Full download Innovative Food Processing Technologies Advances in Multiphysics Simulation 1st Edition Kai Knoerzer pdf docx

Download the full version of the ebook at ebookfinal.
com
Innovative Food Processing Technologies Advances

in Multiphysics Simulation 1st Edition Kai
Knoerzer
https://ebookfinal.com/download/innovative-food-processing-
technologies-advances-in-multiphysics-simulation-1st-
edition-kai-knoerzer/
OR CLICK BUTTON
DOWNLOAD EBOOK
Download more ebook instantly today at https://ebookfinal.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...
Innovative Food Processing Technologies Advances in

Multiphysics Simulation 1st Edition Kai Knoerzer
https://ebookfinal.com/download/innovative-food-processing-
technologies-advances-in-multiphysics-simulation-1st-edition-kai-
knoerzer/
ebookfinal.com
Electron Beam Pasteurization and Complementary Food

Processing Technologies 1st Edition Suresh Pillai
https://ebookfinal.com/download/electron-beam-pasteurization-and-
complementary-food-processing-technologies-1st-edition-suresh-pillai/
ebookfinal.com
Advances in Food Biotechnology 1st Edition Rai
https://ebookfinal.com/download/advances-in-food-biotechnology-1st-
edition-rai/
ebookfinal.com
Information Technologies in Medicine Medical Simulation

and Education 1st Edition Metin Akay
https://ebookfinal.com/download/information-technologies-in-medicine-
medical-simulation-and-education-1st-edition-metin-akay/
ebookfinal.com
Ohmic Heating in Food Processing 1st Edition Hosahalli S.
Ramaswamy
https://ebookfinal.com/download/ohmic-heating-in-food-processing-1st-
edition-hosahalli-s-ramaswamy/
ebookfinal.com
Proteins in Food Processing 2nd Edition Rickey Y. Yada
https://ebookfinal.com/download/proteins-in-food-processing-2nd-
edition-rickey-y-yada/
ebookfinal.com
Advances in Battery Technologies for Electric Vehicles 1st

Edition Bruno Scrosati
https://ebookfinal.com/download/advances-in-battery-technologies-for-
electric-vehicles-1st-edition-bruno-scrosati/
ebookfinal.com
Enzymes in Food and Beverage Processing 1st Edition

Muthusamy Chandrasekaran (Editor)
https://ebookfinal.com/download/enzymes-in-food-and-beverage-
processing-1st-edition-muthusamy-chandrasekaran-editor/
ebookfinal.com
Environmentally friendly food processing Berit Mattsson
https://ebookfinal.com/download/environmentally-friendly-food-
processing-berit-mattsson/
ebookfinal.com
Innovative Food
Processing Technologies:
Advances in
Multiphysics Simulation
Kai Knoerzer, Pablo Juliano, Peter Roupas, and
Cornelis Versteeg EDITORS
A John Wiley & Sons, Inc., Publication

The IFT Press series reflects the mission of the Institute of Food Technologists—to advance the science of
food contributing to healthier people everywhere. Developed in partnership with Wiley-Blackwell, IFT
Press books serve as leading-edge handbooks for industrial application and reference and as essential texts
for academic programs. Crafted through rigorous peer review and meticulous research, IFT Press publica-
tions represent the latest, most significant resources available to food scientists and related agriculture
professionals worldwide.
Founded in 1939, the Institute of Food Technologists is a nonprofit scientific society with 22,000 individual
members working in food science, food technology, and related professions in industry, academia, and
government. IFT serves as a conduit for multidisciplinary science thought leadership, championing the use
of sound science across the food value chain through knowledge sharing, education, and advocacy.
IFT Press Advisory Group
Casimir C. Akoh
Christopher J. Doona
Jung Hoon Han
David B. Min
Ruth M. Patrick
Syed S.H. Rizvi
Fereidoon Shahidi
Christopher H. Sommers
Yael Vodovotz
Mark Barrett
Karen Nachay
Margaret Kolodziej
IFT Press Editorial Board
Malcolm C. Bourne
Dietrich Knorr
Theodore P. Labuza
Thomas J. Montville
S. Suzanne Nielsen
Martin R. Okos
Michael W. Pariza
Barbara J. Petersen
David S. Reid
Sam Saguy
Herbert Stone
Kenneth R. Swartzel
A John Wiley & Sons, Inc., Publication

This edition first published 2011 © 2011 by John Wiley & Sons, Ltd. and Institute of Food Technologists
Wiley-Blackwell is an imprint of John Wiley & Sons, formed by the merger of Wiley’s global Scientific, Technical
and Medical business with Blackwell Publishing.
Registered office: John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial offices: 2121 State Avenue, Ames, Iowa 50014-8300, USA

The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
9600 Garsington Road, Oxford, OX4 2DQ, UK
For details of our global editorial offices, for customer services and for information about how to apply for
permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell.
Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is
granted by Blackwell Publishing, provided that the base fee is paid directly to the Copyright Clearance Center, 222
Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license by CCC,
a separate system of payments has been arranged. The fee codes for users of the Transactional Reporting Service are
ISBN-13: 978-0-8138-1754-5/2011.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and
product names used in this book are trade names, service marks, trademarks or registered trademarks of their
respective owners. The publisher is not associated with any product or vendor mentioned in this book. This
publication is designed to provide accurate and authoritative information in regard to the subject matter covered.
It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional
advice or other expert assistance is required, the services of a competent professional should be sought.
Library of Congress Cataloging-in-Publication Data
Innovative food processing technologies: advances in multiphysics simulation / [edited by] Kai Knoerzer ... [et al.].
p. cm.—(IFT Press series)
Includes bibliographical references and index.
ISBN 978-0-8138-1754-5 (hardback)
1. Food industry and trade–Mathematical methods. 2. Food industry and trade–Simulation
methods. I. Knoerzer, Kai.
TP370.9.M38M85 2011
664'.020113–dc22
2010044300
A catalogue record for this book is available from the British Library.
This book is published in the following electronic formats: ePDF 9780470959411; Wiley Online Library
9780470959435; ePub 9780470959428
Set in 10/12 Times by Toppan Best-set Premedia Limited, Hong Kong
Disclaimer
The publisher and the author make no representations or warranties with respect to the accuracy or completeness of
the contents of this work and specifically disclaim all warranties, including without limitation warranties of fitness
for a particular purpose. No warranty may be created or extended by sales or promotional materials. The advice and
strategies contained herein may not be suitable for every situation. This work is sold with the understanding that the
publisher is not engaged in rendering legal, accounting, or other professional services. If professional assistance is
required, the services of a competent professional person should be sought. Neither the publisher nor the author
shall be liable for damages arising herefrom. The fact that an organization or Website is referred to in this work as a
citation and/or a potential source of further information does not mean that the author or the publisher endorses the
information the organization or Website may provide or recommendations it may make. Further, readers should be
aware that Internet Websites listed in this work may have changed or disappeared between when this work was
written and when it is read.
1 2011
Titles in the IFT Press series
• Accelerating New Food Product Design and Development (Jacqueline H. Beckley, Elizabeth J. Topp, M. Michele
Foley, J.C. Huang, and Witoon Prinyawiwatkul)
• Advances in Dairy Ingredients (Geoffrey W. Smithers and Mary Ann Augustin)
• Bioactive Proteins and Peptides as Functional Foods and Nutraceuticals (Yoshinori Mine, Eunice Li-Chan, and
Bo Jiang)
• Biofilms in the Food Environment (Hans P. Blaschek, Hua H. Wang, and Meredith E. Agle)
• Calorimetry in Food Processing: Analysis and Design of Food Systems (Gönül Kaletunç)
• Coffee: Emerging Health Effects and Disease Prevention (YiFang Chu)
• Food Carbohydrate Chemistry (Ronald E. Wrolstad)
• Food Ingredients for the Global Market (Yao-Wen Huang and Claire L. Kruger)
• Food Irradiation Research and Technology (Christopher H. Sommers and Xuetong Fan)
• Foodborne Pathogens in the Food Processing Environment: Sources, Detection and Control (Sadhana Ravishankar,
Vijay K. Juneja, and Divya Jaroni)
• High Pressure Processing of Foods (Christopher J. Doona and Florence E. Feeherry)
• Hydrocolloids in Food Processing (Thomas R. Laaman)
• Improving Import Food Safety (Wayne C. Ellefson, Lorna Zach, and Darryl Sullivan)
• Innovative Food Processing Technologies: Advances in Multiphysics Simulation (Kai Knoerzer, Pablo Juliano,
Peter Roupas, and Cornelis Versteeg)
• Microbial Safety of Fresh Produce (Xuetong Fan, Brendan A. Niemira, Christopher J. Doona, Florence E. Feeherry,
and Robert B. Gravani)
• Microbiology and Technology of Fermented Foods (Robert W. Hutkins)
• Multivariate and Probabilistic Analyses of Sensory Science Problems (Jean-François Meullenet, Rui Xiong, and
Christopher J. Findlay
• Nanoscience and Nanotechnology in Food Systems (Hongda Chen)
• Natural Food Flavors and Colorants (Mathew Attokaran)
• Nondestructive Testing of Food Quality (Joseph Irudayaraj and Christoph Reh)
• Nondigestible Carbohydrates and Digestive Health (Teresa M. Paeschke and William R. Aimutis)
• Nonthermal Processing Technologies for Food (Howard Q. Zhang, Gustavo V. Barbosa-Cánovas, V.M.
Balasubramaniam, C. Patrick Dunne, Daniel F. Farkas, and James T.C. Yuan)
• Nutraceuticals, Glycemic Health and Type 2 Diabetes (Vijai K. Pasupuleti and James W. Anderson)
• Organic Meat Production and Processing (Steven C. Ricke, Michael G. Johnson, and Corliss A. O’Bryan)
• Packaging for Nonthermal Processing of Food (Jung H. Han)
• Preharvest and Postharvest Food Safety: Contemporary Issues and Future Directions (Ross C. Beier,
Suresh D. Pillai, and Timothy D. Phillips, Editors; Richard L. Ziprin, Associate Editor)
• Processing and Nutrition of Fats and Oils (Ernesto M. Hernandez and Afaf Kamal-Eldin)
• Processing Organic Foods for the Global Market (Gwendolyn V. Wyard, Anne Plotto, Jessica Walden, and Kathryn
Schuett)
• Regulation of Functional Foods and Nutraceuticals: A Global Perspective (Clare M. Hasler)
• Resistant Starch: Sources, Applications and Health Benefits (Yong-Cheng Shi and Clodualdo Maningat)
• Sensory and Consumer Research in Food Product Design and Development (Howard R. Moskowitz,
Jacqueline H. Beckley, and Anna V.A. Resurreccion)
• Sustainability in the Food Industry (Cheryl J. Baldwin)
• Thermal Processing of Foods: Control and Automation (K.P. Sandeep)
• Trait-Modified Oils in Foods (Frank T. Orthoefer and Gary R. List)
• Water Activity in Foods: Fundamentals and Applications (Gustavo V. Barbosa-Cánovas, Anthony J. Fontana Jr.,
Shelly J. Schmidt, and Theodore P. Labuza)
• Whey Processing, Functionality and Health Benefits (Charles I. Onwulata and Peter J. Huth)
Contents
Preface, ix
Contributors, xiii
1. Introduction to Innovative Food Processing Technologies: Background, Advantages,

Issues, and Need for Multiphysics Modeling, 3
Gustavo V. Barbosa-Cánovas, Abdul Ghani Albaali, Pablo Juliano, and Kai Knoerzer
2. The Need for Thermophysical Properties in Simulating Emerging Food Processing

Technologies, 23
Pablo Juliano, Francisco Javier Trujillo, Gustavo V. Barbosa-Cánovas, and Kai Knoerzer
3. Neural Networks: Their Role in High-Pressure Processing, 39

José S. Torrecilla and Pedro D. Sanz
4. Computational Fluid Dynamics Applied in High-Pressure Processing Scale-Up, 57

Cornelia Rauh and Antonio Delgado
5. Computational Fluid Dynamics Applied in High-Pressure High-Temperature Processes:

Spore Inactivation Distribution and Process Optimization, 75
Pablo Juliano, Kai Knoerzer, and Cornelis Versteeg
6. Computer Simulation for Microwave Heating, 101

Hao Chen and Juming Tang
7. Simulating and Measuring Transient Three-Dimensional Temperature Distributions

in Microwave Processing, 131
Kai Knoerzer, Marc Regier, and Helmar Schubert
8. Multiphysics Modeling of Ohmic Heating, 155

Peter J. Fryer, Georgina Porras-Parral, and Serafim Bakalis
9. Basics for Modeling of Pulsed Electric Field Processing of Foods, 171

Nicolás Meneses, Henry Jaeger, and Dietrich Knorr
10. Computational Fluid Dynamics Applied in Pulsed Electric Field Preservation

of Liquid Foods, 193
Nicolás Meneses, Henry Jaeger, and Dietrich Knorr
vii
viii Contents
11. Novel, Multi-Objective Optimization of Pulsed Electric Field Processing

for Liquid Food Treatment, 209
Jens Krauss, Özgür Ertunç, Cornelia Rauh, and Antonio Delgado
12. Modeling the Acoustic Field and Streaming Induced by an Ultrasonic Horn Reactor, 233
Francisco Javier Trujillo and Kai Knoerzer
13. Computational Study of Ultrasound-Assisted Drying of Food Materials, 265

Enrique Riera, José Vicente García-Pérez, Juan Andrés Cárcel, Victor M. Acosta,
and Juan A. Gallego-Juárez
14. Characterization and Simulation of Ultraviolet Processing of Liquid Foods Using

Computational Fluid Dynamics, 303
Larry Forney, Tatiana Koutchma, and Zhengcai Ye
15. Multiphysics Modeling of Ultraviolet Disinfection of Liquid Food—Performance Evaluation

Using a Concept of Disinfection Efficiency, 325
Huachen Pan
16. Continuous Chromatographic Separation Technology—Modeling and Simulation, 335

Filip Janakievski
17. The Future of Multiphysics Modeling of Innovative Food Processing Technologies, 353
Peter J. Fryer, Kai Knoerzer, and Pablo Juliano
Index, 365
Color plate section appears between pages 208 and 209.

Preface
The food industry is an increasingly competitive and carbon and water footprint of food processing, thus
dynamic arena, with consumers now more aware of playing an important role toward environmental sus-
what they eat and, more importantly, what they want tainability and global food security.
to eat. Important food quality attributes such as Apart from the underlying thermo- and fluid-
taste, texture, appearance, and nutritional content dynamic principles of conventional processing, these
are strongly dependent on the way the foods are innovative technologies incorporate additional
processed. Multiphysics dimensions, for example, pressure
In recent years, with the aim to improve, or waves, electric and electromagnetic fields, among
replace, conventional processing technologies in others. To date, some of them still lack an adequate,
order to deliver higher-quality and better consumer- complete understanding of the basic principles of
targeted food products, a number of innovative tech- intervening in temperature and flow evolution in
nologies, also referred to as “emerging” or “novel” product and equipment during processing. Their
technologies have been proposed, investigated, proper application, development and optimization of
developed, and in some cases, implemented. These suitable equipment and process conditions still
technologies take advantage of other physics phe- require a significant amount of further knowledge.
nomena such as high hydrostatic pressure, electric Computational Fluid Dynamics (CFD) is already
and electromagnetic fields, and pressure waves. established as a tool for characterizing, improv-
Some of the most promising innovative technolo- ing, and optimizing traditional food processing
gies, in various stages of development and adoption, technologies. Innovative technologies, however,
are discussed in this book, namely high-pressure provide additional complexity and challenges for
processing (also in combination with heat), micro- modelers because of the concurrent interacting
wave processing, ohmic heating, pulsed electric Multiphysics phenomena. In order to differentiate
field processing, ultrasound processing (liquid- and Multiphysics modeling from CFD modeling, the word
airborne), ultraviolet light (UV) processing, and “Multiphysics” will be capitalized throughout the
enhanced continuous separation. book.
These innovative technologies provide the oppor- Four symposia were organized at two consecu-
tunity not only for the development of new foods tive Annual Meetings of the Institute of Food
but also for improving the safety and quality of Technologists (IFT) in 2008 and 2009 (New Orleans
conventional foods through milder processing. and Anaheim, respectively) to gather Multiphysics
Different physical phenomena, utilized by these modeling experts in innovative technologies to
technologies, can potentially reduce energy and water present and discuss the latest advances in their
consumption and therefore assist in reducing the respective fields. These symposia highlighted the
ix
x Preface
importance and key role of Multiphysics modeling generalized visual programming approach referred
to further advance the development of each innova- to as macroscopic modeling. In Chapters 4 and 5,
tive technology and facilitate their introduction into “conventional” CFD modeling approaches for high-
the food industry. pressure processes at both low and high temperatures
Written by international experts from world-class are discussed and their application for equipment
research centers, academia, and industry, this book design, scale-up, and optimization are highlighted.
explains and discusses how Multiphysics model- Also described is their application to present the
ing—that is, the simulation of the entire process process outcomes in terms of safety and quality
comprising the actual equipment, varying process of the processed foods. Chapter 6 and 7 covers
conditions, and the thermophysical properties of the the extension of classical CFD with a further
food to be treated—can be applied in the develop- Multiphysics dimension, electromagnetic radiation,
ment, optimization, and scale-up of innovative food and the implementation for designing and character-
processing technologies. The most recent research izing microwave heating processes. Chapter 7 also
outcomes are shown to demonstrate benefits to discusses various temperature mapping techniques
process efficiency and the impact on scalability, and introduces the use of magnetic resonance
safety, and quality. imaging (MRI) for the determination of microwave-
The first part of this book includes two chapters induced three-dimensional heating patterns. In
introducing the rationale of the book and some Chapter 8, historical and new developments of
common themes to all chapters. Chapter 1 is the Multiphysics modeling applied to ohmic heating are
introductory chapter outlining the range of innova- presented. Chapters 9, 10, and 11 are devoted to
tive processing technologies covered, briefly describ- modeling of pulsed electric fields processing, cover-
ing the technologies and making the case for the ing the basics of the technology, its application for
necessity of Multiphysics modeling for their design, predicting liquid food pasteurization, and the “mul-
development, and application. Chapter 2 discusses tiobjective” optimization of the technology for liquid
the importance of determining the relevant (common food processing. Chapters 12 and 13 present two
and technology-specific) thermophysical properties distinctly different ultrasound applications. Chapter
and their essential role for accurate model 12 covers liquid-borne ultrasound, including a review
prediction. on its use in food processing, followed by an exten-
The second part of the book is an extensive col- sive review of the mathematics and physics involved
lection of chapters devoted to the various case studies in this technology, and this is concluded with a novel
on the modeling of innovative food processing approach of modeling ultrasound-induced stream-
technologies. For clarity and convenience, they are ing. Chapter 13 details the use of airborne ultrasound
divided into subsections focusing on high-pressure for the improvement of drying processes at low tem-
processing (Chapters 3–5), technologies utilizing peratures. The complex mathematics is described
electric and/or electromagnetic effects (microwave, and the chapter is concluded by experimental studies,
ohmic heating, and pulsed electric field processing; highlighting the advantages and commercial poten-
Chapters 6–11), processes using ultrasound waves tial of this innovative drying technology. Chapters
(in liquids or air) (Chapters 12 and 13), ultraviolet 14 and 15 both describe UV processing for liquid
light (UV) processing (Chapters 14 and 15), and food disinfection/pasteurization as an effective alter-
finally, one chapter on innovative chromatographic native to thermal treatments. Chapter 14 focuses on
separation technologies (Chapter 16). the characterization of several alternative reactor
Chapter 3 discusses two fundamentally different designs by Multiphysics modeling, whereas Chapter
modeling approaches to characterize high-pressure 15 compares the performance of different commer-
(low-temperature) systems. It introduces the reader cially available reactors using Multiphysics model-
to the very promising modeling technique known as ing and the introduction of the concept of “disinfection
artificial neural networks (ANN), as well as the more efficiency.” The final technology chapter (Chapter
Preface xi
16) introduces an innovative continuous separation The editors wish to thank all collaborators in this
process based on the chromatographic simulated book for their excellent contributions, and the time
moving bed principle. It outlines the procedure of and effort they have devoted to making this book a
modeling these types of technologies and highlights comprehensive interdisciplinary reference source
the advantages over conventional column or bed- for engineers, technologists and scientists, and
based separation processes. researchers from academia and industry alike. We
Chapter 17 is the take-home message of this believe that the value of this book is not limited to
book, which concludes with a summary on what was food engineering; it is also useful for other branches
presented in the chapters before and provides an of process and chemical engineering. We would also
outlook on future trends in Multiphysics simulation like to thank the Institute of Food Technologist’s
of innovative food processing technologies. Three Nonthermal Processing Division, the International
questions are posed: (1) What can be usefully Division, and the Food Engineering Division for
modeled today?; (2) What extra data is needed?; sponsoring the session symposia that led to the
and (3) How much detail is needed, or Where development of this book.
shall we stop? This chapter is not intended to
provide definitive answers to these questions, but Kai Knoerzer
it suggests some future research directions and Pablo Juliano
places where research ought to or is expected to Peter Roupas
arrive. Cornelis Versteeg
Contributors
Víctor M. Acosta Hao Chen

Grupo de Ultrasonidos de Potencia Department of Biological Systems Engineering
Consejo Superior de Investigaciones Científicas Washington State University
(CSIC) Pullman, WA 99164-6120
Serrano, 144, E28006 (Currently with Microsoft, Redmond, WA)
Madrid, Spain
Antonio Delgado
Abdul Ghani Albaali Institute of Fluid Mechanics
Princess Sumaya University for Technology Friedrich-Alexander University
P.O. Box 1438 Erlangen-Nuremberg
Al-Jubaiha 11941 Cauerstrasse 4, D-91058 Erlangen
Jordan Germany
Serafim Bakalis Özgür Ertunç

Centre for Formulation Engineering Institute of Fluid Mechanics
School of Chemical Engineering Friedrich-Alexander University
University of Birmingham Erlangen-Nuremberg
Birmingham B15 2TT Cauerstrasse 4, D-91058 Erlangen
United Kingdom Germany
Gustavo V. Barbosa-Cánovas Larry Forney

Department of Biological Systems Engineering School of Chemical and Biomolecular Engineering
Washington State University Georgia Institute of Technology
Pullman, WA 99164-6120 311 Ferst Drive, N.W.
Atlanta, GA 30332
Juan Andrés Cárcel
Grupo de Análisis y Simulación de Procesos Peter J. Fryer
Agroalimentarios (ASPA) Centre for Formulation Engineering
Departamento de Tecnología de Alimentos School of Chemical Engineering
Universidad Politécnica de Valencia University of Birmingham
Camí de Vera s/n, E46022, Valencia Birmingham B15 2TT
Spain United Kingdom
xiii
xiv Contributors
Juan A. Gallego-Juárez Jens Krauss

Grupo de Ultrasonidos de Potencia Institute of Fluid Mechanics
Consejo Superior de Investigaciones Científicas Friedrich-Alexander University
(CSIC) Erlangen-Nuremberg
Serrano, 144, E28006, Madrid Cauerstrasse 4, D-91058 Erlangen
Spain Germany
José Vicente García-Pérez

Kai Knoerzer
Grupo de Análisis y Simulación de Procesos
CSIRO Food and Nutritional Sciences
Agroalimentarios (ASPA)
671 Sneydes Road
Departamento de Tecnología de Alimentos
Werribee, VIC 3030
Universidad Politécnica de Valencia
Australia
Camí de Vera s/n, E46022, Valencia
Spain
Nicolás Meneses
Henry Jaeger Department of Food Biotechnology and Food
Department of Food Biotechnology and Food Process Engineering
Process Engineering Technische Universität Berlin
Technische Universität Berlin Koenigin-Luise-Str. 22
Koenigin-Luise-Str. 22 D-14195 Berlin
D-14195 Berlin Germany
Germany
Huachen Pan
Filip Janakievski
Institute of Mechatronic Engineering
Hangzhou Dianzi University
671 Sneydes Road
310018 Hangzhou
Werribee, VIC 3030
China
Australia
Pablo Juliano Georgina Porras-Parral

CSIRO Food and Nutritional Sciences Centre for Formulation Engineering
671 Sneydes Road School of Chemical Engineering
Werribee, VIC 3030 University of Birmingham
Australia Birmingham B15 2TT
United Kingdom
Dietrich Knorr
Department of Food Biotechnology and Food
Cornelia Rauh
Process Engineering
Institute of Fluid Mechanics
Technische Universität Berlin
Friedrich-Alexander University
Koenigin-Luise-Str. 22
Erlangen-Nuremberg
D-14195 Berlin
Cauerstrasse 4, D-91058 Erlangen
Germany
Germany
Tatiana Koutchma
Guelph Food Research Centre, Agriculture and Marc Regier
Agri-Food Canada Fachhochschule Trier
93 Stone Road West University for Applied Sciences
Guelph, ON, N1G 5C9 Schneidershof, 54293 Trier
Canada Germany
Contributors xv
Enrique Fernando Riera Franco de Sarabia José S. Torrecilla

Grupo de Ultrasonidos de Potencia Department of Chemical Engineering
Consejo Superior de Investigaciones Científicas Universidad Complutense de Madrid
(CSIC) Avenida Complutense s/n
Serrano, 144, E28006, Madrid 28040 Madrid
Spain Spain
Pedro D. Sanz Francisco Javier Trujillo

Malta Consolider Team CSIRO Food and Nutritional Sciences
Department of Processes 11 Julius Avenue
ICTAN, CSIC North Ryde, NSW 2113
c/ José Antonio Novais, 10 Australia
28040 Madrid
Spain Cornelis Versteeg
Helmar Schubert 671 Sneydes Road
Universitaet Karlsruhe (TH)/Karlsruhe Institute of Werribee, VIC 3030
Technology (KIT) Australia
Institute of Engineering in Life Sciences
Dept. I: Food Process Engineering Zhengcai Ye
Karlsruhe, Germany Bechtel Oil, Gas and Chemicals, Inc.
3000 Post Oak Blvd
Juming Tang Houston, TX 77056
Department of Biological Systems Engineering
Washington State University
Pullman, WA 99164-6120
Chapter 1
Introduction to Innovative Food Processing

Technologies: Background, Advantages, Issues,
and Need for Multiphysics Modeling
Gustavo V. Barbosa-Cánovas, Abdul Ghani Albaali, Pablo Juliano, and Kai Knoerzer
1.1. Introduction concept development to implementation. In particu-

lar, proper application, development, and optimiza-
In a world that is demanding environmental sustain- tion of suitable equipment and process conditions
ability and food security, innovation is a key require- require a significant amount of further knowledge
ment for the sustained growth of the food industry. and understanding. In this book, the basic principles,
Furthermore, product innovation is the response to current research, challenges, and commercial appli-
the growing demand for value addition along with cations of the respective technologies, as well as the
more sophisticated and diverse food products. development and application of computational fluid
Modern food technology provides a handful of novel dynamics (CFD) and, more broadly, Multiphysics
processing options to explore, which could provide modeling as a tool for characterizing, improving, and
more diverse food industry products and more com- optimizing innovative food processing technologies
petitive and efficient processes. Many of these inno- are covered.
vative technologies can provide new opportunities Most innovative processing technologies have a
for the development of new foods and for the improve- common challenge, that is, to achieve a sufficient
ment of safety and quality of more conventionally uniformity of the treatment or the process. This chal-
manufactured foods through milder processing. lenge is often already an issue at laboratory scale
This book discusses innovative technologies that and it can become progressively worse when scaling
take advantage of physical forces and phenomena up to pilot plants and, subsequently, to commercial
such as high hydrostatic pressure, electric and elec- equipment. Among other potential technology-
tromagnetic fields, and pressure waves, for example, specific issues, nonuniformity of the treatment is
high-pressure processing (also in combination with most commonly encountered. In fact, the nonunifor-
heat), microwave processing, ohmic heating, pulsed mities of the process and the lack of process valida-
electric field (PEF) processing, ultrasound processing tion of innovative processes are the greatest
(liquid and airborne), and ultraviolet light (UV) pro- limitations for industrial uptake.
cessing. Innovative processing technologies present Nonuniform treatment is, however, not specific to
a number of hurdles that need to be addressed from innovative processing technologies; conventional
Innovative Food Processing Technologies: Advances in Multiphysics Simulation, First Edition.

Edited by Kai Knoerzer, Pablo Juliano, Peter Roupas, and Cornelis Versteeg.
© 2011 by John Wiley & Sons, Ltd. and Institute of Food Technologists. Published by John Wiley & Sons, Ltd. ISBN: 978-0-813-81754-5
3
4 Innovative Food Processing Technologies: Advances in Multiphysics Simulation
processing technologies often encounter the same 1.2. Multiphysics Modeling

problem. For example, in conventional heat treat-
ment processes such as canning, the temperature
1.2.1. Definition
at the product surface is significantly higher than Multiphysics modeling is an extension of classical
at the product center during most of the processing CFD. By definition, CFD is one of the branches of
time, and only after prolonged holding times are fluid mechanics that uses numerical methods and
temperature gradients throughout the product dimin- algorithms to solve and analyze problems that
ished. Another clear example of nonuniformity in involve fluid flows. The geometry of the modeled
conventional processing is the drying process of scenario, including all components, is discretized
particulates. In this case, spatial and temporal het- into finite cells on which the governing partial dif-
erogeneities in temperature and water content in the ferential equations (PDEs), namely the continuity,
food product can be even more pronounced. The momentum, and energy conservation equations, are
product goes (1) through an initial linear drying solved. This is detailed in the chapters specific to the
phase with water removal from the product surface, respective technologies. Because these are PDEs,
(2) over the falling rate period with moisture flux they cannot be solved analytically. Numerical tech-
from the inside of the product to its surface, and (3) niques, such as finite differences, finite volumes, or
to a stage of product and drying medium (moisture) finite element methods, must be applied to achieve
equilibrium with almost no further change in water an approximated solution (Sun 2007).
content. In drying food products other important Multiphysics modeling is based on the same prin-
factors often come into play, increasing the degree ciples as conventional CFD, that is, geometry dis-
of nonuniformity: product shrinkage and reduced cretization, and solving the PDEs is performed in a
moisture transport (increasing viscosity of contained similar manner. However, Multiphysics modeling
liquids) up to a stage where pores are blocked. In the comprises additional physical phenomena such as
case of many innovative processing technologies as electromagnetic waves, electrical fields, and acous-
described throughout this book, nonuniformities tic waves related to the innovative technologies dis-
may be reduced through technology-specific effects. cussed further in this chapter. These phenomena can
However, these nonuniformities may be more pro- also be described by physically based PDEs (specific
nounced due to increased complexities influenced by to each innovative technology), which have to be
additional Multiphysics phenomena. solved simultaneously with the ones from classical
This introductory chapter outlines the range of CFD. In some cases, the expression of the process
innovative food processing technologies covered in outcome based on the attributes of the processed
this book and gives a short overview of their benefits food, that is, the remaining microbial load, enzyme
and advantages over traditional technologies. Some activity, and chemical reaction products, is required.
additional background information on the technolo- Within Multiphysics modeling, reaction kinetics
gies, not covered in the respective technology- (i.e., microbial inactivation, quality degradation,
specific chapters, is provided. Furthermore, this chemical reaction, and structural responses) can be
chapter makes a case for the need for applying coupled with the specific differential equations to
Multiphysics modeling in these technologies for provide the spatial distributions of reaction response.
their design, including scale-up and optimization. Multiphysics models that concurrently solve the
The chapter summarizes the problems and chal- PDEs of classical CFD and the additional technology-
lenges faced by the modelers, particularly with specific physical phenomena and the differential
respect to the prediction of temperature, flow and equations describing the reaction response require
technology-specific field distributions (e.g., sound significantly greater computational resources. The
intensity and electric or electromagnetic fields), and increase in affordable computational power in recent
the extent of microbial or enzymatic inactivation years has allowed the simulation of innovative
and their distribution in equipment and products. processes.
Chapter 1 Introduction to Innovative Food Processing Technologies 5
Figure 1.1. Number of commercial high-pressure equipment units around the world as of 2009 (Tonello 2010).
1.3. Innovative Food Processing used not only for preservation, but also for modify-
Technologies ing the physical and functional properties of some
foods.
1.3.1. Background
More than 70 companies currently utilize HPP,
This section presents a brief description of each producing more than 170,000 tons of products
technology covered in this book. The major design (Tonello 2010). Several HPP-treated food products,
problems and application limitations of these tech- including juices, jams, jellies, yogurts, ready-to-eat
nologies are highlighted as an introduction to sub- meat, and oysters, are already widely available in
sequent chapters. Ways in which Multiphysics the United States, Europe, Japan, New Zealand, and
modeling of innovative food processing technologies Australia. These successful applications have led to
can assist in their development will be discussed. a pronounced increase in commercial-scale HPP
units around the world during the past 10 years, as
1.3.1.1. High-Pressure Processing (HPP) and shown in Figure 1.1.
High-Pressure Thermal Sterilization (HPTS) HPP In addition to inactivation of microorganisms and
has demonstrated wide applicability for producing some spoilage enzymes (Seyderhelm et al. 1996;
high-quality foods. HPP has become accepted as Yen and Lin 1996), promising results have been
an attractive alternative to traditional preservation obtained with respect to the application on gelation
methods utilizing preservatives or thermal process- of food proteins (Ohshima et al. 1993), improvement
ing (Hernando Saiz et al. 2008, Chapters 3–5). of digestibility of proteins, and tenderization of meat
HPP is commonly referred to as a nonthermal products (Ohmori et al. 1991; Jung et al. 2000a,
process of liquid and solid foods through application 2000b; Buckow et al. 2010b). These changes in pro-
of high pressure in the order of 100–800 MPa teins have been used successfully in fish meat; in
(1,000 to 8,000 bar) and holding times of several Carpaccio and Carpaccio-like products, high pres-
minutes. HPP of foods is of increasing interest sure allows the “processing” of the product, while
because it allows the inactivation of vegetative still maintaining its raw characteristics. However,
organisms at low or moderate temperature with because of the application of high pressures, these
minimum degradation (Abdul Ghani and Farid products have retained “fresh-like” qualities and
2007). HPP offers opportunities for increased shelf texture compared with heat-processed food, are
life and preservative-free stabilization of meats, microbiologically safe, and have an extended shelf
seafood, vegetable products, and juices. HPP can be life compared with raw food. Gomez-Estaca et al.
(2009) investigated HPP on fish products (such as where pressure-induced protein denaturation may be
salmon, tuna, and cod), showing superior sensory visually evident. High pressures can also induce sig-
results. nificant structural changes (or damages) in some
If the aim of the process is the inactivation of sensitive foods, such as strawberries or lettuce. Cell
microbial spores, high pressure alone is not suffi- deformation and cell membrane damage can result
cient. However, a combination of high pressure and in softening and cell serum loss. Usually, these
elevated temperatures, also referred to as HPTS or changes are undesirable because the food will appear
pressure-assisted thermal sterilization, can result in to be processed and no longer fresh or raw.
synergistic inactivation of these spores at potentially
lower temperatures or shorter processing times, Limitations of HPP and HPTS Although great
thus improving the quality of the processed foods progress has been made in the development of eco-
while potentially reducing energy consumption nomically viable high-pressure applications, the sci-
(Bull et al. 2009). In this application, the increase in entific community and the food industry recognized
pressure is used as a means to increase the tempera- in the early 2000s that engineering fundamentals,
ture evenly and fast in the product. including CFD models, were required to design,
There are two approaches to achieve high- evaluate, optimize, and scale up high-pressure pro-
pressure conditions. In the direct approach, a piston cesses of foods (Hendrickx and Knorr 2001).
is utilized, which compresses the content of the The limitation of HPP to date mainly lies in the
high-pressure chamber. In the indirect approach, a limited throughput and, relative to heat processing,
pressure-transmitting liquid (e.g., water) is pumped the high cost of equipment, labor (HPP is not yet a
into the treatment chamber (high-pressure vessel) fully automated process), and maintenance. High
using a high-pressure pump followed by a “pressure maintenance costs are caused mainly by the extreme
intensifier.” Liquids at extremely high pressures are processing conditions. Furthermore, there are only
compressible, requiring extra fluid to be pumped a few large-scale commercial high-pressure equip-
into the vessel. ment suppliers worldwide that have expertise in the
During compression, the temperature of the pro- food industry, including Avure Technologies, Inc.
cessed food and the pressure-transmitting fluid (Kent, WA), Kobelco (Kobe Steel Ltd., Kobe,
increases due to the compression force working Japan), and NC Hyperbaric (Burgos, Spain).
against intermolecular forces. The magnitude of the A common issue in both HPP and HPTS is the
adiabatic temperature increase depends on a number nonuniformity of some aspects of the treatment. HPP
of factors, such as the pressure medium and food generates pressure waves in liquids, which travel
product thermophysical properties (density, thermal at the speed of sound (sound in water travels at
expansion coefficient, and specific heat capacity) and 1,500 m/s). Therefore, pressure is commonly assumed
initial temperature (see, e.g., Chapters 2, 4, and 5). to be transmitted instantaneously and uniformly.
Higher fat content of the food and higher initial However, treatment nonuniformities can occur
temperature, for example, lead to an increase in during HPP not only as a result of different compress-
compression heating. The phenomenon of increas- ibilities of the various substances in the food product,
ing compression heating at elevated initial tempera- including trapped air (also headspace), but also
tures is important; for example, in HPTS, the product because of the food packaging material. In addition,
and the pressure medium are preheated to achieve if the purpose of the process is the inactivation of the
higher process temperatures, which in turn allows vegetative microorganisms, a nonuniform treatment
inactivation of microbial spores (Wilson et al. 2008). can occur because some microorganisms are suppos-
In HPP, the greater the pressure level and time of edly more resistant to the pressure when embedded
application, the greater the potential for changes in in a fat matrix. Foods with higher fat or oil content
the structure and appearance of the treated foods. may, therefore, protect the microorganisms in some
This is especially true for raw high-protein foods, areas in the food where fat is contained.
In the case of processing above room temperature Microwave heating has been proposed as an
(initial temperature), for example, in HPTS, nonuni- alternative to traditional heating methods in many
form treatment temperature is likely to be more pro- food manufacturing processes, such as (re)heating,
nounced. In addition to pressure, temperature is an baking, (pre)cooking, tempering of frozen food,
important process variable. In heterogeneous food blanching, pasteurization, sterilization, and dehy-
materials, with the contents exhibiting differences in dration (Metaxas and Meredith 1983; Decareau
compression heating, temperatures may not be uni- 1985; Buffler 1993; Metaxas 1996; Schubert and
formly distributed in the food products. Furthermore, Regier 2005; Tang et al. 2008).
the packaging material, the material of the product Microwave and radio frequency heating for pas-
carrier, and the steel of the high-pressure vessel are teurization and sterilization are rapid; therefore, less
not heated to the same extent as the food; therefore, time is required for come-up to the desired process
temperature gradients are developed throughout the temperature compared with conventional heating.
system, leading to heat flux from the products to the This is particularly true for solid and semisolid
cooler areas (which are mainly the steel walls). foods that depend on slow thermal diffusion process
These spatial temperature heterogeneities increase in conventional heating. Microwave and radio fre-
over the process time. Although, theoretically, the quency heating can approach the benefits of high-
preheated product heats up uniformly during temperature short-time (HTST) processing, whereby
compression to sterilization temperatures, during bacterial destruction is achieved, while thermal deg-
pressure holding time temperatures may decrease in radation of the desired components is reduced.
certain areas of the vessel. This can affect spore Heating with microwaves primarily involves two
inactivation, and spores may survive the process if mechanisms. Water in the food is often the main
temperature loss is not prevented. Product carriers component responsible for dielectric heating. Due to
have been developed as a means of retaining their dipolar nature, water molecules follow the
heat throughout the vessel during both pressure alternating electric field associated with electromag-
come-up and holding times (Chapter 5). Multiphysics netic radiation. The second major mechanism is
modeling can greatly assist in the characterization of through the oscillatory migration of ions in the food
temperature distribution, subsequent microbial dis- under the influence of the alternating electric field.
tributions, and other quality changes as a result of Such oscillatory motion of water molecules and ions
temperature inhomogeneities. These models can also and the associated intermolecular friction lead to a
be applied to the redesign and optimization of equip- conversion of electromagnetic energy to thermal
ment and determination of adequate processing con- energy.
ditions for optimum process/product performance. The dielectric properties, namely the dielectric
constant and the loss factor (Chapter 2), determine
1.3.1.2. Microwave and Radio Frequency the strength of the electric field inside the food and
Processing Microwave heating refers to the use of its conversion into heat. These properties strongly
electromagnetic waves of certain frequencies to depend on the composition (or formulation) of the
generate heat in a material (Metaxas and Meredith food, with moisture and salt being the two primary
1983; Roussy and Pearce 1995; Metaxas 1996). determinants of interest (Mudgett 1985, 1986; Sun
Typically, microwave food processing uses frequen- et al. 1995; Nelson and Datta 2001). The subsequent
cies of 2,450 and 915 MHz. In domestic ovens, temperature rise in the food depends on the duration
2,450 MHz frequency is commonly utilized, while of heating, the location in the food, convective heat
in industrial heating application both frequencies are transfer at the surface, and the heat conduction and
used, depending on the product to be treated, that is, extent of evaporation of water inside the food and
product size and composition, associated with the at its surface.
relevant thermophysical properties (Chapters 2, 6, Although the final objective of each process
and 7). differs, an increase in product temperature is seen as
a common theme. There has also been some specu- uniform. Placing dielectrics (i.e., food products) into
lation on the so-called nonthermal effects of electro- the microwave field leads to a change in the field
magnetic waves in the microwave frequency range. distribution. Therefore, differences in the products,
Four theories have been proposed to explain “non- for example, product size, shape, and particularly
thermal” or nondirect thermal effects of microwaves composition with varying dielectric properties, will
on, for example, microorganisms: selective heating, almost certainly lead to changes in process outcomes.
electroporation, cell membrane rupture, and magnetic However, not only do the field variations in the cavity
field coupling (Kozempel et al. 1998). The selective cause nonuniform processing, the field characteris-
heating theory states that solid microorganisms are tics inside the product are also heterogeneous.
heated more effectively by microwaves than the sur- The heterogeneous composition of the different
rounding medium and are thus killed more readily. food components (and different dielectric proper-
Electroporation is caused when pores form in the ties) is an important factor in the heating of foods.
membrane of the microorganisms due to electrical Differences in dielectric properties lead to differ-
potential across the membrane, resulting in leakage ences in temperature increases, even in a perfectly
(this is similar to one of the theories on the effect of homogeneous microwave field. As these properties
PEF processing for cold pasteurization). Cell mem- are in most cases strongly temperature-dependent,
brane rupture is related to the voltage drop across changes in temperature may compensate or may
the membrane, which causes it to rupture, which is increase the nonuniformity. In particular, in cases
also a theory in PEF processing. In the fourth theory, where increasing temperatures lead to increasing
cell lysis occurs due to coupling of electromagnetic loss factors (the imaginary part of the complex
energy with critical molecules within the cells, dis- dielectric permittivity; Chapter 2), a so-called
rupting vitally important internal cell components. thermal “runaway” phenomenon can occur. With
Although researchers have repeatedly reported increasing temperature the rate of converting the
nonthermal effects of microwave processing, the electromagnetic energy into thermal energy increases
general consensus (Heddleson and Doores 1994; as well; therefore, the gradients between hot and cold
Heddleson et al. 1994) is that the reported nonther- areas in the product become more pronounced.
mal effects are likely to be due to the lack of precise Another important factor in heating is the so-
measurements of the time–temperature history and called focusing effect of the microwaves into spe-
its spatial variations. A number of studies have cific areas in the product. This phenomenon is
shown that thermal effect is the essential contributor strongly dependent on the geometrical properties of
to the destruction of microorganisms (Goldblit and the product. For example, a spherical product that
Wang 1967; Rosen 1972; Fujikawa et al. 1992). does not exceed a certain size (due to limited pen-
Therefore, to date, it is presumed that only thermal etration) can exhibit a pronounced hot spot in its
effects on microbial inactivation are effective, and geometrical center.
microbial inactivation caused by microwave pro- Other phenomena causing uneven heating pat-
cessing is essentially the same as in conventional terns include edge and corner overheating (caused
thermal processing. Of course, the rates of heating by the penetration and absorption of the microwaves
and temperature distributions are quite different. from more than one direction) and the development
of standing waves inside the product (which is mainly
Limitations of Electromagnetic Heating Volumetric dependent on the dielectric constant (the real part of
microwave and radio frequency heating is theoreti- the complex dielectric permittivity; Chapter 2).
cally more uniform than conventional heating (Datta The time–temperature history at the coldest point
and Hu 1992). There are, however, a number of for a conventional thermal process is generally pre-
microwave-specific factors that induce nonuniform dictable for a food that is all solid or all fluid. For
heating patterns. First, electromagnetic field distribu- example, for a conduction-heated (solid) food, it is
tion inside a microwave cavity is, in most cases, not usually the geometric center. In microwave heating,
even for a solid food, it is less straightforward to 1.3.1.3. Ohmic Heating Ohmic heating is defined
predict the coldest point and it can change during as a process wherein electric currents are passed
the heating process depending on temperature- through foods or other materials with the primary
dependent material properties and oven characteris- purpose of heating them. The heating occurs in the
tics (Fleischman 1996; Zhang et al. 2001). form of internal electric energy dissipation within
A number of approaches have been proposed to the material. Ohmic heating is distinguished from
improve the uniformity associated with microwave other electrical heating methods by the presence of
heating. These include rotating and oscillating the electrodes contacting the food, the frequency of the
food in the microwave cavity (Geedipalli et al. current, or the waveform. The main purpose for the
2007), providing an absorbing medium (such as hot development of ohmic heating processes was to
water) surrounding the product (Chen et al. 2008; allow for HTST sterilization of solid–liquid mix-
Chapter 6), equilibrating after heating (Fakhouri and tures (Chapter 8).
Ramaswamy 1993), and cycling the power (Chapter Applications of ohmic heating in the food indus-
7). Success to date is limited due to the dependence try to date are scarce, although there are a number
of the materials’ properties on temperature and the of advantages over other (conventional) heating
nonuniform distribution of the electromagnetic field methods. The main advantages for ohmic heating
inside the food and the microwave cavity. Utilizing are the associated rapid and relatively uniform
a lower microwave frequency of 915 MHz and radio heating of the food product, depending on the elec-
frequencies to improve uniformity of heating have trical conductivity of the food components. This is
the potential to improve the evenness of heating expected to reduce unwanted thermal effects on the
(Chen et al. 2008), as the penetration depth into the product that often occur in conventional heating
food is greater and the field nonuniformities are less applications, caused by the need to heat the product
pronounced. Combinations of microwave and con- by the transfer of thermal energy from a heating
ventional technologies in many different configura- medium to a low temperature product, where exces-
tions (e.g., hot air, vacuum, or infrared heating) have sive treatment times are necessary for sufficient
also been used to improve treatment uniformity; heat penetration from the surface of a solid product
(Contreras et al. 2008; Turabi et al. 2008; Abbasi and to its core.
Azari 2009; Kowalski and Mierzwa 2009; Kowalski Potential applications for ohmic heating include
and Rajewska 2009; Seyhun et al. 2009; Uysal et al. its use in blanching, evaporation, dehydration, fer-
2009). These approaches can be successful for some mentation, and extraction. At present, the primary
applications, especially where the cold spot is located type of application is a heat treatment for microbial
at the food surface (Chapter 7); however, in food control, for example, for the pasteurization of milk,
products with high salt or sugar content, the cold spot and also for processing of sauces, fruits, and toma-
is usually within in the food, as the penetration depth toes (Chapter 8).
of the electromagnetic waves is reduced. The principal mechanisms of microbial inactiva-
It remains a challenge to uniformly treat food tion in ohmic heating are thermal in nature. Recent
products with microwaves and to achieve the tar- literature, however, indicates that a mild electro-
geted process outcomes; Multiphysics models, poration mechanism may occur during ohmic
however, will greatly assist in designing microwave heating (similar to the effects utilized in PEF pro-
processes by evaluating process performance and cessing (Lebovka et al. 2005; Kulshrestha and
developing appropriate control strategies (Chapters Sastry 2006). The principal reason for the additional
6 and 7). Accordingly, Multiphysics models (includ- microbial inactivation effect to heating of ohmic
ing temperature-dependent properties of foods) need treatment may be its low frequency (50–60 Hz),
to be developed and subsequently validated to ascer- which allows cell walls to build up charges and form
tain the location of the point of lowest integrated pores. This is in contrast to high-frequency methods
time–temperature history (Chapter 7). such as microwave or radio frequency heating,
where the electric field is essentially reversed before throughout the product. Multiphysics modeling
sufficient charge buildup occurs at the cell walls. (including the temperature-dependent properties of
Nevertheless, temperature is the principal critical the foods: mainly the electrical conductivity) can
process factor in ohmic heating. As in conventional greatly assist the evaluation and optimization of
thermal processes, the key issue is identifying the ohmic heating systems to achieve heating unifor-
slowest heating zone. Fundamentally, there is only mity (Chapter 8).
one critical factor: the temperature–time history of
the coldest point. Since the primary critical process 1.3.1.4. PEF PEF processing is an innovative non-
factor is the thermal history and location of the cold thermal processing technology mainly for liquid and
spot, the effects on microbial inactivation are the pumpable foods (including emulsions, suspensions,
same as for thermal processes. Locating the slowest and semisolids such as sausage meat), predominantly
heating zones during ohmic heating, however, cannot used for the inactivation of microorganisms at
be extrapolated from current knowledge of conven- ambient or mild temperatures, thereby preserving the
tional heating, and requires special consideration. fresh flavor, color, functional properties, and integ-
Several factors significantly affect the tempera- rity of heat-sensitive compounds (Chapters 9–11).
ture within an ohmic process. The critical parame- PEF can also be used to enhance extraction yield of
ters in continuous flow ohmic heating systems juices and bioactives from plant sources. PEF is one
include electrical conductivities of the respective of the most appealing nonthermal technologies for
phases of the food, temperature dependence of the preservation of liquid foods due to reduced heating
electrical conductivity, design of the heating device effects compared with traditional pasteurization
(e.g., location and orientation of the electrodes), methods (Barbosa-Cánovas et al. 1999).
extent of interstitial fluid motion, residence time In PEF processing, a liquid or other pumpable
distribution, thermal properties of the food, and material is passed through an electrode arrangement
electric field strength (Chapter 8). where the PEF is applied. For microbial inactiva-
tion, foods are processed by means of brief pulses
Limitations of Ohmic Heating The main limita- of a strong electric field with field strengths of
tion of ohmic heating is the heterogeneous nature (in around 15–40 kV/cm. For extraction of plant materi-
composition) of the food products and their corre- als and pretreatment of meat for processing, only
sponding electrical conductivities that leads to dif- about 0.7 to 3 kV/cm is required (Toepfl et al. 2006).
ferences in the conversion of the electrical current The utilization of PEF leads to the formation of
into thermal energy. As in microwave heating, in pores (the so-called electroporation [temporary or
ohmic heating, thermal runaway can also occur, permanent]), in the membranes of microbial or plant
because electrical conductivity, which is the property cells, which disturbs and damages the membrane’s
that influences electrical energy dissipation, usually functionality, leading to inactivation of the cells and
increases with increasing temperature. Therefore, the partial release of the cell contents to make
especially in stationary (i.e., not moving in a stream) extraction or other processing more efficient.
solid products, there may be areas that are very hot Membrane disruption occurs when the induced
(usually areas close to the electrodes), which in some membrane potential exceeds a critical value of 1 V
instances may even be burned, while in other areas in many cellular systems, which, for example, cor-
(with initially lower electrical conductivities, or responds to an external electric field of about 10 kV/
farther away from the electrodes) almost no heating cm for Escherichia coli (Castro et al. 1993). The
occurs. most relevant factor affecting microbial inactivation
Uniform heating with ohmic processing is theo- and extraction enhancement by PEF is, therefore,
retically possible, but at the same time challenging the electric field intensity. The combination of elec-
due to the various factors impacting on the slowest tric field intensity, total treatment time during PEF
heating zone and the time–temperature history and pulse shapes, and the associated temperature
increase determine the extent of membrane disrup- temperature increase (caused by ohmic heating).
tion in bacterial and plant cells (Hamilton and Sale Several factors play a role here, including the mate-
1967). Other factors affecting the performance of rial’s electrical conductivity, the frequency of the
the PEF process include the microbial entity to be pulses, their duration (width), adequacy of deaera-
inactivated (type, concentration, and growth stage tion, back pressure, and the flow rate of the liquid
of microorganism) and the treatment media (pH, (laminar or turbulent flow regime; residence time in
antimicrobials, and ionic compounds, electric con- the treatment chamber). Because the pulse duration
ductivity, and medium ionic strength). is only in the range of microseconds and, therefore,
PEF produces products with slightly different the overall treatment time is short, temperature
properties from conventional pasteurization treat- increases during treatment are often assumed to be
ments. Most enzymes are not affected by PEF. The minimal and temperature effects neglected in inac-
fact that the maximum temperature reached is lower tivation studies.
than in thermal pasteurization means that some of In processing liquids with PEF, a nonuniformity
the flavors associated with the raw material are not of the treatment can be a result of the interaction
destroyed. Spores, with their tough protective coats, between the flow, heat transfer, electric field phe-
and dehydrated cells are mostly able to survive PEF nomena, and effects on microbial or plant cells.
processing. The survival of spores and enzymes Predictions of the increase in temperature caused by
means that products have to be refrigerated after the electric field are similar to ohmic heating and
passing through PEF processing in order to slow the less complicated compared with the dissipation of
action of the enzymes and keep pathogens from electromagnetic energy in microwave processing.
growing; PEF alone is generally not capable of pro- Moreover, the property influencing this dissipation
ducing ambient shelf-stable products. However, effect, that is, the electrical conductivity, is easier to
acidic well-packaged products may have a useful measure, and usually shows a less complex behavior
ambient shelf life. with temperature than the two dielectric properties
As indicated before, another potential application in microwave processing, that is, the dielectric con-
of PEF, which is gaining increasing interest, is the stant and the loss factor (Chapter 2).
utilization of the technology for enhanced extraction However, the purpose of the pulsed (potentially
of plant cell material. Because PEF induces electro- alternating) electric field is, unlike in microwave
poration in cell walls at relatively low energy inputs, processing, not an increase in temperature. The
allowing the cell contents to leak out, it holds temperature increase should be minimized in
promise as an efficient way of getting useful com- most PEF applications. The main aim is a nonther-
ponents out of cells and cell membranes (Corrales mal inactivation of vegetative microorganisms for
et al. 2008; Lopez et al. 2009a, 2009b; Loginova et cold pasteurization or a nonmechanical means of
al. 2010; Puertolas et al. 2010). opening cells for enhanced extraction. In particular,
To date, however, PEF has been mainly researched for the purpose of cold pasteurization, a great
to preserve the quality of foods, such as to improve degree of electric field uniformity is needed to
the shelf life of orange juice, apple juice, milk, and ensure a similar treatment of the entire liquid
liquid eggs, as well as the fermentation properties product. Ideally, the same number of electric pulses
of brewer ’s yeast. Martín-Belloso and Soliva- and electric field strength is applied to all microor-
Fortuny (2010) have summarized the work of several ganisms present in the liquid. Typically, pasteuriza-
researchers on food-borne pathogenic microorgan- tion requires inactivation of up to 99.999%, that
isms in different food products. is, 5 log of the target organism. If only a small frac-
tion of microorganisms bypass proper treatment
Limitations of PEF Processing Issues that may through regions of low electric field strength, it
arise with PEF include electric arcing, dielectric is not possible to reach the required extent of
breakdown of the treated food, and a pronounced inactivation.
Achieving this uniformity, however, is very chal- numerical simulations can be applied to improve the
lenging; the electric field distribution is strongly fundamental understanding of the physical phenom-
dependent on the configuration of the treatment ena in the process and to optimize it with respect to
chamber (and to a lesser extent on the electrical the chamber design and operating conditions
conductivity and other thermophysical properties of (Gerlach et al. 2008; Chapters 9–11).
the processed media). PEF chamber designs such as
co-field, coaxial, or colinear electrode arrangements 1.3.1.5. Ultrasound Processing This technology
(Chapters 9–11) exhibit pronounced nonuniformi- is based on pressure waves at frequencies exceeding
ties in flow, temperature, and electric field distribu- 20 kHz, that is, more than 20,000 vibrations per
tions. Uniform fields can be achieved in parallel second. It is considered as another innovative
plate configurations, which are mainly applied for process that has been investigated for many different
batch processing. If the field is not uniform, the purposes over the last decades. While in the earlier
induced temperature increase is also uneven across work mainly the lower frequencies of around 20 kHz
the volume of the treatment chamber. Often, several were studied, research and applications currently
treatment cells are arranged to process in series, include frequencies of several hundred kHz, to
which reduces the effects of imperfections in single several MHz (Chapter 12).
treatment cells. Ultrasound systems consist of a generator for
Thus, in processes for inactivation of specific turning electrical energy into high-frequency alter-
microorganisms that show synergistic effects of nating current, a transducer for converting the alter-
temperature and electric field on inactivation, tem- nating current into mechanical vibrations, and a
perature nonuniformities will lower the performance delivery probe for conveying the sonic vibrations
of the process. Nonuniformities can be minimized, into a medium to couple sonic vibrations to the
but to some extent will always occur. treated material. The transducers may take the shape
To enable a comparable treatment history of the of a rod, plate, bar, or sphere, and are usually manu-
entire product, the flow pattern is very important. factured from titanium, aluminum, or steel.
Laminar flow conditions, which can be found in The ultrasonic transducer can be mounted outside
low-throughput laboratory-scale systems, are to be on the wall of a vessel or flow cell and be in indirect
avoided. In laminar flow, each microorganism contact with foods, or it can be inserted into a treat-
follows a more or less straight path through the ment chamber or flow cell of specified geometry to
treatment chamber; therefore, pronounced differ- transmit energy directly into a food system with
ences in exposure to varying electric field strengths better energy efficiency (Feng and Yang 2005).
and temperatures will occur. Modifying the treat- There are also transducers that are designed for
ment chamber with grids (Chapter 10) or increasing effective transmission into air (Chapter 13).
the flow rate to give turbulent flow (Buckow et al. Ultrasound has attracted considerable interest
2010a) can improve the uniformity of exposure of in the food industry due to its useful effects in
the product to the important treatment variables food structure modification (e.g., emulsification,
(e.g., temperature and electric field strength) and extraction, crystallization, and viscosity alteration),
furthermore improve temperature uniformity due to food preservation, and enzyme modulation (Patist
increased (turbulent) thermal conduction and con- and Bates 2008). As one of the innovative and
vective flows. advanced food processing technologies, it can be
For characterizing process performance, informa- applied to develop gentle but targeted processes to
tion on the field distributions is essential. However, improve the quality and safety of processed foods
such local information inside the chambers is diffi- and, thus, offers the potential for improving existing
cult and near to impossible to obtain experimentally. processes as well as for developing new process
For further development of the PEF technology, options.
Ultrasound alone has some effects on the inacti- ment of dewatering, drying and filtration, inactiva-
vation of vegetative organisms in liquid food prod- tion of microorganisms and enzymes, disruption of
ucts. The bactericidal effect of ultrasound is cells, degassing of liquids, emulsification, accelerat-
generally attributed to intracellular cavitation ing heat transfer and extraction processes (Patist
(Hughes and Nyborg 1962). It is proposed that and Bates 2008; Vilkhu et al. 2008), enhancement
micro-mechanical shocks and jet streaming are of processes dependent on diffusion (e.g., enzyme
created by microscopic cavitation bubbles induced activity, targeted infusion of small compounds
by the fluctuating pressures under the ultrasonica- into porous food matrices), and also targeted move-
tion process (Chapter 12). These shocks and micro- ment of two-phase systems, such as oil droplets
jets disrupt cellular structural and functional or particles dispersed in a continuous aqueous
components up to the point of cell lysis. phase (Doblhoffdier et al. 1994; Hawkes et al. 1997;
Positive effects have been observed when ultra- Groschl 1998). It is evident that ultrasound technol-
sound is used in combination with temperature ogy has a wide range of actual and future applica-
(thermo-sonication) or pressure (mano-sonication) tions in the food industry.
or both (mano-thermo-sonication) in the inactiva- More recently, research activities related to the
tion of pathogenic bacteria, spoilage microorgan- sonochemistry in certain foods products have gained
isms, and enzymes (Cameron et al. 2009; Demirdoven interest, involving the reactions that ultrasound gen-
and Baysal 2009; Lee et al. 2009). The use of tem- erates in food during processing. Jambrak et al.
perature and ultrasound together has been successful (2009) show that these chemical reactions can be
in reducing the enzymatic activity in some target used to generate new compounds in food for specific
products such as juices, providing better stability purposes such as the modification of proteins.
during storage (Terefe et al. 2009). Sonicated milk Hydroxylation of phenolic compounds to enhance
is the most explored product; it shows positive their antioxidant properties has also been studied by
results in pasteurization standards, better homogeni- Ashokkumar et al. (2008).
zation and color, as well as new physical properties Another interesting application is the use of air-
for the development of dairy products (Chouliara borne ultrasound for enhanced drying of food prod-
et al. 2010). ucts. Difficulties in the propagation of ultrasound
Most developments of ultrasound for food appli- waves in air and the impedance mismatch at the
cations are nonmicrobial in nature, that is, their main transducer/air interface have led to the development
aim is not inactivation of microorganisms (Hoover of especially adapted transducers that have been
1997). High frequencies in the range of 0.1 to applied, for example, to drying of carrots, lemon
20 MHz, pulsed operation, and low power levels peel, and other food products (Garcia-Perez et al.
(100 mW) are used for nondestructive testing 2009; Chapter 13).
(Gunasekaran and Ay 1994). These industrial appli-
cations include texture, viscosity, and concentration Limitations of Ultrasound Processing Although
measurements of many solid and fluid foods; com- potential applications of ultrasound processing are
position determination of eggs, meats, fruits and many and diverse, the uptake by industry to date is
vegetables, dairy, and other products; thickness, not widespread. Reasons for this include a lack of
flow level, and temperature measurements for moni- knowledge of ultrasound intensity distribution in
toring and control of several processes; and nonde- tank systems and, particularly, in flow-through
structive inspection of egg shells and food systems, where the forced convection disturbs the
packages. ultrasound field, as well as the effect of the pressure
Apart from testing applications, process improve- waves on the food product. Depending on the equip-
ments have been observed in applications such as ment, with the generators, the transducers, treatment
cleaning surfaces (Tolvanen et al. 2009), enhance- cells, the frequency and power of the ultrasound
waves, and the product properties, the effect of in process design, scale-up and optimization and
ultrasound can differ significantly. subsequent uptake of the technology by the food
Ultrasound processing comprises another Multi- industry.
physics phenomenon: the acoustic field. Several
factors must be considered, for example, ultrasound
frequency, intensity, the associated speed of sound 1.3.1.6. UV Processing UV light for food pro-
and sound absorption, and impact on the acoustic cessing has been investigated for many years but is
field. Although the speed of sound in a homoge- still considered as an innovative technology in food
neous medium is independent of the sound wave processing. In this technology, UV-C light (wave-
frequency and intensity, varying composition of length of 254 nm) is predominantly being used as a
the treated product strongly impacts the speed. The disinfection method to inhibit or inactivate food-
sound absorption is dependent on the composition borne microorganisms, mainly in liquid food prod-
as well as the frequency and intensity of the ultra- ucts (Chapters 14 and 15). Fresh produce can be
sound waves. In addition to this, occurring cavita- processed using UV light, which has a germicidal
tion (Chapter 12) significantly influences the speed effect on many types of microorganisms (bacteria,
of sound, the sound absorption and, therefore, the viruses, protozoa, molds, and yeasts). However, the
acoustic field distribution. The speed of sound in a effect of UV light on microorganisms in liquids
cavitating medium can, for example, decrease from depends on variables such as density of the liquid,
a value of 1,500 m/s to values as low as 20 m/s. types of microorganisms, UV-C absorptivity of the
As discussed in Chapter 12, the ultrasound waves liquid, and the solids (suspended or soluble) in the
in a cavitating medium can be completely absorbed liquid.
by the cavitation bubbles in the close vicinity of the Although the use of UV light is well established
ultrasound transducer and, therefore, large parts of for air and water treatment and surface decontami-
the sonoreactors may not undergo ultrasound treat- nation, its use for treating liquid foods is still limited.
ment. Although this pronounced absorption leads to Recently, interest in using UV has increased as a
the conversion of the sound energy into motion and viable alternative to thermal pasteurization for a
the formation of a turbulent jet, which could in turn range of liquid foods and ingredients (fresh juices,
result in the treated liquid being well-mixed and, fruit purees, soft drinks, raw milk, liquid eggs, liquid
therefore, undergoing a similar treatment over time, sugars and sweeteners, etc.) (Koutchma 2009).
the presence of solid products, which due to size and Pumpable fruit and vegetable products are generally
different densities cannot completely follow the very suitable for processing by UV light to reduce
flow, may induce further treatment nonuniformity. the microbial load (Guerrero-Beltran and Barbosa-
During ultrasound processing, standing waves Cánovas 2004) as long as sufficient fluid mixing
(the so-called bands) can occur. This formation of allows the entire product to be exposed to a certain
bands can be an intended desirable effect, for required dose of UV radiation. The germicidal prop-
example, for separating multiphase products such as erties of UV irradiation are mainly due to DNA
emulsions. In other cases, where the sound waves damage induced through absorption of UV light by
are meant to induce other effects, such as cell DNA molecules. This mechanism of inactivation
disruption, sono- or biochemical reaction, the stand- results in a sigmoidal curve of microbial population
ing waves can unintentionally impair the process reduction (Bolton 1999).
performance. UV treatment can be used for primary disinfection
Hence, generic Multiphysics models, including or as a backup for other purification methods such as
acoustics, heat and fluid flow and, potentially, cou- carbon filtration, reverse osmosis, or pasteurization.
pling to the kinetics of food transformation, enhanced As UV has no residual effect, the best position for a
diffusion, microbial interaction, and enzyme modu- treatment system is immediately prior to the point of
lation need to be developed. Such models can assist use. This ensures that any incoming microbiological
contaminants are destroyed and that there is little substances is limited at these wavelengths. Therefore,
chance of post-treatment contamination. microorganisms on the surface of products can be
In addition to UV-C light, UV light with wave- protected by the so-called shadowing effect, caused,
lengths other than 254 nm can also be used as a for example, by overlapping parts of the products.
radiation source to inactivate microorganisms in Treating opaque liquids is impossible under laminar
foods (liquids or solids). In general, wavelengths flow conditions as the product flowing through the
ranging from 100 (UV-V, vacuum UV light) to center of a UV transparent glass tube will not “see”
400 nm (UV-A) are suitable for UV light processing the UV light. Providing a highly turbulent flow,
(Bintsis et al. 2000; Sastry et al. 2000). however, can allow sufficient treatment uniformity,
UV disinfection has many advantages over alter- as all particles will likely be close to the glass walls
native methods. Unlike chemical treatment, UV at least for a certain period of time. Residence time
does not introduce toxins or residues into the process in such a flow reactor must be sufficiently long to
and mostly does not alter the chemical composition, ensure similar treatment histories of the entire liquid
taste, odor, or pH of the water or liquid being disin- product. Multiphysics modeling can assist in the UV
fected. As a physical method, UV irradiation has a chamber design and optimizing process conditions
positive consumer image and is of interest to the according to the absorptivity and other properties of
food industry as a low-cost nonthermal method of the fluid, while assuring a similar treatment history
preservation. Recent advances in the science and of all portions of the liquid or dispersion.
engineering of UV light irradiation have demon- In all technologies and their associated specific
strated that this technology holds considerable issues regarding nonuniformity discussed in the pre-
promise as an alternative method to traditional vious sections, Multiphysics modeling can assist in
thermal pasteurization for liquid foods and ingredi- providing insights into the internal distribution of
ents, fresh juices, soft drinks, and beverages. processes in treatment chambers and products. It can
be utilized to improve the systems design, perfor-
Limitations of UV Processing Compared with mance, optimization, and scale-up to commercial
water, liquid foods have a range of optical and phys- applications by reducing inhomogeneities and for
ical properties, diverse chemical compositions, and the process to become acceptable and viable.
solid-phase characteristics (particle size and size
distribution, shape and volume fraction), influencing
1.4. Modeling Challenges
UV light transmittance (UVT), dose delivery,
momentum transfer (laminar or turbulent flow), and Previous sections have described a number of limi-
consequently microbial inactivation (Koutchma tations encountered in innovative food processing
2009; Chapter 14). As there is no practical method technologies and how Multiphysics modeling can
for evaluating the spatially resolved performance assist in overcoming them. However, there are prac-
experimentally and predictions of the process per- tical complexities in modeling and validating models
formance are not straightforward for liquid foods for these technologies that will be covered in this
(compared with water), Multiphysics modeling is section.
essential for evaluating particles and fluid velocities
in the UV reactor, particle mixing, particle location,
1.4.1. Modeling Complexity in
residence times, UV fluence rate (irradiance) distri-
Innovative Processing
bution and resulting changes in bacterial count.
As mentioned, UV is mainly useful for surface As mentioned earlier in this chapter, modeling inno-
decontamination (e.g., on fresh produce) and for vative processing involves additional physics phe-
disinfection of liquids transparent to the UV light to nomena to conventional CFD. This implies that the
a certain extent. Although consisting of electromag- fundamental conservation equations from thermo-
netic waves, the penetration of UV light into opaque fluiddynamics need to be coupled with the PDEs to
describe the respective field phenomena (i.e., elec- expressed as a function of the process variables
tromagnetic, electric, and acoustic fields), thereby needed for model accuracy. As will be shown,
providing considerably increased complexity. more properties are needed to model innovative
Developing a Multiphysics model requires the processes than conventional processes.
same steps as developing a CFD model (Sun 2007), • Each boundary condition will have to be defined
for example, the geometry definition, where all for each Multiphysics phenomenon as a require-
objects in the model scenario are constructed and ment to solve PDEs at the interface of each
assembled in the modeling software package or subdomain.
imported from a computer-aided design (CAD) • In most cases, when relatively high frequencies
drawing of the system (treatment chamber, periph- are involved (PEF, ultrasonic processing, ohmic
eral devices, piping, food, packages, etc.). In particu- heating, microwave, UV), time resolution of the
lar, the materials forming the complete computational respective waves is not feasible due to the short
domain (i.e., the processing system), commonly time scales (i.e., a higher frequency gives smaller
referred to as subdomains, may include solids, time scales) across the domain. Therefore, an
liquids, and gases. The next step is discretization, integrated value from the steady-state solution of
that is, approximating the computational domain by the wave equation is often used as a source term in
finite cells. Next, material specific thermophysical the conservation equations for momentum and
properties need to be allocated to each subdomain as energy.
functions of the process variables (e.g., temperature
Multiphysics models are highly nonlinear mathe-
and pressure; Chapter 2) and initial conditions and
matical problems. With each additional PDE, the
boundary conditions need to be defined. Furthermore,
degree of nonlinearity increases. Depending on the
the model needs to specify time dependency, whether
mesh, the thermophysical properties (as functions of
they are transient or stationary.
the process variables), and the number of physics
The main differences between conventional CFD
phenomena being coupled, the difficulty for conver-
and Multiphysics modeling are as follows:
gence of the model may increase. When that is the
case, models may not be as robust as those devel-
• The geometry discretization step often needs more
oped utilizing classical CFD PDEs.
details than in conventional CFD. For example, in
electromagnetics modeling with the finite differ-
ence method 15 cubic cells per wavelength have
1.4.2. Validation of Multiphysics Models
been recommended (QWED Sp.z o.o. 2003). In
microwave processing at 2.45 GHz, with a wave- Validation is an essential step to complete the mod-
length in vacuum of around 12 cm, the maximum eling process. Models used for prediction of process
mesh cell size should therefore be below 1 mm variables and their distributions may converge,
(edge length). Furthermore in modeling an acous- suggesting solutions that might be plausible, but
tic field with the finite element method, the resolved in fact are not accurate (Nicolaï et al. 2001).
three-dimensional (3D) mesh should have at least Therefore, particularly in the case of highly nonlin-
12 degrees of freedom per wavelength for each ear Multiphysics problems, the numerical solutions
possible direction of the wave (i.e., the degrees of must always be validated before using them
freedom of the complete mesh in three dimensions for further studies, such as equipment and process
should be 1,728 times the model volume in wave- redesign, optimization, or scale-up. The validation
lengths). Higher frequencies, with shorter wave- process involves the comparison of predicted data
lengths, therefore, limit the feasible volume of the (i.e., temperature, velocities, inactivation extent,
model scenario (COMSOL Multiphysics 2007). and chemical or physical change) with measured
• As discussed in Chapter 2, there is a lack of ther- data. This can be done using two approaches: (1)
mophysical property data of foods, in particular direct validation of process variables; or (2) indirect
Table 1.1. Tools for the validation of Multiphysics models.
Process variable or Direct (D)/

Technology outcome to validate Indirect (I) Method Chapter or reference
High-pressure Temperature D Thermocouples, wireless temperature Chapter 5
processing logger
I Enzymatic or other temperature time (Pehl et al. 2000;
integrator (TTI), liquid crystals Grauwet et al.
2010a, 2010b)
Fluid velocity D High-pressure PIV (Pehl and Delgado
1999)
Indicator D/I Enzymatic, microbial, colorimetric (Denys et al. 2000)
determination methods among others
Microwave, Temperature D Thermocouples, fiber-optic probes Chapters 7 and 8
ohmic I MRI, infrared thermography, Chapters 6–8
heating microwave radiometry, time
temperature integrators, liquid
crystals
Fluid velocity D PIV (particle tagging), LDA Chapter 8
Indicator D/I Enzymatic, microbial, colorimetric Chapter 6
Pulsed electric Temperature D Thermocouples (not in the area of Chapter 10;
field (PEF) high-electric field strength), (Buckow et al.
fiber-optic probes 2010a)
I Enzymatic TTI —
Indicator D/I Enzymatic, microbial, colorimetric Chapter 11
Ultrasound Temperature D Thermocouples (type K) —
Fluid velocity D PIV, LDA Chapter 12
Acoustic intensity D Hydro- and microphones Chapter 13
I Qualitative visual (cavitation fields, (Klima et al. 2007;
band formation of particles), Sutkar and Gogate
chemical markers 2010)
Indicator D/I Enzymatic, microbial, colorimetric, —
Ultraviolet Fluid velocity D PIV, LDA (Hofman et al. 2007)
(UV) Indicator D/I Microbial (also referred to as Chapters 14, 15
biodosimetry), colorimetric,
validation by means of a (bio)chemical or microbial • sound intensity by means of hydro- or microphones

indicator. Table 1.1 classifies the indirect and direct • particle (size) change by measuring particle size
validation tools to determine the process variables distribution, for example, with a Focused Beam
or outcomes for each technology. Reflectance Measurement (FBRM) device.
Direct measurements include:
The direct measurement of other process variables,
• temperature measurements by utilizing resistance such as electric and electromagnetic field distribu-
thermometers, thermocouples, or fiber-optic sensors tion, including, for example, microwave and UV
• flow measurement by means of Laser Doppler light, is often not feasible in the constrained space of
Anemometry (LDA) or Particle Imaging Velocimetry the respective processing equipment. However, the
(PIV) process outcomes, such as microbial inactivation,
nutrient degradation, structural changes, and mate- 1.5. Concluding Remarks

rial separation, are the true process indicators. Unless
After examining the literature, we note that only
they are spatially resolved, these only give overall
some of the Multiphysics models developed to
outcomes and are of no or limited value to validate
describe innovative processing technologies have
local variations in a treatment cell or zone.
been thoroughly validated. The overarching aim of
An indirect measurement involves the evaluation
the models is to represent a process outcome that will
of enzymatic, microbial, color, chemical, or any
provide certain design or optimization aids. However,
other biological or physical change that represents a
in practice, not all of the validation variables or
change in temperature or any other process variable
outcomes in Table 1.1 have been measured to match
when typical measurement devices cannot be uti-
a certain model. In the case of HPP, PEF, ultrasound,
lized. For example, temperature distribution changes
and ohmic heating, more outcome-related models
after microwave sterilization can be measured using
need adequate validation to establish, for example,
whey protein and measuring Maillard reaction com-
accurate predictions of microbial, enzymatic or
ponents at different locations (Chapter 6). Another
chemical reaction distributions, or other outcome-
example is the use of magnetic resonance imaging
related parameters. On the other hand, more com-
(MRI) to measure the change in proton resonance
plete outcome-related validated models have been
frequency, referred to as chemical shift, to establish
developed for microwave processing. For example,
3D temperature distribution changes (Chapter 7).
the Multiphysics models presented in Chapter 6 have
Methods for validation will be covered in more
assisted in the filing of microwave sterilization pro-
detail in the following chapters.
cessing in the U.S. Food and Drug Administration.
Once the data of the measured process variables
As such, Multiphysics models will mainly be useful
or process outcomes are gathered, there are different
when reaching a stage of predicting process out-
ways of comparing simulated with measured data.
comes that leverage technologies to industrial levels.
A common method to validate transient simulations
In order to achieve this, a more direct collaboration
is the comparison of profiles at specific points of the
between processing equipment manufacturers, inter-
modeling domain, which are measurable. Another
ested industry partners, and researchers is needed for
approach is to compare model and measured data at
successful design and implementation of innovative
several specific time and location coordinates (in a
food processing technologies.
2D or 3D grid) throughout the process period in a
parity plot, for example, represent measured tem-
perature versus simulated ones at identical locations References
and selected times in a plot (e.g., Knoerzer et al.
Abbasi S, Azari S. 2009. Novel microwave-freeze drying of
2007, 2008). Comparison of 2D or 3D distributions onion slices. Int J Food Sci Technol 44(5):974–979.
requires working in matrices. Abdul Ghani AG, Farid MM. 2007. Numerical modeling of heat
Distributions of inactivation or chemical or phys- transfer in high pressure processing using computational fluid
ical changes cannot be validated through direct mea- dynamics. Proceedings of the 4th International Conference on
High Pressure Bioscience and Biotechnology, 1, 205–212.
surements at selected points. The study of a certain
Ashokkumar M, Sunartio D, Kentish S, Mawson R, Simons L,
volume in packages containing an initial amount of Vilkhu K, Versteeg C. 2008. Modification of food ingredients
substance can partially resolve this problem. By this by ultrasound to improve functionality: A preliminary study on
method overall averages for the whole vessel or a model system. Innov Food Sci Emerg Technol 9(2):155–160.
vessel areas where packages are located are calcu- Barbosa-Cánovas GV, Gongora-Nieto MM, Pothakamury UR,
Swanson BG. 1999. Preservation of Foods with Pulsed Electric
lated from the predictions in the model. For example,
Fields. San Diego, CA, USA: Academic Presss.
Chapter 5 shows a case where a relative activity Bintsis T, Litopoulou-Tzanetaki E, Robinson RK. 2000. Existing
ratio (Eq. 5.16) is utilized to determine enzyme and potential applications of ultraviolet light in the food
inactivation distributions after HPP. industry—A critical review. J Sci Food Agric 80(6):637–645.
Bolton JR. 1999. Ultraviolet Applications Handbook. Ayr, ON, Fakhouri MO, Ramaswamy HS. 1993. Temperature uniformity
CAN: Bolton Photosciences, Inc. of microwave heated foods as influenced by product type and
Buckow R, Schroeder S, Berres P, Baumann P, Knoerzer K. composition. Food Res Intern 26(2):89–95.
2010a. Simulation and Evaluation of Pilot-Scale Pulsed Feng H, Yang W. 2005. Power ultrasound. In: YH Hui, ed.,
Electric Field (PEF) Processing. J Food Eng 101(1):67–77. Handbook of Food Science, Technology, and Engineering,
Buckow R, Truong BQ, Versteeg C. 2010b. Bovine cathepsin D 121-1–121-9, New York: CRC Press.
activity under high pressure. Food Chem 120(2):474–481. Fleischman GJ. 1996. Predicting temperature range in food slabs
Buffler CR. 1993. Microwave Cooking and Processing: undergoing long term low power microwave heating. J Food
Engineering Fundamentals for the Food Scientist. New York: Eng 27(4):337–351.
AVI Book. Fujikawa H, Ushioda H, Kudo Y. 1992. Kinetics of escherichia-
Bull M, Olivier S, Diepenbeek R, Kormelink F, Chapman B. coli destruction by microwave irradiation. Appl Environ
2009. Synergistic inactivation of spores of proteolytic Microbiol 58(3):920–924.
Clostridium botulinum strains by high pressure and heat is Garcia-Perez JV, Carcel JA, Riera E, Mulet A. 2009. Influence
strain and product dependent. Appl Environ Microbiol 75(2): of the applied acoustic energy on the drying of carrots and
434–445. lemon peel. Drying Technol 27(2):281–287.
Cameron M, McMaster LD, Britz TJ. 2009. Impact of ultrasound Geedipalli SSR, Rakesh V, Datta AK. 2007. Modeling the heating
on dairy spoilage microbes and milk components. Dairy Sci uniformity contributed by a rotating turntable in microwave
Technol 89(1):83–98. ovens. J Food Eng 82(3):359–368.
Castro AJ, Barbosa-Cánovas GV, Swanson BG. 1993. Microbial Gerlach D, Alleborn N, Baars A, Delgado A, Moritz J, Knorr D.
inactivation of foods by pulsed electric-fields. J Food Process 2008. Numerical simulations of pulsed electric fields for
Preserv 17(1):47–73. food preservation: A review. Innov Food Sci Emerg Technol
Chen H, Tang J, Liu F. 2008. Simulation model for moving food 9(4):408–417.
packages in microwave heating processes using conformal Goldblit SA, Wang DIC. 1967. Effect of microwaves on esche-
FDTD method. J Food Eng 88(3):294–305. richia coli and bacillus subtilis. Appl Microbiol 15(6):1371.
Chouliara E, Georgogianni KG, Kanellopoulou N, Kontominas Gomez-Estaca J, Lopez-Caballero ME, Gomez-Guillen MC,
MG. 2010. Effect of ultrasonication on microbiological, chem- de Lacey AL, Montero P. 2009. High pressure technology
ical and sensory properties of raw, thermized and pasteurized as a tool to obtain high quality carpaccio and carpaccio-like
milk. Int Dairy J 20(5):307–313. products from fish. Innov Food Sci Emerg Technol 10(2):
COMSOL Multiphysics. 2007. Acoustics Module—User ’s 148–154.
Guide. Stockholm, Sweden: COMSOL AB. Grauwet T, Van der Plancken I, Vervoort L, Hendrickx ME, Van
Contreras C, Martin-Esparza ME, Chiralt A, Martinez-Navarrete Loey A. 2010a. Mapping temperature uniformity in industrial
N. 2008. Influence of microwave application on convective scale HP equipment using enzymatic pressure-temperature-
drying: Effects on drying kinetics, and optical and mechanical time indicators. J Food Eng 98(1):93–102.
properties of apple and strawberry. J Food Eng 88(1):55–64. Grauwet T, Van der Plancken I, Vervoort L, Hendrickx ME, Van
Corrales M, Toepfl S, Butz P, Knorr D, Tauscher B. 2008. Loey A. 2010b. Protein-based indicator system for detection
Extraction of anthocyanins from grape by-products assisted of temperature differences in high pressure high temperature
by ultrasonics, high hydrostatic pressure or pulsed electric processing. Food Res Intern 43(3):862–871.
fields: A comparison. Innov Food Sci Emerg Technol 9(1): Groschl M. 1998. Ultrasonic separation of suspended particles—
85–91. Part I: Fundamentals. Acustica 84(3):432–447.
Datta AK, Hu W. 1992. Optimization of quality in microwave- Guerrero-Beltran JA, Barbosa-Cánovas GV. 2004. Review:
heating. Food Technol 46(12):53–56. Advantages and limitations on processing foods by UV light.
Decareau RV. 1985. Microwaves in the Food Processing Industry. Food Sci Technol Int 10(3):137–147.
Orlando: Academic Press Inc. Gunasekaran S, Ay C. 1994. Evaluating milk coagulation with
Demirdoven A, Baysal T. 2009. The use of ultrasound and com- ultrasonics. Food Technol 48(12):74–78.
bined technologies in food preservation. Food RevInt 25(1): Hamilton WA, Sale AJH. 1967. Effects of high electric fields on
1–11. microorganisms .2. mechanism of action of lethal effect.
Denys S, Van Loey AM, Hendrickx ME. 2000. A modeling Biochim Biophys Acta 148(3):789.
approach for evaluating process uniformity during batch high Hawkes JJ, Limaye MS, Coakley WT. 1997. Filtration of bacteria
hydrostatic pressure processing: Combination of a numerical and yeast by ultrasound-enhanced sedimentation. J Appl
heat transfer model and enzyme inactivation kinetics. Innov Microbiol 82(1):39–47.
Food Sci Emerg Technol 1(1):5–19. Heddleson RA, Doores S. 1994. Factors affecting microwave-
Doblhoffdier O, Gaida T, Katinger H, Burger W, Groschl M, heating of foods and microwave-induced destruction of
Benes E. 1994. A novel ultrasonic resonance field device for foodborne pathogens—A review. J Food Prot 57(11):1025–
the retention of animal-cells. Biotechnol Prog 10(4):428–432. 1037.
Heddleson RA, Doores S, Anantheswaran RC. 1994. Parameters Lebovka NI, Praporscic I, Ghnimi S, Vorobiev E. 2005. Does
affecting destruction of salmonella spp by microwave-heating. electroporation occur during the ohmic heating of food? J Food
J Food Sci 59(2):447–451. Sci 70(5):E308–E311.
Hendrickx M, Knorr D. 2001. Ultra High Pressure Treatments of Lee H, Zhou B, Feng H, Martin SE. 2009. Effect of pH on
Foods. New York: Kluwer Academic / Plenum Publishers. Inactivation of Escherichia coli K12 by Sonication,
Food engineering series. Manosonication, Thermosonication, and Manothermosonication.
Hernando Saiz A, Tarrago Mingo S, Purroy Balda F, Tonello J Food Sci 74(4):E191–E198.
Samson C. 2008. Advances in design for successful commer- Loginova KV, Shynkaryk MV, Lebovka NI, Vorobiev E. 2010.
cial high pressure food processing. Food Aust 60(4):154–156. Acceleration of soluble matter extraction from chicory with
Hofman J, Shao L, Wols B, Uijtewaal W, Ijpelaar G, Beerendonk pulsed electric fields. J Food Eng 96(3):374–379.
E, van Dijk H. 2007. Design of UV reactors by CFD: Model Lopez N, Puertolas E, Condon S, Raso J, Alvarez I. 2009a.
development and experimental validation. Proceedings of the Enhancement of the extraction of betanine from red beetroot
COMSOL User Conference 2007, Grenoble, France. by pulsed electric fields. J Food Eng 90(1):60–66.
Hoover DG. 1997. Minimally processed fruits and vegetables: Lopez N, Puertolas E, Condon S, Raso J, Alvarez I. 2009b.
Reducing microbial load by nonthermal physical treatments. Enhancement of the solid-liquid extraction of sucrose from
Food Technol 51(6):66. sugar beet (Beta vulgaris) by pulsed electric fields. LWT Food
Hughes D, Nyborg W. 1962. Cell disruption by ultrasound. Sci Technol 42(10):1674–1680.
Streaming and other activity around sonically induced bubbles Martín-Belloso O, Soliva-Fortuny R. 2010. Pulsed electric fields
is a cause of damage to living cells. Science 138:108–114. processing basics. In: H Zhang, GV Barbosa-Cánovas, VM
Jambrak AR, Lelas V, Mason TJ, Kresic G, Badanjak M. 2009. Balasubramaniam, P Dunne, D Farkas, J Yuan, eds., Nonthermal
Physical properties of ultrasound treated soy proteins. J Food Processing Technologies of Food, Ames, IA: Wiley-Blackwell,
Eng 93(4):386–393. 157–175.
Jung S, de Lamballerie-Anton M, Ghoul M. 2000a. Modifications Metaxas AC. 1996. Foundations of Electroheat. Chicester:
of ultrastructure and myofibrillar proteins of post-rigor beef treated J. Wiley & Sons.
by high pressure. LWT Food Sci Technol 33(4):313–319. Metaxas AC, Meredith RJ. 1983. Industrial Microwave Heating.
Jung S, Ghoul M, de Lamballerie-Anton M. 2000b. Changes in London: Peter Peregrinus Ltd.
lysosomal enzyme activities and shear values of high pressure Mudgett RE. 1985. Dielectric Properties of Foods, Microwaves
treated meat during ageing. Meat Sci 56(3):239–246. in the Food Processing Industry, 15–56. Orlando: Academic
Klima J, Frias-Ferrer A, Gonzalez-Garcia J, Ludvik J, Saez V, Press Inc.
Iniesta J. 2007. Optimisation of 20 kHz sonoreactor geometry Mudgett RE. 1986. Microwave properties and heating character-
on the basis of numerical simulation of local ultrasonic inten- istics of foods. Food Technol 40:84–93.
sity and qualitative comparison with experimental results. Nelson SO, Datta AK. 2001. Dielectric properties of food
Ultrason Sonochem 14(1):19–28. materials and electric field interactions. In: AK Datta, RC
Knoerzer K, Juliano P, Gladman S, Versteeg C, Fryer P. 2007. A Anantheswaran, eds., Handbook of Microwave Technology for
computational model for temperature and sterility distributions Food Application, 69–114. New York: Marcel Dekker Inc.
in a pilot-scale high-pressure high-temperature process. AIChE Nicolaï BM, Scheerlinck N, Verboven P, Baerdemaeker JD.
J 53(11):2996–3010. 2001. Stochastic finite-element analysis of thermal food pro-
Knoerzer K, Regier M, Schubert H. 2008. A computational model cesses. In: J Irudayarai, ed., Food Processing Operations
for calculating temperature distributions in microwave food Modeling. Design and Analysis, 265–304. New York: Marcel
applications. Innov Food Sci Emerg Technol 9(3):374–384. Dekker.
Koutchma T. 2009. Advances in ultraviolet light technology for Ohmori T, Shigehisa T, Taji S, Hayashi R. 1991. Effect of high-
non-thermal processing of liquid foods. Food Bioprocess pressure on the protease activities in meat. Agric Biol Chem
Technol 2(2):138–155. 55(2):357–361.
Kowalski S, Mierzwa D. 2009. Convective drying in combination Ohshima T, Ushio H, Koizumi C. 1993. High-pressure processing
with microwave and IR drying for biological materials. Drying of fish and fish products. Trends Food Sci Technol
Technol 27(12):1292–1301. 4(11):370–375.
Kowalski SJ, Rajewska K. 2009. Effectiveness of hybrid drying. Patist A, Bates D. 2008. Ultrasonic innovations in the food indus-
Chem Eng Process 48(8):1302–1309. try: From the laboratory to commercial production. Innov Food
Kozempel MF, Annous BA, Cook RD, Scullen OJ, Whiting RC. Sci Emerg Technol 9(2):147–154.
1998. Inactivation of microorganisms with microwaves at Pehl M, Delgado A. 1999. An in situ technique to visualize tem-
reduced temperatures. J Food Prot 61(5):582–585. perature and velocity fields in liquid biotechnical substances
Kulshrestha SA, Sastry SK. 2006. Low-frequency dielectric at high pressure. Adv High Press Biosc Biotechnol 519–522.
changes in cellular food material from ohmic heating: Effect Pehl M, Werner F, Delgado A. 2000. First visualization of tem-
of end point temperature. Innov Food Sci Emerg Technol perature fields in liquids at high pressure using thermochromic
7(4):257–262. liquid crystals. Experiments Fluids 29(3):302–304.
Puertolas E, Lopez N, Saldana G, Alvarez I, Raso J. 2010. and polygalacturonase in tomato juice by thermosonication.
Evaluation of phenolic extraction during fermentation of red Food Chem 117(1):20–27.
grapes treated by a continuous pulsed electric fields process at Toepfl S, Mathys A, Heinz V, Knorr D. 2006. Review: Potential
pilot-plant scale. J Food Eng 98(1):120–125. of high hydrostatic pressure and pulsed electric fields for
QWED Sp.z o.o. 2003. QuickWave-3D, Manual for the software, energy efficient and environmentally friendly food processing.
Warszawa. Food Rev Int 22(4):405–423.
Rosen CG. 1972. Effects of microwaves on food and related Tolvanen R, Lunden J, Horman A, Korkeala H. 2009. Pilot-Scale
materials. Food Technol 26(7):36. Continuous Ultrasonic Cleaning Equipment Reduces Listeria
Roussy G, Pearce JA. 1995. Foundations and Industrial monocytogenes Levels on Conveyor Belts. J Food Prot
Applications of Microwaves and Radio Frequency Fields. 72(2):408–411.
Chichester: J. Wiley & Sons. Tonello C. 2010. Commercial high pressure plants around the
Sastry SK, Datta AK, Worobo RW. 2000. Ultraviolet light. J Food world; personal communication (13/04/2010). NC Hyperbaric,
Sci Suppl 65(12):90–92. Spain
Schubert H, Regier M. 2005. The Microwave Processing of Turabi E, Sumnu G, Sahin S. 2008. Optimization of baking of
Foods. Cambridge, England: Woodhead Publishing Limited. rice cakes in infrared-microwave combination oven by
Seyderhelm I, Boguslawski S, Michaelis G, Knorr D. 1996. response surface methodology. Food Bioprocess Technol
Pressure induced inactivation of selected food enzymes. J Food 1(1):64–73.
Sci 61(2):308–310. Uysal N, Sumnu G, Sahin S. 2009. Optimization of microwave-
Seyhun N, Ramaswamy H, Sumnu G, Sahin S, Ahmed J. 2009. infrared roasting of hazelnut. J Food Eng 90(2):255–261.
Comparison and modeling of microwave tempering and infra- Vilkhu K, Mawson R, Simons L, Bates D. 2008. Applications
red assisted microwave tempering of frozen potato puree. J and opportunities for ultrasound assisted extraction in the food
Food Eng 92(3):339–344. industry—A review. Innov Food Sci Emerg Technol 9(2):
Sun D-W. 2007. Computational Fluid Dynamics in Food 161–169.
Processing. Boca Raton, FL: Taylor & Francis. Wilson DR, Dabrowski L, Stringer S, Moezelaar R, Brocklehurst
Sun E, Datta A, Lobo S. 1995. Composition based prediction of TF. 2008. High pressure in combination with elevated tempera-
dielectric properties of foods. J Microw Power Electromagn ture as a method for the sterilisation of food. Trends Food Sci
Energy 30(4):205–212. Technol 19(6):289–299.
Sutkar VS, Gogate PR. 2010. Mapping of cavitational activity in Yen GC, Lin HT. 1996. Comparison of high pressure treatment
high frequency sonochemical reactor. Chem Eng J 158(2): and thermal pasteurization effects on the quality and shelf life
296–304. of guava puree. Int J Food Sci Technol 31(2):205–213.
Tang ZW, Mikhaylenko G, Liu F, Mah JH, Pandit R, Younce F, Zhang H, Datta AK, Taub IA, Doona C. 2001. Electromagnetics,
Tang JM. 2008. Microwave sterilization of sliced beef in gravy heat transfer, and thermokinetics in microwave sterilization.
in 7-oz trays. J Food Eng 89(4):375–383. AIChE J 47(9):1957–1968.
Terefe NS, Gamage M, Vilkhu K, Simons L, Mawson R, Versteeg
C. 2009. The kinetics of inactivation of pectin methylesterase
Chapter 2
The Need for Thermophysical Properties in Simulating

Emerging Food Processing Technologies
Pablo Juliano, Francisco Javier Trujillo,

Gustavo V. Barbosa-Cánovas, and Kai Knoerzer
2.1. Introduction Working with thermophysical properties in

Multiphysics modeling of innovative food process-
Multiphysics modeling of any engineering problem ing can result in a number of challenges, mainly:
comprises simultaneously solving partial differen- (1) understanding the nature of the food materials,
tial equations (PDEs) of mathematical expressions (2) accurately determining the properties of the food
representing different types of physical phenomena, materials under specific process conditions, and (3)
that is, in a coupled form (Chen 2006). To solve the understanding the functional dependence of a food
corresponding PDE, it is essential that the underly- material on the process variables (e.g., temperature,
ing equations, thermophysical properties, and pressure, electric field strength, and (ultra)sound
boundary conditions are individually set up for each intensity). Food materials not only have varied
physical phenomenon. As such, the correct expres- composition and structure, but can also change due
sion of the thermophysical properties as functions to processing conditions and during storage (e.g.,
of the process variables affecting them is important biochemical reactions in fresh and living products,
for accurate model prediction (Knoerzer et al. 2007, chemical changes provoked by process variables,
2008, 2010a; Juliano et al. 2009). In turn, accurate changes in moisture content, micro- and macrostruc-
model prediction is challenged by the lack and tural changes, and rheological changes). Thus, the
the uncertainty (variability) of thermophysical overarching challenge is to identify expressions
property values for many materials, including foods for each food material, including the composition
and expressions of the specific process variables, variables (e.g., moisture, ash content, lipids, carbo-
such as temperature, pressure, and concentration hydrates, and protein), structural variables (e.g.,
(Knoerzer et al. 2010a). This chapter is an overview porosity and tortuosity), and processing variables
of the thermophysical properties required for (e.g., temperature and pressure). This chapter will
Multiphysics modeling, and the dependence of these show specifically that thermophysical properties do
properties on technology-specific variables for accu- not vary significantly with process variables other
rate prediction of physical phenomena manifested than temperature and pressure. Seasonal variations
during novel processing of foods. in living food materials such as fruits, vegetables,

23
and meats can hardly be predicted. Therefore, the the mass, momentum, and energy balance processes.
variability of food materials may be unpredictable Fundamental governing equations included in CFD
and difficult to include in a single model, which models comprise the following properties: density
might affect the end result of a Multiphysics model (mass, momentum, and energy conservation equa-
or its predictability for different materials. However, tion), specific heat capacity and thermal conductivity
as long as the composition of the food is included (energy conservation equation), and viscosity
in the expression of thermophysical properties, and (momentum conservation equation). As shown in
models are properly validated, these variations may Table 2.1, equations for other (multi)physics phe-
be negligible. nomena to be coupled with CFD models require
further thermophysical properties such as the com-
2.2. Definitions and Methods to pression heating properties (relevant in high-pressure
Determine Thermophysical Properties processing), dielectric properties (radio frequency
and microwave processing), electrical conductivity
This section will define the thermophysical proper-
(ohmic heating and pulsed electric fields), sound
ties required to establish Multiphysics models for
absorption coefficient and velocity of sound (ultra-
innovative food processes. In particular, suitable
sound processing), and absorptivity (ultraviolet
properties will be selected for processes requiring
processing).
the use of temperature, elevated pressure, electric
fields, ultrasound irradiation, microwave processing
2.2.1. Density, Porosity, and
and ultraviolet light, and the eventual inclusion of
Related Properties
material and process variables. When food compo-
nents or other materials are included in the model, The density, ρ, of a material is defined as mass per
empirical expressions that have been adjusted to unit volume (SI unit of density is kg/m3). Indeed,
measured data may include other material variables there are different forms of density that can be used,
(e.g., composition and porosity) as they intervene in such as true, material, particle, apparent, and bulk
Table 2.1. Summary of essential properties for emerging food processing technologies.
Technology Property Equation

All food processing Density Mass, momentum,
technologies Specific heat capacity Energy conservation equation
Thermal conductivity Momentum conservation equation
Viscosity
High-pressure processing Thermal expansion coefficient Compression heating (Eq. 2.10; Chapters 4
(HPP) and High-pressure Compressibility and 5)
thermal sterilization (HPTS) Compression heating coefficient
Microwave and radio Dielectric constant Maxwell’s equations including the constitutive
frequency Loss factor relations (Chapters 6 and 7) and energy
Electrical conductivity conservation including source term
Ohmic heating and pulsed Electrical conductivity Charge conservation and energy conservation
electric fields including source term (Chapters 8, 9, 11)
Ultrasound Sound absorption coefficient Wave equation, Helmholtz equation,
Velocity of sound momentum, and energy conservation
including source terms (Chapter 12)
Ultraviolet Absorptivity Radiation intensity (Chapter 14)
Absorption coefficient (or
spectral absorption coefficient)
Optical density
Chapter 2 Thermophysical Properties in Simulating Food Processing Technologies 25
density, depending on its application in process cal- volume, and is termed “expansion.” Two types of
culations or product characterization. However, the shrinkage, isotropic and anisotropic, are usually
apparent density is more commonly used as input observed in the case of food materials. Isotropic
into the model equations. The volume measurement shrinkage is described as the uniform shrinkage of
method is what determines the difference between materials under all geometric dimensions, whereas
them. True and material densities are calculated by anisotropic (or nonuniform) shrinkage develops
excluding volumes occupied by internal and exter- in different geometric dimensions. The former is
nal pores within the food, while particle, apparent, common in fruits and vegetables, while the latter is
and bulk densities are determined from less accurate known in animal tissue, as in meat and fish (Sahin
measurement methods that include pore volume and Sumnu 2006). Most of the density, shrinkage,
(Barbosa-Cánovas et al. 2005). and porosity prediction models for liquid and solid
In most engineering designs, solids and liquids foods are empirical in nature. Recent models have
are assumed to be incompressible—in other words, been developed to predict porosity during air drying
density changes moderately with changes in tem- based on drying temperature, moisture content,
perature and pressure. In food engineering, however, initial porosity, and product type (Lozano 2007).
the density of solid and liquid foods changes with Volume change and porosity are important parame-
temperature and pressure, and composition changes ters in estimating diffusion coefficients for shrinking
as well. In the case of liquid foods, no generic equa- systems. Furthermore, porosity and tortuosity are
tions exist to predict the density. In the literature most used to calculate effective diffusion during the mass
of the density data are correlated empirically as a transfer process (Knoerzer et al. 2004).
function of temperature, pressure, water, solids, and A material’s volume can be measured by buoyant
fat content. Different types of nonlinear correlation force, liquid, gas or solid displacement, gas adsorp-
such as exponential, quadratic, and cubic are used to tion, or by estimating the material’s geometric dimen-
relate density and moisture content (Lozano 2007). sions. The buoyant force method for apparent or
Another way of accounting for a material’s struc- particle volume determination utilizes sample weight
ture is in its porosity, which indicates the volume differences in air and water, while the liquid displace-
fraction of void space or air space inside the material. ment method measures the increase in liquid volume
Volume determination is relative to the amount of (material is immersed in a non-wetting fluid such as
internal (or closed) or external (or open) pores mercury or toluene). A gas pycnometer is a gas dis-
present in the food structure. Therefore, like density, placement device that uses air pressure differences in
different forms of porosity are also used in food a sample cell connected to a manometer to determine
processing studies, namely open pore, closed pore, material volume. Apparent or particle density can be
apparent, bulk, and total porosities (Rahman 1995; determined by coating particles in order to include
Barbosa-Cánovas et al. 2005, 2007). Porosity in internal pores in the volume measured. For solid
foods is mainly predicted from empirical correla- displacement, sand or glass beads can be used instead.
tions, which are valid for individual foods under Porosity can be measured by direct and microscopic
given processing conditions. Fundamental models methods, or can be estimated from density data.
exist that are based on the conservation of mass and
volume, as well as a number of other terms that
2.2.2. Viscosity
account for interaction of components and formation
or collapse of air or the void phase during processing The viscosity ( μ ) is a physical property of fluid
(Lozano 2007). materials (gases, liquids, or semisolid foods) and
Shrinkage or the reduction in volume or geomet- represents the internal friction of a fluid or its resis-
ric dimensions during processing is also important tance to flow (the SI unit of dynamic viscosity is Pa·s)
in solid materials. During post-processing the (Bourne 2002). Similar to the friction that occurs
volume of the material is larger than its initial between moving solids, viscosity transforms the
kinetic energy of motion of the fluid into heat. For Bingham model
instance, highly viscous food materials like honey
τ = τ o + η pγ (2.4)
offer higher resistance to flow than lower viscosity
liquids such as water. Thus, more power is consumed Herschel-Bulkley model
by the pump during pumping of honey to achieve
τ = τ o + K (γ )
n
(2.5)
the same flow rate of pumping water. In addition,
honey exhibits a higher temperature increase under where γ = dv dy is the shear rate and η p is apparent
flowing conditions than water due to higher viscous viscosity.
heat dissipation (caused by friction). For a pure substance, the viscosity is highly
Modeling food-processing operations often dependent on temperature and to a lesser extent on
involves solving the momentum transport equation, pressure. For complex materials like food, the vis-
which accounts for the balance between the forces cosity also depends on composition. Viscosity is
(e.g., pressure gradients and stress tensor) applied to usually determined by measuring the resistance to
a differential volumetric element of a fluid, and the flow in a capillary tube, or the torque produced by
resulting acceleration of the fluid. Viscosity is the movement of an element through the fluid. There
usually incorporated into the rheological model as are two main categories of viscometers applicable
representing the stress tensor, which accounts for the to foodstuffs: capillary, and falling ball, as well as
shear stress acting upon the fluid. Shear stress (τ ) commercial rheometers able to provide viscosity
is a force per unit of area acting parallel or tangential according to rheological models.
to the fluid element: For Newtonian liquid foods it is sufficient to
measure μ at a single value of γ . In order to describe
F
τ= (2.1) a non-Newtonian food, additional properties must be
A measured by attaining flow curves with a rheometer
Simple liquids and gases such as water and air are and determining yield stress. Viscoelastic and semi-
called Newtonian fluids, given that they exhibit an solid foods have been extensively studied during the
ideal correlation between the shear stress (force per last decades. Rheological characterizations of non-
unit area acting tangential to the fluid) and the Newtonian foods have been in the form of τ versus
resulting velocity gradient perpendicular to the γ curves, dynamic characteristics, time effect on
direction of shear: viscosity at constant temperature, and others. Values
of these parameters have been compiled by different
dv authors (Kokini 1992; Steffe 1992; Rao 2007).
τ=μ (2.2)
dy Liquid foods, such as beer, tea, coffee, clarified
fruit juice, wines, cola drinks, vegetable oils, and
For a Newtonian fluid, viscosity represents a linear
milk exhibit Newtonian behavior. As an approxima-
correlation between shear stress and velocity gradi-
tion, viscosity of Newtonian foods can be estimated
ent. However, foods are structurally complex mate-
as the weighted average between the viscosity of
rials that frequently exhibit non-Newtonian behavior
water ( μ w ) and that of the prevalent soluble sub-
(Tabilo-Munizaga and Barbosa-Cánovas, 2005;
stance. Different empirical equations relating liquid
Welti-Chanes et al. 2005). Non-Newtonian fluids are
food viscosity with both soluble solids and tempera-
usually divided into the following general classes:
ture have been published (Rao 2007). The viscosity
(1) those with properties independent of shear rate
of salt and sugar solutions (two major food solutes)
and (2) those with properties dependent on shear
are also available (Kubota et al. 1981). Vitali and
rate (Steffe 1992). Some common rheological
Rao (1984) reported that the effect of concentration
models of liquid foods are
on viscosity of fruit juices at constant temperature
Power law model
can be represented by an exponential-type relation-
τ = K (γ )
n
(2.3) ship. Hydrolytic enzymes present in natural fruit and
vegetable juices, or purees, degrade polysaccharide Several methods are known for measuring
chains and therefore alter the viscosity. specific heat capacity experimentally. Cp can be
determined by methods of mixtures and differential
2.2.3. Specific Heat Capacity scanning calorimetry (DSC). For methods of mix-
tures, a calorimeter of known specific heat is used
The specific (isobaric) heat capacity, Cp, is the amount and CP is determined from a heat exchange balance.
of heat (in joules) needed to raise the temperature of In the DSC method, the sample is put in a special
1 kg of matter by 1 K at a given temperature. The SI cell where the temperature is increased at a constant
unit for Cp is therefore J/kg/K. Specific heat capacity heating rate. The specific heat of the food is obtained
of solids and liquids depends on temperature but does from a single heat thermogram, which records heat
not generally exhibit pronounced pressure dependence. flow as a function of time or temperature.
It is common to use the constant pressure specific
heat, Cp, which thermodynamically represents the 2.2.4. Thermal Conductivity and Diffusivity
change in enthalpy H (kJ/kg) for a given change in
temperature T when it occurs at pressure P: Thermal conductivity, k, is the property of a material
indicating its ability to conduct heat. It represents
⎛ ∂H ⎞ the quantity of heat Q that flows per unit time
CP = ⎜
⎝ ∂T ⎟⎠ P
(2.6)
through a food of a certain thickness and area with
Assuming there is no phase change, the amount of a specific temperature difference between faces; the
heat Q that must be added to a unit mass M (kg of SI unit for k is W/m/K. The rate of heat flow Q
mass or specific weight kg/m3) to raise the tempera- through a material by conduction can be predicted
ture from T1 to T2 can be calculated using the fol- by Fourier ’s law of heat conduction. A simplified
lowing equation: approximation follows:
Q = M ⋅ CP ⋅ (T2 − T1 ) k ⋅ A ⋅ (T2 − T1 )
(2.7) Q = (2.8)
x
The specific heat of foods is drastically influenced
by water content. For example, specific heat has where A is the surface area of the food, x is its thick-
been found to vary exponentially with water content ness, T1 is the temperature at the outer surface where
in fruit pulps above ambient temperatures. heat is absorbed, and T2 is the temperature at the
Furthermore, nonaqueous components show lower inner surface.
Cp (Barbosa-Cánovas et al. 2007; Lozano 2007). Thermal diffusivity α (SI unit, m2/s) defines the
The specific heat of oils and fats is usually about rate at which heat diffuses by conduction through a
one-half the specific heat of water, while the specific food composite and is related to k and Cp through
heat of dry materials in grains and powders is density ρ as follows:
approximately one-third to one-fourth that of water k
(Rahman 1995). As a result of solute–water interac- α= (2.9)
ρ ⋅ CP
tions, the Cp of each individual component in a food
differs from the Cp of a pure component, and usually Thermal diffusivity establishes the speed of heat of
changes with the concentration of soluble solids. Cp three-dimensional propagation or diffusion through
has been measured at different temperatures in fresh the material. It is represented by the rate at which
and dried fruits, meats, cereal grains and cereal temperature changes in a certain volume of food
products, oils and fats, powders, and other dry foods material, while transient heat is conducted through it
(Lozano 2007). Although linear correlations of Cp in a certain direction in or out of the material (depend-
with concentration are known in liquid foods, varia- ing on whether the operation involves heating or
tions are often neglected for engineering calcula- cooling). Equation 2.9 shows that α is directly
tions at near room temperature. proportional to the thermal conductivity at a given
density and specific heat. Physically, it relates the use of a thermal conductivity probe to measure a
ability of the material to conduct heat to its ability to temperature–time relation on a thin cylindrical food
store heat (Barbosa-Cánovas and Rodríguez 2005). piece to which constant heat is applied.
The thermal conductivity of food materials is Thermal diffusivity α is usually either found by
greatly influenced by the water content. Water shows direct experimental methods or estimated through
greater relative magnitudes in comparison to other Equation 2.9. Several direct methods for α determi-
food constituents. Thus, k increases with increased nation can be based on a one-dimensional heat
moisture content. It is common to find a linear rela- conduction equation where geometrical boundary
tionship between thermal conductivity and moisture conditions are defined. For instance, an apparatus
content at ambient conditions, but also a quadratic can be used where the sample is located in a special
relationship, as well as multiple correlations of cylinder and immersed in a water bath at constant
moisture, temperature, and composition can be temperature. Thermocouples located at the center of
found for k in food materials (Lozano 2007). Some the sample (axis) and surface of cylinder measure
models consider that different components of foods temperature at different heating times. Transient
(e.g., fibers) are arranged in layers either parallel or temperature variations are used for the analytical
perpendicular to the heat flow (Salvadori et al. solution. Indirect methods, although they might
1997). For example, in products such as meat, heat yield more accurate diffusivity values, require more
is usually transferred parallel to the fibers and k is time and instrumentation for the three-parameter
dependent on the direction of the heat flow. More determination (ρ, k, and CP; Lozano 2007).
general in nature are the randomly distributed It is worth mentioning the role of the surface heat
models, which consider that the food is composed transfer coefficient, as it is one of the important
of a continuous phase with a discontinuous phase parameters necessary for design and control of food
dispersed within (solid particles being in either processing and associated equipment where fluids
regular or irregular arrangement; Mattea et al. 1989; (air, nitrogen, steam, water, or oil) participate.
Lozano 2007). In porous materials, porosity must be Although the surface heat transfer coefficient is not
included in the model because air has a thermal a property of food, it is used to quantify the transfer
conductivity much lower than that of other food rate of heat by convection from a liquid or a gas
components. Models including density or porosity, (especially boiling liquids and condensing vapors)
and pressure, have been developed for fruits and to the surface of foods. It plays an important role
vegetables, meat and meat products, dairy products, when evaluating the effectiveness of heat transfer in
cereals, and starch (Lozano 2007). Several models processes where hot water or steam is applied
for predicting α in foods have also appeared in the through the evaluation of the overall resistances
literature; however, most are product-specific and a during heat transfer (Juliano et al. 2008).
function of water content (sometimes water activity)
or temperature. Although the influence of carbohy-
2.2.5. Compression Heating Coefficient
drates, proteins, fat, and ash on thermal diffusivity
and Related Properties
has also been investigated, it was found that tem-
perature and water content are the major factors All materials change their volume when subjected to
affecting α (Rahman 1995). temperature or pressure change. For food processing
Experimental methods used to determine k are, operations it is commonly assumed that pressure
for example, the Fitch method and the line source does not appreciably affect the volume of a liquid or
method. In the Fitch method, a solid slab of a certain the solid objects; however, in the case of food pro-
food receives heat from one layer and conducts it cessing under high hydrostatic pressure up to several
to a copper plug. Conductivity k is obtained from hundred MPa, this assumption does not hold true. A
the food’s temperature as a function of heat conduc- significant compression of fluids and some solids
tion time. The line source method is based on the (especially polymeric materials used, e.g., as food
packaging) occurs when materials are subjected to the thermal expansion coefficient of food materials
these pressure levels. The adiabatic temperature is available in the literature. For example, values for
change of an isotropically compressed or a decom- sunflower and olive oils, as well as tomato paste and
pressed material can be expressed as follows: pressure-transmitting fluids, have been reported by
Guignon et al. (2009, 2010) and Aparicio et al. (2010),
dT αP
= ⋅ T = kC ⋅ T (2.10) respectively. Min et al. (2010) have determined the
dP ρ ⋅ CP compressibility (and density) of selected liquid and
where α P is the thermal expansion coefficient and solid foods as they vary with increasing pressure.
kC is the compression heating coefficient. The coef- As shown in Equation 2.10, all compressible
ficient of thermal expansion describes how the materials undergo a change in temperature when
volume of an object changes with change in tem- subjected to pressure. The degree of temperature
perature. In particular, it measures the fractional change is hereby dependent on a complex interaction
change in volume per degree change in temperature of the thermal expansion coefficient, density, and
at a constant pressure; the SI unit is K−1. Likewise, specific heat capacity of the material. It is challenging
the compressibility is a measure of the relative to determine these properties separately under high-
volume change of a fluid or solid as a response to a pressure conditions. Therefore, for predicting the
pressure (or mean stress) change; the SI unit is Pa−1. extent of compression heating during a high-pressure
As indicated earlier, all materials expand or con- process the pressure–temperature-dependent proper-
tract when their temperature or pressure changes, ties can be combined into one pressure–temperature-
and the expansion or contraction always occurs in dependent parameter, referred to as the compression
all directions. This change in volume occurs at the heating coefficient kC (Knoerzer et al. 2010a, 2010b)
same rate in any direction in isotropic materials. with the SI unit Pa−1. For modeling high-pressure
Some materials expand when cooled, such as freez- processes, knowledge of these properties is impera-
ing water, and therefore have negative thermal tive for all materials involved in the modeled scenario
expansion coefficients (Knoerzer et al. 2010a). (see Chapters 4 and 5 for further information).
The adiabatic compressibility β S is a thermody- The variation of compression heating of water
namic property and can be written at constant based on thermophysical properties from the NIST
entropy as: database for water and steam (Harvey et al. 1996)
has been summarized in the literature (Ardia et al.
1 ⎛ ∂V ⎞ 2004; Knoerzer et al. 2007; Mathys and Knorr
βS = −
V ⎜⎝ ∂p ⎟⎠ S
(2.11)
2009). Limited information is available on the com-
pression heating for food materials. Studies on
The thermal expansion coefficient α P is a thermo- selected food materials have been published in the
dynamic property and can be written at a constant last few years (Otero et al. 2000, 2006; Rasanayagam
pressure as: et al. 2003; Ardia et al. 2004; Patazca et al. 2007;
1 ⎛ ∂V ⎞ Shao et al. 2007; Zhu et al. 2007).
αP = ⎜ ⎟ (2.12) Until recently, the extent of compression heating
V ⎝ ∂T ⎠ P
of nonfood solid materials in high-pressure process-
where V is the volume of the material and ∂V/∂T is ing was unknown and assumed negligible. This
the rate of change of that volume with pressure and assumption holds true for metals, which not only
temperature, respectively. have lower thermal expansion coefficients, but at the
In general, the thermal expansion coefficients same time exhibit significantly higher densities,
increase from solids over liquids to gases (Harvey et making the kC values small compared with those of
al. 1996). The variation of this parameter has been liquids. However, it was shown by Knoerzer et al.
studied for water from the NIST database by Juliano (2010b) that some polymeric insulating plastics
et al. (2008). However, very limited information on undergo pronounced heating under pressure, often
exceeding the adiabatic heating of water. This can • the real part, dielectric constant, ε', related to the
be explained by the fact that these materials often capacitance of a substance and its ability to store
have lower densities, but particularly lower specific electrical energy; and
heat capacities, and exhibit greater compressibility, • the imaginary part, dielectric loss factor, ε",
associated with greater thermal expansion capacity. related to the attenuation of the electromagnetic
Also for liquids, pronounced differences in kC energy, dissipated into thermal energy when the
values can be found, as shown in Knoerzer et al. food is subjected to an alternating electrical field
(2010a), associated with differences in molecular (i.e., dielectric relaxation and ionic conduction).
bonds, that is, hydrogen and van der Waal’s bonds
affecting the thermal expansion coefficients. Water, Both parameters are dimensionless because they are
for example, shows less pronounced heating than relative values (Chapters 6 and 7).
nonpolar liquids, such as propylene-glycol, with the Dielectric properties (ε', ε") are primarily deter-
difference being that the values are increasing for mined by the food’s chemical composition (pres-
water with increasing pressures and temperatures; ence of mobile ions and permanent dipole moments
whereas nonpolar liquids show greatest compres- associated with water and other molecules) and, to
sion heating at low pressures, and almost no tem- a much lesser extent, by their physical structure
perature dependency (Knoerzer et al. 2010a, 2010b). (Barbosa-Cánovas et al. 2007). The influence of
The compression heating coefficient of food water and salt (or ash) content largely depends on
materials can be derived from the temperature– the manner in which they are bound or restricted in
pressure profiles obtained in an adiabatic high-pressure movement by other food components. Free water
system. There are certain types of thermocouples and dissociated salts have high values of the dielec-
(i.e., K and T types), which can reliably measure tric properties, while bound water, associated salts,
temperature at high-pressure conditions. A detailed and colloidal solids exhibit lower values. Power
description of this method can be found elsewhere dissipation is directly related to the dielectric loss
(Knoerzer et al. 2010a, 2010b). factor ε" (see also Chapters 6 and 7) and temperature
Thermodynamic equations are able to predict the increase during microwave processing (as per the
thermal expansion coefficient and the compressibil- energy conservation equation) further depends on
ity of pure water and gases. However, food materials the specific heat of the food, the thermal conductiv-
are more complex and their values cannot be derived ity, and the density of the material.
theoretically. As long as density and heat capacity Permittivity also strongly depends on the fre-
are known as functions of temperature and pressure, quency of the applied alternating electric fields.
the thermal expansion coefficient can be calculated Frequency contributes to the polarization of (polar)
from the compression heating of the food material molecules such as water. In general, the permittivity
according to Equation 2.10. Very recently, a volume increases with temperature, whereas the loss factor
piezometer has been integrated into a high-pressure may either increase or decrease depending on
system to provide in situ data for compressibility the operating frequency (Mohsenin 1984; Regier
and density (Min et al. 2010). and Schubert 2005). In microwave processing at
frequencies of either 915 MHz or 2.45 GHz, the loss
factor decreases with increasing temperature, caused
2.2.6. Dielectric Properties
by a greater mobility of ions and dipoles due to
The dielectric permittivity, ε, is a complex number lower viscosities (Regier and Schubert 2005).
used to explain interactions of foods with electric Comprehensive tabulations of electrical property
fields. It determines the interaction of electromag- data are available for foods (Zhang 2007).
netic waves with matter and defines the charge The electrical field inside the food is determined
density under an electric field. In solids, liquids, and by the dielectric properties and geometry of the load
gases the complex permittivity comprises two values: and the food processing chamber configuration.
Dielectric properties are of great importance in in thermal properties, the porosity of the food plays
measuring and heating applications, and also in the an important role in the conduction of electrons
selection of proper packaging materials and in the through the food.
design of microwave and radio frequency heating Electrical properties are important when process-
equipment since they influence how the material ing foods with pulsed electric fields, ohmic heating,
interacts with the electromagnetic waves. They induction heating, radio frequency, and microwave
furthermore determine the penetration depth of elec- heating (Chapters 6–11). Electrical conductivity
tromagnetic waves into foods. plays a fundamental role in ohmic heating, a process
Known methods for measuring dielectric proper- in which electricity is transformed into thermal
ties are the cavity perturbation, open-ended coaxial energy when an alternating current (AC) is applied
probe, and transmission line methods. Since modern to the food. Ohmic heating has potential use in
microwave network analyzers have become avail- fluid pasteurization; hence, knowing the effective
able, the methods of obtaining dielectric properties electrical conductivity or the overall resistance of
over wide frequency ranges have become more effi- liquid–particle mixtures is important. Liquids and
cient. Computer control of impedance and network liquid–particle mixtures can also be pasteurized with
analyzers has facilitated the automatic measurement pulsed electric fields technology. In this case, prod-
of dielectric properties; special calibration methods ucts with low electrical conductivity are better and
have also been developed to eliminate errors caused more energy-efficient to process, unless a synergy of
by unknown reflections in the coaxial line systems. heat and electric field strength is assumed; then
Distribution functions and empirical relationships higher electric conductivities assist in heating up the
can be used to express the temperature dependence liquid from moderate initial temperatures to the
of dielectric properties. process target temperature.
The electrical conductivity of a material is gener-
ally measured by passing a known current at constant
2.2.7. Electrical Conductivity voltage through a known volume of the material and
by determining resistance. The total conductivity is
The electrical conductivity, σ, is a measure of how then calculated simply by taking the inverse of the
well electric current flows through a material with a total resistivity. Basic measurements involve bridge
certain cross-sectional area A, length L, and resis- networks (such as the Wheatstone bridge circuit) or
tance R. It is the inverse value of electrical resistivity a galvanometer. There are other devices that measure
(measure of resistance to electric flow), as expressed electrical conductivity of foods under ohmic or con-
in the following equation (Regier and Schubert ventional heating conditions, using thermocouples
2005; Zhang 2007): and voltage and current transducers to measure
L voltage across and current through samples (Zhang
σ= (2.13) 2007).
A⋅ R
The electrical conductivity (SI unit, S/m) of foods
has been found to increase with temperature and
2.2.8. Acoustic Properties
also with water and ionic content. Mathematical
relationships have been developed to predict the Modeling of sound fields is gaining importance in
electrical conductivity of food materials (Buckow the food industry given several new applications of
et al. 2010). Below freezing temperatures, electrical power ultrasound, such as enzyme activity modula-
conductivity shows a pronounced decrease, since ice tion and enhanced extraction. The most important
conducts less well than water. Phase transitions in properties of a fluid, utilized to model sound fields,
foods (such as starch gelatinization) and cell struc- are speed of sound ( c ) and attenuation coefficient
tural changes also affect electrical conductivity. As ( α ).
2.2.8.1. Speed of Sound The speed of sound is of sound. More realistic approaches include the
the rate at which an acoustic wave travels through a effect of sound scattering, a deviation or reflection
fluid. This thermodynamic property of the fluid of sound at the phase boundary. The two most
depends on the following equilibrium conditions: important sources of sound scattering, in the long
wavelength limit (particle radius of dispersed phase
⎛ ∂p ⎞ r << λ wavelength of ultrasonic wave), are visco-
c2 = ⎜ ⎟ (2.14)
⎝ ∂ρ ⎠ adiabatic inertial and thermal scattering (McClements et al.
1990). Visco-inertial scattering occurs due to differ-
For an ideal gas, Equation 2.14 takes the form:
ences of density between the phases, whereas thermal
c 2 = γ RT (2.15) scattering is caused by differences in compressibil-
ity, heat capacity, and thermal diffusion between the
where γ = ( c p cv ) , R is the universal constant, and
phases. This will be explained further in Section
T is the absolute temperature (Lighthill 1978). For
2.2.8.2. Ament (1953) developed an equation of
an ideal gas, c depends only on temperature. For
sound propagation through emulsions, accounting
liquids, the speed of sound can be expressed as:
for visco-inertial scattering. The Ament formula can
1 be plotted in a three-dimensional figure to visualize
c2 = (2.16)
βρ the dependence of the speed of sound with the
volume fraction and Γr . The latter is the product of
where β is the adiabatic compressibility. β and ρ the shear wave skin depth ( Γ ), at which the ampli-
vary with the equilibrium temperature and pressure tude of the wave has diminished to 1/e of its original
of the liquid. Hence, the speed of sound of pure value, and particle radius of the dispersed phase
liquids depends on temperature and pressure. Since (r). It has been shown that for the region where
there is no simple theory for predicting these vari- Γr ≈ 1 the speed of sound is frequency-dependent
ables, this dependence must be measured experi- and in those regions where this condition does not
mentally; the resulting values are usually expressed apply ( Γr ≠ 1) the speed of sound is frequency-
as empirical equations (Shoitov and Otpushchennikov independent (Povey and McClements 1998).
1968). As any other thermodynamic property, the The speed of sound in liquids can be strongly
speed of sound is a function of composition for influenced by the presence of bubbles, especially for
mixtures. cavitating systems, where bubbles are generated by
In nondispersive media, acoustic waves travel at pressure changes during the propagation of high-
the same speed of sound regardless of frequency intensity ultrasonic waves in liquids. For example,
(Leighton 1997). However, in a dispersive medium the speed of sound in a cavitating liquid may decrease
the speed of sound can vary with frequency from 1,500 m/s (in the case of pure water) to 20 m/s,
(McClements and Povey 1989). This may occur in which is even less than the velocity of the speed of
food emulsions where a secondary phase is dis- sound in air (340 m/s; Servant et al. 2001). Systems
persed in a continuous phase. In an ideal case, that containing bubbles may exhibit resonant scattering,
is, a non-scattering two-phase system, the volume where attenuation is sharply increased near the reso-
average values for compressibility and density can nant frequency. In resonant scattering the magnitude
be used (McClements et al. 1990): of the scattered wave is of the same order as the
β = (1 − φ ) β1 + φβ2 (2.17) incident wave. Multiple scattering of the bubbles has
a pronounced effect on the speed of sound. In this
ρ = (1 − φ ) ρ1 + φρ2 (2.18)
case, the speed of sound and the attenuation are a
where φ is the dispersed phase volume fraction and complex function of the frequency and number and
the subscripts 1 and 2 represent the continuous and size of bubbles (McClements and Povey 1989).
dispersed phases, respectively. Then, the mixture The speed of sound is generally determined by
values of β and ρ can be used to calculate the speed measuring the acoustic transit time over a known
path length. For instance, a receiver acoustic intermolecular level processes, such as conversion of
ferometer allows measuring the speed of sound by kinetic energy of molecules into stored potential
knowing the distance traveled by the sound and the energy formed by structural rearrangement of clus-
phase shift between received and transmitted signals ters, and rotational and vibrational energies of mol-
(Barthel and Nolle 1952; McCartney and Drouin ecules (Kinsler et al. 2000).
1977). The pulse-echo-overlap (PEO) method sends For multiphase systems such as emulsions, the
a pulsed sound from the transmitter to a reflector. energy of the ultrasound wave may be scattered
The signal on its way back again excites the trans- at the interface between primary and secondary
ducer and a new echo is sensed. The process is phases. When ultrasound is scattered, the ultrasound
repeated until the wave is completely damped. The energy is redirected away from the incident wave
speed of sound is obtained by measuring the time (McClements and Povey 1989). A first ideal
between the maximum of two consecutive echoes approach assumes that there is no interaction
(Junquera et al. 2002). Broadband resonance tech- between the phases. Thus, the attenuation of a two-
niques have also been widely used to determine phase non-scattering system is equal to the volume
speed of sound and to characterize liquids (Eggers average of the absorption coefficients:
and Kaatze 1996). Unfortunately, in spite of current
α = (1 − φ ) α1 + φα 2 (2.20)
technologies capable of accurately determining the
speed of sound in fluids, and the increased use of However, in real emulsions, scattering of ultrasound
ultrasound to characterize food materials, there is a has a pronounced effect on both speed of sound and
lack of reported data on speed of sound of food attenuation (McClements et al. 1990). The difference
materials in the public domain. between the overall attenuation ( α ) and that caused
by absorption alone ( α o ) is termed “excess attenu-
2.2.8.2. Sound Attenuation Coefficient Sound ation” ( α exc ), which is further defined as a combina-
attenuation is the loss occurring during the propagation of various scattering mechanisms (McClements
tion of sound waves. The attenuation coefficient and Povey 1989):
represents the damping of the initial pressure ampli-
α = α o + α exc (2.21)
tude P0 of an ultrasonic wave propagating along a
distance r (Lamberti et al. 2009): Overall attenuation is usually described with the
simplified equation below (Eggers and Kaatze 1996):
P = P0 e −α r (2.19)
α ( f ) = Bf 2 + α exc ( f ) (2.22)
The sound attenuation at ultrasonic frequencies in a
pure liquid is caused by absorption mechanisms that where f is frequency, Bf = α o is the attenuation due
2
convert energy from the ultrasonic wave into heat to absorption alone, and B is a constant. As men-
(McClements et al. 1990). In liquid, the most impor- tioned before, the two main sources of scattering in
tant mechanisms are viscosity, thermal conduction, the long wavelength limit ( r << λ ) are visco-inertial
and molecular relaxation. As explained in Section ( α vis ) and thermal scattering ( α th ). Visco-inertial
2.2.2, viscosity transforms the fluid kinetic energy scattering occurs when droplets have a different
of motion into heat. Hence, viscous losses occur density compared with the surrounding fluid. In the
when there is a relative motion between adjacent presence of an ultrasonic wave a net force causes
portions of the fluid upon passage of sound waves. the droplet to oscillate; the oscillation is damped by
Besides that, the temperature of the fluid locally the viscosity of the surrounding fluid. In thermal
increases during compression and decreases during scattering, differences in compressibility, heat capac-
expansion. Thermal conduction losses occur when ity, and thermal diffusion between the droplets of the
heat flows from the compressed and “hot” parts of dispersed phase and the continuous primary phase
the medium to the adjacent and expanded “cold” cause the ultrasound wave to be scattered (Povey and
parts. Molecular relaxation losses are caused by McClements 1998). When the compressibility of the
particles differs from the surrounding liquid, the par- absorber of electromagnetic radiation (c, mol/L),
ticles and the liquid expand differently under the and path length of light (d, m).
influence of the ultrasonic wave. This difference
A
creates a temperature gradient at the particle–liquid ε= (2.23)
interface. Energy flowing at this boundary is c⋅d
irreversible according to thermodynamic principle In spectroscopy, the absorbance A (optical density)
(Dukhin et al. 2005). The thermal scattering of a is defined as:
dispersion is a function of the thermal expansion,
I
heat capacity, and thermal conductivity of both A = − log (2.24)
phases, and also depends on the size distribution of I0
the particles. where I is the intensity of light at a specified wave-
As explained in Section 2.2.8.1, systems contain- length that has passed through a sample (transmitted
ing bubbles may exhibit multiple strong scattering, light intensity) and I0 is the intensity of the light
which not only has a pronounced effect on the speed before it enters the sample (incident light intensity).
of sound, but also affects the overall attenuation of The liquid itself and the concentration of the sus-
the sound wave. Therefore, sound fields can be pended units can be transparent if A << 1, opaque if
strongly attenuated a few millimeters from the A >> 2 or semitransparent for anything in between
source in high-power horn reactors due to the pres- these extremes. In a majority of cases, liquid foods
ence of cavitation bubbles (see Chapter 12; Trujillo will absorb UV radiation.
and Knoerzer 2009). The absorptivity of a pure substance is deter-
Similar to the speed of sound, attenuation can also mined by means of spectrophotometric analysis. In
be experimentally determined via standard methods particular, since this property is utilized to model
such as ultrasonic pulse techniques (McSkimin UV processing, the absorptivity is determined
1957). Chapter 12 also explains a method of deter- according to the operational frequency. Generally, a
mining the absorption coefficient (only absorption standard curve of the absorbance versus concentra-
without scattering) based on measuring acoustic tion of a pure substance is determined and the slope
streaming. However, also for the absorption coeffi- of the curve defines the absorptivity. Chemical hand-
cients, there is a lack of experimental data reported books such as the Merck Index (Merck Publishing
in the public domain. Hence, attenuation measure- Group, Rahway, NJ) also provide values of absorp-
ments are needed for future modeling of ultrasound- tivity for standard materials and food additives. In
based processes. For cavitating systems, there are the case of liquids containing heterogeneous mix-
still many complexities that need to be solved, given tures or several substances absorbing at the same
the strong scattering produced by bubbles formed frequency, an overall absorptivity can be determined
during cavitation. either empirically or by application of mixture rules.
2.2.9. Molar Absorptivity

2.3. Final Remarks and Future
The molar absorptivity of a substance has a particu-
Recommendations
lar application in the modeling of UV light process-
ing. It is the main optical characteristic observed in Accurate model prediction requires implementation
all liquid foods undergoing UV treatment. The molar of the thermophysical properties of all materials
absorptivity (ε, L/mol/m) of the absorbing species involved in a Multiphysics model scenario. The
is a measure of the amount of light absorbed per unit use of constant values for the respective properties
concentration, or optical density and forms part of generally yields only poor agreement with predic-
the Lambert-Beer ’s law (Eq. 2.23) in a linear relations from actual measured data. Therefore, includ-
tionship between absorbance A, concentration of an ing the functional dependence of the properties on
temperature, pressure, and product-specific param- emulsions), the speed of sound and attenuation
eters and variables (e.g., composition and structure) depend on size and distribution of particles or sec-
is essential for accurate model prediction. It has ondary phase droplets. For systems under cavitation,
been shown that during processing (by technologies the strong scattering caused by bubbles dramatically
explored in this book) thermophysical properties are changes both the speed of sound and attenuation.
mainly affected by temperature, pressure, and Therefore, these two properties are a complex func-
changes in the process and product parameters, tion of frequency, and cavitation extent. Given that
although product parameters such as composition cavitation bubbles are produced (grow and collapse)
and structural porosity, as a matter of simplification, close to the acoustic source where sound intensity
are generally assumed to be constant. is high, the speed of sound and the attenuation
When modeling these processing technologies become dependent on the sound field.
the governing equations of mass, momentum, and In addition, the relevant properties in UV pro-
energy conservation have a number of properties in cessing included in governing conservation equa-
common, including density, specific heat capacity, tions vary with temperature and material composition,
thermal conductivity, and viscosity. while the absorptivity is only dependent on material
For high-pressure processing the main variables composition. Chapters 14 and 15 provide some of
acting are pressure, temperature, and initial compo- the values of absorptivity (absorption coefficient)
sition of the chamber components, which affect for various liquid foods.
the compression heating properties such as density, A number of publications have tabulated data and
specific heat capacity, and thermal expansion coef- empirical equations describing the variation of ther-
ficient, as well as viscosity and thermal conductivity. mophysical properties, including the governing con-
This has been covered in Chapters 4 and 5, which servation equations at atmospheric conditions as a
summarize the changes in these properties at high function of temperature and composition of specific
pressure under low and high temperatures. food materials and polymers (product and compo-
During microwave processing, properties included nent related). Properties related to other physics phe-
in the governing conservation equations and the nomena have not been determined to the same extent.
dielectric properties are only dependent on the tem- As pointed out in this chapter, the main hurdle in
perature, the material, and the selected frequency; developing accurate Multiphysics models is the lack
they are not affected by the electromagnetic field of thermophysical property values. In addition, little
strength. Chapters 6 and 7 present models wherein information exists in the literature on thermophysi-
these properties vary as a function of temperature and cal properties at high-pressure conditions. It would
initial product composition. Similarly, for ohmic be highly desirable that sufficient property data be
heating and pulsed electric fields processing, the determined in the form of a database, one that at least
governing conservation equation properties and the shows the values for carbohydrates, proteins, lipids,
electrical conductivity are only affected by the tem- fiber, and ions present in different foods, as well as
perature of a specific material without changing with values for selected food materials, which may also
the variable electric field. Chapters 9 and 11 provide represent a group of materials. Once this information
more details on the variation of these thermophysical becomes available, the rule of mixtures can be
properties with temperature and composition. applied to most of the thermophysical properties
Ultrasound processing models (Chapter 12) reviewed in this chapter, and overall values for par-
require governing conservation equation properties, ticular foods or food blends can be determined.
as well as the speed of sound and the sound attenu- However, it should be pointed out that the speed of
ation coefficient, which are both dependent on tem- sound and sound attenuation coefficients in disper-
perature and pressure, and frequency in dispersive sions do not necessarily obey simple rules of mix-
media. In the case of mixtures they also depend tures, given that they may also depend on particle
on composition. For dispersions (suspensions and size and distribution of the dispersed phase.
To satisfy the need for these properties, empirical V Volume m3

equations as functions of temperature and pressure x Thickness m
are essential for obtaining reliable Multiphysics y Direction of shear —
models. Furthermore, these models will need to y Distance m
account for the composition and structure of the food
materials. A way of overcoming this challenge is by
providing an open platform for the collection of Greek Symbols
property data, which will enable the development of
accurate models for various food categories. A web αP Thermal expansion coefficient K−1
site approach similar to the Wikipedia principle could α Thermal diffusivity m2/s
assist in developing an extensive database accessible α Sound attenuation coefficient m−1
to the general public. Inter-laboratory validation would αo Sound attenuation coefficient due m−1
then be needed to assure data consistency and accuracy. to absorption alone
α exc Excess attenuation m−1
Notations α th Thermal scattering sound m−1
attenuation
Latin Symbols α vis Visco-inertial sound attenuation m−1
A Absorbance — β Compressibility Pa−1
A Area m2 ρ Density kg/m3
B Constant m−1/Hz2 ε Porosity —
CP Specific heat capacity J/kg/K ε Dielectric permittivity —
c Concentration mol/L ε Molar absorptivity L/mol/m
c Speed of sound m·s−1 ε' Real part of complex permittivity; —
d Path length m dielectric constant
f Frequency Hz ε" Imaginary part of complex —
F Force N permittivity, dielectric loss
H Enthalpy J/kg factor
I, I0 Light intensity W/m2 φ Volumetric fraction —
kC Compression heating Pa−1 γ = ( c p cv ) Ratio between constant —
coefficient pressure and constant volume
k Thermal conductivity W/m/K heat capacities
K Constant γ Shear rate s−1
L Length m η Viscosity Pa·s
M Mass and specific mass kg or kg/m3 ηp Apparent viscosity Pa·s
n Exponential constant λ Wavelength m
P, p Pressure Pa τ Shear stress N/m2
Q Heat/energy J σ Electrical conductivity S/m1
Q Heat/energy flow J/s
R Electrical resistivity m/S
R Gas constant = 8.3142 J/K/mol
Abbreviations
r Distance m
r Radius of dispersed particles m AC Alternating current
S Entropy J/K PDE Partial differential equation
T Temperature K CFD Computational fluid dynamics
t Time s SI International System of Units (Système
u, v Velocity m/s international d’unités)
DSC Differential scanning calorimetry ethylene glycol, propylene glycol, castor oil, silicon oil, and
NIST National Institute for Standards and some of their binary mixture. J Chemical Eng Data 55(9):
3017–3023.
Technology Harvey AH, Peskin AP, Sanford AK. 1996. NIST/ASTME—
UV Ultraviolet IAPSW Standard Reference Database 10, version 2.2.
Juliano P, Knoerzer K, Barbosa-Cánovas GV. 2008. High pres-
sure thermal processes: thermal and fluid dynamic modeling
Operators principles 35. In: R Simpson, ed., Engineering Aspects of
Thermal Processing, 91–158. Boca Raton, FL: CRC Press/
d Differential
Taylor & Francis.
∂ Partial differential Juliano P, Knoerzer K, Fryer P, Versteeg C. 2009. C. botulinum
inactivation kinetics implemented in a computational model
of a high pressure sterilization process. Biotechnol Prog 25:
References
163–175.
Ament WS. 1953. Sound propagation in gross mixtures. J Acoust Junquera E, Ruiz M, Lopez S, Aicart E. 2002. A technique and a
Soc Am 25:638–641. method for the continuous, simultaneous, and automatic mea-
Aparicio C, Otero L, Sanz PD, Guignon B. 2010. Specific volume surement of density and speed of sound in pure liquids and
and compressibility measurements of tomato paste at moder- solutions. Rev Sci Instrum 73:416–421.
ately high pressure as a function of temperature. J Food Eng. Kinsler LE, Frey AR, Coppens AB, Sanders JV. 2000.
DOI: 10.1016/j.jfoodeng.2010.10.021. Fundamentals of Acoustics. New York: John Wiley & Sons.
Ardia A, Knorr D, Heinz V. 2004. Adiabatic heat modelling for Knoerzer K, Regier M, Erle U, Pardey KK, Schubert H. 2004.
pressure build-up during high-pressure treatment in liquid- Development of a model food for microwave processing and
food processing. Food Bioproducts Proc 82:89–95. the prediction of its physical properties. J Microw Power
Barbosa-Cánovas GV, Rodríguez JJ. 2005. Thermodynamic Electromagn Energy 39:167–177.
aspects of high hydrostatic pressure food processing. In: GV Knoerzer K, Juliano P, Gladman S, Versteeg C, Fryer P. 2007.
Barbosa-Cánovas, MS Tapia, MP Cano, eds., Novel Processing A computational model for temperature and sterility distribu-
Technologies, 183–206. New York: CRC Press. tions in a pilot-scale high-pressure high-temperature process.
Barbosa-Cánovas GV, Ortega-Rivas E, Juliano P, Yan H. AIChE J 53:2996–3010.
2005. Food Powders. Physical Properties, Processing and Knoerzer K, Regier M, Schubert H. 2008. A computational model
Functionality. New York: Springer-Verlag. for calculating temperature distributions in microwave food
Barbosa-Cánovas GV, Juliano P, Peleg M. 2007. Engineering applications. Innov Food Sci Emerg Technol 9:374–384.
properties of foods. In: GV Barbosa-Cánovas, ed., Food Knoerzer K, Buckow R, Sanguansri P, Versteeg C. 2010a.
Engineering. Encyclopedia of Life Support Systems, 25–44. Adiabatic compression heating coefficients for high-pressure
Oxford: EOLSS Publishers Co Ltd. processing of water, propylene-glycol and mixtures—A combined
Barthel R, Nolle AW. 1952. A precise recording ultrasonic inter- experimental and numerical approach. J Food Eng 96:229–238.
ferometer and its application to dispersion tests in liquids. J Knoerzer K, Buckow R, Versteeg C. 2010b. Adiabatic compres-
Acoust Soc Am 24:8–15. sion heating coefficients for high pressure processing—A
Bourne MC. 2002. Texture, viscosity, and food. In: Bourne MC, study of some insulating polymer materials. J Food Eng
ed., Food Texture and Viscosity, 1–32. London: Academic 98:110–119.
Press. Kokini JL. 1992. Measurement and simulation of shear and shear
Buckow R, Schroeder S, Berres P, Baumann P, Knoerzer K. 2010. free (extensional) flows in food rheology.
Simulation and Evaluation of Pilot-Scale Pulsed Electric Field Kubota K, Kurisu S, Suzuki K, Matsumoto T, Hosaka H. 1981.
(PEF) Processing. J Food Eng 101(1):67–77. Study on the flow equations of sugar, salt and sugar-salt skim
Chen XD. 2006. Modeling thermal processing using computa- milk solutions. J Jpn Soc Food Sci Technol-Nippon Shokuhin
tional fluid dynamics (CFD). In: DW Sun, ed., Thermal Food Kagaku Kogaku Kaishi 28:186–193.
Processing, 133–151. Boca Raton: Taylor & Francis. Lamberti N, Ardia L, Albanese D, Di Matteo M. 2009. An ultra-
Dukhin AS, Goetz PJ, Travers B. 2005. Use of ultrasound for sound technique for monitoring the alcoholic wine fermenta-
characterizing dairy products. J Dairy Sci 88:1320–1334. tion. Ultrasonics 49:94–97.
Eggers F, Kaatze U. 1996. Broad-band ultrasonic measurement Leighton TG. 1997. The Acoustic Bubble. New York: Academic
techniques for liquids. Meas Sci Technol 7:1–19. Press.
Guignon B, Aparicio C, Sanz PD. 2009. Volumetric properties of Lighthill J. 1978. Waves in Fluids. Cambridge: Cambridge
sunflower and olive oils at temperatures between 15 and 55C University Press.
under pressures up to 350MPa. High Press Res 29:38–45. Lozano JE. 2007. Thermal properties of foods. In: GV Barbosa-
Guignon B, Aparicio C, Sanz PD. 2010. Volumetric properties of Cánovas, ed., Food Engineering. Encyclopedia of Life Support
pressure-transmitting fluids up to 350 MPa: Water, ethanol, Systems, 45–64. Oxford: EOLSS Publishers Co Ltd.
Mathys A, Knorr D. 2009. The properties of water in the pressure- Sahin S, Sumnu SG. 2006. Physical Properties of Foods. New
temperature landscape. Food Biophys 4:77–82. York: Springer.
Mattea M, Urbicain MJ, Rotstein E. 1989. Effective thermal- Salvadori VO, DeMichelis A, Mascheroni RH. 1997. Prediction
conductivity of cellular tissues during drying—Prediction by of freezing times for regular multi-dimensional foods using
a computer-assisted model. J Food Sci 54:194. simple formulae. Food Sci Technol Lebensmitt Wiss Technol
McCartney ML, Drouin G. 1977. Moderate-accuracy interferom- 30:30–35.
eter for speed of sound measurement in liquids. Rev Sci Schubert H, Regier M. 2005. Introducing microwave processing
Instrum 48:214–218. of food: principles and technologies. In: H Schubert, M Regier,
McClements DJ, Povey MJW. 1989. Scattering of ultrasound by eds., The Microwave Processing of Foods. Cambridge: Woodhead
emulsions. J Phys D Appl Phys 22:38–47. Publishing Limited.
McClements DJ, Povey MJW, Jury M, Betsanis E. 1990. Servant G, Laborde JL, Hita A, Caltagirone JP, Gerard A. 2001.
Ultrasonic characterization of a food emulsion. Ultrasonics Spatio-temporal dynamics of cavitation bubble clouds in a low
28:266–272. frequency reactor: Comparison between theoretical and experi-
McSkimin HJ. 1957. Ultrasonic pulse technique for measuring mental results. Ultrason Sonochem 8:163–174.
acoustic losses and velocities of propagation in liquids as a Shao YW, Zhu SM, Ramaswamy H, Marcotte M. 2007.
function of temperature and hydrostatic pressure. J Acoust Soc Compression heating and temperature control for high pressure
Am 29:1185–1192. destruction of bacterial spores: An experimental method for
Min S, Sastry SK, Balasubramaniam VM. 2010. Compressibility kinetics evaluation. Food Bioprocess Technol. doi: 10.1007/
and density of select liquid and solid foods under pressures up s11947-008-0057-y
to 700 MPa. J Food Eng 96:568–574. Shoitov YS, Otpushchennikov NF. 1968. Pressure dependence
Mohsenin NN. 1984. Electromagnetic Radiation Properties of of the speed of sound in liquid. Izvestiya VUZ Fisika 11:
Food and Agricultural Products. New York: Gordon and 137–139.
Breach Science. Steffe JF. 1992. Rheological Methods in Food Process
Otero L, Molina-Garcia A, Sanz P. 2000. Thermal effect in foods Engineering. East Lansing, MI: Freeman Press.
during quasi-adiabatic pressure treatments. Innov Food Sci Tabilo-Munizaga G, Barbosa-Cánovas GV. 2005. Rheology for
Emerg Technol 1:119–126. the food industry. J Food Eng 67:147–156.
Otero L, Ousegui A, Guignon B, Le Bail A, Sanz PD. 2006. Trujillo FJ, Knoerzer K. 2009. CFD modelling of the acoustic
Evaluation of the thermophysical properties of tylose gel under streaming induced by an ultrasonic horn reactor. In: Seventh
pressure in the phase change domain. Food Hydrocolloids International Conference on CFD in the Minerals and Process
20:449–460. Industries. Melbourne, Australia.
Patazca E, Koutchma T, Balasubramaniam VM. 2007. Quasi- Vitali AA, Rao MA. 1984. Flow properties of low-pulp concen-
adiabatic temperature increase during high pressure processing trated orange juice—Serum viscosity and effect of pulp
of selected foods. J Food Eng 80:199–205. content. J Food Sci 49:876–881.
Povey MJW, Mcclements DJ. 1998. Ultrasonics in food engineer- Welti-Chanes J, Vergara-Balderas F, Bermudez-Aguirre D. 2005.
ing. Part I: Introduction and experimental methods. J Food Eng Transport phenomena in food engineering: Basic concepts and
8:217–245. advances. J Food Eng 67:113–128.
Rahman S. 1995. Food Properties Handbook. Boca Raton, FL: Zhang H. 2007. Electrical properties of foods. In: GV Barbosa-
CRC Press. Cánovas, ed., Encyclopedia of Life Support Systems (EOLSS).
Rao MA. 2007. Rheology of Fluid and Semisolid Foods: Developed under the Auspices of the UNESCO, 115–125.
Principles and Applications. New York: Springer. Oxford, UK: Eolss Publishers.
Rasanayagam V, Balasubramaniam VM, Ting E, Sizer CE, Bush Zhu S, Ramaswamy HS, Marcotte M, Chen C, Shao Y, Le Bail
C, Anderson C. 2003. Compression heating of selected fatty A. 2007. Evaluation of thermal properties of food materials
food materials during high-pressure processing. J Food Sci at high pressures using a dual-needle line-heat-source method.
68:254–259. J Food Sci 72:E49–E56.
Chapter 3
Neural Networks: Their Role in High-Pressure Processing
José S. Torrecilla and Pedro D. Sanz
The main objective of a mathematical model is to research tasks were delayed for nearly a decade. In
reproduce the corresponding process performance 1969, Minsky and Papert published a book titled
and/or estimate important variable values. Because Perceptrons, which was partially responsible for the
of this, the modelization of processes is useful reinitiation of NN research activity (Minsky and
mainly as a previous stage in the design of equip- Papert 1969). Shortly afterwards, the backpropaga-
ment. With this aim, although detailed knowledge tion (BP) algorithm, one of the most important
of the process is not always required, the input and components of NN algorithms, was presented by
output variables of the system to be studied should Werbos and developed by Rumelhart and his col-
be known. Here, two such methods usually used to leagues among other research groups (Werbos 1974;
design this type of models are described, viz. models Rumelhart et al. 1986). In the 1980s, Holpfield’s
based on neural networks (NNs) and macroscopic publications stimulated NN investigations (Hopfield
models. Although successful applications will be 1982, 1984).
shown here, the presented models have to be adjusted In the last two decades, algorithms that handle
prior to applying them to new systems. and process information in a way similar to the
human brain such as NNs and fuzzy logic have been
developed (Chen 1996). In particular, the rapid
development of NN algorithms and their numerous
3.1. Brief History of NNs
applications in commercial and scientific sectors are
In the nineteenth century, Alexander Bain and mainly based on the notable growth in computer
William James established the foundation of modern technology. Now, NN technology is used in food
NNs, which later, in the middle of the twentieth science, chemical engineering, process control,
century, were developed by Warren McCulloch and medical diagnosis, forensic analysis, weather fore-
Walter Pitts (McCulloch and Pitts 1943). These algo- casting, financial applications, investment analysis,
rithms were further developed during the ensuing and others. In all these applications, depending on
two decades, but in the 1960s, as a result of the exag- their requirements, NNs can be applied to tasks
geration of successful results, bitter critique focused related to classification and/or prediction activities
mainly on the capability of NNs was published and (Huang et al. 2007).

39
ACTIVATION TRANSFER
INPUTS WEIGHTS OUTPUTS
FUNCTION FUNCTION
0.5 1
x I iWi f ( x)
1
x= –0.9 x f(x) = –0.289
i 1 e
0.7 –2
Figure 3.1. Schema and an example of the calculation procedure of an artificial neuron (x, I, i, and f(x) are the activation
function, input value, layer, and sigmoid function as transfer function, respectively).
3.2. Basis of NNs values must be optimized using real data from the
system to be modeled (Huang et al. 2007). The opti-
Although biological neurons are much more com-
mization of the weights is commonly called the
plicated than artificial neurons, the design and
learning process, which is the most important
development of artificial neurons have been inspired
process in the design of NN.
by biological cells. A biological neuron consists of
Depending on the NN type, their weights can be
dendrites, soma, axon, and synapses. The signal is
optimized by supervised or unsupervised methods.
received by dendrites, processed in the soma or cell
The most common learning process used is the
body and the output is emitted by the axon and
supervised method, which requires a database com-
transmitted to other neurons by synapses. An artifi-
prising input and target output values. For instance,
cial neuron receives signals, which are processed in
to optimize the weights of a feedforward network, a
the neurons by mathematical algorithms (a combi-
supervised learning process such as a BP algorithm
nation of linear or nonlinear algorithms), and the
is normally used. The unsupervised learning process
output information is then sent to another neuron as
requires only input information. Kohonen’s self-
input or is used directly as a network result. A sche-
organization map is a network that follows an unsu-
matic of an artificial neuron and an example of a
pervised learning process (vide infra). Throughout
calculation procedure are shown in Figure 3.1.
the chapter, every statistical result has been calcu-
The connectivity of artificial neurons emulates
lated using estimated values worked out by the NN
the biological neurological system, which is in most
models and the real values from the external valida-
cases to design an NN; artificial neurons are joined
tion samples.
and depending on the connections between them,
different types of NNs can be defined, for example,
3.2.1. BP Algorithm
neurons may or may not receive information back
from the next layers, forming NNs called feedback The BP (also called the retro-propagation of error)
or feedforward networks, respectively, whereas the is a common method used to optimize the weights
recurrent networks, once the signal is received in a of supervised networks. These algorithms can be
neuron from a given layer, are communicated to used to optimize the weights of NNs with fully or
other neurons in the same layer before sending it partially interlayer connections (no intra-layer
onto the next or last layer. In most cases, the artifi- connection).
cial neurons are arranged in layers. Basically, the information is inputted into an NN
Each connection is controlled by parameters and is fed forward throughout hidden and output
called weights. As some input signals may be more layers to calculate a response. The calculation
important than others, to obtain the best result (the process in each neuron of the hidden and output
maximum accuracy possible, vide infra), the weight layers consists of activation and transfer functions.
Chapter 3 Neural Networks: Their Role in High-Pressure Processing 41
The activation function, Equation 3.1, means that calculated. When every data set of the learning
the input data to each neuron are multiplied by sample has been used, a cycle is finished and another
weight; the result, xk, is fed into a transfer function. begins with the first data set of the aforementioned
The sigmoid or hyperbolical tangent function is the database (Torrecilla et al. 2009a). A more detailed
most commonly used transfer function, Equations 3.2 explanation can be found elsewhere (Fine 1999;
and 3.3, respectively. The calculated value, yk, is the Arbib 2003).
output of the considered neuron. This calculation
procedure is shown in Figure 3.1. 3.2.2. Unsupervised Learning Process
xk = ∑w
j =1
jk ⋅ yj (3.1) The networks optimized by the unsupervised learn-
ing process are able to learn from the input signals
to ascertain their regularity without knowing in
⎛ 1 ⎞
yk = f ( x k ) = ⎜ advance what the desired outputs are. Although
⎝ 1 + e − xk ⎟⎠
(3.2)
unsupervised learning may provide more effective
⎛ 1 − e −2 xk ⎞ solutions than supervised learning (Torrecilla et al.
yk = f ( x k ) = ⎜ (3.3)
⎝ 1 + e −2 xk ⎟⎠
2009b), unsupervised learning processes are still in
an early trial stage, which can only be used for some
where i, j, and k subscripts are the input, hidden, and simple applications (Huang et al. 2007). Due to the
output layers, respectively. Given that the optimum fact that one of the most representative, interesting,
weights are calculated by differential calculations, and influential unsupervised networks is Kohonen’s
all algorithms that compose the BP algorithm (acti- self-organization map (Kohonen 1987), the learning
vation, transfer function) must be differentiable process of this type of NN is exposed here.
(vide infra). Then, this output is compared with the Self-organizing maps (SOMs), Kohonen’s self-
real or known value (rk) and the mean square error organization map, or Kohonen’s NN can learn to
value (MSE) is calculated, Equation 3.4. detect irregularities and correlations in their input
N data and adapt their future responses to that input
∑ (r − y )
1
MSE = k k
2
(3.4) accordingly, that is, they are able to recognize groups
N k with similar characteristics (Kohonen 1987; Demuth
et al. 2005). The architecture of SOM models is
where N is the number of neurons in the output
shown in Figure 3.3. This is composed of two layers,
layer. Once the MSE has been calculated, the opti-
viz. input and competitive layers. Every circle and
mization of the weight begins. The weights of the
arrow represents a neuron and weight, respectively,
hidden and output layer connections are first opti-
that is, there are as many weights as arrows and the
mized. An example of the forward calculation pro-
number of neurons is equal to the product of the
cedure and the equations required to optimize every
width and length of the competitive layer.
weight are shown in Figure 3.2. As can be seen, two
In order to represent the whole input database, the
different groups of equations are used to optimize
neurons, defined by weight vectors, look for the best
the weights from hidden-output layers and input-
place in the competitive layer during the learning
hidden layers.
process (Figure 3.3). The weights of the SOM model
The determination of the level of generalization
optimization can be summarized in five stages:
of the NN is performed once the weights have been
optimized. For that purpose, the NN is tested by 1. Assign random values to the weights.
new databases commonly called verification. In 2. A data set from the learning sample is presented
this way a testing or an internal validation sample to the SOM.
and an external validation sample is obtained. That 3. The neuron with the least Euclidean distance
is, the optimized NN calculates the output values between its weights and data set (D) is selected
using a new database and the prediction error is then (Eq. 3.5). This is called the winning neuron.
Exploring the Variety of Random
Documents with Different Content
Händen, und sie erwachte wie aus einem schönen Traum, als es
draußen lebendig wurde und die Ausgeflogenen heimkehrten.
»Aber Frida! Da sitzt Du noch! Es ist ja Mitternacht vorbei, weißt
Du das nicht?«
Sie bekam freundliche Schelte, nur Gisela sah entrüstet aus.
»Wäre sie mit uns gewesen, Tante, so wäre sie schon längst müde
geworden.«
Sie war in einen liebenswürdigen, lustigen und eleganten Kreis
geraten, der kleine Märchenfritz aus dem Wind- und Wolkenturm von
Hohen-Leucken – aber sie hätte wohl noch in einen besseren
geraten können. Es kamen hin und wieder Leute in den ihren hinein,
die sahen sich nach ihr um und konnten sie danach eine ganze
Zeitlang nicht vergessen. Und das nicht darum, weil sie sie als die
Einfalt vom Lande amüsierte.
Im ganzen spielte sie ja hier die unvorteilhafteste Rolle, die solch
ein unbehauener junger Menschenblock zwischen all den
gehobelten, glatten und strahlenden Figuren und Figürchen spielt. Es
ist eine gar ehrliche Tragik um diese Rolle.
Wo sollte sie nun hin? Wo paßte sie nun hin? Immer nur in ihr
altes ödes Heimatshaus, das hier so sichtlich verachtet wurde? Was
jeder hier konnte: glatt in dieser Gesellschaft aufgehen, das konnte
nur sie nicht? Woran konnte es nur liegen als an ihrer
hoffnungslosen Dummheit, daß sie sich hier stets zur Freude
gewaltsam zwingen mußte und erst wieder los und ledig fühlte,
wenn sie allein mit sich war wie an den schönen Kopfwehabenden?
Der kleine Märchenfritz konnte es nicht wissen, daß er nur falsch
gelaufen war, daß es für ihn noch schöne und lustige Wege gab,
auch außerhalb des Nebelrings von Hohen-Leucken. Er kam im
nächsten Frühjahr, ziemlich zerbrochen in seinem Selbstgefühl und
zerfallen mit sich und der ganzen Welt ins Vaterhaus zurück.
* * *
Gisela war mitgekommen. Deren glänzendes Leben hatte jetzt
vielleicht für immer ein Ende. Das war eine harte Nuß für das
verwöhnte Kind der Welt.
Drei Menschen sitzen im kalten, mürrischen Hause und warten,
daß es Sommer wird über dem Moor. Für Fritzchen freilich ist der
Sommer im Grunde heute schon da, trotz Schnee, Hagel, Aprilsturm
und Nässe. Aber sie will es nicht – nichts will sie wissen, hören,
fühlen. Sie will hier nicht glücklich sein, weil es doch nur ein neues
Zeugnis ihrer Dummheit ist. Aber was hilft's, daß sie nicht will? Sie
sieht das Moor und sieht die Wolken, sie riecht Papas Zigarren und
sieht sein rundes, rotes, brummiges Gesicht, sie rennt durchs Haus
und über die zugigen Treppen, es zieht und pfeift aus allen Ecken,
Jakob klappert mit dem Mittagsgeschirr, ihr alter Tisch steht noch am
Fenster – alles ist, wie es war – ach Fritz, Fritz, was tut man mit all
der Freude, und wenn man auch noch so klug sein möchte!
»Papa, was hast Du den ganzen Winter angefangen?«
Herr v. Dörfflin sieht nicht wohler aus seit dem Herbst, als
Fritzchen abreiste, auch durchaus nicht lustiger. Ja, was hat er
angefangen?
»Nichts, Fritz.«
Nichts. Der Fritz wird langsam ernst und seine Blicke werden
verwirrt. Was weiß ein siebzehnjähriges Geblüt von dem Nichts, in
das ein armseliges, verloddertes Leben versinkt?
Fräulein Miller war nicht mehr da, ihr Amt in diesem Hause war
beschlossen. Sie war nie eine liebenswürdige, weitherzige Gefährtin
gewesen, aber nun fehlte sie Fritzchen doch. Sie sollte ja nun
vollständig erwachsen sein. Ach, dieser Wirrkopf hatte wohl noch
manches Jahr vor sich, ehe man ihn für erwachsen nehmen konnte.
Also sprach auch Fräulein Gisela. Sie hatte hier keine Freude an
dem klappernden Jakob, an Zigarrenrauch und Wolken. Sie nahm
Anstoß an allem, besonders auch an Fritzchen. Es verging kein Tag,
an dem sie nicht um die versunkene Herrlichkeit klagte.
Wie fein war sie geworden! Ja, sie hatte schon Grund, hier
unzufrieden zu sein. Ihre kühlen, schlanken Hände waren so weiß,
ihr blondes schlichtes Haar hatte durch sorgsame Pflege einen
sanften Glanz erhalten, auch verstand sie, sich prächtig zu frisieren.
Alle ihre Kleider hatten einen eleganten Sitz, ihre Bewegungen, ihre
Sprechweise waren klar, vornehm und ruhig. Was war dagegen der
Struwwelkopf aus der Turmstube?
Fritzchen bewunderte dies feine, sichere Wesen. Ach, wer jemals
so werden könnte! Aber das zu wünschen, war wohl hoffnungslos.
Der schlimme Winter saß noch wie brennendes Gift im Blut. Aber
nun kam der Sommer über das Moor.
Wißt Ihr denn, wie der Sumpf blühen kann, Ihr Stadtmenschen,
ihr Lampenmenschen! Wie es sich da liegt, da hinten in der Lichtung
hinter dem alten Graben, wo man sich im Gras verstecken kann, so
hoch steht es. Kennt Ihr das Zirpen und Schwirren und tausendfache
Leben um einen her, und die Sonnenstrahlen flirrend durch die
Zweige?
»Gisa, willst Du mit an den alten Graben?«
»Was willst Du da?«
»Im Gras liegen. Seerosen bring' ich auch mit.«
Die Frage war recht überflüssig. Gisa – Gisa sollte in den
durchlöcherten Kahn steigen, der bei jeder Fahrt rapide Wasser zog,
dann landen – Fritzchen nannte das nämlich landen! – an einer
sumpfigen Stelle, wo man nur von einer Baumwurzel zur anderen
springend, schließlich auf eine feste Grasfläche gelangen konnte –
und das alles, um schließlich ein paar Stunden im Gras zu liegen mit
krabbelnden Würmern und Ameisen im traulichen Bunde. »Danke,
liebe Frida. Fahre nur allein.«
Ja, Gisa, Prinzessin Unmut, wie soll denn das werden? Das sind
doch die höchsten Freuden, die Hohen-Leucken bieten kann!
Fritzchen grübelte angestrengt. Sie ehrte Giselas Unmut und fühlte
sich brennend verantwortlich, ihn zu zerstreuen.
»Gisa, soll ich Wilhelm sagen, daß er anspannt? Ich kutschiere
Dich über die Felder.«
»Bei diesem ewigen Wind? Und meine Haare? Und immer nur
über die Felder? Nein, Fritzchen, Du meinst es gut, aber das ist
wirklich kein Vergnügen.«
»O, jetzt weiß ich etwas! Willst Du bei mir reiten lernen?«
Bei Fritzchen reiten lernen. Eine zweifelhafte Gunst. »Bei wem
hast Du es denn gelernt?«
»Bei mir selbst, natürlich.«
»So? Und welches Pferd hast Du dazu?«
»Ach, Möt. Eigentlich heißt es Erdmuthe. Papa hat sie mal als
zurückgesetztes Remontepferd gekauft, aber sie geht nicht an der
Deichsel. Wilhelm sagt, es ist nichts mit ihr zu machen, sie ist
verdammelt. Da habe ich voriges Jahr mit ihr losgelegt. Aber fein!
Über den Graben hinter dem Böllinger Kreuzweg setzt sie wie ein
Pfeil. O, wenn Du mal mitkönntest. Du nimmst vorläufig den
Schecken vom zweiten Gespann. Der geht wie ein Lamm und hat
keine Mucken.«
»Ja, wenn ich einen richtigen Reitlehrer hier hätte! Aber auch
dann! Es muß doch schrecklich stuckern! Nein, laß mich nur. Das ist
alles nichts für mich. Aber da nun die Leute wissen, daß ich wieder
hier bin, muß sich doch am Ende wohl etwas Verkehr hier anfinden
auf dem alten Räubernest.«
Fritzchen schlich sich zum Papa. »Papa, Gisa langweilt sich hier so.
Kannst Du es nicht machen, daß manchmal wieder Besuch
herkommt?«
»Ja, wie soll ich das machen?«
»Wenn Du es nur willst, kannst Du es schon machen.«
Es war ein sonnenleuchtender Junitag. Vor dem Fenster im Hof
blühten die alten Linden und ihr Duft strömte in die beiden offenen
Fenster herein. Herr v. Dörfflin war in Joppe und Reitstiefeln, er
wollte auf die Entenjagd gehen. Was ist es für ein anderes Ding um
solch ein Landjunkergesicht zur Sommers- als zur Winterszeit! Heute
sieht es frisch, gespannt, gebräunt, unternehmend aus und hockt
nicht in Dumpfheit und im Gefühl des Nichts.
»Ja, Fritz, das denkst Du Dir so.«
Am Abend kam er zurück, und beim Abendessen warf er hin, als
mache er eine Bemerkung über das Wetter: »Morgen kommt Hans
Henning v. Zülchow. Die Tannenwalder wollen nächste Woche auch
einmal kommen.«
Die Tannenwalder waren im Grunde ziemlich langweilige und
herkömmliche Leute. Aber es war ein junger Sohn dabei, der Jura
studiert hatte und sich jetzt in das väterliche Gut einarbeitete, und
eine Tochter, die gleich Gisela mehrere Jahre in Berlin gewesen war.
Das Ding legte sich also recht vielversprechend an.
Fritzchen blieb der Mund offen stehen. »Wie hast Du das so
schnell gemacht, Papa?«
Diese Fragerei paßte ihm nicht. »Ich hab' gar nichts gemacht!«
schnauzte er sie an. »Wir haben uns getroffen, wo der alte Graben
in den Tannenwalder See geht. Der Zülchow war mit dem alten und
dem jungen Euler da auf Entenjagd.«
So weit ist er wegen der Enten gerudert? dachte Fritzchen, aber
sie hütete ihre Zunge. Nach dem Abendessen ging sie in die Küche,
die Enten zu sehen, die der Herr geschossen hatte.
»Er hat gar keine abgegeben, gnädiges Fräulein«, sagte die
Mamsell.
* * *
Damit fing der Verkehr im Herrenhause von Hohen-Leucken
wieder an.
Es war jetzt doch alles anders wie ehedem. Alte, unliebsame
Geschichten waren vergessen, Herr v. Dörfflin erschien als völlig
unschädlich, und zwei junge, aufblühende Töchter waren im Hause.
Giselas Ruf als Weltdame machte Karriere, sie hatte die Art,
gleichzeitig zu imponieren und zu gefallen. Trotz ihrer Sicherheit und
Gewandtheit war sie auch für die plumpesten Junker handlich,
verstand auf die trivialsten Gegenstände mit entzückender
Leichtigkeit einzugehen und ihnen dadurch den Glanz von etwas
ganz Besonderem zu verleihen.
Es blieb nicht bei Hans Henning und den Eulers aus Tannenwalde.
Es kamen die Bärs, die Leisewitzens, die Winkels dazu, ja eines
Tages hatten Herr und Frau August Schultze mit dem Sohn und
Erben Leopold, die drüben das Böllinger Rittergut dem verschuldeten
Baron Laue abgenommen hatten, Besuch gemacht und waren nicht
wieder los zu werden. Herrn v. Dörfflins Adelsstolz entsetzte sich, er
war geradezu schmählich ungezogen zu diesem Besuch. Aber es war,
als ob er mit aller Wucht seines Knüppels auf eine leere Haut
schlüge, statt auf den Esel, so unschuldig blickte Herr Schultze drein.
Nachher – lange Auseinandersetzungen mit Gisela. Die hatte sich mit
Herrn Leopold sehr nett unterhalten, fühlte sich von seinen großen
Reisen und seinem flotten Weltleben angeheimelt und wünschte
durchaus, diesen Verkehr festzuhalten und: »lächerlich veraltete
Vorurteile« beseitigt zu sehen.
Jawohl, es kam denn auch zu Tage, daß dieser Prozeß beseitigter
Vorurteile und demnach einer Aufnahme Herrn Schultzes in den
Verkehrskreis bei den Winkels, den Leisewitzens und verschiedenen
anderen bereits längst in aller Stille vor sich gegangen sei, und Herr
v. Dörfflin hatte jetzt nicht mehr Mark und Ausdauer genug, um eine
so völlig isolierte Wut- und Abwehrstellung festzuhalten. Herr August
Schultze mit Familie gehörte danach also auch zu den Besuchern von
Hohen-Leucken.
Es kam noch bunter. Die beiden jungen Töchter wurden
eingeladen, und Gisela fand, obwohl es ihr selber
Unbequemlichkeiten machte, daß eine Gesellschaftsdame hier jetzt
unumgänglich nötig sei. Herr v. Dörfflin sagte: »Verdammter Unsinn,
da wird nichts draus!« Fritzchen machte ganz entsetzte Augen und
rebellierte dagegen. Aber Gisela war die einzige, die etwas von
solchen Dingen verstand, die Dame wurde verschrieben, und im
nächsten Winter war sie da. Es war die Witwe eines Offiziers, von
Adel und außerordentlich mit den Formen der feinen Welt vertraut.
Sie hieß Frau v. Pohle, war energisch und trotz aller Weltförmigkeit
voll tiefer, ruhiger Güte. Ein stürmisches Leben hatte sie hart
geschüttelt, so daß sie nicht mit den Ansprüchen eines verwöhnten
Herzens nach Hohen-Leucken kam. Das nüchterne, häßliche Haus,
der verbummelte Mensch, der hier Hausherr war, die beiden
verschieden gearteten und verschieden geleiteten Töchter, die
stumme, kahle Einsamkeit der Gegend, alles sprach ihr stark zum
Herzen und bewog sie, hier ihre beste Kraft und Liebe, ihren feinsten
Takt einzusetzen, um auf diesem verwilderten Felde doch noch eine
gute Saat zu ziehen.
Fritzchen begriff es schlecht, was für sie da kam. Sie hatte bisher
auch nur dürftige Erfahrungen mit den Gestalten ihrer Umgebung
gemacht. Es war ein zur Not mit ihnen Fertigwerden gewesen, sonst
nichts. Wo war die Hand, die sie behütet hatte, als sie ihren
Träumen bis in die Wolken nachlief, oder ihnen auf einem
unerzogenen Pferde über Gräben und Brachen nachjagte – die ihr
gegeben hätte, als sie hungrig und durstig war, die ihr den wirren
Kopf mit seinem tollen Phantastenkram gestreichelt hätte, die sie
geführt hätte, als die Wege sich verwirrten?
Immer sich selbst war dieser junge Vogel überlassen worden. Nun
duckt er sich, nun huscht er davon, als eine feine Hand ihn fangen
möchte. Er haßt die Käfige, die er vom vorigen Winter her kennt.
Sechstes Kapitel.
Gregor v. Zülchow trat aus Fritzchens Märchenbüchern heraus und

stand in Fleisch und Blut vor ihr da.
In der Stunde, da dies geschah, da sie ihn, der ihr schon fast zu
einer Sagengestalt verschollen war, in der Blüte seiner jungen
Herrlichkeit wiedersah, da fielen alle ihre selbsterdichteten Märchen
und Träume wie Schatten hin, wie Nebel, wenn der Morgen der
leuchtenden Wirklichkeit kommt. Und von dieser Stunde an bis zu
den Jahren, die ganz, ganz anders aussahen, dichtete sie kein
Geschichtchen mehr.
Es war in Rummelshof zur Sommerszeit. Seit Frau v. Pohle im
Hause war, hatte sich in der Auffassung der Gegend viel verändert.
Selbst die Freifrau v. Zülchow, diese exklusiveste und empfindlichste
aller Landedeldamen, fand es jetzt ganz natürlich, mit den Dörfflins
zu verkehren, und sie selber betrat dieses Haus, das sie einst so tief
mißachtet hatte, mit ihrem Sohn Hans Henning, dem jungen Offizier.
Hans Henning war gerade wie Fritzchen seiner ersten Kinderliebe
treu geblieben, nur daß es bei ihm etwas weniger phantastisch
zuging, aber nicht sehr viel. Er war ein offener, liebenswürdiger und
starker Junge, mit der Einseitigkeit und dem Idealismus eines kräftig
empfindenden, wenig philosophisch angelegten Gemüts. Er liebte
mit großer Hingabe und stürmischer, vollkommen blinder
Parteinahme alles, was ihm Natur und Verhältnisse nahegebracht
hatten: seinen Beruf, seine Mutter, den Rummelshof und über alles
seinen klugen, stolzen Bruder. Aber seine Liebe war von der Art, daß
sie denen, die ihre Bequemlichkeit lieben, manchmal lästig fallen
konnte. Er war zu gern in ihrem Dienst ein Raufbold. Es war
vorgekommen, daß er einem Jungen, der das theologische Studium
seines Bruders verspottet hatte, das Nasenbein eingeschlagen hatte.
Seine Mutter hatte eine Menge Unannehmlichkeiten, Ängste und
Kosten davon. Seine Lehrer klagten häufig über ihn, er war faul, wild
und händelsüchtig.
Die Baronin Zülchow liebte solche Berührungen mit der Außenwelt
nicht. Sie bekam dadurch eine Gereiztheit gegen ihren tollen Hans.
Daß er im Grunde der weichherzigste Junge und der liebevollste
Sohn war, konnte sie nicht recht versöhnen. Ihr wäre es angenehmer
gewesen, er hätte ihr seine Liebe in Gehorsam und Wohlverhalten
bewiesen, statt sie nur wie eine Sonne über seine Untaten leuchten
zu lassen.
Für Fritzchen war er immer der beste Herzenskamerad und
Mitwisser all ihrer Erlebnisse, Betrachtungen und Phantastereien.
Nur um ihr schönstes Märchenland, in dem Gregor regierte, hing sie
einen dichten Schleier.
Da kam der Sommertag auf dem Rummelshof.
Es war eine große Gesellschaft dort. Zwischen den Bäumen des
Gartens waren Drähte gezogen, denn am Abend sollte Feuerwerk
sein. Die Dienerschaft war verstärkt, aus dem tiefsten Dunkel des
Kellers kamen die ältesten Weinflaschen ans Tageslicht. Die Baronin
trug violette Seide, sie sah wie eine Königsmutter aus.
Alles war zu Ehren ihres ältesten Sohnes Gregor, der vor einigen
Monaten in das Predigtamt in einer kleinen Residenz eingeführt war
und heute zu seinem ersten Besuch seit Jahresfrist kam.
Wie es so kommt, der erste Moment, als Fritzchen ihn wiedersah,
war ein Erstaunen: Ach – so sieht er aus? Das ist er?
Es ist wie eine leise Enttäuschung, oder wie eine Erlösung, ein
Abfinden zwischen »war« und »ist«, zwischen Traum und
Wirklichkeit. Es ist kein Wunder, wenn das Alte, das so strahlend und
herrschend war, sich wehrt, zuckt, das Neue schlägt. Aber das Neue
ist doch mächtiger. Es hat Klang und Farbe und Raum. Du stutzest,
du läßt die alten Fäden fahren und schaust nur und schaust – und
dein Herz, nach einem kurzen leeren Stillstand, setzt jählings mit
einem Wirbel wieder ein. –
»Gnädiges Fräulein –«
Das war der junge Pfarrer, Gregor v. Zülchow, der im
Gesellschaftsrock in einem der Rummelshöfer Zimmer ihr seine
Verbeugung machte. »Ich habe schon die Ehre der Bekanntschaft
aus früheren Jahren.«
»Ja gewiß –«, stammelte Fritzchen.
Sie schämte sich plötzlich zum Umfallen. Von diesem feinen,
kühlen, wildfremden Herrn hatte sie Märchen ohne Ende gedichtet
und noch dazu aufgeschrieben? O nein, o nein, das war ja ein ganz
anderer!
Sie ging umher wie wirr unter all den Menschen. Vielleicht hatte
sie in ihrem Leben noch nicht so reizend ausgesehen. Sie trug ein
weißes Kleid mit Matrosenbluse, als sollte immer noch das
Jungenshafte an ihr betont werden, aber ihr rotbraunes Knabenhaar
hatte man jetzt wachsen lassen und es war in einen Knoten
geschlungen. Nur zwei, drei rote Röschen aus dem Garten zu Hause
steckten ihr im Haar und vor der Brust. Im ganzen sah sie zwischen
all den sorgsam angezogenen Damen aus, als habe man sie eben in
der Wiese eingefangen und mit hergebracht.
Sie sah und hörte nichts von all den Leuten um sie her. Wenn man
sie ansprach, mußte sie sich erst besinnen, ehe sie eine Antwort
zustande brachte. Auch das stand ihr reizend, es gab ihr den Anflug
einer süßen Verträumtheit, der ihr ganzes Wesen dämpfte und
reizvoll verschleierte.
Die wundersame Umwandlung der Traumwelt in die Wirklichkeit
hatte sich vollzogen. Alles Vergangene war versunken und verhallt.
Er trug keinen Panzer und keinen Helm, sondern einen schwarzen
Theologenrock und einen hohen weißen Kragen mit schwarzem
Schlips. Er hatte ein kühnes, schmales, bartloses Gesicht von
niederländischem oder englischem Typus, und vor den düster
ernsten, durchdringenden Augen trug er eine goldene Brille.
Nachdem der erste Wirbel vorüber war, fühlte Fritzchen sich unter
all den Menschen so geborgen wie in einer lichten, goldnen Wolke.
Sie schwatzte, lachte, und wußte schon in nächster Minute nicht
mehr, was sie gesagt hatte. All ihr Tun war wie das Schwirren eines
fröhlichen kleinen Vogels, der eben fliegen lernt. Es lag ihr gar nichts
daran, immer in Gregors Nähe zu sein. Ja, sie spielte mit ihrem
Glück, indem sie mit Hans Henning oder sonst einem lustigen
Menschen viertelstundenlang in den Garten lief, Erdbeeren naschte
oder in dem romantischen Parkteich herumruderte. Das bloße
Gefühl, daß er ja da war, daß sie ihn in jedem Augenblicke, wenn sie
nur wollte, sehen konnte, das krönte alles mit überirdischem Glanz.
Herr Pfarrer Gregor war ja auch der Mittelpunkt des ganzen
Festes. Wie sehr ihm alles huldigte, und wie wichtig er, der blutjunge
Mensch, hier genommen wurde, das erschien Fritzchen als die
natürlichste Sache von der Welt.
Beim Abendessen wurden pompöse Toaste ausgebracht, alle auf
den jungen Pfarrer. Es war im Grunde eine ziemlich alberne
Lobhudelei, die man teils der Mutter, teils seiner allerdings stark
verheißungsvollen Karriere wegen in Szene setzte. Fritzchen hätte zu
jedem Toast jauchzen mögen, nur war ihr es immer noch nicht
genug. Der Umschmeichelte dagegen sah unter seiner tadellosen
Maske der Höflichkeit ein wenig gelangweilt und leicht angewidert
aus. Er hatte ein Lächeln um den Mund, das nicht jeder
vierundzwanzigjährige Junge hat.
Nach dem Essen wurde das Feuerwerk abgebrannt. Das war für
die junge Welt ein großes Entzücken. Aber Gregor gehörte nicht zu
der jungen Welt, er stand und saß bei den Alten auf der Terrasse
oder in den Zimmern. Hans Henning sollte das Abbrennen besorgen,
und Fritzchen, die für allen Firlefanz des Lebens sehr veranlagt war,
bewies sich als unentbehrlichste Hilfskraft. Was hatte sie auch vom
Zuschauen? Selber das Feuer wecken, die Spiel-, Sprüh- und
Lärmgeister des Feuers wecken, daß sie hoch in den stillen
Nachthimmel auffliegen, prasselnd niederstürzen, ihm, dem Herrn
des Tages zur Ehre!
Sie verbrannte sich ein paarmal die Fingerspitzen; was tat das?
Bebend, mit glühenden Wangen sah sie ihren Raketen, Leuchtkugeln
und farbigen Garben nach. »Heil Gregor!« rief ihr Herz. O, hätte sie
es nur laut rufen dürfen.
»Seht die beiden Feuergeister!« wurde gerufen. Damit meinte
man Hans Henning und Fritzchen. Sie sehen auch wohl geisterhaft
genug aus. Hans, der selige Narr, strahlte. »Hören Sie das, Fräulein
Fritz?«
Es stieg viel Torheit, Unsinn und Gaukelei mit diesen Leuchtkugeln
und Raketen auf. Man könnte sich wohl ebenso gut an eine
Feuergarbe hängen, um mit ihr in die Luft zu gehen, als sich von
seiner Liebe und ihren Verheißungen betören zu lassen. Aber die
Beiden wollten nicht hören, was ihr feuriges Spielzeug sie lehren
möchte.
Aus, vorbei. Der letzte Effektknall verhallt, dunkle Nacht. Lachend
und tappend suchen die Gäste in den Steigen des großen Gartens
sich den Weg, um ins Haus zu kommen, oder auch nicht ins Haus zu
kommen. Die Feuergeister haben manch tollen Spuk geweckt. Herr
Gregor, der junge Pfarrer, ist nun doch nicht nur für alle der
Mittelpunkt und der Zweck dieses Festes.
»Fräulein Fritz«, sagte Hans Henning. »Ich glaube gar, Sie haben
sich die Finger verbrannt.«
»Ja, das habe ich!« sagte sie mit der Verzückung eines Märtyrers.
»Ach! Tut es Ihnen weh?«
»Ja, fürchterlich!«
Er geriet gebührendermaßen außer sich. »Wir müssen sofort Öl
anwenden!« rief er.
»Nehmen Sie bitte meinen Arm, Fräulein Fritzchen. Sie stolpern
sonst über die eingeschlagenen Pflöcke. Ach, wie war ich
ungeschickt, das zuzulassen!«
Sie ließ ihn jammern, und überlegte dabei, wie es anzustellen
wäre, daß Gregor ihre Verletzung sähe. Ihr Mut war jetzt so
ungeheuer geschwellt, daß sie seine Ansprache, sein Staunen, sein
Bemitleiden gut hätte ertragen können, ohne umzufallen. Ja, sie
wünschte sich nichts heißer als das.
Aber als sie auf die erhellte Terrasse kamen, fing ihr Herz doch
wieder an zu klopfen. Da stand er, und gerade in der Glastür, durch
die sie gehen mußte!
»Fräulein v. Dörfflin hat sich beim Feuerwerk verbrannt«, sagte
Hans in aufgeregtem Ton zu ihm.
»Verbrannt«, rief Gregor. »Hans, wie konntest Du das zulassen?
Darf ich einmal sehen, gnädiges Fräulein? Da muß aber gleich für
Hilfe gesorgt werden.«
Fritzchen reichte ihm mit einem entzückten Lachen ihre Hände
hin, die Brandstellen waren in der Tat sichtbar. Aber sie wünschte
sich, daß ihre ganzen Hände verbrannt sein möchten, damit er doch
auch etwas Rechtes zu sehen bekomme. Doch es war schon vorbei,
ein paar Damen hatten sich sofort herangeschoben, sie wurde von
mehreren Armen umschlungen, geküßt, getröstet, bedauert bis zur
Unerträglichkeit. Selbst Hans Henning hatten sie von ihr
fortgedrängt, und in dieser Kohorte gelangte sie ins Schlafzimmer,
Öl, Watte, Läppchen waren da, Frau v. Zülchow, die Königin-Mutter,
rauschte auch herein und verband sie sorglich mit eigener Hand.
Solche Anstellerei um das bißchen Verbrennen! dachte Fritzchen
entsetzt.
Ja, aber wer hatte sich denn angestellt da draußen im Garten und
als man die Terrasse heraufkam?
Na, es ist auch so ganz gut. Nun hat man zwei verbundene
Hände, das sieht entsetzlich großartig aus und muß unendliches
Mitleid wecken!
Förmlich strahlend vor Prahlerei betrat das blessierte Fritzchen
wieder den hellen Bannkreis der Gesellschaft. Gregor kam zu ihr,
bedauerte sie, daß sie um seinetwillen leide, faßte ihre Hand und
berührte mit den Lippen die freie Stelle oberhalb der Verbandlappen.
Da wurde das Glücksgefühl so übermächtig in ihr, daß sie am
liebsten die Arme ausgebreitet hätte und ihm um den Hals geflogen
wäre.
* * *
Eine lange Rückfahrt im stockdunklen Wagen durch die graue
Sommernacht, die eintönig auf den weiten Feldern lag, nur am
Rande in dämmernden Strahlen umspielte das Sonnenlicht, das um
diese Jahreszeit nie ganz verlischt, den Horizont.
Man hatte die große Kutsche genommen, weil Gisela bei der Fahrt
durch den Wind Haar und Toilette geschont haben wollte. Nun sah
man nur durch die Glasfenster die grauen Felder vorübergleiten.
Fritzchen saß auf dem Rücksitz; sie hätte ihren heißen Kopf gerne
draußen in der kühlen Nachtluft gebadet, aber auch daß sie dies
nicht konnte, bedrückte sie heute nicht. Sie nahm den Hut ab, lehnte
den Kopf in die Ecke, und so zurückgezogen ins Dunkel, hörte sie,
was Frau v. Pohle und Gisa miteinander sprachen.
Frau v. Pohle sagte: »Welch ein eigentümlicher frühreifer Mensch
dieser älteste Sohn ist!«
Gisela fragte, ob er ihr gefiele. In der Betonung der Frage lag
schon die Annahme, daß jeder unbedingt Ja sagen müsse. Aber Frau
v. Pohle zögerte.
»Gefallen? Liebe Gisela, ich möchte das kaum sagen. Es ist mir zu
wenig Einfachheit und Lebensfrische an diesem jungen Menschen. Er
kommt mir vor, wie ein künstliches, wunderschönes Gebäude aus
Eis, das aber nur in einer Eisatmosphäre existieren kann. Vor der
Sonne müßte er schmelzen.«
Fritzchen fuhr auf, doch Gisela nahm ihr schon das Wort aus dem
Munde.
»Aber Frau v. Pohle! Er hat doch so viel Sonne um sich. Seine
Mutter, seine Freunde –«
»Nein, mein Kind, das ist nichts, das durchdringt seine
Atmosphäre nicht. Es ist eine seltsame Tragik um diesen jungen
Pfarrer, aber freilich fühlt er sie selber wohl am wenigsten.«
»Tragik?« sagte Gisela lächelnd. Sie fühlte sich manchmal
bewogen, über Frau v. Pohle zu lächeln. »Die wäre doch wohl nur
konstruiert. Haben Sie gehört, daß die Möglichkeit vorliegt, daß er
nächstens Hofprediger wird?«
»Hofprediger?«
»Ja, man sagte allgemein so. Das heißt, man flüsterte es sich zu.
Die jüngste Prinzessin, Maria, soll eine starke Vorliebe für ihn haben,
und sie vermag alles über ihren Vater. Sie hat Gregor Zülchow einige
Male predigen gehört und ihn auch zu sich befohlen. Sie soll ihm
große Avancen machen, man spricht in der ganzen Residenz davon.«
»Man spricht in der Residenz wohl so gern und viel wie überall«,
sagte die alte Dame leise mahnend. »Doch kann es ja so sein. Der
Weg, den Herr v. Zülchow geht, wird sicherlich nicht der
gewöhnliche aller Alltagsmenschen sein. Er wird viele Abenteuer
haben, die ihn »interessant« machen. Ach, meine lieben, jungen
Freundinnen, es ist eine bange Sache um einen Menschen, der sich
also vermißt, mit dem Leben und seinen Gestalten zu spielen, wie
dieser Freiherr und Pfarrer! Denn im letzten Grunde läßt das Leben
doch nicht mit sich spielen!«
Fritzchens Gesicht glühte. Hatte sie es nicht gewußt?
Prinzessinnen kamen und suchten seine Gunst! Ja, wohl war sein
Weg nicht der, den andere Leute gingen. Frau v. Pohles sonstige
Randglossen hörte sie nicht. In ihren Ohren klang es wie süße,
rauschende Musik.
»Der liebste aus diesem Rummelshöfer Hause ist mir der junge
Leutnant«, sagte Frau v. Pohle. »Da ist Frische und lebendige Kraft.
Der stellt sich des Lebens Dinge nicht wie Schachfiguren hin, damit
beliebig nach rechts oder links, vor- oder rückwärts zu ziehen. Der
lebt seinen Tag als frischer, unbekümmerter Junge. Der nimmt sich
nicht selbst auseinander und setzt sich wieder zusammen. Der ist
aus einem Guß!«
»Ach – Hans Henning«, sagte Gisela wegwerfend, in dem üblichen
Ton, in dem man gewöhnlich von ihm sprach, von der Mutter herab
bis zu den Bekannten des Hauses und den Reitknechten im Stall.
Frau v. Pohle lachte. »Das ist eine Geschichte zum Händeringen«,
rief sie aus. »Wißt Ihr wohl, woran es liegt? Diesem Jungen schadete
die Nachbarschaft seines glänzenden Bruders! Ihr alle habt, wenn
Ihr ihn ansaht, noch das Blenden im Auge von dem anderen. Ja –
Gregor! heißt es, und dann ganz sanft und barmherzig: Ach, der
Hans! Ich möchte Euch hier wohl mal ein bißchen eine Predigt halten
über Menschen und Menschenwert. Ich sage Euch, meine Lieben,
dieser Herr Gregor, so totenernst, würdevoll, eisig und alt er
aussieht, der versteht das Leben nicht und nimmt es im Grunde
nicht so ernst wie dieser lustige, prachtvolle Schlingel, der Hans. Der
wird seinerzeit verstehen, damit zu ringen und es sich untertan zu
machen. Was dieser künftige Hofprediger damit anfangen wird – ach
ja, da wird das Zuschauen kein großes Vergnügen sein. Er wird sich
daran vorbeidrücken, denke ich.«
»So denken Sie?« rief Fritzchen mit ausbrechender Wildheit. Sie
zerknitterte ihren Hut, daß das arme Stroh laut krachte und
knirschte. »Ich denke anders! Alle denken anders! Herr Gregor wird
sich nie an etwas vorbeidrücken! Sie haben ihn einmal gesehen, ich
kenne ihn, als er noch Junge war. Er ist der bedeutendste und
größte Mensch auf Erden! Ich lasse kein Wort auf ihn kommen. Ich
habe ihn lieber als Himmel und Erde!«
Da war's heraus! Was tat es? Die ganze Welt konnte es hören, daß
sie hier stand, bereit zu leben und zu sterben für ihn.
Sie hatte auch bei ihrem letzten Ruf aufrecht gestanden in der
Kutsche, aber das Rad fuhr über einen Stein, da fiel sie unrühmlich
auf ihren Sitz. Das machte nichts aus. Sie gab noch einen
flammenden Satz dazu:
»Ich verlache jede Beschuldigung über ihn!«
»Aber Fritzchen! Wie ungezogen!« rief Gisela entsetzt. »Wie
kannst Du so zu Frau v. Pohle sprechen! Da sehen Sie wieder, wie sie
ist!«
»Lassen Sie doch, Kind«, sprach Frau v. Pohle. »Liebes Fritzchen,
ich habe nicht gedacht, Ihr Herz zu kränken. Um Gott, Kind, nein.
Sehen Sie einer alten Frau solche Gedankenspielerei nach. Das
Leben wird ja erst beweisen, was richtig und was barer Unsinn an
meinem Geschwätz war.«
»Aber wie kannst Du Deine Neigung so ausschreien!« rief Gisela,
noch immer aufgeregt vor Entrüstung. »Das tut man doch nicht. Du
blamierst Dich ja grenzenlos, Frida!«
»Was kümmert's mich!« entgegnete Fritzchen trotzig.
Frau v. Pohle dachte: Jawohl – Dich, starkes, junges Herz,
kümmert's in der Tat nicht, ob Du Dich blamierst. Das ist das
zweifelhafte Vorrecht derer, die Dich tadeln. Glückauf, Du freie
Menschenseele!
Aber sie sagte das nicht. Für alle ihre Predigten war ihre
Zuhörerschaft in der dunklen Kutsche doch noch nicht reif genug.
Behalte Du nur Deinen Gregor! dachte sie ohne Sorge, solange
wie Dein Herz dieses Bild tragen mag, Du schönes, wildes Kind. Ich
traue: eines Tages siehst Du Dich verwundert um, wo es geblieben
ist.
Siebentes Kapitel.
Aber es ist ein undankbares Amt, Prophet zu sein.

Fritzchen schloß ihre Blumenblätter fester um sich zusammen. –
Hier war eine Gelegenheit, so groß und stark, so weit und sonnig,
eine Mutter zu haben, eine edle Freundin, ein fröhliches und doch
weises Herz, stets aufmerksam und zur Stelle.
Nein, Fritzchen will das nicht. Sie geht in ihre Turmstube und
spricht mit den Wolken, die über das Moor gehen. Das sind ihre
Freunde und Gesellschafter.
Sie gehen massig, grauschwarz, mit hellen, blendenden Rändern.
Unten sind es nur noch ziehende Schneeberge. Sie ziehen vorüber,
andere kommen und ziehen auch. Schon wieder ist das Bild
verändert. Die Sonne kämpft mit der schwarzen Wand, sie kämpft
umsonst, sie erlischt. Horch, wie der Wind in Stößen kommt!
Dies Kind kennt die Wolken wie keines und den Wind ebenso. Ist
es nicht mit ihm um die Wette geritten über die bräunliche Ebene?
Vielleicht ist es Gregor v. Zülchows eigenes Leben, sein kräftigstes,
schönstes Leben, das da mit fliegendem Haar auf der schwarzen Möt
über den Graben setzt hinter dem Böllinger Kreuzweg! Halt fest! –
Oder läßt er es vorbei?
Nun ja, der Hans muß ja auch etwas haben, wenn er Hofprediger
wird!
Gregor steht am Graben neben dem Kreuzweg. Welcher Teufel
oder welcher Engel führte ihn hierher, da der Wind sauste und dies
tolle, herrliche Bild heranführte?
Er stand still, am Fußsteige neben einem Baum.
»Hussa, Möt!«
Er aber war nicht mehr als ein Baum am Wege. Herrin, Königin in
diesem Bilde war sie, die daher kam, sie, die Schwester und Braut
des Sturmwinds!
Sie war vorüber, ohne ihn gesehen zu haben!
Wer sieht die Bäume an, wenn er mit den Wolken Haschen spielt?
Er stand und sah ihr nach, der Staub flog hinter ihr auf.
* * *
Hans Henning kam zu seiner Mutter, die mit Gregor auf der
Veranda saß. Morgen hieß es für beide Söhne wieder scheiden. Hans
Henning mußte, Gregor wollte, so war es schon manch liebes Mal
gewesen.
Der Abend dämmerte. Der Himmel stand regendrohend über den
Bäumen des Gartens, sie raunten leise wie in bangem Vorgefühl. So
bange war auch der Baronin zu Mut. Ach, dies immer neue
Scheideweh! Und wer hielt des Scheidenden Herz, daß ihr
wenigstens das blieb?
Ja, Gregor lieben, das hieß, täglich sterben. Liebte sie diesen Sohn
am höchsten auf der Welt, so litt sie auch um ihn am tiefsten. Er
rächte unbewußt den anderen, den Übersehenen, den immer matt
und halb Geliebten.
Wie war dieses Jüngeren Herz jetzt voll! Stand es nicht auf seiner
Stirn geschrieben, auf seinem Mund, seinen Augen, seinen Händen,
in jeder Bewegung, die er tat? Aber man hatte keine Zeit, diese
Ziffern zu lesen. Es war ja nur Hans!
Die Mutter sprach mit Gregor. Sie tippte an etwas, das auch sie
hatte von fern nur läuten hören, ohne daß er ein Wort darüber
verloren hatte: an seine Hofprediger-Aussichten.
»Gregor – ist etwas daran? Ist das möglich?«
Wie kalt und stolz blickten die blauen Augen! »Möglich? Ja, Mama.
Aber lassen wir das, Du erfährst jede Tatsache, sobald sie vollendet
ist!«
Gewiß, gewiß, sie erfuhr jede vollendete Tatsache. Das war die
Speise, die ihr bester Sohn ihr gab, wenn ihr Mutterherz verhungern
wollte.
Hans Henning aber brannte das Herz in der Brust. Ach, er wollte
etwas anderes geben, als eine vollendete Tatsache. Ihn riß es, vor
der Mutter hinzuknien, den Blondkopf in ihren Schoß zu legen.
Mutter, bist Du zufrieden und froh, wenn ich mir das Fritzchen hole?
Leg' mir Deine kühle, weiße Hand auf, Mutter. Ich will reiten,
morgen früh, ehe wir reisen, und das Fritzchen im Garten, oder in
der Wiese oder in der Turmstube suchen und es nach seinen
verbrannten Fingerchen fragen. Und dann auch noch nach etwas
anderem. Gib mir einen Kuß, liebe Mutter, ich brauche sehr viel Mut.
Mein kleiner Fritz träumt und fabuliert noch gar soviel, der merkt
noch gar nichts von dem großen wilden Strom. Mutter, Du bist so
klug, sage mir doch, ob es schon Zeit ist, sie aufzuwecken, oder ob
es noch zu früh ist –.
»Lieber Gregor, wenn Du hin und wieder ein klein wenig mehr
schreiben könntest – ich meine natürlich nur, wenn Du Zeit und Lust
hast – aber manchmal ein bißchen ausführlicher, weißt Du. Ich kann
mir oft so gar kein Bild von Deinem Leben machen. Nur so
manchmal erzählen: Ich war dort oder dort, und wir haben dies und
das gesprochen. Oder von dem, was Dich innerlich beschäftigt,
Gregor, daß ich ein ganz klein wenig teilnehmen kann –«
Gregor lächelte. Es war ein flüchtiges, durchaus höfliches Lächeln,
aber seine Mutter fürchtete sich davor. Auch jetzt wieder stieg es ihr
heiß in die Wangen.
»So wird es sich schwer machen lassen, liebe Mama. Es fehlt die
Zeit und die Unbeschäftigtheit, auf Deine Ideen einzugehen. Aber ich
will möglichst daran denken.«
Er stand auf und sah über die Brüstung auf den umzogenen
Himmel. In der Ferne wetterleuchtete es. Er dachte an ein anderes
fernes Wetterleuchten, das an ihm vorbeigefahren war, heute am
Kreuzweg, wo der breite Graben war.
Hans Henning sagte leise: »Mutter!«
Die fuhr aus schwerem Sinnen auf. »Ja, was willst Du?«
Der Ton der Frage war scharf und klar. Er klang nach einer
Antwort wie die: »Mir fehlt noch Geld, Mutter.« Oder: »Ich brauche
noch dies und das, wenn ich in die Garnison zurückkomme.« Aber
nimmermehr nach einer so leisen, unbeholfenen Bitte: »Mutter, gib
mir Rat. Ich bin Dein dummer, kleiner Junge und weiß nicht, wie ich
mein Glück anfassen soll.«
Er war neben sie getreten, auch sie stand auf. Die Luft wehte
kühler und schärfer, sie zog fröstelnd ihr leichtes Tuch über den
Schultern zusammen. Sie war sehr groß, schlank und von stolzer
Haltung wie ihr anderer Sohn.
Als Hans nicht gleich antwortete, sagte sie nervös: »Nun, was
gibt's denn? Sprich doch schnell.«
In dem großen Jungen stieg eine jähe Bitterkeit auf. Noch nie
hatte er empfunden wie heute, daß seine Mutter eigentlich niemals
Zeit für ihn hatte. »Sprich schnell!« Ja, so war es immer gewesen.
Ihm war ja auch sonst damit gedient, lang schleppende
Auseinandersetzungen waren wahrlich nicht sein Fach.
Heute abend hätten sie vielleicht doch gepaßt. Oder auch nicht.
Vielleicht hätten drei Worte es getan, aber davon konnte er im
voraus nichts wissen. Doch dazu muß man stillsitzen und Zeit haben.
Aber wer hat für ihn Zeit, er ist ja nur der Hans.
»Laß nur, Mama. Es war nichts.«
Sie sah schon wieder von ihm fort. »Kommst Du mit herein,
Gregor?« bat sie.
Der kam von der Brüstung der Veranda, wo er in die Blitze
gesehen hatte.
»Das Gewitter kommt nicht herauf«, sagte er in einem
beruhigenden Tone.
Wen beruhigte er denn? Die Mutter und Hans hatten andere Dinge
im Kopf, als ein Gewitter, das kommen könnte. Ach ja, es ist ein
wunderlich trauriges Ding um solch einsames Nebeneinander, wo der
eine Blitze sieht, der andere aber in den Blitzen ein schönes wildes
Kind, oder wo eine arme Seele stumm ihr Leid anschaut, und es
nicht verstehen kann. –
Hans Henning sah den beiden nach, wie sie durch die Glastüren
gingen. Der helle Schein der Zimmerlampen überströmte sie. Da
ging die kurze Bitterkeit in dem Jüngling unter.
Wem gab er denn Schuld, wenn er nur der dumme, lustige,
beiseite geschobene Hans war? Wollte er mit dem Bruder hadern,
weil er größer war, oder mit der Mutter, weil sie das fühlte und sah?
Such' Dir doch Deine Wege selber, Narr. Wer heißt Dich, noch an
der Mutter Schürzenband zu hängen?
Er blieb auf der dunklen Veranda allein. Die Bäume rauschten
stärker, Träume umfingen ihn. – Morgen in der Frühe, Du, mein
wilder kleiner Vogel, werde ich Dich da fangen können? O, habe
nicht Angst, ich will Dein feines Gefieder nicht zerdrücken und
verletzen. Wie wir gelacht und fabuliert haben als Kinder, so wollen
wir es weiter tun. Weißt Du noch die Schaukel in Eurem Garten,
Fritz? Du wolltest ja immer so gerne fliegen. O, Du mein
Prinzeßchen, ich könnte die Welt zerschlagen, um sie nach Deinem
Gefallen aufzubauen. Und weißt Du noch, wie wir gestern abend die
Feuergeister waren? Liebling, kleiner, Deine verbrannten Fingerchen
möchte ich wiedersehen!
Sieh mal die Blitze da hinten, wie sie den Himmel aufreißen.
Wollen wir hineingehen, Frida? Ach, wie ist das Glück so bange – ich
habe es nie gewußt.
»Hans, sage mal, was treibst Du hier draußen! Mama ist schon zu
Bett. Seit einer Stunde sitzest Du hier!«
Hans Henning fuhr auf. »Seit seiner Stunde –?«
»Junge, fehlt Dir etwas? Wie Deine Hände heiß sind! Warum
kamst Du nicht mit herein?«
»Ich – weiß nicht. Ich habe mich wohl hier verträumt.«
»Hans – sag mir mal die Wahrheit. Du bist wohl verliebt?«
Sie standen jetzt beide beieinander. Es war zu dunkel, als daß sie
ihre Gesichter sehen konnten. Wie stark der Wind gewachsen war!
Es war ein Sausen und Brausen in den Bäumen des Gartens.
»Ich bin Dein Bruder, mein Junge«, sagte Gregor. »Du kannst Dich
mir vertrauen.«
»Ja!« rief Hans Henning aus. Es war ein Ton des lautersten
Frohlockens.
»Gregor – ich – siehst Du –«
Nein, es ging doch nicht. Hans sah verzweifelt zur Seite. Wie
macht man es denn, daß man so etwas sagt?
»Es ist wohl Frida Dörfflin, um die es sich handelt«, sagte Gregor
in völliger Gelassenheit.
Hans Henning antwortete nicht, das Herz schlug ihm bis zum
Halse. Wie hatte die Nennung dieses Namens ihn durchschnitten!
Wie konnte es nur so unsinnig weh tun, den Namen so kühl wie
geschäftsmäßig nennen zu hören.
Gregor genügte wohl diese stumme Antwort. Er schwieg einen
Moment.
»Das Kind!« sagte er dann in wegwerfendem Ton.
Hans Henning schoß das Blut wild ins Gesicht.
»Was meinst Du damit? Ich lasse nicht an ihr rühren.«
»Wer tut denn das?« sagte Gregor nachlässig. »Meinst Du etwa,
sie sei kein Kind mehr? Verstehst Du Dich so wenig auf
Menschenaugen? Bei diesem Mädchen ist alles noch Klarheit,
Harmlosigkeit und ein vollständiges Spielen dem Leben gegenüber.
Ich spreche das nicht als Tadel aus, sondern nenne es einen
Vorzug.«
»Ich habe das auch schon gedacht«, murmelte Hans. Da stand er
ja mit einem Male mitten in Frage und Antwort, Bitte und Rat, wie er
es sich noch vorhin so sehr gewünscht hatte. Nur daß es nicht die
Mutter war. Aber vielleicht wußte Gregor noch mehr von diesen
Dingen.
»Ich dachte sonst – ich wollte morgen, vor der Abreise –«,
stotterte der große Junge. »Ich will ja auch gar nichts an ihr
zerstören, Gregor. Ich will ihr gar nicht viel von Liebe vorschwatzen.
Nur wissen – ob sie mein Kamerad sein will – diese Ungewißheit,
Gregor, die ist ja zu gräßlich –«
Gregor wandte sich ab und ging mit starken Schritten zweimal die
Veranda auf und nieder. Dann blieb er vor Hans stehen, und als er
sprach, klang seine Stimme wie geschliffener Stahl.
»Ihr seid beide noch Kinder. Durchbrich diesen Zustand nicht aus
Übermut. Du schadest ihr und Eurem ganzen Verhältnis, Du
veranlassest sie, sich zu verschenken, ehe sie sich kennt. Glaubst Du
nicht, Hans, daß das eine Sünde an ihrem Geist und Wesen ist?
Lerne warten, mein Junge, überlaß diese Sache der Zeit, bis die
Blumen von selber aufbrechen.«
»Ist das so –?« stotterte Hans.
»Ja, Hans Henning, das ist so!« sagte Gregor.
Sein Ton legte sich wie eine eiskalte Hand dem Jüngling aufs Herz.
– Hatte er vielleicht doch nur nach Hilfe verlangt, um die Antwort zu
hören, die er wünschte?
Oder ahnte sein erwachtes Herz das Schwert in des Bruders
weisem und wohlbegründetem Rat? Seine Lippen bebten, als er
sagte: »Ich danke Dir. Ich will jetzt schlafen gehen!«
Er ging. Die Sporen klirrten leise. In dem erleuchteten Zimmer sah
er im Spiegel sein Bild vorübergleiten. – »Überlaß es der Zeit –«
Vielleicht hat er recht, und ich handle verrucht, seine Worte zu
verachten. Vielleicht bin ich ein Narr, wenn ich es nicht tue. Ich bin
nicht weiter als zuvor.
Überlaß es der Zeit!
Ja, Du kaltes Herz, Dir steht es wohl an, so zu sprechen. – Aber
hat ein heißes Herz mehr Recht und weiteren Blick?
Das wird heute eine friedvolle Nacht!
Gregor stand noch draußen. Das Gewitter hatte sich verzogen
oder war landeinwärts niedergegangen. Es blitzte nicht mehr. Nur
ein leichter Regen, durch den Wind hin- und hergetrieben, sprühte
durch die Bogenöffnung der Veranda und tanzte oben auf dem
Glasdach. Die Luft war stark abgekühlt.
Er stand eine kurze Weile und sah hinaus. Noch lagen seine
eigenen klingenden Worte ihm im Ohr. »Überlaß es der Zeit, bis die
Blumen von selber aufbrechen.«
Als er nach oben kam, fand er die Verbindungstür nach Hans

Hennings Schlafzimmer geschlossen. Der Knabe grollt, dachte er,
weil ich ihm sein Spielzeug fortnahm. Aber es ist nun einmal so. Ich
tat recht, den Fuß darauf zu setzen. Das alles muß erst werden, und
die Zeit ist mächtig. Gute Nacht, mein trotziger Junge, es kann sein,
daß wir uns noch einmal an dieser Stelle treffen.
Vielleicht! Wer weiß es?
Er legte noch die letzte Hand an seinen Koffer. Den Rock, den er
heute getragen hatte, zog er aus und packte ihn dazu. Dabei nahm
er seine Brieftasche heraus und öffnete sie. Nur eine Kerze brannte
auf dem Tische. An die Fenster trieb der Regen und die seitwärts
geschlossenen Läden klapperten im Wind.
Gregor trat an das Licht und sah auf das Mädchenbild nieder, das
er in Händen hielt. Es war ein feines Gesicht mit nervösem Mund
und großen Augen: Prinzessin Maria, sie hatte es ihm selber
geschenkt.
Nun beginnt das wieder! dachte der junge Pfarrer. Ja, die Erde ist
weit, und wir haben Zeit zu vielen Dingen. Aber sie sollen mir alle
dem größesten dienen!
* * *
Hans Henning hatte nicht aus Groll und Trotz die Verbindungstür
zugemacht. Aber er empfand ein unklares, quälendes Gefühl der
Furcht dem Bruder gegenüber, ein eisiges Widerstreben und
Ablehnen, das er früher nie gekannt hatte. Eine Art Mißtrauen, das
er sich dennoch nicht erklären konnte. Es wäre ihm fürchterlich
gewesen, ihn heute noch sehen oder sprechen zu müssen.
Aber was nun weiter? Er hat gesprochen wie einer, der die
Menschen und Dinge kennt. Ach ja, er wußte immer von jeher
Bescheid.
Tue ab diesen miserablen Wust eigenen Wünschens und
Parteinahme für Dich selbst! sagte Hans Henning zu sich und
drückte die heiße Stirn an die beregnete Fensterscheibe. Mach's
gefällig ein bißchen klar in Dir, oder Du bist ein Hund! Es kommt
hierbei doch auf das Wohl, das Glück und die Seele meines kleinen
Fritz an – sonst auf nichts. Und warum wollte ich sie fragen? Um
mich zu beruhigen!
Also gut. Gregor hat recht. Gregor hat recht. Ganz egal, wie's mir
eingeht oder einleuchtet, Gregor und die Vernunft haben recht.
Ach ja. Man muß auch manchmal Steine auf der Brust haben, das
wird wohl nicht anders gehen.
Ich reite also morgen früh nicht. Nein – nicht. Laß es doch
brennen.
Nein, jetzt nicht wieder anders denken. Ruhe!! Ich reite nicht.
Aber eins kann ich doch noch.
Er ging an einen alten Sekretär, schloß ihn auf, und aus dem
hintersten Geheimfach, das wohl sonst schon Gott mochte wissen,
welchen Familiengeheimnissen gedient hatte, holte er ein altes,
verblichenes Kinderbildchen hervor. So hatte der Fritz ausgesehen,
als er ihn zum ersten Male sah.
Hans hatte das Bild in einer Rumpelkammer auf Hohen-Leucken
gefunden und ohne Gewissensbisse gestohlen. Wie zärtlich er es
liebte! Es war sein kostbarster Schatz.
Die Augen sahen etwas trotzig und düster drein. Es war dem
kleinen Kopf wie eine arge Zumutung erschienen, sich so dem
Photographen hinzustellen. Sogar der süße Kindermund schien leise
zu zucken. Das Hälschen, die runden kleinen Arme waren bloß, ein
einfaches schottisches Kleidchen hing bis auf die Knie nieder.
»Damals hast Du die Tasse auch schon übergeschwappt, Fritz!«
sagte Hans Henning.
Er küßte das liebe kleine Bild.
»Leb wohl, mein Liebling. Du mußt noch ein bißchen wachsen.
Hörst Du, Unband, wachse schnell. Sonst halte ich das Warten doch
nicht aus.«
Tür an Tür waren sie miteinander, und jeder hielt ein Bild. Der
Hans mit seiner einzigen armen, starken jungen Liebe, dieser Hans,
der nicht warten konnte, der mit seiner wilden, herrlichen Ungeduld
rang wie mit einem durchgehenden Pferde – und der andere, der
durch Blumenbeete ging und zu den Knospen sagte: Blüht nur erst
auf, dann werden wir sehen. Es ist Zeit da zu vielen Dingen – –
Achtes Kapitel.
– – – – Fritzchen saß bei der schwarzen Hede Marusch im Dorf,

die im Bett lag und die Auszehrung hatte. Vor zehn Jahren und mehr
hatte sie auf dem Schlosse gedient, Fritzchens schmutzige Kleider
gewaschen und den Boden auch nicht reinmachen wollen. Sie war
gerade so brav und nichtsnutzig, geschäftig und faul gewesen, wie
die anderen Mägde auf Hohen-Leucken auch. Jetzt hatte sie einen
Hofgänger geheiratet, hatte vier lebendige und zwei tote Kinder, und
ging mit eiligen Schritten, als könne sie es kaum abwarten, dem
Sterben zu. Sie hustete sich die arme Lunge heraus, und Fritzchen
hatte Gelegenheit, in ein Bild des Jammers zu sehen, das von
keinem freudigen Strahl erhellt wurde.
Sie kannte aus ihrem trübseligen Dorf diese Bilder von Jugend auf.
Da liefen diese Mädchen und Frauen herum, arbeiteten sich ab,
fingen an zu husten, und dann sahen sie mit heißen, starren,
verzweifelten Augen aus ihren bunten Federkissen heraus.
Solange noch Lebenshoffnung glimmte, waren sie untertänig und
voll schmeichelnder Dankbarkeit gegen ihr junges Fräulein, das sie
zu besuchen und ihnen Erfrischungen zu bringen kam. Alte Frauen
erzählten ihr geläufig von der Engelsgüte ihrer seligen Mutter. Aber
wenn die Hoffnung erlosch, hörte dies alles auf. Dann sah der nackte
Menschheitsjammer, dem hoch und niedrig zu leeren Lauten werden,
ihr unverhüllt in das junge Gesicht.
»Wissen Sie noch, Fräulein«, sagte Hede Marusch, »als man mir
meinen lütten Auta begrub? Da habe ich immerfort gebarmt, ich
wollt mit. Nee, nee, Fräulein, man soll sich das nicht wünschen. Das
Grab ist kalt, und all die Leute, die hier schon unter die Erde

Full download Innovative Food Processing Technologies Advances in Multiphysics Simulation 1st Edition Kai Knoerzer pdf docx

Uploaded by

Document Informationclick to expand document informationInnovative

Document Informationclick to expand document information

Copyright:

Available Formats

Full download Innovative Food Processing Technologies Advances in Multiphysics Simulation 1st Edition Kai Knoerzer pdf docx

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

Full download Innovative Food Processing Technologies Advances in Multiphysics Simulation 1st Edition Kai Knoerzer pdf docx

Uploaded by

Copyright:

Available Formats

Download the full version of the ebook at ebookfinal.

Innovative Food Processing Technologies Advances

Download more ebook instantly today at https://ebookfinal.com

Innovative Food Processing Technologies Advances in

Electron Beam Pasteurization and Complementary Food

Advances in Food Biotechnology 1st Edition Rai

Information Technologies in Medicine Medical Simulation

Proteins in Food Processing 2nd Edition Rickey Y. Yada

Advances in Battery Technologies for Electric Vehicles 1st

Enzymes in Food and Beverage Processing 1st Edition

Environmentally friendly food processing Berit Mattsson

A John Wiley & Sons, Inc., Publication

A John Wiley & Sons, Inc., Publication

Editorial offices: 2121 State Avenue, Ames, Iowa 50014-8300, USA

Library of Congress Cataloging-in-Publication Data

Set in 10/12 Times by Toppan Best-set Premedia Limited, Hong Kong

1. Introduction to Innovative Food Processing Technologies: Background, Advantages,

2. The Need for Thermophysical Properties in Simulating Emerging Food Processing

3. Neural Networks: Their Role in High-Pressure Processing, 39

4. Computational Fluid Dynamics Applied in High-Pressure Processing Scale-Up, 57

5. Computational Fluid Dynamics Applied in High-Pressure High-Temperature Processes:

6. Computer Simulation for Microwave Heating, 101

7. Simulating and Measuring Transient Three-Dimensional Temperature Distributions

8. Multiphysics Modeling of Ohmic Heating, 155

9. Basics for Modeling of Pulsed Electric Field Processing of Foods, 171

10. Computational Fluid Dynamics Applied in Pulsed Electric Field Preservation

11. Novel, Multi-Objective Optimization of Pulsed Electric Field Processing

13. Computational Study of Ultrasound-Assisted Drying of Food Materials, 265

14. Characterization and Simulation of Ultraviolet Processing of Liquid Foods Using

15. Multiphysics Modeling of Ultraviolet Disinfection of Liquid Food—Performance Evaluation

16. Continuous Chromatographic Separation Technology—Modeling and Simulation, 335

Color plate section appears between pages 208 and 209.

Víctor M. Acosta Hao Chen

Serafim Bakalis Özgür Ertunç

Gustavo V. Barbosa-Cánovas Larry Forney

Juan A. Gallego-Juárez Jens Krauss

José Vicente García-Pérez

Pablo Juliano Georgina Porras-Parral

Enrique Fernando Riera Franco de Sarabia José S. Torrecilla

Pedro D. Sanz Francisco Javier Trujillo

Introduction to Innovative Food Processing

1.1. Introduction concept development to implementation. In particu-

Innovative Food Processing Technologies: Advances in Multiphysics Simulation, First Edition.

processing technologies often encounter the same 1.2. Multiphysics Modeling

Table 1.1. Tools for the validation of Multiphysics models.

Process variable or Direct (D)/

validation by means of a (bio)chemical or microbial • sound intensity by means of hydro- or microphones

nutrient degradation, structural changes, and mate- 1.5. Concluding Remarks

The Need for Thermophysical Properties in Simulating

Pablo Juliano, Francisco Javier Trujillo,

2.1. Introduction Working with thermophysical properties in

Innovative Food Processing Technologies: Advances in Multiphysics Simulation, First Edition.

Technology Property Equation

2.2.9. Molar Absorptivity

To satisfy the need for these properties, empirical V Volume m3

Neural Networks: Their Role in High-Pressure Processing

José S. Torrecilla and Pedro D. Sanz

Innovative Food Processing Technologies: Advances in Multiphysics Simulation, First Edition.