compounds are considered NIAS
[45–47]. NPs can be analyzed by direct injection GC-MS
(DI-GC-MS), and NPEOs can be analyzed by ultra-high-perfor-
mance liquid chromatography coupled to quadrupole time-of-
flight with MSE technology (UPLC-Q-TOF-MSE) [48].
The non-target analysis is the most suitable analytical approach
for the determination of NIAS. It is based on the screening of all
chemical compounds detectable by the specific analytical method in
one sample injection. By using non-target analysis, it is possible to
determine new compounds that have not been yet studied, which is
highly desirable in NIAS analysis [51, 52]. The reported findings of
new NIAS [30, 51–53] prove that non-target analysis is an indis-
pensable tool for the qualitative analysis of recycled polymers for
food packaging applications.
On the other hand, the analysis of IAS is easier when compared
to NIAS. For this reason, the target analysis can be used success-
fully. In this case, the target analysis is based on the detection and
quantification of the list of known compounds. It would be conve-
nient to use an internal standard containing a similar compound to
IAS that is currently investigated. Furthermore, it would be perfect
if the quantitative analysis of the detected IAS in recycled polymers
for food packaging applications contained internal standards at
similar concentrations to the analyte [33, 54, 55] (Fig. 2).
When analytical standards of chemical compounds are not
available in the market, the quantitative analysis of IAS and NIAS
is more complex [41, 56]. Another case is when the analyte is a
complex mixture of different compounds giving a single signal, and
its origin cannot be determined to match the best standards (e.g.,
mineral oils, as shown in Fig. 3) [43, 57]. Therefore, the concen-
trations of analytes are estimated by semi-quantification. Calibra-
tion is based on a standard that has a similar chemical structure to
the analyzed NIAS, or on a standard of a known compound that is
eluted in the center of the chromatogram [58]. In exceptional
cases, small-scale synthesis of missing standards can be per-
formed on a laboratory scale.
Finally, highly sensitive analytical techniques capable of separ-
ating and detecting IAS and NIAS need to be used. A wide range
of analytical methods with well-establish protocols of identification
and quantification are required. Successful analysis of IAS and
NIAS demonstrates the effectiveness of applying practical guidance
on performance criteria and validation procedures for analytical
methods used in the control of FCMs [59], the performance of
migration assays [29], and the risk assessment [60].
Nowadays, new analytical methods are developed with lower
limits of detection and quantification. Also, the future develop-
ments in the field of highly sensitive analytical methods will lead
Fig. 2 Dependency diagram of the safety and quality of recycled polymers in terms of analysis of IAS and NIAS
and analytical methods
Fig. 3 Chromatograms of (a) olive oil sample with a concentration of 157.42 μg g-1 of mineral oil; (b) pure
olive oil sample with a concentration of 57.42 μg g-1 compared with the standard of alkanes C7–C40 with
a concentration at 10 μg g-1. (Reproduced from Ref. [18] with permission from Elsevier, Copyright 2013)
to determining new IAS and NIAS that are present at the ultra-
traces (ppb and sub-ppb) level. Those compounds will probably not
cause health hazards due to a very low concentration. Although it
should be highlighted that the toxicity of migrants depends not
only on the dose but also on the exposure time.
Nevertheless, qualitative and quantitative analyses of migrating
compounds from FCMs must comply with European legislation
and all its amendments [61] in order to be available on the market.
Therefore, proper characterization of recycled food packaging is
critical.
Moreover, there is a strong tendency toward applying compu-
tation in the building of chemical libraries [62] that will allow the
identification of chemical structures and named NIAS that are
currently referred to as “unknown.” Modern libraries of screening
compounds are already at the forefront of innovative chemical
fingerprint matching design supporting the needs of scientists in
their pursuit of novel molecules and NIAS identification.
2 Methods
2.1 Determination of The assurance of quality and safety in the application of recycled
IAS and NIAS polymers for food contact involves performing complex analyses
utilizing a wide range of analytical methods. Only a broad spectrum
of studies will give enough information on the plastic risk of con-
taining toxic chemicals dangerous to human health. Both qualita-
tive and quantitative research is crucial in the study of IAS and
NIAS. The selection of the appropriate techniques depends on
the type of sample being analyzed, whether it is a direct analysis of
the polymer or a migration study to ensure polymer safety.
Moreover, target or non-target analysis can be performed depend-

ing on the information about the recycled polymer
composition [33].
During the investigation of IAS and NIAS from recycled poly-
mers, the application of modern analytical methods for determining
and quantifying different analytes in the complex matrix is required.
A dependency diagram describing the relationships between NIAS,
IAS, risk assessment, recycled polymer development and safety, and
analytical methods is shown in Fig. 2. The following analytical
techniques are abbreviated: surface-enhanced Raman spectroscopy
(SERS), atmospheric solids analysis probe (ASAP), headspace-solid
phase microextraction gas chromatography-olfactometry-mass
spectrometry (HS-SPME-GC-O-MS), direct injection gas
chromatography-mass spectrometry (DI-GC-MS), solid-phase
microextraction gas chromatography-mass spectrometry (SPME-
GC-MS), atmospheric pressure gas chromatography coupled to
quadrupole time-of-flight with high energy mass spectrometry
(APGC-Q-TOF-MSE), ultra-high-performance liquid chromatog-
raphy coupled to quadrupole time-of-flight with high energy mass
spectrometry (UPLC-Q-TOF-MSE), ultra-high-performance liq-
uid chromatography coupled to triple quadrupole with mass spec-
trometry (UPLC-QqQ-MS), hybrid linear ion trap–high-
resolution mass spectrometry (LTQ-orbitrap), and ultra-high-per-
formance liquid chromatography coupled to ion-mobility quadru-
pole time-of-flight mass spectrometry (UPLC-IM-Q-TOF-MS).
Analysis of IAS and NIAS can be performed by (i) direct analy-
sis of the polymer surface, (2) odors analysis, and (3) migration
study. As a result, volatile, semi-volatile, and/or non-volatile com-
pounds can be detected depending on the analytical technique
applied. In the sequence, the possibility of identifying NIAS and
IAS will be discussed by (i) direct analysis of polymer surface,
especially using ASAP and SERS, (ii) migration assays, and the
analyses of (iii) volatile and (iv) non-volatile compounds.
2.2 Direct Analysis of Regarding the ASAP, the sample is taken with a glass rod
Polymer Surface (immersed, rubbed) and then is entered into the ionization cham-
ber, where it is evaporated and ionized under atmospheric pressure.
The produced ions are analyzed by the MS detector. The advantage
of ASAP analysis is that the analyte concentration in the sample
analyzed directly is much higher than, for example, in samples after
migration assays. This allows better identification of IAS and NIAS.
It is a screening technique that directs the researcher to the appro-
priate choice for further analysis [63]. Still, ASAP has no necessity
for treatment and manipulation of the sample and there is a lack of a
precleaning step. Those additional tasks are very often time-
consuming, expensive, and environmentally unfriendly due to the
usage of a high number of solvents. It should be highlighted that
there is no separation of the analytes on the chromatographic
column. This results in the simultaneous determination of all ana-

lyses from the recycled polymer matrix. Therefore, ASAP is com-
monly used for target analysis, because it is a rapid tool for
screening purposes [12, 64].
A novel application of ASAP for the analysis of NIAS and IAS is
the determination of characteristic markers of mineral oils hydro-
carbons (MOH) [65]. MOH is a mixture of hydrocarbons contain-
ing thousands of chemical compounds of different structures and
sizes derived from petroleum. Therefore, their presence in recycled
packaging can come from polyolefins, paper, board, traces of adhe-
sives, and industrial contamination [61, 66].
It was confirmed that the influence of mineral oils on human
health is negative, and their migration to food should be avoided
[67–69]. As qualitative and quantitative analysis is complicated,
MOHs have been divided into two groups to facilitate the analysis,
namely: mineral oil saturated hydrocarbons (MOSH) and mineral
oil aromatic hydrocarbons (MOAH). The signal of both, com-
monly reported by gas chromatography with flame ionization
detector (GC-FID), has the shape of a characteristic hump
[18, 30], as can be seen in Fig. 3a. Injection of n-alkanes [70]
under the same chromatographic conditions as samples of MOH
can help the identification of individual fractions of mineral oils, as
can be seen in Fig. 3b. Guidance on sampling, analysis, and data
reporting for the monitoring of MOHs in food and FCMs can be
helpful in MOH analysis [54]. Chemical markers are very useful
tools to verify the origin of MOAH. They also avoid misinterpreta-
tion of the analysis by providing detailed and reliable chemical
evidence of MOAH contamination. Samples of recycled PET,
recycled paperboard, and packaging of couscous and semolina
were analyzed in search for MOAH markers by ASAP coupled to
Q-TOF-MSE, and APGC-Q-TOF-MSE [35].
Another technique that allows the direct analysis is surface-
enhanced Raman spectroscopy or surface-enhanced Raman scatter-
ing (SERS). It is a surface-sensitive technique that results in the
enhancement of Raman scattering by nanoparticles such as silver,
gold, and copper or a mixture of them adsorbed on the rough
support surface. Near a rough metal surface, the Raman cross-
section can be dramatically enhanced by a factor of up to 1014.
This allows very sensitive measurements of the analyte adsorbed on
the surface [45, 71]. The enhancement mechanisms are brooadly
divided into chemical and electrochemical enhancement. The
chemical theory claims that, when molecules are adsorbed on the
surface, their electronic states can interact with the states in the
metal and produce new transitions. The true nature of this theory is
still not fully understood. The electrochemical theory is based on
the enhancement of the local electromagnetic field on the surface of
a metal. If the wavelength of the incident electromagnetic field is
close to the plasma wavelength of the metal, electrons can be
excited into an extended surface electronic state (surface plasmon

resonance). This leads to extensive local fields. On the other hand,
there is the formation of charge-transfer complexes between the
analyte and the surface (resonance enhancement) [58]. SERS mea-
surement is carried out using a confocal microscope. The Raman
microscope is by far one of the best instrumentation enhancements
one can make. The new generation of Raman microscopes can offer
a powerful nondestructive and noncontact method of sample anal-
ysis. One of the most incredible benefits is the use of the confocal
Raman microscope design. This enables a very small sample area or
volume to be analyzed down to the micron scale. Combine this
micro Raman analysis with automatic focusing XYZ motion, and it
becomes possible to produce “chemical” images of a sample
[59]. It is worth mentioning that mapping of the sample can also
be performed by applying NIR hyperspectral imaging (see
Chap. 10).
An example of a rapid Raman approach for the detection of
NIAS can be the determination of titanium dioxide, calcium car-
bonate, and calcium sulfate as contaminants in polymer pellets and
food packaging [29]. This technique allows the verification of the
distribution of analytes in the analyzed sample. In this case, it was
found that calcium carbonate and calcium sulfate were environ-
mental contaminants. At the same time, titanium dioxide was a
contaminant from the origin of the production process since it
was also found in the bulk material. Nevertheless, some of those
additives (e.g., titanium dioxide) can also be IAS as they are com-
monly used as polymer additives (fillers and white pigments). Fig-
ure 4 shows the Raman spectrum of oxo-biodegradable
polyethylene for food applications with marked shifts at 609 and
449 cm-1 that are characteristic of titanium dioxide [72].
Additionally, the distribution of NIAS and its clustering can be
better understood by Raman imaging. Moreover, image analysis
can be used as a semi-quantitative analysis of NIAS on the sample
surface [29]. Figure 5 shows a 2-D map and a 3-D representation of
an oxo-biodegradable polyethylene sample for food packaging
applications imaged with a shift of titanium dioxide (609 cm-1).
Different colors indicate different intensities of Raman shifts.
2.3 Migration Assays Although food packaging is designed to contain food products and
protect them from the environment during transport and storage,
it can also be a very significant chemical contamination source
[73]. Therefore, mass transfer between the packaging material
and packaged food under certain conditions is called migration.
This process is essential from an analytical point of view because
migration assays based on different types of simulants determine
the analytical methods applied in the analysis of migrants [15, 12,
46, 74].
1100
1050
1000
950
900
850
800
750
700
Raman Intensity (cps)
650
600
550
500
450
400
350
300
250
200
150
609 449
100
50
0
3500 3000 2500 2000 1500 1000 500
Raman Shift (cm-1)
Fig. 4 Raman spectrum of oxo-biodegradable polyethylene for food packaging applications with marked shifts
corresponding to titanium dioxide
Fig. 5 Raman (a) 2-D map and (b) 3-D representation (collected at 609 cm-1) of an oxo-biodegradable
polyethylene sample for food packaging applications
NIAS as migrants are compounds of low molar mass (consid-

ered to be less than 1000 Da) that are present in the recycled
polymer. Food quality and safety can be compromised, and there-
fore, the health of consumers when these compounds reach a
specific limit. Consequently, research on the migration of chemical

compounds from packaging to food is fundamental [47–50, 53,
75].
The European Union (EU) controls food-grade materials and
articles distributed in the EU market. It is unacceptable that food
packaging materials release substances in amounts hazardous to the
consumers’ health and change the composition of the food prod-
uct. This is why strict formal requirements are imposed on plastics.
The European Commission Regulation EU 10/2011 [10] on
plastic materials and articles intended to come into contact with
food indicates in detail conditions for migration testing. There is no
specific legislation for recycled polymers for food packaging appli-
cations. Standardized time-temperature conditions representing
the particular food application and covering the maximum shelf-
life of the packaged food are applied to the migration study. While it
would be best to perform migration tests using real food, migration
is usually tested using simulants. This is because food is a very
complex matrix that is challenging to analyze. Simultaneously,
food simulants imitate the behavior of food in contact with a
sample of recycled plastic and are easier to analyze [12, 51,
66]. Table 1 shows types of food simulants, the proposition of
analytical technique for analysis of migrants in each matrix, indica-
tion if its pretreatment is necessary, and what type of pretreatment
may be successfully used.
The migration tests on NIAS delectation depend on whether
the measurable material is mono or multilayer. Total immersion of
the sample in food simulants is used for monolayer materials.
Simultaneously, a special migration cell is needed for multilayer
materials where only one side of the material is analyzed. Another
option for multilayer packaging is migration assay performed by
filling. In this case, a bag is made from the analyzed polymer and
filled with food simulant. Only polymers with the property of
thermosealing can be analyzed in this way.
2.4 Analysis of When migrants are determined in food simulants, both volatile and
Volatile Compounds non-volatile compounds need to be analyzed. The group of volatile
organic compounds (VOCs) includes organic compounds with
boiling points less than or equal to 250 °C at a pressure of
101.3 kPa [86].
The method most frequently used to determine NIAS as VOCs
in recycled polymer migration research is gas chromatography. It
enables the separation of a mixture of compounds and, in combi-
nation with an appropriate detection system, gives information
about the type and concentration of determined compounds after
the calibration that precedes it. The essence of the chromatographic
separation is the multiple division of the mixture components
between two immiscible phases: the stationary phase and the
mobile phase, the latter being a gas called a carrier gas. Together
Table 1
Summary of analytical techniques and sample pretreatment procedures for food simulants that are in EU 10/2011
Simulant Food simulant Analytical technique Sample pretreatment Reference

A Ethanol 10 vol% SPME-GC-MS, AQ-TOF-TOF-MSE, UPLC-Q- Not necessary; possible reconcentration of [15, 30, 34, 50–
TOF-MSE, UPLC-QqQ-MS, LTQ-orbitrap, analytes may be done by HFLPMEa or FPSEb 52, 55, 60, 66,
UPLC-IM-Q-TOF 76]
B Acetic acid 3% (w/v) SPME-GC-MS, APGC-Q-TO-MSE, UPLC- Not necessary; possible reconcentration of [15, 30, 34, 50–
QQ-TOFMSE, UPLC-QqQ-MS, analytes may be done by HFLPMEa or FPSEb 52, 55, 60, 66,
LTQ-orbitrap, UPLC-IM-QQ-TOF 76]
C Ethanol 20 vol% SPME-GC-MS, APGC-QQ-TOFMSE, UPLC- Not necessary; possible reconcentration of [15, 50–52, 55,
Q-Q-TOFSE, UPLC-QqQ-MS, analytes may be done by HFLPMEa or FPSEb 62, 66, 77]
LTQ-orbitrap, UPLC-IM-Q-TOF
D1 Ethanol 50 vol% SPME-GC-MS, APGC-Q-TOF-MSE, UPLC-Q- Dilution in case of SPME-GC-MS analysis; [15, 50–52, 55,
TOF-MSE, UPLC-QqQ-MS, LTQ-orbitrap, possible reconcentration of analytes may be 64–66, 77, 78]
UPLC-IM-Q-TOF done by HFLPMEa or FPSEb
D2 Vegetable oil or Vegetable oil: Direct analysis with SERS and after Extraction in case of vegetable oil [15, 34, 50–52,
ethanol 95 vol% extraction with methanol or ethanol the same 55, 60, 66, 75,
techniques as for ethanol 95% 79, 80]
Ethanol 95%: DI-GC-MS, APGC-Q-TOF-MSE, Ethanol 95%: Possible reconcentration of
UPLC-Q-TOF-MSE, UPLC-QqQ-MS, analytes may be done by HFLPMEa or FPSEb
LTQ-orbitrap, UPLC-IM-Q-TOF
E Poly(2,6- Direct analysis with ASAP Extraction with ethanol or methanol; possible [39, 50–53, 55,
diphenyl-p- After extraction with ethanol or methanol: reconcentration of analytes may be done by 61, 75, 81–85]
phenylene oxide) DI-GC-MS, APGC-Q-TOF-MSE, UPLC-Q- HFLPMEa or FPSEb
called Tenax® TOF-MSE, UPLC-QqQ-MS, LTQ-orbitrap,
UPLC-IM-Q-TOF
a
IAS and NIAS in Recycled Packaging
HFLPME automatic multiple dynamic hollow fiber liquid-phase microextraction system

b
FPSE fabric phase sorptive extraction
87
with the carrier gas, the analyzed substances move in the chro-
matographic column; those with higher affinity for the stationary
phase move more slowly along the column and later reach the
detector [87, 88].
Liquid simulants such as 95% ethanol is commonly analyzed by
GC-MS. This method analyzes small molecules after electron-
impact ionization (EI) to obtain mass spectra. Subsequent compo-
nent identification is made by comparing the obtained spectrum,
characteristic for each NIAS, with spectra in volatile compound
libraries (e.g., NIST). The coefficient determining the agreement
between the analyzed NIAS with recycled polymers for food appli-
cations and the standard is the match percentage. The greater the
match percentage, the greater the agreement between the spectra.
Confirmation (100%) of the qualitative analysis can be done by
injecting pure standards of NIAS [66].
EI is a type of hard ionization using high energy and completely
breaks down the molecules. Therefore, if the molecular ion charac-
teristic of NIAS cannot be identified, it is recommended to try to
analyze it with a complementary analytical method based on chem-
ical ionization (CI) or atmospheric pressure ionization (APGC)
using quadrupole and time-of-flight coupled to high-resolution
MS. This technique allows direct analysis of all liquid simulants
(Table 1). An example of such research can determine volatile
NIAS from a starch-based polymer with a new formula for food
packaging applications [89].
For the analysis of food simulants with high water content,
SPME-GC-MS can be used. It is a high-speed technique that does
not require the use of a solvent and therefore allows the analysis of
volatile NIAS in aqueous solutions. Solid-phase microextraction
(SPME) is based on adsorption of analytes in a hot fiber coated
with different polymers or sorbents and their thermal desorption in
Fig. 6 Representative total ion chromatograms (TIC) of (a) virgin and (b) recycled EPS containers with
determined NIAS compounds. (Reproduced from Ref. [30] with permission from Elsevier, Copyright 2019)
the injection port of the chromatograph, where they are detected

by MS. Figure 6 shows an example of the SPME-GC-MS spectra of
volatile NIAS identified in recycled expanded polystyrene contain-
ers and their migration into food simulants such as 3% acetic acid
and 10% ethanol. Chromatograms have market-detected NIAS
with numbers with lists that can be seen in the literature [30].
Furthermore, a technique used to identify odor-active com-
pounds that can result in NIAS [32] in a complex matrix such as
recycled plastics is the headspace-solid phase microextraction-gas
chromatography-olfactometry-mass spectrometry (HS-SPME-
GC-O-MS). This technique contributed to the revolution in the
analysis of the odors by the specific correlation with the chro-
matographic peaks of analytes and their aroma [40, 90]. Paiva
et al. [31] analyzed the presence of volatiles and odoriferous com-
pounds in samples of recycled polypropylene. Forty-five com-
pounds were extracted by headspace solid-phase microextraction
(HS-SPME) and detected by GC-MS and a sniffing port (GC-MS-
O). Nine of these compounds interfere with the quality of food and
had odoriferous characteristics, such as apple, vinegar, heavy-
scented smell, hot oil, vinegar vapor, and burnt synthetics. There-
fore, this work amplifies the importance of the detected odorous
compounds for food applications and, concurrent with other liter-
ature [40, 78, 85], highlights the development of processes ade-
quate to separate and decontaminate recycled polymers for food
contact.
2.5 Analysis of Non- Hyphenated techniques use the resolution capability of the analyti-
volatile Compounds cal method and the capability of MS to identify the separated
components. Liquid chromatography coupled to mass spectrome-
try (LC-MS) is such an analytical technique. The most important
advantage of this technique is the ability to determine polar and
macromolecular compounds; therefore, it is widely used in NIAS
analysis. Unlike gas chromatography, liquid chromatography allows
the analysis of non-volatile compounds.
As a result of the soft ionization of the analytes, a molecular ion
is obtained. For liquid chromatography, there is no standard spec-
tral library as the acquisition of samples can be performed using
different experimental conditions (different energies of ionization).
UPLC-Q-TOF-MSE is the sensitive, fast, and effective tech-
nique used successfully for IAS and NIAS identification. Qualitative
analysis is based on the simultaneous application of low and high
collision energy for spectral acquisition. This mode provides accu-
rate precursor and fragment ion mass information simultaneously.
Therefore, this is a possible identification and pattern recognition
of compounds such as aromatic amines [40], oligomers from
starch-based polymer [89], and migrants from adhesives for food
packaging applications [91]. Additionally, target analysis in single
ion recording (SIR) mode, using even more sensitive equipment
UPLC-QqQ-MS, can be applied to make a quantitative analysis

of the determined IAS and NIAS [15].
High-resolution ion trap hybrid linear mass spectrometry
(LTQ-Orbitrap) compared to UPLC-Q-TOF-MSE can perform a
multi-level NIAS fragmentation. This device is comprised of an MS
linear ion trap and an Orbitrap mass analyzer. It has already been
applied for the qualitative analysis of FCM non-volatile
migrants [15].
Ion-mobility quadrupole time-of-flight mass spectrometry,
coupled to the ultra-high-performance liquid chromatography
(UPLC-IM-Q-TOF-MS), allows obtaining a very clean spectrum
of the analyzed NIAS, because retention time together with
the drift time is used to determine ions. Therefore, a novel collision
cross-section (CCS) value connected directly with the shape and
size of the analyzed NIAS is determined. Thus, the application of
this analytical technique for analysis of migrants from recycled food
packaging confirms much better-determined compounds. An
example of its application can be the determination of polyamide
6 and polyamide 66 oligomers from kitchenware utensils to food.
Oligomers are part of the polymer structure and are currently not
legislated by the EU. They became NIAS in case of migration into
food products from packaging [47].
3 Risk Assessment
Risk assessment of the negative impact of chemicals from recycled

polymers on human health is based on the threshold of toxicologi-
cal effects obtained from tests on animals. The threshold of toxico-
logical concern (TTC) is defined as the level of the chemicals
analyzed below which there would be no significant hazard to
human health [92].
A positive list of migrants that can be detected in samples of
migration assays has been presented in the European Regulation
No 10/2011 on plastic materials and articles intended to contact
foodstuffs. Therefore, these chemical substances can be used in
recycled polymers for food packaging applications. Migration limits
are a theoretical mathematical value that must not exceed the
concentration of a given compound. Migration limits keep recycled
plastics safe. Substances on the positive list of EU 10/2011 have
assigned a specific migration limit (SML). These values were estab-
lished based on the toxicity data of each chemical studied by the
European Food and Safety Authority (EFSA). It should be empha-
sized that the term of global migration with the migration process
is also connected. In this case, the total concentration of all sub-
stances migrating to the food (detected in the food simulant) must
not exceed the overall migration limit (OML) of 60 mg kg-1 of
food or 10 mg dm-2 of recycled material [66].
Fig. 7 Decision tree from Toxtree software for NIAS bisphenol A diglycidyl ether
Nevertheless, NIAS are very often new substances boasting

toxicities with no prior study. The toxicity of chemicals not
included in the positive list is estimated using Cramer’s rules and
the open-source application Toxtree in the TTC approach. Toxtree
can assess toxic hazards by applying a decision tree approach.
Cramer’s classification assigns chemicals to one of three toxicity
classes and proposes a maximum daily intake. Theoretical maxi-
mum migration amounts of 1.80, 0.54, and 0.09 mg kg-1 are
applied to classes I, II, and III, respectively [89]. Figure 7 presents
a decision tree from Toxtree software for bisphenol A diglycidyl
ether (BADGE). This toxic NIAS is an endocrine-disrupting sub-
stance and has been classified as class III [78, 89, 93].
Equation 1 is used to calculate the theoretical maximum migra-
tion amounts (mg kg-1) of NIAS coming from recycled polymer
and not present in the positive list of EU 10/2011:
EDI = migration ‧ food intake ‧ CF (1)
where EDI is the estimated daily intake (maximum daily intake for
each substance per person, considered 1 kg person-1 d-1 in Eur-
ope); CF = consumption factor [94].
In the case of BADGE, the endocrine-disrupting NIAS that has
been assigned by Toxtree to Cramer class III EDI would be
0.09 mg kg-1. This is because the maximum daily intake for each
substance per person in Europe is considered 1 kg person-1 d-1,
and the consumption factor is not applied in Europe. CF is fraction
of the daily diet for a specific material that is part of the food contact
materials. It is used for the analysis of recycled polymers in the USA.
Therefore, obtained values of migrating BADGE concentration
from recycled food packaging cannot exceed the calculated EDI
value.
Difficulties and challenges found in the analysis of IAS and NIAS in

recycled food packaging have been present. The main challenges
are summarized into the following: (i) the lack of information on
the actual composition of the different ingredients and materials
used for packaging production; (ii) extensive use of additives as
stabilizers, antioxidants, plasticizers, among others, depending on
the application; (iii) the need for high-sensitivity and precision
analytical techniques; and (iv) the high shear rates and temperatures
employed in recycling process that results in the formation of new
molecules. Therefore, the identification and unequivocal confirma-
tion of IAS and NIAS can be puzzling. In most cases, the
non-target analysis is the most suitable analytical approach for the
determination of NIAS, while the analysis of IAS is much easier,
and the target analysis can be used favorably.
Successful analysis of IAS and NIAS is based on applying highly
sensitive analytical techniques capable of their separation and detec-
tion. Migration assays, international legislation, and appropriate
application of risk assessment are crucial for ensuring the quality
and safety of recycled polymers to food contact. The use of analyti-
cal techniques has been shown for the direct analysis of surface.
They are commonly used for target analysis and being a rapid tool
for screening purpose. However, they present a lower limit of
detection. Additionally, the importance of odorous research and
samples from migration assays (volatile and non-volatile IAS and
NIAS) are addressed in this chapter. Therefore, techniques such as
SERS, ASAP, HS-SPME-GC-O-MS, DI-GC-MS, SPME-GC-MS,
GC-FID, APGC-Q-TOF-MSE, UPLC-Q-TOF-MSE, UPLC-
QqQ-MS, LTQ-orbitrap, and UPLC-IM-Q-TOF have been dis-
cussed, and examples of the analysis of real IAS and NIAS in the
complex matrix have been added. Future developments in the field
of super sensitive analytical techniques will probably lead to new
IAS and NIAS. Simultaneously, quick identification of those com-

pounds will be possible due to the development of modern chemi-
cal libraries.
Acknowledgments
The authors want to thank the PACMI group (Program for the
Quality Assurance of Individualized Medications) of the Pharmacy
area of the San Jorge University. Also, the authors wish to thank the
Government of Aragon and the European Social Fund for financial
support (T53-20R) to the GUIA group. Also, the authors would
like to acknowledge FAPESP 2016/25703-2.
References
1. Groh KJ, Geueke B, Martin O, Maffini M, barriers. Food Addit Contam 23:202–211.
Muncke J (2020) Overview of intentionally h t t p s : // d o i . o r g / 1 0 . 1 0 8 0 /
used food contact chemicals and their hazards. 02652030500373661
Environ Int 106225:106225. https://doi. 8. Vitrac O, Fractionnement UMR, Cedex R,
org/10.1016/j.envint.2020.106225 Hayert M (2005) Risk assessment of migration
2. Beneventi E, Tietz T, Merkel S (2020) Risk from packaging materials into foodstuffs. 51.
assessment of food contact materials. EFSA J https://doi.org/10.1002/aic.10462
18:1–11. https://doi.org/10.2903/j.efsa. 9. Eaves D (2021) Consultant Regulatory I
2020.e181109 European Food Contact Compliance:
3. Frans FV, Jyrki V (2019) Circular economy: Non-Intentionally Added Substances [NIAS].
Commission welcomes Council final adoption intertek.com/regulatory/food-contact/.
of new rules on single-use plastics to reduce Accessed 22 Mar 2021
marine plastic litter. 30–31 10. Commission Regulation (EU) No 10/2011
4. Groh KJ, Backhaus T, Carney-Almroth B, (2011) Commission Regulation (EU) No
Geueke B, Inostroza PA, Lennquist A, Leslie 10/2011 of 14 January 2011 on plastic mate-
HA, Maffini M, Slunge D, Trasande L, War- rials and articles intended to come into contact
hurst AM, Muncke J (2019) Overview of with food. Off J Eur Union:1–136
known plastic packaging-associated chemicals 11. NIST (National Institue of Standards and
and their hazards. Sci Total Environ 651: Technology) Libro del Web de Quı́mica del
3253–3268. https://doi.org/10.1016/j. NIST, SRD 69. Retrieved March 26, 2020,
scitotenv.2018.10.015 from https://webbook.nist.gov/
5. Arvanitoyannis IS, Bosnea L (2004) Migration 12. Nerin C, Alfaro P, Aznar M, Domeño C (2013)
of substances from food packaging materials to The challenge of identifying non-intentionally
foods. Crit Rev Food Sci Nutr 44:63–76. added substances from food packaging materi-
h t t p s : // d o i . o r g / 1 0 . 1 0 8 0 / als: a review. Anal Chim Acta 775:14–24.
10408690490424621 https://doi.org/10.1016/j.aca.2013.02.028
6. Hahladakis JN, Velis CA, Weber R, 13. Skjevrak I, Brede C, Steffensen IL, Mikalsen A,
Iacovidou E, Purnell P (2018) An overview of Alexander J, Fjeldal P, Herikstad H (2005)
chemical additives present in plastics: migra- Non-targeted multi-component analytical sur-
tion, release, fate and environmental impact veillance of plastic food contact materials: iden-
during their use, disposal and recycling. J Haz- tification of substances not included in EU
ard Mater 344:179–199. https://doi.org/10. positive lists and their risk assessment. Food
1016/j.jhazmat.2017.10.014 Addit Contam 22:1012–1022. https://doi.
7. Dole P, Feigenbaum AE, De La Cruz C, org/10.1080/02652030500090877
Pastorelli S, Paseiro P, Hankemeier T, 14. Tawakkal ISMA, Cran MJ, Miltz J, Bigger SW
Voulzatis Y, Aucejo S, Saillard P, Papaspyrides (2014) A review of poly(lactic acid)-based
C (2006) Typical diffusion behaviour in pack- materials for antimicrobial packaging. J Food
aging polymers – application to functional
Sci 79:R1477. https://doi.org/10.1111/ on contaminated post-consumer PET: thermal

1750-3841.12534 and crystallization behavior. J Polym Environ
15. Wrona M, Nerı́n C (2020) Analytical 28:91–99. https://doi.org/10.1007/s10924-
approaches for analysis of safety of modern 019-01579-9
food packaging: a review. Molecules 25. 25. Cruz SA, Scuracchio CH, Fitaroni LB, Oliveira
h t t p s : // d o i . o r g / 1 0 . 3 3 9 0 / C (2017) The use of melt rheology and solu-
molecules25030752 tion viscometry for degradation study of post-
16. Bach C, Dauchy X, Severin I, Munoz JF, consumer poly(ethylene terephthalate): the
Etienne S, Chagnon MC (2013) Effect of tem- effects of the contaminants, reprocessing and
perature on the release of intentionally and solid state polymerization. Polym Test 60:236–
non-intentionally added substances from poly- 241. https://doi.org/10.1016/j.polymer
ethylene terephthalate (PET) bottles into testing.2017.03.026
water: chemical analysis and potential toxicity. 26. Research GV (2020) Plastic market size, share
Food Chem 139:672–680. https://doi.org/ & trends analysis report by product (PE, PP,
10.1016/j.foodchem.2013.01.046 PU, PVC, PET, polystyrene, ABS, PBT, PPO,
17. Ibarra VG, Rodrı́guez A, Quirós B, De Losada epoxy polymers, LCP, PC, polyamide), by
PP, Sendón R (2018) Identification of inten- application, by region, and segment forecasts,
tionally and non-intentionally added sub- 2020–2027
stances in plastic packaging materials and their 27. Brouwer MT, Alvarado Chacon F, Thoden van
migration into food products. 3789–3803 Velzen EU (2020) Effect of recycled content
18. Wrona M, Pezo D, Nerin C (2013) Rapid ana- and rPET quality on the properties of PET
lytical procedure for determination of mineral bottles. Part III: Modelling of repetitive recy-
oils in edible oil by GC-FID. Food Chem 141: cling. Packag Technol Sci 33:373–383.
3993–3999. https://doi.org/10.1016/j. https://doi.org/10.1002/pts.2489
foodchem.2013.06.091 28. Geueke B (2018) Dossier – non-intentionally
19. Dabrowska A, Borcz A, Nawrocki J (2003) added substances (NIAS). Food Packag Forum
Aldehyde contamination of mineral water 7 . h t t p s : //d o i . o r g / 1 0 . 5 2 8 1 / z e n o d o.
stored in PET bottles. Food Addit Contam 1265331
20:1170–1177. https://doi.org/10.1080/ 29. Portesi C, Visentin D, Durbiano F, Abete MC,
02652030310001620441 Rizzi M, Maurino V, Rossi AM (2019) Devel-
20. Bach C, Dauchy X, Chagnon MC, Etienne S opment of a rapid micro-Raman spectroscopy
(2012) Chemical compounds and toxicological approach for detection of NIAS in LDPE pel-
assessments of drinking water stored in poly- lets and extruded films for food packaging
ethylene terephthalate (PET) bottles: a source applications. Polym Test 80:106098. https://
of controversy reviewed. Water Res 46:571– doi.org/10.1016/j.polymertesting.2019.
583. https://doi.org/10.1016/j.watres.2011. 106098
11.062 30. Song X-C, Wrona M, Nerin C, Lin Q-B,
21. Özlem KE (2008) Acetaldehyde migration Zhong H-N (2019) Volatile non-intentionally
from polyethylene terephthalate bottles into added substances (NIAS) identified in recycled
carbonated beverages in Türkiye. Int J Food expanded polystyrene containers and their
Sci Technol 43:333–338. https://doi.org/10. migration into food simulants. Food Packag
1111/j.1365-2621.2006.01443.x Shelf Life 20:100318. https://doi.org/10.
22. Horodytska O, Cabanes A, Fullana A (2020) 1016/j.fpsl.2019.100318
Non-intentionally added substances (NIAS) in 31. Paiva R, Wrona M, Nerı́n C, Bertochi I,
recycled plastics. Chemosphere 251:126373. Gavril L, Andrea S (2021) Importance of pro-
https://doi.org/10.1016/j.chemosphere. file of volatile and off-odors compounds from
2020.126373 different recycled polypropylene used for food
23. Yusà V, López A, Dualde P, Pardo O, Fochi I, applications. Food Chem 350:129250.
Pineda A, Coscolla C (2020) Analysis of https://doi.org/10.1016/j.foodchem.2021.
unknowns in recycled LDPE plastic by 129250
LC-Orbitrap Tribrid HRMS using MS3 with 32. Bauer A, Jesús F, Gómez Ramos MJ,
an intelligent data acquisition mode. Micro- Lozano A, Fernández-Alba AR (2019) Identi-
chem J 158:105256. https://doi.org/10. fication of unexpected chemical contaminants
1016/j.microc.2020.105256 in baby food coming from plastic packaging
24. Fitaroni LB, de Oliveira ÉC, Marcomini AL, migration by high resolution accurate mass
Paranhos CM, Freitas FL, Cruz SA (2020) spectrometry. Food Chem 295:274–288.
Reprocessing and solid state polymerization https://doi.org/10.1016/j.foodchem.2019.
05.105
33. Borzi F, Torrieri E, Wrona M, Nerı́n C (2019) Vitrac O, Schaefer A, Simoneau C (2015) Prac-
Polyamide modified with green tea extract for tical guidelines on the application of migration
fresh minced meat active packaging applica- modelling for the estimation of specific migra-
tions. Food Chem 300:125–242. https://doi. tion. Publications Office of the European
org/10.1016/j.foodchem.2019.125242 Union
34. Gavril GL, Wrona M, Bertella A, Świeca M, 43. PlasticsEurope (2011) Risk assessment of
Râpă M, Salafranca J, Nerı́n C (2019) Influence non-listed substances (NLS) and
of medicinal and aromatic plants into risk non-intentionally added substances (NIAS)
assessment of a new bioactive packaging based under article 19. 32
on polylactic acid (PLA). Food Chem Toxicol 44. Center for Food Safety and Applied Nutrition
132:1–11. https://doi.org/10.1016/j.fct. (2006) Guidance for industry: use of recycled
2019.110662 plastics in food packaging (chemistry consid-
35. Jaén J, Domeño C, Alfaro P, Nerı́n C (2021) erations). https://www.fda.gov/regulatory-
Atmospheric Solids Analysis Probe (ASAP) and information/search-fda-guidance-
Atmospheric Pressure Gas Chromatography documents/guidance-industry-use-recycled-
(APGC) coupled to Quadrupole Time of plastics-food-packaging-chemistr y-con
Flight Mass Spectrometry (QTOF-MS) as siderations. Accessed 23 Mar 2020
alternative techniques to trace aromatic mar- 45. Santos EDB, Sigoli FA, Mazali IO (2013)
kers of mineral oils in food packaging. Talanta Metallic Cu nanoparticles dispersed into
227:122079. https://doi.org/10.1016/j. porous glass: a simple green chemistry
talanta.2020.122079 approach to prepare SERS substrates. Mater
36. Anvisa RDCN, Nacional A, Sanitária DV Lett 108:172. https://doi.org/10.1016/j.
(2004) RDC N° 216_ANVISA – Agência matlet.2013.06.110
Nacional de Vigilância Sanitária 46. Koster S, Rennen M, Leeman W, Houben G,
37. Food and Agriculture Organization of the Muilwijk B, van Acker F, Krul L (2014) A novel
United Nations (FAO), World Health Organi- safety assessment strategy for non-intentionally
zation (WHO) (2015) CODEX ALIMEN- added substances (NIAS) in carton food con-
TARIUS COMMISSION. 2005. Procedural tact materials. Food Addit Contam Part A
manual, 24th ed. Rome Chem Anal Control Expo Risk Assess 31:
38. Magnuson B, Munro I, Abbot P, Baldwin N, 4 2 2 – 4 4 3 . h t t p s : // d o i . o r g / 1 0 . 1 0 8 0 /
Lopez-Garcia R, Ly K, McGirr L, Roberts A, 19440049.2013.866718
Socolovsky S (2013) Review of the regulation 47. Canellas E, Vera P, Song XC, Nerin C,
and safety assessment of food substances in Goshawk J, Dreolin N (2021) The use of ion
various countries and jurisdictions. Food mobility time-of-flight mass spectrometry to
Addit Contam Part A Chem Anal Control assess the migration of polyamide 6 and poly-
Expo Risk Assess 30:1147–1220. https://doi. amide 66 oligomers from kitchenware utensils
org/10.1080/19440049.2013.795293 to food. Food Chem 350:129260. https://doi.
39. Canellas E, Nerı́n C, Moore R, Silcock P org/10.1016/j.foodchem.2021.129260
(2010) New UPLC coupled to mass spectrom- 48. Asensio E, Peiro T, Nerı́n C (2019) Determi-
etry approaches for screening of non-volatile nation the set-off migration of ink in
compounds as potential migrants from adhe- cardboard-cups used in coffee vending
sives used in food packaging materials. Anal machines. Food Chem Toxicol 130:61–67.
Chim Acta 666:62–69. https://doi.org/10. https://doi.org/10.1016/j.fct.2019.05.022
1016/j.aca.2010.03.032 49. Asensio E, Montañés L, Nerı́n C (2020)
40. Wrona M, Vera P, Pezo D, Nerı́n C (2017) Migration of volatile compounds from natural
Identification and quantification of odours biomaterials and their safety evaluation as food
from oxobiodegradable polyethylene oxidised contact materials. Food Chem Toxicol 142:
under a free radical flow by headspace solid- 111457. https://doi.org/10.1016/j.fct.
phase microextraction followed by gas 2020.111457
chromatography-olfactometry-mass spectrom- 50. Otoukesh M, Nerı́n C, Aznar M, Kabir A, Fur-
etry. Talanta 172:37–44. https://doi.org/10. ton KG, Es’haghi Z (2019) Determination of
1016/j.talanta.2017.05.022 adhesive acrylates in recycled polyethylene tere-
41. Bratinova S, Raffael B, Simoneau C (2009) phthalate by fabric phase sorptive extraction
Guidelines for performance criteria and valida- coupled to ultra performance liquid
tion procedures of analytical methods used in chromatography – mass spectrometry. J Chro-
controls of food contact materials matogr A 1602:56–63. https://doi.org/10.
42. Hoekstra E, Brandsch R, Dequatre C, 1016/j.chroma.2019.05.044
Mercea P, Milana MR, Stoermer A, Trier X,
51. Aznar M, Alfaro P, Nerin C, Kabir A, Furton Anal Bioanal Chem 412:5999. https://doi.
KG (2016) Fabric phase sorptive extraction: an org/10.1007/s00216-020-02779-2
innovative sample preparation approach 60. Su Q-Z, Vera P, Nerı́n C (2020) Direct
applied to the analysis of specific migration immersion–solid-phase microextraction cou-
from food packaging. Anal Chim Acta 936: pled to gas chromatography–mass spectrome-
97–107. https://doi.org/10.1016/j.aca. try and response surface methodology for
2016.06.049 nontarget screening of (semi-) volatile
52. Salafranca J, Pezo D, Nerı́n C (2009) Assess- migrants from food contact materials. Anal
ment of specific migration to aqueous simu- Chem 92:5577–5584. https://doi.org/10.
lants of a new active food packaging 1021/acs.analchem.0c00532
containing essential oils by means of an auto- 61. Vera P, Canellas E, Nerı́n C (2018) Identifica-
matic multiple dynamic hollow fibre liquid tion of non volatile migrant compounds and
phase microextraction system. J Chromatogr NIAS in polypropylene films used as food pack-
A 1216:3731–3739. https://doi.org/10. aging characterized by UPLC-MS/QTOF.
1016/j.chroma.2009.03.001 Talanta 188:750–762. https://doi.org/10.
53. Otoukesh M, Vera P, Wrona M, Nerin C, 1016/j.talanta.2018.06.022
Es’haghi Z (2020) Migration of dihydroxyalk- 62. Hoppe M, de Voogt P, Franz R (2018) Oligo-
ylamines from polypropylene coffee capsules to mers in polyethylene naphthalate and polybu-
Tenax® and coffee by salt-assisted liquid–liquid tylene terephthalate – identification and
extraction and liquid chromatography–mass exploring migration. Food Package Shelf Life
spectrometry. Food Chem 321:126720. 17:171–178. https://doi.org/10.1016/J.
https://doi.org/10.1016/j.foodchem.2020. FPSL.2018.07.001
126720 63. Pietropaolo E, Albenga R, Gosetti F, Toson V,
54. Bratinova S, Hoekstra E (2019) Guidance on Koster S, Marin-Kuan M, Veyrand J, Patin A,
sampling, analysis and data reporting for the Schilter B, Pistone A, Tei L (2018) Synthesis,
monitoring of mineral oil hydrocarbons in food identification and quantification of oligomers
and food contact materials in the frame of from polyester coatings for metal packaging. J
Commission Recommendation (EU) 2017/ Chromatogr A 1578:15–27. https://doi.org/
84 10.1016/J.CHROMA.2018.10.002
55. Pezo D, Salafranca J, Nerı́n C (2007) Develop- 64. Carrizo D, Domeño C, Nerı́n I, Alfaro P, Nerı́n
ment of an automatic multiple dynamic hollow C (2015) Atmospheric pressure solid analysis
fibre liquid-phase microextraction procedure probe coupled to quadrupole-time of flight
for specific migration analysis of new active mass spectrometry as a tool for screening and
food packagings containing essential oils. J semi-quantitative approach of polycyclic aro-
Chromatogr A 1174:85–94. https://doi.org/ matic hydrocarbons, nitro-polycyclic aromatic
10.1016/j.chroma.2007.08.033 hydrocarbons and oxo-polycyclic aromatic.
56. Rycenga M, Camargo PHC, Li W, Moran CH, Talanta 131:175. https://doi.org/10.1016/j.
Xia Y (2010) Understanding the SERS effects talanta.2014.07.034
of single silver nanoparticles and their dimers, 65. Garcı́a Ibarra V, Sendón R, Garcı́a-Fonte XX,
one at a time. J Phys Chem Lett 1:696–703. Paseiro Losada P, Bernaldo R, de Quirós A
https://doi.org/10.1021/jz900286a (2019) Migration studies of butylated hydro-
57. Koster S, Bani-Estivals M-H, Bonuomo M, xytoluene, tributyl acetylcitrate and dibutyl
Bradley E, Chagnon M-C, Garcia ML, phthalate into food simulants. J Sci Food
Godts F, Gude T, Helling R, Paseiro-Losada P, Agric 99:1586. https://doi.org/10.1002/
Pieper G, Rennen M, Simat T, Spack jsfa.9337
L. Guidance on best practices on the risk asses- 66. Nerı́n C, Wrona M (2018) Polymers/food
ment of non-intentionally added substances contact and packaging materials – analytical
(NIAS) in food contact materials and articles. aspects, 3rd edn. Elsevier
Commissioned the ILSI Europe Packaging 67. Morbeck DE, Fredrickson JR, Walker DL
Materials Task Force (2012) Factors that affect mineral oil toxicity:
58. Sharma B, Frontiera RR, Henry A, Ringe E, role of oxygen and protein supplement. Fertil
Van Duyne RP (2012) SERS: materials, appli- Steril 98:S29. https://doi.org/10.1016/j.fer
cations, and the future Surface enhanced tnstert.2012.07.108
Raman spectroscopy (SERS) is a powerful 68. Nygaard UC, Vege Å, Rognum T, Grob K,
vibrational. Mater Today 15:16 Cartier C, Cravedi JP, Alexander J (2019)
59. Jahn IJ, Mühlig A, Cialla-May D (2020) Appli- Toxic effects of mineral oil saturated hydrocar-
cation of molecular SERS nanosensors: where bons (MOSH) and relation to accumulation in
we stand and where we are headed towards?
rat liver. Food Chem Toxicol 123:431. chromatography-olfactometry-mass spectrom-

https://doi.org/10.1016/j.fct.2018.11.022 etry. Talanta 207:120301. https://doi.org/
69. Otunga GN, Maiyoh GK, Macharia BN, Tuei 10.1016/j.talanta.2019.120301
VC (2019) Transformer mineral oil ingestion 79. Gelbke H-P, Banton M, Block C, Dawkins G,
induces systemic sub-acute toxicity in Wistar Eisert R, Leibold E, Pemberton M, Puijk IM,
rats. Heliyon 5:e02998. https://doi.org/10. Sakoda A, Yasukawa A (2019) Risk assessment
1016/j.heliyon.2019.e02998 for migration of styrene oligomers into food
70. Pezo D, Wrona M, Rodriguez-Lafuente A, from polystyrene food containers. Food Chem
Nerin C (2012) A sulphuric acid-impregnated Toxicol 124:151–167. https://doi.org/10.
silica gel clean-up procedure for the determina- 1016/J.FCT.2018.11.017
tion of n-alkanes migration from paraffin based 80. Wrona M, Salafranca J, Rocchia M, Nerı́n C
paper packaging into cheddar cheese. Food (2015) Application of SERS to the determina-
Chem 134:405. https://doi.org/10.1016/j. tion of butylated hydroxyanisole in edible and
foodchem.2012.02.076 essential oils Spectrosc (Santa Monica) 30
71. Santos EDB, Sigoli FA, Mazali IO (2013) 81. Rubio L, Valverde-Som L, Sarabia LA, Ortiz
Surface-enhanced Raman scattering of MC (2019) The behaviour of Tenax as food
4-aminobenzenethiol on silver nanoparticles simulant in the migration of polymer additives
substrate. Vib Spectrosc 68:246. https://doi. from food contact materials by means of gas
org/10.1016/j.vibspec.2013.08.003 chromatography/mass spectrometry and PAR-
72. Wrona M, Salafranca J, Nerı́n C (2017) Fast AFAC. J Chromatogr A 1589:18–29. https://
assessment of oxo-biodegradable polyethylene doi.org/10.1016/j.chroma.2018.12.054
film oxidation by surface-enhanced Raman 82. de Poças MF, Oliveira JC, Pereira JR,
scattering with in situ formation of a silver Brandsch R, Hogg T (2011) Modelling migra-
nanoparticle substrate. J Mater Chem C 5: tion from paper into a food simulant. Food
463. https://doi.org/10.1039/c6tc04401h Control 22:303. https://doi.org/10.1016/j.
73. Trinetta V (2018) Obsolete: definition and foodcont.2010.07.028
function of food packaging. In: Reference 83. Aznar M, Domeño C, Nerı́n C, Bosetti O
Module in Food Science (2015) Set-off of non volatile compounds
74. Wrona M, Nerin C (2019) Chapter 7: Risk from printing inks in food packaging materials
assessment of plastic packaging for food and the role of lacquers to avoid migration.
applications Dyes Pigments 114:85–92. https://doi.org/
75. Liu YQ, Wrona M, Su QZ, Vera P, Nerı́n C, Hu 10.1016/j.dyepig.2014.10.019
CY (2021) Influence of cooking conditions on 84. Garcı́a Ibarra V, Bernaldo R, de Quirós A,
the migration of silicone oligomers from sili- Paseiro Losada P, Sendón R (2019)
cone rubber baking molds to food simulants. Non-target analysis of intentionally and non
Food Chem 347:128964. https://doi.org/10. intentionally added substances from plastic
1016/j.foodchem.2020.128964 packaging materials and their migration into
76. Oliveira ÉC, Echegoyen Y, Cruz SA, Nerin C food simulants. Food Packag Shelf Life 21:
(2014) Comparison between solid phase 100325. https://doi.org/10.1016/j.fpsl.
microextraction (SPME) and hollow fiber liq- 2019.100325
uid phase microextraction (HFLPME) for 85. Vera P, Canellas E, Nerı́n C (2020) Com-
determination of extractables from post- pounds responsible for off-odors in several
consumer recycled PET into food simulants. samples composed by polypropylene, polyeth-
Talanta 127:59–67. https://doi.org/10. ylene, paper and cardboard used as food pack-
1016/j.talanta.2014.03.042 aging materials. Food Chem 309:125792.
77. Kim HS, Lee YJ, Koo YJ, Pack EC, Lim KM, https://doi.org/10.1016/j.foodchem.2019.
Choi DW (2021) Migration of monomers, 125792
plastic additives, and non-intentionally added 86. European Parliament and the Council (2004)
substances from food utensils made of Directive 2004/42/EC on the limitation of
melamine–formaldehyde resin following ultra- emissions of volatile organic compounds due
violet sterilization. Food Control 125:107981. to the use of organic solvents in certain paints
https://doi.org/10.1016/j.foodcont.2021. and varnishes and vehicle refinishing products
107981 87. Kupiec T (2004) Quality-control analytical
78. da Oliveira WS, Monsalve JO, Nerin C, methods: gas chromatography. Int J Pharm
Padula M, Godoy HT (2020) Characterization Compd 8:305
of odorants from baby bottles by headspace
solid phase microextraction coupled to gas
88. Hopfer H (2020) Analytical methods: gas materials. Food Chem Toxicol 75:79–87.
chromatography. Reference Module in Food https://doi.org/10.1016/j.fct.2014.10.029
Science 92. Patlewicz G, Jeliazkova N, Safford RJ, Worth
89. Osorio J, Dreolin N, Aznar M, Nerı́n C, Han- AP, Aleksiev B (2008) An evaluation of the
cock P (2019) Determination of volatile non implementation of the Cramer classification
intentionally added substances coming from a scheme in the Toxtree software. SAR QSAR
starch-based biopolymer intended for food Environ Res 19:495. https://doi.org/10.
contact by different gas chromatography-mass 1080/10629360802083871
spectrometry approaches. J Chromatogr A 93. Dreolin N, Aznar M, Moret S, Nerin C (2019)
1599:215–222. https://doi.org/10.1016/j. Development and validation of a LC–MS/MS
chroma.2019.04.007 method for the analysis of bisphenol a in poly-
90. Ubeda S, Aznar M, Nerı́n C (2019) Determi- ethylene terephthalate. Food Chem 274:246.
nation of volatile compounds and their sensory https://doi.org/10.1016/j.foodchem.2018.
impact in a biopolymer based on polylactic acid 08.109
(PLA) and polyester. Food Chem 294:171– 94. Dewhurst I, Renwick AG (2013) Evaluation of
178. https://doi.org/10.1016/j.foodchem. the Threshold of Toxicological Concern
2019.05.069 (TTC) – challenges and approaches. Regul
91. Canellas E, Vera P, Nerı́n C (2015) Risk assess- Toxicol Pharmacol 65:168. https://doi.org/
ment derived from migrants identified in sev- 10.1016/j.yrtph.2012.03.007
eral adhesives commonly used in food contact
Chapter 5
Poly- and Perfluorinated Alkyl Substances in Food

Packaging Materials
Rachel C. Scholes, William Hart-Cooper, Gregory M. Glenn,
and William J. Orts
Abstract
Poly- and perfluorinated alkyl substances (PFAS) are commonly used additives in food packaging materials
that impart water and grease resistance. However, this class of compounds is coming under increased
scrutiny due to human health and environmental concerns. As a result, regulatory agencies are developing
limits on PFAS in food packaging. The development and enforcement of such limits highlights the need for
robust PFAS detection methods. Unfortunately, targeted methods that detect specific PFAS compounds
can measure only a small subset of PFAS. Thus, total fluorine methods are preferred for food packaging
applications. Commercially available total fluorine methods include combustion followed by ion chroma-
tography or fluoride ion-selective electrodes. Surface measurement techniques are also under development,
which may be particularly useful for nondestructive, rapid screening of food packaging materials. This
chapter provides a discussion of the various methods available, and under development, for quantifying
PFAS in food packaging materials. Alternative strategies to impart water and grease resistance to food
packaging are also discussed.
Key words PFAS, Perfluorinated compounds, Total fluorine, Detection methods, Food packaging,
PFAS alternatives
1 Introduction
Food packaging materials rely on additives to improve gas and

moisture barriers, oil resistance, and other properties. Among the
most common, and recently scrutinized, classes of additives are
poly- and perfluorinated alkyl substances (PFAS), which impart
water- and oil-repellency to paper and molded fiber products
[1]. PFAS contain multiple carbon-fluorine bonds, which are
incorporated into alkyl chains to produce hydrophobic and oleo-
phobic compounds [2]. In food packaging, the unique ability of
these compounds to impart low surface tension has led to their
widespread use in food contact materials that come in contact with
oil and grease [3]. For instance, PFAS compounds are commonly
https://doi.org/10.1007/978-1-0716-3613-8_5,
99
100 Rachel C. Scholes et al.
used in microwave popcorn bags [4] and fast-food packaging [1]

including pizza boxes and French fry wraps.
However, PFAS have recently been linked to a wide range of
toxic effects, raising questions about the safety of these additives
[5, 6]. Studies conducted with one PFAS compound, perfluorooc-
tanoic acid (PFOA), indicate likely links between exposure and
several adverse outcomes, including high cholesterol, thyroid dis-
ease, reproductive and developmental toxicity, and cancer
[6]. PFOA has been classified as a possible human carcinogen by
the International Agency for Research on Cancer due to studies
linking the compound to kidney and testicular cancers. Immuno-
toxicity has also been observed in children with elevated serum
levels of three PFAS compounds. Most toxicity studies thus far
focus on a handful of compounds, particularly PFOA and perfluor-
ooctane sulfonic acid (PFOS), which were among the first PFAS to
be widely detected in consumer products and the environment.
The toxicity of other PFAS, including those used as replacements
for PFOS and PFOA in industry applications, requires further study
and the development of robust methodologies for read-across from
structurally similar PFAS compounds [7]. In addition, although
PFAS are often present as mixtures, the toxicity of PFAS mixtures is
complex and poorly understood [8]. As a result of the growing
evidence for PFAS toxicity, an increasing array of environmental
experts recommend that PFAS should be treated as a hazardous
chemical class and avoided wherever possible [9].
Food packaging can contribute to human exposure to PFAS
through multiple routes. PFAS can leach from packaging into food
during typical use of items such as microwave popcorn bags
[10, 11]. PFAS have been detected in food items such as packaged
meats [12], popcorn [13], and fast-food items [13, 14], which has
raised concerns about dietary exposure to these compounds result-
ing from their use in food packaging. In addition, PFAS can con-
taminate the food supply through indirect pathways because the
strong carbon-fluorine bonds in PFAS make these compounds
recalcitrant to degradation, resulting in their introduction into
the environment at the end-of-life of food packaging products
[6]. For instance, PFAS can be introduced to food crops when
food packaging-derived compost is applied to agricultural lands
[15]. Indirect PFAS contamination also occurs due to the uptake
of PFAS from contaminated irrigation water and from land-applied
biosolids, which can result in exposures that exceed EPA health
guidelines [16, 17]. The widespread use and recalcitrance of PFAS
have resulted in these compounds becoming ubiquitous environ-
mental contaminants that have now been detected on all continents
[2, 18] including in arctic ice cores [19], in widespread aquatic
environments [20], and in the atmosphere [21]. The full implica-
tions of PFAS contamination to our food chain is still unknown.
PFAS in Food Packaging 101
Regulatory agencies are now taking notice and beginning to

restrict the use of PFAS in food packaging. For instance, in 2015,
the Danish Ministry for Environment and Food issued a guideline
level for organic fluorine in food contact materials in order to
reduce consumer exposure. The limit was revised in 2018 to its
current level, 20 μg g-1. In the USA, several states have passed or
are considering legislation to ban PFAS-containing food packaging
[22]. In 2018, San Francisco became the first USA city to ban the
addition of fluorinated chemicals to all single-use compostable food
packaging, effective in 2020 [23].
Regulatory action has also been taken to reduce the input of
PFAS into compost. The EU Directive on Packaging and Packag-
ing Waste, which provides a standard for compostable and biode-
gradable packaging, now includes a limit of 100 mg kg-1 fluorine
[24]. The Biodegradable Products Institute, which provides com-
postability certification in the USA, has adopted this limit as of
2019 [25]. These actions effectively designate PFAS-containing
packaging as non-compostable, further motivating the move away
from PFAS in food packaging.
Enacting regulations that limit PFAS in food packaging is
challenging, in part because PFAS can contaminate food contact
materials (FCMs) even when not intentionally added. Regulations
often limit the intentional addition of PFAS but allow for the
possibility that PFAS may be present in the final product due to
unintentional sources [26]. Readers are invited to refer to Chap. 4
for further details on intentionally and non-intentionally added
substances. In addition to the use of PFAS in base materials or
linings, PFAS are introduced to food packaging materials inciden-
tally when used as release agents and lubricants during the
manufacturing of FCMs [27]. PFAS can also be present in recycled
fiber used in the manufacture of new food packaging products
[26]. PFAS from these sources cannot be readily distinguished
from “intentionally added” PFAS based on chemical structure
and are instead determined based on the concentration of PFAS
in the final product, following the assumption that high PFAS
concentrations correspond to intentional use. However, the levels
of PFAS imparted during manufacturing are not well characterized.
For example, the US Food and Drug Administration allows for
levels up to 2000 mg kg-1 of some PFAS compounds used as
manufacturing aids for food packaging products [28], but the
actual concentrations imparted during manufacturing are unknown
and may vary widely. These challenges highlight the need for robust
quantitative methods to measure PFAS in food packaging materi-
als, in order to better constrain potential health effects and to
inform and enforce future regulations.
2 Methods
2.1 Background Reliably quantifying PFAS in food packaging requires particular

attention to inclusivity, selectivity, and ease-of-use. The first chal-
lenge is inclusivity: while analytical methods for food contaminants
typically quantify single compounds, PFAS comprises a broad class
of thousands of individual chemicals, which cannot all be included
in existing food safety workflows. Methods that measure total
fluorine have been identified as a strategy to address this particular
challenge. However, the possible presence of inorganic fluoride
interferes with PFAS determination by total fluorine methods,
resulting in a tradeoff between selectivity and inclusivity [29]. Fur-
thermore, extensive sample preparation requirements, for instance,
multi-step extractions from solid materials, limit the usefulness of
some methods for regulatory, product screening purposes or online
process monitoring. These considerations provide important con-
text for considering the breadth of methods under development,
and the promise of new innovations in solids analyses and rapid
screening.
Initial methods for measuring PFAS in the environment used
liquid chromatography and mass spectrometry to quantify specific
PFAS compounds using targeted methods. Standard targeted
methods exist for subsets of PFAS compounds found in water and
soil (e.g., USEPA Method 537.v1; ISO Method 251010; ASTM
D7979; ASTM D7968), but these methods detect only approxi-
mately 20 of the thousands of existing PFAS compounds, and do
not target some of the compounds most commonly found in food
packaging, such as the dialkyl and trialkyl phosphate esters (diPAPs
and triPAPs, respectively). When compared side-by-side, targeted
PFAS methods detect a small portion of total organofluorine com-
pounds, indicating that existing targeted methods poorly represent
total potential PFAS exposure [3, 30]. Importantly, compounds
that are not detected in targeted analyses likely still pose health risks
and, in some cases, are known precursors to contaminants with
well-characterized toxicity. For instance, diPAPs found in food
packaging materials act as endocrine disruptors [31] and are also
metabolized to toxic perfluorinated carboxylic acids [13].
The total oxidizable precursor (TOP) assay was developed to
address a broader suite of PFAS compounds compared to targeted
methods. The TOP assay involves oxidizing a PFAS-containing
sample to form PFAS oxidation end-products, specifically perfluori-
nated carboxylic acids (PFCAs), which are then analyzed with a
targeted LC-MS/MS method. In this way, the TOP assay provides
a total concentration of PFCAs and their precursors. This method
was developed for aqueous samples where the source of PFAS was
fire-fighting foams [32], and it has not been tested on polymers
used in food packaging or newer ether-linked PFAS. The TOP assay
Fig. 1 Classification of poly- and perfluorinated alkyl substances (PFAS) detection methods
PVDF
core cladding coating
18800 1.0
18775 resultant
0.8 reflection
Normalized Intensity
18750
refractive index (n), thickness (L)
18725
OPD/nm
0.6
water
18700 broadband circulator
0.05 ppb
0.4 light source
18675 0.1 ppb
0.15 ppb
PVDF
18650 0.19 ppb
0.2 0.23 ppb
coated
18625 y= 177.87x+ 18669
0.26 ppb fibre
R² = 0.9388 0.3 ppb computer 22°c
18600 0.0 PFOA
0 0.1 0.2 0.3 1530 15451560 15751590 optical spectrum solution incubator
[PFAS]/ppb Wavelenght/nm analyzer
Fig. 2 Portable poly- and perfluorinated alkyl substances (PFAS) sensor device. (Reproduced from Ref. [50]
with permission from Elsevier)
also cannot detect short-chain compounds (e.g., C2 and C3 com-

pounds) that are not retained by HPLC columns, although short-
chain PFCAs could be quantified by pairing the TOP assay with ion
chromatography [33]. The diPAPs commonly used in food contact
materials were effectively converted to PFCAs detectable by the
TOP assay in spiked soil samples [33], indicating that this tech-
nique could provide evidence for the presence of PFAS in food
packaging [34]. However, further development of extraction meth-
ods for polymer-bound PFAS, along with verification that food
packaging-relevant PFAS are effectively converted to PFCAs,
would be required in order to implement the TOP assay for food
packaging materials. With these uncertainties it is not clear that the
TOP assay can be used to enforce regulations on PFAS in food
packaging applications.
The accuracy, ease of use, and, thus, usefulness of mass
spectrometry-based techniques depend on the pretreatment of
the samples, which necessarily includes an extraction protocol

[35]. PFOA and PFAS from liquid matrices are typically extracted
using solid-phase extraction [36], but other methods are also
reported (Fig. 1). These methods include solid-phase [37] or
liquid-phase microextraction [38] and ion-pair extraction
[39]. Online solid phase extraction processes [40, 41] are appealing
because they reduce sample preparation time, although extraction
from solid products to liquid solutions would still be necessary in
order to use this technique within the food packaging process
chain.
In contrast to the above techniques, total F methods quantify
the sum of all fluorine-containing compounds without identifying
specific chemicals and can measure fluorine directly in a solid sam-
ple rather than relying on extraction. These techniques can be used
for quantifying total PFAS if no inorganic fluoride is present or if
inorganic fluoride can be separately measured and subtracted from
total F. Combustion ion chromatography (CIC) is a relatively
common, sensitive, and automated technique for total fluorine
analysis. CIC is reportedly able to measure fluorine with detection
limits as low as 0.8 μg g-1 [3] or 16 nmol cm-2 [1]. Further
analytical details of CIC detection methods are provided below.
Extraction techniques—for example, extractable organofluorine
(EOF) or adsorbable organofluorine (AOF)—have also been paired
with CIC to determine the organic fraction of total F, in cases
where inorganic F is significant. However, further development
of these techniques for solid samples is needed, since the current
methods have limited ability to extract fluoropolymers from solids,
and therefore may underestimate PFAS concentrations. For
instance, EOF extracted <5% of total F from food contact materials
in a recent methods assessment where the contribution of inorganic
F was unknown [3]. The EOF procedure also has low recovery for
some nonpolymer compounds, such as fluorosulfonamides [42],
and increases the total CIC method time from approximately
20 min to over 8 h. The regulatory agency for food packaging
materials in Denmark is developing a method for CIC with inor-
ganic F subtraction to be used for compliance testing of paper
and board matrices [43], which may set a precedent for the use
of a CIC method in regulatory screening processes. Combustion
can alternatively be combined with detection using a fluoride
ion-selective electrode (ISE). This method relies on readily avail-
able laboratory equipment [44, 45] and is currently conducted by
commercial labs in the USA [46].
Surface characterization techniques provide an alternative to
combustion-based methods for total fluorine. Multiple surface
techniques have been used for food packaging, such as particle-
induced γ-ray emission (PIGE) and instrumental neutron activa-
tion analysis (INAA) [3]. PIGE in particular has recently emerged
as a promising analysis technique for total F on solid surfaces
[47]. However, to date, no adaptation exists to subtract inorganic F

from the total F measured using PIGE, so this technique relies on
the assumption that inorganic F is negligible. As a surface charac-
terization technique, PIGE could be used to rapidly and nonde-
structively screen for PFAS-containing surface coatings. This
technique may, however, overestimate bulk PFAS content because
it measures F concentrations only within approximately 200 μm of
the surface, where PFAS concentrations may be higher [3]. Efforts
are currently underway to miniaturize PIGE analysis so that it can
be more easily conducted in the field. If these efforts are successful,
PIGE may become a convenient technique for food packaging
product screening.
Additional surface techniques that could be used for rapid
testing are also under development but require significant further
research. Spectroscopic techniques, including Fourier-transform
infrared (FTIR) and Raman spectroscopies, are promising because
they may require minimal sample preparation. However, most
other spectroscopic techniques do not have the sensitivity to
match mass spectrometry. Surface-enhanced Raman spectroscopy
(SERS) has been proposed as a method with high sensitivity
[48, 49] whereby Raman scattering is measured from molecules
adsorbed onto nanostructured surfaces, such as gold or silver
nanoparticles or activated carbon, thus enhancing the scattering
through surface concentration of the target molecule. This
surface-sensitive technique can theoretically provide increases in
sensitivity of 1010 to 1014 compared to standard Raman spectros-
copy, resulting in the ability to detect compounds at parts per
billion (ppb) levels. For instance, Fang et al. detected PFAS via
SERS using silver nanoparticles deposited onto a graphene surface,
in conjunction with cationic Raman dyes (ethyl violet or methylene
blue) that formed ion pairs with PFAS compounds. They detected
PFOA, PFOS, and 1H,1H,2H,2H-perfluorooctanesulfonic acid
(6:2FTS), with a detection limit of approximately 50 ppb for
PFOA [49]. It should be noted that the Raman signal intensity
was boosted by the dye, rather than the target PFOA molecule,
making this an indirect method of detection. In addition, the
applicability of this technique to solid samples has not been
investigated.
Finally, there is a desire for robust, portable sensing systems
that can be used for testing potentially contaminated materials in
the field or to be used in industrial facilities for ensuring online
compliance. Recently, Faiz et al. developed optical fiber sensors
capped with poly(vinylidene fluoride) (PVDF) that rely on changes
in optical interference to detect the presence of PFOA (Fig. 2)
[50]. The PVDF interacts specifically with PFAS molecules via
hydrophobic and dipole–dipole interactions, resulting in an offset
interference pattern corresponding to the concentration of PFOA
in an aqueous test solution. Work is underway to correlate
interference patterns that could differentiate other PFAS com-

pounds, and further consideration of sensitivity is needed.
The advantage of sensors based on optical fibers is their com-
patibility with monitoring devices. Fiber optic monitoring devices
can be coupled with standard smartphone apps, which could allow
for real-time information from field tests or process
monitoring [51].
A handful of other phone app-compatible sensors and kits for
PFAS detection have been introduced [52]. These include optical
test kits based on methylene blue active substances (MBAS) [53],
surface-enhanced Raman scattering [49], and the use of molecular
imprinted polymer-based ISE [54]. These systems usually require a
color change in response to specific interactions between PFAS and
indicator molecules.
Many PFAS surfactants are anionic and will interact with cat-
ionic dyes such as methylene blue or ethyl violet to form an ion-pair
that results in color changes detectable by calibrated optical systems
or even the naked eye. These surfactant-dye ion-pairs are often
hydrophobic (since their hydrophilic ends have been blocked via
their electrostatic interaction) which means they can be concen-
trated in a nonaqueous phase, which increases method sensitivity.
However, color detection from these systems can vary with pH,
temperature, and weather conditions, indicating a need for further
development of these methods to ensure they are robust enough
for field implementation. Additionally, ion-pair methods have pri-
marily been developed to test for PFAS in water, not in solid
matrices. Accordingly, assaying food packaging papers or plastic
wraps with such methods would require standardizing extraction
methods to achieve reproducible concentrations of fluorinated
compounds over varying packaging samples.
2.2 Selection and Overall, each of the methods discussed above has unique advan-
Implementation tages for different applications. Table 1 indicates the analytes
detected, sensitivity, and sample preparation requirements of the
relatively established methods. Targeted methods are useful for
identifying the presence of specific PFAS compounds and quantify-
ing concentrations at low levels. In contrast, total fluorine methods
such as CIC, C-ISE, and PIGE can provide sensitive detection of
fluorine with low sample preparation requirements, but do not
provide information on the specific fluorinated compounds
present.
In a regulatory context where PFAS are treated as a chemical
class, the use of nontargeted methods is likely acceptable for food
packaging products at the screening stage. For instance, total fluo-
rine methods can be used for product screening, and can be fol-
lowed by requests for material data sheets, additive lists, and
information about manufacturing processing aids when fluorine is
detected [46]. The use of total fluorine methods could also be
Table 1
Method characteristics
Method Analyte(s) Sensitivity Sample preparation

-1
LC-MS/MS Specific PFAS compounds ng L in extract Solvent extraction
included in targeted method
TOP assay PFCAs and PFCA precursors ng L-1 in extract Solvent extraction and oxidation
-1
CIC Total F, total organic F μg g Total F: None
TOF: Extraction of organic F
C-ISE Total F μg g-1 None—Uses solid material
PIGE Total F μg g-1 None—Uses solid material
Table 2
Commercial labs for PFAS in food packaging products
Lab Analytical technique Location

Galbraith C-ISE USA
Eurofins LC-MS/MS, TOP, CIC USA
SGS Total F Hong Kong
ALS LC-MS USA
Intertek CIC or C-ISE USA, Belgium
supported by commercial laboratories that are beginning to offer

total fluorine screening in food packaging products. A noncompre-
hensive list of commercial labs that advertise their services in offer-
ing PFAS measurements in food packaging products is provided in
Table 2. These analyses are likely to become more widely available
as demand for PFAS screening increases. As rapid detection meth-
ods and sensors are further developed, these techniques may also be
incorporated into routine monitoring or screening processes.
3 Combustion Ion Chromatography (CIC)
3.1 Materials • Furnace with water supply (e.g., Auto Quick Furnace AQF-100,
Dia Instruments Co. Ltd.)
• High purity Ar, O2.
• Ion chromatograph (e.g., ICS-3000, Dionex Co. Ltd. with
conductivity detector).
• IonPac AS20 column (77.5 μeg/column; 2 mm i.d. × 250 mm

length, 7.5 μm).
• NaOH, KOH, NaF.
• Silica boats for solid samples.
3.2 Setup The ion chromatography (IC) system must be set up specifically to
run total fluorine samples because sources of fluorine contamina-
tion exist within typical IC systems. Sources of contamination must
be removed or replaced, including substituting certain components
in the flow path of the IC (i.e., polytetrafluoroethylene-containing
tubing, gaskets, gas lines, valves, regulator) with non-fluorinated
materials (e.g., stainless steel, polyetheretherketone, polyethylene
tubing). A gas purifier with activated carbon should be used to
remove trace fluorine from gases. After modification, background
levels of fluorine <1 ng-F can be achieved [42]. For additional
details on minimizing background contamination, see [55].
3.3 Procedure 1. Prepare quantification standards using sodium fluoride.

2. Set food packaging sample on a silica boat and place into a
furnace with water supply at 900–1000 °C for 5 min. Prefera-
bly, use a solids autosampler to automate analyses of multiple
samples.
3. Online IC should be set to analyze liberated fluoride from the
combustion chamber.
(a) IC software (e.g., Chromeleon if using Dionex IC) can be
used to integrate peak areas and provide report of sample
concentrations.
4. It is recommended to analyze samples in duplicate [56].
4 Alternatives to PFAS
The health and environmental hazards of PFAS, combined with

their ability to migrate into food, have led to a surge in interest in
developing alternatives for fiber and paper-based packaging. Alter-
native strategies include applying non-PFAS substances as sizing
agents or replacing fiber products with compostable plastics or
bamboo-based materials with inherent grease resistance. The appli-
cation of non-PFAS substances can be achieved either as (1) external
sizing agents (e.g., laminated films, coatings), which are added after
molding, or (2) internal sizing agents, which, like small molecule
PFAS additives, are added to the wet pulp before molding.
4.1 External Sizing Laminated films have been widely used to provide paper and paper-
Agents (Laminated board with oil and water barriers suitable for food packaging. A
Films, Coatings) disadvantage of laminated films is that they increase material and
processing costs to production, do not homogenously confer oil

and water resistance to the entire paper/paperboard, and may crack
or form pinholes, leading to leaks. Traditional petroleum-derived
films include polymers and waxes, such as polyolefins, polystyrene,
and hydrophobic acrylates [57, 58]. These films are inexpensive,
widely available, provide good water, grease, and gas barriers, and
remain stable at relatively high temperatures. A major disadvantage
of most polymers is their end-of-life outcomes because paper-based
multilayer materials containing recalcitrant plastic films are usually
impossible to recycle economically, particularly when soiled with
food residue. Furthermore, some polymer coatings contribute to
human health concerns because they can leach endocrine disrupt-
ing compounds such as plasticizers and unpolymerized
monomers [59].
Biodegradable thermoplastic materials, such as polylactide
(PLA), polyhydroxyalkanoates (PHA; including polyhydroxybutry-
rate, PHB), polybutylenesuccinate (PBS), polycaprolactone (PCL),
and thermoplastic starch (TPS) are less recalcitrant to environmen-
tal degradation than other polymers and can provide sufficient
barrier properties for food. PBS and PLA degrade readily under
industrial composting conditions, but effectively do not degrade
under conditions representative of natural environments, such as
ambient soil and aquatic conditions [60]. Despite its synthetic
nature, PCL degrades under both industrial and home composting
and soil conditions. PHB and TPS are biodegradable even when
subjected to home composting, marine, fresh water, and soil envir-
onments [60]. PLA is widely used due to its low cost but requires
additives in order to be stable in the presence of hot foods. PHAs
are more expensive than PLA but tend to be more thermally stable
and provide good moisture barriers, approaching those of polyole-
fins [61]. Although they are often sourced from sugar feedstocks,
PLA and PHA can be prepared from food and agricultural waste
streams [62, 63]. In addition to being used as sizing agents, both
traditional and biodegradable plastics can be used as substitutes for
the bulk packaging. However, polymer-based materials usually have
lower biodegradability rates than comparable fiber-based products
and may be more expensive. See Chaps. 1 and 2 for more on
biodegradation.
Film coatings made of polysaccharides are another PFAS alter-
native that can be sourced from renewable feedstocks and can be
readily biodegraded, although covalent modifications (e.g., the
introduction of functional groups) can decrease their biodegrad-
ability. The barrier properties of chitosan films and paper coatings
have been extensively investigated. Chitosan alone provided mod-
est water repellency (contact angles: 55–85°), and excellent grease
resistance when a heavy coating was applied [64]. Higher water
contact angles, that is, greater hydrophobicity, were observed with
chitosan that was functionalized with polydimethylsiloxane
(PDMS) and zein (up to 110° for water and 40–70° for castor
oil) [65]. Moderate water absorbency was achieved in both exam-

ples, accompanied by a modest reduction in water vapor transmis-
sion rates. Optimized compositions of chitosan-PDMS-zein films
achieved strong grease repellency. However, the addition of silox-
anes to the polysaccharide-based films diminishes their environ-
mental and health advantages. Although they exhibit excellent
water repellency, siloxanes are persistent in the environment, poten-
tially bioaccumulative, and may confer hazards such as endocrine
disruption [66]. Another alternative combines high-molecular
weight cationic and anionic starches, which slightly improved the
water resistance of paper while retaining oil resistance comparable
to polyethylene film [67]. Other high-molecular weight starch
treatments increased water absorption but improved oil absorption.
Commercial versions of alternatives with external sizing agents are
already in use and include natural waxes, PLA, and clay coatings
(e.g., tradenames: Practiv’s Earthchoice, Ecotainer®, Eco-Pro-
ducts®, PrimeWare®, Bare®, Solo®, Eco-Forward®, Ecowax®, and
World Centric®) [46].
4.2 Internal Sizing Internal sizing agents are advantageous in their ability to provide
Agents water and oil resistance throughout the packaging, as opposed to
just the coated surface. These additives provide aesthetic advan-
tages by enabling a more natural look to the paper. Internal sizing
agents can also be directly substituted for PFAS in the
manufacturing process and are therefore a simpler solution than
laminated films or external sizing agents, which may introduce
additional steps to production. Alkyl succinic anhydride, styrene
acrylic emulsion, alkyl ketene dimer, and rosin function as liquid
water barriers but may not impart similar enhancements to water
vapor permeability [68]. The micro- and nanostructures of nano-
materials may also provide grease and water barriers, obviating the
use of hazardous, persistent PFAS. Nanocellulose incorporated
during the wet end of paper manufacture was shown to afford an
excellent grease barrier [69].
5 Conclusions
Several alternatives have been developed that can improve the

grease and oil resistance of paper adequately to substitute for
PFAS in food packaging. Laminated films and other external sizing
agents span traditional petroleum-derived plastics, biodegradable
bioplastics, and polysaccharides. Alternative internal sizing agents
have also been identified, although fewer options have been
reported in the literature to date. While the development of
PFAS-free food packaging is an active area of research and more
development is needed, adequate solutions exist that can achieve
effective grease and water barriers without the significant human
and environmental hazards of PFAS.
References
1. Schaider LA, Balan SA, Blum A, Andrews DQ, 9. Kwiatkowski CF, Andrews DQ, Birnbaum LS,
Strynar MJ, Dickinson ME, Lunderberg DM, Bruton TA, DeWitt JC, Knappe DRU, Maffini
Lang JR, Peaslee GF (2017) Fluorinated com- MV, Miller MF, Pelch KE, Reade A, Soehl A,
pounds in U.S. fast food packaging. Environ Trier X, Venier M, Wagner CC, Wang Z, Blum
Sci Technol Lett 4(3):104–111 A (2020) Scientific basis for managing PFAS as
2. Buck RC, Franklin J, Berger U, Conder JM, a chemical class. Environ Sci Technol Lett 7(8):
Cousins IT, de Voogt P, Jensen AA, Kannan K, 532–543
Mabury SA, van Leeuwen SP (2011) Perfluor- 10. Begley TH, White K, Honigfort P, Twaroski
oalkyl and polyfluoroalkyl substances in the ML, Neches R, Walker RA (2005) Perfluoro-
environment: terminology, classification, and chemicals: potential sources of and migration
origins. Integr Environ Assess Manag 7(4): from food packaging. Food Addit Contam
513–541 22(10):1023–1031
3. Schultes L, Peaslee GF, Brockman JD, 11. Begley TH, Hsu W, Noonan G, Diachenko G
Majumdar A, McGuinness SR, Wilkinson JT, (2008) Migration of fluorochemical paper
Sandlom O, Ngwenyama RA, Benskin JP additives from food-contact paper into foods
(2019) Total fluorine measurements in food and food simulants. Food Addit Contam
packaging: how do current methods perform? 25(3):384–390
Environ Sci Technol Lett 6(2):73–78 12. Jogsten IE, Perelló G, Llebaria X, Bigas E,
4. Zabaleta I, Negreira N, Bizkarguenaga E, Martı́-Cid R, K€arrman A, Domingo JL
Prieto A, Covaci A, Zuloaga O (2017) Screen- (2009) Exposure to perfluorinated compounds
ing and identification of per- and polyfluor- in Catalonia, Spain, through consumption of
oalkyl substances in microwave popcorn bags. various raw and cooked foodstuffs, including
Food Chem 230:497–506 packaged food. Food Chem Toxicol 47(7):
5. Knutsen HK, Alexander J, Barregârd L, 1577–1583
Bignami M, Brüschweiler B, Ceccatelli S, 13. Gebbink WA, Shahid U, Oskar S, Berger U
Cottrill B, Dinovi M, Edler L, Grasi-Kraupp B, (2013) Polyfluoroalkyl phosphate esters and
Hogstrand C, Hoogenboom LR, Nebbia CS, perfluoroalkyl carboxylic acids in target food
Oswald IP, Petersen A, Rose M, Roudot AC, samples and packaging-method development
Vleminckx C, Vollmer G, Wallace H, Bodin L, and screening. Environ Sci Pollut Res 20(11):
Cravedi JP, Halldorsson TI, Haug LS, 7949
Johansson N, van Loveren H, Gergelova P, 14. Tittlemier SA, Pepper K, Edwards L (2006)
Mackay K, Levorato S, van Manen M, Concentrations of perfluorooctanesulfona-
Schwerdtle T (2018) Risk to human health mides in Canadian Total Diet Study composite
related to the presence of perfluorooctane sul- food samples collected between 1992 and
fonic acid and perfluorooctanoic acid in food. 2004. J Agric Food Chem 54(21):8385–8389
EFSA J 16(12):5194 15. Lazcano RK, Choi YJ, Mashtare ML, Lee LS
6. Sunderland EM, Hu XC, Dassuncao C, Tokra- (2020) Characterizing and comparing per- and
nov AK, Wagner CC, Allen JG (2019) A review polyfluoroalkyl substances in commercially
of the pathways of human exposure to poly- available biosolid and organic non-biosolid-
and perfluoroalkyl substances (PFASs) and based products. Environ Sci Technol 54:
present understanding of health effects. J 8640–8648
Expo Sci Environ Epidemiol 29:131–147 16. Blaine AC, Rich CD, Hundal LS, Lau C, Mills
7. Fenton SE, Ducatman A, Boobis A, DeWitt MA, Harris KM, Higgins CP (2013) Uptake of
JC, Lau C, Ng C, Smith JS, Roberts SM perfluoroalkyl acids into edible crops via land
(2020) Per- and polyfluoroalkyl substance tox- applied biosolids: Field and greenhouse stud-
icity and human health review: current state of ies. Environ Sci Technol 47:14062–14069
knowledge and strategies for informing future 17. Blaine AC, Rich CD, Sedlacko EM, Hundal
research. Environ Toxicol Chem 40:606. LS, Kumar K, Lau C, Mills MA, Harris KM,
https://doi.org/10.1002/etc.4890 Higgins CP (2014) Perfluoroalkyl acid distri-
8. Goodrum PE, Anderson JK, Luz AL, Ansell bution in various plant compartments of edible
GK (2020) Application of a framework for crops grown in biosolids-amended soils. Envi-
grouping and mixtures toxicity assessment ron Sci Technol 48:7858–7865
of PFAS: a closer examination of dose- 18. Yamashita, N; Yeung, L.W.Y.; Taniyasu, S.;
additivity approaches. Toxicol Sci 179:262. Kwok, K.Y.; Petrick, G.; Gamo, T.; Guruge,
https://doi.org/10.1093/toxsci/kfaa123 K.S.; Lam, P.K.S.; Loganathan, B.G. (2012)
Global distribution of PFOS and related che- 31. Chen P, Yang J, Chen G, Yi S, Liu M, Zhu L
micals. B.G. Loganathan, P.K.S. Lam (Eds.), (2020) Thyroid-disrupting effects of 6:2 and 8:
Global contamination trends of persistent 2 polyfluoroalkyl phosphate diester (diPAPs) at
organic chemicals, Taylor & Francis Group. environmentally relevant concentrations from
pp. 593–628 integrated in silico and in vivo studies. Environ
19. Muir D, Bossi R, Carlsson P, Evans M, De Sci Technol 7(5):330–336
Silva A, Halsall C, Rauert C, Herzke D, 32. Houtz EF, Sedlak DL (2012) Oxidative con-
Hung H, Letcher R, Rigét F, Roos A (2019) version as a means of detecting precursors to
Levels and trends of poly- and perfluoroalkyl perfluoroalkyl acids in urban runoff. Environ
substances in the Arctic environment – an Sci Technol 46(17):9342–9349
update. Emerg Contam 5:240–271. ISSN 33. Janda J, Nödler K, Scheruer M, Happel O,
2405-6650 Nürenberg G, Zwiener C, Lange FT (2019)
20. Houde M, De Silva A, Letcher RJ, Muir DCG Closing the gap – inclusion of ultrashort-
(2011) Monitoring of perfluorinated com- chain perfluoroalkyl carboxylic acids in the
pounds in aquatic biota: an updated review. total oxidizable precursor (TOP) assay proto-
Environ Sci Technol 45(19):7962–7973 col. Environ Sci: Processes Impacts 21:1926–
21. Vierke L, Berger U, Cousins IT (2013) Esti- 1935
mation of the acid dissociation constant of per- 34. Yuan G, Peng H, Huang C, Hu J (2016) Ubiq-
fluoroalkyl carboxylic acids through an uitous occurrence of Fluorotelomer alcohols in
experimental investigation of their water-to- eco-friendly paper-made food-contact materi-
air transport. Environ Sci Technol 47:11032– als and their implication for human exposure.
11039 Environ Sci Technol 50(2):942–950
22. Safer States Bill Tracker. https://www. 35. Rodriguez KL, Hwang J-H, Esfahani AR, Sad-
saferstates.org/bill-tracker/FilterBills. mani AHMA, Lee WH (2020) Recent devel-
Accessed 8/3/2020 opments of PFAS-detecting sensors and future
23. Hogue C (2018) San Francisco moves to ban direction: a review. Micromachines 11:667
food containers made with fluorinated chemi- 36. Shoemaker JA, Grimmett PE, Boutin BK
cals. Chem Eng News 96(32) https://cen.acs. (2009) Method 537, determination of selected
org/policy/legislation-/San-Francisco- perfluorinated alkyl acids in drinking water by
moves-ban-food/96/i32 solid phase extraction and liquid chromatogra-
24. UNE 13432 (2001) Requirements for packag- phy/tandem mass spectrometry (LC/MS/
ing recoverable through composting and bio- MS), Version 1.1, September 2009, National
degradation. Test scheme and evaluation Exposure Research Laboratory, Office Of
criteria for the final acceptance of packaging Research And Development, U. S. Environ-
25. Biodegradable Products Institute. Position on mental Protection Agency, Cincinnati, Ohio
fluorinated chemicals. https://bpiworld.org/ 45268: 600-R-08/092, Ver 1.1
page-1857568. Accessed 6/20/2020 37. Huang Y, Li H, Bai M, Huang X (2018) Effi-
26. Curtzwiler GW, Silva P, Hall A, Ivey A, Vorst K cient extraction of perfluorocarboxylic acids in
(2021) Significance of perfluoroalkyl sub- complex samples with a monolithic adsorbent
stances (PFAS) in food packaging. Integr Envi- combining fluorophilic and anion-exchange
ron Assess Manag 17(1):7–12 interactions. Anal Chim Acta 1011:50–58
27. Glüge J, Scheringer M, Cousins IT, DeWitt JC, 38. Wang J, Shi Y, Cai Y (2018) A highly selective
Goldenman G, Herzke D, Lohmann R, Ng dispersive liquid–liquid microextraction
CA, Trier X, Wang Z (2020) An overview of approach based on the unique fluorous affinity
the uses of per- and polyfluoroalkyl substances for the extraction and detection of per- and
(PFAS). engrXiv preprint. https://doi.org/10. polyfluoroalkyl substances coupled with high
31224/osf.io/2eqac performance liquid chromatography tandem–
mass spectrometry. J Chromatogr A 1544:1–7
28. U.S. Food and Drug Administration. Inven-
tory of Effective Food Contact Substances 39. Villaverde-de-Sáa E, Racamonde I, Quintana
(FCS) Notifications. Accessed 7/15/2020 JB, Rodil R, Cela R (2012) Ion-pair sorptive
extraction of perfluorinated compounds from
29. McDonough CA, Guelfo JL, Higgins CP water with low-cost polymeric materials: poly-
(2019) Measuring total PFASs in water: the ethersulfone vs polydimethylsiloxane. Anal
tradeoff between selectivity and inclusivity. Chim Acta 740:50–57
Curr Opin Environ Sci Heal 7:13–18
40. Wilson SR, Malerød H, Holm A, Molander P,
30. Borg D, Ivarsson J (2017) Analysis of PFASs Lundanes E, Greibrøkk T (2007) On-line
and TOF in products. Nordic Council of SPE-Nano-LC-Nanospray-MS for rapid and
Ministers sensitive determination of perfluorooctanoic
acid and perfluorooctane sulfonate in river appbased/portable sensor for the detection of
water. J Chromatogr Sci 45(3):146–152 fluoro-surfactant PFOA. Chemosphere 191:
41. Barreca S, Busetto M, Vitelli M, Colzani L, 381–388
Clerici L, Dellavedova P (2018) Online solid- 53. Megharaj M, Ravendra N, Mercurio P (2011)
phase extraction LC-MS/MS: a rapid and valid Anionic surfactant detection. AG01N2162FI
method for the determination of perfluori- I. PCT, Australia. AG01N2162FI
nated compounds at sub ngL-1 level in natural 54. Fang C, Zuliang C, Megharaj M, Naidu R
water. J Chem 2018:1 (2016) Potentiometric detection of AFFFs
42. Miyake Y, Yamashita N, So MK, Rostkowski P, based on molecular imprinting polymer. Envi-
Taniyasu S, Lam PKS, Kannan K (2007) Deter- ron Technol Innov 5:52–59
mination of trace levels of total fluorine in 55. Wagner A, Raue B, Brauch HJ, Worch E,
water using combustion ion chromatography Lange FT (2013) Determination of adsorbable
for fluorine: a mass balance approach to deter- organic fluorine from aqueous environmental
mine individual perfluorinated chemicals in samples by adsorption to polystyrene-
water. J Chromatogr A 1143(1–2):98–104 divinylbenzene based activated carbon and
43. Trier X, Taxvig C, Rosenmai AK, Pedersen GA combustion ion chromatography. J Chroma-
(2017) PFAS in paper and board for food con- togr A 1295:82–89
tact: options for risk management of poly- and 56. Koch A, Aro R, Wang T, Yeung LWY (2020)
perfluorinated substances. Nordic Council of Towards a comprehensive analytical workflow
Ministers for the chemical characterization of organo-
44. Galbraith Laboratories. GLI method summary: fluorine in consumer products and environ-
determination of total fluorine by oxygen flask mental samples. TrAC Trends Anal Chem
combustion and ion-selective electrode. 123:115423
Retrieved from: http://galbraith.com/wp-con 57. V€ah€a-Nissi M, Kervinen K, Savolainen A,
tent/uploads/2015/08/E9-3-Total-Fluo Egolf S, Lau W (2006) Hydrophobic polymers
rine-by-Oxygen-Flask-Combustion-ISE-GLI- as barrier dispersion coatings. J Appl Polym Sci
Method-Summary.pdf 101:1958–1962
45. EPA Method 340.2. Fluoride (potentiometric, 58. Krook M, G€allstedt M, Hedenqvist MS (2005)
ion selective electrode). Revised 1974 A study on montmorillonite/polyethylene
46. Safer Chemicals, Healthy Families (2020) A nanocomposite extrusion-coated paperboard.
guide for quick-service restaurant chains: ban- Packag Technol Sci 18:11–20
ning PFAS in food-contact materials. https:// 59. Muncke J (2009) Exposure to endocrine dis-
saferchemicals.org rupting compounds via the food chain: is pack-
47. Ritter EE, Dickinson ME, Harron JP, Lunder- aging a relevant source? Sci Total Environ
berg DM, DeYoung PA, Robel AE, Field JA, 407(16):4549–4559
Peaslee GF (2017) PIGE as a screening tool for 60. Narancic T, Verstichel S, Reddy Chaganti S,
per- and polyfluorinated substances in papers Morales-Gamez L, Kenny ST, De Wilde B,
and textiles. Nucl Instrum Methods Phys Res Babu Padamati R, O’Connor KE (2018) Bio-
407:47–54 degradable plastic blends create new possibili-
48. Ong TTX, Blanch EW, Jones OAH (2020) ties for end-of-life management of plastics but
Surface enhanced Raman spectroscopy in envi- they are not a panacea for plastic pollution.
ronmental analysis, monitoring and assess- Environ Sci Technol 52:10441–10452
ment. Sci Total Environ 720:137601 61. Dilkes-Hoffman LS, Pratt S, Lant PA, Levett I,
49. Fang C, Megharaj M, Naidu R (2016) Surface- Laycock B (2018) Polyhydroxyalkanoate coat-
enhanced Raman scattering (SERS) detection ings restrict moisture uptake and associated loss
of fluorosurfactants in firefighting foams. RSC of barrier properties of thermoplastic starch
Adv 6(14):11140–11145 films. J Appl Polym Sci 135:46379
50. Faiza F, Baxter G, Collins S, Sidiroglou F, 62. Serafim LS, Lemos PC, Albuquerque MGE,
Crana M (2020) Polyvinylidene fluoride coated Reis MAM (2008) Strategies for PHA produc-
optical fibre for detecting perfluorinated che- tion by mixed cultures and renewable waste
micals. Sens Actuators B Chem 312:128006 materials. Appl Microbiol Biotechnol 81:615–
51. Wang F, Lu Y, Yang J, Chen Y, Jing W, He L, 628
Liu Y (2017) A smartphone readable colori- 63. Fahim IS, Chbib H, Mahmoud HM (2019)
metric sensing platform for rapid multiple pro- The synthesis, production & economic feasibil-
tein detection. Analyst 142:3177–3182 ity of manufacturing PLA from agricultural
52. Fang C, Zhang X, Dong Z, Wang L, waste. Sustain Chem Pharm 12:100142
Megharaj M, Naidu R (2018) Smartphone
64. Li Z, Rabnawaz M, Khan B (2020) Response characteristics of siloxanes in biosolids. Ecotox-

surface methodology design for biobased and icol Environ Saf 102:79–83
sustainable coatings for water- and oil-resistant 67. Chi K, Wang H, Catchmark JM (2020) Sus-
paper. ACS Appl Polym Mater 2(3): tainable starch-based barrier coatings for pack-
1378–1387 aging applications. Food Hydrocoll 103:
65. Hamdani SS, Li Z, Rabnawaz M, Kamdem DP, 105696
Khan BA (2020) Chitosan-graft-poly 68. Hubbe MA (2007) Paper’s resistance to
(dimethylsiloxane)/Zein coatings for the fabri- wetting – a review of internal sizing chemicals
cation of environmentally friendly oil- and and their effects. Bioresources 2:106–145
water-resistant paper. ACS Sustain Chem Eng 69. Tayeb AH, Tajvidi M, Bousfield D (2020)
8(13):5147–5155 Paper-based oil barrier packaging using lignin-
66. Surita SC, Tansel B (2014) A multiphase anal- containing cellulose nanofibrils. Molecules 25:
ysis of partitioning and hazard index 1344
Chapter 6
Migration of Building Blocks, Additives, and Contaminants

from Food Packaging Materials
Victor G. L. Souza, Regiane Ribeiro-Santos, Patricia F. Rodrigues,
Carolina Rodrigues, João R. A. Pires, Ana T. Sanches-Silva,
Isabel Coelhoso, Fátima Poças, and Ana L. Fernando
Abstract
Packaging plays an important role in the maintenance of the quality and safety of food products. It is also
the link between the industry and the consumer, through which information is provided concerning
nutritional composition, shelf-life, and storage conditions, in addition to playing the role of product
advertising. Despite all these important functions of food packaging, it can pose a risk to consumers’ health
due to the possible migration of building blocks, additives, degradation products, and contaminants to the
packaged food. In this regard, migration assays are designed to assess the safety of food packaging materials.
This chapter provides a guideline of these assays, as well as some case studies on this topic and an insight on
the safety of food contact materials and additives.
Key words Diffusion, Food simulants, Nanoparticles, Nanoforms, Analytical techniques, Migration,
Food contact materials
1 Introduction
The main goal of food packaging is to protect the food from

tampering or (re)contamination from chemical, physical, and/or
biological sources [1]. Glass, metal, paper and paperboard, and
plastics are the most important groups of materials used in the
food packaging industry [2]. Traditionally, conventional packaging
materials should be completely inert, that is, should not interact
with the packaged food. Thus, it is important to assess if any
additives, building blocks, or other contaminants migrate from
the food contact materials (FCMs) to the food contained in it and
at which levels. It is recognized that chemicals from packaging and
other FCMs can migrate into the food itself and thus be ingested by
the consumer. Monitoring this migration has become an integral
part of ensuring food safety [3]. Some of the major recent chemical
https://doi.org/10.1007/978-1-0716-3613-8_6,
115
116 Victor G. L. Souza et al.
hazards related to migrants from packaging are plasticizers, sub-

stances used in printing inks, hydrocarbons from mineral oils, and
non-intentionally added substances—see Chap. 4 for details on the
latter [4].
According to Regulation (EC) No 1935/2004 [5], a proper
food packaging should be “the container that protects the food
from dirt or dust, oxygen, light, pathogenic microorganisms, mois-
ture and a variety of other destructive or harmful substances. Pack-
aging must also be safe under its intended conditions of use, inert,
cheap to be produced, lightweight, easy to dispose of or to reuse,
able to withstand extreme conditions during processing or filling,
impervious to a host of environmental storage and transport con-
ditions and resistant to physical abuse.”
However, with the developments in this sector, novel functions
have been attributed to the food packaging further than those
traditional ones (i.e., to contain, inform, transport, sell, and pro-
tect). For instance, modified atmosphere packaging (MAP) and
active/intelligent packaging are technologies recently created that
beyond those basic functions are capable to increase the product
quality and safety, by extending the food shelf-life [6]. Nanotech-
nology has also been increasingly used in the packaging field to
improve physical properties and as a support to active and intelli-
gent systems, nanosensors, and smart labels [7].
Concerning those new packaging technologies, as any FCM,
they must also comply with the current FCM legal framework and
guarantee that any substances migrating from the material into the
food could endanger human health, change negatively the compo-
sition of food, or damage its organoleptic characteristics [8]. In the
case of active packaging, where the active compound may be
intended to migrate to the packaged food, such compounds must
also be listed as generally recognized as safe (GRAS) by the United
States Food and Drug Administration (FDA), being labelled as a
food ingredient [9].
The partial replacement of oil-based and/or nonbiodegradable
polymers by those from renewable resources and/or
biodegradable—for some specific applications, for example,
single-use plastics—represents a major trend in the packaging
industry to address sustainability and circularity [10]. Thus, biopo-
lymers emerge as alternative substitutes to petrochemical polymers;
however, their use is often limited by their frequently poorer
mechanical and barrier properties [6]. To overcome such hurdles,
nanostructures are incorporated into these biopolymers for rein-
forcement purposes, which arouses another concern in terms of
their safety to the consumers [11]. Yet, the implications of the use
of engineered nanomaterials (ENMs) in FCM are not well estab-
lished, and further research is demanded [12].
In this chapter, a guideline to evaluate the safety of food
packaging materials is summarized, focusing on the migration of
contaminants such as building blocks and additives. It presents a
Migration from Food Packaging Materials 117
description of the testing conditions, diffusion processes, and quan-

tification techniques, followed by some recent studies in this field,
and finishes with a short insight regarding the safety of nanomater-
ials intended for food contact (see Note 1).
2 Methods to Study the Migration of Contaminants from FCMs and Associated

Analytical Techniques
Migration is defined as the mass transfer process (Fig. 1) from any

constituent initially present in the packaging material to the pack-
aged food product (food or beverages) [13]. The main mechanism
ruling this phenomenon is the diffusion, which can be defined as
the mass transfer resulting from the movement of molecules with-
out the action of external forces (e.g., agitation), due to the differ-
ences in the chemical potential (i.e., from high concentration to the
lower concentration) until the equilibrium is reached [14]. The
migration is, therefore, often a diffusion process of undesirable
compounds (in major cases, not intended) that may impact the
food in two ways, namely (i) safety, due to the migration of harm-
ful/toxic compounds, and (ii) quality, owing to the migration of
substances that impart taint or flavor and odor.
There are also other phenomena related to the migration of
substances from FCMS, such as set-off or leaching [15]; however,
the most common is through diffusion, thus these two concepts are
used interchangeably despite not being completely the same.
Single-Layer Packaging Multiplayer Packaging
Migration
Migration
Printing
ink
Food or Food or
Packaging Packaging Packaging
Food simulant Food simulant
material material 1 material 2
Adhesive
Fig. 1 Migration process of contaminants, additives, nanostructures, and building blocks from food packaging
material. Arrows indicate the direction of diffusion due to the concentration gradient. Color objects represent
the migrants/contaminants, which can be additives, building blocks, or nanostructures. Food simulant to be
assigned according to the regulation applied. Scheme can be used for both single- and multilayer packaging
materials
In general, the diffusion process is governed by models based

on Fick’s second law, which can be simplified in Eq. 1 [6, 16–18]:
0,5
M F ,t 2 Dt
= ð1Þ
M P,0 L P π
where, MF,t is the amount of migrant in the food (simulant) at time
t, MP,0 is the initial amount of migrant in the packaging, D [cm2 s-
1
] is the diffusion coefficient of the migrant in the packaging, and
LP [cm] is the thickness of the packaging [18].
Overall, migration, specific migration, and diffusion and parti-
tion coefficients are important parameters to be determined, and
are defined as follows:
(a) Overall migration (also known as total migration): is the sum
of all substances that can migrate from the FCM to the food
(or food simulant) [19].
(b) Specific migration: is the total of an individual and identified
substance that can migrate from the FCM to the food (or food
simulant); it is generally associated with toxicological
studies [19].
(c) Diffusion coefficient (D): is the parameter that represents the
speed of diffusion of substances/compounds from the pack-
aging into the food or food simulant [20].
(d) Partition coefficient (K): describes the relation between the
concentration in the packaging material and the food, at equi-
librium, or the amount of migrant that is transferred to the
food [21].
Each country or group of countries (e.g., European Union)
defines their implemented legal procedures of the migration limits,
in terms of overall migration limit (OML) and specific migration
limit (SML) [22]—see Note 2.
The investigation of migration is important and should be
carried out case by case, once the level of migration depends on
several intrinsic and extrinsic factors, such as the physical-chemical
nature of migrants (substances that tend to migrate) and foods, the
type of packaged food, the exposure temperature and time of
contact, the interfacial area between food and FCM packaging,
and the characteristics of the packaging material itself [4, 18, 19].
Migration tests can be performed in vitro using food simulants
or in situ (quantified in the food after its direct contact with the
packaging material). The assay can be divided into two steps:
(i) migration process; and (ii) identification/quantification of the
migrant [22].
The identification/quantification of the diffused compounds is
done using analytical techniques, chromatography and spectros-
copy being the most widespread. Preferably, this quantification
should be done directly in the food (in situ assay), after contact with
the packaging. However, this procedure can be difficult due to the
complexity of food matrices that pose analytical difficulties
[6, 22]. Different approaches ranging from estimation based on
the assumption of total transfer to measurements in real food may
be proposed [4]. In vitro methods have been standardized by
regulatory agencies in rules/protocols such as the Regulation
(EC) no. 10/2011 from European Union [19], which establishes
the authorized food simulants and the assay conditions (time and
temperature) to be followed—see Note 3.
The conditions of the migration tests are also defined in Regu-
lation (EU) no. 10/2011 and its amendments [19] and take into
consideration the contact time and temperature in the worst fore-
seeable use of the material (see Note 4).
In the case of paper and cardboard, one of the main concerns is
regarding the possible migration of mineral oil hydrocarbons
(MOH). MOH are mixtures of nonidentified substances that may
have carcinogenic potential and can migrate from this class of
packaging material [23]. For this purpose, Tenax® (commercial
name for Simulant E), a porous polymer absorbent, is used as a
food simulant for examining the migration of volatile and semi-
volatile substances from paper and cardboard into dry, nonfatty
foods [19, 23].
The safety of food products contained in metal packaging is
generally assessed by in situ studies, which means that the quantifi-
cation/identification of the migrants is done directly in the food
after the contact time. Studies on long-term migration from metal
packaging, corresponding to long storage times characteristic of
the normal shelf-life of the packaged food have been reported,
aiming at the search for bisphenol (present in the varnishes) and
metal ions [24, 25].
When the test is done using food instead of food simulants, an
extra step is required to extract/separate the migrant compounds
from the food matrix, enabling its ideal characterization
[25, 26]. The same is needed for some simulants, such as vegetable
oil.
The path to a specific migration assay is summarized in the
scheme illustrated in Fig. 2.
Once the goal of this test is the identification and quantification
of the compounds diffused from the FCM toward the food, the
application of analytical techniques with low detection limits is
mandatory, and the choice of the technique and protocols to be
followed will depend on the type of the compound searched, as
summarized in Fig. 3 (see Notes 5–7).
Some studies based on the migration of contaminants are
described in the following section.
Migration from food packaging
In vitro study In situ study
Step 1 Identification of the food Step 1 Identification of the food
Choise of food simulant, Migration assay - Keep the

Step 2 temperature and time of food in contact with the FCM
contact * Step 2 at storage temperature and
commercial shelf-life, or in
Migration assay - FCM in accelerated conditions*
contact with simulant at
Step 3
temperature and time
established Extraction of target migrants
Step 3
(separation from food matrix)
Characterization of the
Step 4 Characterization of the
migrants Step 4
migrants
Risk assessment and

Step 5 conclusion on the compliance Risk assessment and
with the regulation Step 5 conclusion on the compliance
with the regulation
* In accordance to the legal procedures implemented for each country or group of countries
Fig. 2 Scheme of the steps/protocols to be followed in a migration assay of contaminants, building blocks,
nanostructures, and additives from food contact material
2.1 Migration of Contaminants are substances that have not been intentionally
Contaminants, added to the food, originating from different sources, which
Building Blocks, and includes the packaging material. However, they can be intentionally
Additives added substances (IAS) to food packaging, such as building blocks
and additives or non-intentionally added substances (NIAS), such
as those obtained from degradation or collateral reactions or impu-
rities. Readers can refer to Chap. 4 for further information on
(N)IAS in food packaging. The transfer of these contaminants
from the packaging into the food can occur at levels that can pose
human health in danger, besides the negative impact on the quality
of food [27]. Therefore, to minimize contaminants in foodstuffs
and protect the population, European Union legislation establishes
maximum limits for certain contaminants in food that are safe for
consumption as well as a list of substances authorized to be used in
the manufacture of FCM (positive list) [5, 19, 28].
With this regard, a variety of foods needs to be inspected and
measured for the presence of contaminants to ensure that the food
placed on the market is safe for the consumer. Examples of poten-
tially migrating substances are: monomers, oligomers, alkanes,
phthalates and other plasticizers, processing aids, photoinitiators
Volatile compounds Semi-volatile compounds Non-volatile compounds
Headspace, purge/trap GC, Liquid extraction, gas

Liquid extraction, HPLC
solid phase microextraction chromatography
Detection by MS, NMR\

spectroscopy, UV/VIS
Detection by MS, semi Detection by HPLC-MS, NMR
spectroscopy, semi
quantification by FID, electron spectroscopy, ICP-MS,
quantification by FID and
ionization-mass spectroscopy UV/VIS spectroscopy, DLS
electron ionization-mass
spectroscopy
Fig. 3 Analytical techniques used in the characterization of migrants: gas chromatography (GC), mass spectros-
copy (MS), flame ionization detection (FID), nuclear magnetic resonance (NMR), inductively coupled plasma mass
spectrometry (ICP-MS), high-performance liquid chromatography (HPLC), and dynamic light scattering (DLS)
(PIs), antioxidants, slipping agents, antimicrobial agents, flame-

retardants, and others found in food depending on the type of
food and FMC, such as PFAS (per- and polyfluoroalkyl
substances)—see Chap. 5 for further details on PFAS [29–
34]. These compounds may be originated from printing inks,
adhesives, and materials used to manufacture glass, plastic, paper,
cardboard, and metal-based food packaging. Several studies have
indicated the potential of migration for certain substances from
packaging materials, as depicted in Table 1. In this table, it is also
possible to check which test conditions (protocols) were applied
and also the analytical techniques used to identify the substances
that migrated from food contact materials.
An overview done by He and Bayen [35] highlights the chemi-
cal contaminants in alcoholic beverages. Monomer, oligomer,
phthalates, PIs such as isopropyl-thioxanthone (ITX) and benzo-
phenone, bisphenol A, and other bisphenols have been detected in
alcoholic beverages after their migration from the FCM—poly(eth-
ylene terephthalate) (PET) and polyethylene (PE) [35]. Parabens
Table 1
Recent studies on contaminants migrated from food packaging materials into food or food simulants
122
Packaging/FMC Test condition Analytical technique Main findings Reference

Multilayer packaging Baby food; food simulants: Acetic acid UHPLC-ESI-QTOF MS. Zorbax In total 39 compounds were detected, [30]
(PET/Al/PE) 3% and ethanol 20%. Migration Agilent eclipse plus C8 column whereby 35 of the compounds were
experiments were performed at 40 ° (2.1 mm × 100 mm, 1.8 μm particle polyester oligomers (29 cyclic and
C/10 d size). Mobile phase: Solvent A ultra 6 linear oligomers) and 4 were
pure water/methanol (98:2, v/v), detected as: Bis(2-hydroxyethyl)
solvent B methanol/ultrapure terephthalate, diethyl 5-(2-((2,4,5-
water (98:2, v/v), both mobile trimethoxybenzoyl)oxy)acetamido)
phases contained 5 mM isophthalate, methoxyeugenol, and
Victor G. L. Souza et al.
ammonium formate and 0.1 vol% bis(2- methoxyethyl) sebacate. The

formic acid. Elution: Gradient. detected chemicals were similar in
Column temperature: 35 °C. both food simulants and baby food.
Injection volume: 5 μL. Flow: Two oligomers were quantified at
300 μL min-1 concentrations above
0.010 mg kg-1, exceeding the
maximum residue levels for
baby food.
Multilayer packaging Food simulants: Ultrapure water, UPLC-MS- QTOF. Chromatography Two cyclic esters, composed by AA, [40]
(polyurethane used ethanol 10%, and ethanol 95%. with a waters UPLC BEH C18 diethylene DEG, and IPA, present
as adhesive) Tests were performed at 60 °C for column (199 2.1 mm × 100 mm in the polyurethane adhesive were
10 d; 40 °C for 3 d; 121 °C for and 1.7 μm). Column temperature: evaluated. AA-DEG migration
30 min (conditions for pasteurized 40 °C. Column flow: 0.3 mL min- values were higher than the
1
materials); and 40 °C for 3 d . Injection volume: 10 μL (QTOF) AA-DEG-IPA-DEG values in all
(condition for biological samples) and 5 μL (QqQ). Mobile phases: samples. For most multilayer
Water (phase A) and methanol materials, migration of the cyclic
(phase B) with 0.1% formic acid. esters exceeded the migration limit
Chromatography started at 98/10 (10 ng g-1) established by
phase A/phase B, changed to EU/10/2011 [19] for not-listed
0/100 in 7 min substances. Both cyclic esters are
classified as high toxicity according
to the TTC. Thus, according to
FDA and EFSA, 6 and
5, respectively, out of 20 multilayer
packaging materials tested could be
used
Plastic packaging (PE, Food simulant: Tenax® and isooctane. GC–MS; Phenomenex ZB-5MS In total 27 compounds were detected [31]
PET, PP, EVA, Migration experiments were column (such as: Slip agent, plasticizers,
polyamide) performed at. 60 °C/10 d and at (30 m × 0.25 mm × 0.25 μm). antioxidant, phthalate, alkanes, and
20 °C/2 d, respectively Carrier gas: Helium. Flow: others) in the migration
1 mL min-1. Injector temperature: experiments that occur to a larger
300 °C. Injection mode: Split less. extent in Tenax® than in isooctane.
Injection volume: 1 μL. Moreover, most of detected
Oven ramp: 50 °C for 3 min compounds in plastic samples are
ramped to 300 °C at 8 °C min-1 not included in the positive list of
and held for 3 min monomers and additives that are
allowed to be used in plastic [19]
Paper-plastic from Food simulant: 10% ethanol, 3% acetic Atomic fluorescence spectrometer Mobility and migration of mercury [54]
outside to inside: acid, 20% ethanol, 50% ethanol. (AFS -810). Wavelength: 196 nm. increase at certain temperatures
PE/PAP/PE and Migration experiments were Atomic height: 10 mm. Negative when the contact time is extended.
PE/PAP/PE/Al/ performed at: 20 °C e 40 °C for high pressure: 210 V. Atomic Migration rates are affected by
PE 1, 2, 4, 5, 7, 8, 10, and 11 d. at 60 ° temperature: 200 °C. Lamp solvent type, temperatures, and
C for 3, 6, 9, 12, 18, 24, 48, and current: 10 mA. Reading time: 10 s. food stimulants. The rates of
60 h delay time:1 s. injection volume: migration were highest to 3% acetic
0.5 mL. Carrier gas flow: acid followed to 10% ethanol, 20%
500 mL min-1. Shielded air flow: ethanol, and lowest migration to
900 mL min-1 50% ethanol. The results showed
that the presence of aluminum film
may hinder the migration of
mercury and that amount of
mercury in the printed sample is
larger than that of the unprinted
sample, it was suggesting that the
mercury in the printed material is
estimated to migrate
Paper and PET Migration experiments were GC-MS. Agilent J&W column: A total of 149 volatile compounds [55]
performed at 40 °C for 10 d to food 60 m × 0.25 mm id × 0.25 μm. were found in migration from on
simulant: Ethanol 50%, ethanol Flow 1 mL min-1. Injection mode paper, and 156 from PET materials,
®
95%, and Tenax ; and at 20 °C for split (1:20). Injection volume: some of them came from inks
Migration from Food Packaging Materials
2 d to isooctane food simulant 1 μL. Oven ramp: Initial 35 °C held

for 5 min, then rose at 10 °C min-1
up to 300 °C held for 1.5 min.
123
Acquisition was carried out in

SCAN mode (40–400 m/z)
(continued)
Table 1
124
(continued)

Paper and cardboard Food simulants: 50% and 95% UFPLC-qOrbitrap. Injection In general, migration of [56]
ethanol, migration experiments volume: 5 μL. SuperPhenylHexyl 10 photoinitiators, 4 phthalates, bis
were performed at 15, 30, or column (2.1 mm × 100 mm, (2- ethylhexyl)adipate,
60 min at room temperature or at 2.5 μm) with a prefilter (2.1 mm acetyltributyl citrate, caprolactam,
60 °C. Food simulants: Tenax, ID, 0.2 μm) from Thermo-fisher and BPA was observed to
migration tests were performed at scientific. Mobile phase: Milli-Q depended, main, on the material
150 °C and 250 °C during 15 min. water (solvent A) and ACN (solvent type and the physicochemical
Foodstuffs: Rice, cereals, and milk B) containing 0.1% HCOOH and parameters of the migrants.
powder. Migration tests were 5 mM NH4OAc in the positive and Whereas the temperatures showed
performed after contact for negative modes, respectively. The minor effects on migration, in the
6 months at room temperature LC gradient method was applied as case of liquid simulants. However,
follows: 0–4 min, 85–70% A; compounds from a baking paper to
4–8 min, 70–50% (hold 12 min); Tenax, at 150 and 250 °C,
20–30 min, 50–10% A. After 5 min evidenced an increment of
at 5% A, the gradient returned to migration when increasing
initial conditions for 5 min, with a temperature, except for the most
flow rate set to 0.3 mL min-1 at a volatile analytes. The
column temperature of 35 °C photoinitiators Michler’s ketone,
4,4′ -bis
(diethylamino)benzophenone and
ethyl-4-dimethylaminobenzoate
did not migrate to any of the food
sample tested. Whereas the
migrations of the photoinitiators:
Benzophenone, 2,2′ - dimethoxy-
2-phenylacetophenone,
4-phenylbenzophenone, and
2,4-diethyl-9H-thioxanthen-9-one
into milk powder were found over
their specific migration limit values
Recycled paper Food simulant: Tenax and Sorb-Star. HPLC-GC-FID. Phenomenex As a replacement for the migration of [23]
Migration experiments were Normal phase column mineral oil hydrocarbons from
performed: 20 °C, 40 °C, and 60 ° (250 mm × 2 mm ID). RESTEK recycled food packaging into food
C for 1–12 d. A longer migration pre-columns (10 m × 0:53 mm ID) products, 16 single substances
time (14 days at 20 °C and 40 °C) and RESTEK separation columns (n-alkanes and 15 aromatic
and short-time (1 and 22 h at 20 ° (15 m × 0.25 mm ID × 0.25 μm). compounds) were used like model
C) were also used Two parallel FIDs. Injection substances
volume: 50 μL Migration was a function of
temperature, time, contact type,
molecular weight, and polarity of
the substances. Alkylated aromatics
represent mineral oil aromatic
hydrocarbons more realistically.
Migration values of the touching
contact experiments were slightly
higher than those of the gas-phase
transfer for sorb-star simulant
Paperboard Food simulant: Modified GC-FID. VF-1 MS column Eight representing classes of [57]
polyphenylene oxide (30 m × 0.25 mm × 0.25 μm) with a component chemicals are known to
Food: Pasta, ground pasta, egg 2 m deactivated fused silica migrate from paperboard (1,3,5-
pasta, biscuit, butter waffle, wheat precolumn. Carrier gas: Trit-butylbenzene, 2,6-diisopropyl
semolina, rice semolina, dark Helium. Flow: 0.713 ml min-1. naphthalene, t-butyl anthracene,
chocolate, and milk chocolate. Injection mode: Split/split less n-hexadecane, n-heptadecane,
Migration experiments were inlet set at 250 °C and 1/40 split n-octadecane, n-eicosane, Di-n-
performed at 22 °C, 60–70% RH, ratio. Oven ramp: 150 °C hold for propyl phthalate). Volatility was
2, 4, 10, and 16 weeks of storage 1 min, ramp to 200 °C at 10 ° identified as the most important
C min-1 and hold for 1 min. characteristic of these migrants. Fat
Finally, ramp to 320 °C at 10 ° content and the conformation of
C min-1 and hold for 14 min starch had an influence on
migration. The highest migration
occurred in the case of paperboard
in contact with fatty foods (biscuits
and chocolate) followed by starchy
and particulate foods (egg-based

wheat pasta, wheat flour, and rice
flour). Low migration was observed
125
to wheat pasta
(continued)
Table 1
126
(continued)

Plastic, paper/board, Food simulant: For plastic—3% acetic GF-AAS and ICP-MS The values of heavy metals content [58]
and glass acid, for 10 d to 40 °C. For paper found into paper and board are
and cardboard—a cold aqueous higher than for plastic. This occurs
extract was obtained (24 h at room due the influence of printing inks
temperature) which was stabilized and dyes which are a source of
with HNO3 65%; for glass: 24 h at metals. On migration from food
room temperature, using as food contact materials, the highest values

simulant 4% acetic acid were obtained for Fe and Zn, while
the lowest values were obtained for
Co. however, the results showed
that the values migrated are lower
than maximum allowed limits for all
metals analyzed (Pb, Cd, Cr, Ba,
Cu, Co, Fe, Mn, Ni, Zn)
Metallic (in electrolytic Canned sardines (oil and tomato ICP OES equipment equipped with a Metallic chromium internal coating [25]
chromium/ sauce conserve) commercialized in double-step nebulization camera from bottom and top (closure) and
chromium oxide Brazil and a sea spray nebulizer. Power of chromium oxide in inner steel
coated steel the radiofrequency generator sheets from packaging showed
packaging) (1200 W). Sample flow rate quality satisfactory. The body of
(0.50 L min-1), argon flow from these cans received two coating
the nebulizer (0.60 L min-1). layers, guaranteeing the metallic
Auxiliary argon flow rate material resistance to production
(1.00 L min-1). Main argon flow process of this type of packaging
rate (12.0 L min-1). Axial mode of and minimized the development of
vision the corrosion in the cans. Levels
above the maximum limits allowed
by Brazilian and MERCOSUR
regulations were observed for
inorganic contaminants: As, Cd,
and Cr
Coated tinplate cans Vegetable foods: Including canned For analysis of bisphenol: UHPLC Among bisphenols, only BPA and [24]
fava beans, canned red beans, equipped with a multi-wave BADGE•2H2O were found in
canned chickpeas, canned okra of fluorescence detector. Auto- canned food, the highest level was
different brands. Storage time: sampler temperature: 10 °C; observed to BADGE•2H2O. Sn
0–730 d or purchase date +493 d. injected volume: 20 μL. C18-PFP was undetected in samples. Heat
storage temperature: 5 °C, 22 °C, column (150 × 2.1 mm ID, 2 μm treatment (main sterilization)
and 40 °C particle size). Column temperature: favors more bisphenol migration
20 °C. Mobile phase: Milli-Q water than trace metals migration.
(A) and HPLC plus gradient ACN However, metal release was
(B). Flow rate: 0.3 mL min-1 with influenced, main, by storage.
the following binary gradient: Canned okra offers conditions that
0 min—43% B, 1 min ramp to 50% favor highest BADGE and Fe
B (maintained for 4 min), 2 min content in food during
ramp to 60% B (maintained for pasteurization due its low pH (3.7)
5 min), 1 min ramp to 100% B and high water content (93.8%).
(maintained for 2 min), and back to Among trace metals, the highest
43% B in 1 min (total duration levels were observed for Fe and
16 min) Zn. BPA and Zn (and Pb to a lesser
For analysis of trace metals, AAS was extent) showed similar migration
used trends, with dependence on food
type (pH, water content), brand,
and storage temperature. Cd, Ni,
and Cu were similarly influenced by
food type and can brand.
Significant release of bisphenol and
metals were observed main in acid
food, probably due to corrosion
result in can damage to the can as
denting. The Fe release was
enhanced by 86–90%
high-performance liquid chromatography–gas chromatography with flame ionization detector (HPLC-GC-FID), gas chromatography coupled with quadrupoletime of flight
mass spectrometry (GC-qTOF-MS), high-performance liquid chromatography coupled to a quadrupole time-of-flight mass spectrometer (UHPLC-ESI-QTOF MS), sequential
windowed acquisition of all theoretical mass spectrometry (SWATH), poly(ethylene terephthalate) (PET), aluminum (Al), polyethylene (PE), ethylene vinyl acetate (EVA),
polypropylene (PP), polyamide (PA), ultrahigh-performance liquid chromatograph (UHPLC) coupled to a Thermo Scientific™ Q Exactive™ Focus quadrupole-Orbitrap mass
spectrometer (UHPLC-qOrbitrap), food contact material (FCM), inductively coupled plasma optical emission spectrometry (ICP OES), bisphenol A (BPA), bisphenol A
diglycidyl ether (BADGE•2H2O), Threshold of Toxicological Concern (TTC), Food and Drug Administration (FDA), European Food Safety Authority (EFSA), adipic acid (AA),
127
diethylene glycol (DEG), isophthalic acid (IPA), atomic absorption spectrometer with graphite furnace (GF-AAS), inductively coupled plasma mass spectrometer (ICP-MS),
atomic absorption spectrometry (AAS)
(as ethylparaben and methylparaben), phenolic antioxidants

(as butylated hydroxytoluene), and plasticizer
(as N-butylbenzenesulfonamide) are also compounds that might
be expected to be found in alcoholic beverages as a result of the
migration from crown cap plastic seals in contact with the
food [29].
In the study done by Wang et al. (2019) [36], the migration
behavior of chlorinated paraffins (CPs) presented in plastic (foil-
lined polypropylene) was related to their molecular weight and
chlorine content. In general, CPs were able to migrate into food
simulants (water, 3% acetic acid, 15% ethanol, and hexane) during
the migration experiment (performed at 40 °C/10 d), with the
highest migration concentration being found in hexane food simu-
lants. However, the study revelated that migration of CPs does not
pose immediate risks to human health. Another study with plastic
packaging, carried out by Garcı́a Ibarra et al. [31], showed that in a
total of 100 compounds detected in plastic materials, 27 were NIAS
migrating from FCM into food. The compounds found were: cap-
rolactam (which was present in an external layer of polyamide of
packaging material), 2,6-di-tert-butyl-1,4-benzoquinone, 2,4-di-
tert-butylphenol, phthalates (as diethyl phthalate, diisobutyl
phthalate and bis(2-ethylhexyl) phthalate), benzenesulfonamide,
n-ethyl-2-methyl, benzenesulfonamide, N-ethyl-4-methyl, alkanes
(tetradecane, hexadecane, heptadecane, octadecane, docosane),
isopropyl myristate, octadecanal, 7,9 di-tert-butyl-1-oxaspiro
(4,5)deca-6-9-diene-2,8-dione, tributyl acetyl citrate, slip agent
(as oleamide, hexadecanamide erucamide and octadecanamide),
plasticizers (as triphenyl phosphate and tributyl aconitate), squa-
lene, butylated hydroxytoluene (BHT), 1-hexadecanol, and bis
(2-ethylhexyl) adipate. However, some detected compounds are
not included in Regulation EU no. 10/2011 [19] on plastic
materials [31].
The migration of PIs from plastic baby bibs into Tenax® was
measured in bibs collected in the European market. Results indicate
that several no authorized PIs are in use to print bibs. The most
commonly detected PIs were benzophenone, detected in nearly
all samples, and isopropylthioxanthone, quantified in 12 out of
22 samples. Several non-evaluated PIs were detected: Triphenyl
phosphate, 2-ethylanthraquinone, 2,2-dimethoxy-2-phenylaceto-
phenone, 4-(4-methylphenyltio)benzophenone,
1-hydroxycyclohexyl phenyl ketone, and 4,4′-bis(diethylamino)-
benzophenone [37].
Xue et al. [38] reported that the migration percentage of
organic pollutants from Kraft paper-based packaging to packaged
dry powdered foods increased at a higher temperature and longer
contact time. Some organic substances introduced during the man-
ufacture of paper packaging materials such as phenol, alkylbenzene,
2,6-dissopropyl naphthalene and phthalates (dibutyl phthalate and
Bis(2-ethylhexyl)phthalate (DEHP)) used as food packaging may

migrate into dry food [38]. Migration of printing inks, which are
printed on carton surface, may also occur. Compounds found in
printing inks, as acrylate monomers and binders in UV offset inks,
may migrate into carton food packaging [39].
Contaminants coming from the adhesive ingredients com-
monly used in multilayer films were detected by migration studies
in baby food samples. Some concentration levels were found
exceeding the maximum residue levels for baby food as revealed
in the study of Bauer et al. [30], which raises the concern for
potential adverse effects of these substances on health. Ubeda
et al. [40] reported that polyurethane adhesives, commonly found
in multilayer materials, may contain cyclic ester oligomers as poten-
tial migrants. The authors showed that the concentration of these
compounds in migration to food simulants exceeded the maximum
level for not-listed substances established by Regulation EU
no. 10/2011 [19] for most samples.
Several metal traces, often used in metal containers, but also in
pigments and catalyzers for polymerization, are prone to migrate
into food, among them lead, iron, cadmium, nickel, chromium, tin,
zinc, and copper are being considered as food contaminants. Thus,
the use of metal cans or packaging with metals has to be carefully
addressed since it occasionally develops integrity problems due to
corrosion, which can lead to the migration of metal ions
[24]. According to Rather et al. (2017) [27], metal ions from
corrosion in metallic cans may migrate to food. By-products from
the epoxy resins (varnish commonly used to coat the inner side of
cans), such as bisphenol A (BPA), or bisphenol A diglycidyl ether
(BADGE), as cyclo-di-BADGE can also migrate to food.
Fengler and Gruber [23] studied the migration of mineral oil
hydrocarbons (mixtures of non-identified substances) from
recycled food packaging into food products and concluded that
migration was a function of contact type and polarity of the sub-
stances. Migration values of the direct contact experiments were
slightly higher than those of the gas-phase transfer for Sorb-Star®
simulant. Additionally, according to these authors, alkanes showed
high migration to the lipophilic food simulants due to their low
polarity.
Migration performance of contaminants from packaging mate-
rials into foods are affected by various factors, namely: the extended
contact with final packaging materials during storage, the tempera-
ture (heating), contact type, characteristics of the migrating sub-
stances/migrants (molecular weight, volatility, and polarity) and
food properties (composition—e.g., fat content—and properties),
which could certainly play a significant role in the migration process
[23, 38]. Therefore, it is important to evaluate how these factors
interfere in the contaminant’s migration in FCM, to minimize the
risk of migration and ensure food safety [41, 42].
Other recent research studies on the detection of contaminants

migrants from the food packaging material into food or food
simulants are summarized in Table 1.
2.2 Migration of The current experimental data shows that the research about the
Nanoparticles migration of ENMs is at young stages, especially for food
packaging [43].
Once the food is a complex matrix, the characterization and
detection of nanoparticles after their migration from the FCM are a
complex task and require several procedures or combined detection
methodologies, which limits the fundamental knowledge on
how/if the ENM interacts with the food components, changing
their properties. Thus the difficulty to predict whether the nano-
particles would pose a risk to the consumers when used in the
reinforcement of FCM [6, 44].
These studies should consider the potential effect of nanopar-
ticles’ migration from FCM to the packaged food. Therefore, this
can minimize the application of these nanocomposites as an alter-
native to conventional packaging materials without exploiting their
possible harmful effects on the consumers [6, 43].
The studies are not conclusive, reporting different patterns, for
example, Simon, Chaudhry, and Bakos [45] evaluated the migra-
tion of engineered nanoparticles from different polymer proposing
to address polymers like highly viscous liquids and to derive diffu-
sion coefficients by the Stokes–Einstein equation from viscosity and
particle radius, and they concluded that the reduction in particle
size and polymer dynamic viscosity leads to an increment in the
migration rate [22, 45, 46]. On the other hand, the research group
headed by Bott et al. [16] reported that when the nanoparticles are
immobilized in the polymer chains the migration does not occur, as
observed in low-density PE (LDPE) and polypropylene
(PS) reinforced with carbon nanotubes, which did not diffuse
toward the food simulant tested. They highlighted that the diffu-
sion will always be smaller than the detection limit of any current
sensitive method. This conclusion can be generalized to other FCM
in which black carbon nanotubes are completely embedded [16].
In addition to the complexity of this topic, another factor that
may also influence the migration of nanoparticles, for example, the
diffusion process of silver nanoparticles (AgNPs), known for their
antimicrobial properties [43, 47], is favorable in an acid environ-
ment (lower pH) [43, 47] and high temperature and time of
storage results [47]. Ultimately, the diffusion process also depends
on the type of polymer/nanoparticle and incorporation process
used, for example, in commercial LDPE packaging incorporated
with AgNPs in different formats presented distinct migration pro-
files [16, 43].
The research increasingly needs studies exclusively to under-
stand the migration behavior of nanocomposites when in contact
with foodstuffs. Moreover, it is also important to understand how
the migration process modifies and influences the structure of the

material, in terms of size and morphology, considering the com-
plexity of the mechanisms involved, due to the relevance of these
factors to assess the risks to human health by exposure to increas-
ingly common ENMs [6, 48].
What is expected is that under normal conditions of use, these
packaging cannot transfer their constituents to food in quantities
that could result in depletion of the food’s organoleptic character-
istics or changes in its composition, neither to pose dangerous
consequences for human health [43].
More information about this topic is available in several reviews
and technical papers such as in Störmer et al. [44], Bott et al. [16],
Huang et al. [17], Souza et al. [49], and Souza et al. [22], to cite
a few.
3 Notes
1. Migration of contaminants from contact materials to food can

occur and may represent a risk to the consumers’ health. To
evaluate if it happens and how it happens is mandatory, thus the
importance of migration assays. Ideally, the study of the
migrants should be done directly in the food after its contact
with the packaging. However, due to the complexity of food
matrices, in vitro studies are done using simulants. It is impor-
tant when planning this type of analysis to fully understand the
type of food and packaging to decide the best method to use.
Moreover, once the migration has occurred, the characteriza-
tion of the migrants should be done using the most sensitive
technique to obtain reliable results to make a proper risk assess-
ment and conclude whether the packaging poses or not any
danger to the consumer.
2. In the case of the European Union, the OML required for
plastic material is 10 mg of substances per 1 dm2 of the surface
area of the FCM, or 60 mg kg-1 of food [5], while SML
[mg kg-1] depends on the risk assessment of each specific
substance, and can also be found in the available regulation
[5, 19, 22]. Some of the substances of the positive list (i.e., list
of substances permitted in the manufacture of plastic materials
intended for food contact) of Reg. no. 10/2011 and its
amendments still do not have a defined SML.
3. Food simulants are the test media used to simulate/mimic the
transfer of substances from the packaging material into food;
thus, represent the major physical-chemical properties exhib-
ited by food [19, 50]. According to Regulation (EC) no.
10/2011 and its amendments [19], there are six assigned
food simulants, namely:
(i) Ethanol 10 vol%—Simulant A: assigned for hydrophilic

foods.
(ii) Acetic acid 3% (w/v)—Simulant B: assigned for hydro-
philic and acidic (pH < 4.5) foods.
(iii) Ethanol 20 vol%—Simulant C: assigned for hydrophilic
and mildly alcoholic (alcohol content ≤20%) foods and
food comprising a relevant amount of organic ingredients
that render it a more lipophilic character.
(iv) Ethanol 50 vol%—Simulant D1: assigned to lipophilic and
alcoholic (alcohol content >20%) foods and for oil-in-
water emulsions.
(v) Vegetable oil or ethanol 95 vol%—Simulant D2: assigned
to lipophilic foods containing free fat at the surface.
(vi) Poly(2,6-diphenyl-p-phenylene oxide), particle size
60–80 mesh, pore size 200 nm—Simulant E: assigned
for specific migration into dry food.
4. In general, migration assays are carried out for 10 d at 40 °C or
at 60 °C (which are the conditions to simulate any long-term
storage at room temperature or below, for products with shelf
lives longer than 6 months). More detailed information on the
choice of simulant and conditions of the assay can be found in
this Regulation, which also classifies the type of food and the
simulant(s) that are recommended for the migration tests.
5. To study the migration of nanoparticles, the protocols can be
even more demanding, once it is important to determine if
after the diffusion process the migrant kept its nanometer
scale [51]. Thus, even more robust techniques are generally
required, such as microscopy techniques (scanning and trans-
mission electron microscopies), dynamic light scattering, X-ray
diffraction, and field-flow fractionation, to mention a few [22].
6. It is important to highlight that the analytical methods used
need to be in accordance with the legal requirements, for
example, for the European countries/market with those set in
Article 11 of Regulation (EC) no. 882/2004 [52], repealed by
Regulation (EU) 2017/625 of the European Parliament and
of the Council of March 15, 2017 [53], which defines the
proper methods of analysis and sampling. Also, the technique
quantification limit needs to be taken into consideration, as
more robust characterization methods should be applied
according to the level of migrants found.
7. Depending on the migrant targeted, after the diffusion pro-
cess, the samples may also need some preparation prior to the
quantification. For example, to quantify the migration of inor-
ganic compounds, such as minerals, all organic matter needs to
be removed from the simulant or the food. This process can be
accomplished by incinerating the sample followed by acid
digestion with nitric acid solution (1:1).
Acknowledgments
This research was funded by national funding by FCT, Foundation

for Science and Technology, through the individual research grant
(SFRH/BD/144346/2019) of J.P and the individual research
grant (2020.04441.BD) of C.R. This work was supported by
national funds from FCT/MCTES within the R&D Units Project
Scope (UIDB/EMS/00285/2020, UIDB/00211/2020,
and UIDB/50016/2020). This work has also been supported by
MEtRICs—Mechanical Engineering and Resource Sustainability
Center, which is financed by national funds from FCT/MCTES
(UIDB/04077/2020 and UIDP/04077/2020). This work has
also been supported by the Associate Laboratory for Green
Chemistry—LAQV which is financed by national funds from
FCT/MCTES (UIDB/50006/2020 and UIDP/50006/2020).
This work has also been financially supported by PRR – Recovery
and Resilience Plan and the Next Generation EU funds, INOV.AM
project “Innovation in Additive Manufactoring” (C628518748-
00464029 j 7999), and POCI-01-0145-FEDER-030446 funded
by FEDER funds through COMPETE2020—Programa Operacio-
nal Competitividade e Internacionalização (POCI) and by national
funds from FCT/MCTES (UIDB/00285/2020 – CEMMPRE ref,
LA/P/0112/2020 ARISE ref).
References
1. Prasad P, Kochhar A (2014) Active packaging potential migration to food – a review. Food
in food industry: a review. IOSR J Environ Sci Packag Shelf Life 8:63–70. https://doi.org/
Toxicol Food Technol 8:01–07. https://doi. 10.1016/j.fpsl.2016.04.001
org/10.9790/2402-08530107 7. Poças F, Franz R (2018) Overview on
2. Tang XZ, Kumar P, Alavi S, Sandeep KP European regulatory issues, legislation, and
(2012) Recent advances in biopolymers and EFSA evaluations of nanomaterials. In: Nano-
biopolymer-based nanocomposites for food materials for food packaging. Elsevier, pp
packaging materials. Crit Rev Food Sci Nutr 277–300
52:426–442. https://doi.org/10.1080/ 8. European Union (EU) (2016) Food contact
10408398.2010.500508 materials – regulation (EC) 1935/2004 –
3. Poças F, Hogg T (2007) Exposure assessment European implementation assessment. Off J
of chemicals from packaging materials in foods: Eur Union 140. https://doi.org/10.2861/
a review. Trends Food Sci Technol 18:219– 162638
230. https://doi.org/10.1016/j.tifs.2006. 9. Dainelli D, Gontard N, Spyropoulos D,
12.008 Zondervan-van den Beuken E, Tobback P
4. Poças F (2018) Migration from packaging and (2008) Active and intelligent food packaging:
food contact materials into foods. In: Refer- legal aspects and safety concerns. Trends Food
ence module in food science. Elsevier, pp 1–18 Sci Technol 19:S103–S112. https://doi.org/
5. European Commission (2004) Regulation 10.1016/j.tifs.2008.09.011
(EU) No 1935/2004 of 27 October 2004 on 10. Wani AA, Singh P, Langowski HC (2014)
plastic materials and articles intended to come Food technologies: packaging. Encycl Food
into contact with food and repealing Directives Saf 3:211–218. https://doi.org/10.1016/
80/590/EEC and 89/109/EEC. J da União B978-0-12-378612-8.00273-0
Eur L338:4–17 11. Handford CE, Dean M, Henchion M,
6. Souza VGL, Fernando AL (2016) Nanoparti- Spence M, Elliott CT, Campbell K (2014)
cles in food packaging: biodegradability and Implications of nanotechnology for the agri-
food industry: opportunities, benefits and from composites into food matrices. In:
risks. Trends Food Sci Technol 40:226–241. Cirillo G, Kozlowski MA, Spizzirri UG (eds)
https://doi.org/10.1016/j.tifs.2014.09.007 Composite materials for food packaging, 1st
12. Chaudhry Q, Scotter M, Blackburn J, Ross B, edn. Scrivener Publishing LLC, Beverly, pp
Boxall A, Castle L, Aitken R, Watkins R (2008) 401–435
Applications and implications of nanotechnol- 23. Fengler R, Gruber L (2020) Mineral oil migra-
ogies for the food sector. Food Addit Contam tion from paper-based packaging into food,
Part A Chem Anal Control Expo Risk Assess investigated by means of food simulants and
25:241–258. https://doi.org/10.1080/ model substances. Food Addit Contam Part A
02652030701744538 Chem Anal Control Expo Risk Assess 37:845–
13. Avella M, De Vlieger JJ, Errico ME, Fischer S, 857. https://doi.org/10.1080/19440049.
Vacca P, Volpe MG (2005) Biodegradable 2020.1714750
starch/clay nanocomposite films for food pack- 24. Noureddine El Moussawi S, Ouaini R, Matta J,
aging applications. Food Chem 93:467–474. Chébib H, Cladière M, Camel V (2019) Simul-
https://doi.org/10.1016/j.foodchem.2004. taneous migration of bisphenol compounds
10.024 and trace metals in canned vegetable food.
14. Sarantópoulos CGL, Oliveira LM, Padula M, Food Chem 288:228–238. https://doi.org/
Coltro L, Aalves RMV, Garcia EC (2002) 10.1016/j.foodchem.2019.02.116
Embalagens plásticas flexı́veis: principais polı́- 25. de Mello Lazarini TE, Milani RF, Yamashita
meros e avaliação de propriedades, Campinas DM, Saron ES, Morgano MA (2019) Canned
15. Schmid P, Welle F (2020) Chemical migration sardines commercialized in Brazil: packaging
from beverage packaging materials – a review. and inorganic contaminants evaluation. Food
Beverages 6:1–19. https://doi.org/10.3390/ Packag Shelf Life 21:100372. https://doi.org/
beverages6020037 10.1016/j.fpsl.2019.100372
16. Bott J, Störmer A, Franz R (2014) Migration 26. Luykx DMAM, Peters RJB, Van Ruth SM,
of nanoparticles from plastic packaging materi- Bouwmeester H (2008) A review of analytical
als containing carbon black into foodstuffs. methods for the identification and characteri-
Food Addit Contam Part A 31:1769–1782. zation of nano delivery systems in food. J Agric
https://doi.org/10.1080/19440049.2014. Food Chem 56:8231–8247. https://doi.org/
952786 10.1021/jf8013926
17. Huang J-YY, Li X, Zhou W (2015) Safety 27. Rather IA, Koh WY, Paek WK, Lim J (2017)
assessment of nanocomposite for food packag- The sources of chemical contaminants in food
ing application. Trends Food Sci Technol 45: and their health implications. Front Pharmacol
187–199. https://doi.org/10.1016/j.tifs. 8. https://doi.org/10.3389/fphar.2017.
2015.07.002 00830
18. Chungy D, Papadakis SE, Yamy KL (2002) 28. Commission of the European Communities
Simple models for assessing migration from (2006) Commission Regulation (EC) No
food-packaging ® lms. Food Addit Contam 118/2006 of 19 December 2006 setting max-
19:611–617. https://doi.org/10.1080/ imum levels for certain contaminants in food-
0265203021012638 stuffs. Off J Eur Union:5–24
19. European Commission (2011) Commission 29. Žnideršič L, Mlakar A, Prosen H (2019) Devel-
Regulation (EU) No 10/2011. Off J Eur opment of a SPME-GC-MS/MS method for
Union, pp 1–89 the determination of some contaminants from
20. Stoll L, Rech R, Flôres SH, Nachtigall SMB, de food contact material in beverages. Food Chem
Oliveira RA (2019) Poly(acid lactic) films with Toxicol 134:110829. https://doi.org/10.
carotenoids extracts: release study and effect on 1016/j.fct.2019.110829
sunflower oil preservation. Food Chem 281: 30. Bauer A, Jesús F, Gómez Ramos MJ,
213–221. https://doi.org/10.1016/j. Lozano A, Fernández-Alba AR (2019) Identi-
foodchem.2018.12.100 fication of unexpected chemical contaminants
21. Poças MF, Oliveira JC, Oliveira FAR, Hogg T in baby food coming from plastic packaging
(2008) A critical survey of predictive mathe- migration by high resolution accurate mass
matical models for migration from packaging. spectrometry. Food Chem 295:274–288.
Crit Rev Food Sci Nutr 48:913–928. https:// https://doi.org/10.1016/j.foodchem.2019.
doi.org/10.1080/10408390701761944 05.105
22. Souza VGL, Ribeiro-Santos R, Rodrigues PF, 31. Garcı́a Ibarra V, Bernaldo R, de Quirós A,
Otoni CG, Duarte MP, Coelhoso IM, Fer- Paseiro Losada P, Sendón R (2019)
nando AL (2018) Nanomaterial migration Non-target analysis of intentionally and non
intentionally added substances from plastic
packaging materials and their migration into 41. Silva AS, Freire JMC, Garcı́a RS, Franz R,
food simulants. Food Packag Shelf Life 21: Losada PP (2007) Time–temperature study of
100325. https://doi.org/10.1016/j.fpsl. the kinetics of migration of DPBD from plas-
2019.100325 tics into chocolate, chocolate spread and mar-
32. Sanches Silva A, Cruz Freire JM, Sendón R, garine. Food Res Int 40:679–686. https://doi.
Franz R, Losada PP (2009) Migration and dif- org/10.1016/j.foodres.2006.11.012
fusion of diphenylbutadiene from packages 42. Silva AS, Freire JMC, Franz R, Losada PP
into foods. J Agric Food Chem 57:10225– (2008) Mass transport studies of model
10230. https://doi.org/10.1021/jf901666h migrants within dry foodstuffs. J Cereal Sci
33. Sanches-Silva A, Pastorelli S, Cruz JM, 48:662–669. https://doi.org/10.1016/j.jcs.
Simoneau C, Paseiro-Losada P (2008) Devel- 2008.02.006
opment of a multimethod for the determina- 43. Metak AM, Nabhani F, Connolly SN (2015)
tion of photoinitiators in beverage packaging. J Migration of engineered nanoparticles from
Food Sci 73:C92–C99. https://doi.org/10. packaging into food products. LWT Food Sci
1111/j.1750-3841.2007.00642.x Technol 64:781–787. https://doi.org/10.
34. Vilarinho F, Sendón R, van der Kellen A, Vaz 1016/j.lwt.2015.06.001
MF, Silva AS (2019) Bisphenol A in food as a 44. Störmer A, Bott J, Kemmer D, Franz R (2017)
result of its migration from food packaging. Critical review of the migration potential of
Trends Food Sci Technol 91:33–65. https:// nanoparticles in food contact plastics. Trends
doi.org/10.1016/j.tifs.2019.06.012 Food Sci Technol 63:39–50. https://doi.org/
35. He NX, Bayen S (2020) An overview of chem- 10.1016/j.tifs.2017.01.011
ical contaminants and other undesirable chemi- 45. Šimon P, Chaudhry Q, Bakoš D (2008) Migra-
cals in alcoholic beverages and strategies for tion of engineered nanoparticles from polymer
analysis. Compr Rev Food Sci Food Saf 19: packaging to food – a physicochemical view. J
3916–3950. https://doi.org/10.1111/ Food Nutr Res 47:105–113
1541-4337.12649 46. Souza VGL, Rodrigues C, Valente S,
36. Wang C, Gao W, Liang Y, Jiang Y, Wang Y, Pimenta C, Pires JRA, Alves MM, Santos CF,
Zhang Q, Jiang G (2019) Migration of chlori- Coelhoso IM, Fernando ALL (2020)
nated paraffins from plastic food packaging into Eco-friendly ZnO/chitosan bionanocompo-
food simulants: concentrations and differences sites films for packaging of fresh poultry meat.
in congener profiles. Chemosphere 225:557– Coatings 10:110. https://doi.org/10.3390/
5 6 4 . h t t p s : // d o i . o r g / 1 0 . 1 0 1 6 / j . coatings10020110
chemosphere.2019.03.039 47. Hetzer B, Burcza A, Gr€a f V, Walz E, Greiner R
37. Galbiati E, Pereira J, Selbourne M, do Poças (2017) Online-coupling of AF4 and single
CF (2020) Photoinitiators use in printed baby particle-ICP-MS as an analytical approach for
bibs and their migration into Tenax® by gas the selective detection of nanosilver release
chromatography–mass spectrometry. Packag from model food packaging films into food
Technol Sci pts.2546:203. https://doi.org/ simulants. Food Control 80:113–124.
10.1002/pts.2546 https://doi.org/10.1016/j.foodcont.2017.
38. Xue M, Chai X-S, Li X, Chen R (2019) Migra- 04.040
tion of organic contaminants into dry pow- 48. Hannon JC, Kerry JP, Cruz-Romero M, Azlin-
dered food in paper packaging materials and Hasim S, Morris M, Cummins E (2017)
the influencing factors. J Food Eng 262:75– Kinetic desorption models for the release of
82. https://doi.org/10.1016/j.jfoodeng. nanosilver from an experimental nanosilver
2019.05.018 coating on polystyrene food packaging. Innov
39. Ateş Duru Ö, Özdemir L, Oktav M (2019) Food Sci Emerg Technol 44:149–158.
Occurrence and effects of migration in carton https://doi.org/10.1016/j.ifset.2017.07.001
food packaging caused by printing inks. In: 3rd 49. Souza VGL, Rodrigues PF, Duarte MP, Fer-
international printing technologies sympo- nando AL (2018) Antioxidant migration stud-
sium, Istanbul, pp 139–144 ies in chitosan films incorporated with plant
40. Ubeda S, Aznar M, Rosenmai AK, Vinggaard extracts. J Renew Mater 6:548–558. https://
AM, Nerı́n C (2020) Migration studies and doi.org/10.7569/JRM.2018.634104
toxicity evaluation of cyclic polyesters oligo- 50. Souza VGL, Rodrigues C, Ferreira L, Pires
mers from food packaging adhesives. Food JRA, Duarte MP, Coelhoso I, Fernando AL
Chem 311:125918. https://doi.org/10. (2019) In vitro bioactivity of novel chitosan
1016/j.foodchem.2019.125918 bionanocomposites incorporated with differ-
ent essential oils. Ind Crop Prod 140:111563.
https://doi.org/10.1016/j.indcrop.2019. multilayer food packaging materials by

111563 GC-MS analysis and pattern recognition with
51. Blasco C, Picó Y (2011) Determining nanoma- chemometrics. Food Addit Contam
terials in food. TrAC Trends Anal Chem 30: Part A:1–12. https://doi.org/10.1080/
84–99. https://doi.org/10.1016/j.trac.2010. 19440049.2016.1155757
08.010 56. Blanco-Zubiaguirre L, Zabaleta I, Prieto A,
52. European Commission (2004) Regulation Olivares M, Zuloaga O, Elizalde MPP (2020)
(EC) No 882/2004 on official controls per- Migration of photoinitiators, phthalates and
formed to ensure the verification of compliance plasticizers from paper and cardboard materials
with feed and food law, animal health and ani- into different simulants and foodstuffs. Food
mal welfare rules. Off J Eur Union 2004:1–141 Chem 344:128597. https://doi.org/10.
53. European Commission (2017) Regulation 1016/j.foodchem.2020.128597
(EU) 2017/625 on official controls and other 57. Baele M, Vermeulen A, Claes M, Ragaert P, De
official activities performed to ensure the appli- Meulenaer B (2020) Migration of surrogate
cation of food and feed law, rules on animal contaminants from paperboard to foods: effect
health and welfare, plant health and plant proof food and surrogate properties. Food Addit
tection products. Off J Eur Union 95:1–142 Contam Part A 37:2165–2183. https://doi.
54. Peng X, Fu H, Hu J, Luo F (2020) Investiga- org/10.1080/19440049.2020.1778184
tion on mercury migration discipline in differ- 58. Ungureanu E, Mustăţea G, Popa ME (2020)
ent paper-plastic food packaging containers. J Heavy metals contamination of food contact
Food Sci 85:1186–1192. https://doi.org/10. materials in Romania. Sci Bull Ser F
1111/1750-3841.15060 Biotechnol XXIV:63–68
55. Clemente I, Aznar M, Nerı́n C, Bosetti O
(2016) Migration from printing inks in
Chapter 7
In Vitro Cytotoxicity Testing of Food Packaging

Arthur B. Ribeiro, Juliana G. F. Silva, Lucas N. F. Trevizan,
Hernane S. Barud, Flávia A. Resende, and Denise C. Tavares
Abstract
Colorimetric assays with tetrazolium salts allow rapid evaluation of cytotoxicity endpoints. These assays are
based on the ability of viable cells to convert tetrazolium salts into formazan products through the succinate
dehydrogenase system in the mitochondrial respiratory chain. In the presence of NADH/NADPH, these
salts are reduced to formazan products characterized by an intense and distinct color that depends on the
original tetrazolium salt used as the substrate for the reaction. Only viable cells, which contain intact plasma
and mitochondrial membranes, will have active dehydrogenase. Agents that break the membranes and
interfere with the respiratory chain will deactivate the enzyme and consequently the formation of formazan
products. Thus, the amount of formazan product can be correlated with the number of viable cells after
exposure to the tested substance. In this chapter, the most common colorimetric cell viability assays with
tetrazolium salts are present to assess the cytotoxicity of food packaging.
Key words Colorimetric assay, XTT, MTS, MTT, WST, Formazan, Respiratory chain, Dehydro-
genase system
1 Introduction
In vitro cytotoxicity assays are becoming increasingly recognized as

an extremely valuable tool to identify compounds that might pose
certain health risks to humans [1, 2]. The need for sensitive, reli-
able, and easy methods has led to the development of several
standard assays that allow rapid evaluation of chemically induced
damages in physiologically normal cells that lead to death or cell
proliferation disturbances [3–5]. Among the most basic types of
in vitro bioassays, colorimetric assays using tetrazolium salts have
been widely used to measure cellular metabolic activity as an indi-
cator of cell viability, proliferation, and cytotoxicity [3, 6, 7].
The most commonly used tetrazolium salts can be classified
into two basic categories: cationic and anionic salts. Cationic salts
(Fig. 1a) are positively charged and easily penetrate viable eukary-
otic cells through electrostatic interactions with the anionic plasma
https://doi.org/10.1007/978-1-0716-3613-8_7,
137
138 Arthur B. Ribeiro et al.
Fig. 1 Schematic model of cellular reduction of (a) cationic salts (MTT) and (b) anionic salts (XTT) in viable cells
membrane, serving as substrates for active cellular dehydrogenases

and reductase enzymes to be reduced to formazan products, as for
example the MTT ((3-(4,5-dimethylthiazol-2-yl)-2,5-diphenylte-
trazolium bromide)) tetrazolium salt, which gives name to the
MTT colorimetric assay [2, 8].
The MTT colorimetric assay has been widely used to measure
cytotoxicity or cytostatic activity of compounds. This methodology
is based on the ability of viable cells to convert MTT into purple
insoluble formazan crystals [8, 9]. Due to its lipophilic side groups
and positive charge, the MTT salt can quickly penetrate viable cell
membranes and be reduced by mitochondrial or cell plasma
enzymes like oxidoreductases, dehydrogenases, oxidases, and per-
oxidases [2, 10]. The MTT formazan product then accumulates as
insoluble needle-shaped crystals that precipitate inside cells and can
also be deposited near the cell surface [2, 11]. The produced
formazan is finally quantified by absorbance measurements using
a spectrophotometer. For this, an organic solvent such as dimethyl
sulfoxide (DMSO), acidified isopropanol, or sodium dodecyl sul-
fate (SDS) are required to solubilize the crystals [12–14]. Viable
cells, with an active metabolism, convert MTT into a purple-
colored formazan product with a maximum wavelength near
570 nm [7, 9, 15].
Tests using anionic tetrazolium salts were developed later. The
anionic salts (Fig. 1b) are negatively charged and do not easily
penetrate cell membranes. These require therefore an electron
coupling reagent capable of penetrating viable cells and transferring
electrons to the tetrazolium salt at the cell surface or at the plasma
membrane level, converting the salt into a formazan product.
Cytotoxicity of Food Packaging 139
Fig. 2 Schematic model of anionic tetrazolium salts cleavage: (a) MST and (b) WST-1—to formazan. (ECR
electron coupling reagent)
Anionic salts include MTS (3-(4,5-dimethylthiazol-2-yl)-

5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium)
(Fig. 2a), XTT (2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-
tetrazolium-5-carboxanilide), and WST-1 4-[3-(4-iodophenyl)-
2-(4-nitrophenyl)-2H-5-tetrazolio]-1,3-benzene disulfonate
(Fig. 2b) [2, 8, 15].
MTS, XTT, or WST tetrazolium salts are widely used in cell
viability and proliferation tests [2, 8, 15]. During the assays, the
tetrazolium salt (MTS, XTT or WST) is reduced to a highly water-
soluble formazan dye in cell culture medium through the succinate
dehydrogenase system of the mitochondrial respiratory chain in
metabolically active viable cells (Fig. 2) [10]. The reaction requires
the presence of an electron coupling reagent, usually phenazine
methosulfate (PMS) or phenazine ethyl sulfate (PES), capable of
penetrating viable cells [7, 16]. The amount of water-soluble prod-
uct generated from anionic salts reduction can be quantified mea-
suring the absorbance at a wavelength of 420–570 nm using a
spectrophotometer [7, 15]. The amount of formazan dye formed
can be correlated with the number of viable cells after exposure to
the tested substance.
In this chapter, the most common colorimetric cell viability

assays with tetrazolium salts are presented. Methodologies are
available to assess the cytotoxicity of food contact materials
(of which potential cytotoxic components may be diffused into
food) or even edible packaging.
2 Materials
2.1 General • Laminar flow clean bench/cabinet

Equipment • 96-Well microtiter plate (flat-bottom)
• Water bath
• Inverse-phase contrast microscope
• 96-Well plate spectrophotometer (microplate reader) equipped
with 420–650 nm filter
• Single channel pipette (10–100 μL)
• Multi-channel pipette (20–200 μL)
• Sterile pipette tips (10–200 μL)
• Filters/filtration devices
• Hemocytometer or cell counter
• CO2 incubator
• Vortex mixer
2.2 Chemicals • Phosphate-buffered saline (PBS) without Ca2+ and Mg2+

• Fetal bovine serum (FBS)
• Ham’s F-10 Nutrient Mixture (HAM-F10) with l-glutamine
(preferably) and without phenol red
• Hydrochloric acid (HCl)
• Sodium dodecyl sulfate (SDS)
2.3 Assay Kits Commercial kits of tetrazolium salt colorimetric assays (MTT,
XTT, MTS, and WST-1) are available from several companies (see
Notes 1 to 3).
2.3.1 MTT Assay • Cell Proliferation Kit I (MTT)—Roche—Cat. No. 11 465

007 001
• MTT Assay Kit (Cell Proliferation)—AbCam—Cat.
No. ab211091
• CyQUANT™ MTT Cell Viability Assay—Invitrogen—Cat.
No. V13154
2.3.2 XTT Assay • Cell Proliferation Kit II (XTT)—Roche—Cat. No. 11 465

015 001
• XTT Assay Kit—AbCam—Cat. No. ab232856
• CyQUANT™ XTT Cell Viability Assay—Invitrogen—Cat.
No. X12223
2.3.3 MTS Assay • MTS Assay Kit (Cell Proliferation)—AbCam—Cat.

No. ab197010
• CellTiter 96® AQueous One Solution Cell Proliferation Assay
(MTS)—Promega—Cat. No. G3580
2.3.4 WST-1 Assay • Cell Proliferation Reagent WST-1—Roche—Cat. No. 11 644

807 001
• WST-1 Cell Proliferation Assay Kit—Abnova—Cat.
No. KA1384
• Ready-to-use Cell Proliferation Colorimetric Reagent
(WST-1)—MyBioSource—Cat. No. MBS841449
3 Methods
3.1 Extraction According to the “International Standard ISO1 10993-1; 5 and

12 (2020)” [17–19], the conditions for the extraction of sub-
stances present in food packaging systems must simulate those
found in clinical use, to determine the real toxicological potential
of the material used. Therefore, the vehicles and extraction condi-
tions used must be appropriate to the nature and use of the final
product and the purpose of the test.
First, to ensure that the extraction containers do not adulterate
the extract from the test sample, the extraction must be carried out
in clean, chemically inert, closed containers using aseptic techni-
ques. Then, the vehicle of choice should reflect the extraction
purpose. The use of polar and non-polar vehicles should be consid-
ered to extract all substances present in the packaging. For assays
with mammalian cells, the preferred extraction vehicle is the culture
medium supplemented with 5–10% FBS due to its ability to sup-
port cell growth and extract substances of both polarities. In addi-
tion to the serum culture medium, using serum-free medium
should be considered to specifically extract polar substances.
An extraction at 37 ± 1 °C for 24 ± 2 h in tissue culture is
acceptable for the cytotoxicity test, since extraction temperatures
higher than 37 ± 1 °C can adversely impact the chemical character-
istics and/or stability of serum and other constituents in the culture
medium. Cultured primary human or rodent cell lines such as
CHO, V79, CHL/IU, and L5178Y cells or human cell lines such
as TK6 can be used, according to the recommendations for in vitro
toxicity tests [20]. For materials that are inherently cytotoxic,
further testing using various dilutions of the tested solution may
be necessary to determine the level at which cytotoxicity no longer
occurs.
3.2 Preparing the • Cell Proliferation Kit I (MTT)—Roche—Cat. No. 11 465

Working Solution of 007 001:
Commercial Kits of I. MTT Labeling Reagent (Non-sterile, Ready-to-Use)
Tetrazolium Salt
Colorimetric Assays – 5 × 5 mL MTT at 5 mg mL-1 in PBS
(See Notes 4 and 5) II. Solubilization Solution (1x—Ready-to-Use)
– 3 × 90 mL 10% SDS in 0.01 M HCl
Filter the solution through a 0.20-μm pore size membrane to
sterilize the MTT solution and to remove all solid particles like
nonspecifically formed formazan crystals. Store the solution in
small aliquots protected from light at -20 °C.
• Cell Proliferation Kit II (XTT)—Roche—Cat. No. 11 465
015 001:
I. XTT Labeling Reagent
– 5 × 25 mL XTT at 1 mg mL-1 in RPMI 1640 medium
II. Electron-Coupling Reagent
– 5 × 0.5 mL PMS (N-methyl dibenzopyrazine methyl
sulfate)
Thaw the XTT labeling reagent and electron-coupling reagent
in a water bath at 37 °C. Mix each vial thoroughly to obtain a clear
solution. For one microplate (96 wells), mix 5 mL XTT labeling
reagent with 0.1 mL electron coupling reagent. Store the solution
in small aliquots protected from light at -20 °C.
• MTS Assay Kit (Cell Proliferation)—AbCam—Cat.
No. ab197010
Thaw the solution and briefly centrifuge small vials at low speed
prior to opening. All kit components are supplied as ready to be
used. It is recommended to add 10 μL/well of the MTS Assay Kit
to the cells already cultured in 100 μL/well (1:10 final dilution).
Keep in ice while using. Store the solution in small aliquots pro-
tected from light at -20 °C (see Note 5).
• Cell Proliferation Reagent WST-1—Roche Cat. No. 11 644
807 001
Thaw the solution by heating at 37 °C for 2–10 min and vortex
to dissolve the precipitates. The WST-1 reagent can be used
without any limitations after thawing. It is recommended to add

10 μL/well of the Cell Proliferation Reagent WST-1 to the cells
already cultured in 100 μL/well (1:10 final dilution). Store at 2–8 °
C, protected from light, for up to 4 weeks, or in aliquots at -20 °C
for longer periods.
3.3 Cytotoxicity 1. Seed cells at a density of 1 × 104 cells/well in a 96-well flat-

Assay bottom microtiter plate and allow cells to adhere for 24 h at
37 °C in a CO2 incubator (see Note 6).
2. The culture medium must be replaced with fresh medium after
24 h of incubation.
3. Add the food packaging extract to the cells and incubate for a
desired period (24, 48, or 96 h) at 37 °C in a CO2 incubator.
The suggested total volume is 100 μL for a 96-well plate (see
Note 7).
4. Include, in each plate, (i) negative control wells (3–6 wells that
contain the same number of cells as the experimental wells and
that are not exposed to the tested substance), (ii) blank control
wells (3–6 wells are needed to measure the blanks, which only
contain the culture medium), and (iii) positive control wells
(3–6 wells that contain the same number of cells as the experi-
mental wells and that are exposed to known cytotoxic
substances).
5. After the treatment period, discard the culture medium and
wash cells with PBS, add 100 μL of HAM-F10 culture medium
(see Note 8).
6. Cell cytotoxicity measurements using the different test:
A. Using the Cell Proliferation Kit I (MTT)—Roche—Cat.
No. 11 465 007 001:
7A. Subsequently, add 10 μL of the MTT Labeling Reagent
(0.5 mg mL-1) per well and incubate for 4 h at 37 °C in a
CO2 incubator (see Note 9).
8A. Following, add 100 μL of the solubilization solution into
each well. Allow the plate to stand overnight at 37 °C in a
CO2 incubator.
9A. Check for complete solubilization of the purple formazan
crystals (see Note 10).
10A. Measure the spectrophotometric absorbance of the sam-
ples using a microplate reader. The wavelength to measure
the absorbance of the formazan product is between
550 and 570 nm according to the filters available for the
microplate reader used. The reference wavelength should
be higher than 650 nm (see Note 11).
B. Using the Cell Proliferation Kit II (XTT)—Roche—Cat.

No. 11 465 015 001:
7B. Subsequently, add 50 μL of the XTT Labeling Mixture
(0.3 mg mL-1) per well and incubate for 4–17 h at 37 °C
in a CO2 incubator (see Note 9).
8B. Measure the spectrophotometric absorbance of the sam-
ples using a microplate () reader. The wavelength to mea-
sure the absorbance of the formazan product is between
microplate reader used. The reference wavelength should
be higher than 650 nm (see Notes 10 and 11).
C. Using the MTS Assay Kit (Cell Proliferation)—AbCam—Cat.
No. ab197010:
7C. Subsequently, add 10 μL of the MTS Labeling Reagent per
well and incubate for 0.5–4 h at 37 °C in a CO2 incubator
(see Note 9).
8C. Shake the plate briefly on a shaker.
9C. Measure the spectrophotometric absorbance of the sam-
microplate reader used (see Note 10).
D. Using the Cell Proliferation Reagent WST-1—Roche Cat.
No. 11 644 807 001:
7D. Subsequently, add 10 μL of the Cell Proliferation Reagent
WST-1 per well and incubate for 4 h at 37 °C in a CO2
incubator (see Note 11).
8D. Shake thoroughly for 1 min on a shaker.
9D. Measure the spectrophotometric absorbance of the sam-
microplate reader. The reference wavelength should be
higher than 600 nm (see Notes 10 and 11).
3.4 Cell Viability A. Subtract the culture medium background (blank group—See
Calculation Note 12) from all samples reading, as described in the formula:
Real Absorbance ðR:AbsÞ = Absorbance Sample

- Absorbance Blank Group
Fig. 3 The fictitious graph represents the cell viability profile of X-cells treated
with different concentrations of different food packaging extracts (FPE)
(10–320 μg mL-1) in 24 h of incubation
B. The cell viability percentage is calculated using the following

equation (See Note 13):
R:Abs Sample
Cell viability ð%Þ = × 100
R:Abs Negative Control
C. Plot the cell viability percentages against the tested food pack-
aging extract (FPE) concentration. The 50% inhibiting con-
centration (IC50) can be determined graphically, as in the
example below (Fig. 3):
4 Notes
1. All available tetrazolium salt kits are only for research; not for
human or veterinary diagnostic or therapeutic use.
2. The reagents have good stability when stored at -15 to -25 °
C, protected from light, until the expiration date printed on the
label.
3. Before use, it is necessary to review the complete Safety Data
Sheet, which is available directly on the website of the manu-
facturer or upon request.
4. To obtain reliable results, thaw the working solutions immedi-
ately before use.
5. The kits described in this topic were used to exemplify the
procedures to be performed, however, without the intention
of recommending.
6. Optimal sensitivity of a colorimetric assay is obtained with near-

confluent cells at the time of the assay. In general, cells seeded
at densities between 5 × 103 and 1 × 104 cells per well should
reach optimal population densities. However, since there are
cell types that have low metabolic activity, it is recommended to
increase the concentration of cells to obtain formazan color
development within the time suggested by the employed kit.
7. Colored compounds can have their own optical density
(OD) that can influence the measurement when diluted.
Thus, it is recommended to wash the cells with PBS before
adding the tetrazolium salts to remove any trace of the tested
substance that could give inaccurate results.
8. The culture conditions used to grow the cells can affect the
results and must be taken into consideration when analyzing
the data. The presence of serum or phenol red in the culture
medium can generate background and seriously affect the sam-
ple absorption reading. It is recommended, if possible, to
culture cells in phenol red or serum-free medium during the
incubation period with the tetrazolium salt.
9. Incubation time with the tetrazolium salt varies according to
the type and concentration of the cells. Incubation time can be
optimized by an initial test reading the absorbance at various
times (i.e., 4, 6, 8, and 12 h) after the addition of the tetrazo-
lium detection solution, using the same plate.
10. The plate should be gently shaken (Shake mode: Slow/
Shake time: 2 s) to evenly distribute the dye in the wells prior
to reading the absorbance with a spectrophotometer.
11. The tetrazolium-specific absorbance can be measured between
420 and 570 nm, depending on which kit is used. The refer-
ence wavelength absorbance reading is used to eliminate the
background signal resulting from cell debris or other
non-specific absorbance.
12. The use of the blank control (only the culture medium without
cells and without treatment) is necessary to exclude the influ-
ence of color of the culture medium in the experiment, sub-
tracting the blank readings from the sample readings.
13. Absorbance values that are lower than the control cells indicate
a cell viability reduction. Conversely, a higher absorbance rate
indicates a cell viability/proliferation increase, as in the exam-
ple in Fig. 4.
Fig. 4 Schematic model of the 96-well plate representing the cell viability profile of X cells exposed to different
tetrazolium salts
Acknowledgments
A.B. Ribeiro was the recipient of a M.Sc. fellowship from the São
Paulo Research Foundation (FAPESP, Brazil; grant # 2018/
25770-7). D.C. Tavares is grateful to the National Council for
Scientific and Technological Development (CNPq, Brazil) for the
fellowship granted. H.S. Barud thanks CNPq (grant # 407822/
2018-6; INCT-INFO), FAPESP (grants # 2018/25512-8 and
2013/07793-6), and TA Instruments Brazil.
References
1. Groh KJ, Muncke J (2017) In vitro toxicity Biotechnol 17:1213–1221. https://doi.org/

testing of food contact materials: state-of-the- 10.2174/1389201017666160808160513
art and future challenges. Compr Rev Food Sci 4. Shokrzadeh M, Modanloo M (2017) An over-
F 16:1123–1150. https://doi.org/10.1111/ view of the most common methods for asses-
1541-4337.12280 sing cell viability. J Res Med Dent Sci 5:33–41.
2. Stockert JC, Horobin RW, Colombo LL, Bláz- https://doi.org/10.5455/jrmds.2017526
quez-Castro A (2018) Tetrazolium salts and 5. Liu X, Rodeheaver DP, White JC et al (2018) A
formazan products in cell biology: viability comparison of in vitro cytotoxicity assays in
assessment, fluorescence imaging, and labeling medical device regulatory studies. Regul Tox-
perspectives. Acta Histochem 120:159–167. icol Pharmacol 97:24–32. https://doi.org/10.
https://doi.org/10.1016/j.acthis.2018. 1016/j.yrtph.2018.06.003
02.005 6. Menyhárt O, Harami-Papp H, Sukumar S et al
3. Adan A, Kiraz Y, Baran Y (2016) Cell prolifer- (2016) Guidelines for the selection of func-
ation and cytotoxicity assays. Curr Pharm tional assays to evaluate the hallmarks of cancer.
Biochim Biophys Acta 1866:300–319. Pharmacol 67:17–30. https://doi.org/10.

https://doi.org/10.1016/j.bbcan.2016. 1016/j.bcp.2003.07.016
10.002 14. Wilson WR, Hay MP (2011) Targeting hyp-
7. Pr€abst K, Engelhardt H, Ringgeler S, Hübner oxia in cancer therapy. Nat Rev Cancer 11:
H (2017) Basic colorimetric proliferation 393–410. https://doi.org/10.1038/nrc3064
assays: MTT, WST, and Resazurin. Methods 15. Kamiloglu S, Sari G, Ozdal T, Capanoglu E
Mol Biol 1601:1–17. https://doi.org/10. (2020) Guidelines for cell viability assays.
1007/978-1-4939-6960-9_1 Food Front 1:332–349. https://doi.org/10.
8. Riss TL, Moravec RA, Niles AL et al (2016) 1002/fft2.44
Cell viability assays. In: Assay guidance manual 16. Dong G, Pan W, Zheng T et al (2006) Colori-
[Internet]. Eli Lilly & Company and the metric assay to determine in vitro antibacterial
National Center for Advancing Translational activity against clinical isolates: enhanced activ-
Sciences ity in damaged Chinese masson pine needles. J
9. van Meerloo J, Kaspers GJ, Cloos J (2011) Cell Integr Plant Biol 48:1034–1046. https://doi.
sensitivity assays: the MTT assay. Methods Mol org/10.1111/j.1744-7909.2006.00310.x
Biol 731:237–245. https://doi.org/10.1007/ 17. ISO 10993-1:2020 (2020) Part 1: evaluation
978-1-61779-080-5_20 and testing within a risk management
10. Berridge MV, Herst PM, Tan AS (2005) Tetra- process. In: Biological evaluation of medical
zolium dyes as tools in cell biology: new devices. International Organization for Stan-
insights into their cellular reduction. Biotech- dardization ISO, Geneva
nol Annu Rev 11:127–152. https://doi.org/ 18. ISO 10993-5:2020 (2020) Part 5: tests for
10.1016/s1387-2656(05)11004-7 in vitro cytotoxicity. In: Biological evaluation
11. Lü L, Zhang L, Wai MS et al (2012) Exocytosis of medical devices. International Organization
of MTT formazan could exacerbate cell injury. for Standardization ISO, Geneva
Toxicol In Vitro 26:636–644. https://doi. 19. ISO 10993-12:2020 (2020) Part 12: sample
org/10.1016/j.tiv.2012.02.006 preparation and reference materials. In:
12. Abbott A (2003) Biology’s new dimension. Biological evaluation of medical devices. Inter-
Nature 424:870–872. https://doi.org/10. national Organization for Standardization
1038/424870a ISO, Geneva
13. Hall MD, Martin C, Ferguson DJ et al (2004) 20. OECD (2016) Test No. 487: in vitro mamma-
Comparative efficacy of novel platinum lian cell micronucleus test, OECD guidelines
(IV) compounds with established chemothera- for the testing of chemicals, section 4. OECD
peutic drugs in solid tumour models. Biochem Publishing, Paris. https://doi.org/10.1787/
9789264264861-en
Chapter 8
In Vitro Genotoxicity/Mutagenicity Testing of Food

Packaging
Flávia A. Resende, Juliana G. F. Silva, Arthur B. Ribeiro,
Lucas N. F. Trevizan, Hernane S. Barud, and Denise C. Tavares
Abstract
To ensure the safe use of packaging materials and food contact materials, genotoxicity assessment is one of
the requirements of regulatory agencies around the globe. Thus, it is essential to carry out preliminary tests
to clarify this possible mechanism. The Salmonella/mammalian-microsome mutagenicity test, widely
known as the Ames test, is a rapid, relatively simple procedure for testing chemicals for mutagenicity as
well as for offering provision for the metabolism of otherwise nonmutagenic chemicals to their potentially
DNA-reactive forms. However, a single test is not sufficient to detect all relevant genotoxic mechanisms in
tumorigenesis. Thus, in order to complement the results in the Ames test and to contribute to the
elucidation of the effects, ensuring their use or not, mutagenicity at the chromosomal level must also be
evaluated. In the micronucleus (MN) assay, chromosomal damages induced by chemical products are
evaluated. The MN is expressed in dividing cultured cells because fragments from damaged chromosomes
or whole chromosomes that lag during anaphase become enveloped by nuclear membrane, independently
from the main nucleus during telophase, prior to cell division. Together, these tests detect the most relevant
events for the multistep process of malignancy, that is, gene mutations, clastogenicity, and aneugenicity.
Detailed descriptions of the protocols used for detection of point mutations and chromosomal damage
induced by food packaging in vitro are given in this chapter.
Key words Micronucleus test, Ames test, Salmonella/microsome assay, Genetic toxicology
1 Introduction
The genotoxic/mutagenic potential tests reveal, at the first level of

evidence, the genotoxicological effect of edible and/or food con-
tact packaging materials, having their safe use in mind. So, these
tests are always required irrespective of the extent of migration (see
Chapter 4 on IAS and NIAS, Chapter 5 on PFAS, and Chapter 6 on
migration) and the resulting human exposure to these materials
[1], because even though human exposure levels may be quantita-
tively low, these substances are considered to be of high toxicologi-
cal concern if they act as DNA reactive mutagens [2].
https://doi.org/10.1007/978-1-0716-3613-8_8,
149
150 Flávia A. Resende et al.
Commonly, the toxicological tests of food contact materials are

focused on single substances and their genotoxicity. However, most
of the time, the material is not available as a pure chemical and the
chemical identity is not known, causing people to be exposed to
mixtures of chemicals [3], that can or cannot interact with genetic
material. This interaction may result, under certain circumstances,
not only in cancer, but also in degenerative conditions such as
accelerated aging, immune dysfunction, cardiovascular and neuro-
degenerative diseases (in case of accumulation of DNA damage in
somatic cells) and in spontaneous abortions, infertility, or heritable
damage in the offspring and/or subsequent generations resulting
in genetic diseases (in case of DNA damage in germ cells) [4]. Thus,
sample preparation procedures need to be optimized and standar-
dized and approaches on the concept of safe level of food packaging
materials should be discussed. Regulatory agencies around the
globe have conducted research and developed both guidance and
regulations for safety assessments of materials intended to contact
food. Although the food packaging safety assessment structures
developed by these agencies have similar principles, they differ in
the application of these principles [5], which is necessary for new
approaches to meet this legal obligation for authorization applica-
tions of packaging materials.
DNA damage is a complex biological process involving several
modes of actions, determining the cellular fate and the severity of
the hazard. Currently, a wide variety of bioassays are available to
assess the genotoxic potential of chemicals and materials. These
assays have been evaluated for their ability to correctly predict the
adverse effects of matter and are often used as screening tools [6].
The Salmonella/Escherichia coli microsome assay (Ames test) is
required by regulatory authorities worldwide in order to identify
substances that can produce genetic damage that leads to gene
mutations (base substitution type mutations—S. Thyphimurium
strains TA1535, TA100, TA102, and TA104, and E. coli strains
WP2 uvrA or WP2 uvrA (pKM101), frameshift mutations—S.
Thyphimurium strains TA1537, TA1538, TA97a, and TA98)
[7]. The bacterial strains and mutagenicity test procedure, devel-
oped by Bruce Ames and published in 1973 [8, 9], still retain a
primary role in the testing of chemicals and materials for commer-
cial use [7].
The Ames test uses amino acid-dependent strains of S. Typhi-
murium and E. coli, each carrying different mutations in various
genes of either histidine operon, in S. Typhimurium bacteria, or
tryptophan operon in E. coli making them auxotrophic for the
corresponding amino acids. These mutations act as hot spots for
mutagens that cause DNA damage by different mechanisms. In
addition, some strains may have (i) rfa mutations, which cause
changes in the lipopolysaccharide barrier of the bacterial cell wall,
thus facilitating the entry of large molecules (all Salmonella strains);
Genotoxicity of Food Packaging 151
(ii) deficiency of the nucleotide excision repair system, preventing

the bacterium from repairing the damage that has been done to its
genetic material (uvrB detection in Salmonella strains, except
TA102, or uvrA mutation in E. coli strains); and (iii) the plasmid
R factor (plasmid pKM101) that confers resistance to ampicillin
and induces an error-prone DNA repair pathway (strains TA97,
TA97a, TA98, TA100, TA102, and WP2 uvrA (pKM 101)).
Together, these mutations give the strains greater sensitivity in
detecting several mutagens [10, 11].
Due to the inability to synthesize histidine, these strains cannot
grow and form colonies in the absence of this essential amino acid
[12]. However, when they are exposed to agents that induce new
mutations in the gene, this function is restored and the auxotrophic
character is reversed, allowing bacteria to grow and form colonies
[11]. Thus, the assay detects the mutational reversion of
his-dependent bacteria to his-independent colonies (Salmonella)
or trp-dependent bacteria to trp-independent colonies (E. coli).
The mutagenic potential of a compound can then be calculated
from the number of colonies that are formed on the plate by the
concentration of the compound used [12].
Another consideration about the test is the addition of the
so-called S9 fraction, obtained from rat liver, which contains xeno-
biotic metabolizing enzymes for the identification of mutagenic
agents of indirect action, which must be metabolized in order to
become active [13]. Therefore, in the absence (-S9) or presence
(+S9) of metabolic activation, the monitoring of the direct and
indirect actions of a compound, respectively, is possible, guarantee-
ing the faithful identification of agents that cause gene mutations.
For the detection of chromosomal damage, the micronucleus
(MN) test is a widely used method that detects chromosomal loss
and breakage, being used as biomarker for the identification of
clastogenic and aneugenic agents [14]. The chromosomal changes
identified in the MN test are verified by counting circular structures
surrounded by nuclear membrane, called MN, which are formed by
chromosomal fragments or whole chromosomes that were delayed
during anaphase and were not incorporated into the nuclei of
daughter cells [15].
This test can be performed in vitro, using cultured primary
human or other mammalian peripheral blood lymphocytes and
several rodent cell lines such as CHO (Chinese hamster ovary
cells), V79 (Chinese hamster lung fibroblasts, male), CHL/IU
(Cricetulus griseus, Chinese hamster lung fibroblasts, female), and
L5178Y (mouse lymphoma) cells, or human cells, such as TK6
(human spleen lymphoblasts). Other cell lines such as HT29
(human colorectal adenocarcinoma), Caco-2 (Caucasian colon ade-
nocarcinoma), HepaRG (hepatic stem cell line), HepG2 cells (liver
hepatocellular carcinoma), A549 (adenocarcinomic human alveolar
basal epithelial cells), and primary Syrian Hamster Embryo cells
have been used for MN testing, but have not been extensively
validated to date. Therefore, the choice of these cells should be
justified [16]. The MN test can still be performed in vivo using
hematopoietic rodent cells [17].
The OECD 487 Test Guideline [16] allows the use of proto-
cols with and without cytochalasin B (cytoB). CytoB inhibits actin
polymerization and blocks cytokinesis, and cells that have com-
pleted one cell cycle after treatment can be distinguished from
undivided cells by their binucleate appearance. The advantage of
using cytoB is that it allows the clear identification that treated and
control cells have divided in vitro, and also provides a simple
assessment of cell proliferation, allowing for the identification and
analysis of MN only in cells that have completed one mitosis. The
use of protocols without cytokinesis block can be accomplished,
provided there is evidence that the cell population analyzed has
undergone mitosis.
Thus, the bacterial reverse gene mutation test and the in vitro
MN assay detect two main genetic endpoints, that is, gene and
chromosome mutations, respectively. Therefore, these tests are
currently considered equally appropriate in a standard genetic toxi-
cology battery for predicting potential human risks [18].
2 Materials
1. General laboratory glassware (flasks, bottles, graduated cylin-

ders, etc.).
2. Petri dishes (100 × 15 mm2).
3. Sterile glass tubes (50 × 16 mm2) with caps.
4. Test tube racks.
5. Pipets (1, 2, 5, and 10 mL).
6. Pipettors (adjustable volumes).
7. Sterile pipette tips.
8. Cryogenic tubes for freezing down permanent and working
cultures.
9. Colony counter (manual or electronic).
10. 6-Well Microtiter Plate (flat-bottom).
11. Conical tubes (15 mL).
12. Microscope slides.
13. Cell culture flasks.
General Apparatuses
1. Autoclave.
2. Shaking incubator set at 120 rpm and 37 °C.
3. Incubator for the GM agar plates.

4. Oven, heating, or water bath set at 43–48 °C to maintain
temperature of top agar.
5. Boiling water bath or microwave oven for melting top agar.
6. Magnetic stirrers.
7. Analytical balances (up to 0.001 g).
8. Water purification system to generate distilled water.
9. Ultrafreezer or liquid nitrogen tank.
10. Refrigerator/freezer.
11. Biological safety cabinet.
12. Inverted light microscope.
13. Incubator with humidified atmosphere of 5% carbon diox-
ide (CO2) and temperature of 37 °C for cell growth.
14. Cytocentrifuge.
15. Microscope with excellent optics for bright-field and fluo-
rescence examination of stained slides at ×1000
magnification.
Chemicals
2.1 Salmonella/ 1. Agar.

Microsome assay 2. Glucose.
(Salmonella Test;
3. D-biotin.
Ames Test)
4. L-Histidine∙HCl.
5. Sodium chloride.
6. Oxoid nutrient broth No. 2.
7. Monobasic sodium phosphate.
8. Dibasic sodium phosphate.
9. Magnesium chloride.
10. Potassium chloride.
11. D-glucose-6-phosphate disodium.
12. Nicotinamide adenine dinucleotide phosphate sodium salt.
13. Mammalian tissue homogenate (S9 fraction).
14. Dimethyl sulfoxide (DMSO) or other solvents that maximize
extraction of the substances present in the food packaging
materials.
15. Glycerol.
16. Positive control chemicals (see Note 1).
2.2 Micronucleus 1. Cell growth media: Eagle’s Minimal Essential Medium

Test (EMEM), Dulbecco’s Modified Eagle’s medium (DMEM),
RPMI, Ham’s F10, Ham’s F12 (check the most suitable one
for each test cell).
2. Phosphate-buffered saline (PBS) without Ca2+ and Mg2+.
3. Fetal bovine serum (FBS).
4. Cytochalasin B.
5. Dyes: Giemsa, acridine orange or panoptic (check the most
suitable one for each test cell).
6. DMSO or other solvents that maximize extraction of the sub-
stances present in the food packaging materials.
7. Positive control chemicals (see Note 1).
3 Methods
Extraction
Water and culture medium are the most commonly used solvents.
In case of water, it is advised not to exceed a maximum concentra-
tion of 10 vol% because of molarity changes on the medium and
dilution of nutrients.
If other than well-established solvents/vehicles are used, their
inclusion should be supported by data indicating their compatibil-
ity with the test system and their ability to maximize extraction of
the substances present in the food packaging materials. They must
not interfere with cell proliferation, metabolic activation, and must
not induce DNA changes.
The International Standard ISO 10993-12 [19] suggests that
the extraction of substances from films (thickness <0.5 mm) should
be carried out at 37 ± 1 °C for 72 ± 2 h. Other conditions for
extraction also are described, but care should be taken that this does
not alter the chemical characteristics of food packaging. The surface
area of the films must be of 6 cm2 to the volume of 1 mL of
extraction vehicle. When surface area cannot be determined, a
mass/volume of extracting fluid shall be used [19].
Fresh preparations should be employed unless stability data
demonstrate the acceptability of storage.
Food packaging extract is added directly to the test systems
and/or diluted prior to treatment if it interferes with bacterial or
cellular growth and survival.
3.1 Salmonella/ According to the OECD Guidelines [20], at least five strains of
Microsome Assay bacteria should be used. The recommended combination of strains
(Salmonella Test; is TA1535; TA1537, TA97a or TA97; TA98; TA100; and TA102
Ames Test) of S. Typhimurium, E. coli WP2 uvrA, or E. coli WP2 uvrA
(pKM101).
Preparation of Permanent Cultures

Receive strains on a small sterile filter disk embedded in nutrient
agar; first wipe the disk across the surface of a nutrient agar plate,
and then transfer the disk to 5 mL of nutrient broth. For lyophi-
lized culture, aseptically add 1 mL of nutrient broth to rehydrate
the culture (a process which should take up to 2 min), then transfer
the rehydrated culture to 4 mL of nutrient broth. Then, transfer a
drop of the culture to a nutrient agar plate and streak the inoculum
for individual colonies across the surface of the plate [11].
If single colonies are observed after overnight incubation at
37 °C, pick one healthy looking colony and restreak it for individual
colonies on minimal agar medium (GM agar) plate supplemented
with biotin and histidine for S. Typhimurium strains, and trypto-
phan for E. coli strains for purification and verification of genotypic
characteristics [11, 20].
Other genotypic characteristics that should be similarly
checked: ampicillin resistance in strains TA98, TA100 and TA97a
or TA97, and WP2 uvrA (pKM101); ampicillin + tetracycline resis-
tance in strain TA102 to assess the presence or absence of R-factor
plasmids; rfa mutation in S. Typhimurium through sensitivity to
crystal violet, and uvrA mutation in E. coli or uvrB mutation in S.
Typhimurium, through sensitivity to ultraviolet light [20]. The
strains should also yield spontaneous revertant colony plate counts
within the frequency ranges expected from the laboratory’s histori-
cal control data and preferably within the range reported in the
literature [11, 13].
The strains grown in nutrient broth at a density of 1 to 2 × 109
colony-forming unit (CFU) mL-1 (Optical Density540 nm between
0.1 and 0.2) are frozen with 0.5 mL of a sterile cryopreserver, such
as glycerol or DMSO for the culture (final concentration, 10 vol%);
mix well and dispense 1-mL aliquots in pre-marked sterile cryo-
genic tubes.
The tester strains should be kept frozen in ultrafreezer (-80 °C)
or liquid nitrogen.
Preparation of Solutions and Plates

1. Minimal agar medium (GM agar) plates: Add 15 g of agar to
900 mL distilled water. Autoclave it for 15 min at 121 °C
(1.5 atm; relative pressure). Add 20 mL of sterile Vogel-Bonner
E medium (VB salts), and mix thoroughly, then add 50 mL of a
sterile glucose (40 or 8 vol%) solution; again, swirl thoroughly.
Dispense the agar medium in 100 × 15 mm2 Petri dishes
(approximately 25 mL per plate). After the solidification, the
plates must be stored in an oven at 37 °C for 48 h (see Note 2).
2. Histidine/biotin solution (0.5 mM): Add 124 mg of D-biotin
and 96 mg of L-Histidine∙HCl to 1000 mL distilled water.
Autoclave it for 30 min at 121 °C (1.5 atm; relative pressure).
3. Top agar supplemented with histidine/biotin: Add 6 g of agar

and 6 g of sodium chloride to 900 mL of distilled water.
Autoclave it for 30 min at 121 °C (1.5 atm; relative pressure).
For the test, melt the top agar in a microwave oven or in boiling
water, and then add 100 mL of limited histidine and biotin
solution (0.5 mM). The top agar is used to deliver the bacteria,
chemical, and buffer or S9 mix to the bottom agar and it is one
of the most critical medium components in the Ames test
because it contains the trace amount of histidine (0.05 mM)
for limited growth and biotin at a concentration of 0.05 mM,
which is in excess of what is needed for the growth of the
Salmonella strains.
4. Nutrient broth: Add 0.75 g of nutrient broth (Oxoid nutrient
broth No. 2) to 30 mL water. Autoclave it for 15 min at 121 °C
(1.5 atm; relative pressure).
5. Sodium phosphate buffer, 0.1 mM, pH 7.4: After mixing 120 mL
monobasic sodium phosphate (0.1 M) and 880 mL of dibasic
sodium phosphate (0.1 M), adjust pH to 7.4 using 0.1 M
dibasic sodium phosphate solution. Autoclave it for 30 min at
121 °C (1.5 atm; relative pressure). The buffer is used for
testing chemicals in the absence of metabolic activation (see
Note 3).
6. Co-factors for S9 mix: A number of commercial vendors provide
S9 preparations, as Molecular Toxicology, in the United States.
Once the S9 mix is prepared, it should be kept on ice for the
duration of the experiment. The metabolic activation system
consisted of 4% S9 fraction, 1% of magnesium chloride at
0.4 mol L-1, 1% of potassium chloride at 1.65 mol L-1, 0.5%
of D-glucose-6-phosphate disodium at 1 mol L-1, 4% of nico-
tinamide adenine dinucleotide phosphate sodium salt (NADP)
at 0.1 mol L-1 in 50% of phosphate buffer at 0.2 mol L-1, and
39.5% of sterile distilled water.
Inoculum
For each experiment, individual culture flasks are inoculated with

each strain. Inoculate 0.1 mL of the tester strain cultures in 30 mL
of Oxoid nutrient broth No. 2 and place on a shaker in the dark and
gently shake (100 rpm) for 11–14 h (overnight) at 37 °C. On the
next morning, remove the cultures from the incubator and keep at
room temperature away from direct fluorescent light. It is essential
that the cultures used in the experiment contain a high titer of
viable bacteria. The titer may be demonstrated either from histori-
cal control data on growth curves, or in each assay through the
determination of viable cell numbers by a plating experiment.
Experimental Procedure
The procedure described here pertains to the preincubation
method.
1. To the sterile glass tubes maintained at room temperature, add
the following components in this order, with mild mixing after
each addition:
• 0.5 mL of metabolic activation (S9) mix or sodium phos-
phate buffer.
• Different volumes of the food packaging extract. Include
untreated, solvent/vehicle and strain-specific positive con-
trols, both with and without metabolic activation, in each
assay (see Notes 1 and 4).
• 0.1 mL overnight culture of the bacterial strain.
2. Incubate the mixture at 37 °C for 20 min.
3. To each tube, add 2 mL of molten top agar supplemented with
histidine/biotin maintained at 43–48 °C. The content of the
test tubes is then mixed and poured onto the surface of GM
agar plates (see Note 5).
4. When the top agar has solidified (2–3 min), the plates are
inverted and incubated at 37 °C for 48 h.
5. The colonies are then counted, and the results are expressed as
the number of revertant colonies per plate. If colonies cannot
be counted immediately after the 48 h incubation, the plates
can be stored in a refrigerator for up to 2 days. All plates must
be removed from the incubator and counted at the same time
(Fig. 1).
Preliminary experiments are useful to determine toxicity and
insolubility of the food packaging samples. Cytotoxicity may be
detected in the final population on the GM agar plate after the
48-h incubation by a thinning of the background lawn, which may
be accompanied by a decrease in the number of revertant colonies,
Fig. 1 Petri dishes with revertant colonies of Salmonella Typhimurium (TA102

strain). (a) Untreated control; (b) Positive control (Mitomycin C)
absence of background lawn (i.e., complete absence of growth), or

by presence of pinpoint non-revertant colonies (generally in con-
junction with an absence of background lawn).
Analysis of the Results

After the plates are removed from the incubator, the colonies are
counted, and the results are expressed as mean revertant colonies
per plate ± standard deviation for each dose of the test sample and
the controls. A sample is considered a mutagen (it induces point
mutations by base substitutions or frameshifts in the genome of
either S. Typhimurium and/or E. coli) if it produces a
concentration-related increase over the range tested and/or a
reproducible increase at one or more concentrations in the number
of revertant colonies per plate in at least one strain with or without
metabolic activation system. The determination of a positive vs. a
negative result is made through evaluation procedures for compar-
ing dosed plates with the concurrent solvent/vehicle control plates,
including a requirement for a specific fold increase (2- or 3-fold,
specific to the bacterial strain).
3.2 Micronucleus Experimental Procedure

Test
1. In 6-well plates, seed 100,000 cell per well in 2 mL of cell
growth media (see Note 6) with 10% of fetal bovine serum
(complete culture medium). Prepare an appropriate number
of wells for the experiment from a single pool of cells. These
cells should be in exponential growth phase at the time of
treatment.
2. Incubate the plates at 37 °C in a 5% CO2 atmosphere for 24 h.
This time is necessary for the adhesion of cells to the plate, for
the formation of a semiconfluent cell monolayer, and for the
progression of cells to the exponential growth phase.
3. After incubation, the solutions for the treatment groups must
first be prepared in a culture medium in an amount sufficient
for the treatment of the test. For treatment, it is optional to
change the culture medium or not.
4. Three non-cytotoxic concentrations of the food packaging
extract should be evaluated. Include untreated, solvent/vehi-
cle, and positive controls in each treatment series (see Notes 1
and 7). Prepare duplicate cultures/wells at each experimental
test point.
5. Smoothly homogenize the cultures with cross movements,
avoiding bubble formation.
6. For the treatment, cells should be exposed to the food packag-
ing extract without metabolic activation for 3–6 h, and sampled
at a time equivalent to about 1.5–2.0 normal cell cycle lengths
after the beginning of treatment. To conclude a negative
outcome, cells should be continuously exposed without meta-

bolic activation until sampling at a time equivalent to about
1.5–2.0 normal cell cycle lengths [19]. After treatment, all
plates are placed in a CO2 incubator at 37 °C and 5% CO2.
7. After the end of the exposure time, analyze the appearance of
the cultures under an inverted microscope, mainly regarding
the presence of precipitation, morphology, and cell death.
8. After the incubation period, remove the culture medium con-
taining the test sample and wash the wells twice with 2 mL of
PBS suitable for cell cultures.
9. Add fresh medium (2 mL per well).
10. Then add 20 μL per well of the 3 μg mL-1 pre-prepared
cytochalasin B solution.
11. The plates must be placed in the CO2 incubator for 1.5–2
normal cell cycle lengths.
Harvest
It is important to cast a water film on the slides, so that they can be

ready and cold (2–8 °C) for use. To do this, wash the slides with
neutral detergent, rinse them under running water, and then in
distilled water. Then, the slides should be immersed, one by one, in
distilled water and raised to check if a water film has formed on
them. If the film does not form, wash again. The vial containing the
slides must be kept and refrigerated until the moment of use. There
is also another possibility of cleaning the slides, this being cleaning
them in 70% ethanol and distilled water.
12. Collect the medium from each well into appropriately labelled
15-mL centrifuge tubes to avoid loss of detached mitotic
cells.
13. The cells must be washed twice with PBS (2 mL), the first
washing being reserved in the centrifuge tube and the second
washing must be discarded. Remove excess PBS from each
well with a pipette.
14. Add 0.3 mL of 0.1% trypsin to each well to bring cell mono-
layers into suspension. Time lapse (approximately 5 min) and
temperature of incubation (37 °C) are indicative and should
be standardized in each laboratory based on visual observa-
tions of cell detachment, as trypsin activity may vary among
different lots.
15. When monolayer cells are completely detached, inactivate
trypsin with 0.7 mL of complete culture medium (the
medium reserved in the centrifuge tubes can be used) to
block enzymatic digestion.
16. Add these cell suspensions to their respective centrifuge tubes

and centrifuge them for 5 min at 1000 rpm.
17. Aspirate the supernatant carefully using a glass pipette, leaving
approximately 0.5–1.0 mL as the pellet protection margin.
18. With the aid of the pipette, resuspend the cells, gently homo-
genizing the pellet (20×) and keep the solution inside the
pipette so that all samples come into contact with the potas-
sium chloride (KCl; 0.075 M) at the same time.
19. Add 3 mL of KCl previously cooled to 2–8 °C and incubate
for approximately 7–10 min at room temperature. During
this period, gently resuspend the cells, mixing 40× with the
pipette.
20. After incubation with KCl, add 0.5 mL of the fixative (3:1 v/v
methanol-glacial acetic acid), prepared at the time of use and
kept at room temperature, and mix gently.
21. Centrifuge the cultures for 5 min at 1000 rpm. Remove the
supernatant, leaving approximately 0.5–1.0 mL.
22. Homogenize the pellet (20×) and keep the solution inside the
pipette.
23. Add 5 mL of fixative and mix immediately.
24. Centrifuge the cultures for 5 min. Remove the supernatant.
25. Add 5 mL fixative again and mix.
26. In this step, cultures should be kept in a refrigerator (2–8 °C)
for at least 1 h and/or until the slides are prepared.
27. At the time of making the slides, centrifuge the tubes, remove
the supernatant, and finally produce an appropriately concen-
trated cell suspension, maintaining approximately 0.5 mL of
fixative.
28. Approximately five drops cellular solution should be dripped
under the identified slide. Prepare at least three slides for each
experimental point, labeled with the identity of the culture.
Leave the slides to dry at room temperature prior to staining
for at least 1 day. After this period, stain the slides or store in
slide boxes.
29. The slides can be stained with 3 vol% Giemsa in tap water,
0.0125% (w/v) acridine orange in PBS, panoptic, among
others, depending on the most suitable one for each test cell
(Fig. 2).
Analysis of the Results (See Note 8)
The frequencies of cells with MN (with one, two, and more than
two MN) are recorded. A total of 6000 binucleated cells are scored
per treatment, corresponding to 2000 cells per treatment per repe-
tition. Attention should be given to ensure that MN are scored only
Fig. 2 Microphotograph (1000× magnification) of a micronucleated binucleated

HepG2 cell. Staining with Giemsa (3%)
in binucleated cells and not in multinucleated cells, because multi-

nucleated cells are not once-divided cells and tend to have greatly
elevated MN frequencies relative to binucleated cells, which would
result in inaccurate genome damage estimates [17].
A sample is considered a mutagen in this assay if statistically
significant increases in the proportion of micronucleated cells over
the negative/solvent controls (reference point for comparison in
the statistical evaluation of the results) are observed at one or more
concentrations.
Furthermore, the determination of the Cytokinesis-Block Pro-
liferation Index (CBPI) may be used to calculate cell proliferation.
CBPI indicates the average number of nuclei per cell. Cells with
well-preserved cytoplasm containing 1–4 nuclei are scored. Analyze
1500 cells per treatment for a total of 500 cells per repetition. CBPI
is calculated using the following formula [16]:
ððNo:mononucleate cellsÞ þ ð2 × No:binucleate cellsÞ þ ð3 × No:multinucleate cellsÞÞ

CBPI =
ðTotal number of cellsÞ
4 Notes
1. Examples of suitable positive controls for Salmonella/micro-

some assay (to confirm the reversion properties and specificity
of each tester strain, and the efficacy of the metabolic activation
system): 4-nitro-o-phenylenediamine (TA98 and TA97a, 10 μg
per plate); sodium azide (TA100, 1.25 μg per plate); mitomy-
cin C (TA102, 0.5 μg per plate), in the absence of S9 and
2-anthramine (TA98, TA100, TA 97a, 1.25 μg per plate),
2-aminofluorene (TA102, 10 μg per plate), in the presence
of S9.
For MN test, the most commonly used positive control
agents are: methyl methanesulfonate, mitomycin C, 4-nitro-
quinoline-N-oxide, cytosine arabinoside, benzo(a)pyrene,
Table 1 Recipe for Vogel-Bonner medium E (50X)
Ingredients Quantity per liter

1. Warm distilled water (about 50 °C) 670 mL
2. Magnesium sulfate (MgSO4.H2O) 10 g
3. Citric acid monohydrate 100 g
4. Potassium phosphate, dibasic, anhydrous (K2HPO4) 500 g
5. Sodium ammonium phosphate (Na2NH2PO4.4H2O) 175 g
cyclophosphamide, colchicine, or vinblastine. Concentrations

should be defined in preliminary tests.
2. Vogel–Bonner (VB salts) medium E.
The ingredients must be added in the order indicated
below. Make sure that each salt is dissolved thoroughly by
stirring it on a magnetic stirrer before adding the next salt
(Table 1).
• The agar should never be autoclaved together with the VB
salts and glucose.
• The plates can be stored at 4 °C for several weeks when
placed in sealed plastic bags to prevent dehydration. Before
use, the plates should be warmed up to room temperature
and examined for excess moisture.
3. Sodium phosphate, monobasic (0.1 M): To 1 L of water, add
13.8 g NaH2PO4.H2O.
Sodium phosphate, dibasic (0.1 M): To 1 L of water, add
14.2 g Na2HPO4.H2O.
4. At least five different concentrations of the food packaging
extract should be selected for the test, and the interval between
each concentration should be approximately half log (√10). At
least three plates for each dose level and for the controls is
recommended.
Extracts obtained from aqueous solvents can be used at
levels up to approximately 1 mL per plate before they interfere
with the gelling of the top agar, while organic solvents are often
used at a maximal dose of 0.1 mL per plate [21].
Untreated control: also called spontaneous control, as it
aims to demonstrate the rate of spontaneous reversion of each
strain and that no deleterious or mutagenic effects are induced
by the chosen solvent.
Solvent/vehicle control: solvent or vehicle alone, without
test substance, and treated in the used maximum volume in the
treatment group. In the Ames test, the solvent/vehicle control

is also called negative control [20].
Positive control (see Note 1).
5. It is important to quickly swirl the plates after the addition of
the top agar to the surface of the GM agar plates to ensure an
even distribution of the top agar that contains the bacteria, test
sample, and S9 mix or buffer.
6. The most common culture media for the cell lines mentioned
in the Introduction section are Eagle’s Minimal Essential
Medium (EMEM), Dulbecco’s Modified Eagle’s medium
(DMEM), RPMI, Ham’s F10, Ham’s F12.
7. When cytochalasin B is used, the most appropriate method to
assess cytotoxicity is to calculate the cytokinesis-block prolifer-
ation index (CBPI). CBPI: the proportion of second-division
cells in the treated population relative to the untreated control.
Untreated control: also called negative control, only 2 mL
of complete culture medium.
Solvent/vehicle control: solvent or vehicle alone, without
test substance, and treated in the used maximum volume in the
treatment group.
Positive control (see Note 1).
8. MN are morphologically identical, but smaller than nuclei. The
diameter of MN usually varies between 1/16th and 1/3rd of
the mean diameter of the main nuclei, which correspond to
1/256th and 1/9th of the area of one of the main nuclei in a
binucleated cell, respectively. MN are non-refractile and they
are not linked or connected to the main nuclei. MN may touch
but not overlap the main nuclei and the micronuclear boundary
should be distinguishable from the nuclear boundary. More-
over, MN usually have the same staining intensity as the main
nuclei, but occasionally staining may be more intense.
For analysis of CBPI, mono-, bi, and multinucleated cells
are viable, with an intact cytoplasm and normal nucleus mor-
phology containing one, two, and three or more nuclei,
respectively.
Acknowledgments
F. A. Resende thanks the São Paulo Research Foundation

(FAPESP; grant # 2017/16278-9). H. S. Barud thanks the
National Council for Scientific and Technological Development
(CNPq; grant # 407822/2018-6; INCT-INFO), FAPESP (grants
# 2018/25512-8 and 2013/07793-6), and TA Instruments Bra-
zil. D. C. Tavares is grateful to CNPq for the fellowship granted.
References
1. Bolognesi C, Castoldi AF, Crebelli R, 10. Tejs S (2008) The Ames test: a methodological
Barthélémy E, Maurici D, Wölfle D, Volk K, short review. Environ Biotechnol 4:7–14
Laurence Castle L (2017) Genotoxicity testing 11. Mortelmans K, Zeiger E (2000) The Ames
approaches for the safety assessment of sub- Salmonella/microsome mutagenicity assay.
stances used in food contact materials prior to Mutat Res 455:29–60. https://doi.org/10.
their authorization in the European Union. 1016/s0027-5107(00)00064-6
Environ Mol Mutagen 58:361–374. https:// 12. Zeiger E (2001) Mutagens that are not carci-
doi.org/10.1002/em.22094 nogens: faulty theory or fault tests? Mutat Res
2. Rainer B, Mayrhofer E, Redl M, Dolak I, 492:29–38
Mislivececk D, Czerny T, Kirchnawy C, 13. Maron DM, Ames BN (1983) Revised meth-
Marin-Kuan M, Schilter B, Tacker M (2019) ods for the Salmonella mutagenicity test.
Mutagenicity assessment of food contact mate- Mutat Res 113:173–215. https://doi.org/10.
rial migrates with the Ames MPF assay. Food 1016/0165-1161(83)90010-9
Addit Contam Part A Chem Anal Control
Expo Risk Assess 36:1419–1432. https://doi. 14. Doherty AT (2012) The in vitro micronucleus
org/10.1080/19440049.2019.1634841 assay. Methods Mol Biol 817:121–141.
https://doi.org/10.1007/978-1-61779-
3. Groh KJ, Muncke J (2017) In vitro toxicity 421-6_7
testing of food contact materials: state-of-the-
art and future challenges. Compr Rev Food Sci 15. Fenech M (2006) Cytokinesis-block micronu-
Food Saf 16:1123–1150. https://doi.org/10. cleus assay evolves into a “cytome” assay of
1111/1541-4337.12280 chromosomal instability, mitotic dysfunction
and cell death. Mutat Res 600:58–66.
4. Cartus A, Schrenk D (2017) Current methods https://doi.org/10.1016/j.mrfmmm.2006.
in risk assessment of genotoxic chemicals. Food 05.028
Chem Toxicol 106:574–582. https://doi.org/
10.1016/j.fct.2016.09.012 16. OECD (2016) Test no. 487: in vitro mamma-
lian cell micronucleus test, OECD guidelines
5. Karmaus AL, Osborn R, Krishan M (2018) for the testing of chemicals, section 4. OECD
Scientific advances and challenges in safety Publishing, Paris. https://doi.org/10.1787/
evaluation of food packaging materials: work- 9789264264861-en
shop proceedings. Regul Toxicol Pharmacol
98:80–87. https://doi.org/10.1016/j.yrtph. 17. Fenech M (2007) Cytokinesis-blocked micro-
2018.07.017 nucleus cytome assay. Nat Protoc 2:1084–
1104. https://doi.org/10.1038/nprot.
6. Pinter E, Rainer B, Czerny T, Riegel E, 2007.77
Schilter B, Marin-Kuan M, Tacker M (2020)
Evaluation of the suitability of mammalian 18. EFSA Scientific Committee (2011) Scientific
in vitro assays to assess the genotoxic potential opinion on genotoxicity testing strategies
of food contact materials. Foods 9:237. applicable to food and feed safety assessment.
https://doi.org/10.3390/foods9020237 EFSA J 9(9):2379. [69 pp.]. https://doi.org/
10.2903/j.efsa.2011.2379
7. Zeiger E (2019) The test that changed the
world: the Ames test and the regulation of 19. ISO 10993-12:2020, Biological evaluation of
chemicals. Mutat Res 841:43–48. https://doi. medical devices – Part 12: Sample preparation
org/10.1016/j.mrgentox.2019.05.007 and reference materials
8. Ames BN, Lee FD, Durston WE (1973) An 20. OECD (2020) Test no. 471: bacterial reverse
improved bacterial test system for the detection mutation test, OECD guidelines for the testing
and classification of mutagens and carcinogens. of chemicals, section 4. OECD Publishing,
Proc Natl Acad Sci U S A 70:782–786. Paris. https://doi.org/10.1787/20745788
https://doi.org/10.1073/pnas.70.3.782 21. Hamel A, Roy M, Proudlock R (2016)
9. Ames BN, Durston WE, Yamasaki E, Lee FD Chapter 4 – The bacterial reverse mutation
(1973) Carcinogens are mutagens: a simple test. In: Genetic toxicology testing: a labora-
test system combining liver homogenates for tory manual. Elsevier, Amsterdam, pp 79–138.
activation and bacteria for detection. Proc Natl https://d oi.org/10.1016 /B97 8-0-12-
Acad Sci U S A 70:2281–2285. https://doi. 800764-8.00004-5
org/10.1073/pnas.70.8.2281
Part II
Microstructural and Barrier Features of Food Packaging

Materials
Chapter 9
Microstructural and Defect Analysis of Food Packaging

Materials Through X-Ray Microtomography
Marcos V. Lorevice, Pedro I. C. Claro, Diego M. Nascimento,
and Rubia F. Gouveia
Abstract
Packaging is an important part of food products, preserving their main components and attracting
consumer’s interest. However, when damage or flaws are present, food stability can be compromised.
Several factors related to the design (composition and function), processing, and production of food
packaging materials may result in defects that, although unavoidable, must be traceable and monitored.
Few protocols have been proposed to effectively identify defects, most of which destructively or invasively,
preventing the in situ characterization of the internal microstructure. We herein propose a general, simple
protocol to detect and quantify defects, and analyze engineered microstructures in food packaging by X-ray
microtomography. The technique requirements, such as sample characteristics and preparation, equipment
setup, data acquisition and processing, as well as image segmentation, are elucidated by showcasing two
common food packaging materials: paper/plastic/metal multilayers (Tetra Pak®) and polyethylene. X-ray is
comprehensively depicted as an easy, helpful, noninvasive, and nondestructive method to improve the
quality control of food packaging materials.
Key words X-ray tomography, Microstructure, Failures, Defects
1 Introduction
Since the beginning of human civilization, packaging influences the

way in which food is stored and consumed. The first record of
packaging used to store and transport food dates of around
200 years before the common era BCE [1]. The main functions
of food packaging are to protect and preserve the quality and safety
of food during transportation, distribution, storage, commerciali-
zation, and consumption. Moreover, packaging should reduce food
losses and contamination during storage and provide relevant infor-
mation (e.g., nutrition facts and shelf-life). Even today, the food
packaging sector plays a crucial role in the world economy [2, 3],
https://doi.org/10.1007/978-1-0716-3613-8_9,
167
168 Marcos V. Lorevice et al.
accounting for USD 228 billion in 2018 globally, value that is

expected to reach USD 441 billion in 2025 [4].
Food products comprise a synergic sum of different harmo-
nized features: food typology, packaging materials, subliminal mes-
sages, and sensory (e.g., color, taste, texture, flavor, etc.)
impression [5]. From the materials science perspective, the right
selection of raw materials for a particular food packaging requires
the best mechanical, optical, thermal, and barrier properties in
order to preserve and prolong food shelf-life. Food packaging
materials are commonly based on metal, glass, paper and paper-
board, plastics, and even combinations among these into complex
and convenient containers, like Tetra Pak® [6]. Depending on the
technical application, packaging materials can be classified as pri-
mary (direct contact with the food), secondary, and tertiary (the
latter being external to the former and providing physical
integrity) [7].
In spite of all technological efforts that have been made in the
past decades to improve food packaging properties or to add new
functions toward the so-called active and smart packaging [8–10],
the occurrence of flaws or defects always overcomes the technolog-
ical advances. The safety hazards or unhealthy appearance caused by
damaged food products (food/packaging) generally leads to unat-
tractive products to the consumers, and consequently, to an unde-
sirable reduction of their selling [11].
Defect and flaws in food packaging are frequently caused dur-
ing packaging and processing operations (structure/composition,
inferior compatibility among the raw materials), product storage
(temperature and relative humidity), or even by consumer handling
and disposal. According to their effect in packaging integrity and
effective applicability, flaws and/or damages have been categorized
in critical (e.g., collapse, rupture, holes, seal cracks, tears, or cracks),
major (e.g., poor mechanical resistance, misplaced lids or crimp
seals), and minor defects (streaky pigmentation from inadequate
pigment blending, scratches on the lamination, beads over surface
of foil, mold, drawing lines, improper identification on label) [6].
The vast and well-known materials characterization, particu-
larly of their morphological (optical and electron microscopies),
mechanical (compression and tensile tests), thermal (thermogravi-
metry analysis, differential scanning calorimetry, and dynamic
mechanical thermal analysis), barrier (oxygen, CO2, and water),
or even active (antimicrobial and sensory) properties, have been
applied to assess packaging performance and to identify possible
failures with considerable accuracy, besides exhibiting sample-
destructive or -invasive characteristics in most of the cases [12, 13].
Apart from defects easily identified by a rapid inspection
(in some cases visual or by analyzing changes in packaging content,
like off-flavor or color), the smaller ruptures or holes, internal and
structural failures, imperfections, or poor phase compatibility in
X-Ray Microtomography of Food Packaging 169
packaging are usually visual-untraceable, and they can interfere

substantially with the final packaging properties. Even for complex
food packaging systems, like Tetra Pak®, which requires phase
compatibility and can otherwise exhibit defects in one or more of
the four layers, tracking imperfections is something difficult via
some of the current materials analyses. In addition, the incorpora-
tion of micro and nanostructures to enhance packaging properties
requires more accurate techniques in resolution, as well as in mor-
phological statistics to identify and to infer the origin of these
undesirable defects [14].
To overcome these drawbacks, nondestructive or noninvasive
techniques, like ultrasound or X-ray tomography [15, 16, 17], have
been widely applied as effective procedures to investigate internal
characteristics of packaging samples, and in most of the cases,
lacking any sample pretreatment or further preparation, allowing
the analyses to be carried out in situ and helpful correlations with
traditional materials characterizations to be established, as
described above [18]. X-ray micro-computed tomography
(Micro-CT) is a powerful, versatile, noninvasive, and nondestruc-
tive technique used for the three-dimensional (3D) structural char-
acterization of varying types of systems [19–37]. This tool enables
correlations between the microstructural and macroscopic charac-
teristics of different types of materials as aforementioned, including
food packaging [38–40], enabling quantifying and analyzing dis-
persions, second phases, and pores throughout the system [41]. To
elucidate the Micro-CT procedure, Fig. 1 shows a schematic illus-
tration of Tetra Pak® packaging projection acquisition, followed by
pre-processing and processing of Tetra Pak® tomography imagens.
Concerning a brief overview of the X-ray microtomography
procedure, Micro-CT is based on the attenuation of X-rays accord-
ing to the Lambert–Beer law of exponential decay [42]. The X-rays
pass through the sample (Fig. 1a) and they are attenuated accord-
ing to the thickness, composition, and density of the analyzed
material [43]. In addition, the attenuated X-rays produce a
two-dimensional (2D) shadow of the material in the detector,
producing several 2D X-ray projections of the object at each rota-
tion step of the sample around its height Z-axis (Fig. 1b). The
detector (Fig. 1a) and the X-ray source are strategically positioned
in relation to the examined material. The source is located opposite
to the detector, considering the Z-axis of the sample as reference
[44]. The detector includes a charge-coupled device (CCD), a
photodetector camera that transforms the attenuated X-ray
photons into projections (Fig. 1a) [45, 46]. The projections
(Fig. 1b) are aligned in the formation of sinograms (Fig. 1c),
which are the origin of computational reconstruction signal of
tomographic images (Fig. 1d) [47]. In terms of materials differen-
tiation, from a high-quality reconstructed image, it is already possi-
ble to observe the layers that compose the Tetra Pak® packaging.
Fig. 1 Schematic illustration of the basic principles of the Micro-CT technique and data processing: (a)
Lambert–Beer’s law, X-ray source and opposite detector to collect the signal; (b) Images projections from Z-
axis 360° rotation; (c) Sinograms; (d) Two-dimensional slice; (e) High-quality reconstructed images of Tetra
Pak® packaging and segmentation of packaging composition: paperboard, polyethylene (PE), and aluminum
(Al) layers; and (f) Tomographic resolution
The lower attenuation for the paperboard and the polyethylene

(PE) layers and the higher attenuation for the Al layer according
to the colors of the X-ray attenuation scale can be noted (Fig. 1d,
color-scale arrow). Considering the Tetra Pak® packaging 3D ren-
dering (Fig. 1e), the compatibility of the interfaces among the
layers and to analyze each 3D segmented layer can be noticed,
assisting manufacturing engineers to minimize flaws/damage in
the packaging of food products.
In the food industry, Micro-CT has risen as a valuable tool to
correlate the common food characteristics (e.g., flavor, texture,
ripening) with processing and storage conditions [48–52]. A 3D
reconstruction of cream cheese from Micro-CT was applied to
comprehend the influence of fat content on microstructures of
cheese spreads [53]. Fat content was also correlated to the texture
(hardness) of pork lean fermented sausages by Micro-CT, and
geometric parameters obtained from this technique [54]. Cellular
structure and porosity profile obtained from Micro-CT showed
dependency of different baking conditions on bread [49], bread
crumbs and crust [55]. Recently, in situ Micro-CT has been devel-
oped, setting new approaches and giving the possibility to observe
the time-lapse collapse of micro-bubbles in the beer foams [18].
However, to date, the studies on Micro-CT as a nondestructive

characterization tool for food packaging (macro and microstruc-
tures, defects, and failures) are scarce. Herein, we fully describe a
useful step-by-step protocol of extruded PE composite characteri-
zation by using Micro-CT, containing tips and examples to identify
interface regions, failures, defects, and imperfections in food pack-
aging from tomography images.
2 Materials
The follow materials, equipment, and software were used for

tomography images illustration on introduction and methods.
1. Commercial Tetra Pak® packaging and extruded low-density
PE (LDPE) (Braskem, MFI = 8.3 g/10 min at 190 °C/
2.160 kg) composite reinforced with cellulose fibers from sug-
arcane bagasse were used as materials models for Micro-CT
characterization.
2. Tetra Pak® and extruded PE composite packaging were ana-
lyzed on a high-resolution Micro-CT equipment (Bruker
1272, Kontich, Belgium).
3. The acquisition was carried out using the SkyScan 1272 soft-
ware provided by Bruker Micro-CT®.
4. The tomography images reconstruction was performed using
Feldkamp algorithm by NRecon software (version 1.6.9.8,
Bruker Micro-CT®).
5. The segmentation and extraction data from the reconstructed
tomographic image were performed on CT analyzer (CTAn®)
software (version 1.14.4.1, Bruker Micro-CT®).
6. The DataViewer software (version 1.5.1.2, Bruker Micro-
CT®) was used to view the 2D images slice by slice, including
the orthoslices of the reconstructed images and a linear X-ray
attenuation.
7. 3D tomographic images were designed, rotated, and cropped
on CTvox® software (version 2.7.0, Bruker Micro-CT®).
3 Methods
The Micro-CT protocol proposed here relies upon the real analysis
of extruded PE composite, but one may extend it to other samples.
The Micro-CT protocol was performed according to the following
steps:
1. Place the sample symmetrically at the rotation stage, within the

detector’s field of view varying the sample rotation by 360° (see
Notes 1–5).
2. Set up the Micro-CT according to the density of the material.
For tomographic images in low-density materials (e.g., paper
and polymers), low-energy and “unfiltered” or fine Al filters
can be used (Table 1). To transmit X-ray radiation through
denser materials (e.g., ceramics, metals, and glass), Al/Cu or
Cu filters with associated higher X-ray source energies should
be used.
3. Obtain images using a Micro-CT; we performed these analyses
with a high-resolution Micro-CT equipment (see Note 6).
4. Pre-processing: Process the images using the NRecon software
to apply the artifact’s corrections, such as alignment, noise,
ring, etc. (Fig. 2). In the case of PE composite packaging
X-ray projections (Fig. 2a), it is recommended that the gray-
scale range should be delimited on X-ray histogram to guaran-
tee a faithful reconstruction image of the object. The example
illustrated in Fig. 2b shows the following tomographic images
problems: x/y drift and/or misalignment, noises, and ringing
artifacts. Figure 2c illustrates the automatic alignment to cor-
rect the x/y drift and/or slight deviation due to misalignment
of the X-ray projections. Following the tomographic image
problem corrections, the noises are attenuated by the smooth-
ing tool, as shown in Fig. 2d. Finally, ringing artifacts are
corrected, extracting a reliable reconstruction image, as pre-
sented in Fig. 2e. The final image is generated by applying a
color-scale (left side presents lower X-ray attenuation), as
shown in Fig. 2f (see Note 7).
5. Processing: Open the reconstructed image and select several
regions of interest (ROIs), previously preset, throughout the
object using the CT analyzer (CTAn) software. For a specific
ROI, establish the number of slices, thereby rendering a vol-
ume of interest (VOI), measuring in this case
1.5 × 1.5 × 1.5 mm3 (Fig. 3a, right). Using the binary selection
command, a preset thresholding segmentation is applied on
X-ray histogram, partitioning the analyzed object image by
grayscale intensity [56], rendering different segmented images,
containing pores/defects, the main matrix and microstructures
agglomerations (Fig. 3b). It is important to delimit on the
X-ray histogram for a specific grayscale range for all samples,
as represented in Fig. 3c, where VOIs must present compara-
tive X-ray attenuation and data extracted. Cleary, after this data
processing defects (pores or matrix imperfections like filler
agglomeration) can be identified in the analyzed packaging, as
shown in Fig. 3b. After application of the previous
Table 1
Detailed information on the Micro-CT parameters for Tetra Pak® and
extruded polyethylene (PE) composite analyses
Parameters Tetra Pak® PE composite

Source voltage (kV) 20 30
Source current (μA) 175 212
Detector resolution (pixels) 2452 × 1640 2016 × 1344
Nominal resolution (μm) 2 1.4
Exposure (ms) 5340 2300
Rotation step (deg) 0.3 0.15
Frame number 4 3
Random movement 30 10
Filter No No
segmentation, quantitative segmentation data can be obtained

as an outstanding and important morphometric analysis for a
great number of images, which is very laborious to obtain via
light-based microscopes. It can be done slice by slice
(2D object) or for the entire VOI (3D object). In this case,
the PE composite packaging presents a heterogeneous phase
distribution along the slices, as illustrated in Fig. 3d (see Notes
8 and 9).
6. Post-processing: Use the DataViewer software to analyze the 2D
images on the 3D perspective, as shown in Fig. 4. In this case,
three different 2D-slice views are combined, such as 2D Z–X
(coronal), 2D Z–Y (sagittal), and 2D X–Y (transversal) on the
3D object. From this perspective, it is possible to observe in
detail the pores/defects (black phase), polymer matrix (red
phase), dispersed microstructures/fibers (orange phase), and
aggregates (white phase) contained in the PE composite for
each 2D image of XYZ axis. Each 2D slice can be analyzed
using a linear X-ray attenuation distribution, discriminating
from lower to higher attenuation, according to pores/defects,
matrix, dispersed microstructures/fibers, and aggregates incre-
ment, respectively (Fig. 4, bottom).
7. Imaging and Analysis: Regarding the representation of Micro-
CT data in terms of 3D analysis of the reconstructed images
and generated VOIs, different 3D geometries, colors and XYZ
rotations can be performed using the CTvox® software. Some
possible tomographic images are illustrated in Fig. 5. Consid-
ering a VOI of PE composite packaging reconstructed from
X-ray projections, this volume can be cut, rotated, colored, and
Fig. 2 (a) X-ray projection and attenuation histogram; (b) Reconstruction tomographic imaging; workflow with
the data correction: (c) alignment; (d) noise, and (e) ring. (f) Corrected image for the PE composite packaging
segmented using the simple CTvox® tools, as shown in Fig. 5.

These tools allow viewing the interior and exterior of the VOI
in different perspectives and color gradients. Furthermore, the
color-scale segmentation (Fig. 5b) shows the 3D dispersion of
the microstructures/fibers (green), of the agglomerates (light
blue) along the PE matrix (brown), permitting to map the 3D
microstructure. Note that those imperfections (agglomeration,
in this case) within the PE matrix observed in orthoslices
projection (Fig. 4) is trackable in most 3D maps (a white region
in 3D-VOI, Fig. 5a, c) (see Note 10).
Fig. 3 The Micro-CT analyzer software functional tools to segmentate a tomographic image. (a) Region of
interest (ROI) selection (left) from the PE composite packaging reconstructed image to obtain a volume of
interest (VOI) rendering (right). (b) Thresholding segmentation of PE composite in pores/defects (green),
polymeric matrix (orange), and agglomerations (dark blue). (c) X-ray histogram partitioning the analyzed
object image by grayscale intensity. (d) Quantitative morphometric data extracted from thresholding segmen-
tation (from lower to higher attenuation) of a VOI: pores/defects, polymeric matrix and agglomerations content
(%) variation in relation to the number of 2D slices
Fig. 4 The orthoslices of polyethylene (PE) composite and the linear X-ray attenuation (green line) varying
according to PE composite phases obtained and processed in the DataViewer® software
Fig. 5 Polyethylene (PE) composite packaging VOI edited in the CTvox® software by different functional tools:
(a) rotation, (b) segmentation, (c) coloration, and (d) cutting
4 Conclusion
The food packaging industry has shown technological advances in

terms of renewable precursors, physical-mechanical performance,
and innovative functions (check Part III for details on nontradi-
tional functions of packaging materials). Nevertheless, this progress
is limited by the occurrence of flaws/damage in the packaging
material itself, decreasing or failing their purpose of ensuring food
shelf-life. In this chapter, an innovative and useful noninvasive
technique is proposed to investigate and point out 2D/3D damage
maps in packaging, correlating morphometric data to possible
causes (e.g., production, storage) and their influence on packaging
properties. Two different packaging compositions were analyzed by
Micro-CT, as showcases for the sake of didactics. Different types of
data treatment (data acquisition, image processing) were consid-
ered in our protocol. This procedure provides as a helpful and
nonconventional technique for predicting and manipulating
end-use properties of food packaging.
5 Notes
The protocol described here was developed for the Bruker system
available to users as an open facility at The Brazilian Nanotechnol-
ogy National Laboratory (LNNano)/Brazilian Center of Research
in Energy and Materials (CNPEM). It can be, however, extended
to other manufacturers, considering the specificities of each appa-
ratus. Should the reader have any query on calls for users, they are
invited to contact the corresponding author or edu@cnpem.br.
1. Samples usually do not require any preparation prior to Micro-
CT analysis.
2. The best shape for samples is a cylinder (symmetrical rotation),
improving the resolution due to the effective reduction in the
diameter and alignment problems.
3. Sample holder with tape or glue can be used to ensure that the
sample does not move during the data acquisition, keeping the
effectiveness of 3D reconstructions procedure. The projections
must be well aligned to ensure the high quality of the three-
dimensional image of the object. It is worth mentioning that,
in some cases, it is also necessary to minimize the noise and
artifacts generated in the characterization process, using certain
tools to not affect the quality of the reconstructed image
[47, 57].
4. Small samples are preferred for enabling better resolution and
less beam-hardening artifact.
5. The partial width option can be used for objects much smaller
than the camera field, providing shorter scan times due to
bigger rotation step and shorter reconstruction due to smaller
datasets. This option is recommended for 4k binning mode
(detector).
6. Other X-ray tomography approaches exist for other different
scales and purposes, such as the computed tomography
(CT) applied to the medical field for macroscale analysis [44]
and the nano-computed tomography (Nano-CT) coupled to
the synchrotron beamline to analyze sub-micrometer scale in
higher resolution (see scale resolution on Fig. 1) [58].
7. The oversize scanning can be applied when the object is larger
than a scan area, making a scout view and appropriated selec-
tion of the area of interest. The NRecon® software recognizes
the object as a vertically connected dataset.
8. Some plug-ins can be applied in the CTAn® software to opti-
mize the quantitative data and time acquisition, such as custom
processing tool, enabling to create a task list for numerous
parameters and samples.
9. Regarding an advanced characterization it is possible to select
irregular surfaces in packaging (like the example of the Fig. 2)
using the ROI shrink-wrap tool at the custom processing
(CTAn® software), creating a VOI rendering.
10. The CTvox® software allows producing movies/animations
using the flight recorder tools by adding different frames,
shapes, rotation, and movements, some animations from
CTvox® software are available on the Internet (https://
dentistry.llu.edu/micro-ct-imaging-gallery).
Acknowledgments
The authors thank The Brazilian National Council for Scientific and
Technological Development (CNPq) for supporting this work
DTI-A/SisNANO (grants 380312/2020-4 and 380173/2022-
0), PCI-postdoctoral (grants 301362/2020-3, 301356/2020-
3 and 302368/2021-3) and Productivity scholarship (grant
303621/2022-2), and the grant 2020/08651-4 provided by São
Paulo Research Foundation (FAPESP).
References
1. Risch SJ (2009) Food packaging history and Food Sci 72:R39–R55. https://doi.org/10.
innovations. J Agric Food Chem 57:8089– 1111/j.1750-3841.2007.00301.x
8092. https://doi.org/10.1021/jf900040r 3. Robertson GL (2016) Food packaging. CRC
2. Marsh K, Bugusu B (2007) Food packaging: Press
roles, materials, and environmental issues. J
4. Knorr D, Froehling A, Jaeger H et al (2011) https://doi.org/10.1016/j.matchar.2010.

Emerging technologies in food processing. 09.012
Annu Rev Food Sci Technol 2:203–235. 17. Claro PIC, Borges EPBS, Schleder GR et al
https://doi.org/10.1146/annurev.food. (2023) From micro- to nano- and time-
102308.124129 resolved x-ray computed tomography:
5. Parisi S (2012) Food packaging and food Bio-based applications, synchrotron capabil-
alterations: the user-oriented approach, 1st ites, and data-driven processing. Appl Phys
edn. Smithers Rapra Technology Ltd, Shaw- Rev 10:021302. https://doi.org/10.1063/5.
bury, Shrewsbury, Shropshire 0129324
6. Onwurah C, Mamah SC (2018) A method for 18. Dewanckele J, Boone MA, Coppens F et al
damage detection in the packaging materials. J (2020) Innovations in laboratory-based
Sci Eng Res 5:235–249 dynamic micro-CT to accelerate in situ
7. Paine FA, Paine HY (1992) A handbook of research. J Microsc 277:197–209. https://
food packaging. Springer US, Boston doi.org/10.1111/jmi.12879
8. Trajkovska Petkoska A, Daniloski D, D’Cunha 19. Stock SR (2008) Recent advances in X-ray
NM et al (2021) Edible packaging: sustainable microtomography applied to materials. Int
solutions and novel trends in food packaging. Mater Rev 53:129–181. https://doi.org/10.
Food Res Int 140:109981. https://doi.org/ 1179/174328008X277803
10.1016/j.foodres.2020.109981 20. Lorevice MV, Mendonça EO, Orra NM et al
9. Sharma R, Jafari SM, Sharma S (2020) Antimi- (2020) Porous cellulose nanofibril–natural
crobial bio-nanocomposites and their potential rubber latex composite foams for oil and
applications in food packaging. Food Control organic solvent absorption. ACS Appl Nano
112:107086. https://doi.org/10.1016/j. Mater 3(11):10954–10965. https://doi.org/
foodcont.2020.107086 10.1021/acsanm.0c02203
10. Bagde P (2021) Biodegradable polymers for 21. de Siqueira L, Gouveia RF, Grenho L et al
food packaging and active food packaging. In: (2018) Highly porous 45S5 bioglass-derived
Advances in the domain of environmental bioglass – ceramic scaffolds by gelcasting of
technology. Springer, Singapore, pp 113–127 foams. J Mater Sci 53:10718–10731. https://
11. Parisi S (2013) Food industry and packaging doi.org/10.1007/s10853-018-2337-x
materials: performance-oriented guidelines for 22. Ferreira FV, Souza LP, Martins TM et al
users. Smithers Rapra Technology Ltd, Shaw- (2019) Nanocellulose/bioactive glass cryogels
bury, Shrewsbury, Shropshire as scaffolds for bone regeneration. Nanoscale
12. Otoni CG, Avena-Bustillos RJ, Azeredo HMC 11:19842–19849. https://doi.org/10.1039/
et al (2017) Recent advances on edible films c9nr05383b
based on fruits and vegetables—a review. 23. Garcia PS, Gouveia RF, Maia JM et al (2018)
Compr Rev Food Sci Food Saf 16:1151. 2D and 3D imaging of the deformation behav-
https://doi.org/10.1111/1541-4337.12281 ior of partially devulcanized rubber/polypro-
13. Han J-W, Ruiz-Garcia L, Qian J-P, Yang X-T pylene blends. Express Polym Lett 12:1047–
(2018) Food packaging: a comprehensive 1 0 6 0 . h t t p s : // d o i . o r g / 1 0 . 3 1 4 4 /
review and future trends. Compr Rev Food expresspolymlett.2018.92
Sci Food Saf 17:860–877. https://doi.org/ 24. Pandoli OG, Neto RJ, Oliveira NR et al (2020)
10.1111/1541-4337.12343 Ultra-highly conductive hollow channels
14. Sarfraz J, Gulin-Sarfraz T, Nilsen-Nygaard J, guided by a bamboo bio-template for electric
Pettersen MK (2020) Nanocomposites for and electrochemical devices. J Mater Chem A
food packaging applications: an overview. 8:4030–4039. https://doi.org/10.1039/
Nanomaterials 11:10. https://doi.org/10. c9ta13069a
3390/nano11010010 25. Zhou G, Pei S, Li L et al (2014) A graphene –
15. Ozguler A, Morris SA, O’Brien WD (1999) pure-sulfur sandwich structure for ultrafast,
Evaluation of defects in the seal region of long-life lithium – sulfur batteries. Adv Mater
food packages using the ultrasonic contrast 26:625–631. https://doi.org/10.1002/
descriptor, ΔBAI. Packag Technol Sci 12: adma.201302877
161–171. https://doi.org/10.1002/(SICI) 26. Laperre K, Depypere M, van Gastel N et al
1099-1522(199907/08)12:4<161::AID- (2011) Development of micro-CT protocols
PTS464>3.0.CO;2-C for in vivo follow-up of mouse bone architec-
16. Landis EN, Keane DT (2010) X-ray microto- ture without major radiation side effects. Bone
mography. Mater Charact 61:1305–1316. 49:613–622. https://doi.org/10.1016/j.
bone.2011.06.031
27. Bultreys T, De Boever W, Cnudde V (2016) aerogels for water remediation. J Chem Eng
Imaging and image-based fluid transport mod- 462:142166. https://doi.org/10.1016/j.cej.
eling at the pore scale in geological materials: a 2023.142166
practical introduction to the current state-of- 37. Silva DB, Nascimento DM, Claro PIC et al
the-art. Earth Sci Rev 155:93–128. https:// Enhancing water resistance in cationic cellulose
doi.org/10.1016/j.earscirev.2016.02.001 nanofibril adhesive with natural rubber latex.
28. Padilla-ortega E, Darder M, Aranda P et al ACS Appl Nano Mater. https://doi.org/10.
(2016) Ultrasound assisted preparation of 1021/acsanm.3c04257
chitosan – vermiculite bionanocomposite 38. Islam SF, Mancini L, Sundara RV et al (2017)
foams for cadmium uptake. Appl Clay Sci Studying model suspensions using high resolu-
130:40–49. https://doi.org/10.1016/j.clay. tion synchrotron X-ray microtomography.
2015.11.024 Chem Eng Res Des 117:756–772. https://
29. Moreira RLPO, Simão JA, Gouveia RF, Strauss doi.org/10.1016/j.cherd.2016.11.034
M (2020) Exploring the hierarchical structure 39. du Plessis A, le Roux SG, Tshibalanganda M
and alignment of wood cellulose fibers for (2019) Advancing X-ray micro computed
bioinspired anisotropic polymeric composites. tomography in Africa: going far, together. Sci
ACS Appl Bio Mater 3(4):2193–2200. Afr 3:e00061. https://doi.org/10.1016/j.
https://doi.org/10.1021/acsabm.0c00038 sciaf.2019.e00061
30. Ferreira FV, Mariano M, Rabelo SC et al 40. Rouholamin D, Hopkinson N (2014) An
(2018) Isolation and surface modification of investigation on the suitability of micro-
cellulose nanocrystals from sugarcane bagasse computed tomography as a non-destructive
waste: from a micro- to a nano-scale view. Appl technique to assess the morphology of laser
Surf Sci 436:1113–1122. https://doi.org/10. sintered nylon 12 parts. Proc Inst Mech Eng
1016/j.apsusc.2017.12.137 Part B J Eng Manuf 228:1529–1542. https://
31. Ferreira FV, Trindade GN, Lona LMF et al doi.org/10.1177/0954405414522209
(2019) LDPE-based composites reinforced 41. Stock SR (1999) X-ray microtomography of
with surface modified cellulose fibres: 3D mor- materials. Int Mater Rev 44:141–164.
phological and morphometrical analyses to h t t p s : // d o i . o r g / 1 0 . 1 1 7 9 /
understand the improved mechanical perfor- 095066099101528261
mance. Eur Polym J 117:105–113. https:// 42. Maire E, Buffière JY, Salvo L et al (2001) On
doi.org/10.1016/j.eurpolymj.2019.05.005 the application of X-ray microtomography in
32. Canencia F, Darder M, Aranda P et al (2017) the field of materials science. Adv Eng Mater
Conducting macroporous carbon foams 3:539–546. https://doi.org/10.1002/1527-
derived from microwave-generated caramel/ 2648(200108)3:8<539::AID-
silica gel intermediates. J Mater Sci 52: ADEM539>3.0.CO;2-6
11269–11281. https://doi.org/10.1007/ 43. Vásárhelyi L, Kónya Z, Kukovecz A, Vajtai R
s10853-017-1227-y (2020) Microcomputed tomography–based
33. Lopes JH, Magalhães JA, Gouveia RF et al characterization of advanced materials: a
(2016) Hierarchical structures of β-TCP/ review. Mater Today Adv 8:100084. https://
45S5 bioglass hybrid scaffolds prepared by gel- doi.org/10.1016/j.mtadv.2020.100084
casting. J Mech Behav Biomed Mater 62:10– 44. Eberhardt CN, Clarke AR (2002) Automated
23. https://doi.org/10.1016/j.jmbbm.2016. reconstruction of curvilinear fibres from 3D
04.028 datasets acquired by X-ray microtomography.
34. Damasceno S, Corrêa CC, Gouveia RF et al J Microsc 206:41–53. https://doi.org/10.
(2020) Delayed capillary flow of elastomers: 1046/j.1365-2818.2002.01009.x
an efficient method for fabrication and nano- 45. Ning R, Kruger RA (1998) Computer simula-
functionalization of flexible, foldable, twista- tion of image intensifier based computed
ble, and stretchable electrodes from pyrolyzed tomography detector: vascular application.
paper. Adv Electron Mater 6:1900826. Med Phys 188:188–192. https://doi.org/10.
https://doi.org/10.1002/aelm.201900826 1118/1.596250
35. Mariano M, Souza SF, Borges AC et al (2021) 46. Karellas A, Harris LJ, Liu H et al (1992)
Tailoring strength of nanocellulose foams by Charge coupled device detector: performance
electrostatic complexation. Carbohydr Polym considerations and potential for small field
256:117547. https://doi.org/10.1016/j. mammographic imaging applications. Med
carbpol.2020.117547 Phys 19:1015–1023. https://doi.org/10.
36. Lorevice MV, Claro PIC, Aleixo NA et al 1118/1.596819
(2023) Designing 3D fractal morphology of
eco-friendly nanocellulose-based composite
47. Berg S, Saxena N, Shaik M, Pradhan C (2018) https://doi.org/10.1111/j.1365-2621.2007.

Generation of ground truth images to validate 01514.x
micro-CT image-processing pipelines. Lead 53. Laverse J, Mastromatteo M, Frisullo P, Del
Edge 37:412–420. https://doi.org/10. Nobile MA (2011) X-ray microtomography
1190/tle37060412.1 to study the microstructure of cream cheese-
48. Haff RP, Toyofuku N (2008) X-ray detection type products. J Dairy Sci 94:43–50. https://
of defects and contaminants in the food indus- doi.org/10.3168/jds.2010-3524
try. Sens & Instrumen Food Qual 2:262–273. 54. Santos-Garcés E, Laverse J, Gou P et al (2013)
https://doi.org/10.1007/s11694-008- Feasibility of X-ray microcomputed tomogra-
9059-8 phy for microstructure analysis and its relation-
49. Cantre D, Herremans E, Verboven P et al ship with hardness in non-acid lean fermented
(2014) Characterization of the 3-D micro- sausages. Meat Sci 93:639–644. https://doi.
structure of mango (Mangifera indica L. cv. org/10.1016/j.meatsci.2012.11.027
Carabao) during ripening using X-ray com- 55. Besbes E, Jury V, Monteau J-Y, Le Bail A
puted microtomography. Innov Food Sci (2013) Characterizing the cellular structure of
Emerg Technol 24:28–39. https://doi.org/ bread crumb and crust as affected by heating
10.1016/j.ifset.2013.12.008 rate using X-ray microtomography. J Food Eng
50. Schoeman L, Williams P, du Plessis A, Manley 115:415–423. https://doi.org/10.1016/j.
M (2016) X-ray micro-computed tomography jfoodeng.2012.10.005
(μCT) for non-destructive characterisation of 56. Iassonov P, Gebrenegus T, Tuller M (2009)
food microstructure. Trends Food Sci Technol Segmentation of X-ray computed tomography
47:10–24. https://doi.org/10.1016/j.tifs. images of porous materials: a crucial step for
2015.10.016 characterization and quantitative analysis of
51. Mulot V, Fatou-toutie N, Benkhelifa H et al pore structures. Water Resour Res 45:1–12.
(2019) Investigating the effect of freezing https://doi.org/10.1029/2009WR008087
operating conditions on microstructure of fro- 57. Zhang Y, Yu H (2018) Convolutional neural
zen minced beef using an innovative X-ray network based metal artifact reduction in X-ray
micro-computed tomography method. J Food computed tomography. IEEE Trans Med
Eng 262:13–21. https://doi.org/10.1016/j. Imaging 37:1370–1381. https://doi.org/10.
jfoodeng.2019.05.014 1109/TMI.2018.2823083
52. Mousavi R, Miri T, Cox PW, Fryer PJ (2007) 58. Withers PJ (2007) X-ray nanotomography.
Imaging food freezing using X-ray microtomo- Mater Today 10:26–34. https://doi.org/10.
graphy. Int J Food Sci Technol 42:714–727. 1016/S1369-7021(07)70305-X
Chapter 10
Mapping the Distribution of Additives Within Polymer Films

Through Near-Infrared Spectroscopy and Hyperspectral
Imaging
Jussara V. Roque, Cı́cero C. Pola, Larissa R. Terra, Taı́la V. Oliveira,
Reinaldo F. Teófilo, Carmen L. Gomes, and Nilda F. F. Soares
Abstract
Near-infrared (NIR) spectroscopy and hyperspectral imaging allow the study of spectral and spatial
distribution of multiple chemical components in large sample areas. This technique is fast,
non-destructive, contactless, and does not require sample preparation. The NIR spectrum of each sample
pixel is acquired, resulting in a data cube that contains two spatial dimensions (x and y) and one spectral
dimension (z), providing the spectral profiles of every part of the sample. This technique, for example, can
provide significant information about the distribution of additives into polymer matrices with potential to
be used as a tool for real-time quality control. Herein, the stepwise application of this method is demon-
strated for determination of spatial and spectral distributions of film components, showcasing the plastici-
zation of a biodegradable packaging.
Key words NIR-HSI, MCR-ALS, Chemical distribution, Chemical mapping, Macropixel, Homoge-
neity index, Polymer characterization
1 Introduction
Pristine polymers often show poor processability, physicochemical

properties, and performance, limiting most of their applications in
packaging. Such limitations are even more prominent when biopo-
lymers are used. Additives, polymer blends, and composites are
some of the most common strategies to overcome these technolog-
ical hurdles [1–3]. The incorporation of additives, such as fillers,
plasticizers, antistatics, stabilizers, and colorants into polymeric
materials may improve their processability and tune their proper-
ties, making them suitable for packaging production [4–6].
Mechanical strength, gas barrier (check Part II for packaging as
barrier materials), and thermal resistance are some of the properties
that can be drastically affected by the incorporation of additives
https://doi.org/10.1007/978-1-0716-3613-8_10,
183
184 Jussara V. Roque et al.
[3]. Moreover, active properties can be acquired by the material

with the incorporation of a specific class of additives, such as anti-
microbial and antioxidant compounds [7] (check Part III for details
on active roles played by food packaging). However, in all cases, the
distribution and dispersion of the additive within the polymer
matrix are critical for reaching the desired properties. Miscibility
and aggregation problems can compromise the distribution of
added compounds into the host polymer causing phase separation
over time, which leads to undesirable changes and diffusion of the
additive toward the material surface [8, 9]. Hence, knowing how
the additive is distributed throughout the material becomes as
important as its incorporation. Several high-sensitivity techniques
have become available to study additives in polymeric matrices [9–
13]. However, most of these techniques are costly, limited to small
sampling areas, and require long sampling times and laborious
sample preparations. Consequently, these are often unsuitable for
real-time analysis [9, 10].
One way to simultaneously obtain spectral and spatial informa-
tion is to use near-infrared spectroscopy (NIR) associated with
hyperspectral imaging (HSI). NIR-HSI is a fast, contactless, non-
destructive, and relatively low-cost technique (see Note 1) that does
not require sample preparation and does not generate chemical
waste [9, 14]. This technique provides chemical and spatial infor-
mation on multiple chemical components in large sampling areas,
even when in movement, making it suitable for real-time analysis
[15–18].
NIR-HSI divides the entire sample area into individual pixels.
For each pixel an NIR spectrum is acquired, which results in a data
cube, containing multiple spatially co-registered spectra, with two
spatial dimensions (x and y) and one spectral dimension (z),
providing the spectral profiles of every part of the sample [9, 14,
16, 18, 19]. Hence, each pixel spectrum can be used to predict a
map of the sample features (physical, chemical, or categorical) at the
corresponding spatial location [20, 21]. Chemometric tools using
multivariate data analysis play a fundamental role in extracting,
treating, and displaying NIR-HSI information from the acquired
data [16, 22]. Specifically, for NIR-HSI data, Multivariate Curve
Resolution—Alternating Least Square (MCR-ALS) is one of the
best choices for extracting chemical and spatial information. MCR
is a bilinear model suitable for solving compound mixtures, in
which the curve resolution method assumes that the observed
spectra are linear combinations of the pure component spectra
[22, 23]. The number of components in the mixture is determined
or given by previous knowledge, then an estimation of the concen-
tration and/or spectral profiles for each component is obtained.
The ALS algorithm finds the optimum convergence among the
components using different constraints (e.g., non-negativity,
NIR Mapping of Multicomponent Packaging 185
unimodality, closure, and local rank) to obtain the optimum reso-

lution and improve interpretation [22].
In this context, NIR-HSI techniques have been applied in
several areas, such as agriculture [13, 24–26], pharmaceutics [27–
29], forensics [30–32], and food science [33–37]. However, stud-
ies involving NIR-HSI applications for packaging are still limited.
Amigo et al. [10] demonstrated the potential of this technique and
the steps required for the real-time detection of plastic materials
containing flame-retardant additives. Recently, Terra et al. [9]
determined the spatial distribution of four different plasticizers
and sorbic acid within cellulose acetate-based biodegradable films.
Herein, the stepwise application and suitability of this method are
demonstrated for the determination of spatial and spectral distribu-
tions of additives into a polymeric matrix with the potential to be
used for food packaging applications. A biodegradable cellulose
acetate-based film incorporated with glycerol as a plasticizer is
used to illustrate the application of this method.
2 Materials
2.1 Film Components NIR-HSI method can be applied to all sorts of polymeric films used
for food packaging applications [9]. Hence, the required materials
are solely dependent on the type of film that will be evaluated. The
use of high-quality grade chemicals is recommended. Each pristine
component of the film is also indicated to be measured as a standard
to provide the spectrum of the pure component required by the
MCR-ALS. In this protocol, cellulose acetate, glycerol, and acetone
are used for film production.
2.2 Instrumentation Several camera systems are available for HSI acquisition. Our pro-
tocol is relied on the SisuCHEMA (Specim®, Oulu, Finland) chem-
ical imaging system. SisuCHEMA is a complete chemical imaging
workstation that combines NIR spectroscopy, optimized for short
wave infrared (SWIR) (1000–2500 nm wavelength region), with
high-resolution imaging, to provide detailed information on the
chemical components, their quantities, and distributions within the
sample [38] (see Note 2). The system consists of a line-scan imag-
ing spectrometer equipped with a two-dimensional HgCdTe detec-
tor, which has 320 space channels and 256 spectral channels
covering the 928–2524 nm wavelength range with a spectral reso-
lution of approximately 6.3 nm [38].
2.3 Software The chemical image workstation is controlled using ChemaDAQ™

data acquisition software (Specim, Oulu, Finland), which allows the
user to export the acquired NIR-HSI hypercubes directly as a
MATLAB Data Cube file (.mat). NIR-HSI processing can be car-
ried out in several ways, which depends on the operator
preferences. Any programming language can be used for this pur-

pose and several algorithms and packages to perform this proces-
sing are available. Here, we propose the use of MATLAB (The
MathWorks, Co., Natick, MA, USA) as the software to process
the images and extract the chemical information. An algorithm
package for MATLAB (HSIAnalyzer) was developed to perform
NIR-HSI analysis. This package is available at https://github.com/
jussararoque/HSIAnalyzer and from the authors upon request.
3 Methods
3.1 Film Production A cellulose acetate film is used here as an example of the NIR-HIS
application. This film is produced by mixing 2.5 g of cellulose
acetate with 25 mL of acetone and 1.2 mL g–1 of glycerol for
24 h in a closed container at room temperature [9]. Then, this
film-forming formulation is poured and spread onto a glass plate
using a paint applicator and allowed to dry for 2 h at room temper-
ature (see Note 3). Finally, the films are peeled off the glass plates,
inspected for defects, and stored in vacuum-sealed polyethylene/
nylon bags until further use (see Notes 4 and 5). Before NIR spectra
acquisition, films are placed onto a polytetrafluoroethylene (PTFE)
plate and securely attached using an adhesive tape (see Note 6).
3.2 Image A HSI system is usually composed of a sample holder, a light unit, a
Acquisition light scattering device (spectrograph), a camera (detector), and a
computer equipped with the image acquisition software. Optical
radiation can interact with the sample through several mechanisms,
depending on the light source arrangement, the sample, and the
detector. The frequently observed mechanisms are transmittance,
transflectance, diffuse reflectance, interactance, and diffuse trans-
mittance [39–41], which are illustrated in Fig. 1.
Transmittance (Fig. 1) is usually possible in transparent sam-
ples, where the incident light is partially absorbed by the sample,
and the remaining light is detected without any scattering. In
transflectance (Fig. 1), the incident light goes through the sample,
reaches a reflective apparatus, and goes back through the sample,
increasing the optical path. For solid samples, the most common
mechanism is diffuse reflectance (Fig. 1). In this mechanism, the
radiation interacts with solid components of the sample, being
dispersed and absorbed by them, which changes the intensity of
the signal analyzed. In the interactance mechanism (Fig. 1), the
incident light presents a higher probability of strong interaction
with the sample, which leads to an emerging beam containing more
information about the actual sample composition. The diffuse
transmittance (Fig. 1) is the transmittance measurement applied
to the dense solid samples, in which the light is internally scattered
due to the sample’s long optical path [40].
Fig. 1 Examples of the interaction mechanisms between optical radiation and sample: transmittance;
transflectance; diffuse reflectance; interactance, and diffuse transmittance
Fig. 2 Acquisition modes of an HSI: point-by-point scanning, area scanning, line scanning, and single shot.
The scanning directions are shown by arrows, and the yellow areas show data acquired each time. Sample
film is shown on top of PTFE platform. Adapted from Wu and Sun [17] with permission from Elsevier
In film analysis, reflectance is the primary mechanism involved;

however, transflectance can also occur since the films are usually
thin and have a certain level of transparency [9]. Depending on the
mechanism applied, the attenuation of light when interacting with
the sample can be reflectance (R) or transmittance (T), which are
easily transformed into absorbance (log 1/R or log 1/T) for
chemometrics analysis. In NIR-HSI, a NIR spectrum is obtained
for each pixel in which the object is divided to obtain the hypercube
(x, y, z). There are four convenient methods to obtain an HSI,
based on the relative movement between the sample and the
detection unit: point-to-point scanning, line scanning, area scan-
ning, and single shot [17]. Figure 2 illustrates these acquisition
modes.
The line scanning (Fig. 2) is the most used in the literature. In
this case, spectra are obtained simultaneously from an entire line of
pixels in the sample, and a complete hypercube can be obtained by
moving the sample or the detector along the x-axis, similar to being
placed in conveyor belt systems, which is ideal for industrial
applications.
In the next section, the focus will be on the step-by-step

analysis of film components’ spatial distribution using the
NIR-HSI obtained by using a SisuCHEMA (Specim®) chemical
imaging system. The system employs an OLE15 lens with a
200 mm field of view resulting in a pixel size of 625 μm
(200 mm/320 spatial channels) in the x-direction (horizontal
direction) (see Note 7). Hence, if 100 frames per second are used,
the acquisition time of each frame is 0.01 s. Thus, to maintain the
pixel proportionality in the y-direction (vertical direction), the
distance of 625 μm must be traversed in 0.01 s. Therefore, the
scanning speed of 62.5 mm s–1 must be chosen. Always ensure that
the edges of the pixels are proportional by choosing the correct
scanning velocity. Then, the hypercubes obtained by this system are
256 images (one image per wavelength) composed of
625 μm × 625 μm pixels. The total number of pixels per image
changes according to the image size. The instrument itself performs
the calibration of incident light, obtained using white and dark
references.
For the study of spatial distribution of films with known com-
position, wherein a standard sample of each component is available,
a NIR-HSI of each pure component standard must be acquired
separately for further analysis by curve resolution methods (see
Note 8).
3.3 Data Once the NIR-HSI images are acquired, the hypercubes may be
Pretreatment imported to MATLAB software. The raw NIR-HSI is a hypercube
(x, y, z) composed of millions of data points (e.g., a film NIR-HSI
of 350 × 200 pixels operating at 256 wavelengths contains more
than 17 million data points). The xy plane is divided into regular
squares, called pixels, containing chemical information represented
by NIR spectra in the z dimension (Fig. 3). Such an amount of
Fig. 3 NIR-HSI data as a three-dimensional array (hypercube), and an unfolded

two-dimensional array
information requires significant storage space and may make the

data analysis more time-consuming. Thus, methods of data com-
pression are frequently applied to reduce the data amount. This
data reduction can be done both in the spectral or spatial
dimension [42].
The application of pretreatment techniques is important to
eliminate unwanted variations present in the raw data without
compromising the analytical information. Initially, a dark and
white correction is recommended and can be performed using a
dark reference image, obtained with the light source turned off, and
a white reference image obtained from a surface with maximum
reflectance. The SisuCHEMA system performs the light calibration
automatically, acquiring both white and dark references (see Note
9). After image collection, the image is usually reduced in the
spatial dimension (x, y) to require less storage space and reduce
data analysis time. In most cases, the use of the entire image is not
required, so regions of interest (ROIs) can be selected from the raw
image to be used in the data analysis. After acquiring the HSIAna-
lyzer toolbox, the following steps can be performed.
• Add the HSIAnalyzer to MATLAB set path.
• Load the SampExample.mat (NIR-HSI of cellulose acetate film
incorporated with glycerol):
≫ load (‘SampExample.mat’)
• Use plothsi and selroi functions to perform the ROIs selection.

First, the image must be visualized (Fig. 4a):
Fig. 4 NIR-HIS image processing. (a) Original NIR-HSI image with selected ROI; (b) Selected region of NIR-HSI
image; and (c) Binned NIR-HSI image
≫ plothsi (HSI)
where HSI is a NIR-HSI image (x, y, z).

• In figure plot, the zoom tool can be used to select the ROI.
Then, the following command can be used:
≫ HSIsel = selroi (HSI);
where HSIsel is a NIR-HSI image (x, y, z) with the ROI

selected (Fig. 4b).
Next, the removal of dead pixels, which are missing values in
some pixels caused, for instance, by a dysfunction of one of the
diodes in the detector array, is recommended since this is a common
issue in NIR-HSI images [42]. Dead pixels can be removed by
simple interpolation using the mean or median of the neighboring
pixels. The rmvdeadpx function can be used to perform the removal
of dead pixels.
• In the MATLAB environment
≫ HSIcorr = rmvdeadpx (HSIsel,type);
where HSIcorr is a NIR-HSI image (x, y, z) with dead pixels

removed and type can be the removal of dead pixels by “mean”
or “median” of the neighboring pixels.
Data binning is another technique frequently applied for spatial
dimension reduction. In this technique the original values or pixels
from a specific interval are replaced with a representative value of
the same interval. The original pixels are replaced by an average
value in our algorithm.
• In the MATLAB environment:
≫ HSIbin = binning (HSIcorr,bin);
where HSIbin is a NIR-HSI image (x, y, z) binned by bin size

defined by bin input (Fig. 4c).
Because NIR spectra usually present overlapping bands, che-
mometric methods are required to extract the chemical information
in NIR-HSI, which is contained in the spectral dimension. How-
ever, before applying these chemometric methods, it is important
to consider the use of spectral pretreatments to remove data
Fig. 5 Spectral pretreatments examples. (a) Spectra of all pixels without any
spectral pretreatment; (b) Spectra with spikes removed; (c) Spectra with MSC
smoothing; and (d) Spectra Savitzky-Golay smoothing (win = 11)
anomalies, correct baseline and scattering to enhance the spectral

characteristics.
Spikes are one of the most common anomalies that can mask
real information and could lead to misleading interpretation. Spikes
can be defined as a sudden and abrupt rise followed by a sharp
decline in the spectrum. Spikes removal can be done by pixel
comparison in the hypercube arrangement, as follows:
≫ HSIspi = rmvspike(HSIbin,win);
where HSIspi is a NIR-HSI image (x, y, z) without spikes

(Fig. 5b) and win (5) is the pixel window to do spike search.
Because the chemical information is present in a spectral
dimension, bilinear chemometric methods are usually applied.
Thus, the three-dimensional array must be unfolded to a bilinear
matrix D (x*y, z), according to Fig. 3. In this new arrangement, the
spectra from pixels are the rows and the wavelengths (z dimension)
are the columns. The unfolding can be performed by the following
command:
≫ D = unfoldhsi(HSIspi,samp);
where D is a matrix containing all pixels (x*y, z), and samp is

the number of samples vertically concatenated.
Several spectral correction methods can be applied to enhance
spectral characteristics [43]. Standard normal variate (SNV) and
multiplicative scatter correction (MSC) are frequently used to cor-
rect NIR light scattering. Smoothing methods are commonly used
to reduce instrumental noise. Baseline and slope offset are usually
corrected by using Savitzky-Golay derivatives, baseline corrections,
and detrend approaches. However, this process must be applied
carefully, without losing important chemical information. All the
spectral pretreatments must be applied in the two-dimensional
array. In the example presented here, MSC and Savitzky-Golay
smoothing are used on the D matrix through the commands pre-
sented below:
≫ Dmsc = msc(D,type);
where Dmsc is the matrix corrected (Fig. 5c) and type can be
the correction by “mean” or “median” of spectra. Here the
median spectrum is used as reference:
≫ Dsmoo = svtgl(Dmsc,win);
where Dsmoo is the corrected matrix (Fig. 5d) and win

(11) is the variables window to be smoothed. Once pretreatment
methods are performed, we can proceed to the image treatment.
After all the required pretreatment, the multivariate methods
must be applied to the D matrix to extract the main information
about the image components.
3.4 Image Treatment Multivariate calibration and curve resolution are the most common
methods used to quantify all the pixels’ constituents in hyperspec-
tral images. Partial least squares (PLS) are the most used method
from multivariate calibration. Initially, the calibration samples are
used to build a model, which is required to quantify the constitu-
ents of all pixels. MCR-ALS is a curve resolution method that can
directly quantify the constituents of all pixels in hyperspectral
images without calibration samples. These two methods are dis-
cussed in the next sections.
3.4.1 Partial Least Initially, a calibration set needs to be created. At least ten calibration
Squares (PLS) samples with known bulk concentrations of all components of
interest must be prepared. Then, each mean spectra of pretreated
and unfolded image (D matrix) must be calculated. Next, the mean
spectra of the entire calibration set and the respective bulk concen-
tration of each component are used to build the PLS model. The
PLS model is based on equation y = Xb, where y can be either a
vector with the concentration of a component or a Y response
matrix with the concentration of several components simulta-
neously [44]. The X matrix is composed of the mean spectra of
each calibration sample. The X and y variables can be mean-
centered to build a PLS model using leave-one-out cross-validation
to choose the number of latent variables. From the developed
models, a regression vector b is obtained and by multiplying an
unfold D matrix with spectra of all pixels by b (D*b), components’
concentration of all pixels can be predicted in new samples. The
concentration is a vector or matrix C (x*y, n), depending on the
number of components (n), which can be used to generate distri-
bution maps. This step will be further explained in
Subsection 3.4.2.
Additionally, a small validation set (e.g., five samples) can be
prepared for external validation of the PLS model. Moreover, vari-
able selection methods can be used to enhance the PLS model and
to reduce the spectral dimension (z). Genetic algorithm (GA) [45],
ordered predictors selection (OPS) [46, 47], interval partial least
squares (iPLS) [48], and successive projection algorithm (SPA)
[49] are variable selection methods described in the literature to
improve PLS model predictions.
Notably, for the current example illustrating the use of
NIR-HSI it is not possible to perform PLS, since there is only
one sample. However, an algorithm package for MATLAB to
build PLS (PLSpack) is also available upon request, as mentioned
previously.
3.4.2 Multivariate Curve The MCR-ALS method is based on the bilinear equation
Resolution with Alternating D = CST + E, where D is the unfolded matrix (x*y, z), C is the
Least Squares (MCR-ALS) concentrations’ matrix, S is the pure components’ matrix, and E is
the error matrix. In order to find the C matrix and build distribu-
tion maps, MCR-ALS requires an initial estimation of S to itera-
tively solve the equations C = DS(STS)-1 and ST = (CTC)-1CTD.
This process is performed until a convergence criterion is achieved,
which can be defined by the maximum number of interactions or
when the difference between results of consecutive iterations is
lower than a predefined value [50].
During the optimization of the MCR-ALS method, the appli-
cation of constraints is highly recommended due to its rotational
ambiguity, that is, more than one response can be found in a
resolution of the bilinear equation. The commonly applied
constraints are non-negativity, unimodality, closure, and local rank

information [50].
MCR-ALS can be applied directly to the samples of interest,
without requiring a previous calibration. However, it is necessary to
know the pure spectra of each component of interest present in the
sample. Such information is critical to find the C matrix and to
build the distribution maps. The MCR-ALS 2.0 toolbox is available
at https://mcrals.wordpress.com/download/mcr-als-2-0-tool
box/ [50].
The sample used as an example here is a thin film placed over a
PTFE support for NIR spectrum acquisition. The NIR radiation
crossed through the sample capturing information of the thin film
and the PTFE support. Therefore, PTFE spectra were also obtained
to be used as pure spectra in MCR-ALS resolution as it is consid-
ered a component of the film NIR image.
The pure spectra are used for initial estimation in MCR-ALS
(S matrix) and are also incorporated into the D matrix (vertically
concatenated).
In the MATLAB environment:
• Add the MCR-ALS 2.0 toolbox in MATLAB set path.
• Load the StandExample.mat (containing NIR-HSI images of all
known individual components, called standards):
≫ load(‘StandExample.mat’)
Sometimes, it is not possible to obtain the standard NIR-HSI

images. In this case, the pretreated D matrix can be used directly in
MCR-ALS and the initial estimation can be done using the
MCR-ALS 2.0 toolbox. Furthermore, the use of column-wise
augmented matrix—that is, multiple matrices obtained from films
with the same components in different proportions—helps to
obtain more accurate results since greater variability is included in
the data set.
When the standards are available, the following steps are
recommended in MATLAB:
• Unfold NIR-HSI images of each standard component.
≫ Dace = unfoldhsi(HSIace,1);
≫ Dgli = unfoldhsi(HSIgli,1);
≫ Dtef = unfoldhsi(HSItef,1);
It is important to highlight the application of spectral pretreat-

ments on the standard spectra. The application of the same spectral
pretreatment in each standard spectra of D matrix is mandatory
only when the spectral profile is modified, for instance, when the
first or second derivative is applied. In this example, the application
of the same pretreatments on D matrix was evaluated. Better results
were obtained without pretreatments. Then, the application of the
spectral pretreatments must be evaluated for each dataset. Here,
better results were obtained without pretreatments. Hence, the
application of the spectral pretreatments must be evaluated for
each dataset:
• Calculate the average spectra to be used as initial estimation:
≫ pure_spectra = [mean(Dace);mean(Dgli); mean(Dtef)];
• Concatenate the film pretreated D matrix with the D matrices of

all standards:
≫ Dmcr = [Dsmoo;Dace;Dgli;Dtef];
Then, the mcr_main command can be used to initiate the

MCR-ALS 2.0 toolbox. The Dmcr is the data matrix to be selected.
The number of components can be informed manually (three in
this example). The pure_spectra are used as the initial estimation.
The non-negativity concentration must be selected by implemen-
tation of fnnls with all components presenting non-negativity pro-
file. The normalization with the spectra dived by Euclidean norm
must be used. The number of interactions must be set as 100 and
convergence criterion as 0.01. The outputs of MCR-ALS analysis
are the recovered spectra (sopt output or S matrix) and concentra-
tion (copt output or C matrix) of all components. More information
about the MCR-ALS 2.0 toolbox can be found in Jaumot
et al. [50].
The MCR-ALS results of the example presented here are avail-
able at ResultExample.mat.
• On MATLAB environment:
≫ load(‘ResultExample.mat’)
The recovered NIR spectra by MCR-ALS for all components

are presented in Fig. 6 (blue dashed line) in comparison with pure
NIR spectra of all constituents (red solid lines).
Fig. 6 MCR-ALS recovered NIR spectra profile (blue dashed lines) compared to the standards’ profile of each
component (red solid lines) and the correspondent correlation coefficient (R) for cellulose acetate (a), glycerol
(b), and PTFE (c)
Fig. 7 Reshape of C matrix and building of distribution maps
This comparison can be obtained by compspec function using

the following command in MATLAB:
≫ spec = compspec (sopt,pure_spectra, w);
where spec is the normalized spectra, and w is the wavelength range.
3.4.3 Distribution Maps Regardless of the method used to obtain the concentrations of the
and Homogeneity Analysis components of interest, distribution maps can be obtained by
reshaping the relative intensity of the C matrix for each component
(Fig. 7).
The reshaping of C matrix into distribution maps can be
obtained by distmaps function using the following command in
MATLAB:
Fig. 8 Distribution maps of cellulose acetate (a), glycerol (b), and PTFE (c)
• Reshaping the C matrix unfold NIR-HSI images of each

standard:
≫ Ximg = distmaps (copt,m,n);
where Ximg is the cell output containing the distribution

maps of all components, m (rows) and n (columns) are the
spatial dimension of NIR-HSI image before unfolding (HSIspi).
Figure 8 shows the distribution maps of each component.
Next, the homogeneity of the image can be evaluated through
macropixel analysis. The homogeneity concept is related to the
random distribution of the components in the image. First, an
image must be split into non-overlapping square pixels. Then, the
macropixel is calculated by the intensity of the average of all pixels
and the homogeneity parameters can be calculated. These homo-
geneity parameters are obtained by comparing a real image with a
random version obtained from the same pixels, which allows the
calculation of a homogeneity index [9].
In the present example, the Ximg of each component was
divided into 10 × 10 pixels, which was then used to calculate the
macropixel:
≫ Macro1 = macroindex (Ximg{1,1}, mpsize);

Fig. 9 Macropixel analysis of cellulose acetate. (a) Distribution map of cellulose

acetate concentration; (b) Macropixel of distribution map of cellulose acetate; (c)
Random distribution map of cellulose acetate; and (d) Macropixel of random
distribution map of cellulose acetate
where Macro1, Macro2, and Macro3 are struct outputs with the
macropixel image and homogeneity indexes of each component.
Macropixel size is defined by mpsize input. Figure 9 shows an
example of the results generated by the macroindex function.
The obtained homogeneity index is the homogeneity ratio of
Poole (H%Poole). As this index gets closer to 100, the distribution
of the components in the image, or in this case in the film, becomes
more random (see Note 10). For the cellulose acetate, glycerol, and
PTFE example, the indexes obtained were 62%, 52%, and 82%,
respectively, emphasizing heterogeneous distribution of cellulose
acetate and glycerol, that is, in the film composition, since PTFE is
only the support.
4 Notes
1. The most significant cost associated with NIR-HIS analysis is

obviously related to the instrument acquisition, such as
NIR-HIS chemical imaging system. High processing capabil-

ities computers might be required for large datasets.
2. One of the main limitations of NIR-HSI is associated with its
sensitivity, which can make difficult to differentiate spectra of
highly homogeneous samples [14]. Hence, some degree of
heterogeneity is required, since the difference in each pixel
spectrum is used to identify and locate the chemical species
present in the polymeric matrix. The limit of detection asso-
ciated with this method depends mainly on the NIR capabil-
ities, and it is generally around 0.1 wt% [40].
3. Film thickness plays an important role in the data acquisition
since it can affect the signal intensity. Polymeric films with
thickness ranging from 45 to 150 μm have been successfully
analyzed using NIR-HSI by our group.
4. Conditioning of the films is recommended before the
NIR-HSI acquisition. Conditioning procedures are carried
out to bring the material to an equilibrium state with repro-
ducible conditions. The American Society for Testing and
Materials (ASTM) D618-13 standard [51] describes the prac-
tice for conditioning plastics for testing and it is recommended
before NIR-HSI data collection. Basically, the samples should
be kept at 23 ± 2 °C and 50 ± 2% of relative humidity for at
least 24 h before analysis.
5. Image treatment requires a uniform film sample. The presence
of holes, bubbles, wrinkles, or other defects in the film sample
limits NIR-HSI analysis. For instance, the stacking sequence
and thickness of the structure can affect the detection and
representability of the method. Combinations of vibrational
modes are important information of NIR spectroscopy, which
are dependent on the compounds distribution and are directly
affected by the presence of defects, which might lead to band
overlapping, displacement, or disappearance [40]. Flat samples
still result in some light loss due to reflections at interfaces,
such as the air-sample interface; however, the magnitude of this
is usually rendered negligible compared to the absorbance of
light by the sample. When scattering occurs, the light that does
not reach the detector is still interpreted as absorbance even
though it was not absorbed by the sample [52]. This scattering
effect can be increased when the infrared light strikes the
defects.
6. Before analysis, film samples can be accommodated in a sup-
port to avoid possible movement and to guarantee that the
sample is lying flat and to avoid any wrinkles or fold marks.
PTFE can be successfully used as support and an adhesive tape
can be used to securely attach the sample onto the support.
Concomitantly, it is important to preserve the film integrity

and avoid excessive stretching tension to the sample.
7. SisuCHEMA system can analyze samples up to
200 × 300 mm2, which allows the evaluation samples with
variables sizes. This system can image samples from 10 mm at
30 microns pixel resolution up to 100 mm at 300 microns
resolution [38].
8. The collection of the pure component spectra is crucial to
perform the NIR-HSI analysis. Film components are available
in the liquid, powder, or pellet form. In this case, blank film
sample (without any additive) could be used. When blank film
samples cannot be produced, the spectra of each component
could be acquired separately. If possible, the pure component
spectra should be collected following the same method as for
the film samples. For that, it is important to prepare this pure
component sample using a similar support on which the indi-
vidual component samples are uniformly placed.
9. Some instruments might require manual acquisition of the dark
and white references. According to Amigo and Grassi [53], this
calibration could be performed by taking an image of a dark
and a white reference. The dark reference can be obtained by
turning off the light sources or covering the camera with a
non-reflective opaque black cap. The white reference can be
acquired by measuring a uniform, high-reflectance standard or
white ceramic. Then, the relative reflectance image can be
obtained by subtracting the raw spectral image from the dark
and dividing the result by the difference between the white and
dark references. Refer to Amigo and Grassi [53] for more
detail.
10. Despite being herein showcased with a biodegradable polymer,
the presented method can be applied to analyze any
polymeric film.
Acknowledgments
The authors are grateful for the financial support of Conselho

Nacional de Desenvolvimento Cientı́fico e Tecnológico (CNPq),
Fundação de Amparo à Pesquisa do Estado de Minas Gerais
(FAPEMIG), and Coordenação de Aperfeiçoamento de Pessoal
de Nı́vel Superior—Brasil (CAPES)—Finance Code 001. We also
thank Cristiane Vidal for the experimental support in the HSI
acquisition and Prof. Celio Pasquini for promptly receiving us in
the laboratory that he coordinates (Grupo de Instrumentação
e Automação em Quı́mica Analı́tica, Instituto de Quı́mica, Uni-
versidade Estadual de Campinas, Campinas-SP, Brazil) to obtain
the images. C. L. G. and C. C. P. gratefully acknowledge funding

support from National Science Foundation under award numbers
CBET-1756999. C. L. G. and C. C. P. also acknowledge the
National Institute of Food Agriculture, US Department of Agricul-
ture, award numbers 2021-67017-33344, 2020-67021-31375,
and 2018-672 67016-27578 awarded as a Center of Excellence
for financial support.
References
1. Pfaendner R (2006) How will additives shape Chemom 34(1):e3193. https://doi.org/10.
the future of plastics? Polym Degrad Stab 1002/cem.3193
91(9):2249–2256. https://doi.org/10.1016/ 10. Amigo JM, Babamoradi H, Elcoroaristizabal S
j.polymdegradstab.2005.10.017 (2015) Hyperspectral image analysis. A tuto-
2. Rabello M (2000) Aditivação de polı́meros. rial. Anal Chim Acta 896:34–51. https://doi.
Artliber org/10.1016/j.aca.2015.09.030
3. Rabello M, De Paoli M (2013) Aditivaçao De 11. Fraser DG, Jordan RB, Künnemeyer R,
Termoplasticos. Artliber McGlone VA (2003) Light distribution inside
4. Ambrogi V, Carfagna C, Cerruti P, Marturano mandarin fruit during internal quality assess-
V (2017) Chapter 4 – Additives in ment by NIR spectroscopy. Postharvest Biol
polymers. In: Jasso-Gastinel CF, Kenny JM Technol 27(2):185–196. https://doi.org/10.
(eds) Modification of polymer properties. Wil- 1016/S0925-5214(02)00058-3
liam Andrew Publishing, pp 87–108. https:// 12. Li Q, Tang Y, Yan Z, Zhang P (2017) Identifi-
doi.org/10.1016/B978-0-323-44353-1. cation of trace additives in polymer materials by
00004-X attenuated total reflection Fourier transform
5. Groh KJ, Backhaus T, Carney-Almroth B, infrared mapping coupled with multivariate
Geueke B, Inostroza PA, Lennquist A, Leslie curve resolution. Spectrochim Acta A Mol Bio-
HA, Maffini M, Slunge D, Trasande L, War- mol Spectrosc 180:154–160. https://doi.org/
hurst AM, Muncke J (2019) Overview of 10.1016/j.saa.2017.03.019
known plastic packaging-associated chemicals 13. Li N, Taylor LS (2016) Nanoscale infrared,
and their hazards. Sci Total Environ 651: thermal, and mechanical characterization of
3253–3268. https://doi.org/10.1016/j. telaprevir–polymer miscibility in amorphous
scitotenv.2018.10.015 solid dispersions prepared by solvent evapora-
6. Harper CA, Harper C (2006) Handbook of tion. Mol Pharm 13(3):1123–1136. https://
plastics technologies: the complete guide to d o i . o r g / 10 . 1 0 21 / a c s . m ol p ha r m ac e ut .
properties and performance. McGraw-Hill 5b00925
Education 14. Manley M (2014) Near-infrared spectroscopy
7. Han J-W, Ruiz-Garcia L, Qian J-P, Yang X-T and hyperspectral imaging: non-destructive
(2018) Food packaging: a comprehensive analysis of biological materials. Chem Soc Rev
review and future trends. Compr Rev Food 43(24):8200–8214. https://doi.org/10.
Sci Food Saf 17(4):860–877. https://doi. 1039/C4CS00062E
org/10.1111/1541-4337.12343 15. Calvini R, Ulrici A, Amigo JM (2020) Growing
8. Avolio R, Castaldo R, Avella M, Cocca M, applications of hyperspectral and multispectral
Gentile G, Fiori S, Errico ME (2018) imaging. In: Data handling in science and tech-
PLA-based plasticized nanocomposites: effect nology, vol 32. Elsevier, pp 605–629
of polymer/plasticizer/filler interactions on 16. De Juan A, Piqueras S, Maeder M,
the time evolution of properties. Compos Hancewicz T, Duponchel L, Tauler R (2014)
Part B 152:267–274. https://doi.org/10. Chemometric tools for image analysis. In:
1016/j.compositesb.2018.07.011 Infrared and Raman spectroscopic imaging.
9. Terra LR, Roque JV, Pola CC, Gonçalves IM, Wiley-VCH, Weinheim, pp 57–110. https://
Soares NFF, Teófilo RF (2020) Study of chem- doi.org/10.1002/9783527678136.ch2
ical compound spatial distribution in biode- 17. Wu D, Sun D-W (2013) Advanced applications
gradable active films using NIR hyperspectral of hyperspectral imaging technology for food
imaging and multivariate curve resolution. J quality and safety analysis and assessment: a
review—Part I: fundamentals. Innovative
Food Sci Emerg Technol 19:1–14. https://doi. 27. Alexandrino GL, Poppi RJ (2013) NIR imag-
org/10.1016/j.ifset.2013.04.014 ing spectroscopy for quantification of constitu-
18. Zhang C, Jiang H, Liu F, He Y (2017) Appli- ents in polymers thin films loaded with
cation of near-infrared hyperspectral imaging paracetamol. Anal Chim Acta 765:37–44.
with variable selection methods to determine https://doi.org/10.1016/j.aca.2012.12.017
and visualize caffeine content of coffee beans. 28. Amigo JM, Cruz J, Bautista M, Maspoch S,
Food Bioprocess Technol 10(1):213–221. Coello J, Blanco M (2008) Study of pharma-
https://doi.org/10.1007/s11947-016- ceutical samples by NIR chemical-image and
1809-8 multivariate analysis. TrAC Trends Anal Chem
19. Prats-Montalbán JM, de Juan A, Ferrer A 27(8):696–713. https://doi.org/10.1016/j.
(2011) Multivariate image analysis: a review trac.2008.05.010
with applications. Chemom Intell Lab Syst 29. Carneiro RL, Poppi RJ (2014) Infrared imag-
107(1):1–23. https://doi.org/10.1016/j. ing spectroscopy and chemometric tools for in
chemolab.2011.03.002 situ analysis of an imiquimod pharmaceutical
20. ElMasry G, Sun D-W, Allen P (2012) Near- preparation presented as cream. Spectrochim
infrared hyperspectral imaging for predicting Acta A Mol Biomol Spectrosc 118:215–220.
colour, pH and tenderness of fresh beef. J https://doi.org/10.1016/j.saa.2013.08.104
Food Eng 110(1):127–140. https://doi.org/ 30. de la Ossa MAF, Amigo JM, Garcı́a-Ruiz C
10.1016/j.jfoodeng.2011.11.028 (2014) Detection of residues from explosive
21. Kamruzzaman M, Barbin D, ElMasry G, Sun manipulation by near infrared hyperspectral
D-W, Allen P (2012) Potential of hyperspectral imaging: a promising forensic tool. Forensic
imaging and pattern recognition for categori- Sci Int 242:228–235. https://doi.org/10.
zation and authentication of red meat. Innova- 1016/j.forsciint.2014.06.023
tive Food Sci Emerg Technol 16:316–325. 31. de la Ossa MAF, Garcı́a-Ruiz C, Amigo JM
https://doi.org/10.1016/j.ifset.2012.07.007 (2014) Near infrared spectral imaging for the
22. Rodrigues e Brito L, Braz A, Saldanha Hon- analysis of dynamite residues on human hand-
orato R, Pimentel MF, Pasquini C (2019) Eval- prints. Talanta 130:315–321. https://doi.org/
uating the potential of near infrared 10.1016/j.talanta.2014.07.026
hyperspectral imaging associated with multi- 32. Silva CS, Pimentel MF, Honorato RS,
variate data analysis for examining crossing ink Pasquini C, Prats-Montalbán JM, Ferrer A
lines. Forensic Sci Int 298:169–176. https:// (2014) Near infrared hyperspectral imaging
doi.org/10.1016/j.forsciint.2019.02.043 for forensic analysis of document forgery. Ana-
23. Piqueras S, Duponchel L, Tauler R, de Juan A lyst 139(20):5176–5184. https://doi.org/10.
(2011) Resolution and segmentation of hyper- 1039/C4AN00961D
spectral biomedical images by multivariate 33. Weinstock BA, Janni J, Hagen L, Wright S
curve resolution-alternating least squares. (2006) Prediction of oil and oleic acid concen-
Anal Chim Acta 705(1):182–192. https:// trations in individual corn (Zea mays L.) ker-
doi.org/10.1016/j.aca.2011.05.020 nels using near-infrared reflectance
24. Adam E, Mutanga O, Rugege D (2010) Mul- hyperspectral imaging and multivariate analy-
tispectral and hyperspectral remote sensing for sis. Appl Spectrosc 60(1):9–16. https://doi.
identification and mapping of wetland vegeta- org/10.1366/000370206775382631
tion: a review. Wetl Ecol Manag 18(3): 34. ElMasry G, Wang N, ElSayed A, Ngadi M
281–296. https://doi.org/10.1007/s11273- (2007) Hyperspectral imaging for nondestruc-
009-9169-z tive determination of some quality attributes
25. Lee H, Kim MS, Song Y-R, Oh C-S, Lim H-S, for strawberry. J Food Eng 81(1):98–107.
Lee W-H, Kang J-S, Cho B-K (2017) https://doi.org/10.1016/j.jfoodeng.2006.
Non-destructive evaluation of bacteria-infected 10.016
watermelon seeds using visible/near-infrared 35. Forchetti DAP, Poppi RJ (2017) Use of NIR
hyperspectral imaging. J Sci Food Agric hyperspectral imaging and multivariate curve
97(4):1084–1092. https://doi.org/10.1002/ resolution (MCR) for detection and quantifica-
jsfa.7832 tion of adulterants in milk powder. LWT Food
26. Sun H, Zhang S, Chen C, Li C, Xing S, Liu J, Sci Technol 76:337–343. https://doi.org/10.
Xue J (2019) Detection of the soluble solid 1016/j.lwt.2016.06.046
contents from fresh jujubes during different 36. Zhao J, Vittayapadung S, Chen Q, Chaitep S,
maturation periods using NIR hyperspectral Chuaviroj R (2009) Nondestructive measure-
imaging and an artificial bee colony. J Anal ment of sugar content of apple using hyper-
Methods Chem 2019:5032950. https://doi. spectral imaging technique. Maejo Int J Sci
org/10.1155/2019/5032950 Technol 3(1):130–142
37. Taghizadeh M, Gowen A, Ward P, O’Donnell in PLS regression: how and when to use them.
CP (2010) Use of hyperspectral imaging for Chemom Intell Lab Syst 41(2):195–207.
evaluation of the shelf-life of fresh white button https://doi.org/10.1016/S0169-7439(98)
mushrooms (Agaricus bisporus) stored in dif- 00051-3
ferent packaging films. Innovative Food Sci 46. Roque JV, Cardoso W, Peternelli LA, Teófilo
Emerg Technol 11(3):423–431. https://doi. RF (2019) Comprehensive new approaches for
org/10.1016/j.ifset.2010.01.016 variable selection using ordered predictors
38. Specim (2015) SisuChema chemical imaging selection. Anal Chim Acta 1075:57–70.
analyzer. https://www.specim.fi/wp-content/ https://doi.org/10.1016/j.aca.2019.05.039
uploads/2020/03/SisuCHEMA_2_2015. 47. Teófilo RF, Martins JPA, Ferreira MMC
pdf. Accessed 12 Sept 2020 (2009) Sorting variables by using informative
39. Cortés V, Blasco J, Aleixos N, Cubero S, Talens vectors as a strategy for feature selection in
P (2019) Monitoring strategies for quality con- multivariate regression. J Chemom 23(1):
trol of agricultural products using visible and 32–48. https://doi.org/10.1002/cem.1192
near-infrared spectroscopy: a review. Trends 48. Leardi R, Nørgaard L (2004) Sequential appli-
Food Sci Technol 85:138–148. https://doi. cation of backward interval partial least squares
org/10.1016/j.tifs.2019.01.015 and genetic algorithms for the selection of rel-
40. Pasquini C (2003) Near infrared spectroscopy: evant spectral regions. J Chemom 18(11):
fundamentals, practical aspects and analytical 486–497. https://doi.org/10.1002/cem.893
applications. J Braz Chem Soc 14:198–219 49. de Araújo Gomes A, Galvão RKH, de Araújo
41. Pasquini C (2018) Near infrared spectroscopy: MCU, Véras G, da Silva EC (2013) The suc-
a mature analytical technique with new cessive projections algorithm for interval selec-
perspectives – a review. Anal Chim Acta 1026: tion in PLS. Microchem J 110:202–208.
8–36. https://doi.org/10.1016/j.aca.2018. https://doi.org/10.1016/j.microc.2013.
04.004 03.015
42. Vidal M, Amigo JM (2012) Pre-processing of 50. Jaumot J, de Juan A, Tauler R (2015)
hyperspectral images. Essential steps before MCR-ALS GUI 2.0: new features and applica-
image analysis. Chemom Intell Lab Syst 117: tions. Chemom Intell Lab Syst 140:1–12.
138–148. https://doi.org/10.1016/j. https://doi.org/10.1016/j.chemolab.2014.
chemolab.2012.05.009 10.003
43. Amigo JM, Santos C (2020) Chapter 2.1 – 51. ASTM (2013) ASTM D618-21: Standard
Preprocessing of hyperspectral and multispec- practice for conditioning plastics for testing
tral images. In: Amigo JM (ed) Data handling 52. Bassan P (2011) Light scattering during infra-
in science and technology, vol 32. Elsevier, pp red spectroscopic measurements of biomedical
37–53. https://doi.org/10.1016/B978-0- samples. The University of Manchester
444-63977-6.00003-1 53. Amigo JM, Grassi S (2020) Chapter 1.2 – Con-
44. Wold S, Sjöström M, Eriksson L (2001) figuration of hyperspectral and multispectral
PLS-regression: a basic tool of chemometrics. imaging systems. In: Amigo JM (ed) Data
Chemom Intell Lab Syst 58(2):109–130. handling in science and technology, vol 32.
https://doi.org/10.1016/S0169-7439(01) Elsevier, pp 17–34. https://doi.org/10.
00155-1 1016/B978-0-444-63977-6.00002-X
45. Leardi R, Lupiáñez González A (1998)
Genetic algorithms applied to feature selection
Chapter 11
Water Vapor Permeability of Hydrophilic Films

Roberto J. Avena-Bustillos, Noah M. Klausner, and Tara H. McHugh
Abstract
The modified procedure for water vapor permeability (WVP) is a modification to the established ASTM
E96 method for measuring the WVP of films. The E96 method works by putting water in cups and
measuring the mass transfer rate of water vapor through films that are secured as lids to the cups. The WVP
is calculated from a formula including this mass transfer rate as well as estimated partial water vapor pressure
under the film lid at the testing constant temperature. Using the E96 method, the partial water vapor
pressure under the film lid is assumed to be the same as the saturated water vapor pressure at the surface of
the water. This assumption is only true for hydrophobic films, which is why the partial water vapor pressure
under the film lid must be calculated in this modified procedure when measuring the WVP of hydrophilic
films. Here, we provide a detailed account of the foundation for this correction and the procedure to
reliably use it to measure the permeability of water vapor through hydrophilic films.
Key words Water vapor permeability, Partial water vapor pressure, Hydrophilic, Hydrophobic, Water
vapor transmission rate, Permeance, Diffusion, Solubility, Barrier
1 Introduction
The water vapor permeability (WVP) of hydrophilic edible films is

used in determining the shelf-life of food products as well as
designing edible films to match specific food applications. The
accuracy of WVP determinations can influence food stability; there-
fore, it is important to use an accurate method to determine WVP.
The established method for determining the WVP of polymeric
films is the ASTM E96 method [1]. This method entails filling a
cup partially with deionized water and sealing the top of the cup
with the film being tested as a lid (Fig. 1).
1.1 Water Vapor In order to calculate WVP, the transport of water vapor through
Permeability Formula the film is modeled under the following assumptions:
• Water vapor transports one-dimensionally in the positive
z direction.
https://doi.org/10.1007/978-1-0716-3613-8_11,
205
206 Roberto J. Avena-Bustillos et al.
film being tested
electric fan pa, 3 pw, 3 z3

z2
pa, 2 pw, 2
air gap
pa, 1 pw, 1 z1
+
water
z=0 Z
Fig. 1 The cup and film are modeled as an ideal cylindrical system. The partial
vapor pressures of air and water vapor at position zi are pa,i and pw,i,
respectively. If the fan is circulating air fast enough to maintain 0% relative
humidity at z3, pw,3 will be negligible. The temperature is kept constant and
homogenous throughout the system
• The mode of transport is diffusion, which can be modeled by

Fick’s laws.
• Water vapor transport reaches steady-state conditions through-
out the film.
• The solubility of water vapor on the film matrix can be modeled
with Henry’s law.
List of Symbols
C(z,t) = mass concentration of water vapor at position z, time
t [g m-2]
J(z,t) = mass flux of water vapor at position z, time t [g m-2 s-1]
A = area of the mouth of the cup [m2]
V = volume [m3]
D = diffusion coefficient of water vapor [m2 s-1]
S = solubility of water vapor in the film [s2 m-2]
p = partial pressure of water vapor [kPa]
Q = mass of water vapor transferred [g]
The WVP derivation begins with a mass balance on a horizontal
differential slice of the film [3]:
δC ðJ - J out Þ * A - δJ * A - δJ
= in = = ð1Þ
δt δV δz * A δz
Invoking Fick’s first law:
dC - δJ δ2 C
J = -D → =D 2 ð2Þ
dz δz δz
WVP of Hydrophilic Films 207
Substituting Eq. 2 into Eq. 1 gives the following partial differ-

ential equation:
δC δ2 C
=D 2 ð3Þ
δt δz
Because the water vapor concentration at any point will have
reached a steady state, δC
δt = 0 and Eq. 3 can be reduced to the
following boundary value problem:
d2C
= 0 : C ðz 2 Þ = C 2 , C ðz 3 Þ = C 3 ð4Þ
dz 2
Solving Eq. 4 yields the following formula for water vapor
concentration at position z:
C3 - C2
C ðz Þ = C 2 þ ðz - z 2 Þ ð5Þ
z3 - z2
Taking the z derivative of Eq. 5 and substituting it back into
Fick’s Law provides the following expression for mass flux of water
vapor:
C2 - C3
J =D
dC C - C2 z3 - z2
= 3 ð6Þ
dz z3 - z2 dC
J = -D
dz
Henry’s Law states that C = Sp, and J, the mass flux of water
Q
vapor, is defined as At where t is the time duration of mass transfer.
Substituting these two expressions into Eq. 6 gives the following
equation:
Q pw,2 - pw,3
= SD ð7Þ
At z3 - z2
Since the WVP of a film is the solubility of water vapor on the
matrix of the film multiplied by the diffusivity of water through the
film, the expression WVP = SD can be substituted, and Eq. 7 can
be rearranged to solve for WVP:
Q ðz 3 - z 2 Þ
WVP = ð8Þ
At pw,2 - pw,3
Because pw,3 is controlled at 0 (Fig. 1) the following equation,

Eq. 9, is the final WVP formula [2]:
Q ðz 3 - z 2 Þ
WVP = ð9Þ
Atpw,2
The basis of the ASTM E96 method for WVP is measuring the
mass loss from the cup over time (Q/t) while controlling the other
parameters of the equation. To do so, the cup is placed in a cabinet
kept at 0% relative humidity (RH) by fans moving air at a speed of
152 m min-1 to assure constant and complete wipeout of water

vapor coming out of the film top surface [2], while the cabinet is
inside a controlled-temperature incubator, chamber, or room. The
loss of mass of the cup is monitored over time, and the data
gathered is analyzed through linear regression to determine the
rate at which water molecules transport through the film in units
of mass per time. This mass loss rate is divided by the area of the film
to obtain the water vapor transmission rate (WVTR). The per-
meance of the film is obtained by dividing the WVTR by the partial
vapor pressure of water on the underside of the film. Finally, the
WVP is determined by multiplying the permeance by the thickness
of the film, as suggested by the derived equation.
Under the ASTM E96-80 method, the assumption is made
that water molecules will diffuse through the air gap between the
water surface and the film underside and being largely trapped by
the film until eventually reaching a homogeneous equilibrium value
throughout the air gap, similar to the saturated partial water vapor
pressure at the constant testing temperature. This assumption is
only correct for hydrophobic films because they have more efficient
moisture barrier properties than hydrophilic films [4]. In the case of
hydrophilic films, water molecules move faster through the film,
and the transport of water molecules causes a gradient of partial
water vapor pressure values through the air gap when equilibrium is
reached [2]. As a result of the partial water vapor pressure gradient,
the partial water vapor pressure at the underside of a hydrophilic
film is lower than the saturated partial water vapor pressure at the
surface of the water at the bottom of the cup.
1.2 Partial Water The WVP correction method [2] to the ASTM E96 method is a
Vapor Pressure correction for the water vapor pressure value used in the WVP
Formula calculation. To determine the correct WVP for a hydrophilic film,
the partial vapor pressure of water at the underside of the lid must
be calculated by modeling the gradient of water vapor pressure
across the air gap. To do so, the transport of water vapor across
the air gap must be modeled under the following assumptions [5]:
• Transport occurs within a binary molecular system of air and
water vapor. Subscripts of a and w will be used to denote the
properties of air and water vapor, respectively.
• Transport of a species occurs via a combination of diffusive flux
and bulk motion.
• Diffusive flux can be modeled with Fick’s Law.
• Transport of air is negligible (Na = 0).
• The ideal gas law holds true.
List of Symbols
Ni = total molecular transport of species i [mol m-2 s-1]
Ci = molar concentration of species i [mol m-3]
n
C = molar concentration of the mixture = C i for n number of
species in the mixture [mol m-3] i=1
vi = velocity of species i [m s-1] n

C i vi
v = molar average velocity of the mixture = i=1
C [m s-1]
yi = molar fraction of species i = CCi *100 [%]
Ji = diffusive molar flux of species i [mol m-2 s-1]
Daw = diffusion coefficient between air and water vapor [m2 s-1]
p = pressure of the mixture [atm]
pj,i = partial pressure of species j at position zi [kPa]
V = volume [m3]
T = temperature (constant) [K]
R = the gas constant [m3 atm K-1 mol-1]
We start by defining the total molecular transport of water
vapor (Nw) as the product of the molar water vapor concentration
(Cw) and the velocity of water vapor in the mixture (vw):
N w = C w vw ð10Þ
The right-hand expression can be manipulated to include the
diffusive velocity of water vapor (vw - v) where v is the molar-
average velocity of the mixture:
C w vw þ C a va
N w = C w ðv w - v Þ þ C w v : v = ð11Þ
Cw þ Ca
Cw
N w = C w ðv w - v Þ þ ðC v þ C a va Þ ð12Þ
Cw þ Ca w w
Given that Cw + Ca is the mixture concentration C, the mole
fraction of substance i (yi) is defined as CCi . Also, air transport is
defined as Na = Cava in the same fashion as water vapor.
Plugging in these definitions to Eq. 12 yields the following
equation:
N w = C w ðvw - vÞ þ y w ðN w þ N a Þ ð13Þ
The first term on the right side of Eq. 13 accounts for the
diffusive molar flux of water (Jw) and the second term accounts for
the flux of water due to bulk motion. Since the transport of air is
negligible (Na = 0), Eq. 13 can be simplified as follows:
Jw
N w = J w þ ywN w → N w = ð14Þ
1 - yw
Invoking Fick’s first law:

dC w - D aw dC w
J w = - D aw →Nw = * ð15Þ
dz 1 - yw dz
- CD aw dy w
C w = Cy w → N w = * ð16Þ
1 - yw dz
Integrating both sides of Eq. 16 across the z-axis of the water
gap progresses as follows:
z2 y w,2
dy w
Nw dz = - CD aw ð17Þ
z1 y w,1 1 - yw
where yw,i = mole fraction of water vapor at position zi

1 - y w,2
N w ðz 2 - z 1 Þ = CD aw ln ð18Þ
1 - y w,1
By definition, yw,i + ya,i = 1 → 1 – yw,i = ya,i.
The expression inside the natural logarithm of Eq. 18 can be
manipulated as follows to obtain an equation that depends on
partial pressures of water at z positions 1 and 2:
C a,2
1 - y w,2 y a,2 C C a,2
= = = ð19Þ
1 - y w,1 y a,1 C a,1 C a,1
C
na,i pa,i
By the ideal gas law, pV = nRT → C a,i = V a,i = RT and C =
n p
V = RT
pa,2
C a,2 RT
pa,2 1 - pw,2
= pa,1 = = ð20Þ
C a,1 pa,1 1 - pw,1
RT
Substituting the expression from Eq. 20 into Eq. 18 gives the

following:
pD aw 1 - pw,2
N w ðz 2 - z 1 Þ = ln ð21Þ
RT 1 - pw,1
Rearranging ends the derivation for partial pressure of water
vapor directly under the film ( pw,2):
RT N w ðz 2 - z 1 Þ
pw,2 = 1 þ pw,1 - 1 exp ð22Þ
pD aw
Eq. 22 is used to approximate the partial pressure of water
vapor on the film underside, which is a key parameter in Eq. 9,
the WVP formula.
This correction is advantageous firstly because more accurate
WVP values can be obtained, and secondly because the effect of film
thickness on WVP in hydrophilic films can be explained. Previous
studies have indicated similar relationships between film thickness
and permeability properties in hydrophilic film systems; however,

those studies could not adequately define the relationships in their
explanations for the thickness effect. WVP was found to increase
when the thickness of a given hydrophilic film was increased, which
is counterintuitive. Examining the conditions on the film’s under-
side was necessary to explain this relationship.
Using the correction method [2], an exponential relationship
between WVP and the RH at the underside of the film (propor-
tional to the water vapor pressure there) can be established by
varying the air gap distance over WVP tests. This exponential
relationship is remarkably similar to the relationship between film
thickness and WVP, so it can be logically deduced that there is a
relationship between film thickness and RH at the film’s underside.
Supporting this logic is the idea that a thicker hydrophilic film will
resist mass transfer more than a thinner one, causing a higher RH
(and consequently a higher water vapor pressure) at the film under-
side. This higher water vapor pressure value causes a higher WVP
value, based on the relationship previously established.
A limitation to the WVP correction method is that it relies on
experimental saturated water pressure and heat of vaporization
values. To calculate the partial water vapor pressure under the
film, the partial water vapor pressure at the surface of the water
(saturated vapor pressure) must be known. This value can be
obtained from a steam table at a given constant temperature—or
calculated using the Clausius–Clapeyron equation [6]. If the
saturated vapor pressure is calculated, other experimentally derived
values must be used. Using experimentally derived values intro-
duces some amount of error between the actual conditions of the
WVP test and the conditions under which those values were
obtained. That said, accurate steam tables such as the 1967
ASME Steam Tables [7] are readily accessible, making the reliance
on tabulated values a minor limitation.
2 Materials
1. Controlled-temperature room or incubator to keep tempera-

ture constant and large enough to house one or more cabinets
kept at 0% RH.
2. Desiccating cabinet—such as catalog no. 08-647-28 from
Fisher Scientific, Inc. (Fair Lawn, NJ, USA) or other models
and manufacturers—to hold at least eight cups.
3. Motor—such as the Bodine motor model no. 574—with vari-
able speed controller—such as the Motor Master Series
20,000, both from Minarik Electric Co. (Fresno, CA, USA)
or other models and manufacturers—as well as a fan—such as
Fig. 2 Water vapor permeability cabinet with motor and fan rotation speed
controller, Drierite in bottom shallow pans and sample cups on top tray, with
cabinet inside a temperature-controller incubator
model no. 607601-01 from Refrigeration Supply House

(Sacramento, CA, USA) or other models and manufacturers—
installed in each hermetically sealed cabinet to control RH at
0% (see Note 1). The motor is attached to the desiccating
cabinet such that the rotating shaft goes through the roof of
the cabinet, and the gap between the shaft and the hole in the
cabinet is sealed by a chamber with oil. The fan is attached to
the shaft of the motor, so that it spins horizontally near the
ceiling of the cabinet (Fig. 2).
4. Calcium sulfate desiccant—for example, 6-mesh Drierite from
Fisher Scientific, Inc. (Fair Lawn, NJ, USA) or other
suppliers—is recommended to equilibrate cabinets to 0% RH
prior to each experiment (see Note 2).
5. Mid-sized plastic trays to hold the desiccant are placed under-
neath the test cups.
6. Cups made of poly(methyl methacrylate) (PMMA), aluminum,
or stainless steel—in a uniform cylindrical fashion. For a typical
cup-shaped configuration, the cup could be 1.25-cm tall and
have a diameter of 8.2 cm, externally. The cup opening could
have a diameter of 5 cm for an area of 19.6 cm2 for the cup
mouth where water will permeate through the testing film, and
the cup interior depth could be 1.05 cm.
7. To seal in the film, an 8.2 cm outside diameter, 5-cm-diameter
ring opening, and 0.60-cm-tall PMMA or other cup material
containing a 19.6-cm2 lid opening are placed on top of the film
through a face previously greased for tight sealing (see Note 3).
8. Silicon sealant—e.g., in the form of high vacuum grease from

Dow Corning (Midland, MI)—is used to create a barrier under
and over the film (see Note 4).
9. Deionized water (6 mL for previously described cup dimen-
sions) was added to the bottom of the cup to provide a source
of water vapor for the film to be exposed to (see Note 5).
10. A volumetric pipette was used to measure water for the
test cups.
11. Screws symmetrically located around the ring circumference
are used to tighten the seal.
12. Electric screwdriver—such as Checkline TSD-400 from Elec-
tromatic Equipment Co. (Cedarhurst, NY, USA) or other
models and manufacturers—to tighten four screws in each
cup. Seal ring tightening can also be done manually with a
regular screwdriver.
13. Anemometer—such as model no. 127MS from Solomat
(Stamford, CT, USA) or other models and manufacturers—is
to measure air velocity in the cabinet.
14. Hygrometer—such as model no. 605 from Airguide (Chicago,
IL, USA) or other models and manufacturers—to measure RH
inside the cabinet.
15. Analytic balance of at least 0.001 g sensitivity—such as Excel-
lence XPE Analytical Balance from Mettler Toledo (Columbus,
OH, USA) or other models and manufacturers—to weigh cups
at different time intervals during testing.
16. Micrometer—such as model no. 515 from Lufkin Rule
Co. (Saginaw, MI, USA) or other models and manufacturers—
used to measure film thickness (see Note 6).
17. Linear regression and subsequent calculations can be done in
an Excel spreadsheet such as in Fig. 3.
3 Methods
1. Place the cabinet in a temperature-controlled room or incuba-

tor with the temperature set at a known value near, usually
20–25 °C (see Fig. 2).
2. Spread out the Drierite in an inch-thick layer on shallow trays at
the bottom of the desiccating cabinet (see Note 2). Place the
hygrometer on the bottom of the cabinet next to the Drierite
(see Fig. 2).
*Steps 3–15 apply to each film individually
Fig. 3 Example of a spreadsheet for calculations of the water barrier properties of a gelatin film. Each sample
and replicate will have its own spreadsheet
3. Inspect eight films of the material to be tested, making sure

they appear identical and uniform in composition and without
pinholes.
4. Label each testing cup so they can be easily identified.
5. Apply a thin line of high-vacuum grease around the rim of each
cup and ring (see Note 7).
6. Using a pipette, transfer 6 mL of deionized water or saturated
salt solution to the bottom of the testing cup.
7. Measure the thickness of films to the nearest 0.001 mm at five
random positions using the micrometer. Average the five values
to obtain the thickness value to be used in the WVP calculation
for the film (see Note 6).
8. Place the film on the rim of the cup, centered.

9. Place the ring on top of the film, centered.
10. Using the screwdriver, tighten four screws symmetrically along
the circumference of the cup (see Note 8).
11. Place all eight testing cups on a tray inside the cabinet near the
walls for the greatest airflow. Take note of each cup’s position,
as they need to be rotated regularly to ensure each cup under-
goes the same air velocity conditions.
12. If the suggested components are used in the fan apparatus, the
motor controller can simply be set to maximum speed to
ensure proper airflow. If other components are used, imple-
ment the anemometer to measure airflow inside the cabinet
and calibrate the motor such that air velocity in the cabinet is at
least 152 m min-1 (see Note 9).
13. Wait an hour for steady-state water vapor conditions to be
achieved.
14. Using the balance, weigh each cup at least once every 2 h, for a
period of 24–28 h (see Note 10). Keep track of the time elapsed
in relation to the start of the experiment.
15. Each time the cups are weighed, rotate the positions of the
cups in the cabinets (see Note 11).
16. The average water gap is estimated using the initial and final
volumes of water added and the dimensions of the cup (see
Note 5).
17. For each cup, the data obtained can be interpreted as a set of
points on a graph consisting of time on the x-axis and cup mass
on the y-axis. Analyze the data with a linear regression, produc-
ing a formula for cup mass as a linear function of time.
18. Divide the slope of the linear mass function by the area of the
mouth of the testing cup to obtain the WVTR, the mass flux of
water vapor through the film.
19. Convert the WVTR into Nw, the molar flux of water vapor, by
dividing the WVTR by the molar mass of water.
20. Calculate the saturated water vapor pressure at the water’s
surface based on the air temperature (see Note 12).
21. Input the parameters into Eq. 22 to calculate the partial water
vapor pressure on the underside of the film pw,2 (see
Subheading 1.2).
22. Divide the WVTR by the partial water vapor pressure under the
film surface to obtain the permeance of the film.
23. Multiply the permeance of the film by its average thickness to
obtain its WVP.
4 Notes
1. The fan systems use airflow to maintain 0% RH in each cabinet

in order to keep the water vapor pressure outside of the film
controlled at 0 for the WVP measurement.
2. To make sure Drierite is fully activated, take a half-pound of
6-mesh, non-indicating Drierite and spread it into a one-
granule-thick layer on a tray and heat the layer at 210 °C for
1 h [8]. Contact with Drierite may cause various bodily irrita-
tions, and prolonged and repeated exposure can result in lung
disease and/or cancer, so it is recommended to use appropriate
PPE (nitrile gloves, respiratory mask, and goggles) [9].
3. The sealant is important because, without it, water vapor can
escape through the gap underneath the film and become a
source of error in the measure of mass transfer through the
film. Sealant should not be applied excessively in a way that will
protrude from the junctures of the film and cup mouth and lid
as it may result in errors during weighing as sealant could smear
fingers after touching cups.
4. Avoid direct eye contact with high vacuum grease, as it may
cause temporary redness and discomfort [10].
5. Initial height from the surface of the water in the cup to the film
underside is calculated from a 6-mL volume right cylinder
subtracted from the cup depth: 1.05–6/
2π (2.5)2 = 1.05–0.153 = 0.897 cm. The final height is
calculated from the final volume of water remaining at the
end of the test, and an average height is used for the calculation
of the stagnant air gap inside the testing cup.
6. For a given film, the thickness may vary at different points. For
this reason, the thicknesses at different points must be averaged
to best describe the overall thickness.
7. Avoid using too much grease, or it will seep across the edges of
the rim once under pressure. Grease on the technician’s fingers
after repeating WVP cup weighing will affect WVTR.
8. Make sure the screws apply even downward pressure on all
points of the ring by tightening each screw in small amounts
at a time and alternating between screws that are across from
each other in an “X” pattern.
9. Air speeds inside the cabinet may not be lower than
152 m min-1. This speed is the minimum airspeed necessary
to maintain 0% RH on the top surface of films.
10. Do not let the experiment run over 30 h long because the cups
can dry out and change the modeling parameters.
11. Rotating the cups after each mass measurement ensures that all
cups experience the same air velocity conditions over the
course of the procedure.
12. The following formula can be used to calculate the saturated
water vapor pressure at the water’s surface [6]:
exp 34:494 - 4924:99

T þ237:1
pw,1 = ð23Þ
ðT þ 105Þ1:57
Acknowledgments
Our gratitude to Dr. John M. Krochta, Emeritus Professor at the

Department of Food Science and Technology and the Biological
and Agricultural Engineering Department at the University of
California, Davis, for guiding authors Avena and McHugh to
develop the water vapor permeability method for hydrophilic
films during their PhD thesis work.
References
1. ASTM (1980) Standard test methods for water and ice. J Appl Meteorol Climatol 57(6):
vapor transmission of materials. Standards Des- 1265–1272. https://doi.org/10.1175/
ignation: E96-80. In: Annual book of ASTM. JAMC-D-17-0334.1
ASTM, Philadelphia, pp 771–778 7. ASME (1968) The 1967 ASME steam tables.
2. McHugh TH, Avena-Bustillos R, Krochta JM Nav Eng J. https://doi.org/10.1111/j.
(1993) Hydrophilic edible films: modified pro- 1559-3584.1968.tb04564.x
cedure for water vapor permeability and expla- 8. W A Hammond Drierite (n.d.) Technical data-
nation of thickness effects. J Food Sci 58:899– regeneration of drierite desiccants. https://
903 secure.drierite.com/catalog3/page19b.cfm#:
3. Zang L (2016) Lecture 4: diffusion: Fick’s sec- ~:text=Exhausted%20Commercial%20
ond law. https://my.eng.utah.edu/~lzang/ DRIERITE%20or%20Du,container%20and%
images/lecture-4.pdf. Accessed 16 July 2020 20sealed%20while%20hot. Accessed
4. Morillon V, Debeaufort F, Capelle M et al 16 July 2020
(2000) Influence of the physical state of water 9. W A Hammond Drierite (2019) Drierite
on the barrier properties of hydrophilic and [Material Safety Data Sheet]. https://secure.
hydrophobic films. J Agric Food Chem 48: drierite.com/RegularDrieriteSDS.pdf.
11–16. https://doi.org/10.1021/jf990809z Accessed 16 July 2020
5. King Abdulaziz University (n.d.) Molecular 10. Dow Corning (2002) High vacuum grease
mass transport. https://www.kau.edu.sa/ [Material Safety Data Sheet]. https://www.
Files/0001424/Subjects/note1%20333.pdf. conncoll.edu/media/website-media/offices/
Accessed 16 July 2020 ehs/envhealthdocs/High_Vacuum_Grease.
6. Huang J (2018) A simple accurate formula for pdf. Accessed 16 July 2020
calculating saturation vapor pressure of water
Chapter 12
Permeation of Oxygen and Carbon Dioxide Through Food

Packaging Materials
Victor G. L. Souza, Carolina Rodrigues, João R. A. Pires, Ana L. Fernando,
Vitor Alves, and Isabel Coelhoso
Abstract
Barrier properties of packaging materials are important requirements when selecting and developing
optimal packaging systems. The determination of the barrier properties of a polymer film is crucial to
estimating and predicting the product-packaging shelf-life. The specific barrier requirements of the package
are related to the product characteristics and the intended end-use application. Much of the design of
barrier packaging involves controlling the exchange of gaseous components (e.g., O2, CO2) between the
external and internal package environments. Thus, this chapter intends to give an overview of the methods
and equipment used for the evaluation of the permeation of oxygen and carbon dioxide through polymer-
based packaging films.
Key words Permeation of gases, Oxygen, Carbon dioxide, Permeability, Transmission rate, Barrier
properties
1 Introduction
Packaging materials may prevent the transport of oxygen (O2) and

keep the desired balance of oxygen and carbon dioxide (CO2)
within the headspace for extended food shelf-life [1]. Thus, when
a polymer film has low oxygen permeability, the oxygen pressure
inside the container drops, retarding oxidation reactions and respi-
ration rate (in the case of fruits and vegetables), extending the shelf-
life of the product. Carbon dioxide transport is also important for
packaging with modified active atmosphere (MAP) technology and
for products that release carbon dioxide over time, as is also the case
with fruits and vegetables [2].
Food can be packaged using different materials, such as plastic,
paper and cardboard, metal, and glass. Polymers show important
advantages over traditional packaging materials (glass, paper, and
metal) namely flexibility, lightweight, toughness, versatility, and
https://doi.org/10.1007/978-1-0716-3613-8_12,
219
cost. However, unlike glass and metals, polymers do not offer an

infinite gas barrier [1]. Combinations of different polymers, in the
form of multilayer structures or blends, can provide sufficient bar-
rier properties for the intended shelf-life of most products [3]. Poly-
mers with inorganic/organic materials in nanoscale dimensions,
such as clays, metal oxides, and cellulose, can boast significantly
enhanced barrier properties [4–7].
Gas transport through polymers is described by the solution-
diffusion model. According to this model, permeation through a
flat sheet or film occurs in three steps: permeant (i) dissolves into
the upstream side of the film (where there is a high permeant partial
pressure or high thermodynamic activity), then (ii) diffuses through
the film, and (iii) desorbs from the downstream side of the film
(where there is a low permeant partial pressure or low thermody-
namic activity). In one dimension, gas diffusion through a polymer
follows the first Fick’s law:
dC
J = -D ð1Þ
dx
where J is the gas flux (mol m-2 s-1), D is the effective diffusion
coefficient for the gas in the polymer (m2 s-1), and dC/dx is the
local concentration gradient of the gas (mol m-4).
Permeability (P) can be expressed as the product of the effective
diffusion coefficient, D, and the solubility coefficient, S (mol m-
3
Pa), the latter being the ratio of the equilibrium gas concentration
in the polymer at the upstream side of the film C (mol m-3) divided
by the permeant partial pressure p (Pa), resulting in the following
equation:
C =S p ð2Þ
Substituting Eq. 2 in Eq. 1 and integrating through the film
thickness (l) results:
P
J= ðp 1 - p 2 Þ ð3Þ
l
The permeability characterizes the steady-state rate of mass trans-
port of gas molecules through polymers. In a dense polymer film,
the permeability, P, is defined as the molar flux of gas through the
polymer relative to a fixed coordinate system, J, normalized by the
film thickness, l (m), and by the difference between the gas partial
pressures in the upstream ( p1) and downstream compartments
( p2).
Accordingly, permeability has dimensions of quantity of gas
(either mass or moles or volume) times thickness divided by area,
time, and pressure. Several units have been used to report the
permeability of gases in the literature (see Note 1). Though SI
units are mol m m-2 s-1 Pa-1, barrer [8] is a commonly used
non-SI unit for the permeability of gases, which was originally
Gas Permeation Through Food Packaging 221
defined for convenience because many polymer membranes had

permeabilities around 1 barrer. It is defined as 10-10 cm3 (STP)
cm/cm-2 s-1 cm Hg-1, where STP refers to standard temperature
and pressure (273.15 K and 105 Pa) and 1 barrer is equal to
3.35 × 10-16 mol m m-2 s-1 Pa-1 [9].
Oxygen and carbon dioxide are two of the main permeants
studied in food packaging applications. The oxygen and carbon
dioxide barriers are quantified by their permeabilities (OP and
CO2P, respectively), which indicate the amount of gas that perme-
ates per unit of area, time, and pressure multiplied by the film
thickness. Together with the permeability, the oxygen and carbon
dioxide transmission rates (OTR and CO2TR, respectively) express
the quantity of gas passing through a unit area of a film per unit
time under the conditions of the test. The SI units of transmission
rate are mol m-2 s-1. The permeability (P) is correlated to the
transmission rate (TR) by the following equation:
TR
P =l ð4Þ
ΔP
where l is the thickness of the film (m), ΔP = p1 - p2 is the
difference between gas partial pressure across the film (Pa), where
p1 is the gas partial pressure at the temperature test on the feed side
and p2 is the gas partial pressure at the temperature test on the
permeate side.
The permeability through plastic materials depends on several
factors, namely, (i) the permeant properties (molecule size and
chemical nature); (ii) material/polymer characteristics (molecular
orientation, degree of crystallinity, free volume, and chain stiffness);
and (iii) external conditions (such as temperature and relative
humidity) [10]. The temperature and humidity conditions to
which the food product will be exposed in the supply chain are
vital for calculating the required barrier, so that it applies to the
conditions expected. Furthermore, the specific barrier requirement
of a packaging system depends upon the food characteristics and
the intended end-use applications [11, 12].
The temperature dependence of permeability and diffusivity are
usually modeled using the Arrhenius equations of the following
forms:
P = P 0 exp - E p =RT ð5Þ
D = D 0 expð- E D =RT Þ ð6Þ
where Ep and ED are activation energies for permeation and diffu-
sion, and P0 and D0 are pre-exponential factors.
The effect of temperature on solubility is usually expressed by
the van’t Hoff relationship:
S = S 0 expð- ΔH s =RT Þ ð7Þ
where S0 is a pre-exponential factor, and ΔHs is the heat of sorption

of permeant in the polymer.
The presence of polar groups in the polymer chains often
increases chain rigidity, which can increase glass transition temper-
ature and improve gas barrier properties. Increasing crystallinity in
a polymer generally decreases gas permeability. Crystallinity influ-
ences both solubility and diffusion coefficients. The absorption of
water vapor can increase, decrease, or have no effect on the gas
permeability of barrier polymers. Hydrophilic barrier polymers,
except for certain amorphous polyamides, lose their barrier proper-
ties with increasing relative humidity (RH) [9]. Thus, this chapter
intends to give an overview of the methods and equipment used for
the evaluation of the permeation of oxygen and carbon dioxide
through polymer-based packaging films.
2 Materials
1. Polymer film with a 6.5-cm diameter for each replicate.

2. Stainless steel cell composed of two cylindrical chambers (e.g.,
7 cm in diameter and 5 cm in length each chamber), with
sealing O rings (e.g., Viton™ rubber) and a porous stainless-
steel support, equipped with high-pressure stainless-steel fit-
tings, valves, and tubing.
3. Two pressure transducers.
4. Thermostatic bath with a 2-L volume, a temperature range of
20–100 °C with a precision of ±0.03 °C.
5. Micrometer (precision of ±0.001 mm).
6. High-purity grade carbon dioxide (99.998%), oxygen
(99.999%), and nitrogen (99.999%).
7. A computer for data acquisition and the LabView software from
National Instruments.
3 Methods
There are two basic methods for measuring permeability: isostatic

and quasi-isostatic. Isostatic methods use a continuous flow on
both sides of the polymer film to provide constant permeant con-
centrations (Fig. 1a). Quasi-isostatic methods, in turn, use a con-
tinuous flow to maintain constant penetrant concentration only on
the upstream side and allow penetrant accumulation on the down-
stream side of the film (Fig. 1b).
The time-lag method is one of the most employed for measur-
ing the permeability of gases through films and membranes [13]. It
allows the determination of both the permeability and diffusion
Fig. 1 Schematic representations of cells for measuring the permeability in

isostatic (a) and quasi-isostatic (b) modes
coefficient of pure gases in a polymer matrix and, indirectly, the

solubility coefficient.
Before the system reaches a steady state, the flux across the film
and the pressure in the permeate compartment vary with time using
a quasi-isostatic method (Fig. 1b).
Representing the pressure of the gas as a function of time, when
t tends toward very long times, a steady state is reached and a
straight line is observed (Eq. 8).
RTApf SD l2
pt = t- ð8Þ
V pV m l 6D
where R is the universal gas constant (J mol-1 K-1), T is the
absolute temperature (K), A is the membrane area (m2), pf is
pressure in the feed compartment (Pa), Vp is the volume of the
permeate compartment (m3), Vm is the molar volume of the gas at
STP conditions (273.15 K and 105 Pa) (m3 mol-1), l is the film
thickness (m), and D (m2 s-1) is the diffusion coefficient.
From the interception of the time axis with the extrapolated
linear steady state, it is possible to obtain the time lag:
l2
θ= ð9Þ
6D
where is the time lag (s), which allows the determination of the
diffusion coefficient. From the slope of the straight line is obtained
the permeability, and since P = SD, the solubility coefficient can be
determined.
Another quasi-isostatic method for measuring gas permeability
is based on the diaphragm cell method for obtaining diffusion
coefficients. It is also known as a pressure decay method
[14]. The experimental apparatus (Fig. 2) is composed of a
stainless-steel cell with two identical chambers separated by the
testing film. During the measurement, two porous plates support
the polymer film on both sides to prevent sagging. However, the
plates allow the gas to freely contact the film.
The permeability is evaluated by pressurizing one of the cham-
bers (feed) up to constant pressure (e.g., 0.7 bar), with pure carbon
dioxide or oxygen followed by the measurement of the pressure
change in both chambers over time, using two pressure transdu-
cers. The measurements are made at a constant temperature, usu-
ally 30 °C, using a thermostatic bath. High-purity grade carbon
dioxide (99.998%) and oxygen (99.999%) are used. The permeabil-
ity is calculated with the pressure data obtained from both com-
partments according to the following equation [14]:
1 Δp0 t
ln =P ð10Þ
β Δp δ
30°C
P P
1 2
Purge Purge
Fig. 2 Experimental setup for measuring the permeability using the pressure decay method ((1) feed
compartment, (2) permeate compartment, (3) water bath, (4) thermostatic head, (5) feed gas)
where Δp0 and Δp (bar) are the pressure differences between feed
and permeate compartments at the beginning of the experiment
and at any time, respectively, P (m2 s-1) is the gas permeability,
t (s) is the time, δ (m) is the film thickness, and β is the geometric
parameter:
1 1
β=A þ ð11Þ
Vf Vp
where A is the film’s area (m2) and Vf and Vp are the volumes (m3)
of the feed and permeate compartments, respectively.
Δp
The data are plotted as 1β ln Δp0 versus δt , and the permeability
is determined from the slope.
The geometric factor of the cell can be obtained by calibration
of the equipment using a film with known permeability to an inert
gas (i.e., N2). It is normally used as a polydimethylsiloxane (PDMS)
film because of its low barrier to gases (e.g., N2), thus a low time
needed for permeation. Figure 3 shows the pressure of N2 in the
permeate and feed compartments during a calibration experiment.
Δp
Using the N2 permeability value (P) and plotting P1 ln Δp0
versus δt , the geometric parameter β is obtained from the
slope (Fig. 4).
The detailed protocol for measuring the O2 or CO2 permeabil-
ity at 30 °C is the following:
3.1 Calibration of the 1. Measure the PDMS film thickness.

Permeation Cell 2. Place the film inside the permeation cell and close it.
3. Introduce the permeation cell in the water bath regulated for
30 °C.
Permeate
0,30
Pressure (bar)
0,25
0,20
0,15
0,10 Pressure profile
0,05 0,8
Permeate Feed
Pressure (bar)
0,00
0 2000 4000 6000 8000 10000 12000 0,6
Time (s)
0,4
0,2
Feed
0,80 0,0
Pressure (bar)
0 2000 4000 6000 8000 10000 12000

0,60
Time (s)
0,40
0,20
0,00
0 2000 4000 6000 8000 10000 12000
Time (s)
Fig. 3 N2 pressure in the feed and permeate compartments of the cell during a calibration experiment
1E+10
9E+09
1/P ln(Dp0/Dp) (bar-1)

8E+09
7E+09
6E+09
5E+09
4E+09
3E+09
2E+09
1E+09
0
0 50000000 100000000
t/d (s/m)
Δp 0
Fig. 4 Representation of P1 ln Δp versus t
δ for obtaining the geometric
parameter β from the slope
4. Open all the valves and the N2 gas bottle.

5. Turn on the computer and start the acquisition program,
LabView.
6. With all the valves opened, open the N2 gas regulator up to
0.1 bar.
7. After 1 min, close the exit valves and increase the N2 pressure
up to 0.7 bar.
8. Close the entrance valves and wait 5 min to see if the pressure is
constant in both compartments.
9. Open and close the exit valve connected to the permeate
compartment.
10. Follow the evolution of pressure in both compartments (feed
and permeate) with time.
11. Using the N2 permeability value for that PDMS film given by
Δp0
the manufacturer and plotting P1 ln Δp versus δt , the geometric
parameter β is obtained from the slope.
3.2 O2 or CO2 1. Measure the film thickness.

Permeability of the 2. Place the film inside the permeation cell and close it.
Film
3. Introduce the permeation cell in the water bath regulated for
30 °C.
4. Open all the valves and the O2 or CO2 gas bottle.
5. Start the acquisition program.
6. With all the valves opened, open the O2 or CO2 gas regulator
up to 0.1 bar.
7. After 1 min, close the exit valves and increase the O2 or CO2
pressure up to 0.7 bar.
Table 1
Water activity of saturated saline solutions at 25 °C
Saturated salt solution Water activity

LiCl 0.115
CH3COOK 0.225
MgCl2·6H2O 0.324
K2CO3 0.447
Mg (NO3)2 0.520
NaNO2 0.649
NaCl 0.769
(NH4)2SO4 0.806
BaCl2 0.920
K2SO4 0.977
8. Close the entrance valves and wait 5 min to see if the pressure is
constant in both compartments.
9. Open and close the exit valve connected to the permeate
compartment.
10. Follow the evolution of pressure in both compartments (feed
and permeate) with time.
Δp
11. Plot the data as 1β ln Δp0 versus δt , and the permeability is
determined from the slope.
3.3 Conditioning of Film equilibration is usually carried out by placing samples in

Hydrophilic Films at desiccators containing at the bottom a saturated salt solution with
Constant Relative a known water activity (aw) value (see Note 2). Different relative
Humidity humidity values may be achieved by using different saturated salt
solutions. The aw of the saturated salt solutions at 25 °C is shown in
Table 1 [15]. The relative humidity, RH = aw × 100. The equili-
bration process is complete when the film mass remains unchanged
over time.
3.4 Standard The most used standard methods for measuring OTR are ASTM
Methods for Oxygen D3985, ASTM F1927, and ASTM F2622 [16–18]. They use an
and Carbon Dioxide isostatic permeation measurement with a flow-through technique
Permeation as represented in Fig. 5.
The differences between standards are related to:
(a) Test gas.
(b) Type of samples.
(c) Method and type of sensor.
(d) Measurement conditions.
Fig. 5 Isostatic permeation measurement of O2 using a flow-through technique
ASTM D3985
This test method covers a procedure for the determination of the
steady-state rate of transmission of oxygen gas through plastics in
the form of film, sheeting, laminates, coextrusions, or plastic-
coated papers or fabrics. It provides the determination of
(i) OTR, (ii) the permeance of the film to oxygen gas (PO2), and
(iii) OP in the case of homogeneous materials. It relies on a coulo-
metric sensor and uses nitrogen as a carrier gas. It applies to dry
conditions at a constant temperature.
ASTM F1927
rate of transmission of oxygen gas, at a steady state, at a given
temperature, and at %RH level, through film, sheeting, laminates,
coextrusions, or plastic-coated papers or fabrics. This test method
extends the common practice of dealing with zero humidity to a
controlled relative humidity. Humidity plays an important role in
the OTR of many materials, in particular hydrophilic polymers.
ASTM F2622
steady-state rate of transmission of oxygen gas through plastics in
the form of film, sheeting, laminates, coextrusions, or plastic-
coated papers or fabrics. It uses various sensors, including coulo-
metric, electrochemical, and zirconium oxide. The standard meth-
ods most used for measuring CO2TR are ASTM F2476 and DIN
53380-4 [19, 20].
ASTM F2476
This method covers a procedure for the determination of the
steady-state rate of transmission of carbon dioxide gas through
plastics in the form of film, sheeting, laminates, coextrusions, or
plastic-coated papers or fabrics. It provides for the determination of
(i) CO2TR, (ii) the permeance of the film to carbon dioxide gas
(PCO2), and (iii) CO2P in the case of homogeneous materials. It
uses an infrared sensor and nitrogen as a carrier gas. It applies to dry

conditions at a constant temperature.
DIN 53380-4
This method covers a procedure for the determination of the
steady-state rate of transmission of carbon dioxide gas through
plastics in the form of film, sheeting, laminates, coextrusions, or
plastic-coated papers or fabrics.
Instruments for Measuring O2 and CO2 Permeation

MOCON [21] has commercial instruments for measuring oxygen
transmission rates of flat films and packages being the Ox-Tran®
analyzers the most popular. Measurements are made following
ASTM method D3985. In this isostatic coulometric method, flat
film samples are clamped into a diffusion cell, which is then purged
of residual oxygen using an oxygen-free carrier gas such as N2. The
carrier gas is routed to the instrument sensor until a stable zero has
been established. Pure oxygen is then introduced into the outside
chamber of the diffusion cell. Oxygen molecules diffusing through
the film to the inside chamber are conveyed to the sensor by the
carrier gas. The Ox-Tran system uses a patented coulometric sensor
(Coulox®) to detect oxygen transmission through both flat films
and packages.
Modern Controls, Inc. (MOCON) also makes instruments for
measuring carbon dioxide permeation. Their Permatran-C® line of
instrument analyzer can test the CO2TR on both films and
packages from low to high barriers. MOCON’s patented modu-
lated infrared sensor is the only modulated sensor system on the
market meeting ASTM F2476 and capable of measuring low
CO2TR detection levels down to 0.5 cm3 m-2 d-1.
Labthink [22] also has gas permeability instruments for O2 and
CO2, which are described in Table 2.
Table 2
Gas permeability instruments for O2 and CO2
Testing instrument Testing method Test gas Type of packaging

C230, OX2/231 Coulometric sensor method O2 Films and sheeting, packages
(equal pressure)
VAC Series Differential pressure O2, N2, and CO2, Films and sheeting
special gases
G2/131, G2/132 Differential pressure O2, N2, and CO2 Films and sheeting, packages
G2/130 Differential pressure O2, N2, and CO2 Packages
4 Notes
1. Permeability units
A wide variety of units are used for gas permeability due to
the variation in industries measuring permeability. Other com-
mon units of gas permeability include:
• cm3 (STP) mil in-2 (100 d)-1 atm-1, where 1 mil is equal to
0.00254 cm and 1 d is 24 h.
• ft3 (STP) mil ft-2 d-1 atm-1, where 1 ft is equal to
0.3048 cm.
• m3 (STP) m m-2 s-1 Pa-1.
The permeability in Eq. 8 is given in m2 s-1 because the
solubility coefficient is dimensionless. In order to convert 1 bar-
rer to m2 s-1, it is necessary to use the molar volume at STP
conditions and multiply by RT resulting in 1 barrer equal to
8.4 × 10-5 m2 s-1 at 30 °C where 1 barrer is equal to
3.35 × 10-16 mol m m-2 s-1 Pa-1.
2. Permeation in hydrophilic films
As the oxygen and carbon dioxide permeabilities of hydro-
philic films are highly dependent on their adsorbed moisture
content, they should be previously equilibrated at constant
relative humidity in desiccators with saturated saline solutions:
LiCl, CH3COOK, MgCl2·6H2O, K2CO3, Mg(NO3)2,
NaNO2, NaCl, (NH4)2SO4, BaCl2, and K2SO4 (Table 1).
The barrier properties of different hydrophilic materials
should be compared under the same relative humidity and
temperature conditions. The CO2 and O2 permeability of
hydrophilic films, namely, from polysaccharides and proteins,
are extremely dependent on their water content, as observed by
several authors. Gontard et al. [23] reported for wheat gluten
films an increase of the oxygen permeability of about 950 times
for a change of the film water content from 7.5% to 42% (dry
basis) and an increase of nearly 36,550 times on the CO2
permeability for a variation of the relative humidity of the
atmosphere in contact with the film from 60% to 95%. It is
difficult to compare the results presented by different authors
unless the exact film water content or testing %RH for each
study is specified.
Acknowledgments
This research was funded by national funding from FCT, Founda-

tion for Science and Technology, through the individual research
grant (SFRH/BD/144346/2019) of J.P. and the individual
research grant (2020.04441.BD) of C.R. It was supported by
national funds from FCT/MCTES (UID/AGR/04129/2020)

and by MEtRICs—Mechanical Engineering and Resource Sustain-
ability Center, which is financed by national funds from
FCT/MCTES (UIDB/04077/2020 and UIDP/04077/2020).
Finally, this work has been supported by the Associate Laboratory
for Green Chemistry—LAQV, which is financed by national funds
from FCT/MCTES (UIDB/50006/2020 and UIDP/50006/
2020).
References
1. Robertson GL (2006) Food packaging: princi- 269–276. https://doi.org/10.1016/j.
ples and practice, 2nd edn. CRC Press, Boca carbpol.2009.08.002
Raton 11. Siracusa V (2012) Food packaging permeabil-
2. Vaclavik VA, Christian EW (2014) Essentials of ity behaviour: a report. Int J Polym Sci 2012:1–
food science, 4th edn. Springer Science+Busi- 11. https://doi.org/10.1155/2012/302029
ness Media, New York 12. Souza V, Pires JJRA, Torrico É et al (2019)
3. Mohamed SAA, El-Sakhawy M, El-Sakhawy Activity of chitosan-montmorillonite bionano-
MAM (2020) Polysaccharides, protein and composites incorporated with rosemary
lipid-based natural edible films in food packag- essential oil: from in vitro assays to application
ing: a review. Carbohydr Polym 238:116178. in fresh poultry meat. Food Hydrocoll 89:241–
https://doi.org/10.1016/j.carbpol.2020. 252. https://doi.org/10.1016/j.foodhyd.
116178 2018.10.049
4. Souza VGL, Pires JRA, Rodrigues C et al 13. Crank J (1975) The mathematics of diffusion,
(2019) Physical and morphological characteri- 2nd edn. Clarendon Press, Oxford
zation of chitosan/montmorillonite films 14. Cussler EL (2009) Diffusion: mass transfer in
incorporated with ginger essential oil. Coatings fluid systems, 3rd edn. Cambridge University
9(11):700. https://doi.org/10.3390/ Press
coatings9110700 15. Hong TD, Edgington S, Ellis RH et al (2005)
5. Souza VGL, Rodrigues C, Valente S et al Saturated salt solutions for humidity control
(2020) Eco-friendly ZnO/chitosan bionano- and the survival of dry powder and oil formula-
composites films for packaging of fresh poultry tions of Beauveria bassiana conidia. J Invertebr
meat. Coatings 10(2):110. https://doi.org/ Pathol 89:136–143. https://doi.org/10.
10.3390/coatings10020110 1016/j.jip.2005.03.007
6. Pires JRA, Souza VL, Fernando AL (2019) 16. ASTM D3985-17 (2017) Standard test
Valorization of energy crops as a source for method for oxygen gas transmission rate
nanocellulose production–current knowledge through plastic film and sheeting using a cou-
and future prospects. Ind Crop Prod 140: lometric sensor. ASTM International, West
111642. https://doi.org/10.1016/j.indcrop. Conshohocken
2019.111642 17. ASTM F1927-20 (2020) Standard test method
7. Tian F, Chen W, Wu CE et al (2019) Preserva- for determination of oxygen gas transmission
tion of Ginkgo biloba seeds by coating with rate, permeability and permeance at controlled
chitosan/nano-TiO2 and chitosan/nano-SiO2 relative humidity through barrier materials
films. Int J Biol Macromol 126:917–925. using a coulometric detector. ASTM Interna-
https://doi.org/10.1016/j.ijbiomac.2018. tional, West Conshohocken
12.177 18. ASTM F2622-20 (2020) Standard test method
8. Barrer RM (1941) Diffusion in and through for oxygen gas transmission rate through plas-
solids. Macmillan, New York tic film and sheeting using various sensors.
9. Mulder M (2003) Basic principles of mem- ASTM International, West Conshohocken
brane technology. Kluwer Academic Publish- 19. ASTM F2476-20 (2020) Standard test method
ers, Springer Netherlands for the determination of carbon dioxide gas
10. Alves VD, Costa N, Coelhoso IM (2010) Bar- transmission rate (CO2TR) through barrier
rier properties of biodegradable composite materials using an infrared detector. ASTM
films based on kappa-carrageenan/pectin International, West Conshohocken
blends and mica flakes. Carbohydr Polym 79:
20. DIN 53380-4:2006-11 (2006) Testing of 22. Packaging Testing Instruments & Testing
plastics – determination of gas transmission Services – Labthink International. https://
rate – Part 4: Carbon dioxide specific infrared www.labthinkinternational.com/
absorption method for testing of plastic films 23. Gontard N, Thibault R, Cuq B, Guilbert S
and plastic mouldings. German Institute for (1996) Influence of relative humidity and film
Standardization, Berlin composition on oxygen and carbon dioxide
21. Gas Measurement Instruments & Testing Ser- permeabilities of edible films. J Agric Food
vices | AMETEK MOCON. https://www. Chem 44:1064–1069. https://doi.org/10.
ametekmocon.com/ 1021/jf9504327
Chapter 13
Microbial Permeation Through Food Packaging Materials

Julia V. Ernesto, Patricia Severino, Anna C. Venturini,
Cristiana M. P. Yoshida, Classius F. da Silva, and Patricia S. Lopes
Abstract
Microbial permeation is an essential property in developing food packaging materials. Here we describe a
simple method to determine it by placing the film on open vials containing nutrient broth. Negative and
positive vials are also provided in this method. The tested vials are placed in an open environment for about
ten days. The turbidity of the nutrient broth in any vial is recorded as microbial contamination, meaning the
microorganism could permeate through the packaging material. This straightforward protocol can be
invoked by material developers targeting microbial-tight packaging.
Key words Permeation of microorganisms, Turbidity, Flexible packaging, Food innocuity, Food
safety, Microorganism-proof packaging
1 Introduction
The microbial permeation assay allows analyzing whether a given

packaging system constitutes a barrier to microorganisms’ passage,
preventing contamination. The primary roles of food packaging
materials are to provide physical protection, prevent post-
processing contamination, extend shelf-life, and communicate
information to consumers [1]. The properties of packaging materi-
als play a significant role in food quality and safety. It is essential to
control the microbial barrier protection to prevent spoilage and
pathogenicity of food products.
Among the aforementioned functions of food packaging, pro-
tecting food from microbiological contamination after processing
stands out. Disrupting the hermeticity of the packaging system may
allow microbial contamination. An example is when sealing or
closing such systems, which could be solved by adjusting machinery
or revising standard protocols. Classic materials such as glass, rigid
plastics, and metals are used in food packaging and usually do not
pose risks of microbial permeation when used according to
https://doi.org/10.1007/978-1-0716-3613-8_13,
233
234 Julia V. Ernesto et al.
pre-established processing standards. Although these traditional

materials have excellent barrier properties against microbial perme-
ation, in addition to suitable mechanical properties, most of these
materials have very slow biodegradation, lasting in nature for cen-
turies. Thus, such materials might represent an environmental
problem for the planet if discarded incorrectly. New biodegradable
materials based on biopolymers have been tested in food packaging
to circumvent these low biodegradation rates.
However, flexible packaging, particularly those from biopoly-
mers, still faces challenges concerning mechanical and barrier prop-
erties (against, e.g., water vapor and microorganisms). The barrier
properties of a packaging system are closely related to the chemical,
physical, sensory, and microbiological stability of the product. Spe-
cifically, in the case of food packaging, the methodologies for the
quantitative determination of gas (e.g., oxygen and carbon dioxide;
see Chap. 12) and moisture (see Chap. 11) permeabilities are well
standardized, even with specific commercial equipment for measur-
ing these properties. On the other hand, there is no standardized
methodology for the quantitative determination of microbial per-
meation through food packaging. Blocking the passage of small
molecules such as gases and water vapor is intuitively harder for
packaging than it is for larger objects such as microbial cells. How-
ever, porous or defective films could allow microorganisms to pass
through packaging, therefore leading to the problems pointed out
above.
As far as biopolymer-based packaging, it is to be considered
that biopolymers are usually hydrophilic. According to Mondal and
Hu [2], membranes (or films) of hydrophilic polymers can absorb
water very quickly and create a “wicking” action that attracts water
molecules and then enables the transmission of steam through an
active diffusion mechanism. In addition, the thickness and chemical
structure are determinants in the permeability of a nonporous
membrane (or film).
Also, biopolymers can present amorphous and crystalline
domains (i.e., semicrystalline) in different proportions. Polymer
chains do not pack perfectly, and there is some unoccupied space
between them. The amount of many small spaces between the
polymer chains in amorphous, noncrystalline materials is the free
volume. Even if this volume accounts for only a small portion of the
overall volume, it is sufficient to allow some rotation of polymer
backbone segments. Therefore, a dense polymer may be thought of
as a “porous medium,” with the local free volume serving as the
“pores.” Fluid transport through porous media is identical to
penetrant transport through a dense polymer membrane (or film).
The distinction between these two scenarios is that the size and
location of “pores” inside the dense polymer membrane change
with time, implying that a penetrant will pass through a “pore” with
a certain probability. As a result, it is fair to consider a dense
Microbial Permeation through Packaging 235
polymer membrane to be a porous material, with pores identified as

gaps in the polymer matrix [3].
The barrier against microbial permeation is a topic of great
importance in applying plastic films for food, and the methodolo-
gies used to evaluate microbial permeation are scarce. However,
methods for determining microbial permeation are described in the
development of films and membranes used as wound dressings
because microbial contamination represents a major problem in
wound healing. Such methods may be extended to flexible packag-
ing. Several articles report methods for determining microbial per-
meation in dressings [4–11], most of these relying on the
methodology described by Wittaya-Areekul and Prahsarn [12],
which has also been used to assess microbial permeation in food
packaging [13, 14].
The method described by Wittaya-Areekul and Prahsarn [12]
consists of fixing the packaging material on the top of a vial contain-
ing a nutrient broth. This vial is exposed to the environment and
analyzed after a few days. As the vial and broth are previously
sterilized, any turbidity of the broth indicates that microorganisms
could permeate through the material. On the other hand, if there is
no broth turbidity, the material is considered impermeable to
microorganisms. These techniques can be applied to (bio)-
polymeric films and even flexible composites or any materials that
can be fixed on the top of the vial. Several advantages can be
mentioned for this technique, such as low cost, unsophisticated
equipment, and the possibility of evaluating the permeation of
specific microorganisms by using different broths, from the sim-
plest to the most sophisticated. Yet, like most microbiological
analyses, this method is not rapid; instead, it takes a few days to
thoroughly assess the microbial permeation through the material.
2 Materials
2.1 Experimental 1. Ultrapure or deionized water (see Note 2).

Devices and Sample 2. Glass flasks (vials) with a 100-mL capacity, sanitized and dried
Preparation (calculate enough for the number of replicates that will be
( See Note 1) carried out per sample). Penicillin flasks (Fig. 1a) are
recommended.
3. Graduated cylinders or glass beakers.
4. Nutrient broth—approximate formula (per liter): beef extract,
3 g; enzymatic digest of gelatin, 5 g; final pH, 6.8 (± 0.2 at
25 °C).
5. Autoclave (see Note 3).
6. Poly(vinyl chloride) (PVC) weldable unions (Fig. 1b)
with 25-mm-diameter opening (in sufficient quantity for the
Fig. 1 Materials for the determination of microbial permeation
number of replicates that will be carried out per sample, one set
per flask).
7. Gamma radiation source (25 kGy) or 70 vol% ethanol solution
plus sterilized phosphate buffer saline.
8. Laboratory auto-sealing, flexible, and moldable film, as
Parafilm®.
9. Tweezers and scissors sterilized by autoclaving.
2.2 Result Evaluation 1. Digital camera.

2. Absorbance reader using 96-well plates, from Biotek Synergy
HT or another manufacturer.
3. 3.5-Triphenyl tetrazolium (TTC) at 0.1% (w/v) diluted in a
sterile 0.9% (w/v) NaCl solution (see Note 4).
3 Methods
It is suggested that at least one triplicate be tested for each type of

packaging sample.
1. Prepare the liquid medium (nutrient broth) for microbial
growth (see Notes 5–8).
2. Perform steam sterilization by autoclaving glass flasks contain-
ing culture medium, usually for 15 min at 121 °C. A typical
standard for steam sterilization is achieved after 15 min under a
pressure of 106 kPa (1 atm) once all surfaces have reached a
temperature of 121 °C.
3. Cut the film samples into 5-cm-diameter disks (in the number
of replicates that will be carried out per sample).
4. Sterilize the sample disks, e.g., by exposure to ultraviolet radia-
tion, in laminar flow for 15 min on each side, or another
suitable sterilization method, such as ionizing radiation or
even ethylene oxide. Please be aware that you must perform

previous tests to ensure that the samples do not undergo
modifications after the sterilization process.
5. Hatch the sample disks between the PVC devices (as shown in
Fig. 2). Ensure that the sample does not present a tear or hole
from the manipulation.
6. Fit the PVC apparatus containing the sample attached to the
mouths of the flasks containing the culture medium (as shown
in Fig. 2).
7. Seal the apparatus and the upper part of the flasks laterally with
Parafilm® so that the packaging samples are the only commu-
nication between the culture medium in the flasks and the
external environment.
8. As controls, set six vial systems (triplicates for positive and
negative controls) comprising the PVC union and the flasks
filled with the culture medium. For the positive control, seal
these only laterally using Parafilm®; that is, these flasks will not
have barriers between the culture medium and the external
environment. For the negative control, seal the flask and
union entirely using Parafilm®; that is, there should be no
communication between the culture medium and the external
environment.
9. Expose the systems to environmental conditions, e.g., in the
lab (for specific microorganisms, see Notes 5–9) for 10 days,
depending on the study design. On days 1, 5, and 10, for
example, macroscopically assess the turbidity of the culture
medium, which indicates microbial growth (see Notes 10–12).
10. Take pictures of all systems throughout the evaluation period
(Fig. 3).
11. At the end of the test period, transfer aliquots from each flask
to 96-well culture plates (Fig. 4) and perform an absorbance
reading at 600 nm. To confirm the result, insert 100 μL of
TTC dye solution in each well. Incubate the plates for 1 h and
evaluate again using the plate reader at 540 nm (see Note 13).
4 Notes
1. Perform all the procedures in a laminar flow cabinet, preferably

a class II A1. The operator must be appropriately dressed,
wearing a cap, mask, gloves, and lab coat. Manipulate the
samples using sterilized tweezers.
2. Prepare all solutions using ultrapure water or deionized water,
according to the reagents and analytical grade.
Fig. 2 Schematic representation of the microbial permeation-measuring device:

poly(vinyl chloride) union (a) before (note the position of the sample specimen)
and (b) after assembling and attachment into the penicillin flask
Fig. 3 Visual aspect of the (a) positive- and (b) negative-control vials after
exposure to the laboratory environment
Fig. 4 Schematic representation of microbial penetration analysis
3. Perform the correct disposal or sanitization of all materials that

come into contact with the biological culture medium. Ensure
that all sterilized material for the experiment (when indicated)
has gone through the correct sterilization procedure, using
autoclaving sterilization tapes, for example.
4. The TTC dye solution must be prepared and immediately used.
The laminar flow cabinet lights must be off, and the plates must
be covered with aluminum foil until the reading process. This
solution is extemporary and must not be stored.
5. Add 50 mL of water to a 100-mL graduated cylinder or a glass
beaker. Weigh 0.4 g of nutrient broth and transfer it to a glass
flask of item 1. Add the water to the glass flask. Perform the
same procedure for each flask used in the test.
6. Should the purpose be the evaluation of a specific class of
microorganisms, like anaerobic as the Clostridium sp. [15],
the culture medium can be switched to, e.g., Fluid Thioglycol-
late Medium or even Reinforced Clostridial Medium. After the
exposure period, the flasks must be incubated in anaerobiosis at
30–35 °C for 48 h. It is important to remember that the
Clostridium species can sporulate; that is, the spores, which
are forms of resistance of the microorganism that are released
to the environment when the conditions are not conducive to
its growth, can be spread around the environment and can
reach the product if this packaging is not adequate. Anaerobi-
osis conditions could be assessed using an anaerobiosis cham-
ber or even a CO2 incubator.
7. The culture medium can be modified according to the type of
microorganisms to be evaluated in the test: if the intention is to
evaluate the permeation of fungi only, one can use the
Sabouraud Dextrose Broth. This medium is poor in nutrients

and limits the growth of bacteria, which are more nutritionally
demanding. For more demanding microorganisms, the brain
and heart infusion broth or even the tryptic soy broth can be
used instead.
8. It is important to note that, if the intention is to evaluate a
specific microorganism that is not usually found in the lab
environment, the target microorganism can be inoculated
directly onto the packaging material at the top of the system
(induced test). When this method is used, it is essential that,
once the microorganism is inoculated, the upper PVC union is
capped with aluminum foil and that the systems are incubated
at an appropriate temperature. This allows the evaluation of a
specific packaging behavior.
9. To evaluate specific microorganisms, the usual concentration is
103–106 CFU cm-2 depending on the microorganism type
[16]. The use of microorganisms obtained from a culture
collection, like ATCC (American Type Culture Collection) or
NIH (National Institute of Health), is preferable to the use of
wild microorganisms or even those isolated from humans or
animals. To obtain standard cultures, use methods like dilution
followed by counting or turbidity assays using an absorbance
reader or even MacFarland scale.
10. The exposure times may vary, depending, for example, on the
shelf-life or transportation conditions of the target products, so
that the real situation of the product is simulated. In this
context, other parameters such as temperature and humidity
can be adjusted to mimic the targeted environmental
conditions.
11. The herein described test was modified from Wittaya-Areekul
and Prahsarn (2006), similar to Augustine et al. [17]. The
macroscopic evaluation of the positive control flasks ensures
that the nutrient broth was suitable for microbial growth and
could represent a free-condition system. In contrast, the nega-
tive control flasks are tested to illustrate the efficiency of the
sterilization process and to ensure that all the microorganisms
observed in the culture medium of the devices containing the
samples arose from the external environment and must have
passed through the packaging material.
12. The packaging is microorganism-proof when microbial growth
is evidenced in the positive control, but all the tested mem-
branes prevent any visible microbial contamination.
13. Other methods can be used to evaluate the results, for example,
the preparation of slides with the contaminated medium for
identifying the microorganism followed by the Gram staining
process and then by molecular assays (polymerase chain reac-
tion) or even mass spectroscopy.
References
1. Singh P, Wani AA, Langowski H-C (2017) methylcellulose and hydroxypropyl cellulose
Food packaging materials: testing & quality polymers for wound dressing. Int J Pharm
assurance, 1st edn. CRC Press Investig 6:86. https://doi.org/10.4103/
2. Mondal S, Hu JL (2006) Segmented shape 2230-973x.177810
memory polyurethane and its water vapor 10. Tomoda BT, Corazza FG, Beppu MM et al
transport properties. Des Monomers Polym (2020) Silk fibroin membranes with self-
9(6):527–550. https://doi.org/10.1163/ assembled globular structures for controlled
156855506778944028 drug release. J Appl Polym Sci 137:48763.
3. Chen C, Han B, Li J, Shang T, Zou J, Jiang W https://doi.org/10.1002/app.48763
(2001) A new model on the diffusion of small 11. Oliveira RN, McGuinness GB, Rouze R et al
molecule penetrants in dense polymer mem- (2015) PVA hydrogels loaded with a Brazilian
branes. J Membr Sci 187(1–2):109–118. propolis for burn wound healing applications. J
https://doi.org/10.1016/S0376-7388(00) Appl Polym Sci 132. https://doi.org/10.
00689-X 1002/app.42129
4. Singh B, Sharma S, Dhiman A (2013) Design 12. Wittaya-Areekul S, Prahsarn C (2006) Devel-
of antibiotic containing hydrogel wound dres- opment and in vitro evaluation of chitosan-
sings: biomedical properties and histological polysaccharides composite wound dressings.
study of wound healing. Int J Pharm 457:82– Int J Pharm 313:123–128. https://doi.org/
91. https://doi.org/10.1016/j.ijpharm.2013. 10.1016/j.ijpharm.2006.01.027
09.028 13. de Silva FM, Lopes PS, da Silva CF et al (2016)
5. Singh B, Kumar A (2016) Radiation formation Active packaging material based on buriti oil –
of functionalized polysaccharide-protein based Mauritia flexuosa L.f. (Arecaceae) incorporated
skin mimicking semi-inter penetrating network into chitosan films. J Appl Polym Sci 133.
for biomedical application. Int J Biol Macro- https://doi.org/10.1002/app.43210
mol 92:1136–1150. https://doi.org/10. 14. Rodrigues MÁV, Bertolo MRV, Marangon CA
1016/j.ijbiomac.2016.08.011 et al (2020) Chitosan and gelatin materials
6. El-Hosary R, El-Mancy SMS, El-Deeb KS et al incorporated with phenolic extracts of grape
(2020) Efficient wound healing composite seed and jabuticaba peel: rheological, physico-
hydrogel using Egyptian Avena sativa chemical, antioxidant, antimicrobial and bar-
L. polysaccharide containing β-glucan. Int J rier properties. Int J Biol Macromol 160:769–
Biol Macromol 149:1331–1338. https://doi. 779. https://doi.org/10.1016/j.ijbiomac.
org/10.1016/j.ijbiomac.2019.11.046 2020.05.240
7. Genevro GM, Gomes Neto RJ, Beppu MM 15. Taclindo C, Nygaard GS, Bodily HL (1967)
et al (2019) Glucomannan asymmetric mem- Examination of prepared foods in plastic
branes for wound dressing. J Mater Res 34: packages for Clostridium botulinum. Appl
481–489. https://doi.org/10.1557/jmr. Microbiol 15:426–430. https://doi.org/10.
2018.463 1128/aem.15.2.426-430.1967
8. Salehi M, Niyakan M, Ehterami A et al (2020) 16. Spanu C, Scarano C, Ibba M et al (2014)
Porous electrospun poly(ε-caprolactone)/gel- Microbiological challenge testing for Listeria
atin nanofibrous mat containing cinnamon for monocytogenes in ready-to-eat food: a practical
wound healing application: in vitro and in vivo approach. Ital J Food Saf 3:231–237. https://
study. Biomed Eng Lett 10:149–161. https:// doi.org/10.4081/ijfs.2014.4518
doi.org/10.1007/s13534-019-00138-4 17. Augustine R, Kalarikkal N, Thomas S (2014)
9. Prabu D, Majdalawieh A, Abu-Yousef I et al Advancement of wound care from grafts to
(2016) Preparation and characterization of bioengineered smart skin substitutes. Prog
gatifloxacin-loaded sodium alginate hydrogel Biomater 3:103–113. https://doi.org/10.
membranes supplemented with hydroxypropyl 1007/s40204-014-0030-y
Part III
Nontraditional Roles Played by Food Packaging Materials

Chapter 14
Do Not “Pack and Pray”: Use Predictive Models to Assess

the Microbial Safety and Shelf-Life of Modified Atmosphere
Packaged Foods
Arı́cia Possas, Fernando Pérez-Rodrı́guez, and Antonio Valero
Abstract
Besides protecting food from contact with the external environment and from airborne contamination,
packaging can also contribute to prolong food shelf-life and increase food safety. The gaseous environment
within a package can delay or inhibit microbial growth or modify the microbial ecology of the product.
Evaluating the microbiological safety and the shelf-life of foods through predictive microbiology tools
implies developing or selecting appropriate models that consider the impacts of different static and/or
dynamic concentrations of O2, CO2, and N2 on the behavior of spoilage and pathogenic microbiota. There
is an increasing demand for predictive models that would allow one to identify in advance the headspace gas
composition and the packaging material suitable for increasing food safety and shelf-life. The development
of such numerical tools would decrease the number of time-consuming challenge-testing experiments
necessary for experimentally quantifying and assessing the microbial fate in terms of changes in headspace
composition. The objective of the present chapter is to provide information on the adequate development,
validation, and interpretation of predictive microbiology models, including the effect of packaging atmo-
sphere to achieve more reliable estimates of microbial behavior in packaged foods. A case study on a
validated predictive model in cooked meat products is presented using the MicroHibro software tool.
Key words Predictive microbiology, Food packaging, Headspace composition, Shelf-life, Food safety,
Modeling, Gas transfer, Carbon dioxide, Predictive modeling
1 Introduction
Besides protecting foodstuffs from contact with external environ-

ments and airborne contamination, food packaging can also help
prolong food shelf-life and increase food safety [1]. The gaseous
mixture present in a modified atmosphere packaging (MAP) can
inhibit or slow down microbial growth in foods and even modify
their ecology [1]. To avoid the transfer of gases through packaging,
high-barrier films with multilayers are usually used in MAP
https://doi.org/10.1007/978-1-0716-3613-8_14,
245
246 Arı́cia Possas et al.
[2]. Such materials in many cases are oversized in terms of barrier

properties, as most of the time there is no knowledge a priori of the
specific requirements of the product regarding the gaseous mixture
necessary to avoid microbial spoilage or proliferation of foodborne
pathogens. Since high-barrier materials are usually expensive, MAP
represents direct costs for the food industry [1].
In this context, identifying the headspace gas composition and
the appropriate packaging material to assure a prolonged shelf-life
and the control of microbial pathogens in MAP is of great interest.
Predictive microbiology is the field of food microbiology aimed at
describing microbial processes in foods—i.e., growth, inactivation/
survival, and transfer—through the application of mathematical
models [3]. These models enable estimating microbial populations
in foods as functions of intrinsic, extrinsic, and implicit factors, such
as pH, aw, temperature, and microbial interactions, denoting rele-
vant tools for shelf-life estimation, product development, microbial
risk assessment, and aid in the compliance with microbiological
criteria established for foods [3–5]. The development of predictive
microbiology models, therefore, represents a breakthrough in the
understanding of the effects of different factors on microbial behav-
ior, including gaseous compounds present in MAP.
Traditionally, predictive models are classified as primary or
secondary models according to their response variable [6]. Primary
models are those describing the microbial kinetics, i.e., the evolu-
tion of microbial levels in foods over time at constant conditions.
These models are used to estimate microbial growth and survival/
inactivation parameters in foods (e.g., growth or survival rates, lag
times, and maximum population density). On the other hand,
secondary models relate changes in the primary model parameters
with intrinsic or extrinsic factors (e.g., temperature, aw, pH). Pri-
mary and secondary models as well as other modeling structures are
usually implemented in user-friendly predictive microbiology soft-
ware tools, the so-called tertiary models, to be applied in simula-
tions for decision making in the food industry. Existing software for
predictive microbiology have been reviewed by Tenenhaus-Aziza
and Ellouze [7] and more recently by Possas et al. [8]. Usually,
software users specify the values of the model factors such as storage
temperature, and the software provides the results of model simula-
tions as growth or survival/inactivation estimates.
Attention has been given to the impacts of the gaseous atmo-
sphere of food packaging on microbial behavior [5, 9–12]. Carbon
dioxide (CO2) is the most important component in the choice of a
gaseous mixture to be applied in MAP [1]. The inhibiting effect of a
constant CO2 concentration distributed homogeneously in the
headspace, together with the effects of food intrinsic and extrinsic
factors, has been explored in predictive microbiology studies [4, 13,
Predictive Models for MAP Foods 247
14]. The choice of using a point-estimate approach when dealing

with MAP gases can be associated with difficulties in quantifying
their dynamics, which may be associated with their solubilization
and diffusion into the food and mass transfer through the packag-
ing materials [4, 11, 15]. Therefore, although the integration of the
impact of gas dynamics in predictive models developed in MAP
foods would result in more realistic predictions of microbial behav-
ior, the development of such models remains a challenge
[4, 11]. Examples of models integrating the impact of gas dynamics
are the ones developed for Pseudomonas spp. and lactic acid bacteria
(LAB) in chicken fillets [11] and the one developed for Listeria
monocytogenes in processed cheese [16].
Comparison of results derived from different investigations
with MAP products is very difficult due to the use of packaging
materials with different permeabilities to O2 and CO2 (see
Chapter 12 for protocols to determine the gas barrier of food
packaging materials) and/or due to the lack of information
provided in studies regarding the gas/product (G/P) ratio used
in MAP [5]. To overcome the latter problem, instead of consider-
ing the concentration of CO2 in the headspace, some authors
considered more reasonable the evaluation of dissolved CO2 in
the water phase of the product in modeling studies [5, 10]. In
addition, the simultaneous effect of both O2 and CO2 on the
behavior of microorganisms in MAP foods has not been frequently
investigated in modeling studies [17–20], due to difficulties in
measuring O2 concentrations [4]. These variability sources imply
that predictive models, including the effect of MAP, should be built
based on extensive challenge testing studies to assure a proper
model validation.
For more information regarding predictive models built with
growth data obtained from MAP foods, see the review by Chaix
et al. [4]. The schematic flow for the development of predictive
microbiology models in foods is shown in Fig. 1. In Subheading
2 of this chapter, methods traditionally performed for data genera-
tion for model development, fitting, and validation are briefly
described based on information compiled from various studies. In
addition, we present an example of how predictive models can be
applied for shelf-life estimation, using our user-friendly predictive
microbiology software MicroHibro (www.microhibro.com) [21].
2 Methods
2.1 Data Generation Assuming that the objective of a study is to evaluate and model the
to Evaluate Microbial effects of MAP on the growth kinetics of a microorganism in a
Growth Kinetics in foodstuff, growth curves expressed as microbial concentrations
MAP Foods over time are developed at different conditions (e.g., temperature,
aw, pH) and various MAP configurations to cover a large range of
Experimental design
Growth data generation
Model fitting
Model validation
Model implementation in user-friendly software
Model application
Fig. 1 Schematic flowchart of the development of predictive microbiology

models in MAP foods
possibilities, following a previously defined experimental design (see

Note 1). For more information regarding guidelines for the per-
formance of challenge tests for model development, see ISO
20976-1 [22]. For data generation, representative samples of the
food product under evaluation are usually inoculated with the
target spoilage or pathogenic microorganisms.
The atmosphere of the packaging containing the inoculated
samples can be modified by means of gas packaging units, which
remove the air and insert the desired food-grade gas mixtures of
CO2, O2, and N2 in selected packaging materials. The proportions
of CO2, O2, and N2 in MAP depend on the product nature and the
ratio between the volume of the package and that of food [23]. For
instance, low levels of O2 are typically applied to reduce respiration
and the related quality loss of vegetables, while high levels of this
gas are used to stabilize the red color of fresh meat [18]. After
sealing, the packages are stored under the environmental condi-
tions set at the experimental design. Storage times are defined based
on the shelf-life of the evaluated product.
Inoculated packaged samples are analyzed immediately after
inoculation (time 0) and withdrawn and microbiologically analyzed
at proper time intervals by applying traditional methods for micro-
bial detection and enumeration, e.g., plate count method [22]. The
physicochemical characteristics of samples that will be included as
explanatory variables in the predictive model and other relevant
parameters are also monitored in samples at different time points
during shelf-life, e.g., aw, pH, and lactic acid content.
The headspace composition is measured immediately after

packing and at each sampling time using gas analyzers. Aliquots
of the headspace gas are collected with syringes after piercing
packaging materials with the aid of a septum, e.g., film cover. In
high-barrier packages, it is reasonable to assume that, once the
gaseous mixture in the package has reached equilibrium, the CO2
concentration in the headspace is proportional to the amount of
CO2 adsorbed by the product [24].
Different methodologies have been applied to estimate the
concentration of CO2 dissolved in the product as functions of the
initial CO2 concentration in the headspace (see Notes 2 and 3)
[9, 12]. For instance, the concentrations of the CO2 adsorbed by
the product samples have been estimated by assessing the volume
changes in the package headspace using a buoyancy technique and
performing calculations based on volumetric measures and Henry’s
constant [12, 24]. A quadratic polynomial model has been devel-
oped to estimate the CO2 dissolved in the water phase of cooked
meat products as a function of the initial CO2 concentration in the
headspace, the storage temperature, and the relation G/P volume
ratio [10].
The microbial data obtained in samples at different time points
under different conditions are then subjected to model fitting.
2.2 Model Selection Predictive models that consider the impact of a gas mixture on
and Fitting microbial growth are needed to predict the growth of microorgan-
isms in a MAP system. These models should be able to describe the
microbial patterns as influenced by the gaseous composition of the
product. An exponential-type law is usually used to describe micro-
bial kinetics in foods. Examples of predictive models used are
shown in Fig. 2. Different primary models have been proposed to
translate microbial growth kinetics in foods, including parameters
with biological meaning, being the Baranyi and Roberts model
[25], the modified Gompertz model [26], and the logistic model
[27], the most commonly applied ones. Four microbial kinetic
parameters can be estimated by fitting these models to growth
kinetic data: the initial and maximum population densities (N0
and Nmax, respectively), the lag time (lag), and the maximum
growth rate (μmax).
The growth parameters estimated through fitting primary
models to growth data are dependent on the environmental and
intrinsic factors evaluated (temperature, pH, aw, CO2, etc.). The
relationship between the estimated kinetic parameters and the eval-
uated factors can be described using secondary models. To this end,
the most applied mathematical equations are the Ratkowsky-type
or square-root models [28], cardinal models [29], polynomial
models [30], and artificial neural network (ANN) models
[31]. While polynomial and ANN models are purely empirical,
square root models and the models belonging to the “cardinal”
A 8
B
0.15
Microbial count (log cfu.g-1)
6 0.12
(h-1)
max
0.09
m
4
0.06
2
0 25 50 75 100 0 500 1000 1500 2000
Time (h) Dissolved CO2(ppm)
Fig. 2 Examples of mathematical models fitted to growth data. (a) Gompertz primary model; (b) Square root
secondary model
family have biological meaning, including parameters with

biological significance such as the minimum temperature required
for the growth of the target microorganism (Tmin). Besides their
mechanistic nature, Ratkowsky type and cardinal models can be
extended to account for the influence of multiple relevant factors
that affect microbial growth using the gamma concept approach
(see Note 4) [32]. More information regarding modeling struc-
tures and approaches and their advantages and drawbacks can be
found in Pérez-Rodrı́guez and Valero [3].
Regression methods are applied to estimate the parameters of
the selected models that result in a better description of the
observed data [3]. The most widely used regression method is the
least square method [3]. In addition, the one-step and two-step
regression analysis methods have been applied to fit models to
microbial data. In the first approach, primary models describing
the microbial kinetics as a function of time can be combined with
secondary functions describing the effect of environmental condi-
tions on microbial fate, and the combined model is fitted to the
observed data. In this way, the parameter estimation can be done
using a single-step procedure [18]. In the standard two-step mod-
eling regression analysis, first the primary models are fitted to the
microbial kinetic data, and then the secondary model is fitted to the
primary model parameters estimates. Both approaches can result in
accurate and precise models depending on the growth data used for
model development (see Note 5). Such parameter estimation
techniques can be applied when a set of static experiments is avail-

able [33]. However, when evaluating dynamic conditions, the
two-step parameter estimation method can no longer be used.
Model fitting can be performed by using different statistical
and programming software, such as the SSP, Excel, MATLAB, and
R. In addition, model fitting tools have been developed to fit
different growth models to microbial data, such as the DMFit
(available online at www.combase.cc) and the biogrowth (available
online at https://foodlab-upct.shinyapps.io/biogrowth4/)
[34]. Different statistical goodness-of-fit indexes can be applied
to assess the goodness-of-fit indexes of predictive models, mainly
the simple Root Mean Square Error (RMSE) (Eq. 1). The lower the
RMSE, the better the fitting of the model to the data.
n 2
Yi -Yi
i=1
RMSE = ð1Þ
n
where Yi corresponds to the observed value; Y i is the predicted
value; and n is the number of data points or observations.
Finally, when comparing two models, the F-ratio and the
so-called corrected Akaike’s information criterion (AIC) [35] are
two indexes frequently used [3]. For more details on how to
perform calculations and the interpretation of these indexes, see
refs. [3, 26].
2.3 Model Validation Predictive models might be validated in foods prior to use, to
evaluate the reliability of their predictions with real data [3]. To
validate a predictive model, additional experiments with a MAP
food artificially contaminated with the microorganism of interest
are usually designed [18]. The experimental conditions set for these
challenge tests (temperature, aw, pH, CO2 concentration, etc.) may
be within the domain used for model development. Although
validation should be performed with data derived from experiments
with a given food, microbial data from the literature are usually
used for validation to avoid the costs of additional challenge tests.
Data on microbial responses in foods for model validation can be
also found in databases, mainly the ComBase database (www.
combase.cc) [36].
Modelers classify the models as fail-safe or fail-dangerous if the
prediction overestimates or underestimates the observed growth,
respectively. From the public health point of view, a fail-safe model
is preferred compared to a fail-dangerous one, as the former yields
more conservative predictions. Graphical representations of model
predictions versus model observations are useful in evaluating
whether a model is fail-safe or fail-dangerous. To determine in
which degree model predictions coincide with the observed data
derived from validation studies, traditional goodness-of-fit indexes
can be calculated, such as the abovementioned RMSE [3]. More-

over, specific validation indexes—namely, the accuracy (Af) and bias
(Bf) factors—can be determined to evaluate the capacity of the
models to predict microbial behavior [37]. The Af indicates how
well the growth model predictions coincide with the observed data
and a value equal to 1 indicates a perfect coincidence. On the other
hand, the Bf = 1 indicates that observations are equally distributed
below and above model predictions, while Bf < 1 and Bf > 1
evidence a fails-dangerous and a fail-safe model, respectively [37].
Once validated, models are implemented in software tools to be
available for prediction purposes.
3 Application of Predictive Models to Evaluate the Shelf-Life of Foods Using the

Software Micro Hibro
The objective of this section is to demonstrate how validated pre-

dictive microbiology models available in the literature can be
applied to assess the shelf-life of a MAP product with respect to
the presence of a foodborne pathogen. The freely available software
MicroHibro (www.microhibro.com), which is a software for pre-
dictive microbiology and risk assessment in foods developed by our
group [21], was used to implement and simulate the predictive
models selected for this case study.
3.1 Case Study A food manufacturer is interested in evaluating if the MAP config-
urations of a processed cooked ham, together with storage temper-
ature and physicochemical characteristics, assure that the product is
safe during its shelf-life with regards to the presence of
L. monocytogenes. Otherwise, another gaseous mixture may be used.
According to the European Regulation 2073/2005, for those
ready-to-eat (RTE) foods supporting the growth of
L. monocytogenes, food manufacturers might demonstrate that the
levels will not exceed 100 cfu g-1 during their shelf-life
[38]. According to the mentioned regulation, durability studies,
challenge tests, or predictive microbiology models can be applied
by manufacturers to demonstrate to the competent authority that
their RTE foods comply with the criteria established with regard to
the presence of the pathogen [38].
3.2 Model Selection After an extensive bibliographic search with the aid of predictive
and Implementation microbiology experts, a predictive model was identified as a candi-
date to be applied for the evaluation of L. monocytogenes behavior
during the shelf-life of MAP cooked ham using the following
criteria: (i) the characteristics of the MAP cooked ham under eval-
uation (pH, aw, Na-lactate (NaL) concentration, and storage tem-
perature) are within the domain used for model development;
(ii) additional experiments for model validation in MAP cooked
ham were performed and a good agreement between observations

and model predictions were noted.
The selected secondary model was constructed with microbial
data obtained in modified Brain Heart Infusion agar and validated
with microbial data obtained in MAP-cooked meat products
[9]. The model describes the dependence of the specific maximum
growth rate (μmax, h-1) of L. monocytogenes in MAP cooked ham to
temperature, aw, NaL, and CO2 dissolved in the water phase, and it
was derived from an extended Ratkowsky model (Eq. 2). For model
application, Eq. 2 was implemented in MicroHibro combined with
the Gompertz primary model [32].
p
μmax = 0:000713ðT þ 3:5419Þ
x ðaw - 0:9295Þð3140 - CO2 Þð5:9547 - NaLÞ ð2Þ
where T is the storage temperature (°C); aw is the water activity of
the product; NaL is the concentration of NaL (wt%) in the product
and CO2 is the concentration of CO2 dissolved in the water phase
of the product (mg L-1).
Once implemented, the combined predictive model is available
in MicroHibro to be applied by users in a friendly interface. The
software has a database gathering model parameter values. Like-
wise, if readers are interested in performing simulations using pre-
dictive models that are not currently available in the software
database, feel free to contact the authors.
3.3 Conditions for According to the manufacturer’s information, the mean character-
Model Predictions istics of the product under evaluation are aw = 0.96, NaL = 2.3%,
pH = 6.18, storage temperature = 4 °C, MAP gaseous mix-
ture = 20% CO2/80% N2, and constant G/P volume ratio of
4/1. The concentration of dissolved CO2 (mg L-1) in the water
phase can be deduced from the empirical equation developed by
Devlieghere et al. [10], based on the initial CO2 (%) in the head-
space, the storage temperature, and the G/P volume ratio. Under
the mentioned conditions, the concentration of CO2 dissolved in
the water phase is 590 mg L-1. The product is packaged in high-
barrier polyethylene packaging, and its shelf-life is 28 days.
Assuming an initial level of contamination of L. monocytogenes
in the product (y0) equal to 1 cfu g-1 (0 log cfu g-1), and that the
pathogen does not require an adaptation time before it starts to
grow in the product, i.e., lag time = 0 h, the safe shelf-life of the
product can be estimated by simulations using the implemented
predictive model. The safe shelf-life is defined as the time required
by L. monocytogenes to achieve the limit established (Nt) by the
current European Regulation for RTE foods, i.e., 100 cfu g-1
(2 log cfu g-1). If this level is reached within 28 days, which
according to the manufacturer is the shelf-life of the product,
another MAP configuration may be selected as a control measure
Fig. 3 Prediction module of the MicroHibro software: definition of the conditions for model simulations
to reduce the exposure of cooked ham consumers to

L. monocytogenes.
After selection of the implemented model in the prediction
module of MicroHibro, the conditions for model simulations may
be defined (Fig. 3), i.e., y0, temperature, product aw, concentration
of CO2 dissolved in water phase, NaL concentration, and Nt.
3.4 Results of The kinetic parameters and the time required by a microorganism
Simulations Using to reach a given concentration in a food product are estimated by
Predictive Models using the predictive models implemented in MicroHibro. Under
the evaluated conditions, the estimated μmax of L. monocytogenes is
0.009 h-1. The time required for L. monocytogenes to reach
2 log cfu g-1 is 23 days. The kinetic growth curve of the evaluated
pathogen under the evaluated conditions can be seen in Fig. 4.
Based on these results, it can be concluded that the current MAP
configuration, together with storage temperature and product
characteristics, does not assure that the product is safe during its
shelf-life under the evaluated conditions.
According to the manufacturer, two new different gaseous
mixtures could be evaluated to increase the safe shelf-life of
MAP-cooked ham with respect to the presence of
L. monocytogenes: 30% CO2/70% N2 and 40% CO2/60% N2.
Under these two MAP conditions, the L. monocytogenes growth
kinetic parameters estimated using the predictive models imple-
mented in MicroHibro are shown in Table 1. Estimated growth
curves are shown in Fig. 4. Based on these results, it can be con-
cluded that the MAP gaseous mixture that would assure a safe shelf-
life of cooked ham would be 40% CO2/60% N2, since under the
Fig. 4 Listeria monocytogenes growth curves in cooked ham under the evaluated conditions: T = 4 °C,
aw = 0.979 and NaL = 1.5 wt%, and MAP configurations: 20% CO2/80% N2 (red curve), 30% CO2/70% N2
(blue curve), 40% CO2/60% N2 (yellow curve) (simulations using the predictive models developed by
Devlieghere et al. [9])
Table 1
Results of simulations using the predictive models developed by Devlieghere et al. [9]
MAP CO2 dissolved (mg L-1) μmax (h-1) Time to reach 2-log (d)
20% CO2/80% N2 590 0.009 23
30% CO2/70% N2 849 0.008 26
40% CO2/60% N2 1098 0.007 29
evaluated conditions the time required by L. monocytogenes to reach

the microbiological limit would be 29 days (Table 1), thus comply-
ing with the established shelf-life of the manufacturer. Therefore, it
is demonstrated through this case study that validated predictive
models available in the literature can be used for decision making in
the food industry. Further, the applied model can be used as a
scientific tool to demonstrate product compliance with the EU
legislation regarding L. monocytogenes levels in RTE foods.
4 Notes
1. It is important to highlight that, before setting up experiments

for data generation and subsequent model development, it is
convenient to perform a bibliographic review of the scientific
literature and research on predictive microbiology software to
check for the availability of published models that, once vali-
dated, could be useful for a given application. This would avoid
the performance of unnecessary, time-consuming challenge
tests.
2. In most of the predictive models developed so far, it is consid-

ered that only the gases present in the packaging headspace are
relevant for food safety, as if microbial contamination were
present only on the surface of the product. Microorganisms
are generally located in the aqueous phase of foods, or spatially
distributed [3], which implies that the concentration of gases
that diffuse through the product would be relevant for food
safety.
3. Although there is a great interest in using predictive microbi-
ology models as tools to support the selection of appropriate
packaging materials based on their barrier properties, the trans-
fer of gases through packaging materials has been quantified in
just a few predictive modeling approaches.
4. In modular modeling approaches, factors affecting microbial
behavior in foods are considered independent from each other.
However, pH changes in the products are induced by the
dissolution or desorption of CO2 present in MAP, which high-
lights that, to obtain more reliable predictive models, the
impacts of CO2 dynamics on the pH must also be
considered [4].
5. Some controversies have arisen between the comparison of
fitting procedures, as the two-step method minimizes the
error of the predicted growth/survival parameters, while the
one-step procedure minimizes the error of the predicted
growth. However, in the last few years, the model fitting has
been carried out using a single-step fitting because it normally
provides more accurate model estimations when validated in
food matrices.
5 Conclusions
In this chapter, an overview of methods used for data generation,

model fitting, and validation was presented in the context of MAP
products. Validated models implemented in user-friendly software
such as MicroHibro can be useful for a series of applications,
including product development, microbial risk assessment, and
food safety management.
Models considering the dynamics of gases present in MAP
associated with their solubilization, diffusion through the food-
stuffs, and transfer through packaging materials are lacking mainly
due to limitations in methods for gas quantification in foods.
Despite that, some currently available models provide valuable
information regarding the impacts of dissolved CO2 in the water
phase against spoilage microbiota or contaminating pathogens in
foods.
Finally, the case study presented in this chapter demonstrates
that validated predictive models can be applied for decision making
and to evaluate whether the MAP configuration of a food product,

its storage temperature, and physicochemical characteristics would
allow the growth of a foodborne pathogen to undesirable levels
according to current regulations, limiting its shelf-life. Likewise,
predictive models could be applied to evaluate the growth potential
of spoilage microorganisms in foods and its consequences on their
shelf-life.
Acknowledgements
The authors are greatly acknowledged by the Council of Economy,

Knowledge, Business and University of Junta de Andalucı́a (Project
AT 2017-5686) and FEDER European funds. The authors are
grateful to the EU PRIMA program and the International Joint
Programming (Project Reference PCI2019-103453) R&D Pro-
jects 2019 from the Spanish Ministry of Science and Innovation
(Plan Estatal de Investigación Cientı́fica y Técnica y de Innovación
2017–2020: State R&D Program Oriented to the Challenges of
the Society) for funding the ArtiSaneFood project (PRIMA/
0001/2018). The authors wish to thank the Spanish Government
(Ministry of Science and Innovation, research project PID2019-
108420RB-C31-ASEQURA) for its financial support.
References
1. Guillard V, Couvert O, Stahl V et al (2016) assessment: a description and comparison of

Validation of a predictive model coupling gas tools presented at the ICPMF8 Software Fair.
transfer and microbial growth in fresh food Food Microbiol 45:290–299
packed under modified atmosphere. Food 8. Possas A, Valero A, Pérez-Rodrı́guez F (2022)
Microbiol 58:43–55 New software solutions for microbiological
2. Hutchings N, Smyth B, Cunningham E et al food safety assessment and management. Curr
(2021) Development of a mathematical model Opin Food Sci 44:100814
to predict the growth of Pseudomonas spp. in, 9. Devlieghere F, Geeraerd AH, Versyck KJ et al
and film permeability requirements of, high (2001) Growth of Listeria monocytogenes in
oxygen modified atmosphere packaging for modified atmosphere packed cooked meat pro-
red meat. J Food Eng 289:110251 ducts: a predictive model. Food Microbiol 18:
3. Perez-Rodriguez F, Valero A (2013) Predictive 53–66
microbiology in foods. In: Predictive microbi- 10. Devlieghere F, Debevere J, Van Impe J (1998)
ology in foods. Springer, New York, p 128 Concentration of carbon dioxide in the water-
4. Chaix E, Couvert O, Guillaume C et al (2015) phase as a parameter to model the effect of a
Predictive microbiology coupled with gas modified atmosphere on microorganisms. Int J
(O2/CO2) transfer in food/packaging sys- Food Microbiol 43:105–113
tems: how to develop an efficient decision sup- 11. Dolan KD, Meredith H, Bolton DJ et al
port tool for food packaging dimensioning. (2019) Coupling the dynamics of diffused
Compr Rev Food Sci Food Saf 14:1–21 gases and microbial growth in modified atmo-
5. Devlieghere F, Van Belle B, Debevere J (1999) sphere packaging. Int J Food Microbiol 292:
Shelf life of modified atmosphere packed 31–38
cooked meat products: a predictive model. Int 12. Meredith H, Valdramidis V, Rotabakk BT et al
J Food Microbiol 46:57–70 (2014) Effect of different modified atmo-
6. Buchanan RL (1993) Predictive food microbi- spheric packaging (MAP) gaseous combina-
ology. Trends Food Sci Technol 4:6–11 tions on Campylobacter and the shelf life of
7. Tenenhaus-Aziza F, Ellouze M (2015) Soft- chilled poultry fillets. Food Microbiol 44:
ware for predictive microbiology and risk 196–203
13. Geeraerd AH, Valdramidis VP, Devlieghere F 25. Baranyi J, Roberts TA (1994) A dynamic
et al (2004) Development of a novel approach approach to predicting bacterial growth in
for secondary modeling in predictive microbi- food. Int J Food Microbiol 23:277–294
ology: incorporation of microbiological knowl- 26. Zwietering MH, Jongenburger I, Rombouts
edge in black box polynomial modeling. Int J FM et al (1990) Modeling of the bacterial
Food Microbiol 91:229–244 growth curve. Appl Environ Microbiol 56:
14. Mejlholm O, Dalgaard P (2009) Development 1875–1881
and validation of an extensive growth and 27. Ratkowsky DA (1983) Nonlinear regression
growth boundary model for Listeria monocyto- modeling: a unified practical approach. In:
genes in lightly preserved and ready-to-eat Marcel Decker I (ed) Nonlinear regression
shrimp. J Food Prot 72:2132–2143 modeling: a unified practical approach. Marcel
15. Simpson R, Almonacid S, Acevedo C (2001) Dekker, New York
Development of a mathematical model for 28. Ratkowsky DA, Olley J, McMeekin TA et al
MAP systems applied to nonrespiring foods. J (1982) Relationship between temperature and
Food Sci 66:561–567 growth rate of bacterial cultures. J Bacteriol
16. Chaix E, Broyart B, Couvert O et al (2015) 149:1–5
Mechanistic model coupling gas exchange 29. Rosso L, Lobry JR, Bajard S et al (1995) Con-
dynamics and Listeria monocytogenes growth venient model to describe the combined effects
in modified atmosphere packaging of non of temperature and pH on microbial growth.
respiring food. Food Microbiol 51:192–205 Appl Environ Microbiol 61:610–616
17. Alfaro B, Hernández I, Le Marc Y et al (2013) 30. Gibson AM, Bratchell N, Roberts TA (1988)
Modelling the effect of the temperature and Predicting microbial growth: growth responses
carbon dioxide on the growth of spoilage bac- of salmonellae in a laboratory medium as
teria in packed fish products. Food Control 29: affected by pH, sodium chloride and storage
429–437 temperature. Int J Food Microbiol 6:155–178
18. Geysen S, Verlinden BE, Geeraerd AH et al 31. Hajmeer MN, Basheer IA, Najjar YM (1997)
(2005) Predictive modelling and validation of Computational neural networks for predictive
Listeria innocua growth at superatmospheric microbiology II. Application to microbial
oxygen and carbon dioxide concentrations. growth. Int J Food Microbiol 34:51–66
Int J Food Microbiol 105:333–345 32. Zwietering MH, Wijtzes T, De Wit JC et al
19. Pin C, Baranyi J, De Fernando GG (2000) (1992) A decision support system for predic-
Predictive model for the growth of Yersinia tion of the microbial spoilage in foods. J Food
enterocolitica under modified atmospheres. J Prot 55:973–979
Appl Microbiol 88:521–530 33. Possas A, Valero A, Garcı́a-Gimeno RM et al
20. Geysen S, Escalona VH, Verlinden BE et al (2021) Combining UV-C technology and caf-
(2006) Validation of predictive growth models feine application to inactivate Escherichia coli
describing superatmospheric oxygen effects on on chicken breast fillets. Food Control 129:
Pseudomonas fluorescens and Listeria innocua 108206
on fresh-cut lettuce. Int J Food Microbiol 111: 34. Garre A, Koomen J, den Besten HMW, Zwie-
48–58 tering MH (2023) Modeling population
21. González SC, Possas A, Carrasco E et al (2019) growth in R with the biogrowth Package. J
“MicroHibro”: a software tool for predictive Stat Soft 107:1–51
microbiology and microbial risk assessment in 35. Akaike H (1974) A new look at the statistical
foods. Int J Food Microbiol 290:226–236 model identification. IEEE Trans Autom Con-
22. ISO/FDIS (2018) ISO/FDIS 20976-1: trol 19:716–723
microbiology of the food chain—require- 36. Baranyi J, Tamplin ML (2004) ComBase: a
mentes and guidelines for conducting chal- common database on microbial responses to
lenge tests of food and feed products—Part 1: food environments. J Food Prot 67:1967–
Challenge tests to study growth potential, lag 1971
time and maximum growth rate. ISO, p 36
37. Ross T (1996) Indices for performance evalua-
23. McMillin KW (2008) Where is MAP going? A tion of predictive models in food microbiology.
review and future potential of modified atmo- J Appl Bacteriol 81:501–508
sphere packaging for meat. Meat Sci 80:43–65
38. European Commission (2005) COMMIS-
24. Rotabakk BT, Lekang OI, Sivertsvik M (2007) SION REGULATION (EC) No 2073/2005
Volumetric method to determine carbon diox- of 15 November 2005 on microbiological cri-
ide solubility and absorption rate in foods pack- teria for foodstuffs
aged in flexible or semi rigid package. J Food
Eng 82:43–50
Chapter 15
Antifungal Activity of Edible Films and Coatings

for Packaging of Fresh Horticultural Produce
Lluı́s Palou and Marı́a B. Pérez-Gago
Abstract
Protocols for in vitro and in vivo evaluation of the antifungal activity of edible films and coatings (ECs) used
for postharvest treatment of fresh fruits and vegetables are described in this chapter. Antifungal ECs are
typically prepared by incorporating particular antimicrobial ingredients into EC matrix formulations.
Different methods and numerous variations can be adopted for both in vitro and in vivo evaluation, mostly
depending on the specific purpose of the assay, the components and properties of the EC matrix and the
antifungal agent(s), the nature of the target fungal pathogen, and the characteristics and usual postharvest
handling of each horticultural product. In any case, however, the inoculum of the target fungi will be used
in the experiments, and its preparation is also detailed in this chapter. In general, while EC solid dry films are
used for in vitro tests, EC liquid emulsions are used for in vivo assays. We describe three of the most
common and, in our opinion, useful antimicrobial in vitro tests specifically intended for use with fungal
strains, i.e., agar diffusion or disk diameter tests, film surface inoculation tests, and plate counting
germination tests. Coating of fresh produce artificially inoculated with the pathogen is commonly used in
laboratory-scale in vivo experiments to assess the ability of ECs to control disease. Further larger-scale
semicommercial or commercial trials conducted in pilot plants or packinghouse facilities with naturally
infected, cold-stored produce can also be considered.
Key words Edible films and coatings, Fresh fruits and vegetables, Postharvest fungal decay, Fungal
inoculum preparation, In vitro antifungal activity, In vivo disease control
1 Introduction
Harvested fresh horticultural products are highly susceptible to

dehydration, physiological changes, mechanical injuries, and path-
ological decay that affect quality attributes, reduce produce stor-
ability, and cause major product losses throughout the supply
chain. Typically, fungi are the most prevalent causal agents of
postharvest diseases, particularly in the case of fresh fruits. Some
of them have a wide range of hosts and attack many different fresh
products, while others are much more specific and only attack
particular commodities. Thus, for example, pathogenic species in
https://doi.org/10.1007/978-1-0716-3613-8_15,
259
260 Lluı́s Palou and Marı́a B. Pérez-Gago
the genera Botrytis, causing gray mold, Alternaria, causing black

spot or black rot, Colletotrichum, causing anthracnose, Fusarium,
causing Fusarium rot, Lasiodiplodia, causing stem-end rot, Penicil-
lium, causing blue or green molds, or Rhizopus and Mucor, causing
soft rot, are important postharvest pathogens of a large variety of
fruits and vegetables, including citrus, pome fruits, stone fruits,
tropical fruits, berries, grapes, persimmons, pomegranates, toma-
toes, peppers, cucumbers, squash, eggplants, and melons, among
others. In contrast, fungal species that only attack particular fruit
families include Penicillium digitatum (Pers.:Fr.) Sacc., Penicil-
lium italicum Wehmer, and Geotrichum citri-aurantii (Ferraris)
Butler, which cause green and blue molds and sour rot, respectively,
of citrus fruit; and Monilinia ssp., which are especially virulent on
stone fruits, causing postharvest brown rot [1–3].
Traditionally, postharvest disease control of fresh fruits and
vegetables involves the use of synthetic chemical fungicides, alone
or incorporated into commercial waxes. However, human health
risks, environmental issues associated with fungicide residues, and
the proliferation of resistant strains of the pathogens have raised
important concerns worldwide, increasing the need to look for safer
alternatives [4].
In the last years, a considerable amount of research work has
focused on the development of edible coatings (ECs) with antifun-
gal activity for fruits and vegetables as a sustainable alternative to
conventional fungicides to control postharvest diseases and main-
tain the quality of fresh horticultural produce. Edible films and
coatings are thin layers of material composed mainly of natural
biopolymers (i.e., proteins or polysaccharides), lipids, or a mixture
of them. ECs for fresh horticultural produce provide important
functions, including a semipermeable barrier to water vapor, oxy-
gen, and carbon dioxide between the coated fruit and the sur-
rounding atmosphere, enhancement of fruit appearance, and
carriers of active ingredients such as antimicrobials, antioxidants,
nutraceuticals, etc. Therefore, ECs can preserve the postharvest
quality of fruits and vegetables by reducing weight loss, respiration
rate, senescence, and also fungal decay if they exert appropriate
antifungal activity [5].
In the particular case of ECs with antifungal activity, two dif-
ferent general types can be considered: (i) film-forming capacity
biopolymers with inherent antifungal activity, such as chitosan coat-
ings and Aloe vera gels, which have received a lot of attention for
postharvest treatment of fruits and vegetables [6–8], and (ii) ECs
designed with the incorporation as additional ingredients to the
coating formulation matrix of food-grade antifungal compounds
able to reduce the growth of spoilage microorganisms and control
postharvest diseases. Thus, antifungal agents of potential use in
edible films and coatings comprise a wide variety of compounds
from natural or synthetic sources, such as mineral salts, organic
Antifungal Activity of Edible Films and Coatings 261
acids and their salts, parabens, enzymes, bacteriocins, polypeptides,

natural extracts, essential oils, and metal-based nanoparticles or
nanocomposites. In addition, antifungal ingredients can also be
antagonistic microorganisms that perform as biocontrol agents
[6, 9].
In general, the development of antifungal ECs requires an
initial optimization of coating formulations based on the chemical
compatibility of the ingredients to achieve stable emulsions capable
of forming films with good structural properties when applied to
the fruit surface [10]. Although films and coatings are the same in
nature and sometimes are used as synonymous, they refer to differ-
ent concepts according to their different purposes and utilization.
Films are defined as stand-alone, solid, thin layers of materials and
are usually prepared from coating emulsions by casting, molding, or
extrusion procedures. Among these, casting is the most commonly
used method for film formation at laboratory and pilot scales,
whereas extrusion is one of the major polymer processing techni-
ques currently in use at commercial scales [11]. Edible films can be
used as covers, wraps, or separation layers in foods, although they
are primarily used as testing structures for the determination of
barrier, mechanical, solubility, structural, and other properties of
interest, such as antimicrobial or antioxidant activity provided by
certain film-forming materials or ingredients. On the other hand,
coatings are applied to fruits and vegetables by dipping or spraying
and involve the formation of films directly on the surface of the
commodity they are intended to protect or enhance, forming part
of the final fresh product [12, 13]. Therefore, from the research
point of view, antifungal EC formulations can be used to form
stand-alone films intended for in vitro studies or can be directly
applied to the commodity for in vivo studies.
Protocols for both in vitro and in vivo evaluation of the anti-
fungal activity of ECs for fresh fruits and vegetables are described in
this chapter. Moreover, methods for fungal inoculum preparation
are also detailed, as this is an essential step to conduct both types of
experiments. In both cases, different methods and numerous varia-
tions can be adopted depending on the specific purpose of the assay,
the components and properties of the EC matrix and the antifungal
agent, the nature of the target fungus, and the characteristics and
usual postharvest handling of the horticultural product [14]. In
vitro studies with films are generally a good approach for the
preliminary evaluation and screening of specific EC formulations
against particular target microorganisms. Although considerable
variations of in vitro assays can be found in the literature, many
refer to general antimicrobial activity, and in practice often explain
methods suitable to test bacterial strains [15, 16]. After describing
the procedure for film formation, we describe here three of the
most common and, in our opinion, useful procedures specifically
intended for use with fungal strains, i.e., agar diffusion or disk
diameter tests, film surface inoculation tests, and plate counting

germination tests.
In any case, the performance of films on agar medium will often
fail to appropriately predict the performance of ECs once applied to
fruits or vegetables because of obvious differences between in vitro
and in vivo conditions. Therefore, the disease control ability of EC
formulations cannot be anticipated by the antifungal activity
in vitro and, in all cases, ECs effective in vitro against the target
pathogen must be tested in vivo with infected fruit, taking into
account all the factors involved in disease development and trying
to simulate as much as possible the actual disease conditions in the
packinghouse.
Although more laborious, time consuming, and expensive than
in vitro tests, in vivo tests are required to determine the effective-
ness of antifungal ECs for each specific pathosystem. Since natural
infection rates on harvested produce can be low or highly variable,
it is common in laboratory assays to artificially inoculate the fruit
host with the pathogen to get high and uniform levels of infection.
Posterior semicommercial or commercial evaluation of selected EC
treatments is usually conducted in pilot plants or packinghouse
facilities using a large sample size of naturally infected fruit. In
general, two different kinds of antifungal activity can be assessed
on inoculated commodities: (i) curative activity, intended to assess
the ability to control fungal infections already established in the
fruit, and for which the antifungal EC is applied after the inocula-
tion of the fruit, and (ii) preventive activity, intended to evaluate the
capacity of the EC to protect the fruit from posterior infections. In
this case, the fruit is coated and artificially inoculated with the
pathogen afterwards [17]. Inoculated and coated fruit can be incu-
bated at 20–25 °C to favor fast fungal development or long-term
stored at low temperatures to resemble commercial fruit handling.
Disease incidence and severity and pathogen sporulation are peri-
odically determined on each fruit during storage and disease con-
trol ability of ECs is usually given as a percentage of disease
reduction with respect to the control treatment. General protocols
for these in vivo assays are also described in this chapter.
2 Materials
2.1 Preparation of • Laboratory laminar flow hood. Bunsen burner. Inoculating

Fungal Inoculum loops.
• Vortex mixer. Sterile test tubes, Erlenmeyer flasks, glass funnel,
gloves, cheesecloth, Pasteur pipettes, and micropipettes.
• Potato dextrose agar (PDA) (see Note 1): typically prepared
from the commercial medium: 4.0 g L-1 potato extract or
potato peptone (equivalent to 200 g infusion from pota-

toes) + 20.0 g L-1 glucose + 15.0 g L-1 agar.
• Petri dishes: typically, 90-mm diameter x 15-mm height plastic
plates are used. These are conveniently autoclavable and
disposable.
• Emulsifier aqueous solution: Polysorbate 80 (Tween® 80) or
Triton™ X-100, at 0.05% (w/v).
• Hemacytometer: Thoma chamber or Fuchs–Rosenthal chamber
is usually used.
• Optical microscope. Laboratory incubator.
2.2 In Vitro • Antifungal EC emulsions to be evaluated.

Antifungal Activity • Spore suspensions of the target fungal pathogen (inoculum).
• Laboratory laminar flow hood. Bunsen burner. Inoculating
loops. Sterile Petri dishes, test tubes, tweezers, scalpels, cork
borers, Erlenmeyer flasks, glass rods, gloves, stirring rods, and
micropipettes.
• Casting plates for film formation.
• PDA (see Note 1). Potato dextrose broth (PDB): liquid culture
medium prepared as PDA without the agar.
• Ruler, caliper, or digital caliper.
• Laboratory incubator. Refrigerator.
2.3 In Vivo • Antifungal EC emulsions to be evaluated.

Assessment of Disease • Spore suspensions of the target fungal pathogen (inoculum).
Control
• Target commodity: Samples of fresh fruits or vegetables.
• Sterile inoculating stainless-steel rod (with probe tip), pricker,
scalpel, cork borer, gloves, test tubes, Erlenmeyer flasks, and
micropipettes.
• Surface disinfection solutions, e.g., 0.5% sodium hypochlorite,
70% ethanol.
• Immersion containers, stirring rods, mesh screen, plastic or
metal grid, plastic cavity sockets, corrugated cartons, and plastic
trays.
• Chronometer, ruler, caliper, digital caliper.
• Incubation cabinets or walk-in rooms. Cold storage rooms.
• Pilot plant-scale or commercial-scale fresh produce packingline.
3 Methods
3.1 Preparation of Suspend 39 g of commercial PDA powder in 1 L of distilled water.

Fungal Inoculum If necessary, bring pH to a final value of 5.6 ± 0.2. Autoclave at
121 °C for 15 min. Leave to cool to 40–45 °C and pour into plates
3.1.1 Preparation of PDA
within the laminar flow hood (about 20 mL in each 90-mm-diam-
Petri Dishes
eter dish). Put the plate lids on and allow to dry within the hood.
3.1.2 Culture of Fungal Strains are generally obtained from known culture collections or
Strains isolated from infected produce, purified, identified, and cultured
and maintained (replated) in artificial media. For replating, work
within the laminar flow hood. Take spores or mycelium with a
sterile inoculating loop (heated to red hot in the burner flame and
allowed to cool) from a grown colony on a PDA dish and transfer
them to a fresh PDA plate by gently touching the agar surface.
Depending on the fungal species, inoculation of the fresh plate can
be more convenient in one central point or three equidistant points
on the agar surface. Incubate inoculated plates at 20–25 °C inside a
laboratory incubator, generally in the dark for 7–21 d (see Note 2).
3.1.3 Obtaining Spore Within the laminar hood, take abundant spores from a 7- to 21-d-
Suspensions old fungal culture and transfer them to a sterile test tube containing
a known volume of aqueous emulsifier solution (0.05% Tween®
80 or Triton™ X-100) to obtain a very high concentrated suspen-
sion. Mix roughly in a vortex mixer for 2 min and filter the content
to another sterile test tube through a glass funnel containing two
layers of sterile cheesecloth. This will allow the separation of spores
from mycelial fragments. Mix the suspension again and use a Pas-
teur pipette to transfer drops to a hemacytomer (see Note 3).
Count the spores in an optical microscope (×10 or ×40) to deter-
mine the spore concentration and calculate the suspension volume
needed to add using a micropipette to a known volume of fresh
emulsifier solution to obtain the final suspension at the desired
inoculum concentration (see Note 4). Mix again this final suspen-
sion before use in the experiments. See Note 5 for an alternative
method to prepare large volumes of certain spore suspensions.
When fruit are going to be inoculated in in vivo tests with weak
fungal pathogens, some additional ingredients can be added to the
spore suspension to favor actual infection rates (see Note 6).
3.2 In Vitro The purpose of these tests is to evaluate the antifungal activity of
Antifungal Activity ECs against the target fungal pathogen in a rapid, easy, cheap, and
simple manner that does not involve the use of fresh produce. For
this, the most practical approach is to work with films instead of
coating liquid emulsions. As a solid thin layer produced by drying
the emulsion, the film will appropriately simulate the characteristics
of the emulsion once applied onto the surface of fresh produce (see
Note 7).
3.2.1 Film Casting For film production from liquid coating formulations, pipette and
evenly spread an appropriate volume of the degassed emulsion on
rimmed, smooth plates (e.g., Petri dishes, Teflon plates, high-
density polyethylene (HDPE) casting plates, etc.; see Note 8) rest-
ing on a leveled slab and allow to dry at ambient conditions,
normally at approximately 21–25 °C and 50% RH until drying is
complete (see Note 9). While whole films dried in Petri dishes can
be directly used in some tests (e.g., plate counting germination
tests), in other cases the dry film is peeled intact from the casting
surface using a sterile scalpel and tweezers and aseptically cut into
smaller disks of the desired diameter using a sterile cork borer. Use
the films immediately or aseptically store them at 4 °C in the
refrigerator until use in the experiments.
3.2.2 Agar Diffusion or These tests are intended to determine the ability of coating films to
Film Disk Diameter Tests inhibit the spore germination and the mycelial growth of a particu-
lar fungal pathogen in an artificial agar culture medium.
Working within the laminar hood, place 100 μL of spore sus-
pension of the target pathogen on the center of a PDA, DRBCA, or
other agar medium plate and spread uniformly over the entire agar
surface by gently rubbing with a sterile glass rod. A series with
different concentrations of inoculum can be used, usually from
103 to 106 spores mL-1, depending on the fungus. Aseptically
transfer the film disk (16-mm diameter) to the center of the agar
surface and lid the plate. In some cases, smaller film disks (5-mm
diameter) can be produced and three of them can equidistantly be
plated in the same Petri dish. Depending on the objective of the test
and the nature of the EC, control disks can be simply of sterile filter
paper or disks of film formulated without the antifungal ingredient
(s). For each pathogen, type of coating film, and inoculum level,
three to five replicated plates are generally prepared. Depending on
the type of film and antifungal ingredient, put the plates in a
standard refrigerator at 4 °C for 3 h to allow, if that is the case,
the diffusion of film ingredients from the disk to the agar medium
[18]. Then transfer them to an incubator set at 20 °C, 25 °C, or the
most adequate temperature (Table 1) and incubate for a variable
period of 4–14 d, depending on the fungal species. Periodically
(every 1, 2, or 3 d, depending on the growth rate of the fungus)
measure, in two or four directions, the length of the inhibition
zone around the film disk (from the perimeter of the film disk until
the edge of the inhibited area; Fig. 1). These quantitative zone
measurements are giving in fact qualitative results, and results from
different studies are difficult to compare because of the many
specific conditions of the experiments including film size and prop-
erties, antifungal agent, temperature, incubation time, target fun-
gus, inoculum concentration, etc. [19].
This method is particularly suitable for fungal pathogens of
easy sporulation in vitro that produce large amounts of small-size
Table 1
Appropriate incubation temperature for optimal growth of common fungal
pathogens causing postharvest disease on fresh horticultural produce
Pathogen Disease Temperature (°C)

Penicillium spp. Blue/green molds 25
Botrytis spp. Gray mold 20
Alternaria spp. Black spot, black rot 25
Monilinia spp. Brown rot 25
Colletotrichum spp. Anthracnose 25
Geotrichum spp. Sour rot 28
Aspergillus spp. Black rot 30
Rhizopus spp. Soft rot, Rhizopus rot 25
Mucor spp. Mucor rot 25
Lasiodiplodia spp. Stem-end rot 28
Fusarium spp. Fusarium rot 25
Variations among different species in the same genus may occur. For example, while 25 °
C is the most appropriate temperature for Monilinia fructicola, it is 20 °C for Monilinia
laxa
Fig. 1 Disk diameter tests on DRBC agar plates for evaluation of the in vitro inhibition of Penicillium digitatum
(a) and Penicillium italicum (b) by control HPMC-lipid films (left-hand side images) and HPMC-lipid films
containing potassium sorbate (PS), a mixture of sodium benzoate and potassium sorbate (SB + PS) and
sodium salt of methyl paraben (SMP) (right-hand side images). (Reproduced from Ref. [10] with permission
from ACS Publications)
spores (e.g., Penicillium spp., Aspergillus, spp., etc.), which allow

the uniform distribution of the fungal inoculum on the agar surface
in the test culture plates. A variation employing mycelial plugs can
be considered with target fungi of difficult sporulation or with
spores of large size (e.g., Lasiodiplodia spp., Alternaria, spp., Rhi-
zopus spp., etc.; see Note 10).
3.2.3 Film Surface These assays are intended to assess the ability of the target pathogen
Inoculation Tests to either grow on the coating material itself or penetrate and pass
through it. The purpose is to simulate coating surface contamina-
tion and determine if, once applied to the fruit, the coating will
provide a barrier functionality and will be able to prevent new
infections caused by external contaminating inoculum, usually air-
borne spores or spores and mycelia from rotten adjacent fruit (i.e.,
fungi causing nests of decay on stored produce). Although different
variations of this type of test have been proposed, we describe here
two of the simplest versions.
Working within the laminar hood, place equidistantly film disk
pieces inside an empty sterile Petri dish and drop 10–20 μL of spore
suspension of the target pathogen on the surface of each disc. Four
16-mm-diameter disks can be used in a four-section compartmen-
talized 100-mm-diameter plastic Petri dishes [18], but less disks of
higher diameter can also be used. A series with different concentra-
tions of inoculum can be used, usually from 103 to 106 spores mL-
1
, depending on the fungus. Transfer the plates to an incubator set
at 20 °C, 25 °C, or the most adequate temperature (Table 1) and
visually assess fungal growth periodically during incubation (every
1, 2, or 3 d) for a variable period of 4–14 d, depending on the
observed rate of fungal growth. No control disks are needed for
this test, although in some cases could be of interest to use disks of
film matrix without the antifungal agent(s). Typically, for each
pathogen, film treatment, and inoculum level, three to five repli-
cated plates are prepared. Results are qualitative and usually given
just as positive (+) or negative (-) growth.
A variation of this test consists in using PDA Petri dishes
instead of empty dishes and carefully inoculate the surface of film
disks placed on the PDA surface. It is important that the inoculum
drop does not move from the disk surface to the surrounding agar
medium. In this case, the artificial media resembles the fruit surface
and, if evaluations during incubation give positive fungal growth on
the agar medium, it will mean that the coating disc is not acting as
an effective barrier for contaminating fungal inoculum since the
spores have been able to pass through the film disk and germinate
and growth on the media. More sophisticated versions of this test
involving plate counting of fungal populations have been
described [14].
3.2.4 Plate Counting As an adaptation from tests designed to work with bacterial cells
Germination Test [20, 21], the purpose of these tests is to indirectly assess the effect
of the antifungal film on the spore germination of the target fungal
pathogen, avoiding the large amount of time and labor typically
needed for observation of germination in the microscope. Plate
counting methods are time, space, and labor consuming, but, in
contrast to the aforementioned in vitro tests, they give quantitative
results that can be used to measure log reductions due to the
antifungal film [16, 19].
Prepare a spore suspension of the target pathogen in a PDB
solution at a final concentration of 102–104 spores mL-1, depend-
ing on the fungus, by adding aseptically with a micropipette the
correspondent volume of aqueous spore suspension of known con-
centration into sterilized and cooled PDB flasks. Working within
the laminar hood, pour 15 mL of PDB spore suspension into a
90-mm-diameter Petri dish containing the dry film to be tested, put
the lid on, and hold the inoculated plates on an orbit shaker at
50 rpm at room temperature. At various time intervals, from 4 to
24 h, depending on the experiment, take 100 μL of PDB spore
suspension on the film surface and plate them homogeneously in
fresh PDA dishes by gently rubbing over the entire agar surface
with a sterile glass rod. Control disks are usually sterile paper disks
or disks of coating matrix formulated without the antifungal agent.
When the objective of the test is to find out the influence of the
concentration of the antifungal ingredient on spore mortality, a
series of increasing antifungal agent concentrations in the film are
prepared. Incubate the plates at 25 °C for 3–5 d and count the
number of fungal colonies growing on each plate. For each patho-
gen, film treatment, and time interval, two replicated plates are
usually prepared, and, if needed, serial dilutions with duplicate
plating can be performed. Depending on the number of grown
colonies, results are given directly as spores mL-1 or as log
spores mL-1 (see Note 11). Results can also be given as a percent-
age of inhibition of germination with respect to control disks (see
Note 12).
3.3 In Vivo Due to the large variety of postharvest fungal diseases that affect
Assessment of Disease fresh horticultural produce, many variations of the general proce-
Control dures described here can be found in the literature. In general, the
objective of these in vivo trials is to evaluate the disease control
ability of antifungal EC formulations after their application to fruits
and vegetables actually infected by the target pathogen. Common
laboratory assays with fruit artificially inoculated with the pathogen
are described. Figure 2 represents a schematic diagram for this type
of experiment, particularly for the evaluation of ECs containing
“generally recognized as safe” (GRAS) salts as antifungal ingredi-
ents for the control of Alternaria black spot on cherry tomato
[22]. Notes in this section will refer to procedural variations,
Fig. 2 Methodological procedure for formulation and in vivo evaluation of the ability of edible coatings
containing GRAS salts to control black spots of tomato caused by the fungus Alternaria alternata. (Reproduced
from Ref. [22]; Open Access)
particularly those to be considered in larger-scale semicommercial

or commercial trials conducted in pilot plants or packinghouse
facilities.
3.3.1 Fresh Produce Select by hand and use in the experiments healthy fresh fruits or
Sample Preparation vegetables of uniform size and good condition. If possible, use
produce from local organic or commercial orchards located in the
area surrounding the laboratory or research facility. If the fruit are
from a packinghouse or store, always acquire them before any
postharvest treatments are applied. Use the fruit the same day or
one day after harvest or, if not possible, store them at the most
adequate commercial cold storage temperature for each commodity
(Table 2) and high relative humidity (RH > 90%) only for several
days. If cold-stored, allow the fruit to warm and dry at room
temperature for several hours before use in the experiments. Before
each experiment, randomize and wash the fruit with fresh water, a
biodegradable detergent solution or, if needed, dip them for
1–2 min in a surface disinfection solution at room temperature,
usually a diluted bleach solution (sodium hypochlorite at 0.5 vol%).
In the case of small sample sizes, spraying with 70% ethanol
Table 2
Recommended conditions for long-term storage of major fresh fruits and vegetables
Commodity Storage temperature (°C) Approximate storage life

Avocado 3–7 2–4 weeks
Banana 13–15 1–4 weeks
Berries
Blackberry, blueberry, raspberry -0.5–0 3–10 days
Strawberry 0 7–10 days
Cherry, sweet -1–0 2–3 weeks
Citrus
Orange 3–5 3–12 weeks
Mandarin 4–7 2–4 weeks
Lemon 10–13 1–6 months
Lime 9–10 6–8 weeks
Grapefruit 10–15 6–8 weeks
Eggplant 10–12 1–2 weeks
Grape -0.5–0 1–6 months
Guava 5–10 2–3 weeks
Kiwifruit 0 3–5 months
Mango 13 2–3 weeks
Melons
Cantaloupes and other melons 2–5 2–3 weeks
Honeydew and Orange-flesh 5–10 3–4 weeks
Papaya 7–13 1–3 weeks
Persimmon 0 1–3 months
Pineapple 7–13 2–4 weeks
Pome fruit
Apple -1.1–1 3–6 months
Pear (European) -1.5–(-0.5) 2–7 months
Pomegranate 5–7 2–3 months
Squash
Summer squash 7–10 1–2 weeks
Winter squash 12–15 2–3 months
Tomato 8–13 1–5 weeks
Stone fruit (apricot, nectarine, peach, plum) -0.5–0 1–4 weeks
Watermelon 10–15 2–3 weeks
Adapted from: http://postharvest.ucdavis.edu/Commodity_Resources/Storage_Recommendations/
solution can also be considered. Rinse the surface-disinfected fruit

with abundant tap water and allow to air-dry at room temperature.
3.3.2 Experimental In order to determine curative activity, wound the surface of

Design, Fungal Inoculation, selected and randomized fruit with a sterile stainless steel pricker,
and Coating Application scalpel, or rod and place with a micropipette a known volume
(10–30 μL) of spore suspension of the target pathogen in the
wound (see Note 13). When the inoculum drop is dried or after a
certain period of time, typically 24 h at room temperature to
resemble incipient field infections, coat the fruit by immersion
(10–30 s) with the corresponding EC emulsion, and leave the
coated fruit to drain on a mesh screen or an adequate plastic or
metal grid and air-dry at room temperature (see Note 14). To test
preventive activity, coat the fruit with the corresponding EC treat-
ment, allow to drain and air-dry, and, once dried or after a particular
period of time (e.g., 24 h), wound inoculate them with the patho-
gens as previously described [17]. In any case, inoculated but
uncoated fruit (immersed in water for the same 10- to 30-s period)
are used as a negative control. Depending on the purpose of the
experiment, in some cases, negative control fruit can be coated with
EC formulated without the antifungal ingredient(s). In other cases,
a positive control treatment can be added, normally fruit treated
with a commercial postharvest chemical fungicide of known and
high efficacy on that pathosystem.
Use a completely randomized design in which each treatment is
applied to a variable number of replicates and fruit per replicate,
depending on fruit size. For instance, common sample sizes for
laboratory experiments are three to five replicates of 10–25 fruits
per treatment (see Note 15).
3.3.3 Storage Conditions After draining and air-drying at room temperature, place coated
and Assessment of Disease fruit on plastic cavity sockets on corrugated cartons or plastic trays
Control and incubate them in a climatic walk-in storage room for 7–21 days
at 20 °C or the most convenient growth temperature for each
target pathogen (Table 1). This procedure will allow to obtain
quick results on the disease control ability of each EC treatment.
Assess periodically (every 1–7 days) disease incidence by counting
the number of infected wounds or fruit in each replicate (Fig. 3,
bottom left) and, for each evaluation date, express mean values as
percentage; disease severity by measuring the lesion diameter of
each infected wound (Fig. 3, bottom right) and express mean
values in mm; and pathogen sporulation by counting the number
of sporulated wounds or fruit in each replicate (Fig. 3, bottom left)
and express mean values as percentage (see Note 16). For each
evaluation date, all these results can also be expressed as a percent-
age of reduction with respect to the control treatment (see Note
17). Disease severity over time can also be directly expressed as the
area under the disease progress curve (AUDPC) or, more accu-
rately, the area under the disease progress stairs (AUDPS [23]).
Fig. 3 Concurrent rind wound and inoculation of a spore suspension of Penicillium sp. on orange fruit (top).
Determination of disease incidence (number of infected fruit) and pathogen sporulation (number of sporulated
fruit) (bottom left), and disease severity (lesion diameter; bottom right) on citrus fruit wound inoculated with
Penicillium sp. and incubated at 20 °C for 7 days
If among the objectives of the experiment there is the assess-

ment of the ability of antifungal ECs to control disease on fresh
produce stored at low temperatures for prolonged periods, after
draining and air-drying at room temperature, place coated samples
on plastic cavity sockets on plastic trays and store them in a cold
storage room at the recommended commercial conditions for each
commodity (Table 2). Assess periodically (typically every
1–2 weeks) disease incidence and severity and pathogen sporulation
in each replicate as described above. For each evaluation date, give
the results as described above.
4 Notes
1. PDA is a universal medium for in vitro fungal growth. In

particular cases, other media can be used that favor the growth
of certain fungi. In other cases, media can be amended with
antibiotics (streptomycin and chlortetracycline are among the
most common) or specific fungicides to avoid the growth of
certain potential contaminating microorganisms. For example,
Dichloran Rose Bengal Chloramphenicol Agar (DRBCA) is

often used to inhibit the growth of bacteria and contaminate
Mucorales fungi (e.g., Rhizopus spp., Mucor spp.).
2. Although optimal incubation conditions of PDA cultures can
vary depending on the fungal species (Table 1) [1–3], the large
majority of postharvest pathogens attacking fresh horticultural
produce will grow well in the range of 20–25 °C and complete
their life cycle (full colony growth) after 7–21 days of incuba-
tion at these temperatures.
3. Common hemacytometers used to count spores of postharvest
pathogens include the Thoma counting chamber, mostly used
for fungal species with a small spore size (e.g., Penicillium spp.,
Aspergillus, spp., Geotrichum spp., Botrytis spp., etc.), and the
Fuchs-Rosenthal chamber, typically used for species with a
larger spore size (e.g., Alternaria spp., Monilinia spp., Rhizo-
pus spp., Mucor spp., etc.). Counting instructions and calcula-
tions to determine the spore concentration are particular for
each type of chamber.
4. The following equation is used to obtain the spore suspension
at the desired final concentration (Cf): Ci × Vi = Cf × Vf, where
Ci is the known high spore concentration determined by
counting in the hemocytometer (initial concentration), Vf is
the final volume of suspension at the desired inoculum concen-
tration that we wish to prepare, and Vi is the unknown volume
of initial suspension that we need to transfer to the volume of
fresh emulsifier solution (Ve). Obviously, Vf = Vi + Ve.
5. If large volumes of spore suspension of fungal species with a
small spore size (e.g., Penicillium spp., Aspergillus, spp., Geo-
trichum spp.) are needed for semicommercial trials, they can be
prepared by rubbing with a sterile glass rod all the spores on the
agar surface of a 90-mm-diameter culture plate after adding
5 mL of 0.05% emulsifier solution. This can be repeated with
additional culture plates, if needed. Then, pass the spore sus-
pension through two layers of cheesecloth and dilute with
sterile water to an absorbance of 0.1 at 420 nm determined
with a spectrophotometer. This density is approximately equiv-
alent to 106 spores mL-1 [24].
6. Additional substances that can be added to the fungal inocu-
lum suspension include fruit juice to enhance the nutrition of
the fungus, hence boosting spore germination and mycelium
development, fungicides and/or antibiotics to kill other con-
taminating microorganisms potentially present in the inocula-
tion site, and cycloheximide or other analog protein inhibitors
to retard the possible healing of the rind wound inflicted for
inoculation [24].
7. The results of in vitro tests will not predict the effectiveness to

control disease on specific fruits and vegetables due to the
complex interactions among host, pathogen, and environment
that occur during in vivo disease development and/or to dif-
ferences on the release capability of antifungal active ingredi-
ents from films located on agar medium and from coatings
located on the surface of the fruit peel [4]. In general, factors
such as temperature, water activity, nutrient availability, com-
ponents and properties of the coating matrix and the antifungal
agent, and potential interactions of the agent or other coating
ingredients with fruit tissue components will greatly influence
the antifungal activity. Furthermore, surface properties of the
cuticle and the whole peel of fruits and vegetables can especially
influence the diffusion rate of antifungal agents in the host peel.
Thus, the release capability of antifungal agents from films
located on agar medium can differ from that of coatings
located on the fresh product peel surface, and the overall
performance of coating formulations is typically highly depen-
dent on the commodity species and even cultivar [25]. Never-
theless, in vitro tests can be very useful in sequential research to
properly identify and select the most promising ECs among a
variety of candidates (screening of ECs with different antifun-
gal agents and/or concentrations) or to investigate the activity
of particular ECs against a variety of fungal pathogens.
8. Material and size of the casting plate will vary depending on the
final purpose of the film. Examples of common use are
9-cm-diameter whole film disks produced in standard Petri
dishes and 14.1-cm-diameter film disks dried on HDPE plates.
In the latter case, the purpose is to cut the original disk into
numerous smaller disks and the material should ensure the
release of the films after drying. Films that are difficult to peel
from the plates can be cast onto plates covered with sterile wax
paper (0.2-mm thickness). Typically, the thickness of the
obtained dry films is 0.1–0.5 mm and it is very important to
minimize the thickness variation among film treatments.
9. Drying rates can also be improved with hot air ovens, micro-
waves, and vacuum driers. However, controlling the drying
conditions is critical since different evaporative levels and tem-
peratures can affect the physical and structural properties of the
resulting film [11]. This is especially relevant in the case of
some antifungal agents such as essential oils and natural
extracts where exposure to high temperatures might be a limi-
tation. If needed, film drying can be performed under aseptic
conditions by placing the emulsion plates inside a previously
sterilized laminar flow hood.
10. Aseptically produce agar plugs (5-mm diameter) with a sterile
cork borer from grown PDA cultures of the target pathogen
and plate three of them equidistantly on the surface of the test

dish containing fresh PDA medium. Afterwards, place the film
disk (16-mm diameter or smaller) on the center of the agar
surface and lid the plate. The rest of the indications are those
described in Subheading 3.2.2. Periodically during incubation,
measure the length of the inhibition zone in the direction from
the disk to each of the fungal plugs. In other cases, the agar
plate surface can be divided mentally into two equal areas, and
while the fungal plug is placed in the center of one of them, the
disk is placed in the center of the other.
11. Spores are often also referred to as colony-forming units
(CFU).
12. The following formula is used to obtain the percentage of
inhibition of spore germination (I): I (%) = [(CFUc-CFUf)/
CFUc] × 100, where CFUc is the number of germinated spores
(grown colonies) on PDA dishes plated with PDB spore sus-
pension from the plates containing the control film disks, and
CFUf is the number of germinated spores on PDA dishes
plated with PDB spore suspension from the plates with film
disks correspondent to each different treatment.
13. Artificial fungal inoculation of fresh produce can be performed
in different ways depending on the target pathogen and the
commodity. When the pathogen is a fungal species with a small
spore size (e.g., Penicillium spp., Aspergillus, spp., Geotrichum
spp., Botrytis spp., etc.), wounding and inoculation can be
performed simultaneously by immersing a stainless-steel rod
with a 1-mm-wide, 2-mm-long probe tip into the spore sus-
pension and wounding the fruit peel (Fig. 3, top). Small- and
medium-size fruit are usually wounded once in the equator
while several equidistant wounds can be inflicted in large-size
fruit. In some experiments with large fruit, it can be convenient
to inoculate the same fruit with two or even more different
target pathogens in order to save fruit and labor. Fruit inocu-
lation with fungal mycelial plugs can be considered with target
fungi of difficult sporulation (e.g., Lasiodiplodia spp. etc.
[26]).
14. Although brief fruit immersion is the most uniform and effec-
tive way to coat the fruit with the EC emulsion, in some cases,
it could be interesting to mimic coating application in indus-
trial packingline roller conveyors by pipetting a small amount
of the emulsion (0.1–0.5 mL, depending on the commodity)
onto each fruit and rubbing manually with gloved hands
[27]. This type of EC application will consume much lower
amounts of emulsion, so it can also be useful when the prepa-
ration for research purposes of a large quantity of the emulsion
is a limitation. In semicommercial or commercial trials, EC
treatments are habitually performed as spray application in

packingline machinery. Research pilot plants intended to
conduct semicommercial trials are generally equipped with
small-scale versions of fruit packinglines used in commercial
packinghouses, whereas commercial trials are directly con-
ducted in packinghouse facilities. When packingline machinery
is used, it is mandatory to thoroughly clean nozzles and roller
conveyors between treatments.
15. Semicommercial or commercial trials are typically conducted
with larger sample sizes. While samples of three to five repli-
cates of 100–300 fruit per treatment, depending on the type of
fruit commodity, are frequently used in semicommercial assays,
several replications per treatment of various entire fruit field
boxes, bins, or packages are used in commercial trials [22].
16. Semicommercial or commercial trials are typically conducted
with naturally infected fruit; i.e., intact fruit from the field not
artificially inoculated with the target pathogen. In this case, it is
important to determine the etiology of the lesions, i.e., the
pathogen(s) causing disease. Results are often reported only as
disease incidence (number of decayed fruit or visible lesions)
and pathogen sporulation, but the severity of natural decay can
also be assessed as disease indexes based on the quantification
of disease severity according to particular quantitative or quali-
tative scales. For instance, common scales give scores as a
function of the relative surface of the fruit covered by the
disease lesion. The following example was developed for pome-
granate postharvest crown decay caused by B. cinerea [28]:
0 = no lesion (visible infected area) or fungal mycelium pres-
ent, 1 = mycelium present in the crown, 2 = lesion ≤25% of
skin surface, 3 = lesion on 26–50% of skin surface, 4 = lesion
>50% of skin surface. In other cases, even more simple qualita-
tive scales are used: 0 = no visible decay, 1 = slight decay
symptoms, 2 = moderate decay symptoms, and 3 = severe
decay symptoms.
17. All three parameters determined for in vivo assessment of
disease control (disease incidence and severity and pathogen
sporulation) can be expressed as reduction percentage with
respect to the control treatment. For example, in the case of
disease incidence: DIR (%) = [(DIc-DIt)/DIc] × 100, where
DIR is the disease incidence reduction, DIc is the disease
incidence (number of infected wounds or fruit) in the control
treatment (uncoated or coated with EC without antifungal
ingredient(s)), and DIt is the disease incidence (number of
infected wounds or fruit) in the corresponding antifungal EC
treatment.
Acknowledgements
Spanish IVIA, INIA, AEI, and the European Union through

the project StopMedWaste (PRIMA Programme) and the
European Regional Development Fund (ERDF) of the Generalitat
Valenciana 2021–2027 are gratefully acknowledged for providing
financial support to conduct research on this topic. In memory of
Dr. Miguel Ángel del Rı́o, for his unconditional friendship, guid-
ance, and support.
References
1. Barkai-Golan R (2001) Postharvest diseases of 9. Palou L, Ali A, Fallik E et al (2016) GRAS
fruits and vegetables: development and con- plant- and animal-derived compounds as alter-
trol. Elsevier, Amsterdam natives to conventional fungicides for the con-
2. Bautista-Baños S (ed) (2014) Postharvest trol of postharvest diseases of fresh
decay: control strategies. Academic Press/Else- horticultural produce. Postharvest Biol Tech-
vier, London nol 122:41–52. https://doi.org/10.1016/j.
3. Palou L, Smilanick JL (eds) (2020) Postharvest postharvbio.2016.04.017
pathology of fresh horticultural produce. CRC 10. Valencia-Chamorro SA, Palou L, del Rı́o MA
Press/Taylor and Francis Group, Boca Raton et al (2008) Inhibition of Penicillium digita-
4. Palou L, Pérez-Gago MB (2021) GRAS salts as tum and Penicillium italicum by hydroxypro-
alternative low-toxicity chemicals for posthar- pyl methylcellulose-lipid edible composite films
vest preservation of fresh horticultural containing food additives with antifungal prop-
products. In: Spadaro D, Droby S, Gullino erties. J Agric Food Chem 56:11270–11278.
ML (eds) Postharvest pathology. Next genera- https://doi.org/10.1021/jf802384m
tion solutions to reducing losses and enhancing 11. Suhag R, Kumar N, Petkoska AT et al (2020)
safety, plant pathology in the 21st century, vol Film formation and deposition methods of edi-
11. Springer, Cham, pp 163–179 ble coating on food products: a review. Food
5. Valencia-Chamorro SA, Palou L, del Rı́o MA Res Int 136:109582. https://doi.org/10.
et al (2011) Antimicrobial edible films and 1016/j.foodres.2020.109582
coatings for fresh and minimally processed 12. Andrade RD, Skurtys O, Osorio FA (2012)
fruits and vegetables: a review. Crit Rev Food Atomizing spray systems for application of edi-
Sci Nutr 51:872–900. https://doi.org/10. ble coatings. Compr Rev Food Sci Food Saf 11:
1080/10408398.2010.485705 323–337. https://doi.org/10.1111/j.
6. de Souza EL, Ramos Berger LR, Marı́n A et al 1541-4337.2012.00186.x
(2020) Chitosan and other edible coatings for 13. Palou L, Valencia-Chamorro SA, Pérez-Gago
the control of postharvest diseases of fresh MB (2015) Antifungal edible coatings for fresh
fruit. In: Palou L, Smilanick JL (eds) Posthar- citrus fruit: a review. Coatings 5:962–986.
vest pathology of fresh horticultural produce. https://doi.org/10.3390/coatings5040962
CRC Press/Taylor and Francis Group, Boca 14. Campos CA, Gerschenson LN, Flores SK
Raton, pp 403–452 (2011) Development of edible films and coat-
7. Martı́nez-Romero D, Guillén F, Castillo S et al ings with antimicrobial activity. Food Biopro-
(2020) Aloe spp. gels to reduce fruit disease cess Technol 4:849–875. https://doi.org/10.
and maintain quality properties. In: Palou L, 1007/s11947-010-0434-1
Smilanick JL (eds) Postharvest pathology of 15. Balouiri M, Sadiki M, Koraichi Ibnsouda S
fresh horticultural produce. CRC Press/Taylor (2016) Methods for in vitro evaluating antimi-
and Francis Group, Boca Raton, pp 713–756 crobial activity: a review. J Pharm Anal 6:71–
8. Rajestary R, Landi L, Romanazzi G (2020) 79. https://doi.org/10.1016/j.jpha.2015.
Chitosan and postharvest decay of fresh fruit: 11.005
meta-analysis of disease control and antimicro- 16. Aloui H, Khwaldia K (2016) Natural antimi-
bial and eliciting activities. Compr Rev Food crobial edible coatings for microbial safety and
Sci Food Saf 20:563–582. https://doi.org/10. food quality enhancement. Compr Rev Food
1111/1541-4337.12672
Sci Food Saf 15:1080–1130. https://doi.org/ 23. Simko I, Piepho HP (2012) The area under the
10.1111/1541-4337.12226 disease progress stairs: calculation, advantage,
17. Valencia-Chamorro SA, Pérez-Gago MB, del and application. Phytopathology 102:381–
Rı́o MA et al (2009) Curative and preventive 389. https://doi.org/10.1094/PHYTO-07-
activity of hydroxypropyl methylcelluloselipid 11-0216
edible composite coatings containing antifun- 24. Eckert JW, Brown GE (1986) Evaluation of
gal food additives to control citrus postharvest postharvest fungicide treatments for citrus
green and blue molds. J Agric Food Chem 57: fruits. In: Hickey KD (ed) Methods for evalu-
2770–2777. https://doi.org/10.1021/ ating pesticides for control of plant pathogens.
jf803534a APS Press, St Paul, pp 92–97
18. Min S, Krochta JM (2005) Inhibition of Peni- 25. Pérez-Gago MB, Palou L (2016) Antimicro-
cillium commune by edible whey protein films bial packaging for fresh and fresh-cut fruits and
incorporating lactoferrin, lactoferrin hydroly- vegetables. In: Pareek S (ed) Fresh-cut fruits
sate, and lactoperoxidase system. J Food Sci and vegetables: technology, physiology, and
70:87–94. https://doi.org/10.1111/j. safety. CRC Press/Taylor and Francis Group,
1365-2621.2005.tb07108.x Boca Raton, pp 403–452
19. Franssen LR, Krochta JM (2000) Edible coat- 26. Guimarães JER, de la Fuente B, Pérez-Gago
ing containing natural antimicrobials for pro- MB et al (2019) Antifungal activity of GRAS
cessed foods. In: Roller S (ed) Natural salts against Lasiodiplodia theobromae in vitro
antimicrobial for minimal processing of foods. and as ingredients of hydroxypropyl
CRC Press, Boca Raton, pp 250–262 methylcellulose-lipid composite edible coat-
20. Padgett T, Han IY, Dawson PL (2000) Effect ings to control Diplodia stem-end rot and
of lauric acid addition on the antimicrobial maintain postharvest quality of citrus fruit. Int
efficacy and water permeability of corn zein J Food Microbiol 301:9–18. https://doi.org/
films containing nisin. J Food Process Preserv 10.1016/j.ijfoodmicro.2019.04.008
24:423–432. https://doi.org/10.1111/j. 27. Gunaydin S, Karaca H, Palou L et al (2017)
1745-4549.2000.tb00429.x Effect of hydroxypropyl methylcellulose-
21. Kristo E, Koutsoumanis KP, Biliaderis CG beeswax composite edible coatings formulated
(2008) Thermal, mechanical and water vapor with or without antifungal agents on physico-
barrier properties of sodium caseinate films chemical properties of plums during cold stor-
containing antimicrobials and their inhibitory age. J Food Qual 2017:8573549. https://doi.
action on Listeria monocytogenes. Food Hydro- org/10.1155/2017/8573549
coll 22:373–386. https://doi.org/10.1016/j. 28. Palou L, Crisosto CH, Garner D (2007) Com-
foodhyd.2006.12.003 bination of postharvest antifungal chemical
22. Palou L (2018) Postharvest treatments with treatments and controlled atmosphere storage
GRAS salts to control fresh fruit decay. Horti- to control gray mold and improve storability of
culturae 4:46. https://doi.org/10.3390/ ‘Wonderful’ pomegranates. Postharvest Biol
horticulturae4040046 Technol 43:133–142. https://doi.org/10.
1016/j.postharvbio.2006.08.013
Chapter 16
Antibacterial Activity of Active Food Packaging Materials

Paula J. P. Espitia and Rejane A. Batista
Abstract
Active food packaging materials with antimicrobial properties have stood out from other technological
products for food preservation due to its practical application and cost-benefit. This analytical protocol was
based on the “Antimicrobial Disk Susceptibility Tests” and adapted to its application for antibacterial food
packaging. It points out in a simplified, careful, and objective way the entire step-by-step methodology to
evaluate antimicrobial activity applied to food packaging incorporated with active compounds against
different microorganisms of interest to the food industry. Therefore, it is an interesting tool for students,
researchers, laboratories, and companies interested in the science of food packaging technology.
Key words Antimicrobial activity, Active food packaging, Disk diffusion method
1 Introduction
Spoilage microorganisms are one of the main reasons for food loss
and waste according to the Food and Agriculture Organization of
the United Nations (FAO), resulting in an estimated yearly loss of
40–50% fruits and vegetables, 35% fish, 30% cereals, and 20% dairy
and meat products [1]. On the other hand, foodborne pathogens
are affecting the lives of consumers, resulting in deaths, hospitaliza-
tions, and illnesses worldwide [2]. In this regard, the World Health
Organization has indicated that around 600 million people get sick
after consuming food that has been contaminated after its
production [3].
Safe production and commercialization of food require the use
of proper packaging materials, which allows for protecting it from
external threats, such as spoilage and pathogenic microorganisms,
oxygen, moisture, light, and heat, among others, as well as increas-
ing its shelf-life, decreasing postcontamination, and preserving its
quality until it reaches the final consumer [2, 4]. Moreover, food
packaging plays the role of unitizing food products and attracting
consumers with a marketing appeal, features that are essential in
different links of the food supply chain [5]. The worldwide market
https://doi.org/10.1007/978-1-0716-3613-8_16,
279
280 Paula J. P. Espitia and Rejane A. Batista
of food packaging has been estimated at over USD 300 billion in

2019, with a predicted increase rate of 5.2% yearly [1]. Thus, this is
a domain of high relevance not only from the food preservation
point of view and the economical perspective but also from the
innovation standpoint. In this regard, consumers’ demand for
healthier and safer food, together with the necessity of conve-
nience, has led to the development of innovative food packaging
technologies, including active food packaging [6].
Active food packaging interacts with the contained food prod-
uct in order to provide it with one or more beneficial function(s),
including scavenging of oxygen, moisture, or ethylene; emission of
flavor; and antioxidant or antimicrobial activities, among others
[7]. Antimicrobial food packaging stands out as a particular type
of active food packaging having in its structure an antimicrobial
agent that might be released into foodstuff, reducing or even
inhibiting the growth of spoilage microorganisms or foodborne
pathogens therein [8]. This results in food safety and quality assur-
ance, as well as increased shelf-life [1, 6].
Polymer-based antimicrobial food packaging can be developed
by direct incorporation of the antimicrobial agent into the polymer
matrix, followed by its progressive diffusion from the packaging to
the food surface. Readers can refer to Chap. 18 for detailed proto-
cols to determine the release kinetics of several active compounds,
including antimicrobials. Alternatives include intrinsically antimi-
crobial polymers or sachets that are previously incorporated with
volatile antimicrobial agents and placed within the packaging head-
space [1, 8]. In the latter strategy, although the antimicrobial agent
is inside the packaging, it does not touch the food surface, but
instead creates a modified atmosphere as a result of its progressive
vapor release.
Most of the tests applied to determine the antimicrobial activity
of food packaging materials include the use of foodborne patho-
gens (e.g., Escherichia coli, Staphylococcus aureus, and Salmonella
enterica) as well as different fungal strains, such as yeasts and
postharvest molds [1]. While antibacterial assays are detailed
herein, protocols for determining the antifungal efficiency of edible
packaging are depicted in Chap. 15. Yet, the recent pandemic of
Sars-CoV-2 served as a wake-up call for the scientific community to
also track the antiviral activity of active packaging materials.
Studying the antimicrobial activity allows researchers to have an
estimated performance of the biological activity of developed films
against selected microorganisms. However, it is highly important to
assess the antimicrobial activity in real food systems, considering
that variations in results might be observed due to the interaction of
the antimicrobial agent with some food components once it has
been released into the food matrix [9].
The antimicrobial activity packaging materials can be deter-
mined in vitro relying on optical density after applying the
Antibacterial Food Packaging 281
antimicrobial packaging material on a simulated solid food model

system, by colony counting, minimum inhibitory concentration
(MIC), or by the disc diffusion method [9]. The latter has been
reported with several names in the literature, including agar diffu-
sion method [9], antibacterial activity Kirby–Bauer test—disc dif-
fusion method [10], disc diffusion assay [11], disc diffusion
method [12], disk—or disc—diffusion method [12–15], Kirby–
Bauer disk diffusion assay [16], and overlay diffusion test
[17, 18]. Although difference in designations, all the reported
tests are based on the “Antimicrobial Disk Susceptibility Tests”
organized and presented by the Clinical and Laboratory Standards
Institute (CLSI)1 in its document “M02-A11—Performance Stan-
dards for Antimicrobial Disk Susceptibility Tests; Approved Stan-
dard” [19]. This protocol presents the methodology necessary to
apply the disk diffusion test in a standardized manner. This protocol
is reviewed and analyzed periodically, and its performance, applica-
tions, and limitations are presented as well. This standardized
method is based on the “Antibiotic susceptibility testing by a
standardized single disk method” originally described by Bauer
and collaborators in 1966 [20], and currently known as the
“Kirby-Bauer Disk Diffusion Susceptibility Test”. This is the most
detailed disk diffusion method, and it is based on the correlation of
the inhibition zone diameter with the MIC of different antimicro-
bial agents against microbial strains in order to determine if they are
susceptible or resistant [19].
Considering that the disk diffusion test is one the most widely
used techniques for determining the biological activity of antimi-
crobial food packaging, this protocol presents its main materials,
methods, and remarks for its successful application and
performance.
1.1 Brief History of The discovery of penicillin and its effects against bacterial infection
the Method caused by Staphylococci and Streptococci by Alexander Fleming was
one of the greatest findings of modern medicine. This led to the
development and application of microbiological analyses to deter-
mine susceptibility or resistance to antimicrobials of antibiotic
compounds, to further being used in treating human pathologies.
Although the original method for determining susceptibility to
antimicrobials was based on broth dilution techniques, the disk
diffusion method was a widely accepted and adopted methodology
among most US clinical microbiology laboratories by the decade of
1950s. However, this methodology was not standardized, and each
1
“The Clinical and Laboratory Standards Institute (CLSI) is an international, voluntary, nonprofit, interdisci-
plinary, standards-developing, and educational organization accredited by the American National Standards
Institute, which develops and promotes the use of consensus-developed standards and guidelines within the
health care community. These consensus standards and guidelines are developed to address critical areas of
diagnostic testing and patient health care, and are developed in an open and consensus-seeking forum” [19].
laboratory had an internal modification concerning methodological

variables (culture media, incubation time, incubation temperature,
among others). Thus, researcher Kirby and colleagues at the Uni-
versity of Washington School of Medicine and the King County
Hospital, after an extensive research regarding this methodology,
presented a single disk diffusion methodology intended to assess
the antimicrobial susceptibility in 1956. However, during the
1960s, the lack of standardization of this methodology was still a
problem. Kirby and collaborators continued to work on organizing
systematically and updating all information related to the disk
diffusion method and finally published their findings. This resulted
in the organization of a committee in 1961 by the World Health
Organization, with the goal of developing a standardized technique
for testing the antimicrobial susceptibility. The result was the
worldwide known Kirby–Bauer disk diffusion test [21]. In this
regard, the standardized method was widely adopted in microbiol-
ogy clinical laboratories by 1972. The advantages of this technique
include its relative simplicity of application from the technological
point of view and its reproducibility [22]. At present, the organiza-
tion in charge of updating and modifying the original procedure of
the Kirby–Bauer disk diffusion test is the CLSI, in order to allow
uniformity and reproducibility of results.
On the other hand, the first studies regarding antimicrobial
food packaging are dated from 1985, with the development of
waxes and other edible coatings incorporated with antimycotic
agents to protect food products [23]. In 1989, the term “active
packaging” was defined by Labuza and Breene as packaging that
“fosters desirable interactions” [24]. In 1990, the first reports
regarding the use of food-grade antimycotic agents incorporated
in cellulose edible films and their effectivity for food preservation
were published. In 1995, Rooney presented a complete literature
review on active and antimicrobial food packaging [25], which was
a compilation of published research in these areas. However, pub-
lished literature indicating the use or the adoption of the Kirby–
Bauer disk diffusion test in the field of antimicrobial food packaging
was scarce in the 1990s. The first studies related to the adaptation
of this technique to study the antimicrobial activity of active food
packaging date from 2000 and 2001, in which researchers referred
usually as “inhibition zone assay.”
Currently, the Kirby–Bauer disk diffusion test with modifica-
tions is extensively used to study the antimicrobial activity of active
food packaging incorporated with organic and inorganic antimicro-
bial agents, which can also be found in the polymer matrix even at
the nanoscale.
2 Materials
The following materials are needed for carrying out the disk
diffusion test:
• Sterile saline in 2-mL tubes.
• 0.5 McFarland standard.
• Wickerham card.
• Mueller–Hinton agar plates (100 or 150 mm).
• Caliper or ruler.
• Antibiotic disks.
• Forceps.
• Vortex.
• Sterile swabs.
• Alcohol 70%.
• Incubator.
• Disk of the developed packaging material intended to be tested.
• Laminar hood.
• Autoclave.
• UV light source.
• Bunsen burner.
• Microorganism to the antimicrobial activity of the developed
food packaging (foodborne or spoilage microorganism is
recommended).
3 Methods
3.1 Preparation of 1. Prepare the Mueller–Hinton agar from dehydrated media

the Mueller–Hinton according to the manufacturer’s instructions.
Agar Plates 2. Pour the Mueller–Hinton agar medium into Petri dishes. The
depth of the prepared agar should be 4 mm. This corresponds
to 25 mL of liquid culture medium in 100-mm Petri dishes or
60 mL of liquid culture medium poured in 150-mm Petri
dishes (see Note 4.1).
3. Once the Mueller–Hinton agar medium is poured into the
Petri dishes, allow the agar to equilibrate at room temperature.
4. If there is liquid on the surface of the poured agar, place the
Petri dishes ajar on their lids, to allow removal of the liquid and
its evaporation. It is important to do this procedure in a
pre-sterilized laminar flux hood for 10–30 min.
Fig. 1 McFarland standards representing different microbial concentrations (left to right: 0.5, 1.0, 2.0, 3.0, and
4.0 McFarland unit), positioned in front of a Wickerham card. (Adapted from Bioanalytic GmbH [www.
bioanalytic.de] with permission)
5. Once dried and ready to use, label each Petri dish regarding
tested microorganisms and antimicrobial agents incorporated
in the food packaging.
3.2 McFarland • The 0.5 McFarland Standard corresponds to 1.5 × 108 colony-
Standard forming unit (CFU) mL-1. The standard should be used
together with the Wickerham card, which is a card with black
and white lines in parallel that allows visual comparison of the
0.5 McFarland Standard with the prepared bacterial suspen-
sion (see Note 4.2) that will be used in the disk diffusion test
as inoculum (Fig. 1).
• Prior to its use, agitate the 0.5 McFarland Standard vigorously
to achieve a homogeneous turbidity.
• In order to determine the adequacy of the prepared bacterial
suspension (inoculum), hold side by side the 0.5 McFarland
Standard and the inoculum in front of the Wickerham card.
• Compare the turbidity of the inoculum and the 0.5 McFarland
Standard with proper light. The comparison should be done by
observing the parallel lines through both suspensions.
• If the prepared inoculum is less turbid than the 0.5 McFarland
Standard, add more microorganisms.
• Add more saline solution to the prepared inoculum if it is denser
(more turbid) than the 0.5 McFarland Standard. However, if the
difference in turbidity is too big, then it is recommended to start
over, instead of continuing to dilute the prepared bacterial
dilution.
3.3 Streaking This technique consists of placing an aliquot of the microbiolo-

Technique gical material on a point on the surface of the culture medium
and then spreading it out in order to obtain progressively smaller
amounts of the targeted microbiological material on the agar
surface. The main aim of the streaking technique is to obtain
sufficient rarefaction of the microbiological material, to the point
of obtaining isolated colonies, of bacteria or yeasts, in the solid
culture medium.
To ensure successful application of this technique, pay atten-
tion to the following recommendations:
• Use a nickel–chrome inoculation loop to seed the material
through streaks on the surface of the culture medium contained
in a Petri dish.
• Always pass the bacteriological loop through a flame or use a
disposable loop.
• Do not return the bacteriological loop over the streaked section
of the culture medium.
• Avoid a large number of streaked sections on the culture
medium.
• Do not pierce the medium with the bacteriological loop.
• Take a small amount of the initial microbiological material to be
isolated.
• Always work with all material close to the flame (Bunsen burner)
to avoid external contamination in your experiment.
• Follow a single direction when streaking the sample, for exam-
ple, from the edge to the center of the plate.
• Passing the bacteriological loop through the flame between each
streak sequence increases the likelihood of obtaining isolated
colonies.
A step-by-step is suggested below and illustrated in Fig. 2:
1. Unload the swab or the inoculating loop in an isolated corner
of the culture medium.
2. Make the first streaking. Following this, make streak sequences
in order to obtain the depletion of the inoculum from the loop
and consequently allow the microorganisms to develop form-
ing isolated colonies.
3. The beginning of the next streaking sequence should overlap
with the end of the previous one and may occupy the rest of the
plate.
Fig. 2 Steps involved in streaking a plate for discrete colonies (right-hand side): Parallel line quadrant streak
followed by undulating line quadrant streak. After each streak, the inoculating loop or needle should be flamed
to red hot and allowed to cool before proceeding to the next step. (Reproduced from Microbe Notes [www.
microbenotes.com/streak-plate-method-principle-methods-significance-limitations] with permission from
the author and owner)
3.3.1 Incubation After applying the streaking methodology, the Petri dishes should
be incubated inverted in a bacteriological incubator at 36 ± 2 °C for
24–48 h (conditions favorable to most microorganisms of interest.
This should be checked according to each study). After 24 h, it is
possible to evaluate the morphological characteristics of isolated
growing colonies.
3.3.2 Interpretation of 1. Observe the appearance of isolated colonies on agar plates and
the Isolation Results from differentiate according to the distinctive characteristics of the
the Exhaustion Methods target microorganism.
2. To obtain pure cultures, remove, with a sterile inoculation
needle, a portion of a chosen colony and transfer it aseptically
to an inclined simple agar tube, making a streak on its surface.
Incubate at 36 ± 2 °C for 24 h.
3. After the incubation period, observe the culture grown in the
tube and check the purity considering its homogeneous aspect
and through Gram stain.
3.4 Inoculum 1. Take 4–5 isolated colonies from the Petri dish with the cultured
Elaboration microorganism to be tested, using an inoculating loop (see
Note 4.3).
2. Suspend the colonies in sterile saline (2 mL).
3. Homogenize the tube with the suspended colonies.
4. Verify the desired turbidity according to the 0.5 McFarland
Standard and adjust it if necessary according to item 3.2.
5. Use the inoculum within 15 min after it is prepared.
3.5 Inoculation of the 1. Immerse a sterile swab in the prepared inoculum.

Petri Dishes 2. Rotate the swab against the internal side of the tube containing
Containing Mueller– the inoculum. This should be done with pressure enough to
Hinton Agar remove the excess liquid contained in the swab, which should
not be dripping wet.
3. Proceed to inoculate the Petri dish containing the Mueller–

Hinton agar by streaking three times the agar surface with the
swab containing the bacterial suspension. Then rotate the plate
(60 ° approximately) and repeat the procedure. Ensure an equal
distribution of the inoculum on the entire agar surface.
4. Verify that there is no excess inoculum liquid on the agar
surface. To avoid this, rim the Petri dish with the swab.
5. Prepare the used swab for disposal by autoclaving beforehand
and discard in the appropriate container.
6. Allow the surface of the agar to dry by placing the lid of the
Petri dish ajar. This procedure should be done at room temper-
ature for a maximum of 15 min.
3.6 Placement of 1. To place the antimicrobial food packaging disks, slightly

Antimicrobial remove the lid from the Petri dish and with the help of a
Packaging Disks in the previously sterilized forceps, place the antimicrobial food pack-
Inoculated Petri Dishes aging disks (6 mm in diameter) on the surface of Mueller–
Hinton agar (Fig. 3).
2. Apply gentle pressure with the forceps on top of the disks to
ensure its contact with the agar surface.
3. Place the lid on the Petri dish to minimize the exposure of the
agar to the external environment and repeat the process
according to the numbers of antimicrobial food packaging
desired to be tested. For each antimicrobial food packaging
tested, a control treatment—food packaging without any
antimicrobial—should be prepared and tested simulta-
neously (see Note 4.4).
Fig. 3 Placement of disk on the surface of inoculated Petri dishes. Reproduced from Microbe Notes with
permission from the author and owner [www.microbenotes.com/kirby-bauer-disc-diffusion/]
4. Once all disks are placed on the agar surface, invert the Petri
dishes and incubate. Ideal incubation conditions for foodborne
pathogens and spoilage microorganisms should be checked in
each specific case.
3.7 Measurement of • After the incubation period, measure the inhibition zone on the
the Inhibition Zone back of each Petri dish with a caliper or a ruler (see Note 4.5).
The diameter of the disk should be considered in this measure-
ment (Fig. 3).
• The results should be determined without any visual aid. The
Petri dish should be held on a black surface, and enough light
should be provided to see the back of the Petri dishes. The
results should be observed using a vertical sight line.
• Microbial growth starting at the edge of the disk should be
reported as 0 mm.
3.8 Results Report The obtained results are presented as a measurement of the diame-
and Presentation ter of the inhibition zone [mm].
Although the original “Kirby-Bauer Disk Diffusion Suscepti-
bility Test” establishes a guideline published by the CLSI to deter-
mine the level of susceptibility of the selected microorganism to the
tested antimicrobial—susceptible (S), intermediate (I), or resistant
(R)—based on an interpretation chart, in the field of antimicrobial
food packaging there is so far no standardization regarding the
microbial susceptibility level. This is in part due to the lack of a
specific institution in charge of that task in this field, as well as due
to the increasing number of antimicrobial agents that are
incorporated in polymeric matrices to later be studied with the
potential application for food preservation.
4 Notes
All procedures described in this protocol should be done in a sterile

environment. Agar and Petri dishes should be sterilized, e.g., in an
autoclave while packaging disks (samples to be tested) should be
sterilized with UV, autoclave, or any other suitable method.
Researchers should be provided with laboratory coats or appropri-
ate clothes, as well as gloves for personal protection.
Besides this general warning, notes on specific steps are listed
below:
4.1 Preparation of 1. Erroneous results might be produced if the Mueller–Hinton

the Mueller–Hinton agar medium poured in Petri dishes has a different depth. If the
Agar Plates agar medium is too shallow in the Petri dishes, the antimicro-
bial agent released from the developed packaging material will
diffuse further than it should, and the resulted inhibition zone
will be bigger than expected. If the depth of the agar medium

in the Petri dishes is higher, it will result in an inaccurate
resistant result.
4.2 McFarland 1. This Standard can be commercially purchased or prepared in

Standard house [26].
2. The 0.5 McFarland Standard should be replaced when large
particle aggregates are observed instead of a homogeneously
turbid suspension.
4.3 Inoculum 1. The selected microorganism for testing the antimicrobial activ-
Elaboration ity of the food packaging should be in the log phase. Thus, the
microorganism should be activated to its log phase the previous
day to performing this technique. This procedure should be
adjusted according to the selected microorganism.
2. Different microorganisms have been reported to be used to test
the antimicrobial activity of active food packaging. Table 1
presents the reported microorganisms used, as well as polymer
matrices and antimicrobial agents tested using the disk diffu-
sion method.
4.4 Placement of 1. It is recommended to place the disks not closer than 24 mm on

Antimicrobial the Mueller–Hinton agar. For Petri dishes of 150 mm in diam-
Packaging Disks in the eter, a maximum of 12 disks (6 mm in diameter each) should be
Inoculated Petri Dishes placed, and for Petri dishes of 100 mm in diameter, a maximum
of five disks should be placed on the surface of the Mueller–
Hinton agar. When the diameter of the disk is larger, less disks
should be used.
2. A template of the positions at which the disks will be placed on
the Mueller–Hinton agar surface should be used, considering
the spacing among disks and the maximum number of disks
according to each type of Petri dish.
3. Prevent the placement of disks close to the border of the Petri
dish, since after incubation, the results are likely not to be
possible to read with accuracy.
4. Avoid altering the physical integrity of the Mueller–Hinton
agar surface (any disruption) due to excessive pressure when
inoculating or placing the disks, or by dipping the disk inside
the agar. In these cases, the procedure should be repeated since
the obtained results will not be possible to read due to the lack
of accuracy.
4.5 Measurement of 1. It is expected that the inhibition zone exhibited should be

the Inhibition Zone circular, with a convergent microbial growth around it, if the
agar was inoculated and the disks were placed properly.
Table 1
Microorganisms used to test the antimicrobial activity of active food packaging using the disk
diffusion method
Food packaging
Microorganism Antimicrobial agent material Reference
Listeria monocytogenes Olive lea extract Fish gelatin [17]
Staphylococcus aureus and Pomegranate peel extract Hydroxypropyl high- [12]
Salmonella amylose starch
Salmonella enterica, ZnO nanoparticles and ascorbic Nanocomposites [10]
L. monocytogenes, Yersinia acid comprising
enterocolitica, Pseudomonas tragacanth gum and
aeruginosa, Escherichia coli, polyvinyl alcohol
S. aureus, and Enterococcus
faecalis
E. coli Carvacrol and thymol Low-density [16]
polyethylene filled
with halloysite
nanotubes
E. coli O157:H7 and S. aureus Rosin-grafted cellulose Bovine gelatin [27]
nanocrystals
E. coli and Bacillus subtilis Rosin used to modify the Halloysite nanotubes [28]
surface of cellulose coated with
nanocrystals to obtain rosin- chitosan
modified CNF (R-CNF)
S. aureus (MRSA) and E. coli Silver Gelatin [14]
(DH5-alpha)
Bacillus cereus, S. aureus, E. coli, Chitosan nanoparticles Tapioca starch [11]
and Salmonella Typhi
E. coli O157:H7 and S. aureus Chitosan Chitosan/ε-PL [29]
bionanocomposites
films
B. subtilis and E. coli Lauric acid Bacterial cellulose [15]
2. If it is not possible to read the complete diameter of the

inhibition zone, it is recommended to measure the radius to
another edge of the circumference of the inhibition zone and
multiply this result by two to determine the diameter.
3. The test must be rerun when the microbial growth is not
consistent on the agar’s surface, but instead individual colonies
are observed across the surface.
4. When measuring the inhibition zone, this should be deter-
mined bearing in mind that this corresponds to a zone without
any evident microbial growth.
Acknowledgments
The authors thank the support provided by the Foundation for

Research and Technological Innovation Support of the State of
Sergipe (FAPITEC/SE).
References
1. Motelica L, Ficai D, Ficai A, Oprea OC, Kaya (ed) Advances in food biotechnology. Wiley,
DA, Andronescu E (2020) Biodegradable anti- New York
microbial food packaging: trends and perspec- 10. Janani N, Zare EN, Salimi F, Makvandi P
tives. Foods 9:1438 (2020) Antibacterial tragacanth gum-based
2. Al-Tayyar NA, Youssef AM, Al-hindi R (2020) nanocomposite films carrying ascorbic acid
Antimicrobial food packaging based on sus- antioxidant for bioactive food packaging. Car-
tainable bio-based materials for reducing food- bohydr Polym 247:116678. https://doi.org/
borne pathogens: a review. Food Chem 310: 10.1016/j.carbpol.2020.116678
1 2 5 9 1 5 . h tt p s : // d o i. o r g /1 0 . 1 0 1 6 / j. 11. Shapi’i RA, Othman SH, Nordin N, Kadir
foodchem.2019.125915 Basha R, Nazli Naim M (2020) Antimicrobial
3. Kim I, Viswanathan K, Kasi G, properties of starch films incorporated with
Thanakkasaranee S, Sadeghi K, Seo J (2020) chitosan nanoparticles: in vitro and in vivo eval-
ZnO nanostructures in active antibacterial food uation. Carbohydr Polym 230:115602.
packaging: preparation methods, antimicrobial https://doi.org/10.1016/j.carbpol.2019.
mechanisms, safety issues, future prospects, 115602
and challenges. Food Rev Int:1–29. https:// 12. Ali A, Chen Y, Liu H, Yu L, Baloch Z, Khalid S,
doi.org/10.1080/87559129.2020.1737709 Zhu J, Chen L (2019) Starch-based antimicro-
4. Kim I, Viswanathan K, Kasi G, Sadeghi K, bial films functionalized by pomegranate peel.
Thanakkasaranee S, Seo J (2019) Poly(lactic Int J Biol Macromol 129:1120–1126. https://
acid)/ZnO bionanocomposite films with posi- doi.org/10.1016/j.ijbiomac.2018.09.068
tively charged ZnO as potential antimicrobial 13. Fuenmayor CA, Espitia PJP (2018) Electro-
food packaging materials. Polymers 11:1427 spun nanofibers: development and potential in
5. Espitia PJP, Otoni CG (2018) Nanotechnol- food packaging applications in nanotechnology
ogy and edible films for food packaging appli- applications in the food industry. In: Rai V, Bai
cations in bio-based materials for food J (eds) Nanotechnology applications in the
packaging. In: Ahmed S (ed) Bio-based mate- food industry. CRC Press, Boca Raton
rials for food packaging. Springer, Singapore 14. Sarwar MS, Niazi MBK, Jahan Z, Ahmad T,
6. Kumar S, Mukherjee A, Dutta J (2020) Chit- Hussain A (2018) Preparation and characteri-
osan based nanocomposite films and coatings: zation of PVA/nanocellulose/Ag nanocompo-
emerging antimicrobial food packaging alter- site films for antimicrobial food packaging.
natives. Trends Food Sci Technol 97:196–209. Carbohydr Polym 184:453–464. https://doi.
https://doi.org/10.1016/j.tifs.2020.01.002 org/10.1016/j.carbpol.2017.12.068
7. Huang T, Qian Y, Wei J, Zhou C (2019) Poly- 15. Zahan KA, Azizul NM, Mustapha M, Tong
meric antimicrobial food packaging and its WY, Abdul Rahman MS, Sahuri IS (2020)
applications. Polymers 11:560 Application of bacterial cellulose film as a bio-
8. Santos JCP, Sousa RCS, Otoni CG, Moraes degradable and antimicrobial packaging mate-
ARF, Souza VGL, Medeiros EAA, Espitia PJP, rial. Mater Today Proc 31:83–88. https://doi.
Pires ACS, Coimbra JSR, Soares NFF (2018) org/10.1016/j.matpr.2020.01.201
Nisin and other antimicrobial peptides: pro- 16. Krepker M, Shemesh R, Danin Poleg Y,
duction, mechanisms of action, and application Kashi Y, Vaxman A, Segal E (2017) Active
in active food packaging. Innov Food Sci food packaging films with synergistic antimi-
Emerg Technol 48:179–194. https://doi. crobial activity. Food Control 76:117–126.
org/10.1016/j.ifset.2018.06.008 https://doi.org/10.1016/j.foodcont.2017.
9. Espitia PJP, Batista RA (2015) Non-thermal 01.014
food preservation: control of food-borne 17. Albertos I, Avena-Bustillos RJ, Martı́n-Diana
pathogens through active food packaging and AB, Du W-X, Rico D, McHugh TH (2017)
nanotechnology in advances in food Antimicrobial olive leaf gelatin films for
biotechnology. In: Ravishankar RV enhancing the quality of cold-smoked Salmon.
Food Packag Shelf Life 13:49–55. https://doi. 25. Rooney ML (1995) Overview of active food
org/10.1016/j.fpsl.2017.07.004 packaging in active food packaging. In: Rooney
18. Espitia PJP, Avena-Bustillos RJ, Du W-X, ML (ed) Active food packaging. Springer,
Chiou B-S, Williams TG, Wood D, McHugh New York
TH, Soares NFF (2014) Physical and antibac- 26. Garcia L (2010) Preparation of routine media
terial properties of Açaı́ edible films formulated and reagents used in antimicrobial susceptibil-
with thyme essential oil and apple skin poly- ity testing in clinical microbiology procedures
phenols. J Food Sci 79:M903–M910. https:// handbook. In: Garcia LS (ed) Clinical microbi-
doi.org/10.1111/1750-3841.12432 ology procedures handbook. Washington, DC,
19. CLSI (2012) Performance standards for anti- ASM Press
microbial disk susceptibility tests; approved 27. Leite LSF, Bilatto S, Paschoalin RT, Soares AC,
standard-eleventh edition. CLSI document Moreira FKV, Oliveira ON, Mattoso LHC,
M02-A11. 76p Bras J (2020) Eco-friendly gelatin films with
20. Bauer AW, Kirby WM, Sherris JC, Turck M rosin-grafted cellulose nanocrystals for antimi-
(1966) Antibiotic susceptibility testing by a crobial packaging. Int J Biol Macromol 165:
standardized single disk method. Am J Clin 2974–2983. https://doi.org/10.1016/j.
Pathol 45:493–496 ijbiomac.2020.10.189
21. Hudzicki J (2009) Kirby-Bauer disk diffusion 28. Niu X, Liu Y, Song Y, Han J, Pan H (2018)
susceptibility test protocol. 23p Rosin modified cellulose nanofiber as a reinfor-
22. Biemer JJ (1973) Antimicrobial susceptibility cing and co-antimicrobial agents in polylactic
testing by the Kirby-Bauer disc diffusion acid/chitosan composite film for food packag-
method. Ann Clin Lab Sci 3:135–140 ing. Carbohydr Polym 183:102–109. https://
doi.org/10.1016/j.carbpol.2017.11.079
23. Hotchkiss JH (1995) Safety considerations in
active packaging in active food packaging. In: 29. Wu C, Sun J, Lu Y, Wu T, Pang J, Hu Y (2019)
Rooney ML (ed) Active food packaging. In situ self-assembly chitosan/ε-polylysine bio-
Springer, New York nanocomposite film with enhanced antimicro-
bial properties for food packaging. Int J Biol
24. Hotchkiss JH (1997) Food-packaging interac- Macromol 132:385–392. https://doi.org/10.
tions influencing quality and safety. Food Addit 1016/j.ijbiomac.2019.03.133
Contam 14:601–607. https://doi.org/10.
1080/02652039709374572
Chapter 17
Antioxidant Activity Assays for Food Packaging Materials

Fabiana H. Santos, Danielle C. M. Ferreira, Julia R. V. Matheus,
Ana E. C. Fai, and Franciele M. Pelissari
Abstract
Antioxidant packaging is an emerging technology that limits deteriorative reactions in oxidation-sensitive
food products. The direct interaction of the antioxidant material with the packaged product may inhibit
oxidation reactions by scavenging free radicals, consequently improving the food stability and extending its
shelf-life. Although these packages represent a promising alternative for preserving food, until now,
standardized procedures to accurately quantify their efficacy have been lacking. The methodologies
employed to assess the antioxidant activity of food packaging are the same as those already used for natural
extracts. These methods measure the ability of the analyzed material to scavenge free radicals. Herein, we
describe in detail the principal methodologies that have been used to evaluate the antioxidant activity of
food packaging materials.
Key words Antioxidant packaging, Methodologies, Free radicals, DPPH, ABTS, FRAP, ORAC,
Spectrophotometry
1 Introduction
Packaging plays a fundamental role in food quality and preserva-

tion. However, in the last years, traditional packaging has failed to
prevent or even delay food degradation reactions, such as lipid
oxidation, enzymatic browning, and microbial growth. Thus, con-
siderable efforts have been made to reduce degradation effects and
meet the growing consumer demand for safe and high-quality
products [1].
Antioxidant packaging is an innovative strategy that limits
deteriorative reactions in oxidation-sensitive food products, conse-
quently improving their stability and extending their shelf-life
[1, 2]. The direct interaction of the antioxidant material with the
packaged product may inhibit oxidation reactions by scavenging
free radicals [2]. Commonly, the deactivation of radicals occurs by
https://doi.org/10.1007/978-1-0716-3613-8_17,
293
294 Fabiana H. Santos et al.
two main mechanisms, namely, single-electron transfer (SET) and

hydrogen atom transfer (HAT) [3]. SET is related to the antioxi-
dant ability to transfer one electron to reduce free radicals, includ-
ing metals and carbonyls [4]. HAT is associated with the
antioxidant ability to quench free radicals by hydrogen
donation [3].
Given that, several in vitro methods have been used to assess
the antioxidant activity of a wide variety of matrices, including food
packaging materials [5]. One of the most important methods is the
oxygen radical absorbance capacity (ORAC), which is based on the
HAT reaction mechanism [3]. ORAC evaluates the antioxidant
capacity to inhibit the dissolution of a fluorescent probe, usually
fluorescein [6]. In this assay, a peroxyl radical reacts with fluores-
cein, forming a nonfluorescent product. However, the presence of
antioxidants inhibits or protects the fluorescein from this reaction.
The protective effect (antioxidant activity) is measured by compar-
ing the area under the fluorescence decay curve of the antioxidant
sample with the area of the control sample [7].
One of the most useful SET-based methods is the ferric reduc-
ing antioxidant power (FRAP) [3]. This assay measures the antioxi-
dant capacity of the material based on the reduction of the FRAP
ferric ion (Fe3+) to ferrous iron (Fe2+). The reduction leads to a
color change in the solution from light blue to dark blue [8]. ABTS
(2,2′-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid)) and
DPPH (2,2-diphenyl-1-picryl-hydrazyl-hydrate) methods are
often related to SET reaction mechanisms. However, their radicals
may be neutralized either by a direct reduction via electron transfer
or by a radical quenching via H atom transfer [3]. Therefore, they
can be classified as SET- or HAT-based methods.
The ABTS assay, also known as TEAC (Trolox equivalent
antioxidant capacity), evaluates the antioxidant capacity to scavenge
the ABTS radical. The presence of antioxidants can change the
color of the solution containing the ABTS radical from dark green
to light green. On the other hand, the DPPH method measures the
capacity of the antioxidant to reduce the DPPH radical [3]. DPPH
is a stable hydrophobic radical with a deep purple color. Reactions
with other radicals, electrons, or hydrogen atoms lead to color loss.
The antioxidant activity is determined according to the color
change [9].
These assays are often low-cost, easy to perform, and do not
require ultra-sensitive equipment [4]. However, most of them are
global and nonspecific, which requires more than one test to assess
the antioxidant activity of a product [10]. This chapter will present
a description of the main methods to evaluate the antioxidant
activity of food packaging materials.
Antioxidant Activity of Food Packaging 295
2 Materials
2.1 DPPH Assay 1. Packaging sample.

2. 2,2-Diphenyl-l-picrylhydrazyl (DPPH•).
3. Methanol or another organic solvent (see Note 1).
4. Grade-1 qualitative filter paper.
5. 6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid
(Trolox).
6. Glass beaker.
7. Scissors.
8. Falcon tube.
9. Analytical balance.
10. Vortex homogenizer.
11. Ultrasound bath.
12. Aluminum foil.
13. Amber bottle.
14. UV-visible spectrophotometer.
2.2 ABTS Assay 1. Packaging sample.

2. 2,2′-Azino-bis(3-ethylbenzothiazoline-6-sulfonic) acid (ABTS)
(PubChem CID: 6871216).
3. Potassium persulfate (K2S2O8) (PubChem CID: 54602120).
4. Distilled water.
5. Absolute ethanol (C2H6O) (PubChem CID: 702).
6. Standard Trolox (PubChem CID: 40634).
7. Glass beaker.
8. Volumetric flask.
9. Amber bottle.
10. Analytical balance.
2.3 FRAP Assay 1. Packaging sample.

2. Hydrochloric acid P.A. (HCl).
3. 2,4,6-Tris(2-pyridyl)-1,3,5-triazine (TPTZ) (PubChem CID:
77258).
4. Ferric chloride (FeCl3).
5. Acetate buffer pH 3.6: sodium acetate (C2H3NaO2) and acetic
acid (CH3COOH).
6. Distilled water.
7. Volumetric flask.
8. Amber bottle.
2.4 ORAC Assay 1. 75 mmol L-1 phosphate buffer pH 7.0: monosodium phos-
phate monohydrate, disodium phosphate heptahydrate, and
1 M hydrochloric acid (HCl).
2. Packaging sample.
3. Distilled water.
4. Fluorescein disodium salt.
5. 2,2′-Azobis(2-amidinopropane) dihydrochloride (AAPH).
6. 6-Hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid
(Trolox).
7. Micropipette.
8. Fluorescence spectrophotometer.
3 Methods
3.1 DPPH Assay 1. Cut the packaging sample into small pieces.
3.1.1 Packaging Sample 2. Weigh 100 mg of the sample using an analytical balance.
Extracts (See Note 2) 3. Place the sample in a Falcon tube and add 5 mL of methanol.
4. Shake the solution for 3 min using a vortex homogenizer.
5. Keep the solution at room temperature for 3 h and shake it
again for 3 min before use.
6. To calculate the inhibitory concentration (IC50), at least three
more concentrations of packaging extracts must be prepared.
3.1.2 0.06 mmol L-1 1. Weigh 4 mg of DPPH in a glass beaker wrapped with
DPPH Solution (See Note 3) aluminum foil.
2. Add 100 mL of methanol (see Note 4).
3. Place the solution in an ultrasound bath for 5 min at room
temperature.
4. Filter the solution using a grade1 qualitative filter paper.
5. Store in an amber bottle at 4 °C in the dark (see Note 5).
3.1.3 Performing the 1. Add 0.1 mL of the package extract to 3.9 mL of a DPPH
DPPH Assay (Fig. 1) methanol solution (0.06 mmol L-1).
(See Note 6) 2. Prepare a control sample only with methanol and a DPPH
solution using the proportions previously described (0.1 mL
of methanol and 3.9 mL of a DPPH solution).
Fig. 1 Simplified flowchart for the DPPH analysis with principle of DPPH radical scavenging by the HAT reaction
mechanism
3. Keep the solutions at room temperature for 30 min in the dark

(see Note 7).
4. While the solutions are incubated, turn on the spectrophotom-
eter and reset the equipment with methanol.
5. Transfer the solution to an optical glass cuvette and measure its
UV absorbance at 517 nm (see Notes 8 and 9).
6. Assays should be performed at least in triplicate for each sample
(see Note 10).
7. The antioxidant capacity of each sample can be calculated using
Eq. 1:
DPPH inhibition (%) = [(Ac - A s )=(A c )] × 100 (1)

where Ac is the absorbance value at 517 nm of the control
solution and As is the absorbance value at 517 nm of the sample
solution.
8. To compare the results obtained with the antioxidant activity of
a known antioxidant sample, build a Trolox standard curve (see
Notes 11 and 12) using the procedure described for the
packaging sample at different concentrations (see Note 13).
The standard curve is obtained by plotting the Trolox absor-
bance as a function of the Trolox concentration (mmol L-1).
The results are expressed as μmol Trolox equivalents (TE)/g
dry sample.
9. The antioxidant capacity of the packaging can also be expressed

as IC50 (inhibitory concentration), which represents the
required concentration of the packaging to reduce the initial
concentration of DPPH by 50% (see Note 14). The results are
expressed as mg of packaging sample/mL of extract.
3.2 ABTS Assay 1. Weigh 1 g of the food packaging material (see Note 15).
3.2.1 Packaging Sample 2. Cut the sample into small pieces and immerse them in 5 mL of
Extracts [11] a hydroalcoholic mixture (see Note 16).
3. Keep the mixture under stirring overnight (see Note 17).
4. Prepare at least three different dilutions of the sample extract
[12, 13] (see Note 18).
5. Store under refrigeration in a dark amber bottle until assay.
3.2.2 7 mmol L-1 ABTS 1. Weigh 192 mg of ABTS and dissolve in distilled water until the
Stock Solution volume is 50 mL in a volumetric flask.
2. Homogenize and transfer the solution to an amber glass bottle.
3. Store under refrigeration for up to a month.
3.2.3 140 mmol L-1 1. Weigh 378.4 mg of potassium persulfate and dissolve in dis-
Potassium Persulfate tilled water until the volume is 10 mL in a volumetric flask.
Solution 2. Homogenize and transfer the solution to an amber glass bottle.
3. Store at room temperature for up to a month.
3.2.4 1 mmol L-1 Trolox 1. Weigh 12.5 mg of Trolox and dissolve in ethyl alcohol until the
Standard Solution volume is 50 mL in a volumetric flask.
Prepare and use only on the day of analysis.
3. Prepare the standard curve using the standard 1 mmol L-1
Trolox solution, as shown in Table 1.
Table 1
Preparation of the calibration curve
Trolox 1 mmol L-1 solution Final concentration

concentration (mL) Ethanol (mL) (μmol L-1)
0 10.0 0
0.5 9.5 50
2.0 8.0 200
3.0 7.0 300
4.0 6.0 400
5.0 5.0 500
6.0 4.0 600
3.2.5 Performing the 1. Mix an ABTS solution (7 mmol L-1) with a potassium persul-
ABTS Assay fate solution (2.45 mmol L-1) (1:0.5 v/v).
2. Keep the mixture at room temperature for 16 h in the dark [15]
(see Note 19).
3. Dilute an aliquot of the ABTS●+ solution with ethanol to
obtain an absorbance value of 0.70 at 734 nm using a UV-Vis
spectrophotometer [15] (see Note 20).
4. Solubilize each sample extract dilution or pure solvent (con-
trol) (60 μL) with the ABTS●+ diluted solution (2940 μL) and
incubate at 37 °C for 10 min in the dark [15–18].
5. Measure the absorbance at 734 nm in a UV-Vis spectropho-
tometer [15–18] (see Note 21).
6. Use Trolox as a standard to obtain the calibration curve (con-
centration ranging from 0 to 600 μmol L-1) (see Note 22) and
express the results as μmol L-1 Trolox equivalent/g of dry
sample (see Note 23).
7. Perform all measurements in triplicate.
3.3 FRAP Assay 1. The food packaging extract in the FRAP assay should be
prepared as described in Subheading 3.2.1.
3.3.1 Packaging Sample
Extracts
3.3.2 40 mmol L-1 HCl 1. Add 3.34 mL of concentrated HCl in a volumetric flask and
Solution add distilled water until the volume is 1 L.
2. Homogenize and transfer the mixture to a glass bottle. Store at
room temperature.
3.3.3 10 mmol L-1 TPTZ 1. Weight 3.12 g of TPTZ and dissolve in 5 mL of an HCl
Solution 40 mmol L-1 solution.
2. Add HCl 40 mmol L-1 until the volume is 1 L in a volumetric
flask.
3.3.4 20 mmol L-1 Ferric 1. Weight 5.4 g of ferric chloride and dissolve in distilled water
Chloride Solution until the volume is 1 L in a volumetric flask.
(FeCl3·6H2O) 2. Homogenize and transfer the mixture to an amber glass bottle.
3.3.5 30 mmol L-1 1. Weight 0.31 g of sodium acetate (C2H3NaO2) and dissolve in
Acetate Buffer Solution 1.6 mL of acetic acid (CH3COOH).
(pH 3.6) 2. Add distilled water until the volume is 1 L in a volumetric flask.
3. Homogenize and transfer the mixture to an amber glass bottle.

4. Store at room temperature.
3.3.6 1 mmol L-1 Trolox 1. Use the procedure described in Subheading 3.2.4 to prepare
Standard Solution the Trolox standard solution.
3.3.7 Performing the 1. Mix 25 mL of acetate buffer 30 mmol L-1 (pH 3.6), 2.5 mL of
FRAP Assay TPTZ solution (10 mmol L-1), and 2.5 mL of a FeCl3·6H2O
solution (20 mmol L-1) and incubate at 37 °C for 30 min in a
water bath [19] (see Note 24).
2. Solubilize each sample extract dilution (150 μL) with a FRAP
solution (2850 μL) and keep at room temperature for 30 min
in the dark [19–22].
3. Measure the absorbance at 593 nm using a spectrophotometer
[19, 20] (see Note 25).
4. Use Trolox as a standard to obtain the calibration curve (con-
centration ranging from 0 to 600 μmol L-1) as described in the
ABTS assay (see Note 26).
5. Calculate the antioxidant activity of the sample by subtracting
the blank sample (see Note 27) and express the results as μmol
L-1 Trolox equivalent/g of dry sample.
6. Perform all measurements in triplicate.
3.4 ORAC Assay (See 1. 75 mmol L-1 aqueous phosphate buffer can be prepared in a
Note 28) 2-L solution with monosodium phosphate monohydrate and
disodium phosphate heptahydrate at pH 7.0 using 1 mol L-1
3.4.1 Phosphate Buffer
hydrochloric acid (HCl).
(pH 7.0)
3.4.2 Final Reaction 1. The samples (packages) must be initially solubilized in distilled
Mixture Based on the 2 mL water (0.0500 g/10 mL of water, 50 °C, 1 h), and, subse-
Volume (See Notes 29 quently, each initial solution must be adequately diluted in
and 30) sodium phosphate buffer (75 mmol L-1, pH 7.0) [14].
3.4.3 Performing the 1. Add 1650 μL of fluorescein sodium salt (0.05 μL) to sodium
ORAC Assay phosphate buffer 0.075 mol L-1 at pH 7.0 (see Note 31).
2. Mix 200 mL of the diluted sample or 50 mmol L-1 of Trolox
(Trolox at five calibration solutions (0–50 μmol L-1) is used as
a standard) with the previous solution.
3. Incubate the mixture at a constant temperature of 37 °C for
15 min (see Note 32).
4. Check the fluorescence intensity (485 nmexcitation/525 nmemis-
sion) every 5 min for 80 min in a quartz cuvette (see Note 33).
Fig. 2 Reaction of 2,2′-Azobis (2-amidinopropane) dihydrochloride (AAPH) during the ORAC assay using
fluorescein as the fluorescent probe. (Adapted from Zulueta et al. [23] with permission from Elsevier, see Note
35)
5. When the stability is achieved, quickly add (see Note 34)

150 μL of 2,2′azobis(2-amidinopropano) dihydrochloride
(AAPH) at 5.55 mmol L-1 concentration (Fig. 2) using a
multichannel pipette.
6. Read the fluorescence until it reaches a value of zero.
7. The control is the 75 mmol L-1 Na phosphate buffer at
pH 7.0, which must be used to reset the equipment (see Note
36).
8. After the analysis, a graph of fluorescence intensity vs. time is
obtained. The area under the fluorescence decay curve (AUC)
can be calculated using Eq. 2:
AUC = 1 + f 1 =f 0 + . . . + f i =f 0 + f 80 =f 0 (2)
where f0 is the fluorescence obtained at time 0 and fi is the fluores-
cence obtained at times between 0 and 80 min (see Note 37).
9. The ORAC value is calculated by Eq. 3:
μmol Trolox equivalents

ORAC
g
= [(A s - A b )=(A t - Ab )] kah (3)
where As is the AUC of the fluorescein in the sample, which was
calculated by an integrating program, At is the AUC of the Trolox,
Ab is the AUC of the control, k is the dilution factor, a is the
concentration of the Trolox expressed as mmol L-1, and h is the
ratio between the liters of extract and the grams of packaging used
for the extraction.
10. To establish the reproducibility of the analytical method, the
sample preparation must be repeated four times and the values
expressed as μmol Trolox equivalent per g of dry sample. At
least three independent tests must be performed for each
sample.
4 Notes
4.1 DPPH Assay 1. Because DPPH is a hydrophobic radical, the reactions must be
carried out in organic solvents [9], such as methanol, ethanol,
and acetone. The solvent depends on the material.
2. The preparation of the packaging extract was based on the
method described by Noronha et al. (2014) [24] and Kam
et al. (2018) [22].
3. A 0.06 mmol L-1 DPPH solution is equivalent to a 0.0024%
(w/v) DPPH solution.
4. Add the solvent used in the preparation of the packaging
sample extract.
5. The DPPH solution must be prepared daily and stored at 4 °C
in amber bottles until the assay. It is recommended to cover the
bottles with aluminum foil.
6. Based on the method described by Brand-Williams, Cuvelier,
and Berset (1995) [25], Piñeros-Hernandez et al. (2017) [26],
and Rodrı́guez et al. (2020) [27].
7. The tubes must also be wrapped with aluminum foil to avoid
any exposure to light. DPPH reactions are highly sensitive to
the reaction environment, i.e., water and solvent, pH, oxygen,
and light exposure.
8. The reduction degree of the DPPH radical during its reaction
with antioxidants is measured at 515–517 nm [28].
9. In the DPPH assay, the antioxidants react with the stable

DPPH free radical, leading to a discoloration of the molecule
from deep violet to light yellow. The degree of discoloration
indicates the antioxidant activity of the analyzed sample.
10. Before measuring each sample, clean the cuvette with distilled
water to avoid interference and inaccurate results.
11. The standard curve is used to quantitatively determine the
antioxidant activity of an unknown sample from a sample
with a known property.
12. A standard curve can also be built with gallic acid and
ascorbic acid.
13. It is recommended to plot the Trolox standard curve with at
least five points. To prepare the Trolox solutions, use the
procedure described in Subheading 3.2.4 with methanol as a
solvent instead of ethanol. The solvent must be the same as in
the preparation of the packaging extract.
14. Lower IC50 values indicate a product with higher antioxidant
activity.
4.2 ABTS Assay 15. Choosing the adequate extraction techniques (maceration,
infusion, vortex, supercritical fluid extraction, ultrasound-
assisted extraction, and others), the solvent type (hydrophilic
or lipophilic), and its proportion to the solids (solid/solvent
ratio) depends on the antioxidant compound present in the
active food packaging material [29, 30].
16. Several solvents can be used in the extraction of antioxidant
compounds. Polar solvents (hydrophilic), such as methanol,
ethanol, and water, are used for extracting phenolic acids,
flavonoids, and ascorbic acid, while nonpolar solvents (lipo-
philic), such as ether, or low-polarity solvents, namely chloro-
form and acetone, are used for extracting carotenes,
xanthophylls, alkaloids, terpenoids, and tocopherols
[30, 31]. The material of the active packaging will indicate
which disintegration procedure is the most appropriate. Sim-
pler procedures can be used for water-soluble packaging, such
as chitosan. In contrast, materials with more complex struc-
tures, such as alginate, require more elaborate procedures to
be dissolved (freezing, grinding, and methanol
extraction) [32].
17. In some cases, it is necessary to perform a more complex
extraction than soaking or immersing [11, 20]. Other tech-
niques may be included in order to enhance the extraction of
antioxidant compounds, such as stirring [33], heating [15],
shaking [19], centrifuging, and filtering with supernatant
removal [16, 19, 34].
Fig. 3 Reaction scheme of ABTS (2,2′-azinobis(3-ethylbenzothiazoline-6-sulphonic acid) and ABTS radical

cation (ABTS●+)
18. Depending on the antioxidant activity of the sample, more

dilutions of the sample extract may be required. The higher
the antioxidant potential of the sample is, the more efficient
the dilution must be. The absorbance of the samples must be
within the limits of the standard curve.
19. This step is fundamental to produce an ABTS radical cation
(ABTS●+), which is visually perceived due to the blue color of
the mixture. The principle of the ABTS method is based on
the antioxidant reaction with ABTS●+ generated by potas-
sium persulfate, followed by the absorbance diminution of
the blue solution at 734 nm [30]. In other words, the antiox-
idant compounds present in the active food packaging act by
eliminating this cationic radical ABTS●+ and converting it
into its colorless neutral form, which is measured by spectro-
photometry (Fig. 3) [35].
20. The ABTS solution must be prepared right before each assay.
Instead of ethanol [13, 15, 20, 21, 36], other solvents must
be used, such as water [11, 16] and methanol [19].
21. Before reading the samples and the standard, use the solvent
as a blank to reset the spectrophotometer.
22. The analytical response of this method is obtained by com-
paring the standard antioxidant with the ability of the sample
to scavenge the ABTS●+ radical in a dose-response curve. The
most common standard antioxidants are Trolox (water-
soluble vitamin E analog), gallic acid (3,4,5-
trihydroxybenzoic acid), and ascorbic acid [30, 35]. To deter-
mine the standard curve, follow the same procedures
performed with the samples using different concentrations of

the standard (in triplicate). It is important to highlight that,
depending on the antioxidant capacity of the sample, different
Trolox concentrations can be used to keep the sample absor-
bance within the limits of the standard curve. First, mix an
aliquot (60 μL) of each Trolox solution (0, 50, 200, 300,
400, 500, and 600 μmol L-1) with the ABTS●+ radical solu-
tion (2940 μL) for 10 min. Then, read the absorbance at
734 nm. Plot the Trolox concentrations (μmol L-1) on the
X axis and the respective absorbance on the Y axis and calcu-
late the line equation. From the line equation, calculate the
Trolox absorbance by Eq. 4:
y = ax + b (4)
where x is the concentration of Trolox and y is the corresponding
absorbance.
Once the standard curve is defined, it is possible to use the
absorbance of the analyzed samples to determine the antioxidant
capacity as μmol L-1 Trolox equivalent. Divide the result by the
mass of film in the sample to obtain the antioxidant capacity per
gram of the film sample. By using 1 g of film in the sample
preparation, it is possible to determine the mass by the dilution in
the assay.
23. The results of the antioxidant capacity may also be expressed
as percentages of inhibition using Eq. 5:
A ABTS - A sample
ABTS●+ scavenging effect (%) = × 100 (5)
A ABTS
where AABTS is the absorbance of the ABTS●+ solution and Asample
is the absorbance of the sample.
4.3 FRAP Assay 24. At this stage, the working solution (FRAP solution) is pro-
duced. When preparing this solution, you should initially mix
the acetate buffer with the TPTZ solution and then add the
FeCl3 solution. It is important to emphasize that this solution
must be prepared on the day of analysis and stored in an
amber bottle. The principle of the FRAP method is based
on the ability of antioxidants to reduce the ferric ion (FeIII)
complexed with TPTZ (ferric tripyridyl triazine complex) to
the ferrous ion (FeII) complex (ferrous tripyridyl triazine
complex) at low pH, in which it is possible to perceive the
blue color of the mixture and spectrophotometrically measure
at 593 nm (Fig. 4). Thus, the change in absorbance is directly
associated with the total reducing capacity of electron-
donating antioxidants present in the sample [37, 38].
Fig. 4 Reduction of the FeIII (ferric ion) complexed with TPTZ (2,4,6-tris(2-pyridyl)-1,3,5-triazine) to FeII
(ferrous ion) complexed with TPTZ. Adapted from Rufino et al. [39] with permission from Embrapa
25. First, use the FRAP solution to calibrate the spectrophotom-

eter and then measure the FeII-TPTZ complex (colored prod-
uct) in the samples.
26. Several standard antioxidants can be used, such as gallic acid
(3,4,5-trihydroxybenzoic acid) [20], ferrous sulfate solutions
(FeSO4·7H2O) [21, 40], and Trolox (water-soluble vitamin
E analog), which is the most common [12, 19, 22, 33].
27. Prepare the blank sample using distilled water in the FRAP
solution instead of the FeCl3·6H2O solution.
4.4 ORAC Assay 28. Based on the method described by Cao et al. (1993) [41],
Ninfali et al. (2005) [42], and Madhujith and Shahidi
(2007) [43].
29. All the solutions must be prepared right before analysis and all
working standard solutions must rigorously dilute the respec-
tive stock solutions in the phosphate buffer (75 mmol L-1,
pH 7.0).
30. In most techniques, phosphate buffer is used to dilute all
extracts since fluorescein is more stable at pH 7.0 [43].
31. Prepare the solution on the day of analysis. In the case of
contact with eyes or skin, fluorescein sodium salt causes severe
eye injuries and skin sensitization. Therefore, wear suitable
gloves and safety glasses with side shields.
32. 37 °C is the ideal temperature to produce peroxyl radicals that

are responsible for oxidizing fluorescein during the test to
produce decay curves [44].
33. The fluorescence determination must be performed using a
fluorescence spectrophotometer.
34. Due to the thermal sensitivity of AAPH, the working solution
must be prepared right before analysis.
35. In this assay, the peroxyl radical reacts with a fluorescent
compound (fluorescein, oxidizable substrate) to form a non-
fluorescent product. The antioxidant when donating hydro-
gen atoms inhibits the loss of fluorescence intensity, in which
the inhibition is proportional to the antioxidant activity.
36. The standard stock solution of the Trolox antioxidant can be
prepared weekly and serial dilutions with stock buffer must be
performed daily to reach the range indicated in the prepara-
tion of the calibration graphs. The standard curve is generated
with five Trolox concentrations, and the Trolox equivalent of
the sample is calculated by the linear (Y = a + bX) or quadratic
(Y = a + bX + cX2) relationships between the Trolox (Y)
concentration (μM) and the liquid area under the fluorescein
X decay curve (AUCsample – AUCwhite) [45].
37. The antioxidant protective effect is determined by calculating
the area formed under the fluorescence decay curve of the
sample vs. time when compared to the blank, where no anti-
oxidants are presented. The area under the curve is calculated
for both the standard and the sample. The AUC is also calcu-
lated for the blank. The control is obtained only from fluores-
cein to check the maintenance of its fluorescence over time.
Acknowledgments
The authors would like to thank the Brazilian research funding

agencies CAPES (Coordination for the Improvement of Higher
Education Personnel) and CNPq (National Council for Scientific
and Technological Development) for providing financial support
for this study.
References
1. Han JW, Ruiz-Garcia L, Qian JP, Yang XT 3. Gómez-Estaca J, López-de-Dicastillo C, Her-

(2018) Food packaging: a comprehensive nández-Muñoz P, Catalá R, Gavara R (2014)
review and future trends. Compr Rev Food Advances in antioxidant active food packaging.
Sci Food Saf 17:860–877 Trends Food Sci Technol 35:42–51
2. Pelissari FM, Sartori T, Santos F, Molina G, 4. Granato D, Shahidi F, Wrolstad R, Kilmartin P,
Menegalli FC (2016) Active Melton LD, Hidalgo FJ, Miyashita K, van
bio-packaging. In: Inamuddin (ed) Green Camp J, Alasalvar C, Ismail AB, Elmore S,
polymer composites technology: properties Birch GG, Charalampopoulos D, Astley SB,
and applications. CRC Press Pegg R, Zhou P, Finglas P (2018) Antioxidant
activity, total phenolics and flavonoids con- antioxidant activity and physical properties of
tents: should we ban in vitro screening meth- films. Food Hydrocoll 40:145–152
ods? Food Chem 264:471–475 16. Go EJ, Song KB (2020) Development and
5. Licciardello F, Wittenauer J, Saengerlaub S, characterization of Citrus junos pomace pectin
Reinelt M, Stramm C (2015) Rapid assessment films incorporated with rambutan (Nephelium
of the effectiveness of antioxidant active lappaceum) peel extract. Coatings 10(8):714
packaging-study with grape pomace and olive 17. Yang HJ, Lee JH, Won M, Song KB (2016)
leaf extracts. Food Packag Shelf Life 6:1–6 Antioxidant activities of distiller dried grains
6. Nerı́n C (2010) Antioxidant active food pack- with solubles as protein films containing tea
aging and antioxidant edible films. In: Oxida- extracts and their application in the packaging
tion in foods and beverages and antioxidant of pork meat. Food Chem 196:174–179
applications. Woodhead, pp 496–515 18. Eça KS, Machado MTC, Hubinger MD,
7. Bentayeb K, Vera P, Rubio C, Nerin C (2009) Menegalli FC (2015) Development of active
Adaptation of the ORAC assay to the common films from pectin and fruit extracts: light pro-
laboratory equipment and subsequent applica- tection, antioxidant capacity, and compounds
tion to antioxidant plastic films. Anal Bioanal stability. J Food Sci 80(11):C2389–C2396
Chem 394:903–910 19. Tongnuanchan P, Benjakul S, Prodpran T
8. Eça KS, Sartori T, Menegalli FC (2014) Films (2012) Properties and antioxidant activity of
and edible coatings containing antioxidants – a fish skin gelatin film incorporated with citrus
review. Braz J Food Technol 17:98–112 essential oils. Food Chem 134:1571–1579
9. Schaich KM, Tian X, Xie J (2015) Reprint of 20. Rambabu K, Bharath G, Banat F, Show PL,
“Hurdles and pitfalls in measuring antioxidant Cocoletzi HH (2019) Mango leaf extract
efficacy: a critical evaluation of ABTS, DPPH, incorporated chitosan antioxidant film for
and ORAC assays”. J Funct Foods 18:782–796 active food packaging. Int J Biol Macromol
10. Sanches-Silva A, Costa D, Albuquerque TG, 126:1234–1243
Buonocore GG, Ramos F, Castilho MC, 21. Oliveira Filho JG, Rodrigues JM, Valadares
Machado AV, Costa HS (2014) Trends in the ACF, Almeida AB, Lima TM, Takeuchi KP,
use of natural antioxidants in active food pack- Alves CCF, Sousa HAF, Silva ER, Dyszy FH,
aging: a review. Food Addit Contam Part A Egea MB (2019) Active food packaging: algi-
Chem Anal Control Expo Risk Assess 31: nate films with cottonseed protein hydroly-
374–395 sates. Food Hydrocoll 92:267–275
11. Bonilla J, Sobral PJA (2016) Investigation of 22. Kam WYJ, Mirhosseini H, Abas F, Hussain N,
the physicochemical, antimicrobial and antiox- Hedayatnia S, Chong HLF (2018) Antioxidant
idant properties of gelatin-chitosan edible film activity enhancement of biodegradable film as
mixed with plant ethanolic extracts. Food active packaging utilizing crude extract from
Biosci 16:17–25 durian leaf waste. Food Control 90:66–72
12. Ruiz-Navajas Y, Viuda-Martos M, Sendra E, 23. Zulueta A, Esteve MJ, Frı́gola A (2009) ORAC
Perez-Alvarez JA, Fernández-López J (2013) and TEAC assays comparison to measure the
In vitro antibacterial and antioxidant properties antioxidant capacity of food products. Food
of chitosan edible films incorporated with Thy- Chem 14:310–316
mus moroderi or Thymus piperella essential 24. Noronha CM, De Carvalho SM, Lino RC,
oils. Food Control 30:386–392 Barreto PLM (2014) Characterization of anti-
13. Re R, Pellegrini N, Proteggente A, Pannala A, oxidant methylcellulose film incorporated with
Yang M, Rice-Evans C (1999) Antioxidant α-tocopherol nanocapsules. Food Chem 159:
activity applying an improved ABTS radical 529–535
cation decolorization assay. Free Radic Biol 25. Brand-Williams W, Cuvelier ME, Berset C
Med 26(9–10):1231–1237 (1995) Use of a free-radical method to evaluate
14. Bondini RB, Pugine SMP, Melo MP, Carvalho antioxidant activity. LWT Food Sci Technol 28:
RA (2020) Antioxidant and anti-inflammatory 25–30
properties of orally disintegrating films based 26. Piñeros-Hernandez D, Medina-Jaramillo C,
on starch and hydroxypropyl methylcellulose López-Córdoba A, Goyanes S (2017) Edible
incorporated with Cordia verbenacea (Erva cassava starch films carrying rosemary antioxi-
baleeira) extract. Int J Biol Macromol 159: dant extracts for potential use as active food
714–724 packaging. Food Hydrocoll 63:488–495
15. Bitencourt CM, Fávaro-Trindade CS, Sobral 27. Rodrı́guez GM, Sibaja JC, Espitia PJP, Otoni
PJA, Carvalho RA (2014) Gelatin-based films CG (2020) Antioxidant active packaging based
additivated with curcuma ethanol extract: on papaya edible films incorporated with
Moringa oleifera and ascorbic acid for food waste for active packaging: multicomponent
preservation. Food Hydrocoll 103:105630 polyhydroxyalkanoate coated nanopapers with
28. Olszowy M, Dawidowicz AL (2018) Is it pos- improved hydrophobicity and antioxidant
sible to use the DPPH and ABTS methods for capacity. Food Packag Shelf Life 21:100356
reliable estimation of antioxidant power of col- 37. Frankel EN, Meyer AS (2000) The problems of
ored compounds? Chem Pap 72:393–400 using one-dimensional methods to evaluate
29. Khan MK, Paniwnyk L, Hassan S (2019) Poly- multifunctional food and biological antioxi-
phenols as natural antioxidants: sources, dants. J Sci Food Agric 80:1925–1941
extraction and applications in food, cosmetics 38. Benzie I, Strain J (1996) The ferric reducing
and drugs. In: Plant based “Green chemistry ability of plasma (FRAP) as a measure of “anti-
2.0”, pp 197–235 oxidant power”: the FRAP assay. Anal Biochem
30. Pisoschi AM, Pop A, Cimpeanu C, Predoi G 239(1):70–76
(2016) Antioxidant capacity determination in 39. Rufino MSM, Alves RE, Brito ES, Morais SM,
plants and plant-derived products: a review. Sampaio CG, Pérez-Jiménez J, Saura-Calixto
Oxidative Med Cell Longev 2016:1-36 FG (2006) Metodologia cientı́fica: Determina-
31. Silva BV, Barreira JCM, Oliveira MBPP (2016) ção da atividade antioxidante total em frutas
Natural phytochemicals and probiotics as bio- pelo método de redução do ferro (FRAP)
active ingredients for functional foods: extrac- Comunicado técnico online EMBRAPA
tion, biochemistry and protected-delivery Agroindústria Tropical 125
technologies. Trends Food Sci Technol 50: 40. Crizel TM, Rios AO, Alves VD, Bandarra N,
144–158 Moldão-Martins M, Flôres SH (2018) Biode-
32. Bonilla J, Atarés L, Vargas M, Chiralt A (2012) gradable films based on gelatin and papaya peel
Edible films and coatings to prevent the detri- microparticles with antioxidant properties.
mental effect of oxygen on food quality: possi- Food Bioprocess Technol 11:536–550
bilities and limitations. J Food Eng 110(2): 41. Cao G, Alessio HM, Cutler RG (1993)
208–213 Oxygen-radical absorbance capacity assay for
33. Sganzerla WG, Rosa GB, Ferreira ALA, Rosa antioxidants. Free Radic Biol Med 14:303–311
CG, Beling PC, Xavier LO, Hansen CM, Fer- 42. Ninfali P, Gloria M, Giorgini S, Rocchi M,
rareze JP, Nunes MR, Barreto PLM, Lima Bacchiocca M (2005) Antioxidant capacity of
Veeck AP (2020) Bioactive food packaging vegetables, spices and dressings relevant to
based on starch, citric pectin and functiona- nutrition. Br J Nutr 93:257–266
lized with Acca sellowiana waste by-product: 43. Madhujith T, Shahidi F (2007) Antioxidative
characterization and application in the posthar- and antiproliferative properties of selected bar-
vest conservation of apple. Int J Biol Macromol ley (Hordeum vulgarae L.) cultivars and their
147:295–303 potential for inhibition of low-density lipopro-
34. Franco GT, Otoni CG, Lodi BD, Lorevice MV, tein (LDL) cholesterol oxidation. J Agric Food
de Moura MR, Mattoso LHC (2020) Escalat- Chem 55:5018–5024
ing the technical bounds for the production of 44. Stockham K, Paimin R, Orbell JD, Adorno P,
cellulose-aided peach leathers: from the bench- Buddhadasa S (2011) Modes of handling Oxy-
top to the pilot plant. Carbohydr gen Radical Absorbance Capacity (ORAC) data
Polym:116437 and reporting values in product labelling. J
35. Arnao MB, Cano A, Acosta M (2001) The Food Compos Anal 24:686–691
hydrophilic and lipophilic contribution to 45. Prior RL, Wu X, Schaich K (2005) Standar-
total antioxidant activity. Food Chem 73:239– dized methods for the determination of anti-
244 oxidant capacity and phenolics in foods and
36. Urbina L, Eceiza A, Gabilondo N, Corcuera dietary supplements. J Agric Food Chem 53:
MÁ, Retegi A (2019) Valorization of apple 4290–4302
Chapter 18
Release of Active Agents from Food Packaging Materials

Murilo S. Pacheco, Mariana A. de Moraes, Mariana A. da Silva,
and Andréa C. K. Bierhalz
Abstract
Active packaging (AP) is an innovative type of food packaging from which active compounds, such as
antimicrobials or antioxidants, may be released in order to enhance food quality and safety and extend its
shelf-life. The efficiency of the AP system depends on the release rate, since too high rates may lead to a
premature loss of the AP activity, while too low rates may not be sufficient to prevent food deterioration. In
this context, the study of the active agent release mechanisms and kinetics is pivotal to determining the
performance of AP systems. Here we describe a general method to evaluate the release of an active
compound from a (bio)polymeric film, including the protocol for the release test, the quantification
methods, and the mathematical models used to estimate diffusivity and elucidate the release mechanism.
As a showcase, we focus on the release of the antimicrobial natamycin entrapped in a biopolymeric matrix to
a food simulant liquid medium (ethanol solution), and we provide suggestions and protocols that can be
extended to other biopolymers and analytes.
Key words Polymeric film, Active compound, Mathematical modeling, Fickian diffusion, Swelling,
Release mechanism
1 Introduction
Active packaging (AP) has been defined, according to the European

FAIR-Project CT 98–4170, as “a type of packaging that changes
the condition of the packaging to extend shelf-life or improve safety
or sensory properties while maintaining the quality of the food”
[1]. AP can act either as absorbing or releasing systems, depending
on their mechanism of action. The former absorbs substances from
the food or the packaging headspace, such as ethylene and oxygen
scavengers. The latter, on the other hand, allow active compounds
to diffuse from the AP to the food surface and includes the release
of antimicrobials, antioxidants, CO2 emitters, flavors, aromas,
enzymes, probiotic bacteria, nutraceuticals, and plant growth
hormones [2].
https://doi.org/10.1007/978-1-0716-3613-8_18,
311
312 Murilo S. Pacheco et al.
The active agent can be applied by several methods to polymer-

based AP systems, namely, (i) physical adsorption onto the surface
of the polymeric material; (ii) immobilization onto polymers
by ionic or covalent bonds; (iii) addition into sachets, pouches, or
pads, which are placed inside the packaging; and (iv) incorporation
within the bulk of the polymer matrix [2]. The latter can be
achieved either by directly adding or encapsulating the active
agent during polymer processing [3], by its adsorption (e.g., by
immersion), or by supercritical impregnation through the use of
solutions or suspensions containing the active agent [4, 5].
The diffusion of the active agent may be achieved by direct
contact between food and packaging material or through gas phase
diffusion (volatile systems) from packaging to the food surface,
which depends on the AP design and the nature of the active
compound [6, 7].
Antimicrobial packaging is a promising and advantageous form
of AP because microbial contamination problems are known to
occur mostly on food surfaces. Thus, the AP system would allow
the incorporated antimicrobial to be continuously delivered to the
food surface, maintaining an effective concentration of the active
agent where it is most needed [7]. Besides, multifunctional AP
materials have also been investigated, combining more than one
active agent or even actives that act both as antioxidant and antimi-
crobial, which is the case of some essential oils [8, 9].
The main mechanisms that describe the release of an active
agent from an AP system are diffusion, swelling, and/or disintegra-
tion. In exclusively diffusive-driven systems, mainly synthetic and
water-resistant polymers, the active agent diffuses through the
polymeric matrix and from the film surface into the food. The
swelling takes place when the polymer matrix is put in contact
with compatible solvents, usually hydrophilic matrices in water or
hydrophobic matrices in organic media. In the swollen state, the
relaxation of the polymer chains increases the diffusion coefficient
and enables the transport of the active molecules to the food. The
disintegration mechanism occurs when there is hydrolysis, cleavage,
or erosion of the matrix according to the medium conditions,
which is typically induced by the absorption of fluids [2].
Matching the release rate with the kinetics of food deteriora-
tion is crucial for food-targeted AP systems. If the migration is too
fast, the active agent will be depleted before the expected storage
period, and the AP system will lose its functionality earlier than
needed. On the other hand, if the release rate is too low, the
released amount of active compound can be insufficient to prevent
food deterioration [9, 10]. The release rate will be dependent on
the permeability, which in turn depends on the characteristics of the
polymer (type, concentration, glass transition temperature, and
molecular weight), fabrication parameters, and chemical interac-
tions among the polymer, active agent, and food. Several strategies
Release of Actives from Food Packaging 313
can tune the film properties and develop ideal controlled release
systems, including crosslinking, UV irradiation, encapsulation,
complexation, multilayer systems, and nanocomposite films,
among others [2].
1.1 Methods to Considering the release of an active compound from a polymeric

Evaluate the Release film to the food, there are three steps involved in the process:
of Active Agents from (i) molecular diffusion through the film, (ii) mass transfer across
Polymeric Films the film/food interface, and (iii) dispersion into food or desorption
into packaging headspace. Typically, the diffusion of the active
compound through the matrix is assumed as the slowest and rate-
controlling step, in which the concentration difference is the
driving force of the mass transfer phenomenon [9].
The diffusion of the active agent and its solubility in the matrix
are the two main parameters that govern the release rate in the
specific matrix. Diffusivity, which indicates how fast the active
compound moves within the film, and partition coefficient, which
indicates the affinity of the active agent for the two phases (film and
food)—and therefore suggests how much active agent is released to
the food at equilibrium—are the two model parameters used to
describe the release behavior [9].
Empirical and semiempirical mathematical models can be fitted
to kinetic data in order to elucidate the active agent release mechan-
isms. The most commonly applied model is the so-called Power
Law Model (Eq. 1) [11].
Mt
= kt n ð1Þ
M1
where t is time, k is a constant related to the structural and geomet-
ric characteristics of the matrix, and n is the release exponent.
From the coefficient n in Eq. 1, it is possible to determine the
release mechanism. For thin films, n = 0.5 indicates that the mass
release is proportional to the square root of time and controlled by
diffusion. This mechanism is known as Fickian transport. On the
other hand, when n = 1.0, the release is controlled by the swelling
or erosion of the polymeric matrix, which is known as case-II
transport. When 0.5 < n < 1.0, the diffusion and swelling mechan-
isms overlap, and the transport is considered anomalous [12].
When the mechanism is predominantly diffusive, the transient
mass transfer process of the active compound through the film
can be described by Fick’s second law (Eq. 2), assuming
one-dimensional diffusion from an infinite flat plate [13, 14]. The
solutions of the diffusional equation can be fitted to the experimen-
tal kinetic release data, enabling the prediction of the diffusivity of
the active compound through the polymeric film.
2
∂C ∂ C
=D ð2Þ
∂t ∂x 2
Several models based on solutions of the Fickian differential

equation with appropriate initial and boundary conditions are used
to predict the release behavior, obtain the concentration profile,
and estimate the diffusivity. However, many assumptions are
required, as follows: (i) the diffusion of the active through the
film is the rate-controlling step; (ii) no structural change occurs in
the film during the release; (iii) the active compound can be readily
desorbed from the film; (iv) the active compound is initially homo-
geneously distributed in the film; (v) the initial concentration of the
active in the food is zero; (vi) no concentration gradient of the
active exists in the food; (vii) the partition coefficient and diffusivity
are constant at a given temperature; (viii) interactions between food
simulant and film are negligible; and (ix) no degradation of the
active compound occurs [9]. Based on these assumptions, the
integrated solution of Eq. 2 in terms of released mass is given by
Eq. 3.
Mt 1 2αð1 þ aÞ - Dq 2n t
=1- exp ð3Þ
M1 n = 1 1 þ α þ α2 q 2
n δ2
where Mt is the mass of active compound that migrated to the food
simulant at time t, M1 is the mass of active compound in the food
simulant at equilibrium, D is the diffusivity of the active compound
in the film, δ is the film thickness, qn is the nonzero positive roots of
Eq. 4, and α is given by Eq. 5 [13, 14].
tan q n = - αq n ð4Þ
VF
α= ð5Þ
KPV P
where VF is the volume of food simulant, VP is the volume of the
film (AP), and KP is the partition coefficient given by Eq. 6.
C P,1
Kp = ð6Þ
C F ,1
where CP,1 and CF,1 are the concentrations of the active com-
pound, respectively, in the film (AP) and the food simulant, both at
equilibrium.
Typically, the volume of the food simulant is larger than the
volume of the AP (VF >> VP), which makes α >> 1. In this case, a
simplified solution is obtained (Eq. 7) [13].
Mt 8 1 1 ð2n þ 1Þ2 π 2
=1- 2 exp - Dt ð7Þ
M1 π n=0
ð2n þ 1Þ2 δ2
For short times, when Mt/M1 ≤ 0.6, Eq. 7 assumes boundary
conditions of a semi-infinite solid model, being simplified to Eq. 8
for a flat plate [15].
Mt Dt
=4 ð8Þ
M1 πδ2
Liquid or food simulant may cause swelling of the (bio)-
polymeric matrix, and in this case, the release kinetics may not
follow the Fickian diffusion model [9]. When case-II or anomalous
transport governs the release, models that include the influence of
matrix swelling are more appropriate. Equation 9 allows the pre-
diction of diffusivity considering both the Fickian and swelling
contributions, as well as the relaxation time associated with polymer
relaxation (τ) and the deviation from the ideal Fickian behavior
(XF) [16]. XF ranges from 0 to 1; for XF equal to 1, Eq. 9 is the
solution of the Fick’s second law (Eq. 7), whereas for XF equal to
0, anomalous diffusion is obtained.
Mt 8 1 1 ð2n þ 1Þ2 π 2
= XF: 1 - 2 exp - Dt
M1 π n=0
ð2n þ 1Þ 2
δ2
t
þð1 - X F Þ: 1 - exp -
τ
ð9Þ
Usually, the methods used to study the release of an active
compound from a polymeric film involve the immersion of the
film in a release (sink) medium, and periodic monitoring of the
cumulative concentration of the compound in the liquid medium.
The fractional mass released is calculated, and the mathematical
models are fitted to the kinetic data. Several food simulants with
different pH and water activity are used as sink media for release
studies allowing standardization, reproducibility, and ease of analy-
sis. These media can predict the active agent diffusion from the AP
to the real food. The Food and Drug Administration (FDA) recom-
mends the use of 10 vol% ethanol solution in water as a simulant of
aqueous and acidic foods; 10 and 50 vol% ethanol solution in water
as simulants of low- and high-alcohol content foods, respectively;
and food oil (e.g., corn or olive oil), HB307 (mixture of synthetic
triglycerides), Miglyol 812 (derived from coconut oil), and others
(such as ethanol solutions—used for specific polymeric matrices) as
simulants of fatty foods [17]. On the other hand, the European
legislation recommends the use of water for aqueous foods with a
pH above 4.5; 3 vol% acetic acid solution in water for acidic
aqueous foods with a pH below 4.5; 10 vol% ethanol solution in
water for alcoholic products; and olive oil for fatty foods
[18]. Other food simulants are reported in the literature, such as
heptane, isooctane, ether, isopropanol [19], cyclohexane, ethyl
acetate, glyceryl tripelargonate, terpenes, tributyrin [13], and sun-
flower oil [20].
When the release test is designed for a solid medium, mathe-
matical modeling becomes much more complex, as the resistance to
mass convection plays an important role in the release and a finite
food volume must be considered. Many studies involving migration

in a food packaging system are related to contaminants and syn-
thetic packaging components. In this approach, it is also necessary
to use solid food simulants, such as agar or gelatin gels, cheeses, or
Tenax®—a dry food simulant recommended by the European Reg-
ulation 10/2011 [21–23].
Thus, based on the authors’ expertise [5, 24–26], the protocols
to evaluate the release of an active compound (natamycin) from an
AP (crosslinked biopolymeric film) to a liquid medium (ethanol
solution as a food simulant) are described in this chapter. This
approach may be extrapolated to other cases, involving different
analytes and release media, according to the proposed application,
and following the required assumptions for suitably fitting the
mathematical models.
2 Materials
2.1 Reagents and • Absolute ethanol.

Film Sample • Ultrapure water.
• Analyte (natamycin) solution to prepare a standard curve.
• Active loaded crosslinked polymeric film sample (4 cm × 4 cm).
2.2 Equipment • Digital micrometer (0–25 mm, 0.001 mm).

• Analytical balance (0.0001 g).
• Glassware to prepare solutions.
• Orbital shaker water bath.
• UV-Vis spectrophotometer.
• Quartz 1-mL cuvettes.
• Covered or screw-capped containers.
• Tweezers.
3 Methods
The method described herein is based on the release into a liquid

medium, with complete renewal of the release medium. The sam-
ples are transferred through recipients, ensuring no resistance to
mass transfer at the film surface (i.e., the active agent concentration
at the film surface is zero). Other methods can be used, such as
collecting aliquots from the release medium during the test, with
partial renewal of the medium or return of the aliquot after quanti-
fication (see Note 1).
3.1 Food Simulant Prepare an ethanol solution at a predetermined concentration, e.g.,

10 vol% (see Notes 2 and 3), adding 100 mL of absolute ethanol in
900 mL of water. Cover the solution until use. The total amount of
solution needed for the test depends on the time intervals selected
to collect the experimental points (see Note 4).
3.2 Sample Prepare the sample of the actively loaded biopolymeric crosslinked
Preparation film with a determined dimension, e.g., 4 cm × 4 cm, and with a
known average thickness (see Note 5). The mass of the active
compound (natamycin) loaded in the film sample must be known
(see Note 6).
3.3 Equipment Setup Prepare a shaker bath with distilled water at a determined tempera-
ture for the experiment, e.g., 25 °C (see Note 7). Prepare a spec-
trophotometer UV-Vis to measure the absorbance of the samples
collected during the release test.
3.4 Quantification Prepare the standard curve of the analyte used. Prepare a solution of
Method the natamycin (analytical standard) at a determined concentration,
e.g., 40 mg L-1, using the food simulant 10 vol% ethanol as
solvent, and prepare several dilutions to obtain a range of different
concentrations. Using a UV-Vis spectrophotometer at 317 nm (see
Note 8), a quartz bucket, and the ethanolic solution as a baseline,
measure the absorbance of the different concentrations of natamy-
cin. If necessary, dilute the natamycin solutions so that the absor-
bance values are between 0 and 1. Plot the absorbance values versus
concentrations and perform a linear regression, R2 > 0.99, to
obtain the analytical standard curve equation.
3.5 Release Test 1. Place 25 mL of the ethanolic solution in covered or screw-

capped containers.
2. Place the containers with the solutions in the shaker bath and
set the appropriate rotation speed to the shaker, e.g., 150 rpm
(see Notes 9 and 10).
3. At time zero, immerse the film sample in the first container and
start time counting.
4. After a predetermined time interval, quickly transfer, using
tweezers, the film sample from the first to the second container
(Fig. 1).
5. Remove the first container from the shaker bath and store it
for later quantification of the released active compound (see
Note 11).
6. Repeat steps 4 and 5 at predetermined time intervals, quickly
transferring the sample to the next container, and storing the
solution from the previous one; prepare all solutions using
ultrapure water and analytical grade reagents. Store all reagents
Polymeric film containing the active compound
Food simulant
Shaker bath
Active compound released
Fig. 1 Schematic representation of the release test: The polymeric active film is placed in the first container
with the food simulant and at determined time intervals it is quickly transferred to the next container. The food
simulant containing the active compound released is stored to further quantification
at room temperature unless indicated otherwise. Diligently

follow all waste disposal regulations when disposing waste
materials.
7. After removing the previous containers, replace the spaces of
the shaker bath with new containers with blank solutions to
ensure temperature preconditioning.
8. The time intervals can be determined according to the total
duration of the test until no further release can be detected,
considering a fast-release behavior in the beginning of the test
and a slow-release behavior at the end of the test. The sample
can be transferred at predetermined test times, e.g., 5 min,
10 min, 15 min, 20 min, 30 min, 45 min, 1 h, 2 h, 3 h, 4 h,
5 h, 10 h, 24 h, 48 h, 72 h, 96 h, 120 h, 200 h, 300 h, 400 h,
500 h, 600 h, 700 h, 800 h, 900 h, and 1000 h (see Note 12).
9. After the last time interval, collect the film sample and store in
an appropriate place.
10. Turn off the shaker bath and remove all distilled water, accord-
ing to the manufacturer’s recommendations.
3.6 Quantification Considering natamycin as the active compound, UV-Vis spectro-

photometry would be indicated for quantification. Other quantifi-
cation methods can be used for other analytes (see Note 13).
1. Read the absorbance at 317 nm of the 10 vol% ethanol solution
used as a food simulant on the UV-Vis spectrophotometer and
set as blank.
2. Measure the absorbances of all solutions stored during the
release test, rinsing the bucket after each reading, and measur-
ing from the lowest to the highest concentration.
3. Using the equation obtained by the standard curve, convert all

the absorbance data to natamycin concentrations.
4. Use the volume of each container (25 mL) to convert all the
concentrations to mass values of natamycin released at
each step.
3.7 Mathematical 1. Sum all the mass values of natamycin released during the test,
Modeling obtaining the total mass of natamycin released from the film to
the food simulant at equilibrium (M1).
2. Sum the mass released at each previous point from the begin-
ning of the test to the time interval being calculated into the
cumulative mass release (Mt) at each time.
3. Divide each cumulative mass (Mt) by the total mass released
(M1) to calculate the cumulative mass fraction released for
each time (Mt/M1).
4. Plot the Mt/M1 values as a function of time t to obtain the
natamycin release profile (exemplified in Fig. 2).
5. Fit the Power Law Model (Eq. 1) to the release kinetic data
when Mt/M1 ≤ 0.6 to elucidate the rate-controlling mecha-
nism of release. Plot ln(Mt/M1) as a function of ln(t), consid-
ering the linearization form of the equation, and obtain a linear
regression (Fig. 3). The diffusional exponent n is obtained
from the angular coefficient (slope), while the constant k is
calculated from the linear coefficient (intercept).
1.2
Fractional natamycin released (M1/Mv)
1.0
0.8
0.6
0.4
0.2
0.0
0 500 1000 1500 2000 2500 3000
Time (h)
Fig. 2 Example of a release profile of natamycin from crosslinked alginate films

to an ethanol solution
ln (t)
6 8 10 12 14 16
0.0
-0.5
-1.0
-1.5
ln (M1/Mv)
-2.0
-2.5
-3.0 Linear fitting of the

experimental data
-3.5
Experimental data
-4.0
Fig. 3 Example of the fitting of linearized Power Law Model to the kinetic data of
natamycin released from crosslinked alginate films to ethanol solution
6. If the diffusional exponent n is close to 0.5, Fickian diffusion is

the rate-controlling mechanism of release. If n is close to 1.0,
swelling or erosion of the matrix is the rate-controlling mecha-
nism. If n falls between 0.5 and 1.0, there is an anomalous
transport, in which both Fickian diffusion and swelling are
important mechanisms that drive the active release from
the film.
7. Fit the series solution of the Fick’s second law (Eq. 7) to the
release of kinetic data (Fig. 4) and use nonlinear regression to
obtain the model parameters, including diffusivity (D) (see
Notes 14 and 15).
8. Fit the semi-infinite solid or short times model (Eq. 8) to the
release kinetic data when Mt/M1 ≤ 0.6 to determine the
diffusivity (D) at the beginning of the release test. Plot Mt/
M1 data as a function of t0.5 [s0.5] and obtain a linear fit. From
the angular coefficient, according to the equation, calculate
diffusivity (Fig. 5).
9. Compare the diffusivity found by the two models. The good
fitting of the short times model and the proximity of the
diffusivity value found by the short times model with that
found by the series solution model may indicate that swelling
has a low influence on diffusivity.
10. If the release models show a great influence on swelling, fit the
model described in Eq. 9 to determine diffusivity and evaluate
the deviation from the ideal Fickian behavior (XF).
11. Evaluate the fitted models using fitting coefficients, e.g., R2
(see Note 16).
1.2

1.0
0.8
0.6
0.4
Fick’s 2nd Law
series solution
0.2
Experimental data
0.0
0 500 1000 1500 2000 2500 3000
Time (h)
Fig. 4 Example of a release profile of natamycin from crosslinked alginate films

to ethanol solution with fitting of Fick’s second Law series solution
0.7
0.6
0.5
0.4
0.3
0.2
Linear fitting of the
experimental data
0.1
Experimental data
0.0
0 5 10 15 20 25 30
t0.5 (s0.5)
Fig. 5 Example of the fitting of linearized short times model to the kinetic data of
natamycin released from crosslinked alginate films to ethanol solution
4 Notes
1. Several methods can be used to carry out the release test to

liquid medium. The complete renewal of the release medium
ensures no resistance to mass transfer at the film surface and
maintains a low concentration of the active in the medium,
avoiding the need for the partition coefficient determination,
which is sometimes difficult to obtain. When using the method
of collecting aliquots during the test with partial renewal of the

medium, the concentration gradient is still favored, but the
partition coefficient will be of greater importance, as well as
the miscibility of the analyte in the release medium. In this case,
it is necessary to consider the changes in the concentration at
each point by the added volume to measure the released mass at
each time interval. On the other hand, when the aliquot is
returned after quantification, the test gets more proximity
with the reality, although it may be necessary to obtain the
partition coefficient due to the increase of the active compound
concentration, once it could play an important role in the
release behavior.
2. Prepare all solutions using ultrapure water and analytical-grade
reagents. Store all reagents at room temperature unless indi-
cated otherwise. Diligently follow all waste disposal regulations
when disposing waste materials.
3. Prepare the solution of food simulant according to the concen-
tration recommended by the FDA or other regulatory agen-
cies, as well as to the type of food that is being simulated, e.g.,
acidic, alcoholic, fatty.
4. Prepare a sufficient volume of food simulant for the complete
release test, e.g., if 30 experimental points are expected to be
collected to construct the release profile, with complete
renewal of the release medium, prepare at least 30 × 25 mL of
solution = 750 mL.
5. The thickness of the film sample will be necessary to fit the data
to the mathematical models. Measure the film thickness at least
at five points along the film sample using a micrometer and use
the mean value. Measure it before and after the release test to
evaluate whether the swelling behavior plays an important role
in the release kinetics.
6. It is important to quantify the initial amount of the active
compound incorporated into the film sample used in the
release test in order to determine the % of active released.
7. FDA recommends that the release test must be carried out
under the most severe conditions of temperature and time for
the proposed use. It is recommended a test temperature of
40 °C for 10 days for room-temperature applications and 20 °
C for refrigerated applications [17].
8. Use the absorption peak related to the quantification of the
specific analyte used in the food simulant. It is recommended to
perform a wavelength scan to determine the correct peak of the
analyte dissolved in the food simulant solution.
9. The continuous stirring during all the release tests decreases
mass convection resistance, which is necessary for the applica-
tion of the mathematical models.
10. Partition coefficient can be obtained using Eq. 6 with M1 and

the amount of active compound remaining in the film at
equilibrium.
11. During storage, all containers must be kept closed to avoid sink
medium evaporation and consequent changes in solute
concentration.
12. Frequently, there is a faster release in the first instants of the
test, being necessary to collect experimental points at shorter
time intervals, in order to allow proper release profiles and a
good fitting of the mathematical models.
13. Besides UV-Vis spectrophotometry, several methods can be
used to quantify natamycin or other active compounds, such
as high-performance liquid (HPLC) and gas chromatography.
For example, nisin can be quantified through reverse-phase
HPLC [27], spectrofluorimetry [28], and colorimetry [22]
when released from films.
14. Typically, the diffusivity is given in cm2 s-1. Use time in [s] and
thickness in [cm] for model fitting to obtain diffusivity.
15. In the case of performing the release test using the release from
only one side of the polymeric film, it is important to correctly
consider the boundary conditions to obtain the correct form of
solution of the diffusional equation.
16. Other fitting coefficients can be used to analyze the fitting
accuracy and compare the models used properly, such as root
mean square error [29].
Acknowledgments
The authors thank the São Paulo Research Foundation (FAPESP,

grants #2017/20274-9 and #2018/25656-0) and the National
Council for Scientific and Technological Development (CNPq,
grants #473972/2011-5 and #140303/2012-0). The authors are
grateful to Professor Theo Guenter Kieckbusch for his long-term
dedication, collaboration, and numerous discussions on the math-
ematics of diffusion for active food packaging. Thank you for always
being a source of inspiration to our career and life.
References
1. Vermeiren L, Devlieghere F, VanBeen M et al 3. Dos Pires ACS, De Soares NFF, De Andrade
(2000) Development in the active packaging of NJ et al (2008) Development and evaluation of
foods. J Food Technol Africa 5:6–13 active packaging for sliced mozzarella preserva-
2. Almasi H, Jahanbakhsh Oskouie M, Saleh A tion. Packag Technol Sci 21:375–383
(2020) A review on techniques utilized for 4. De Souza AC, Dias AMA, Sousa HC et al
design of controlled release food active packag- (2014) Impregnation of cinnamaldehyde into
ing. Crit Rev Food Sci Nutr 0:1–21 cassava starch biocomposite films using
supercritical fluid technology for the develop- 19. Feigenbaum AE, Riquet AM, Scholler D
ment of food active packaging. Carbohydr (2000) Fatty food simulants: solvents to
Polym 102:830–837 mimic the behavior of fats in contact with pack-
5. Bierhalz ACK, Da Silva MA, De Sousa HC et al aging plastics. ACS Symp Ser 753:71–81
(2013) Influence of natamycin loading meth- 20. Silva AS, Garcı́a RS, Cooper I et al (2006)
ods on the physical characteristics of alginate Compilation of analytical methods and guide-
active films. J Supercrit Fluids 76:74–82 lines for the determination of selected model
6. Mastromatteo M, Mastromatteo M, Conte A migrants from plastic packaging. Trends Food
et al (2010) Advances in controlled release Sci Technol 17:535–546
devices for food packaging applications. Trends 21. Cruz JM, Sanches Silva A, Sendón Garcı́a R
Food Sci Technol 21:591–598 et al (2008) Studies of mass transport of
7. Yam KL, Zhu X (2012) Controlled release model chemicals from packaging into and
food and beverage packaging. Woodhead Pub- within cheeses. J Food Eng 87:107–115
lishing Limited 22. Ripoche AC, Chollet E, Peyrol E et al (2006)
8. Appendini P, Hotchkiss JH (2002) Review of Evaluation of nisin diffusion in a
antimicrobial food packaging. Innov Food Sci polysaccharide gel: influence of agarose and
Emerg Technol 3:113–126 fatty content. Innov Food Sci Emerg Technol
9. Chen X, Chen M, Xu C et al (2019) Critical 7:107–111
review of controlled release packaging to 23. Van Den Houwe K, Van LJ, Lynen F et al
improve food safety and quality. Crit Rev (2017) The use of Tenax® as a simulant for
Food Sci Nutr 59:2386–2399 the migration of contaminants in dry food-
10. Han JH (2000) Antimicrobial food packaging. stuffs: a review. Packag Technol Sci 31:781–
Woodhead Publishing Limited 790
11. Siepmann J, Peppas NA (2001) Modeling of 24. da Silva MA, Bierhalz ACK, Kieckbusch TG
drug release from delivery systems based on (2012) Modelling natamycin release from algi-
hydroxypropyl methylcellulose (HPMC). Adv nate/chitosan active films. Int J Food Sci Tech-
Drug Deliv Rev 48:139–157 nol 47:740–746
12. Serra L, Doménech J, Peppas NA (2006) Drug 25. Bierhalz ACK, da Silva MA, Kieckbusch TG
transport mechanisms and release kinetics from (2012) Natamycin release from alginate/pec-
molecularly designed poly(acrylic acid-g-ethyl- tin films for food packaging applications. J
ene glycol) hydrogels. Biomaterials 27:5440– Food Eng 110:18–25
5451 26. Bierhalz ACK, da Silva MA, Braga MEM et al
13. Poças MF, Oliveira JC, Oliveira FAR et al (2014) Effect of calcium and/or barium cross-
(2008) A critical survey of predictive mathe- linking on the physical and antimicrobial prop-
matical models for migration from packaging. erties of natamycin-loaded alginate films. LWT
Crit Rev Food Sci Nutr 48:913–928 Food Sci Technol 57:494–501
14. Crank J (1975) The mathematics of diffusion. 27. Rivera-Hernández L, Chavarrı́a-Hernández N,
Oxford University, Bristol del López Cuellar MR et al (2020) Pectin-
gellan films intended for active food packaging:
15. Siepmann J, Siepmann F (2008) Mathematical release kinetics of nisin and physico-mechanical
modeling of drug delivery. Int J Pharm 364: characterization. J Food Sci Technol
328–343
28. Imran M, Klouj A, Revol-Junelles AM et al
16. Flores S, Conte A, Campos C et al (2007) Mass (2014) Controlled release of nisin from
transport properties of tapioca-based active HPMC, sodium caseinate, poly-lactic acid and
edible films. J Food Eng 81:580–586 chitosan for active packaging applications. J
17. FDA (2007) Guidance for industry: prepara- Food Eng 143:178–185
tion of premarket submissions for food contact 29. Lavoine N, Guillard V, Desloges I et al (2016)
substances (Chemistry Recommendations), Active bio-based food-packaging: diffusion and
U.S. release of active substances through and from
18. Grob K (2008) The future of simulants in cellulose nanofiber coating toward food-
compliance testing regarding the migration packaging design. Carbohydr Polym 149:40–
from food contact materials into food. Food 50
Control 19:263–268
Chapter 19
Bioactive Properties of Probiotic and Prebiotic Edible Films

Jackson A. Medeiros, Carolina M. Niro, Mateus K. Salgaço, Kátia Sivieri,
and Henriette M. C. Azeredo
Abstract
Edible films have been increasingly studied as carriers for bioactive components, including probiotic
microorganisms and/or prebiotic compounds. Whereas the incorporation of prebiotic compounds does
not usually require wide modifications to the usual processing protocols to produce edible films, the
addition of live microorganisms requires special attention when it comes to processing conditions, in
order to assure that their viability is preserved as much as possible. In this chapter, we describe the basic
procedures for obtaining probiotic and/or prebiotic edible films as well as the specific determinations that
are required to properly characterize them in terms of their bioactive properties and stability.
Key words Probiotic bacteria, Film casting, Gastrointestinal tract, In vitro digestion, Lactobacillus
1 Introduction
The global probiotic market has been evaluated at USD 49.4 billion
in 2020 and estimated to grow to USD 69.3 billion by 2025.
Although probiotics may also be consumed as dietary supplements,
the market for probiotic foods outweighs that of supplements [1].
Edible films and coatings containing probiotics may be used as
primary packaging materials, but also function as part of the food
itself, since they are designed to be consumed with the food prod-
uct. Since those edible films are usually based on biopolymer matri-
ces capable of including probiotic cultures, providing some degree
of protection against external factors (including the low pH during
the passage through the stomach), they have been suggested as
interesting carriers of probiotic microorganisms, acting thus as
bioactive edible films by contributing to consumers’ health, even-
tually also contributing to food stability due to the competitive
effects of probiotics against deteriorating microorganisms [2]. Pre-
biotic compounds (such as inulin and several oligosaccharides) are
https://doi.org/10.1007/978-1-0716-3613-8_19,
325
326 Jackson A. Medeiros et al.
nondigestible components capable of selectively stimulating the

growth and/or activity of probiotics. They have also been sug-
gested as bioactive components of edible films containing probio-
tics [3, 4] or not [5, 6].
The ability of an edible film to keep a good probiotic stability
upon processing and storage depends on the ability of its compo-
nents to protect the bacteria and also on the ability of the bacteria
themselves to survive processing and storage conditions.
2 Materials
2.1 Film Preparation 1. Autoclave.

2. Incubation chamber.
3. Centrifuge.
4. Hot plate and magnetic stirrer (when the film dispersion
requires heating).
5. 50-mL sterile tubes.
6. Petri dishes (or a flat glass plate with a drawdown bar).
7. Ethanol 70 vol%.
8. NaCl.
9. Probiotic culture.
10. Prebiotic compound (if it is to be included).
11. Culture medium for the probiotic culture of choice (e.g.,
DeMan–Rogosa–Sharpe [MRS] broth).
12. Film matrix (polysaccharide or protein).
13. Plasticizer (e.g., polyols such as glycerol or sorbitol).
2.2 Estimating 1. Lab scale with an accuracy of 0.1 mg.

Probiotic Viability on 2. Incubation chamber.
Film-Forming
3. Air oven.
Dispersions and Films
4. Vortex.
5. Hot plate and magnetic stirrer.
6. Aluminum pans.
7. Petri dishes.
8. 10-μL and 1000-μL micropipettes and sterile tips.
9. Solidified culture medium suitable for the bacteria of choice.
11. NaCl.
Probiotic and Prebiotic Edible Films 327
2.3 Evaluating the 1. Incubation chamber.

Prebiotic Activity of 2. UV-vis spectrometer.
Films
3. Standard bacterial strain.
4. Culture medium (e.g., MRS broth), with and without glucose.
5. Prebiotic compound.
6. 0.5 McFarland standard (see Note 1).
2.4 In Vitro Digestion 1. Lab scale with an accuracy of 0.1 mg.

Method 2. Stomacher.
3. Metabolic bath.
4. Shaking water bath.
5. 500-mL threaded vials.
6. Volumetric flasks.
7. Sterile or ultrapure water.
8. KCl.
9. KH2PO4.
10. NaHCO3.
11. NaCl.
12. MgCl2(H2O)6.
13. (NH4)2CO3.
14. CaCl2(H2O)2.
15. CaCl2.
16. NaOH.
17. Pepsin (e.g., Sigma-Aldrich P7000 or similar).
18. Bile (e.g., Sigma-Aldrich B8631 or similar).
19. Pancreatin (e.g., Sigma-Aldrich P1750 or similar).
3 Methods
3.1 Film Preparation See Note 2 for precautions to avoid the growth of any contaminant
microorganisms.
1. Activate the probiotic culture, according to the specific growth
requirements of the strain of choice. When it comes to bacteria
from the Bifidobacterium genus and those from the Lactoba-
cillaceae family, proceed as follows [7]: Inoculate the culture
into MRS broth and incubate it at 37 °C for 48 h. Aseptically
transfer the cell suspensions into 50-mL sterile tubes and cen-
trifuge them at 4000 g for 10 min. Discard the supernatant,
wash the bacteria pellet with 35-mL sterile saline solution

(0.85% w/v NaCl), and centrifuge it again. Repeat the
discarding–washing–centrifuging steps one more time and
add the resulting bacteria pellet directly to the film
formulation.
2. Prepare the film-forming dispersion by adding the matrix to
the solvent (usually distilled water), as well as the plasticizer and
the prebiotic component (if the formulation includes a prebi-
otic). Homogenize the dispersion as required for the chosen
formulation. When using a matrix that requires heating (e.g.,
starch, which is typically heated at 80–90 °C for gelatinization),
the film-forming dispersion should be cooled down to no more
than 40 °C before adding the probiotic (see Note 3).
3. Add the probiotic pellet in such an amount (calculated from a
previous viable cell count on the pellet) to provide the film-
forming dispersion with the desired probiotic counting (see
Note 4). Stir the probiotic-containing dispersion (see Note 5).
4. Set aside at least 3 g of the dispersion (for plating and counting)
and spread the remaining dispersion onto the casting surface
(e.g., Petri dish or a flat glass plate) for a uniform wet thickness
(defined by using a drawdown bar when on a flat surface).
5. Dry the film-forming dispersion (see Note 6).
3.2 Estimating the The viable probiotic counts should always be taken on a dry basis,
Viability Loss of the and that is why the total solid content of the sample to be analyzed
Probiotics on Film should be previously determined. The reference method to deter-
Drying mine the solid content is AOAC 968.11 [8], for which we suggest
some adaptations:
3.2.1 Total Solid
Contents 1. Weigh a sample of the film-forming dispersion (about 5 g) or
the dried film (about 0.5 g), with an accuracy of 0.1 mg, in a
previously dried aluminum pan (in triplicate).
2. Dry the aluminum pan + sample in an air oven at 105 °C for as
long as it takes to obtain three consecutive weighings (with no
less than 1 h between them) with differences of no more than
2 mg.
3. Determine the total solid content (g/g) as the weight of the
sample that was kept during drying.
3.2.2 Viable Cell Count 1. Transfer 1 g of the film-forming dispersion into 9 mL of a

on the Film-Forming sterile saline solution (0.85% NaCl), obtaining thus a 10-1
Dispersion dilution.
2. Homogenize the 10-1 dilution and make serial dilutions by
transferring 1 mL of the original dilution into 9 mL of the
sterile saline solution for the next dilution (Fig. 1).
Fig. 1 Scheme of the serial dilutions from a film-forming dispersion, followed by drop plating in a petri dish
divided into quadrants
3. Plate the different dilutions by the drop plate method [9] (see
Note 7). Use Petri dishes containing the solidified culture
medium, e.g., MRS agar for bacteria from the Lactobacillaceae
family [7], or Tryptone Glucose Yeast (TGY) agar for Bacillus
coagulans [3], and divide each agar plate into four quadrants,
each quadrant for a dilution. Pipette out 10 μL (“drop”) from
each dilution on the proper quadrant (at least in triplicate)
(Fig. 1). Leave the plates open until the drops are absorbed
by the culture medium.
4. Incubate the plates under the prescribed conditions for bacte-
rial growth (e.g., 37 °C for 72 h in anaerobic conditions for
Lactobacillus acidophilus).
5. Search for a dilution (quadrant) in which the number of colo-
nies per drop is in the 3–30 range, and count the colonies [10].
6. Calculate the viable cell count (on a dry basis), using Eq. 1 (see
Note 8).
Cell countðcf u g - 1 , d:b:Þ

number of colonies × dilution f actor
= ð1Þ
sample volume ðmLÞ × solid contentðg=gÞ
3.2.3 Viable Cell Count 1. Transfer a 0.1 g sample of the film into 9.9 mL of a sterile saline
on the Dried Film solution (0.85% NaCl). Vortex it for 1 min, then stir it at
80 rpm for 1 h at the temperature of incubation (which
depends on the microorganism), obtaining a 10-2 dilution.
2. Proceed as in Subheading 3.2.2 from step 2. With those results
(from Subheadings 3.2.2 and 3.2.3), you will be able to check
the viability loss derived from drying.
3.3 Estimating the 1. Store the film samples aseptically under previously defined
Viability Loss of the temperatures (e.g., storage at 4 and 25 °C, as proposed by
Probiotics on Films Kanmani and Lim [11]).
During Storage 2. Proceed to periodic viable counts (e.g., at 10-days intervals), as
described in Subheading 3.2.3.
3.4 Assessing the The presence of probiotic microorganisms may provide the films
Antimicrobial Capacity with some antimicrobial activity due to the competitive advantages
of Probiotic Films of the probiotics over other microorganisms. A recent review has
addressed the antimicrobial testing methods to be applied to films
[12], including disk diffusion method, viable cell count method,
and optical density-based methods. Antifungal and antibacterial
assessments of food packaging materials are also addressed in
Chaps. 15 and 16 of this volume, respectively.
3.5 Evaluating the The prebiotic activity of a prebiotic-containing film may be assessed
Prebiotic Activity of by measuring the effect of the presence of the film on the growth of
Films probiotic bacteria. The following example is based on assessing the
growth of bacteria that use MRS broth as a culture medium (e.g.,
Bifidobacterium genus and Lactobacillaceae family).
1. Transfer a sample of a previously plated standard bacterial strain
into 10 mL of MRS broth, leave it to grow at 37 °C for 24 h,
and read its optical density (OD) in a UV-vis spectrometer at
425 nm. Read also the OD of a 0.5 McFarland standard at the
same wavelength, which corresponds to 1 × 108 cfu mL-1, and
assess the bacterial concentration in the culture by using Eq. 2.
OD sample × 108
Bacterial concentration cfu mL - 1 = ð2Þ
OD McFarland
2. Transfer 2 mL of the cultured broth into each of the following:

(A) 10 mL of MRS broth (positive control), (B) 10 mL of MRS
broth without glucose containing 20 g L-1 of the prebiotic
film, (C) 10 mL of MRS broth without glucose containing
20 g L-1 of a control film without the prebiotic (negative
control), and (D) 10 mL of MRS broth without glucose con-
taining only the prebiotic agent (at the same amount as in the
film sample in B). Leave them to grow at 37 °C for 48 h under
anaerobic conditions.
3. Read the OD from all the treatments at 425 nm, as well as the
OD of the 0.5 McFarland standard, to assess the bacterial
concentration in each treatment by using Eq. 2, in order to
evaluate the prebiotic effect of the prebiotic film (B) when
compared to the control film (C) and the free prebiotic (D).
3.6 Evaluating the When a prebiotic agent is added to the formulation of a film
Prebiotic Effect on containing a probiotic microorganism, the prebiotic may have a
Probiotic Viability on positive effect on the probiotic viability during processing and/or
Film Processing and storage. This effect (when it occurs) may be assessed by using Eq. 3.
Storage Prebiotic effect = P 0 - P f - P0 - Pf ð3Þ
nonpreb preb
P0: initial (i.e., before processing or storage) probiotic cell count

(log cfu g-1, on a dry basis); Pf: final (i.e., after processing or
storage) probiotic cell count (log cfu g-1, on a dry basis); nonpreb:
film containing the probiotic but not the prebiotic; preb: film
containing both probiotic and prebiotic.
3.7 In Vitro Digestion In vitro digestion models have been widely used to overcome the
Method constrictions associated with in vivo methodologies. COST Action
INFOGEST has developed a harmonized international protocol for
static simulation of digestion in the upper gastrointestinal tract of
adults [13]. This protocol is easily accessible, relatively inexpensive,
and widely used for digestion assessments. However, it has some
restrictions, especially in the gastric phase. Therefore, recently,
COST Action INFOGEST proposed a semi-dynamic model
including kinetic aspects, gradual acidification, secretion, and emp-
tying of fluid and enzyme [14]. These adaptations were made to
provide kinetic data on digestion and absorption of nutrients. Since
our objective here is to propose methods for the simulated diges-
tion of probiotic films, not involving digestion and absorption of
nutrients, we detail below the protocol of Minekus et al. [13], with
some adaptations.
3.7.1 Preparation of 1. Perform all weightings described in Table 1 for solution prepa-
Solutions for In Vitro ration. All dilution processes should be in sterile and/or ultra-
Digestion Protocol pure water (see Notes 9 and 10).
2. Prepare three digestive fluids: (A) salivary simulated fluid—
SSF, (B) gastric simulated fluid—SGF, and (C) intestinal
simulated fluid—SIF (Table 2), in 500-mL threaded vials.
3.7.2 In Vitro Digestion 1. Oral simulated phase: Transfer 200–300 mg of probiotic film
Process into 4 mL of a simulated salivary solution—SSF (25 μL of
0.3 mmol L-1 CaCl2 and 975 μL of ultrapure water) and
homogenize it in a Stomacher for 1 min at 150 rpm. Incubate
the mixture for 2 min at 37 °C in a metabolic bath (Fig. 2).
Prepare three vials, one of which to be used to assess this phase
itself, and the others to be conducted to the subsequent phases.
2. Gastric simulated phase: Add 7.5 mL of simulated gastric
solution—SGF [1.6 mL of pepsin solution 25,000 U mL-1,
5 μL of 0.3 mmol L-1 CaCl2; 0.2 mL of 1 mmol L-1 HCl and
0.695 mL of ultrapure water] into each of the vials coming
from the simulated oral phase. Adjust the pH to 3 and incubate
Table 1
Stock solutions for simulated digestion fluids
Constituent g L-1 mol L-1

KCl 37.3 0.5
KH2PO4 68.0 0.5
NaHCO3 84.0 1.0
NaCl 117.0 2.0
MgCl2(H2O)6 30.5 0.15
(NH4)2CO3 48.0 0.5
CaCl2(H2O)2 44.1 0.3
Table 2
Recommended concentrations of electrolytes for Simulated Salivary Fluid
(SSF), Simulated Gastric Fluid (SGF), and Simulated Intestinal Fluid (SIF)
Constituent SSF (pH 7) SGF (pH 3) SIF (pH 7)

Volume of stock solution (mL)
KCl 15.10 6.90 6.80
KH2PO4 3.70 0.90 0.80
NaHCO3 6.80 12.50 42.50
NaCl – 11.80 9.60
MgCl2(H2O)6 0.50 0.40 1.10
(NH4)2CO3 0.06 0.50 –
HCl 0.09 1.30 0.70
H2O (Milli-Q) 373.75 365.70 338.50
Total (mL) 400 400 400
the mixture at 37 °C for 2 h to simulate gastric digestion (see

Note 11) (Fig. 2).
3. Intestinal Simulated phase: Add 11 mL of simulated intestinal
solution—SIF (6.8 mmol L-1 KCl; 0.8 mmol L-1 KH2PO4;
85 mmol L-1 NaHCO3; 38.4 mmol L-1 NaCl; 0.33 mmol L-1
MgCl2), 5 mL of pancreatin 800 U mL-1 solution, 2.5 mL of
160 mmol L-1 bile solution, 40 μL of 0.3 mmol L-1 CaCl2,
0.15 mL of NaOH 1 mmol L-1, and 1.31 mL of ultrapure
water) to the gastric chyme (see Note 12) (Fig. 2).
4. A complete experiment should be conducted in parallel with
three incubation vials (white/control).
Fig. 2 Illustration of the in vitro digestion process
4 Notes
1. The 0.5 McFarland standard is a mixture of 0.5 mL of BaCl2 at

1% and 99.5 mL of H2SO4 at 1%.
2. Edible films should be free from pathogenic and deteriorating
microorganisms for safety reasons. Moreover, it is important to
avoid the growth of any contaminant microorganisms, which
may impair the probiotic viable cell counting along the process.
All surfaces that will come into contact with film components,
film-forming dispersions, or the dried films should be thus
previously sanitized. Steam-sterilizable glassware and equip-
ment components should be autoclaved at 121 °C for
15 min, whereas non-sterilizable surfaces should be sanitized
with ethanol 70 vol%.
3. Because each strain has its own susceptibilities, an ideal tem-
perature for adding the probiotic should be determined on a
case-by-case basis.
4. Although there is not a standard recommendation for probiotic
viable counts in edible films, we propose here a final count (for
the dried film) of about 9 log cfu g-1. Considering that
non-spore-forming and spore-forming strains have been

reported as presenting viability losses of about 2 and 1 log cfu
g-1 (respectively) on processing [3, 15], it may be established
as a general rule that the film-forming dispersions should con-
tain about 11 log cfu g-1 (for formulations with nonspore-
forming strains) or 10 log cfu g-1 (for formulations with spore-
forming strains), on a dry basis. Ideally, however, we recom-
mend that the actual losses in each case are assessed by prelimi-
nary tests, since the losses depend not only on the used strain
but also on interactions of the bacteria with the formulation
components, as well as the processing conditions.
5. Spore-forming bacteria may be subjected to a stirring speed of
about 10,000 rpm for no longer than 20 min. For nonspore-
forming bacteria, the stirring speed should be no more than
2000 rpm for no longer than 20 min.
6. The drying (time and temperature) conditions depend not only
on the drying equipment used (e.g., air circulation speed), but
also on the solid content and thickness of the dispersion. The
definition of the drying conditions of probiotic-containing
films should also take the thermal stability of the probiotic
strain into consideration. Nonspore-forming bacteria should
ideally be dried at no more than 30 °C, whereas spore-forming
ones may be subjected to about 50 °C (although films contain-
ing Bacillus coagulans have been reported to be dried at 80 °C
for 150 min) [3]. We recommend preliminary tests in order to
establish the drying conditions; eventual adjustments in disper-
sion formulation (solid content) and/or wet thickness may be
necessary in order to enhance probiotic viability.
7. The plating may be carried out by traditional methods such as
pour plate or spread plate [9]. We recommend the drop plate
method because it is time-saving and cheaper (requiring less
petri dishes and culture medium). On the other hand, the
micropipette should be very well calibrated, since the sample
volume is too small.
8. The counts are usually expressed as log cfu g-1, since changing
the values to their logarithms makes it easier to compare values.
9. In order to obtain better volumetric accuracy, it is suggested
that each reagent should be prepared in volumetric flasks.
10. All materials are of standard analytical grade. Sodium bicarbon-
ate (0.5 M) should be filtered through a 0.22-μm membrane
under vacuum.
11. Add 1 g of pepsin into 10 mL of ultrapure water.
12. Add 0.8 g of pancreatin into 10 mL of ultrapure water. Add
0.7 g of bile into 10 mL of ultrapure water.
References
1. Market Data Forecast Probiotics Market 9. da Silva N, Taniwaki MH, Junqueira VCA et al
(2020). https://www.marketdataforecast. (2018) Microbiological examination methods
com/market-reports/probiotics-market. of food and water: a laboratory manual. CRC
Accessed 8 Apr 2021 Press
2. Espitia PJP, Batista RA, Azeredo HMC et al 10. Herigstad B, Hamilton M, Heersink J (2001)
(2016) Probiotics and their potential applica- How to optimize the drop plate method for
tions in active edible films and coatings. Food enumerating bacteria. J Microbiol Methods
Res Int 90:42. https://doi.org/10.1016/j. 44:121–129. https://doi.org/10.1016/
foodres.2016.10.026 S0167-7012
3. Oliveira-Alcântara AV, Soares Abreu AA, Gon- 11. Kanmani P, Lim ST (2013) Development and
çalves C et al (2020) Bacterial cellulose/cashew characterization of novel probiotic-residing
gum films as probiotic carriers. https://doi. pullulan/starch edible films. Food Chem 141:
org/10.1016/j.lwt.2020.109699 1041–1049. https://doi.org/10.1016/j.
4. Pruksarojanakul P, Prakitchaiwattana C, Setta- foodchem
chaimongkon S et al (2020) Synbiotic edible 12. Abdollahzadeh E, Nematollahi A, Hosseini H
film from konjac glucomannan composed of (2021) Composition of antimicrobial edible
lactobacillus casei-01® and Orafti®GR, and its films and methods for assessing their antimi-
application as coating on bread buns. J Sci crobial activity: a review. Trends Food Sci Tech-
Food Agric 100:2610–2617. https://doi. nol 110:291–303. https://doi.org/10.1016/
org/10.1002/jsfa.10287 j.tifs.2021.01.084
5. Fernandes LM, Guimarães JT, Silva R et al 13. Minekus M, Alminger M, Alvito P et al (2014)
(2020) Whey protein films added with galac- A standardised static in vitro digestion method
tooligosaccharide and xylooligosaccharide. suitable for food-an international consensus.
Food Hydrocoll 104:105755. https://doi. Food Funct 5:1113–1124. https://doi.org/
org/10.1016/j.foodhyd.2020.105755 10.1039/c3fo60702j
6. Bersaneti GT, Garcia S, Mali S et al (2019) 14. Mulet-Cabero A-I, Egger L, Portmann R et al
Evaluation of the prebiotic activities of edible (2020) A standardised semi-dynamic in vitro
starch films with the addition of nystose from digestion method suitable for food – an inter-
Bacillus subtilis natto. 116:108502. https:// national consensus. Food Funct 11:1702–
doi.org/10.1016/j.lwt.2019.108502 1 7 2 0 . h t t p s : // d o i . o r g / 1 0 . 1 0 3 9 /
7. Ebrahimi B, Mohammadi R, Rouhi M et al C9FO01293A
(2018) Survival of probiotic bacteria in carbox- 15. Soukoulis C, Singh P, Macnaughtan W et al
ymethyl cellulose-based edible film and assess- (2016) Compositional and physicochemical
ment of quality parameters. LWT Food Sci factors governing the viability of Lactobacillus
Technol 87:54–60. https://doi.org/10. rhamnosus GG embedded in starch-protein
1016/j.lwt.2017.08.066 based edible films. Food Hydrocoll 52:876–
8. Association of Official Analytical Chemists 887. https://doi.org/10.1016/j.foodchem.
(AOAC) (2000) Official methods of analysis. 2014.03.008
AOAC International, Arlington
Chapter 20
Sensory Acceptance Test of Edible Packaging Using Hedonic

Scale
Suzana Maria Della Lucia and Tarcı́sio Lima Filho
Abstract
Food packaging has various functions, including protecting food and food products from potential damage
and degradation and providing information to consumers. Conventional packaging is commonly a one-
time-use item that is then discarded. Therefore, there is a demand to achieve more sustainable, higher-
quality, and healthier food production systems. In this context, edible packaging can be a good alternative
because they can be manufactured from biobased, biodegradable, and/or edible materials. Studies aiming
at developing edible packaging must be based on tests involving packaging formulations, physical-
mechanical properties, microbiological safety, and others. However, one must have in mind that, because
edible packaging can be eaten, it is essential to study their sensory characteristics and the influence of such
on consumers’ acceptance. In this chapter, a protocol for evaluating the sensory acceptance test of edible
packaging using a nine-point hedonic scale is presented. We expect to assist researchers and professionals
that are involved in the development and production of edible packaging, which is a permanent tendency in
the food industry.
Key words Affective methods, ANOVA, Food packaging, Principal component analysis
1 Introduction
Food packaging has various functions. Its primary role is to provide

protection to food, extending food shelf-life by reducing exposure
to spoilage agents (such as microorganisms, oxygen, water vapor,
and off-flavors) and avoiding losses of food compounds, such as
flavor volatiles [1, 2]. Therefore, food packaging plays an important
role in society, protecting food products from potential damage
and degradation while ensuring safety and hygiene, and reducing
food waste. Other important functions include providing informa-
tion on nutritional content, storage, handling of packaging material
after use, and marketing [1–3].
https://doi.org/10.1007/978-1-0716-3613-8_20,
337
338 Suzana Maria Della Lucia and Tarcı́sio Lima Filho
Conventional packaging is commonly a one-time-use item that

is discarded soon after the use of the product by the consumer. As
such, conventional, single-use packaging poses a terrible environ-
mental problem when it is not biodegradable [3].
In this context, recently, there has been a demand for more
sustainable, higher-quality, and healthier food production systems.
This includes food packaging that does not increase pollution, that
is, packaging that is environmentally friendly, manufactured from
renewable natural resources and of biodegradable character. Thus,
the search for sustainable packaging that is biodegradable and/or
edible has led to the development of various edible packaging
systems [2, 3].
The materials of the edible packaging are derived from edible
ingredients, such as natural polymers that can be directly consumed
by humans without any potential health risk. Most proteins and
polysaccharides are edible and can be used as matrices for edible
films and coatings.
Polysaccharides (such as starches, cellulose and its derivatives,
chitosan/chitin, and gums), polypeptides (animal or plant-based
proteins), and lipids can be used to manufacture edible packaging.
The utility and the value of edible packaging are seen in its capacity
to maintain food quality, extend shelf-life, reduce waste, and con-
tribute to the economic efficiency of packaging materials [1, 3].
In recent years, a strong research effort has been driven toward
edible food packaging for different foods and beverages. Each
packaging has its peculiar characteristics in order to be as suitable
as possible for each product. Many edible packaging systems are still
in the stages of laboratory tests, but numerous are already being
commercialized. Examples are as follows: coffee cups made of white
chocolate coated wafer; seaweed-based cups made with natural
sweeteners, gluten, and gelatin; cookie-based cups; starch-based
films for sandwiches printed with vegetable-based inks; seaweed-
based sachets used to pack pasta seasoning that can be dissolved and
consumed along with the pasta; trays, pots, and cups made from
starch, natural fibers, sugarcane, bamboo, and rice; seaweed-based
films used to pack hamburgers that can be eaten along with the
meat; edible films made from food industry tailings, tomato,
spinach, papaya, and guava, ensuring sustainability and the reduc-
tion of food waste; seaweed-based film with seasonings that can be
dissolved in warm water to prepare a soup; and many other exam-
ples of edible packaging [4–7].
It is important to emphasize that the studies aiming at devel-
oping edible packaging must be based on tests involving packaging
formulations, physical-mechanical performance, barrier and ther-
mal properties, microbiological safety, nutritional characteristics,
among others. However, one must have in mind that, because
edible packaging is expected to be eaten, it is essential to study
Acceptance Test of Edible Packaging 339
their sensory characteristics and the influence of such on consu-

mers’ acceptance. It is useless for an edible packaging to have the
desirable physical-chemical, microbiological, and other qualities, if
it does not have support from the consumers in terms of its appear-
ance, aroma, flavor, and texture. Edible packaging can be sensorially
inert, acting just as a protection layer, without any sensory alter-
ation, or have sensory appeal, which is desirable in many products,
since they can complement the sensory characteristics of the food
and increase the sensory acceptance of the product by the con-
sumer. From this point of view, sensory acceptance tests are essen-
tial tools for providing the optimal answers about the acceptance of
edible packaging by the market.
Sensory acceptance tests are part of one of the groups of
classical sensory evaluation methods: the Affective Methods,
which express consumers’ subjective reactions to food, beverages,
and other materials, such as degree of liking or disliking, accepting
or rejecting, and preferring order. They are divided into qualitative
and quantitative tests. Qualitative tests aim to obtain subjective
information in an exploratory way, studying the thoughts, atti-
tudes, perceptions, and behaviors of the consumers in relation to
the product. Quantitative Affective Methods aim to obtain direct
information from consumers in relation to their preferences and/or
acceptance of sensory attributes or the product as a whole [8, 9].
Acceptance tests are among the quantitative Affective Methods.
These tests aim to assess the degree to which consumers like or
dislike the product. Among the most widely used acceptance tests is
the Hedonic Scale test. In this test, the consumer expresses the
degree of liking or disliking a given product, globally or in relation
to a specific attribute. The most used scales are those of seven and
nine points, which contain the defined terms located, for example,
between “like extremely” and “dislike extremely” [8–10].
In this chapter, a protocol for evaluating the sensory acceptance
test using a nine-point Hedonic Scale of edible packaging is pre-
sented. We expect to assist researchers and professionals that are
involved in the development and production of edible packaging,
which is a permanent tendency in the food industry.
2 Materials
• Edible packaging samples.

• Laboratory with individual booths for Sensory Analysis.
• White light.
• Answer sheet.
3 Methods
The acceptance test described in this protocol can be performed to

evaluate only the edible packaging, to evaluate the acceptance of
food consumption together with the edible packaging, or to evalu-
ate the acceptance of only the food. In cases where the packaging is
designed to be consumed together with the food, without making
sense to evaluate only the packaging, the protocol can be applied in
two assays: (i) to analyze the sensory acceptance of the food con-
sumed together with the edible packaging and (ii) to analyze the
sensory acceptance of the food. Thus, it is possible to investigate
whether the edible packaging has a positive or negative influence on
the sensory acceptance of the food.
For standardization purposes, the protocol was described for
the evaluation of edible packaging only. However, the protocol can
be applied to the other cases mentioned.
3.1 Defining the In acceptance tests, global acceptance (overall impression; i.e., the
Sensory Attributes to edible packaging as a whole) can be assessed or also the acceptance
be Evaluated of specific sensory attributes of the edible packaging, such as color,
appearance, thickness, aroma, texture, taste, and flavor. These attri-
butes can be evaluated globally or by specifying a certain aroma,
taste, and flavor (e.g., banana aroma, vinegar aroma, sweet taste,
salty taste, acid taste, strawberry flavor, and chocolate flavor).
It is noteworthy that, when specifying a sensory attribute to be
evaluated, the consumer will pay more attention to this attribute
than he/she would do if he/she were evaluating global acceptance
(without targeting a specific attribute). Therefore, researchers
should analyze what is most interesting for the research, whether
to draw the consumer’s attention to specific attributes or to simu-
late the real consumption process, when no attribute is targeted.
The greater the number of attributes to be analyzed, the more
complex the analysis will be for the evaluators (consumers). There-
fore, researchers should limit the analysis to only those attributes
that are central to the research.
3.2 Defining the There are several scales used to measure the sensory acceptance of
Scale food and packaging. Among the available scales, the Hedonic Scale
stands out, which was described in detail by Jones et al. [11] and by
Peryam and Pilgrim [12].
Among the existing Hedonic Scales, the nine-point scale, rang-
ing from the terms 1 = “dislike extremely” to 9 = “like extremely”
is the most used by the scientific community and industry (Fig. 1).
Therefore, it will be the scale showcased in this protocol.
This scale is simple and easily understood by consumers.
Through this scale, the consumer expresses sensory acceptance of
the product, stating how much he/she liked or disliked the
product.
Fig. 1 Model score sheet for acceptance test with a nine-point balanced Hedonic Scale
For more information on the other scales that can be used in

acceptance tests, readers are recommended to refer to Lawless and
Heymann [10], specifically Chaps. 7 and 14.
3.3 Presentation of The packaging samples can be served monadically (one at a time)
Edible Packaging and sequentially (one after the other), that is, a response required
after each packaging. The samples may also be served all at once,
but this requires the panelist to match the correct test sample to the
correct three-digit code on the answer sheet. Therefore, for testing
with untrained evaluators (consumers), it is safer to serve products
one at a time and retrieve the sample after each response [10] (see
Note 1).
3.4 Defining the In affective tests, the team of evaluators should be composed of a
Number of Evaluators group of people selected as a representative sample of a population.
Usually, the population is the usual or potential consumer market
for the product [8] (see Note 2).
Acceptance tests, performed in the laboratory, are used to
preliminarily select samples for future tests and in the stages of
new product development [8]. For laboratory tests, studies carried
out recently and published in scientific journals have used teams
composed of 80–150 consumers. Within this range, the greater the
number of samples to be evaluated, the greater the number of
consumers needed.
In addition to the laboratory, tests can also be carried out in
central locations (e.g., supermarkets and restaurants) and house-
holds. In these places, there is less control of the environment and
the analysis conditions (noise, inadequate lighting, absence of
individual cabin for analysis); therefore, a greater number of eva-

luators is necessary, when compared with laboratory tests. For more
information, readers are recommended to refer to Meilgaard,
Civille and Carr [13], Chap. 13 specifically.
3.5 Procedure for 1. Code the samples with random three-digit numbers.
Analyzing Packaging 2. Prepare and print the evaluation answer sheet (Fig. 1) for all
Samples evaluators (one form per packaging sample).
3. Provide pens to fill in the answer sheet.
4. The analysis site must be quiet, with a pleasant temperature and
white light (if the researchers are interested in the visual analysis
of the packaging).
5. In another room, welcome the evaluators who will participate
in the study.
6. Deliver, explain, and request a signature on the Free and
Informed Consent Form (mandatory in research with human
beings, consult the Ethics Committee for Research with Human
Beings of the place where the research will be carried out).
7. Collect information about research participants. In the sensory
evaluation, the evaluators are the instruments of analysis; there-
fore, characterizing the profile of the participants is necessary.
The basic information needed is the age, gender, and frequency
or intention of consumption of the product under study. Some
information that can also be requested, according to the
research interest, is the level of education, monthly family
income, and occupation.
8. Deliver the evaluation form (Fig. 1) to the consumers.
9. Explain to the evaluators the number of samples to be analyzed,
one at a time, and the analysis procedure: the evaluators (con-
sumers) must taste the product and inform how much they
liked or disliked the appearance, aroma, texture, flavor, and the
overall impression of the product. Between evaluations, evalua-
tors should rinse their mouths with water and wait for 3 min
before testing another sample.
10. Clear the doubts of the evaluators.
11. Invite the evaluators to enter the packaging evaluation room.
12. Serve the edible packaging samples in a monadic and random
manner.
13. As soon as the evaluator finishes analyzing the first packaging
sample, collect the completed answer sheet and ask him/her to
rinse the mouth with water.
14. After 3 min, serve the next sample to be analyzed (chosen
randomly). Perform this procedure until all samples are
analyzed.
15. Repeat steps 5–14 for all evaluators, until the total number of
evaluators is completed.
3.6 Data Analysis The responses of each sensory attribute must be tabulated and
analyzed separately. The data from the nine-point scales are
assigned values one through nine, nine usually being the “like
extremely” level.
The results can be analyzed by descriptive statistics, by means of
frequency distribution graphs of the hedonic scores of each attri-
bute. The frequency distribution graphs allow the researchers to
visualize the segmentation of the hedonic scores of each sample
within the scale. Thus, it is possible to compare the performance of
two or more packaging samples, checking those that had the high-
est mean hedonic scores, that is, the most accepted one.
The results can also be analyzed using parametric statistics,
t-tests on means for two packaging samples, or analysis of variance
followed by comparisons of means for more than two packaging
samples [10]. According to Lawless and Heymann [10], even
though the scale may not achieve a true interval level of measure-
ment, the parametric approach is usually justified based on the
larger sample size in a consumer test.
For each sensory attribute, an analysis of variance (ANOVA) in
completely randomized blocks (consumers) can be performed to
determine whether there is a significant effect ( p ≤ 0.05) of the
packaging samples on the hedonic scores. The mathematical model
that represents the analysis is shown in Eq. 1. The hypothesis of
nullity of zero variability is tested between the packaging samples
(H 0 : σ 2R = 0).
Y ij = m þ P i þ C j þ e ij ð1Þ
Yij—hedonic scores of packaging i attributed by the consumer j
m—constant inherent to the model or the overall average value
Pi—random effect of packaging i
Cj—random effect of consumer j
eij—normal random error, independent and equally distributed
(0,σ 2).
When necessary, comparisons of means should be performed.
The Preference mapping method can be used in order to
evaluate the individual responses of each consumer and not just
the average response of the group of consumers who analyzed the
packaging. Through the Preference mapping, in a single graph,
hedonic information for each consumer participating in the study
is simultaneously presented in a multidimensional space represent-
ing and containing the evaluated packaging [10]. Preference
mapping can give a clear idea of which changes must be made in
product reformulation.
The Internal Preference mapping, sometimes called MDPREF,
is usually a principal component analysis (PCA), in which the
hedonic scores are arranged in a matrix of product (packaging)
(in p lines) for consumers (in n columns), which is reduced in a

small number of independent components, minimizing the loss of
original information (variation). The purpose of the internal Pref-
erence mapping is to find a small number of principal components
(usually two or three) that explain a large percentage of the varia-
tion in the consumer hedonic responses [9, 10]. For internal Pref-
erence mapping, all consumers should evaluate all the products.
According to Lawless and Heymann [10] and Lavine et al. [14], in
order to have a reasonable perceptual map, the researcher should
have the consumers evaluate at least six products that span the
perceptual space.
3.7 Presentation and The results can be presented using tables and graphs. As an exam-
Interpretation of the ple, the results of a fictitious research on edible packaging will be
Results presented.
3.7.1 Frequency In Fig. 2, frequency distribution graphs of the hedonic scores are
Distribution presented for the aroma and flavor attributes of three edible pack-
aging samples (A, B, and C).
In Fig. 2, packaging sample A presented a higher frequency of
higher hedonic scores (from 7 to 9), when compared with the other
packaging, for the two attributes under study (aroma and flavor). In
addition, packaging sample B presented the highest distribution of
negative hedonic scores (from 1 to 4) for the aroma attribute, that
is, sample B was the one that presented aromas with greater sensory
rejection, while sample C presented the highest distribution of
negative hedonic scores for the flavor attribute. The same analysis
can be done for the other sensory attributes and the overall impres-
sion of the packaging samples.
3.7.2 ANOVA Table 1 presents the results of the ANOVA and the comparison of
means test of three edible packaging samples (A, B, and C). The
samples did not differ in terms of acceptance as for the appearance
and texture attributes ( p > 0.05). A significant effect of the pack-
aging samples ( p ≤ 0.05) in the ANOVA (Table 1) was verified in
Fig. 2 Examples of frequency distribution graphs of the hedonic scores for the aroma and flavor attributes of
three packaging samples
Table 1
Summary of ANOVA, average hedonic scores, and Tukey’s test for each packaging sample and
sensory attribute
Mean hedonic scoresC
Attribute MSpacA MSresB p-value Packaging A Packaging B Packaging C

Appearance 0.36 1.34 0.7543 7.2 a 7.1 a 7.1 a
Aroma 4.52 1.44 0.0440 7.9 a 4.9 b 7.0 a
Texture 5.01 1.83 0.0649 6.4 a 6.1 a 6.3 a
Flavor 6.81 1.91 0.0228 8.2 a 6.7 b 4.3 c
Overall impression 6.43 1.75 0.0238 7.8 a 6.3 b 6.1 b
A
Mean-square of the packaging sample
B
Mean-square residue
C
Means followed by at least one equal letter within the same row do not differ ( p > 0.05) by Tukey’s test
the hedonic scores of the aroma and flavor attributes and overall
impression. Packaging sample B presented the least accepted
aroma. Sample C presented the least flavor acceptance. Packaging
samples B and C showed the lowest overall impression, while
sample A had the highest hedonic scores for flavor attribute and
overall impression.
3.7.3 Internal Preference The Internal Preference Map obtained from the overall impression
Mapping analysis, of six edible packaging samples, is shown in Fig. 3.
The first principal component (PC1) accounted for 71.4% of
the variation in the results, whereas the second principal compo-
nent (PC2) accounted for 10.2% of the variation in the results, and
the two together explained 81.6% of the variation in the results
(Fig. 3) (see Note 3).
The data for each consumer are shown on the map as a black
dot, which corresponds to the endpoint of the fitted vector. Most
consumers are located to the right of the map, where packaging
samples 1 and 4 are located. Therefore, the packages with the
highest acceptance are 1 and 4, followed by 3. The sample
6 obtained the least sensory acceptance. The same graph can be
generated for the results of acceptance of sensory attributes (e.g.,
appearance, aroma, texture, and flavor).
In this protocol, the acceptance test was described using a nine-

point Hedonic Scale and performed in the laboratory. However,
there are many scales that can be used and different locations for
running the test. A variety of useful methods are available to
Fig. 3 Internal preference map based on six edible packaging samples
researchers, providing vital information for edible packaging devel-

opers and marketers alike. As mentioned by Lawless and Heymann
[10], high sensory acceptance does not guarantee that the product
will be a market success. The likelihood of purchase (and most
importantly, the repurchase) depends on the concept, price,
brand, positioning, promotions, advertising, label information,
consumer awareness, nutritional characteristics, and many other
factors. However, sensory appeal is the essential “platform” with-
out which the product is unlikely to succeed [10].
5 Notes
1. The packaging samples must be served randomly, so the effect of

the order of presentation and the residual effect, characterized by
the influence of a treatment in the evaluation of the subsequent
one, is eliminated.
2. Potential consumers are individuals who do not have the
habit to consume a product, but are consumers of the
product in potential. The use of potential consumers is
recommended when the product under evaluation is an
unusual product and thus it does not have an established
consumer Market.
3. Generally, an explanation of the variability of results by the

first principal components (first and second principal com-
ponents, for example) that is greater than 70% is considered
satisfactory for the explanation of the results.
References
1. Otoni CG, Avena-Bustillos RJ, Azeredo HMC, 7. Food Connection (2021) Empresa cria emba-
Lorevice MV, Moura MR, Mattoso LHC, lagem comestı́vel a partir de fibras naturais.
McHugh TH (2017) Recent advances on edi- ht t p s : // w w w. f o o dc o n n e c t i o n . c o m . b r /
ble films based on fruits and vegetables – a empreendedorismo/empresa-cria-embalagem-
review. Compr Rev Food Sci Food Saf 16: comestvel-partir-de-fibras-naturais. Accessed
1151–1169. https://doi.org/10.1111/ 23 apr 2021
1541-4337.12281 8. Della Lucia SM, Minim VPR, Carneiro JDS
2. Daniloski D, Petkoska AT, Lee NA, Bekhit (2018) Análise sensorial de alimentos. In:
AED, Carne A, Vaskoska R, Vasiljevic T Minim VPR (ed) Análise sensorial: estudos
(2021) Active edible packaging based on milk com consumidores, 4th edn. Editora UFV,
proteins: a route to carry and deliver nutraceu- Viçosa, pp 13–49
ticals. Trends Food Sci Technol 111:688–705. 9. Reis RC, Minim VPR (2018) Testes de
https://doi.org/10.1016/j.tifs.2021.03.024 aceitação. In: Minim VPR (ed) Análise senso-
3. Petkoska AT, Daniloski D, D’Cunha NM, rial: estudos com consumidores, 4th edn. Edi-
Naumovski N, Broach AT (2021) Edible pack- tora UFV, Viçosa, pp 69–85
aging: sustainable solutions and novel trends in 10. Lawless HT, Heymann H (2010) Sensory eval-
food packaging. Food Res Int 140:109981. uation of food: principles and practices, 2nd
https://doi.org/10.1016/j.foodres.2020. edn. Springer, New York. 596p
109981 11. Jones LV, Peryam DR, Thurstone LL (1955)
4. Retratos da Economia (2015) Copo de café Development of a scale for measuring soldier’s
comestı́vel é lançado para reduzir impacto food preferences. J Food Sci 20:512–520.
ambiental. https://economia.estadao.com.br/ https://doi.org/10.1111/j.1365-2621.1955.
blogs/retratos-da-economia/copo-de-cafe- tb16862.x
comestivel-reduz-impacto-ambiental/. 12. Peryam DR, Pilgrim FJ (1957) Hedonic scale
Accessed 22 apr 2021 method of measuring food preferences. Food
5. Sociedade Nacional de Agricultura (2016) Technol 1957:9–14
Embalagens comestı́veis. https://www.sna.agr. 13. Meilgaard MC, Civille GV, Carr BT (2016)
br/embalagens-comestiveis/. Accessed Sensory evaluation techniques, 5th edn. CRC
24 apr 2021 Press, Boca Raton. 588p
6. As melhores soluções sustentáveis (2017) 14. Lavine BK, Jurs PC, Henry DR (1988) Chance
Embalagem comestı́vel é feita de algas. classifications by non-linear discriminant func-
https://catracalivre.com.br/as-melhores- tions. J Chemom 2:1–10. https://doi.org/10.
solucoes-sustentaveis/embalagem-comestivel- 1002/cem.1180020103
e-feita-de-algas/. Accessed 22apr 2021
Chapter 21
Consumer Choice Probabilities for Food Packaging

Tarcı́sio Lima Filho, Suzana Maria Della Lucia,
and Valéria Paula Rodrigues Minim
Abstract
Packaging is of extreme importance as it represents the very first contact among consumers and food or
beverage. Packaging therefore plays a silent seller role and stimulates consumers to decide whether to buy
the product or not. In this chapter, we discuss the modified choice-based conjoint analysis (MCBCA), a
quantitative method that has been used to assist in the clarification of consumer behavior, especially when
seeking to analyze the attributes of product packaging guiding consumer choice. The protocol for
determining the consumer choice probabilities for food packaging based on MCBCA is presented. We
intend to assist everyone involved in the process of designing new products, especially in the stages of
developing the marketing strategy, production, and market testing, by studying the modification and choice
of packaging and labels, contributing to increasing the product’s competitiveness in the market.
Key words Modified choice-based conjoint analysis, Non-sensory characteristics, Likelihood of

choice
1 Introduction
Most food and beverage products are sold packaged, with a label
that has been designed, then printed or attached onto it [1]. In this
context, studying the packaging itself is extremely important as it
represents the consumer’s first contact with the product, thus
denoting the primary object for defining the choice and purchase
of the packaged good [2]. Packaging therefore acts as a silent seller
by providing information about the product, which is evaluated by
the consumer to support the decision to whether buy the product
or not.
Thus, it can be concluded that, the decision to purchase a
certain product for the first time usually depends on the extrinsic
information or characteristics, that is, the non-sensory characteris-
tics related to this product, which are normally present on the
packaging [3]. Packaging characteristics can lead the consumer to
https://doi.org/10.1007/978-1-0716-3613-8_21,
349
350 Tarcı́sio Lima Filho et al.
purchase a product, while sensory characteristics confirm the accep-

tance and can determine recurring purchases [4, 5].
Among the information or characteristics presented in the
packaging, the following can be highlighted: Label, brand, price,
material, color, texture, design, format, illustrations, origin, prepa-
ration method, nutritional facts, expiry date, net weight, ingredient
list, health-related label messages, and nutritional claims, among
others. All these characteristics have specific influences on the con-
sumer’s intention to purchase or choose the product. This is
because, from the consumer’s perspective, a food is always asso-
ciated with a packaging and is often selected through the informa-
tion provided [2, 6, 7]. Therefore, packaging plays a vital role in
decision making, as a specific combination of quality attributes
therein presented determines the expected quality. An informed
consumer aggregates knowledge about food from various available
sources and compares it with the information on the product label
[5]. Thus, it is expected that the presence of a well-designed
packaging will have a powerful influence on the formation of the
sensory expectations by the consumer, also influencing the choice
and finally the purchase of the product. The expectation generated
by the information contained in the packaging is particularly impor-
tant because it can either improve or worsen the perception of the
product even before its consumption [8].
While some of this information appearing on packaging is
enforced by the law (e.g., in Brazil, the label must at least display
the product’s name, weight, and use by/best before date), further
information—both textual and graphical—is often added to help
inform the consumer, encourage favorable product expectations,
and enhance the consumption process [1].
Studies have been developed with the aim of evaluating the role
of packaging and/or factors contained in it on consumer behavior,
because, as previously stated, it is of crucial importance to the
choice of product during purchase [2, 6]. The Conjoint analysis
technique has been widely and successfully used to carry out this
type of study, in order to understand the attitudes and behaviors of
consumers toward the packaging of food and beverages.
The Conjoint analysis was developed in the fields of psychome-
try and consumer research, being used as a support to understand
how the consumer evaluates the quality of products. It allows one
to understand how an individual develops a preference for products
or services [9], based on their different characteristics. Its develop-
ment dates from 1964 and its introduction in marketing research
took place through Green and Rao, in 1971 [10]. This technique is
one of the most important tools in assisting product development
and decision making in marketing.
More specifically, the Conjoint analysis aims to investigate the
joint effect of two or more independent variables on the evaluation
of a dependent variable [2]. It is a quantitative method that has
Non-Sensory Characteristics of Food Packaging 351
been used to assist in the clarification of consumer behavior regard-

ing a product, especially when seeking to analyze the attributes of
product packaging on consumer choices and purchase [11]. In
packaging studies, the application of the Conjoint analysis is
based on data collection by combining specific levels of each factor
or characteristic to be studied in the packaging to obtain a set of
different treatments (possible packaging for the product); these
investigated packaging systems are presented to consumers for the
global assessment of preference, purchase intention, or choice
[2]. The use of such a method enables the assessment of the
packaging characteristics that are essential for increasing consumer
intent to purchase or choose a product [11].
Researchers have developed over the years different types of
data collection within the Conjoint analysis, as well as different data
analysis techniques from this type of study. One of the types of
analysis, the ratings-based conjoint analysis, uses data collection to
mark the preference/purchase intention/acceptance of a packaging
or product using scales. The rankings-based Conjoint analysis, in
turn, relies on the ordering of treatments (such as packaging)
according to preference/purchase intention/acceptance to obtain
the data. In the third type, called choice-based, consumers choose a
treatment (e.g., a packaging) from several options instead of assign-
ing notes separately or ordering them [9].
In this chapter, the Modified choice-based conjoint analysis
(MCBCA) is discussed, as presented by Della Lucia [6], Lima
Filho et al. [11], and Carneiro et al. [12]. MCBCA outstands as
the most realistic approach to collect data in the simulation of
consumer purchasing behavior, which can lead to a greater validity
of the results.
In this type of analysis, consumers must choose a packaging
among several alternatives, assembled from a set of factors (the
studied packaging characteristics) and their levels (different types
that the characteristics can take), without having to assign notes of
intention to purchase or rank the packaging treatments, which
makes the data collection protocol closer to the reality of these
consumers. The consumer’s choice behavior is therefore investi-
gated through the so-called “choice-based” Conjoint analysis
[6, 13].
MCBCA allows one to estimate the probabilities of choice
associated with the evaluated packaging, as well as compare the
probabilities of choosing a specific packaging for any two levels of
the same factor or characteristic of which they are constituted
[6, 13]. MCBCA is particularly important when studying the con-
sumer behavior. In this vein, the technique presented in this chapter
is expected to assist those involved in the process of designing new
products, especially in the stages of developing the marketing strat-
egy, production, and market testing, supporting improvements,
modifications, and choices of packaging, therefore contributing to

the product’s commercial competitiveness.
2 Materials
• Packaging samples.
• Table for displaying all samples.
• White light.
• Stopwatch.
• Answer sheet and consent form (see Note 1).
3 Methods
3.1 Factors Related First, it is necessary to determine which characteristics of the pack-
to Packaging and Their aging (factors and their respective levels) are the most relevant to be
Respective Levels evaluated. This is a very important step since the purpose of the
method is to determine how these characteristics will influence the
likelihood of consumer choices.
Therefore, the factors inherent to the packaging, which may
interfere with the probability of choice, must be chosen (see Note
2).
The determination of these characteristics must be carried out
considering the objectives and the experience of the researchers,
considering information from the literature and, or, through Focus
Group sessions (see Note 3).
Table 1 shows examples of factors and their levels, for studies
with packaging, which can be investigated through the MCBCA.
3.1.1 Case Study
Table 1
Examples of factors and levels of food or beverage packaging
Factors Levels
Material Flanders, aluminum, glass, poly(ethylene terephthalate) (PET), or biaxially oriented
polypropylene (BOPP)
Size Small, medium, or large (specifying dimensions)
Volume 350, 600, or 1000 mL
Format Rectangular, square, round, or triangular
Color Red, black, gray, blue, or green
A case study is presented to exemplify the application of the method

and allow a more didactic explanation of all stages of the MCBCA.
In this fictitious study, the effect of beer packaging characteristics
on the likelihood of Consumer choices is investigated. The packag-
ing has three factors with two levels each, namely: material factor,
with levels “glass” and “PET”; volume factor, with levels “350 mL”
and “600 mL”; and color factor, with levels “amber” and “green.”
3.2 Data Collection MCBCA encompasses an experiment with several factors, each with
its own levels. The factors can be qualitative or quantitative vari-
ables. Each treatment (packaging) is obtained by combining the
levels of the factors; therefore, the treatments are obtained by a
factorial array.
The complete profile is the data collection method to be used in
the MCBCA. In the complete profile, each package (treatment) is
formed by the combination of all factors, that is, it is formed by the
combination of a level of each factor.
3.3 Determining the After defining the factors, their levels, and the method of data
Packaging Samples to collection, it is necessary to define which treatments will be ana-
Be Analyzed lyzed by the evaluators. One can adopt the complete factorial or the
fractional factorial.
In the complete factorial, all possible combinations of the levels
of the factors will be analyzed. Therefore, the complete factorial
should be adopted whenever the number of factors and levels to be
evaluated is small. When there are a large number of factors and
levels, fractional factorials should be adopted (see Note 4).
The MCBCA calculates the probability of choosing each pack-
age, seeking to investigate which package has the highest likelihood
of choice. When using fractional factorial, leaving some packaging
samples without evaluation, there is a risk of not analyzing the
packaging that would have the greatest likelihood of consumer
choices. Therefore, it is recommended, whenever possible, to give
preference to the use of the complete factorial array.
3.3.1 Case Study The data will be collected using the complete profile method [15],
and a complete factorial treatment array will be used [12]. There-
fore, eight treatments will be applied, as outlined in Table 2.
3.4 Preparing The packaging samples to be studied must be prepared. Prototypes

Packaging Samples or photographs of the packaging can be used. The use of prototypes
is recommended because it is a better representation of reality
(in three dimensions) when compared with photographs (in two
dimensions).
3.4.1 Case Study The eight packaging samples (treatments) of the case study are
shown in Fig. 1.
Table 2
Treatments under study
Treatment Material Volume (mL) Color

1 Glass 600 Amber
2 Glass 600 Green
3 Glass 350 Amber
4 Glass 350 Green
5 PET 600 Amber
6 PET 600 Green
7 PET 350 Amber
8 PET 350 Green
3.5 Defining the All packaging samples must be presented simultaneously to the eva-
Order of Presentation luators. The order of disposal of the packages must follow a prede-
of the Packaging fined experimental design (see Note 5).
Samples
3.5.1 Case Study Annex 1 shows the design proposed by MacFie et al. [16] to present
eight treatments, which would be the design used in the case study.
There are 48 possible orders for the presentation of eight samples.
All these orders must be considered during the evaluation of the
packaging.
3.6 Defining the In MCBCA, the evaluation of packaging must be carried out by
Number of Evaluators traditional or potential consumers of the product. Assessors do not
need to be previously trained. It is only necessary to explain, on the
day of the analysis, how the evaluators should proceed to analyze
the samples.
The number of evaluators that will carry out the analyses
depends on the number of possible orders of the design and the
number of repetitions defined. Equation 1 should be used to
calculate the number of evaluators.
nevaluators = norder :r ð1Þ
where nevaluators is the number of evaluators; norder is the number of
possible orders of presentation of packaging in the design; and r is
the number of repetitions (see Note 6).
3.6.1 Case Study In the case study, for eight packages, there are 48 possible packag-
ing presentation orders (Annex 1). If the researcher chooses to
perform three repetitions, 144 evaluators (three repetitions ×
48 orders) will be needed to complete the analysis. In this case,
three evaluators will analyze the packaging samples in the same
order of presentation. To facilitate and streamline the analysis
Fig. 1 Example of treatments. (Image elements by vectorpocket on Freepik)
procedure, these three evaluators can carry out the analysis of the
samples at the same time.
3.7 Procedure for 1. Code the packages with random three-digit numbers.
Analyzing the 2. Prepare and print the answer sheet in sufficient numbers for all
Packaging Samples evaluators (one sheet per evaluator) (Fig. 2).
3. Provide pens to fill in the answer sheet.
4. The place where the analysis will be taken must be quiet, with a
pleasant temperature and white light. In the analysis room,
arrange the packages on a table or gondola in the correct
order of presentation. All packaging samples must be displayed
simultaneously, and the order must follow the design of MacFie
et al. [16].
5. In another room, welcome the evaluators who will participate
in the study. The number of repetitions will be the number of
evaluators who will carry out the analysis at the same time.
Therefore, if it was decided to perform three repetitions,
three evaluators should be invited at a time.
6. Deliver, explain, and request a signature on the Free and
Informed Consent Form (see Note 1).
7. Collect information about research participants (see Note 7).
Consider that you want to buy (product name). Please write the product code you would
buy.
Code:
Comments:
Fig. 2 Answer sheet
8. Deliver the answer sheet to consumers (Fig. 2).

9. Explain the analysis procedure: the evaluators (consumers)
must simulate the product purchase process at the supermar-
ket. Evaluators will have 3 min to analyze all packages. At the
end of this time, they must inform the product they would
choose to buy.
10. Clear the doubts of the evaluators.
11. Invite the evaluators to enter the packaging evaluation room
(the packaging must already be arranged in the correct presen-
tation order).
12. Time 3 min.
13. Ask the evaluators to mark, on the answer sheet, the code of the
package they would choose to buy.
14. Repeat steps 4 through 13 until the packaging is arranged in
all presentation orders [16].
3.8 Tabulation of The evaluators choose only one package among those presented. In
Data the tabulation of the results of the sheets, the value 1 is assigned to
the chosen packaging, and the value 0 to the others. To perform the
analysis of the results, the levels of the factors must also be coded.
3.8.1 Case Study The coding of the levels of the case study is shown in Table 3.
3.9 Data Analysis To carry out the MCBCA, the following considerations were made:
,
Let Y ,k = y 1k , y 2k , . . . , y Nk be the vector of answers for the
kth consumer, with:
yjk = 0 for not chosen packaging and yjk = 1 for the chosen package
As each appraiser chooses only one package, Eq. 2 applies:
N
y jk = 1 ð2Þ
j =1
For j = 1, 2, . . ., N treatments (packaging samples) analyzed by

each of the k = 1, 2, . . ., E evaluators.
Equation 3 represents the matrix notation of the model
Y = Xβ ð3Þ
where
Table 3
Coding of the levels of factors (case study)
Factor Level Codification

1—Material 1—Glass 0
2—PET 1
2—Volume 1—600 mL 0
2—350 mL 1
3—Color 1—Amber 0
2—Green 1
Y is the vector of the evaluators’ answers for the analyzed packages.

X is the matrix with the coded values of the factor levels (Table 3).
β is the vector of parameters to be estimated, with only one coeffi-
cient being estimated per factor.
The notation Xjβ is used to indicate packaging j, where
Xjβ = (X1j, X2j, . . ., Xsj)β (Eq. 4):
β = ðβ1 β2 . . . βs Þ0 ð4Þ
where
Xsj represents the level of the sth factor present in the jth treatment.
The coding Xsj = 0, 1, . . ., l - 1 is adopted for l levels.
For two levels, we have Xsj = 0, 1, as done in the coding presented
in Table 3.
Pj is the probability associated with jth packaging, satisfying
Eq. 5:
0 ≤ Pj ≤ 1
with
N ð5Þ
Pj = 1
j =1
The model proposed by McFadden [17], termed multinomial

logit, is adopted to estimate the likelihood of the choice of a
treatment (Eq. 6)
eX j β
Pj = N
ð6Þ
eX j β
j =1
where X is the matrix of encoded values of the factor levels and β is

the vector of estimated parameters through iterative numerical
methods to maximize the likelihood function (L) of the sample
or, similarly, the logarithms of the function L. For more details, we
recommend consulting Agresti [18], who presents an approach to

estimation and inferences using this model.
When applying the MCBCA, the main objective is to estimate
Pj for each package under study. In this way, it is possible to
investigate which packaging is more likely to be chosen by the
consumer.
The MCBCA also allows calculating the effect of choosing a
treatment at one level of a factor over another level of the same
factor (hazard ratio value), according to Eq. 7 [4, 11].
P ðlevel 2Þ
Hazard ration = = e βs ðX level 2 - X level 1 Þ ð7Þ
P ðlevel 1Þ
For the case study, with three factors and two levels each, s = 1,
2, 3 factors, Xlevel 2 = 1 and Xlevel 1 = 0 (according to the encoding
levels of each factor, Table 3).
3.10 Presentation The results can be presented using tables and graphs. As an exam-
and Interpretation of ple, the results of the case study will be presented. It is important to
Results (Case Study) highlight that these data are fictitious, we use them only for didactic
purposes.
The estimated coefficients (β′) and the hazard ratio values are
shown in Table 4.
The volume and color factors had a significant effect on the
consumer choices according to the model used ( p ≤ 0.001). The
material factor showed no significant effect on the consumer evalu-
ation ( p > 0.001) (Table 4).
The hazard ratio value for the material factor is
P ðlevel 2Þ P ðPET materialÞ
Hazard ration = = = 0:931
P ðlevel 1Þ P ðGlass materialÞ
The hazard ratio value for the volume factor is
P ðlevel 2Þ P ð360 mLÞ
P ðlevel 1Þ P ð600 mLÞ
Table 4
Summary of the analysis for estimating the model coefficients by
maximum likelihood
Factor Estimated coefficient (β) Hazard ratio value

Material -0.12401ns 0.931
Volume -1.05313* 2.436
Color -1.52820* 3.449
*Significant according to the chi-square test ( p ≤ 0.001)
ns
Not significant according to the chi-square test ( p > 0.001)
The hazard ratio value for the color factor is

P ðlevel 2Þ P ðGreenÞ
P ðlevel 1Þ P ðAmberÞ
The hazard ratio value is a ratio of estimated probabilities. The
hazard ratio value of 0.931 for the material factor means that the
probability of the consumers choosing a package with the material
“glass” was 1.07 times greater than the probability of them choos-
ing a package with the material “PET.” The value was quite close to
1.0, showing that the effect of the factor was weak (nonsignificant,
as shown in Table 4).
The probability of the consumers choosing a package with the
volume “360 mL” was 2.436 times greater than the probability of
them choosing a package with the volume “600 mL.” The proba-
bility of the consumers choosing a package with the color “green”
was 3.449 times greater than the probability of them choosing a
package with the color “amber.”
Table 5 and Fig. 3 show the congruence between the observed
Choice Probabilities values and those estimated by the MCBCA for
each study treatment.
The results recorded for the MCBCA show that treatment
4 (Fig. 3), referring to the packaging with the characteristics
“glass,” “350 mL,” and “green,” possessed the highest estimated
probability of Consumer choices ( p = 0.1840), followed by treat-
ment 8 ( p = 0.1803, Table 5). Both treatments generated esti-
mated probabilities with rather close values and only differed
regarding the material factor; that is, package 8 was made of PET
material and package 4 was made of glass material. Therefore, these
packages are likely to have a greater positive impact on consumers’
choice of beer.
Table 5
Observed and estimated probabilities by Modified choice-based conjoint
analysis (MCBCA) for the treatments under study
Treatment Observed probability Estimated probability by MCBCA

1 0.0724 0.0582
2 0.1004 0.1065
3 0.1406 0.1368
4 0.2028 0.1840
5 0.0806 0.0821
6 0.0839 0.1031
7 0.1405 0.1490
8 0.1788 0.1803
0.25
0.2
Probability
0.15
0.1
0.05
0
0 1 2 3 4 5 6 7 8
Packaging (treatment)
Probability observed Probability estimated by MCBCA
Fig. 3 Observed and estimated probabilities for the study packaging
4 Notes
1. All research with human beings must be previously approved by

an Ethics Committee. Typically, participants are required to
sign a Consent Form. Consult the Ethics Committee for
Research with Human Beings of the place where the research
will be carried out.
2. Care should be taken to select only the relevant characteristics
because a large number of factors and levels result in a large
number of treatments (packaging samples) to be evaluated,
which makes the analysis more complex for the evaluators and
can affect the reliability of the results.
3. For more information on Focus Groups, read Chap. 16 by
Lawless and Heymann [14].
4. The use of the complete factorial may become impracticable
with the increase in the number of factors and levels due to the
increase in the number of treatments to be analyzed by the
evaluators. For example, in the case study presented, with three
factors and each of them with two levels, we will have eight
treatments. If we add one more factor with two levels, the
number of treatments increases to 16. In this case, the evalua-
tors would have to analyze 16 packaging samples and indicate
which one would be chosen, and it is a more complex and
laborious task than analyzing only eight samples. The increase
in the number of treatments increases the complexity of the
analysis to be made by the evaluators and, consequently,
decreases the reliability of the results obtained.
As a suggestion, the use of three factors, each with two
levels, as in the case study presented, has been quite satisfactory
in packaging choice studies.
5. The experimental design proposed by MacFie et al. [16] is

recommended because it ensures that each sample appears the
same number of times in each position, in addition to being
successive, and preceded, the same number of times by the
other packages. In this way, the effect of the order of presenta-
tion of the packages and the residual effect, characterized by
the influence of a treatment in the evaluation of the subsequent
one, are eliminated.
6. Generally, three repetitions are sufficient to estimate the exper-
imental error.
7. In the sensory evaluation, the evaluators are the instruments of
analysis; therefore, characterizing the profile of the participants
is necessary. The basic information needed is the age, gender,
and frequency of consumption of the product under study.
Some information that can also be requested, according to
the research interest, is the level of education, monthly family
income, occupation, and consumption habits of the product to
be analyzed.
Annex 1 – Presentation Design for 8 Treatments
Order of presentation
Session 1 2 3 4 5 6 7 8
1 5 4 8 7 1 2 6 3
2 8 5 1 4 6 7 3 2
3 4 7 5 2 8 3 1 6
4 3 6 2 1 7 8 4 5
5 7 2 4 3 5 6 8 1
6 6 1 3 8 2 5 7 4
7 2 3 7 6 4 1 5 8
8 1 8 6 5 3 4 2 7
9 1 7 5 8 3 4 6 2
10 7 8 1 4 5 2 3 6
11 6 3 2 5 4 1 8 7
12 8 4 7 2 1 6 5 3
13 5 1 3 7 6 8 2 4
14 2 6 4 3 8 5 7 1
15 4 2 8 6 7 3 1 5
16 3 5 6 1 2 7 4 8
(continued)
Order of presentation
Session 1 2 3 4 5 6 7 8
17 2 6 5 1 7 3 4 8
18 7 5 4 2 8 6 3 1
19 5 2 7 6 4 1 8 3
20 1 3 6 8 2 4 5 7
21 6 1 2 3 5 8 7 4
22 4 7 8 5 3 2 1 6
23 8 4 3 7 1 5 6 2
24 3 8 1 4 6 7 2 5
25 6 1 3 4 8 5 7 2
26 8 3 7 6 2 1 5 4
27 5 2 4 7 1 8 6 3
28 2 7 5 8 4 3 1 6
29 1 4 6 5 3 2 8 7
30 3 6 8 1 7 4 2 5
31 7 8 2 3 5 6 4 1
32 4 5 1 2 6 7 3 8
33 4 1 2 7 5 6 8 3
34 7 6 1 3 4 8 2 5
35 3 8 6 5 7 2 1 4
36 1 7 4 6 2 3 5 8
37 5 2 8 4 3 1 6 7
38 6 3 7 8 1 5 4 2
39 2 4 5 1 8 7 3 6
40 8 5 3 2 6 4 7 1
41 6 1 5 8 3 4 2 7
42 3 5 2 6 7 1 4 8
43 5 6 3 1 2 8 7 4
44 1 8 6 4 5 7 3 2
45 8 4 1 7 6 2 5 3
46 7 2 4 3 8 5 1 6
47 4 7 8 2 1 3 6 5
48 2 3 7 5 4 6 8 1
Source: MacFie et al. [16]

References
1. Simmonds G, Woods AT, Spence C (2019) 9. Hair Junior JF, Anderson RE, Tatham RL,
‘Shaping perceptions’: exploring how the Black WC (1995) Conjoint Analysis. In: Hair
shape of transparent windows in packaging Junior JF, Anderson RE, Tatham RL, Black
designs affects product evaluation. Food Qual WC Multivariate data analysis with readings,
Prefer 75:15–22. https://doi.org/10.1016/j. 4th ed. Englewood Cliss: Prentice Hall,
foodqual.2019.02.003 556–615
2. Della Lucia SM (2005) Conjoint analysis no 10. Steenkamp JBEM (1987) Conjoint measure-
estudo de mercado de café orgânico. Disserta- ment in ham quality evaluation. J Agric Econ
tion (Master in Food Science and 38:473–480. https://doi.org/10.1111/j.
Technology)—Brazil: Federal University of 1477-9552.1987.tb01065.x
Viçosa, Viçosa—MG. 86 pp 11. Lima Filho T, Della Lucia SM, Lima RM,
3. Choi Y, Lee J (2019) The effect of extrinsic Minim VPR (2015) Conjoint analysis as a tool
cues on consumer perception: a study using to identify improvements in the packaging for
milk tea products. Food Qual Prefer 71:343– irradiated strawberries. Food Res Int 72:126–
353. https://doi.org/10.1016/j.foodqual. 132. https://doi.org/10.1016/j.foodres.
2018.08.004 2015.03.023
4. Della Lucia SM, Minim VPR, Silva CHO, 12. Carneiro JDS, Silva CHO, Della Lucia SM,
Minim LA, Ceresino EB (2010) Expectativas Minim VPR (2018) Análise conjunta de
geradas pela marca sobre a aceitabilidade de fatores. In: Minim VPR (ed) Análise sensorial:
cerveja: Estudo da interação entre caracterı́sti- estudos com consumidores, 4th edn. Editora
cas não sensoriais e o comportamento do con- UFV, Viçosa, pp 191–242
sumidor. Bol Cent Pesqui Process Aliment 28: 13. Della Lucia SM, Minim VPR, Silva CHO,
11–24. https://doi.org/10.5380/cep.v28i1. Minim LA (2018) Caracterı́sticas não sensoriais
17893 e o comportamento do consumidor: Conceitos
5. Pinto VRA, Freitas TBO, Dantas MIS, Della e métodos estatı́sticos de avaliação. In: Minim
Lucia SM, Melo LF, Minim VPR, Bressan J VPR (ed) Análise sensorial: estudos com con-
(2017) Influence of package and health-related sumidores, 4th edn. Editora UFV, Viçosa, pp
claims on perception and sensory acceptability 148–190
of snack bars. Food Res Int 101:103–113. 14. Lawless HT, Heymann H (2010) Sensory eval-
https://doi.org/10.1016/j.foodres.2017. uation of food: principles and practices, 2nd
08.062 edn. Springer, New York. 596p
6. Della Lucia SM (2008) Métodos estatı́sticos 15. Green PE, Srinivasan V (1978) Conjoint anal-
para avaliação da influência de caracterı́sticas ysis in consumer research: issues and outlook. J
não sensoriais na aceitação, intenção de compra Consum Res 5:103–123. https://doi.org/10.
e escolha do consumidor. Thesis (PhD in Food 1086/208721
Science and Technology)—Brazil: Federal Uni- 16. MacFie HJ, Bratchell N, Greenhoff K, Vallis
versity of Viçosa, Viçosa—MG. 116 pp LV (1989) Designs to balance the effect of
7. Ballco P, De-Magistris T, Caputo V (2019) order of presentation and first-order carry-
Consumer preferences for nutritional claims: over effects in hall tests. J Sens Stud 4(2):
an exploration of attention and choice based 129–148. https://doi.org/10.1111/j.
on an eye-tracking choice experiment. Food 1745-459X.1989.tb00463.x
Res Int 116:37–48. https://doi.org/10. 17. McFadden D (1974) Conditional logit analysis
1016/j.foodres.2018.12.031 of qualitative choice behaviour. In: Zarembka P
8. Della Lucia SM, Minim VPR (2018) Grupo de (ed) Frontiers in econometrics. Academic,
foco. In: Minim VPR (ed) Análise sensorial: New York, pp 105–142
estudos com consumidores, 4th edn. Editora 18. Agresti A (1990) Categorical data analysis.
UFV, Viçosa, pp 86–112 Wiley, New York. 558 p
Chapter 22
Thermal Performance of Food Packaging Containing Phase

Change Materials
Bianca C. N. Fernandes and Ana S. Prata
Abstract
The thermal performance of thermo-active food packaging is a key factor in ensuring its successful
application. The enthalpy (latent heat) and temperature of fusion of the phase change materials (PCMs)
must be certified to match the payload requirements. The PCM-functionalized material also must be
characterized through kinetic experiments in a near-real application experimental setup. Reliable and
comparable protocols for these characterizations are proposed in this chapter.
Key words Thermo-active packaging, Phase transitions, Temperature control, Thermal conductivity,
Melting point, Thermal storage, Phase change packaging
1 Introduction
Phase change materials or PCMs are employed for developing

temperature control packaging systems for the shipment or con-
sumption of temperature-sensitive goods in various temperature
ranges for food, pharmaceutics, and life science industries
[1–5]. They reliably keep the temperature inside the packaging
stable, preventing it from falling below or exceeding a certain mark.
The PCM’s state transition between liquid and solid states
maintains a constant temperature equal to their melting/freezing
points—mind that these are often temperature ranges instead of
single temperatures. Several PCMs can be found as candidates to
develop packaging systems, chosen to change phases at specific
temperatures to match the payload requirements, but some com-
mercial PCMs are unsuitable for food packaging or cold chain
technologies due to the pungent odors, flammability, reduced
cycling stability, and corrosive nature [6, 7].
Many organic compounds are suitable PCMs as they have
good latent heat (160–190 J g-1), thermal and chemical stabilities,
recyclability, noncorrosiveness, no subcooling properties, and
https://doi.org/10.1007/978-1-0716-3613-8_22,
365
366 Bianca C. N. Fernandes and Ana S. Prata
operating temperature within the major applications for human

consumption. The melting point can be chosen for frozen (-18 °
C), refrigerated (2–8 °C), room temperatures (25 °C), or warm
products for consumption (50 °C). The PCM-based packaging
systems developed nowadays are focused on the cold chain and
large containers [8–13], but the absorption or encapsulation of
the PCM into structured materials has been used to confine the
PCMs, increasing their heat storage capacity. In the case of capsules,
the reduction of particle dimensions allows achieving a high surface
area-to-volume ratio, increasing the heat transfer [14–20], and
hence thermal conductivity [21–24].
Normally, differential scanning calorimetry (DSC) measure-
ments are employed to determine the main thermo-physical prop-
erties of PCM-loaded particles [16–18, 25–27]. Thermal
conductivity, on the other hand, is of extremely difficult measure-
ment due to the phase change and the limitations of available
equipment for measurement, being many times determined (i) by
numerical simulations, which are normally simplified due to the
complexity [28], or (ii) indirectly by determining other thermo-
physical properties, which also do not consider the differences from
the solidification process driven by heat conduction and the melt-
ing process, dominated by the natural convection [29]. Some
experimental alternatives for measuring the thermal conductivity
in the liquid phase include the hot disk [23, 30] and hot wire
[8, 12, 31–34] instruments.
The best way to experimentally determine the thermal perfor-
mance of packaging systems is to reproduce the real systems and
follow the temperature evolution at different points through a data
acquisition system [8, 12, 31–34]. Then, the effective thermal
management solutions will depend on (i) the thermal conductivity
of both phases (i.e., liquid and solid), (ii) the optimum starting
temperature condition determined by the melting point, (iii) the
thermal stability, and (iv) the overall heat gained or lost by the
packaging material, determined by the experimental setup that is
depicted below.
2 Materials
2.1 Packaging 1. PCM particles with a melting point of 79 °C (we showcase

Functionalization packaging materials functionalized with 2-mm-diameter parti-
cles made up of commercial type-3 pale-yellow carnauba wax,
extruded up to the softening point (60 °C), and coated with an
alginate solution in a fluidized bed [14], but this protocol can
be extended to other encapsulated phase change packaging
systems).
Thermo-Active Packaging 367
2. Two polymer (e.g., cellulose) cups with the same dimensions,

one of which with the bottom removed.
3. Plastic adhesive.
2.2 Thermal Stability 1. Thermogravimetric (TG), derivative TG (DTG), and DSC

of the PCM Particles apparatuses.
2. Gases: oxygen and nitrogen.
3. Scale with precision of 10 mg.
4. Standard aluminum sample pans and lids.
2.3 Thermal 1. Pencil, scissors, ruler, and caliper.

Conductivity of the 2. Heat-generating device.
Packaging Material
3. Heavy marble plate with low thermal conductivity
(Heat-Flow-Meter
(k = 2.5 W m-1 °C-1).
Method)
4. T-type thermocouples (precision of ±0.2 °C).
5. Heat flux sensor (D = 0.06 m).
6. Insulator (e.g., ceramic fiber blanket).
7. Voltage variator (single phase, 20 A current, 2.5 kVA capacity).
8. Data Logger (AHLBORN model 2390-5).
2.4 Thermal 1. K-type thermocouples (precision ±0.5 °C).

Performance of the 2. Thermal tape.
Packaging Material
3. Thermostatic bath (DC-6515).
4. Data logger (Testo 177-T4, Brazil).
5. Commercial soybean oil (or other food according to the
application).
3 Methods
3.1 Packaging 1. Distribute uniformly a determined mass of particles (e.g., 10 g

Functionalization of particles) onto the external surface of the cup (e.g., diameter,
4 cm; height, 5 cm)—Fig. 1a.
2. Fix them with the adhesive.
3. Keep the system stand at least for 3 h at 25 °C, for drying.
4. Remove the bottom of the other cup and place the
PCM-containing cup (serves as a shell) inside the bottomless
cup (serves as outer shell)—the PCM particles will be enclosed
within the annular space between the two concentric cups
(Fig. 1b).
Fig. 1 Packaging functionalization: (a) Particles distributed onto the external surface of the cup and the cup
without bottom; (b) final packaging
3.2 Thermal Stability 1. Switch on the calorimeter and allow it to equilibrate for at least
of the PCM Particles 30 min before the analysis; the calibration should be realized
prior the measurements.
2. Adjust the pressure of the oxygen (50 mL min-1) and nitrogen
gas supplies according to the manufacturer’s
recommendations.
3. Set the run parameters: heating rate of 10 °C min-1, from 25 to
600 °C (see Note 1);
4. Accurately weigh the sample pans and lids. An empty reference
set of pan+lid must be employed (see Note 2).
5. Accurately weigh 10 mg of PCM particles in the pans, hermeti-
cally seal them, and record the final mass (pan + sample + lid)
(see Note 3).
6. Enter the sample information (e.g., mass of particles, name) in
the acquisition software.
7. Retrieve raw data from the experiment and treat as follows:
(a) Perform peak integration in order to obtain the values for
enthalpy of fusion (ΔH) (see Note 4).
(b) From the endothermic peak, the minimum point (far
from baseline) is taken as the melting temperature,
Tm. For mixtures, this point defines the liquidus curve
in the phase diagram (see Note 5).
(c) From the simultaneous TG and DSC apparatus, TG/
DTG curves, it is taken the thermal stability and degrada-
tion temperature of the material, i.e., onset temperatures
corresponding to the temperature to which 5% of the mass
of the sample has evaporated or decomposed (see Note 6).
Fig. 2 Schematic experimental setup
3.3 Thermal 1. Shape the packaging material into 0.04-m-side squares with
Conductivity of the homogeneous thickness (see Note 7).
Packaging Material 2. Measure sample thickness at least three different points and
(Heat-Flow-Meter calculate the average value.
Method) 3. Calibrate the T-type thermocouples at five different tempera-
tures (e.g., 0, +15, +25, +70, and +90 °C).
4. Place the samples between the heat-generating device on the
bottom and the marble plate on top to compress the sample set
and ensure contact (Fig. 2).
5. Install thermocouples at the interfaces (bottom and top) of the
sample with the paper, and the flowmeter (on the top).
6. Enclose the system in a thermal insulator to eliminate lateral
heat losses and guarantee the heat flow in the axial direction of
the sample.
7. Turn on the heat-generating device using a voltage variator
until a steady flow is reached (constant power supply of
700 W).
8. Record the data of temperature using the data logger coupled
to the heat flux sensor and the thermocouples; the temperature
is constant at temperatures around Tm.
9. Validate the measurement system using reference materials (see
Note 8).
10. Calculate the thermal conductivity (k) of the sample [W m-1 °
C-1] that comes from unidimensional and steady-state Fourier
equation (Eq. 1). From a known electric power supply (q) and
dT, the thermal conductivity (k) is determined.
e:q
k= ð1Þ
dT
where dT is the temperature difference at the top and the bottom of

the sample [°C]; q is the heat flow [W m-2]; and e is the thickness
[m]. Uncertainty of experiments was found to be in the range of
±2.5%.
3.4 Thermal 1. Calibrate K-type thermocouples at five different temperatures

Performance (e.g., 0, +15, +25, +70, and +90 °C).
3.4.1 Experimental Setup 2. Place them in different positions at the middle height of the
package: inner surface, external surface, geometric center, and
one in the external environment (see Note 7).
3. Attach them with a thermal tape to ensure the contact with
the wall.
3.4.2 Activation of PCM 1. Place water in a thermostatic bath (DC-6515) and keep the
in Empty Cups temperature constant at 90 °C.
2. Put the empty cup in a way that the water reaches almost the
border of the cup (Fig. 3).
3. Record the temperature evolution through a Datalogger by
measuring the temperature every 3 s. When the thermocouple
placed on the internal wall reaches 90 °C, keep the system in
the water for an additional 20 min.
4. Remove the cup from the bath and place it at room tempera-
ture (ca. 25 °C).
5. Record the temperature until it reaches 40 °C, i.e., far from the
phase transition temperature of carnauba wax (79 °C).
6. Repeat the steps above with the control (double cups without
PCM particles).
3.4.3 Activation of PCM 1. Fill the volume of the cup with commercial soybean oil.
in Filled Cups 2. Employ an external resistance (127 V, titanium) into the oil for
heating it to ca. 100 °C (Fig. 4).
3. Remove the external resistance from the cup.
4. Record the temperature of cooling through a Datalogger

(Testo 177-T4, Brazil) by measuring the temperature every 3 s.
5. Record the temperature until it reaches 25 °C, i.e., far from the
phase transition temperature of carnauba wax (79 °C).
6. Repeat the steps above with the control (double cups without
PCM particles).
4 Results
1. DSC is performed in a restricted temperature range to high-

light the phase transition thermal event (the crystallization
peak) of the PCM.
2. TG is useful for obtaining information on the physicochemical
properties of different materials and its decomposition as a
function of temperature.
3. The PCM application depends on the thermal conductivity.
The greater the conductivity of the PCM, the greater the
heat transfer from the particles to the packaging.
4. Comparing the behavior of the temperature in the empty cups
and in the cups with oil, it is possible to verify the similarity in
the cooling curves. Besides that, it is possible to observe that
the packaging maintains the heat in the interior for a longer
period when compared to the cups without PCM.
5 Notes
1. This rate was employed because of its good compromise

between time and signal-to-noise ratio. Rates varying from
0.5 to 10 °C min-1 were reported [35–38]. The operator
may decrease it for more accurate signals. The temperature
range may also be shifted depending on the PCM system and
the target application.
2. Check the compatibility of the pan material with the PCM.

3. Both the mass of the empty pan and the poured sample have
been measured using a scale with a maximum tolerance of
±0.001 mg. This weight is suggested and should cover the
bottom of the pan to ensure proper heat transfer. The heating
rate is reported to be dependent on the mass and type of sample
[39]. Larger amounts of particles improve the performance of
measurement. If the sample is liquid, a syringe, or a disposable
pipette (made of glass or plastic) can be used to measure the
weight in the pan. Estimative value of 5 μL. Thermal degrada-
tion may be higher under different exposition conditions.
Open pan may be employed in nitrogen/oxygen mixture or
inert atmosphere to evaluate the thermal behavior of
samples [35].
4. Enthalpy of transition is obtained by integrating the area
formed under the peak by placing a stable baseline in the
endothermic event.
5. The melting point should also be taken from the interception
of the extrapolated slope of the melting curve and the baseline
of the peak (onset temperature—Tonset). Tonset is normally
employed for pure substances. The same procedure can be
used for the crystallization temperature in the exothermic
peak for the evaluation of the subcooling effect.
6. The mass loss is determined by the difference in mass observed
by the thermal event observed. The limit for this quantification
is established by tracing two tangent horizontal lines over the
TG curve, which defines the region of the thermal event. The
first derivative of the mass loss curve, DTG curve, is used to
locate the thermal event [36].
7. This dimension is suggested. Alternatively, place the PCM
particles in a set of two external layers of paperboard or other
packaging material. Transient hot wire (THW) method has
been employed in the literature [37].
8. The Heat-Flow-Meter method is suitable for thermal conduc-
tivity determinations of packaging material, but it is always
important that the experimental setup is validated. This can
be done by repeating the procedure described in Subheading
3.3 using materials featuring the same thickness as the analyzed
sample as well as known thermal conductivities. For example,
polyurethane foam (low conductivity: 0.035 W m-1 K-1) and
rigid poly(vinyl chloride) films (intermediate conductivity:
0.19 W m-1 K-1).
9. Thermocouples must be employed to measure the environ-
ment (ca. 25 °C) and the bulk temperature of the liquid as
well. Attach them with a thermal tape to ensure the contact
with the wall.
References
1. Elliott MA, Halbert GW (2005) Maintaining 14. Hawlader MNA, Uddin MS, Khin MM (2003)
the cold chain shipping environment for Microencapsulated PCM thermal-energy stor-
Phase I clinical trial distribution. Int J Pharm age system. Appl Energy 74:195–202
299:49–54 15. Paulo BB, Andreola K, Taranto O et al (2019)
2. Singh S, Gaikwad KK, Suk Y (2018) Phase Coating approach for a Phase Change Material
change materials for advanced cooling packag- (PCM). Powder Technol 341:147–156
ing. Environ Chem Lett 16:845–859 16. Chalco-Sandoval W, Fabra MJ, López-Rubio A
3. Kahwaji S, Johnson MB, Kheirabadi AC et al et al (2015) Optimization of solvents for the
(2017) Fatty acids and related phase change encapsulation of a phase change material in
materials for reliable thermal energy storage at polymeric matrices by electro-hydrodynamic
moderate temperatures. Sol Energy Mater Sol processing of interest in temperature buffering
167:109–120 food applications. Eur Polym J 72:23–33
4. Carson JK, East AR (2017) The cold chain in 17. Chalco-Sandoval W, Fabra MJ, Lopez-Rubio A
New Zealand—a review. Int J Refrig 87:185– et al (2017) Use of phase change materials to
192 develop electrospun coatings of interest in food
5. Liu M, Saman W, Bruno F (2012) Review on packaging applications. J Food Eng 192:122–
storage materials and thermal performance 128
enhancement techniques for high temperature 18. Pérez-mMasiá R, López-Rubio A, Lagarón JM
phase change thermal storage systems. Renew (2013) Development of Zzein-based heat-
Sust Energ Rev 16:2118–2132 management structures for smart food packag-
6. Sharma A, Tyagi VV, Chen CR et al (2009) ing. Food Hydrocoll 30:182–191
Review on thermal energy storage with phase 19. Mccann JT, Marquez M, Xia Y (2006) Melt
change materials and applications. Renew Sust coaxial electrospinning: a versatile method for
Energ Rev 13:318–345 the encapsulation of solid materials and fabri-
7. Cabeza LF, Roca J, Nogués M et al (2002) cation of phase change nanofibers. Nano Lett
Immersion corrosion tests on metal-salt 6:2868–2872
hydrate pairs used for latent heat storage in 20. Johnston JH, Grindrod JE, Dodds M et al
the 48 to 58 °C temperature range. Mater (2008) Composite nano-structured calcium
Corros 53:902–907 silicate phase change materials for thermal
8. Oró E, Cabeza LF, Farid MM (2013) Experi- buffering in food packaging. Curr Appl Phys
mental and numerical analysis of a chilly bin 8:508–511
incorporating phase change material. Int J 21. Wang Y, Zhang X, Ji J et al (2019) Thermal
Refrig 58:61–67 conductivity modification of n-octanoic acid-
9. Lu YL, Zhang WH, Yuan P et al (2010) Exper- myristic acid composite phase change material.
imental study of heat transfer intensification by J Mol Liq 288:111092
using a novel combined shelf in food refriger- 22. Sheikholeslami M, Keshteli AN, Babazadeh H
ated display cabinets (Experimental study of a (2020) Nanoparticles favorable effects on per-
novel cabinets). Appl Therm Eng 30:85–91 formance of thermal storage units. J Mol Liq
10. Oró E, Miró L, Farid MM et al (2012) Improv- 300:112329
ing thermal performance of freezers using 23. Wang J, Xie H, Xin Z et al (2010) Enhancing
phase change materials. Int J Refrig 35:984– thermal conductivity of palmitic acid based
991 phase change materials with carbon nanotubes
11. Oro E, de Gracia A, Cabeza LF (2013) Active as fillers. Sol Energy 84:339–344
phase change material package for thermal pro- 24. Sahan N, Paksoy HO (2014) Thermal
tection of ice cream containers. Int J Refrig 36: enhancement of paraffin as a phase change
102–109 material with nanomagnetite. Sol Energy
12. Hoang HM, Leducq D, Pérez-Masia R et al Mater Sol 126:56–61
(2015) Heat transfer study of submicro- 25. Chalco-Sandoval W, Fabra MJ, López-Rubio A
encapsulated PCM plate for food packaging et al (2014) Electrospun heat management
application. Int J Refrig 52:151–160 polymeric materials of interest in food refriger-
13. Alzuwaid F, Ge YT, Tassou SA et al (2015) The ation and packaging. J Appl Polym Sci 131:
novel use of phase change materials in a refri- 40661(1–11)
gerated display cabinet: an experimental inves- 26. Chalco-Sandoval W, Fabra MJ, López-Rubio A
tigation. Appl Therm Eng 75:770–778 et al (2015) Development of polystyrene-based
films with temperature buffering capacity for 33. Vennapusa JR, Konala A, Dixit P et al (2020)
smart food packaging. J Food Eng 164:55–62 Caprylic acid based PCM composite with
27. Pérez-Masiá R, López-Rubio A, Fabra MJ et al potential for thermal buffering and packaging
(2014) Use of electrohydrodynamic processing applications. Mater Chem Phys 253:123453
to develop nanostructured materials for the 34. Kozak Y, Farid M, Ziskind G (2017) Experi-
preservation of the cold chain. Innov Food Sci mental and comprehensive theoretical study of
Emerg Technol 26:415–423 cold storage packages containing PCM. Appl
28. Wang G, Wei G, Xu C et al (2019) Numerical Therm Eng 115:899–912
simulation of effective thermal conductivity 35. Haillot D, Bauer T, Kröner U et al (2011)
and pore-scale melting process of PCMs in Thermal analysis of phase change materials in
foam metals. Appl Therm Eng 147:464–472 the temperature range 120–150 °C. Thermo-
29. Karim Y el, Grosu Y, Faik A et al (2019) Inves- chim Acta 513:49–59
tigation of magnesium-copper eutectic alloys 36. Müller L, Rubio-Pérez G, Bach A et al (2020)
with high thermal conductivity as a new PCM Consistent DSC and TGA methodology as
for latent heat thermal energy storage at basis for the measurement and comparison of
intermediate-high temperature. J Energy Stor- thermo-physical properties of phase change
age 26:100974 materials. Materials 13:1–20
30. Xie H, Gu H, Fujii M et al (2006) Short hot 37. Harish S, Orejon D, Takata Y et al (2015)
wire technique for measuring thermal conduc- Thermal conductivity enhancement of lauric
tivity and thermal diffusivity of various acid phase change nanocomposite with gra-
materials. In: Measurement science and tech- phene nanoplatelets. Appl Energy 80:205–211
nology. Institute of Physics Publishing, pp 38. Pielichowski K, Flejtuch K (2002) Differential
208–214 scanning calorimetry studies on poly(ethylene
31. Lu W, Tassou SA (2013) Characterization and glycol) with different molecular weights for
experimental investigation of phase change thermal energy storage materials. Polym Adv
materials for chilled food refrigerated cabinet Technol 13:690–696
applications. Appl Energy 112:1376–1382 39. Castellon C, Gunther E, Mehling H et al
32. Oró E, Miró L, Farid MM et al (2012) Thermal (2008) Determination of the enthalpy of
analysis of a low temperature storage unit using PCM as a function of temperature using a
phase change materials without refrigeration heat-flux DSC—a study of different measure-
system. Int J Refrig 1709–1714 ment procedures and their accuracy. Int J
Energy Res 33:1258–1265
INDEX
A B
Absorbance Barrier properties
ORAC (see Oxygen Radical Absorbance Capacity barrier to gases ........................................................ 225
(ORAC)) barrier to microorganism........................................ 233
Acceptance (sensory) ........................................... 337–346 barrier to moisture ......................................... 109, 208
Acetyl tributyl citrate ...................................................... 76 Bioactive
Active (differs from bioactive) compound ............................................................... 326
agent ............................................................... 311–323 packaging
packaging .............................................. 116, 280, 282, prebiotic.....................................................325–334
303, 311, 312, 314–316 probiotic ....................................................325–334
Adhesive...................................................... 75, 79, 83, 88, Bioassimilation ................................................................ 28
121, 122, 129, 186, 199, 367 Biobased packaging....................................................... 3, 4
Affective method ........................................................... 339 Biodegradable packaging/plastic
Agglomeration ............................................ 172, 174, 175 oxo-biodegradable packaging
Alternating least square (ALS) ............................ 107, 184 (see Oxo-biodegradable)
Analysis of variance (ANOVA) ............................ 343–345 Biodegradation
Antimicrobial activity in compost .................................................... 29, 34, 57
antibacterial assays in marine water......................................... 5, 12–14, 29
disk susceptibility test ....................................... 281 in soil.................................................. 5, 11, 29, 30, 34
Kirby-Bauer disk diffusion test........281, 282, 288 Biodeterioration/bioerosion ....................................28, 29
antifungal assays Biofilm ................................................................ 28, 65, 69
Agar diffusion test............................261, 265–267 Biofragmentation ......................................................28, 29
disk diameter test .............................261, 265–267 Bisphenol ...............................................91, 119, 121, 127
film surface inoculation test..................... 262, 267
plate counting germination test ...................... 262, C
265, 268 Calorimetry ..................................................................... 23
Antioxidant activity
Carbon dioxide (CO2)
ABTS assay/TEAC ....................................... 294, 295, headspace (see Modified atmosphere packaging)
297–300, 303–305 permeability ............................................225–227, 230
DPPH assay ................................... 295–298, 301–303
respiration (see Respirometry)
FRAP assay .................................... 295–296, 304–306 transmission rate ..................................................... 221
ORAC assay .......................... 296, 300–302, 306–307 Cellular structure ................................................... 28, 170
Arrhenius ....................................................................... 221
Cellulose, see Natural polymer
Assimilation, see Bioassimilation Chemometrics .............................. 67, 184, 187, 190, 191
Attribute (sensory) Chitosan, see Natural polymer
appearance ............................................. 340, 344, 345
Choice probability/consumer choice ................. 349–359
aroma ..................................................... 340, 344, 345 Chromatography
color ................................................................ 168, 340 combustion ion chromatography (CIC) ..... 103, 104,
flavor ..............................................168, 340, 344, 345 107–108
overall impression.................................. 340, 344, 345
gas chromatography (GC).......................... 30, 67, 85,
taste ................................................................. 168, 340 88, 121, 323
texture............................................168, 340, 344, 345 liquid chromatography (LC) ...................88, 102, 124
size-exclusion chromatography (SEC)..................... 29
https://doi.org/10.1007/978-1-0716-3613-8,
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Science+Business Media, LLC, part of Springer
Nature 2024
375
FOOD PACKAGING MATERIALS: CURRENT PROTOCOLS
376 Index
Clausius–Clapeyron equation ....................................... 211 film ................................................................... 19, 205,
Compostable packaging................................................ 101 259–276, 325–334, 338
Conjoint analysis .................................................. 350, 351 packaging ....................................... 140, 280, 337–346
Consumer Emulsion...............................................70, 109, 132, 261,
acceptance (see Acceptance) 263–265, 271, 274, 275
behavior .......................................................... 350, 351
choice .............................................................. 349–359 F
Contaminant ........................................58, 68, 76, 77, 84, Ferric reducing antioxidant power (FRAP),
100, 102, 115–132, 316, 327, 333
see Antioxidant activity
Cramer’s rules ................................................................. 91 Fiber............................................................ 22, 60, 65, 66,
Culture medium............................ 30, 32, 34, 36, 39–45, 68, 70, 87, 88, 99, 101, 105, 106, 108, 171,
47, 48, 50, 51, 139, 141, 143, 144, 146, 154,
173, 174, 338, 367
158, 159, 163, 236, 237, 239, 240, 263, 265, Fick’s law ..................................................... 207, 208, 220
282, 283, 285, 326, 327, 329, 330, 334 Flexible packaging................................................ 234, 235
Cumulative mass ........................................................... 319
Foam ............................................... 22, 65, 102, 170, 372
Curve resolution ......................................... 184, 188, 189 Food simulant ........................................ 85–90, 117–119,
121–126, 128–131, 314–319, 322
D
Foodborne pathogen .......................................... 246, 252,
Damage.......................................... 62, 63, 116, 127, 137, 257, 279, 280, 288
150, 151, 161, 168, 170, 177, 337 Food-contact material (FCM).......................... 75–79, 81,
Data 83, 90, 101, 115–131
acquisition ............................ 177, 185, 199, 222, 366 Formaldehyde............................................................60, 77
binning..................................................................... 190 Fruits and vegetables ................................. 219, 260, 261,
processing ....................................................... 170, 172 268, 270, 274, 279
Defect/imperfection ....................................167–178, 199 Fungus/fungi ....................................................4, 27, 239,
Degradation................................................ 16, 17, 22, 23, 259, 261, 265, 267–269, 272, 273, 275
27–29, 58, 62, 65, 68, 76, 77, 100, 109, 120,
293, 314, 337, 368, 372 G
Detector........................................... 67, 82, 88, 107, 127, Gamma radiation, see Radiation
169, 170, 172, 173, 178, 185–187, 190, 199 Generally recognized as safe (GRAS) .........................116,
Dialkyl phosphate esters (diPAPs) ...................... 102, 103
268, 269
Differential scanning calorimetry (DSC), see Calorimetry Grayscale ............................................................... 172, 175
Diffusion Grease resistance .................................................. 108, 109
anomalous................................................................ 315
coefficient .......................................77, 118, 121, 206, H
209, 220, 222–224, 312
diffusional exponent ...................................... 319, 320 Headspace................................................... 12, 14, 23, 89,
diffusivity .................................................76, 207, 221, 219, 246, 247, 249, 253, 256, 280, 311, 313
313–315, 320, 323 Heat transfer................................................ 366, 371, 372
Digestion Hedonic scale/score ............................................ 337–346
anaerobic................................................ 5, 6, 9, 10, 15 Henry’s law .......................................................... 206, 207
chemical ............................................................ 69, 132 Homogeneity .............................................. 194, 197, 198
enzymatic.......................................................... 62, 159 Hydrophilic ........................................106, 132, 205–217,
in vitro ....................................................327, 331–333 222, 227, 228, 230, 234, 303, 312
model ....................................................................... 331 Hydrophobic ................................................ 99, 105, 106,
Disintegration..................................................7, 8, 11–13, 109, 208, 294, 301, 312
29, 30, 303, 312 Hypercubes..........................................185, 187, 188, 191
Distribution map.................................193, 194, 197, 198 Hyperspectral imaging (HSI) ............... 84, 184–187, 190
DPPH assay, see Antioxidant activity
I
E Imperfection, see Defect
Edible Inhibition zone ................................................... 265, 275,
coating ....................................................259–276, 282 281, 282, 288–290
Intelligent packaging .................................................... 116
Index 377
Intentionally added substances (IAS) .....................75–93, Modified choice-based conjoint analysis
101, 120, 149 (MCBCA) ................. 351–354, 356, 358, 359
Interactance .......................................................... 186, 187 Multivariate calibration................................................. 189
Interface............................. 170, 171, 199, 253, 313, 369 Multivariate curve resolution (MCR) .......................... 184
K N
Kinetics Nanotechnology
microbial growth kinetics .............................. 247–249 nanoparticle ............................ 83, 105, 121–132, 261
mineralization/biodegradation kinetics/rate.... 16–17 nanotube......................................................... 121, 290
release kinetics ...................... 280, 315, 319, 320, 322 Natamycin............................................314, 316–321, 323
Natural polymer, see Polymers
L Near-infrared spectroscopy (NIR), see Spectroscopy
Label ............................................................. 77, 145, 168, Noise ...........................................172, 174, 177, 192, 341
Non-intentionally added substances (NIAS)..........75–93,
214, 284, 346, 349, 350
Lactic acid bacteria (LAB) ............................................ 247 101, 116, 120, 128, 149
Lambert–Beer law ......................................................... 169 Non-sensory characteristics .......................................... 349
Layer Nonylphenols (NPs) ....................................................... 78
monolayer .................................................86, 158, 159 Nutrient broth................................... 153, 155, 156, 235,
multilayer .................................................86, 109, 117, 236, 239, 240
122, 129, 220, 245, 313
O
Likelihood of choice ..................................................... 353
Line scanning ................................................................ 187 Oleophobic, see Hydrophilic
Listeria ......................................................... 247, 255, 290 Orthoslice .................................................... 171, 174, 176
Overall migration, see Migration
M Oxo-biodegradable ...................................................84, 85
Oxygen
Macropixel ............................................................ 197, 198
Mass transfer...........................................76, 83, 117, 207, oxygen permeability ....................................... 219, 230
211, 216, 247, 313, 314, 321 oxygen radical absorbance capacity
McFarland standard ..................283, 284, 286, 289, 327, (ORAC) .................... 294, 300–302, 306–307
oxygen scavenger..................................................... 311
330, 333
Microbial growth kinetics, see Kinetics oxygen transmission rate ........................................ 229
Micronucleus test ................................................. 154, 158
P
Microorganism-proof packaging.................................. 240
Microplastics Paperboard ................................................... 83, 108, 109,
primary.......................................................... 57, 58, 65 115, 125, 168, 170, 372
secondary .............................................................58, 65 Partial least squares (PLS) ................................... 189, 193
Microscopy Partition coefficient.................... 118, 313, 314, 321–323
confocal microscopy.................................................. 84 Pellet ............................................................. 7, 10, 14, 30,
electron microscopy ................................................ 168 60, 65, 84, 160, 200, 328
fluorescence microscopy ........................................... 65 Permeability
infrared microscopy................................................... 30 carbon dioxide (CO2) (see Carbon dioxide
optical microscopy ..................... 60, 65, 71, 168, 264 permeability)
Migration oxygen (see Oxygen permeability)
overall migration limit (OML) ........................ 90, 118 water vapor (see Water vapor permeability)
specific migration limit (SML) ........... 76, 78, 90, 131 Permeance ................................................... 208, 215, 228
Mineral oil hydrocarbons (MOH) Permeation .......................................... 219–230, 233–240
mineral oil aromatic hydrocarbons Phase change material (PCM)............................. 365–372
(MOAH)............................................... 83, 125 Phthalate ..........................................................38, 76, 120,
mineral oil saturated hydrocarbons (MOSH) ......... 83 121, 123, 128
Mineralization ....................................... 16–17, 22, 28–30 Pixel
Misalignment................................................................. 172 dead.......................................................................... 190
Modified atmosphere packaging (MAP) ....................116, size ........................................................................... 188
219, 245–257
378 Index
Poly- and perfluorinated alkyl substances Q
(PFAS).................................. 99–109, 121, 149
Polymer Quantitative segmentation data ................................... 173
natural polymer .............................................. 260, 338 Quasi-isostatic method ........................................ 223, 224
cellulose .............................................7, 10, 14, 19,
R
21, 23, 32, 43, 50, 66, 171, 185, 186, 189,
196, 198, 220, 282, 338, 367 Radiation
chitin .................................................................. 338 gamma ..................................................................... 236
polysaccharide.......................................... 109, 110, ionizing .................................................................... 236
230, 260, 326, 338 optical ............................................................. 186, 187
starch....................................................... 50, 88, 89 ultraviolet (UV)....................................................... 236
zein...............................................................19, 109 X-ray......................................................................... 172
poly(2,6-diphenyl-p-phenylene oxide) .................. 132 Raw data ............................................................... 189, 368
poly(ethylene terephthalate) (PET) ........................ 63, Ready-to-eat (RTE) food ........................... 252, 253, 255
69, 77, 83, 121, 123, 352, 353, 357, 359 Reconstruction ........................................... 169, 171, 172,
poly(lactic acid)/polylactide (PLA) ....................3, 16, 174, 177, 178
17, 31–33, 63, 109, 110 Recycling ...................................................................77, 92
poly(vinylidene fluoride) (PVDF) .......................... 105 Reflectance.................................... 67, 186, 187, 189, 200
polyamide ................................... 60, 69, 90, 128, 222 Region of interest (ROI) .................................... 172, 175,
polybutylenesuccinate (PBS) ........................... 17, 109 178, 189, 190
polycaprolactone (PCL).............................17, 18, 109 Relative humidity (RH) .........................11, 29, 125, 168,
polydimethylsiloxane (PDMS) ............. 109, 225, 226 199, 206, 207, 211–213, 216, 221, 222, 227,
polyetheretherketone (PEEK)................................ 108 228, 230, 265, 269
polyethylene Release
high-density (HDPE) .......................... 36, 77, 265 mechanism ............................................................... 313
low-density (LDPE)........................17, 20, 36, 39, profile ......................................................319, 321–323
77, 121, 171, 290 rate .................................................................. 312, 313
polyhydroxyalkanoate (PHA) ................................. 109 Rendering ............................................170, 172, 175, 178
polyhydroxybutryrate (PHB) ....................17, 18, 109 Reshaping ............................................................. 194, 197
polypropylene .............................................. 36, 37, 41, Respirometry
49, 77, 89, 121, 128 automated.................................................................. 30
polystyrene Bartha ........................................................... 34, 46–50
expanded polystyrene (EPS).................. 77, 88, 89 nonautomated ..................................................... 27–54
polytetrafluoroethylene (PTFE) ...... 66, 70, 108, 186, Risk assessment .......................................... 79, 82, 90–92,
187, 194, 197–199 131, 246, 252, 256
Pore.........................................................28, 70, 132, 142,
169, 172, 173, 175, 234, 235 S
Porosity............................................................... 5, 66, 170
Salmonella/Escherichia coli microsome assay
Positive list...................................... 90, 91, 120, 123, 131
(Ames test) ..................................150, 153–159
Postharvest disease control........................................... 260
Saturated saline solution ...................................... 227, 230
Powder................................................................. 7, 10, 14,
Seal/sealing .............................................. 34, 40, 46, 128,
23, 32, 58, 124, 200, 264
168, 205, 212, 213, 222, 233, 237, 248, 368
Power law .................................................... 313, 319, 320
Segmentation ............................. 170–172, 174, 176, 343
Prebiotic, see Bioactive packaging
Sensor ................................................... 12, 103, 105–107,
Predictive microbiology............. 246–248, 252, 255, 256
227–229, 367, 369
Preference mapping ............................................. 343–345
Sensory
Pretreatment......................................................63, 86, 87,
acceptance (see Acceptance)
103, 169, 187, 189–192, 195
analysis/evaluation......................................... 339, 342
Principal component analysis (PCA)............................ 343
attribute (see Attribute)
Probiotic, see Bioactive packaging
Shelf-life ................................................................. 86, 116,
Produce..............................................................28, 50, 51,
119, 167, 168, 177, 205, 219, 220, 233, 240,
53, 58, 83, 84, 99, 150, 158, 160, 169, 259–
245–257, 311
276, 304, 307
Short times model................................................ 320, 321
Projection ............................................169, 170, 174, 177
Simulant, see Food simulant
Index 379
Single-use packaging..................................................... 338 immunotoxicity ....................................................... 100
Sinogram............................................................... 169, 170 mutagenicity ............................................. 78, 149–163
Smart label..................................................................... 116 phytotoxicity................................................... 7, 11, 13
Smart packaging, see Intelligent packaging Transflectance ....................................................... 186, 187
Smoothing ................................................... 172, 191, 192 Transmission rate
Solubility.............................................................. 206, 207, carbon dioxide (see Carbon dioxide transmission rate)
220–224, 230, 261, 313 oxygen (see Oxygen transmission rate)
Spatial dimension ......................... 67, 184, 189, 190, 197 water vapor (see Water vapor transmission rate)
Specific migration........................................ 118, 119, 132 Transmittance .................................................67, 186, 187
Spectral dimension ..............................184, 190, 191, 193 Trialkyl phosphate esters (triPAPs) .............................. 102
Spectroscopy/spectrometry Trolox equivalent antioxidant capacity (TEAC),
infrared (vibrational) ................................................. 67 see Antioxidant activity
nuclear magnetic resonance (NMR) ............... 29, 121 Turbidity.............................................................. 235, 237,
Raman (vibrational) .................................................. 67 240, 284, 286
SERS .....................................................82–84, 92, 105 Two-dimensional (2D) slice ......................................... 173
X-ray.....................................................................59, 71
Spikes ............................................................................. 191 V
Spoilage microorganisms .................................... 257, 260,
van’t Hoff relationship.................................................. 221
279, 280, 283, 288 Vegetable oil ...................................................87, 119, 132
Standard curve............................................ 297, 303–305, Vegetables, see Fruits and vegetables
307, 314, 317, 319 Volatile organic compound (VOC) ............................... 85
Starch, see Natural polymer
Volume of interest (VOI) ............................172–176, 178
Steam sterilization ......................................................... 236
Storage temperature................................... 127, 246, 249, W
252–254, 257, 269, 270
Surfactant................................................................ 79, 106 Water
Swelling.......................................312, 313, 315, 320, 322 activity............................................227, 253, 274, 315
repellency ........................................................ 109, 110
T resistance......................................................... 109, 110
vapor permeability (WVP) .....................109, 205–217
Tetrazolium salt vapor pressure (WVP)................... 208–211, 215–217
MTS assay ................................................................ 141
vapor transmission rate (WVTR) ................. 110, 208,
MTT assay ............................................................... 140 215, 216
WST-1 assay............................................................. 141 Wickerham card.................................................... 283, 284
XTT assay................................................................. 141
Thermal conductivity..........................366, 367, 369–372 X
Thermogravimetry (TG) .................................... 168, 367,
368, 371, 372 X-ray
Three-dimensional (3D) mapped microstructure....... 174 attenuation .............................................170–173, 176
Three-dimensional (3D) reconstruction ..................... 170 diffraction ................................................................ 132
Three-dimensional (3D) structural characterization .. 169 histogram........................................................ 172, 175
Threshold of toxicological concern (TTC) ........... 90, 91, spectroscopy (see Spectroscopy)
122, 127, 236, 237, 239 tomography (see Tomography)
Thresholding segmentation................................. 172, 175 See also Radiation
Tomography ......................................................... 169, 171
Z
Toxicity
cytotoxicity ............................. 78, 137–146, 157, 163 Zein, see Natural polymer
genotoxicity .............................................. 78, 149–163

Food Packaging Materials

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Food Packaging Materials

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Food Packaging Materials

Uploaded by

Copyright:

Available Formats

Methods and Protocols

Caio Otoni Editor

For further volumes:

ISSN 2662-950X ISSN 2662-9518 (electronic)

Printed on acid-free paper

Campinas, Brazil Anderson S. Sant’Ana

Sao Carlos, Brazil Caio Otoni

Preface to the Series . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

PART I ENVIRONMENTAL AND TOXICOLOGICAL ASPECTS OF FOOD PACKAGING

1 Biodegradability of Biodegradable Plastics in Compost, Marine,

PART II MICROSTRUCTURAL AND BARRIER FEATURES OF FOOD PACKAGING

9 Microstructural and Defect Analysis of Food Packaging Materials

10 Mapping the Distribution of Additives Within Polymer Films

PART III NONTRADITIONAL ROLES PLAYED BY FOOD PACKAGING MATERIALS

BIANCA C. N. FERNANDES • Department of Food Engineering, School of Food Engineering,

FABIANA H. SANTOS • Institute of Science and Technology, Federal University of

Environmental and Toxicological Aspects of Food Packaging

Biodegradability of Biodegradable Plastics in Compost,

Key words Biodegradability, Biodegradation, Industrial composting, Marine environment, Standard

1.3 Biodegradation Biodegradation is an important feature of biodegradable plastics.

microorganisms and enzymes. The bacteria that degrade the poly-

1.4 Biodegradation Certification is needed for biodegradable plastics to ensure that

1.5 Biodegradation Biodegradation can occur under a variety of conditions: anaerobic,

Despite the different nomenclatures, all performance specifica-

Disposal Specification Disintegration Respirometry Biogas

Recommendations for the type of inoculum used for the test

2.1 Biodegradation . Compost soil

2.3 Biodegradation . Blank anaerobic digester inoculum

1. Disintegration: Sufficient disintegration during composting.

The second test procedure for D6400-04 standard specifies a bio-

3. The level of regulated heavy metals (see Note 3) can be

2. Biodegradation rate: Adequate level of inherent biodegrada-

Biodegradation of biodegradable plastics in marine environ-

Marine biodegradation standards require that the plastic sam-

3.5.2 Procedure 1. Place 1 kg of inoculum derived from properly operating anaer-

Average biodegradation rates for various materials are listed in

home compositing conditions. Polycaprolactone (PCL) and poly-

½A] = ½A ]0 e - kobs t ð3Þ

Biodegradation rates of bioplastics, natural, and synthetic materials from literature

Temperature Average biodegradation per Residence

Soil 25–37 <0.01a >20 yra [19]

Wastewater >20 yra

Temperature Average biodegradation per Residence

Low-density polyethylene (LDPE) Soil 15–25 0.01 20 yr [36]

80.00 Aerobic wastewater

40.00 Industrial compost

kobs is the observed pseudo first-order rate constant, and

1. Compost needs to be either newly purchased or no more than

7. With every sampling time, 50% of the headspace gas is replaced

Biodegradability of Polymers by Relatively Low-Cost

Key words Biodegradation, Nonautomated respirometry, Titrimetry, Simplified Bartha, Standards

Biodegradable polymers have shown to be a promising alternative

obtained both from renewable (biobased) and fossil (fossil-based)

from reality. Natural environments, into which plastic waste is often

mesophilic, up to 30 °C). Thermophilic temperatures are normally

Kale et al. [18] assessed the biodegradation of PLA bottles

Comparing respirometry results of samples that do not belong

Biodegradation Time Compost/ Thickness