Caracterización química y estructural

de lignina y lípidos de materiales

lignocelulósicos de interés industrial

Pepijn Prinsen
Sevilla, 2013
Caracterización química y estructural de lignina y
lípidos de diversos materiales lignocelulósicos de
interés industrial

Memoria que presenta

Pepijn Prinsen
para optar al título de Doctor en Ciencias
Químicas por la Universidad de Sevilla.

Sevilla, 24 de Mayo de 2013.

Caracterización química y estructural de lignina y
lípidos de diversos materiales lignocelulósicos de
interés industrial

Visado en Sevilla, a 24 de Mayo de 2013.

LOS DIRECTORES

Prof. José C. del Río Andrade Dra. Ana Gutiérrez Suárez

Profesor de Investigación del CSIC Investigador Científico del CSIC
IRNAS-CSIC IRNAS-CSIC

EL TUTOR

Prof. José Mª Fernández-Bolaños Guzmán

Profesor Titular de la Universidad de Sevilla

Memoria que presenta

Pepijn Prinsen
para optar al grado de Doctor en
Ciencias Químicas por la Universidad
de Sevilla.
DOCTOR D. JOSÉ MANUEL PARDO PRIETO, DIRECTOR DEL
INSTITUTO DE RECURSOS NATURALES Y AGROBIOLOGIA DE
SEVILLA DEL CONSEJO SUPERIOR DE INVESTIGACIONES
CIENTÍFICAS

CERTIFICA: Que la presente Memoria de Investigación titulada

“Caracterización química y estructural de lignina y lípidos de diversos
materiales lignocelulósicos de interés industrial“, presentada por Pepijn Prinsen
para optar al grado de Doctor en Ciencias Químicas, ha sido realizada en el
Departamento de Biotecnología Vegetal, bajo la dirección de los Drs. José C.
del Río Andrade y Ana Gutiérrez Suárez, reuniendo todas las condiciones
exigidas a los trabajos de Tesis Doctorales.

En Sevilla, a 24 de Mayo de 2013.

 AGRADECIMIENTOS

En primer lugar, quiero agradecer a los institutos, instancias y fondos de

financiación que han permitido realizar esta Tesis. Este trabajo se ha llevado a
cabo en el Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS-
CSIC). Ha sido financiado por una beca FPI del Ministerio de Educación y
Ciencia, asignada al proyecto de investigación ´´Utilización de cultivos agrícolas y
forestales para la producción de pasta de papel: Tratamientos enzimáticos para la
eliminación de lípidos y lignina´´ (AGL2008-00709), y por el proyecto europeo
´´Optimised pre-treatment of fast growing woody and nonwoody Brazilian crops by
detailed characterisation of chemical changes produced in the lignin-carbohydrate
matrix (LIGNODECO)´´ (KBBE-244362).

Quiero expresar mi sincero agradecimiento a las personas que tanto directa

como indirectamente han hecho posible la realización de esta Tesis:

A la Dra. Ana Gutiérrez y al Prof. José Carlos del Río, directores de esta
Tesis, por todo lo que me han aportado, a nivel científico y personal, por sus
conocimientos, enseñanzas y consejos, por la confianza en su trato personal, por
mostrar su paciencia con mis dudas científicas y lingüísticas, y por estar siempre
cuando los he necesitado. Su esfuerzo y dedicación han sido decisivo en esta
Tesis.

Al Prof. Ángel T. Martínez, del Centro de Investigaciones Biológicas (CIB-

CSIC, Madrid), cuya capacidad de gestión y organización de los recursos
científicos, técnicos y humanos ha sido una gran aportación a esta Tesis.

A la Dra. Tarja Tamminen, del ´´VTT Technical Research Centre of Finland´´

(Espoo, Finlandia), por ofrecerme la posibilidad de realizar una estancia en su
grupo de investigación, por su trato familiar desde el primer día, y por aportarme
valiosos conocimientos científicos y sugerencias. También quiero agradecer a las
Dras. Tiina Liitiä, Stella Rovio y Harri Heikkinen por su ayuda con diferentes
análisis durante mi estancia en el VTT.

Al Prof. José Ma Fernández-Bolaños, Profesor Titular de la Universidad de

Sevilla, tutor de esta Tesis, por su apoyo en la parte burocrática, y por su tiempo
para intercambiar ideas interesantes.

A mis compañeros, las Dras. Gisela Marques y Edith Marleny, y en particular

al Dr. Jorge Rencoret, por su dedicación y tiempo a enseñarme técnicas y
métodos, intercambiar conocimientos y compartir momentos valiosos.

A mis compañeros Esteban Babot, Alejandro Rico, Antonio Pereira y

Andrés Olmedo, con los que he coincidido durante el transcurso de esta Tesis y
quienes han estado más cerca de mi en el día a día.
 Al Dr. Manuel Angulo y al Prof. José Luis Espartero, por ofrecernos su
servicio y asistencia en los análisis de NMR en el Centro de Investigación
Tecnológica e Innovación de la Universidad de Sevilla (CITIUS).

Al Prof. Jesús Jiménez-Barbero, a la Dra. Lidia Nieto y al Dr. M. Álvaro

Berbis del CIB-CSIC, por su servicio y asistencia en los múltiples análisis de
NMR.

A mis compañeros Asunción Castro y Álvaro Ramos del servicio de análisis

del IRNAS, por los múltiples análisis y por su trato personal.

A mi madre y padre, mis hermanos y mis abuelos por estar cerca de mí

aunque estén lejos en distancia. Y por último, a quién más quiero en este mundo,
Susana González, sin olvidar el apoyo de sus familiares Carolina, Laura,
Carmen y Félix, quienes me han acogido como un hijo más.
 ABREVIATURAS

AFEX Explosión de fibra con amonio

´´Ammonia Fiber Explosion´´
AQ Antraquinona
ARP Percolación recirculativa con amonio
´´Ammonia Recycle Percolation´´
BSTFA N,O-bis-(trimetilsilil)-trifluoroacetamida
DFRC Derivatización seguida de rotura reductora
´´Derivatization Followed by Reductive Cleavage´´
DMSO Dimetilsulfóxido
ECF Libre de cloro elemental
´´Elementary Chlorine Free´´
G Unidad guayacilpropano (o guayacilo)
GAX Glucuronoarabinoxilanos
GC Cromatografía de gases
´´Gas Chromatography´´
GC/MS Cromatografía de gases/espectrometría de masas
´´Gas Chromatography/Mass Spectrometry´´
GPC Cromatografia de permeación en gel
´´Gel Permeation Chromatography´´
H Unidad 4-hidroxifenilpropano (o 4-hidroxifenilo)
h Horas
HBT 1-Hidroxibenzotriazol
HPLC Cromatografia líquida de alta resolución
´´High Performance Liquid Chromatography´
HMBC Espectroscopía 2D de correlación de múltiples enlaces
´´Heteronuclear Multiple Bond Correlation´´
HSQC Espectroscopía 2D de correlación heteronuclear de cuanto simple
´´Heteronuclear Single-Quantum Correlation´´
ICP-OES Espectrometría de emisión óptica con plasma acoplado inductivamente
´´Inductively Coupled Plasma-Optical Emission Spectrometry´´
ID Diámetro interno
´´Internal Diameter´´
ITD Detector de trampa de iones
´´Ion Trap Detector´´
K Índice Kappa
LCC Complejo de lignina-carbohidrato
´´Lignin-Carbohydrate Complex´´
Min Minutos
MWL Lignina de madera molida
´´Milled Wood Lignin´´
NMR Resonancia magnética nuclear
´´Nuclear Magnetic Resonance´´
PCA p-Cumarato
PDA Detector UV de matriz de diodos
´´Photodiode Array´´
ppm Partes por millón
PTFE Politetrafluoroetileno
Py-GC/MS Pirólisis-cromatografía de gases/espectrometría de masas
´´Pyrolysis-Gas Chromatography/Mass Spectrometry´´
rpm Revoluciones por minuto
S Unidad siringilpropano (o siringilo)
SEC Cromatografia de exclusión de tamaño molecular
´´Size Exclusion Chromatography´´
SPE Extracción en fase sólida
´´Solid Phase Extraction´´
TAPPI ´´Technical Association of the Pulp and Paper Industry´´
TCF Totalmente libre de cloro
´´Totally Chlorine Free´´
TMAH Hidróxido de tetrametilamonio
TMSi Trimetilsililo
UV Ultravioleta
ξ Coeficiente de extinción molar
 INDICE

RESUMEN .…….……………………………………..……..…….…….. i
ABSTRACT ............................................................................................... iii

I. INTRODUCCIÓN …………………………...………….................. 1
I.1. Cultivos lignocelulósicos de interés industrial …………………… 3
I.1.1. Cultivos forestales ……………………………...…………… 3
………………………...
I.1.2. Cultivos agrícolas …………….……..……………………….. 4
I.2. Estructura y composición química de los materiales
lignocelulósicos …………………………………………………... 5
I.2.1. Celulosa ……………..….......................................................... 6
I.2.2. Hemicelulosas ………………………………..…………….… 8
I.2.3. Lignina ……………………………..……….…………...…… 10
I.2.4. Compuestos minoritarios ………………………........………. 19
I.3. Biorrefinería de materiales lignocelulósicos ……………………… 23
…………………...…...
I.3.1. Producción de pasta de celulosa ………………………..…… 24
I.3.1.1. Procesos de producción de pasta de papel ……….…..….. 26
I.3.2. Producción de bioetanol …………………..………..……...… 28
I.3.2.1. Pretratamientos del material lignocelulósico ………..…... 30
I.3.2.2. Hidrólisis y fermentación ……………………....…..……. 35
I.3.3. Productos a partir de la lignina ……………………...……….. 36

II. OBJETIVOS ………...……………………..……....……................. 39

III. MATERIALES Y MÉTODOS …………………....………....… 43
III.1. Reactivos ……………..…….….………...………....…..………..... 45
III.2. Materiales lignocelulósicos ……………………........…........…..... 46
III.2.1. Cultivos forestales: Eucalipto ….....…..........………………… 46
III.2.2. Cultivos agrícolas …………………...….....……………......... 47
III.2.2.1. Cultivos energéticos: Hierba elefante ……...……...…… 47
III.2.2.2. Residuos agrícolas: Paja de trigo ………….....………… 49
III.2.3. Pastas de eucalipto ……………………………..………...…... 50
III.2.4. Ligninas precipitadas de licores de cocción ……….............… 51
III.3. Métodos ……….…..……………………….……………............... 51
III.3.1. Preparación de las muestras lignocelulósicas ………….…….. 51
III.3.2. Composición química ………………...…..………………...... 52
III.3.2.1. Determinación del contenido en cenizas ……...……...... 53
III.3.2.2. Análisis de metales y otros elementos …….…..….……. 53
III.3.2.3. Determinación del contenido y de la composición
química de extraíbles lipofílicos ..................................... 55
III.3.2.4. Determinación del contenido en extraíbles hidrosolubles 57
III.3.2.5. Determinación del contenido en proteínas …………….. 57
III.3.2.6. Determinación del contenido en carbohidratos ………... 58
III.3.2.6.1. Determinación del contenido en holocelulosa .. 58
III.3.2.6.2. Determinación del contenido en α-celulosa ….. 58
III.3.2.7. Determinación del contenido en lignina total ……..…... 59
III.3.2.7.1. Determinación del contenido en lignina Klason 60
.
III.3.2.7.2. Determinación del contenido en lignina ácido-
soluble .............................................................. 60
III.3.3. Aislamiento de ligninas de materiales lignocelulósicos …..…. 60
 III.3.3.1. Aislamiento de lignina de maderas y fibras …….....…... 61
III.3.3.2. Aislamiento de lignina residual de pastas alcalinas …… 63
III.3.3.3. Aislamiento de lignina de licores negros …………..….. 64
III.3.4. Técnicas analíticas para el análisis de lignina .…..……….….. 65
III.3.4.1. Py-GC/MS …………………………………….…….…. 65
III.3.4.2. 2D-NMR ………………………………..……...………. 67
31
III.3.4.3. P-NMR …………………………………….……..….. 70
III.3.4.4. DFRC …………………………………………..….…... 73

III.3.4.5. SEC …………………………………………………... 75

IV. REFERENCIAS …………………………………………………... 77

V. RESULTADOS Y DISCUSIÓN …………................................. 99

Publicación I: Prinsen P., Gutiérrez A., Rencoret J., Nieto L.,
Jiménez-Barbero J., Burnet A., Petit-Conil M., Colodette J.L.,
Martínez A.T. and del Río J.C. (2012). Morphological characteristics
and composition of lipophilic extractives and lignin in Brazilian
woods from different eucalypt hybrids. Industrial Crops and
Products, 36, 572-583 …………………….……………………….… 103

Publicación II: Prinsen P. utiérrez A. and del Río J.C. (2012).

Lipophilic extractives from the cortex and pith of elephant grass
(Pennisetum purpureum Schumach.) stems. Journal of Agricultural
and Food Chemistry, 60, 6408-6417 …………………………..…….. 135
….
Publicación III: del Río J.C., Prinsen P. Rencoret . Nieto L.
iménez- arbero . Ralph . Martínez A.T. and utiérrez A. 1 .
Structural characterization of the lignin in the cortex and pith of
elephant grass (Pennisetum purpureum) stems. Journal of
Agricultural and Food Chemistry, 60, 3619-3614 …………….…….. 159
Publicación IV: del Río J.C., Prinsen, P. and Gutiérrez, A. (2013). A
comprehensive characterization of lipids in wheat straw. Journal of
Agricultural and Food Chemistry, 61, 1904-1913 …………………... 195
…………………....
 Publicación V: del Río J.C., Rencoret J., Prinsen P., Martínez A.T.,
Ralph J. and Gutiérrez A. (2012). Structural characterization of
wheat straw lignin as revealed by analytical pyrolysis, 2D-NMR, and
reductive cleavage methods. Journal of Agricultural Food and
Chemistry, 60, 5922–5935 …………………………….……………... 219
Publicación VI: Prinsen P., Rencoret J., Gutiérrez A., Liitiä T.,
Tamminen T., Martínez A.T. and del Río J.C. (2013). Modification
of the lignin structure of a eucalypt feedstock during chemical
deconstruction by kraft-, soda-AQ and soda-O2 pulping. Industrial
& Engineering Chemistry Research (enviado) ……………...……….. 257
……………………...

VI. CONCLUSIONES ……………………………………………..…. 285

VII. ……………...
 RESUMEN

En la presente Tesis se ha estudiado la composición química de diversos materiales

lignocelulósicos, que incluyen cultivos de crecimiento rápido, tanto madereros
(madera de eucalipto) como no madereros (hierba elefante), y residuos agrícolas
(paja de trigo). Se puso especial énfasis en el estudio de la lignina y los compuestos
extraíbles lipofílicos, cuya composición y estructura influyen decisivamente en los
procesos industriales de los materiales lignocelulósicos. Por otro lado, se estudió la
modificación de la estructura de la lignina de eucalipto durante diversos procesos
de deconstrucción alcalina. Esta información constituye un punto de partida
importante para optimizar procesos de biorrefinería, en particular los procesos de
cocción para la producción de pasta de celulosa y los procesos de producción de
bioetanol.

La caracterización estructural de las ligninas se llevó a cabo mediante técnicas

analíticas avanzadas (Py-GC/MS, 2D-NMR, DFRC y SEC). Tanto el contenido en
lignina de las maderas de diferentes especies híbridas de eucalipto (27-28%), como
la composición y las características estructurales de sus ligninas fueron similares,
con una relación S/G entre 2.6 y 3.1, y con un predominio de enlaces alquil-aril éter
β-O-4´, y en menor proporción enlaces carbono-carbono β-β´ en subestructuras
resinol, β-5´ en subestructuras fenilcumarano, y β-1´ en subestructuras
espirodienona. La hierba elefante y la paja de trigo presentaron un contenido en
lignina más bajo (20% y 18%, respectivamente), y la composición de sus ligninas
está más enriquecida en unidades G y H, con una relación S/G de 1.4 y 0.5,
respectivamente, y una relación H/G de 0.1 para ambas. Presentaron una
proporción de enlaces β-O-4´ parecida a la de las ligninas de madera de eucalipto,
sin embargo se caracterizaron por una proporción más alta de enlaces β-5´ en
subestructuras fenilcumarano, y más baja de enlaces β-β´ en subestructuras resinol.
También se caracterizaron por la presencia de unidades aciladas, con un
predominio de unidades S p-cumaroiladas y unidades G acetiladas, en
subestructuras eterificadas de la lignina de hierba elefante y paja de trigo,
respectivamente. Es de destacar el descubrimiento en estas ligninas del flavonoide
tricina, que está unido al polímero mediante enlaces β-O-4´ a unidades G.

El contenido de extraíbles lipofílicos en la madera de las diferentes especies

híbridas de eucalipto fue muy parecido (0.2-0.3%), con un predominio de
compuestos esteroidales, incluyendo tanto esteroles libres como conjugados (en
forma de ésteres y glicósidos). La hierba elefante y paja de trigo presentaron un

i
contenido de extraíbles lipofílicos bastante más elevado (0.9% y 2.0%,
respectivamente), con un predominio de compuestos alifáticos lineales,
principalmente ácidos y alcoholes grasos, ésteres de glicerol, ceras y β-dicetonas.

Finalmente, se estudió la modificación estructural de la lignina de eucalipto

durante la deconstrucción química mediante diferentes procesos de cocción alcalina
(kraft, sosa-AQ y sosa-O2). Para ello, se aislaron las ligninas residuales de pastas
procedentes de las diferentes cocciones alcalinas con distintos grados de
deslignificación (distintos números kappa), y junto a las ligninas precipitadas de los
correspondientes licores de cocción, se analizaron en detalle mediante diferentes
técnicas analíticas (SEC, Py-GC/MS, HSQC y 31P-NMR). La composición y las
características estructurales de las ligninas residuales fueron parecidas a las de la
lignina nativa de la madera, mientras que las ligninas precipitadas presentaron un
grado de degradación mucho mayor, en cuanto a la proporción de enlaces éter β-
O-4´ y enlaces condensados, el contenido de grupos carbonilos e hidroxilos en la
cadena lateral, el contenido de grupos hidroxilos en la parte aromática (fenólicos),
y en cuanto al peso molecular. Se concluyó que el proceso sosa-O2 presentaba una
mayor degradación de la lignina, y por tanto se podría utilizar como pretratamiento
para la deconstrucción de madera de eucalipto con vistas a la producción de
bioetanol.

La presente Tesis incluye los siguientes apartados: i) una introducción acerca de

los constituyentes de los materiales lignocelulósicos, con un enfoque en la lignina,
y su utilización en procesos industriales; ii) los objetivos que se han perseguido en
la Tesis; iii) una descripción detallada de los materiales y métodos utilizados; iv)
las correspondientes referencias citadas; v) los principales resultados obtenidos y la
discusión de los mismos, que se presentan en forma de publicaciones; y finalmente
vi) las principales conclusiones obtenidas.

ii
 ABSTRACT

In the present PhD Thesis, the chemical composition of diverse lignocellulosic

materials, including fast growing eucalypt trees (woody) and elephant grass crops
(non-woody) as well as agricultural residues (wheat straw) was studied. Special
emphasis was put on the composition and structural characteristics of the lignin and
lipophilic extractives, which play a decisive role in the industrial processes of these
materials. On the other hand, the structural modifications of the lignin from
eucalypt wood during different alkaline deconstruction processes were studied.
This information constitutes an important starting point in order to optimize
biorefinery processes, particularly in the cooking processes used for the production
of cellulose pulp or in the production of bioethanol.

The structural characterization of the lignins was assessed by an array of

advanced analytical techniques (Py-GC/MS, 2D-NMR, DFRC and SEC). Both the
content (27-28%) and the structural characteristics of the lignin in the wood from
different eucalypt hybrid species were similar, with a S/G ratio between 2.6 and 3.1
and a predominance of β-O-4´ alkyl-aryl ether linkages, and smaller amounts of the
condensed β-β´ resinol, β-5´ phenylcoumarane and β-1´ spirodienone substructures.
Elephant grass and wheat straw presented lower lignin contents than eucalypt
woods (20% and 18%, respectively), but their lignins were enriched in G- and H-
units, with a S/G ratio of 1.4 and 0.5, respectively, and a H/G ratio of 0.1 in both
lignins. The abundance of β-O-4´ ether linkages was similar to the eucalypt lignins,
however, the abundance of β-5´ phenylcoumarane substructures was higher while
the amounts of β-β´ resinol substructures were lower. Furthermore, elephant grass
and wheat straw lignins were found to be partially acylated at the γ-carbon, with a
predominance of p-coumaroylated S-units and acetylated G-units in etherified
substructures of elephant grass and wheat straw lignin, respectively. A major
finding was the discovery of the flavone tricin, which is apparently incorporated
into the lignin of wheat straw (and in other grasses) by a β-O-4´ ether linkage to a
G-unit, which may have a major impact on the definition of lignin in grasses in
terms from which monomers it may be constituted.

The content of lipophilic extractives in the woods from the different eucalypt
hybrids was very similar (0.2-0.3%), with a predominance of steroid compounds,
including free and conjugated (as esters and glycosides) sterols. Elephant grass and
wheat straw presented higher contents of lipophilic extractives (0.9% and 2.0%,

iii
respectively), with a predominance of aliphatic compounds, mainly fatty acids and
alcohols, glycerol esters, waxes and β-ketones.

Finally, we studied the structural modifications of the lignin from eucalypt wood
during the chemical deconstruction by different alkaline cooking processes (kraft,
soda-AQ and soda-O2). For this purpose, the residual lignins isolated from eucalypt
pulps produced by these alkaline cooking processes at different delignification
degrees (different kappa numbers), and the lignins recovered from the
corresponding cooking liquors, were analyzed by several analytical techniques
(SEC, Py-GC/MS, HSQC and 31P-NMR). The residual lignins from eucalypt pulps
and the corresponding native lignin showed similar composition and structural
characteristics, while the lignins precipitated from the cooking liquors exhibited
much higher degradation, with regards to the proportion of β-O-4´ ethers and
condensed linkages, the content of hydroxyl and carbonyl groups in the side chain,
the hydroxyl group content in the aromatic part (phenol group content) and the
molecular weight distributions. It was concluded that the soda-O2 cooking showed
a higher lignin degradation degree, and may be a promising alkaline pretreatment
for the chemical deconstruction of eucalypt wood in order to obtain higher
hydrolysis and fermentation efficiencies for the production of bioethanol.

The present Thesis includes the following sections: i) an introduction to the

constituents of lignocellulosic materials, with a focus on the lignin and lipophilic
extractives, and the utilization of these materials in industrial processes; ii) the
main objectives of the Thesis; iii) a detailed description of the materials and
methods used; iv) the corresponding references cited; v) the main results obtained
and their discussion, which are presented as scientific publications; and finally vi)
the main conclusions drawn from the results.

iv
I. INTRODUCCIÓN
 I. INTRODUCCIÓN

I.1. Cultivos lignocelulósicos de interés industrial

La biomasa vegetal representa la principal fuente de materiales renovables de la

Tierra, y tiene por tanto un gran potencial como materia prima para la producción
de celulosa, bioenergía y otros productos de interés industrial. La biomasa vegetal,
y en particular los materiales lignocelulósicos, están disponibles en grandes
cantidades y a bajo coste, como cultivos o residuos agroforestales, y podrían ser
una fuente barata de biocombustibles y bioproductos como alternativa a los
combustibles fósiles (Ragauskas et al., 2006; Sarkar et al., 2012; Somerville et al.,
2010). Actualmente existe una gran tendencia a reducir el uso de combustibles
fósiles para disminuir tanto las emisiones de gases de efecto invernadero, como
nuestra dependencia energética, y el uso de la biomasa vegetal puede tener un papel
importante en este sentido. Esta necesidad de un mayor y mejor aprovechamiento
de la biomasa vegetal está fuertemente respaldado por la Unión Europea. En este
sentido, se están considerando estrategias agrícolas alternativas centradas en la
producción de cultivos agroforestales para la obtención de productos no
alimenticios, tales como papel, productos textiles, biomateriales, biocombustibles y
una amplia gama de productos químicos, como alternativa al petróleo, en el
contexto de las llamadas biorrefinerías. La principal fuente de materiales
lignocelulósicos la constituyen los cultivos forestales y agrícolas, aunque los
residuos generados a partir de estos cultivos también pueden ser una materia prima
de interés.

I.1.1. Cultivos forestales

La principal fuente de celulosa la constituyen los cultivos forestales madereros

que se utilizan principalmente para la producción de pasta y papel. La amplia
concentración de madera en zonas de fácil acceso, el elevado contenido en fibras,
el coste de manejo, transporte y facilidad de almacenamiento, así como la
estabilidad de la materia prima y su comportamiento durante el proceso de
obtención de celulosa, han apoyado el uso de la misma en la industria papelera. Las
coníferas constituyen el primer cultivo forestal a escala mundial para la obtención
de pasta de papel, aunque también existe un importante mercado de pastas de
frondosas. En términos económicos generales, las coníferas son más valiosas que
las frondosas, ya que sus troncos son más largos y rectos, su madera es uniforme,
ligera y blanda, presentan fibras largas (3 a 5 mm) que ofrecen una alta resistencia
mecánica, y son más adecuadas para la mayoría de las calidades papeleras (García
Hortal, 2007). Las principales coníferas usadas para la fabricación de pasta de

3
I. INTRODUCCIÓN

papel son la Picea y el pino. La madera de frondosas, por otro lado, es una madera
más dura, con fibras cortas (entre 0.75 y 2 mm). Las características físicas de las
fibras influyen tanto en el proceso de producción de papel, como en la calidad del
producto.

Para la obtención de productos químicos a partir de celulosa, por ejemplo el

bioetanol, las características físicas no influyen en el producto final, pero sí pueden
influir en la resistencia a la deconstrucción química del matriz lignocelulósico. Los
cultivos madereros de crecimiento rápido, tales como el eucalipto, chopo, abedul y
álamo, son atractivos para la obtención de bioetanol, y también una amplia gama de
otros productos químicos derivados, tanto de los carbohidratos (celulosa y
hemicelulosas) como de la lignina.

I.1.2. Cultivos agrícolas

Los cultivos agrícolas constituyen una excelente materia prima alternativa a los
cultivos forestales para la obtención de celulosa, principalmente por su gran
abundancia y su coste relativamente bajo, especialmente en países con baja
disponibilidad de madera. La obtención de bioetanol de primera generación a partir
de caña de azúcar y grano de cereales, influye en el mercado de materias primas y
compite por el uso de suelo. Determinados cultivos que se caracterizan por una alta
velocidad de crecimiento (cultivos energéticos) y/o que presentan ciclos de
crecimiento más cortos, alcanzando la madurez más rápidamente que las especies
madereras, como pasto varilla (Panicum) y hierba elefante (Miscanthus y
Pennisetum), son aptos para la obtención de bioetanol de segunda generación.
Además, se valorizan aquellas especies que muestran una gran robustez en cuanto a
sus necesidades hídricas y de fertilizantes, ya que su cultivo en suelos marginales
evita la competencia del suelo destinado a la obtención de productos alimenticios.
Por otro lado, los cultivos agrícolas destinados a la obtención de productos
alimenticios, generan una cantidad considerable de residuos, como el rastrojo de
maíz, la paja de trigo, de arroz y de otros cereales, que constituyen una fuente
importante de material lignocelulósico (Kim y Dale, 2004). Algunos cultivos
herbáceos, como lino, cáñamo, kenaf, yute, bambú, abacá y sisal son una
alternativa a los cultivos forestales para la producción de papel, y se utilizan
también para la producción de papeles especiales. Así, el uso de estos cultivos para
la producción de pasta de celulosa ha ido aumentando, especialmente en los países
en vías de desarrollo, como India, China y algunos países latinoamericanos.

4
 I. INTRODUCCIÓN

I.2. Estructura y composición química de los materiales

lignocelulósicos

Los materiales lignocelulósicos están constituidos mayoritariamente por tres

polímeros estructurales, la celulosa y las hemicelulosas (carbohidratos) y el
polímero aromático lignina, presentes en la pared celular vegetal (Figura 1).

a
Célula vegetal S3
S2
Pared S1
celular
P
LM

b Lignina

Hemicelulosas

Celulosa

Enlaces de Celulosa Lignina Hemicelulosas

hidrógeno

Figura 1. (a) Visualización esquemática de la pared celular vegetal, constituida por la

lámina media (LM), una pared primaria (P) y una pared secundaria (S), que puede
diferenciarse en la capa externa (S1), media (S2) e interna (S3) (adaptado de Côté, 1967), y
(b) representación esquemática de los principales constituyentes en la pared celular de una
angiosperma no leñosa (adaptado de Bidlack et al. 2008).

5
I. INTRODUCCIÓN

Las propiedades extraordinarias de las fibras lignocelulósicas y la morfología y

funcionalidad de la pared celular son el resultado del alto grado de organización,
estructuración y coordinación de estos polímeros. Una parte minoritaria de los
materiales lignocelulósicos consiste en compuestos de bajo peso molecular solubles
en agua o en solventes orgánicos, así como pequeños contenidos de proteínas y
sales minerales (Fengel y Wegener, 1984; Sjöström, 1993).

I.2.1. Celulosa

La celulosa es el polímero más abundante en la Tierra (Klemm et al., 2005). En

maderas, la celulosa representa cerca del 50% del material lignocelulósico seco
(Aitken et al., 1988), mientras en materiales lignocelulósicos no madereros el
contenido presenta una mayor variación (20-50%) (Knauf y Moniruzzaman, 2004).
Algunas excepciones son algodón (> 95%), abacá (60%), remolacha (18%), etc. La
celulosa está constituida por cadenas lineales de un polímero de unidades de β-D-
glucopiranosa, unidas entre sí por enlaces glicosídicos β-(1→4) (Figura 2).
Durante la formación de este enlace, la posición β de los grupos hidroxilos en el
C1 necesita un giro de 180º de la siguiente molécula de glucosa alrededor del eje
C1-C4 del anillo piranósico, por lo que la unidad que se repite es la celobiosa
(disacárido), con una longitud de 1.03 nm (Sjöström, 1993). En la pared primaria,
la celulosa se presenta en cadenas cortas (grado de polimerización (n) = 500-2000),
y en la pared secundaria, el grado de polimerización es mayor (n ~ 10000).

OH OH
1 4
OH O OH
HO β O HO O O

O HO β O O
O O HO
OH 1 4 OH
OH OH

Celobiosa

Figura 2. Estructura de una cadena de celulosa.

La estructuración de las cadenas de celulosa empieza a nivel de biosíntesis,

mediante interacciones inter- e intramoleculares no covalentes (Gómez et al.,
2008), facilitado por complejos macromoleculares proteicos (rosetas) en la pared
celular. Cuando el crecimiento de los oligosacáridos en las rosetas ocurre
simultáneamente, las cadenas de celulosa se pueden orientar de forma paralela, lo

6
 I. INTRODUCCIÓN

que permite la manifestación de interacciones entre las cadenas (puentes de

hidrógeno). Este alto grado de estructuración da lugar a la formación de las
microfibrillas, con 30-36 cadenas de celulosa y con diámetro 2-10 nm, que a su vez
se ordenan en las macrofibrillas, y finalmente en la fibra (Figura 3). La ausencia de
estas interacciones produce zonas amorfas (con una estructura semicristalina). Se
cree que en estas zonas amorfas se asocian las hemicelulosas.

Las microfibrillas se pueden depositar en estructuras cristalinas altamente

densas. Se cree que la densidad está regulada por la abundancia de dos fases
cristalinas, Iα y Iβ, que se diferencian por diferentes modos de empaquetamiento
(Ding y Himmel, 2006; Somerville et al., 2004). La cristalinidad protege la celulosa
contra la degradación (hidrólisis de los enlaces β-1,4 entre las unidades). La
abundancia de las zonas amorfas, que son más susceptibles a la hidrólisis, juega un
papel importante en la digestibilidad de la pared celular. En la pared secundaria de
fibras madereras, parte de la celulosa se encuentra altamente orientada para dar
estructura a las células del sistema vascular. Sin embargo, en la pared secundaria de
muchas plantas herbáceas, parte de la celulosa no forma fibrillas sino laminillas
cristalinas, como en la paja de trigo (Yu et al., 2005).

Fibra de celulosa a b

Macrofibrilla

Microfibrilla

Cadenas
de celulosa

Figura 3. (a) Estructuración de cadenas de celulosa en micro- y macrofibrillas, y (b)

interacciones mediante puentes de hidrógeno en y entre las cadenas.

7
I. INTRODUCCIÓN

I.2.2. Hemicelulosas

A diferencia de la celulosa, cuya estructura es la misma en los diferentes

materiales lignocelulósicos, en las hemicelulosas existe una gran variación entre las
diferentes especies. Los monosacáridos que las constituyen (Figura 4a), incluyen
pentosas (D-xilosa y L-arabinosa), hexosas (D-glucosa, D-galactosa, L-galactosa, D-
manosa, L-ramnosa y L-fucosa), y ácidos urónicos (ácido D-glucurónico y ácido D-
galacturónico). Generalmente, las unidades de la cadena principal están unidas
entre sí por enlaces β-(1→4). En algunas especies, se alternan los enlaces β-(1→4)
con β-(1→3). De la cadena principal pueden partir diversas ramificaciones
(Scheller y Ulvskov, 2010; Sjöström, 1993). Además, una fracción de la cadena
principal puede estar acetilada. Las hemicelulosas son más solubles y más fáciles
de degradar que la celulosa, ya que tienen un grado de polimerización entre 200 y
300, y una estructura ramificada y amorfa (Sjöström y Westermark, 1999).

La fracción hemicelulósica representa 25-30% de la madera en especies de

coníferas, y 20-43% en la madera de frondosas (Aitken et al., 1988). Las
hemicelulosas se encuentran a lo largo de la pared celular, desde la lámina media
hasta la capa S3 de la pared secundaria, siendo más abundantes en las capas S1 y
S3 y hallándose en menor proporción en la capa S2. Actúan junto con la lignina
como matriz soporte para las microfibrillas de celulosa para proporcionar rigidez a
la pared celular. Interactúan con la celulosa por interacciones no-covalentes. La
unión directa mediante enlaces covalentes sólo ocurre vía pectinas (Singh et al.,
2009). Las maderas de coníferas suelen presentar una mayor cantidad de
hexosanos, como la manosa y galactosa, siendo predominantes los
galactoglucomananos y los glucomananos (Figura 4b) (García Hortal, 2007,
Sjöström y Westermark, 1999), aunque también contienen los
arabinoglucuronoxilanos. En las frondosas predominan los pentosanos,
principalmente xilosa, como en los glucuronoxilanos (Fengel, 1989, Sjöström,
1993, Shimizu, 2001), aunque también se observan glucomananos (García Hortal,
2007). En las plantas no madereras, los xilanos son las hemicelulosas
predominantes. Su composición varía mucho según la especie. Una parte de las
unidades de xilosa puede estar monosustituida con α-L-Araf en posición 2 o 3, y
otra parte está disustituida en las mismas posiciones (Figura 4b). En los xilanos de
muchas especies monocotiledóneas, como en las cereales, la palma, el abacá, etc.,
el átomo C5 de residuos de α-L-Araf puede estar acilado con ácido ferúlico,
dehidrodiferulato y ácido p-cumárico (Hatfield et al., 1999; Scheller y Ulvskov,
2010; Wende, 1997).

8
 I. INTRODUCCIÓN

a
PENTOSAS HEXOSAS ÁCIDOS HEXURÓNICOS
COOH
O O O

D-Xylp β-D-Xilosa D-Glcp β-D-Glucosa D-GlcAp Ácido β-D-Glucurónico

COOH
O O O
MeO

4-O-Me-D-GlcAp
L-Arap β-D-Arabinopiranosa D-Manp β-D-Manosa Ácido α-D-4-O-Metilglucurónico

O COOH
O O

L-Araf β-D-Arabinofuranosa D-Galp α-D-Galactosa D-GalAp Ácido α-D-Galacturónico

DESOXI-HEXOSAS

O O

L-Rhap α-L-Ramnosa L-Fucp α-L-Fucosa

b
O O O O O O O O O O O O O O O O O O

3 2 COOH
COOH
COOH
COOH

COOH

Ac
O
O

O
O

5 Fer: ferulato
MeO

Fer
MeO

Fer Ac: acetato

(Arabino)glucuronoxilanos (Glucurono)arabinonoxilanos

O O O O O O O O O O O O O O O O

Galactomananos O Galactoglucomananos
Ac Ac

O O O O
O 1 O O O
6
O O O O O O O O
4 1

X X F G X X L G
Xiloglucanos

Figura 4. (a) Monosacáridos presentes en las hemicelulosas (adaptado de Fengel y

Wegener, 1984). (b) Tipos y variaciones estructurales simplificadas de las principales
hemicelulosas en diversos materiales lignocelulósicos (adaptado de Scheller y Ulvskov,
2010).

9
I. INTRODUCCIÓN

I.2.3. Lignina

La lignina es un polímero aromático compuesto por unidades de 4-fenilpropano.

Tiene una estructura macromolecular heterogénea y puede contener tres tipos de
unidades aromáticas: unidades p-hidroxifenilo (H), guayacilo (G) y siringilo (S),
que no tienen ningún grupo metoxilo (H) o bien tienen uno (G) o dos (S) grupos
metoxilo en las posiciones C3 y C5 de la unidad aromática (Figura 5).

γ OH γ OH γ OH

β β β
α α α

1 1 1
6 2 6 2
5
4
3 5
4
3
6
5
4
2
3 Monolignoles
OMe MeO OMe
OH
MH MG OH MS OH

H G S Unidades
OMe MeO OMe
O O O

Figura 5. Las unidades p-hidroxifenilo (H), guayacilo (G) y siringilo (S) en la lignina se
forman a partir del acoplamiento de los monolignoles MH, MG y MS (adaptado de Ralph
2004b).

La composición de la lignina depende de su origen botánico. Así, la lignina de

coníferas presenta principalmente unidades G, las frondosas unidades G y S en
diversas proporciones, y la lignina de plantas herbáceas contiene unidades H, G y S
en diversas proporciones. La lignina representa entre el 25 y 33% de la biomasa
seca en maderas coníferas, y entre el 18 y 34% en maderas frondosas (Aitken et al.,
1988). Es el segundo polímero más abundante en la Tierra, detrás de la celulosa. La
lignina es de vital importancia para el organismo vegetal; proporciona rigidez y
resistencia a los tallos, y desempeña funciones en el transporte de agua, nutrientes y
metabolitos en el sistema vascular. También juega un papel importante en el
sistema de defensa de la planta frente a patógenos (Hückelhoven, 2007; Sarkanen y
Ludwig, 1971). Aunque la concentración de lignina en la lámina media y en las
esquinas de la pared primaria es más alta que en la pared secundaria, la mayor
fracción de lignina se encuentra en la pared secundaria, debido al volumen que
ocupa la pared secundaria (Saka y Goring, 1985; Terashima et al., 1993). La

10
 I. INTRODUCCIÓN

lignina es muy abundante en la lámina media, donde se puede considerar como el

cemento entre las paredes celulares. Las interacciones entre la lignina, y la celulosa
y las hemicelulosas son tanto de carácter no-covalente como covalente (Gómez et
al., 2008; Singh et al., 2009).

Biosíntesis de la lignina
Las unidades H, G y S se forman cuando sus respectivos monolignoles
(precursores), los alcoholes 4-hidroxicinamílicos (alcohol p-cumarílico, MH), 4-
hidroxi-3-metoxicinamílico (alcohol coniferílico, MG) y 4-hidroxi-3,5-
dimetoxicinamílico (alcohol sinapílico, MS), se incorporan mediante acoplamiento
en la estructura de lignina en crecimiento. La biosíntesis de los monolignoles
empieza a partir de fenilalanina y está regulada por diferentes genes, cuya
expresión determina la actividad de las enzimas involucradas (Boerjan et al., 2003).
La lignificación empieza por una deshidrogenación de los monolignoles, que da
lugar a la formación de radicales libres de tipo fenoxilo y radicales de carbono,
estabilizados por resonancia (Figura 6a), que se acoplan entre sí mediante diversos
tipos de enlaces.

a Deshidrogenación
OH OH OH OH OH OH

Lacasa/O2
Peroxidasa/H2O2

OMe - (e-+ H+) OMe OMe OMe OMe OMe

OH O O O O O

MG

b Deshidrodimerización
OH

OMe
HO OH
OH
HO
O O
HO OMe
OMe
O

OMe O

OH
HO
β-O-4´ OH OMe β-5´ OMe β-β´

Grupos terminales cinamílicos Pinoresinol

Figura 6. (a) La deshidrogenación enzimática del alcohol coniferílico (y los otros

monolignoles) da lugar a la formación de radicales fenoxilo y sus estructuras resonantes
(adaptado de Adler, 1977). (b) Deshidrodimerización por acoplamiento radicalario
oxidativo a partir del alcohol coniferílico (adaptado de Ralph et al., 2004b).

11
I. INTRODUCCIÓN

La deshidrogenación ocurre mediante la transferencia de electrones, catalizada

por peroxidasas o lacasas en presencia de H2O2 u O2, respectivamente (Adler,
1977; Baucher et al., 1998; Boerjan et al., 2003; Freudenberg y Neish, 1968; Ralph
et al., 2004b). Este mecanismo de acoplamiento da lugar a la formación de diversos
tipos de enlaces, encontrados en la lignina de diversas especies (Figura 7), que
pueden ser de tipo éter (β-O-4´, α-O-4´ y 4-O-5´) y de tipo carbono-carbono (los
denominados enlaces condensados β-β´, β-1´, β-5´ y 5-5´).

RO γ HO
6´ 1´ γ
5´
β β
HO 3´ 2´ O
α 4´
O 1´ α OH 2 1
6´ 2´ 1 3 6 2
5´ 3´ 6 5 3
4´ 5 4 4
1
6 2 1 O
6 2
5 3
4 5 3
4

O O
A B C

O HO γ
3´
2´ 4´ 1´
6´ 2´
1´ 5´ β
O 6´ HO 1´ 5´ 3´
HO
γ α´ α 2´
3´ 4´
6´ γ β
β β´ 4´
1 5´
α
O
α γ´ 6 2 O
6 1 5 3 6´ 1´
5 O 4 5´ 2´
4 2 4´ 3´
3
O O
D E O B

O
1 6 6 1
4 5
2 2 6
5 5 3
3 4 4 3
2 1

O O α O O
α´´
β α´ Ar
β´´ γ β´
γ´´
OH
6 1 1 γ´
OH
5 2 HO 6 2
4 3 5 3
4
O G H
O

Figura 7. Enlaces éter (A-C) y carbono-carbono (D-H) más abundantes encontrados en

subestructuras de (A) alquil-aril éter β-O-4´, (B) α-O-4´, (C) aril-aril éter 4-O-5´ (D)
resinol β-β´/α-O-γ´/γ-O-α´, (E) alquilo-arilo β-1´, (F) fenilcumarano β-5´/α-O-4´, (G)
dibenzodioxocina 5-5´/α´´-O-4/β´´-O-4´, y (H) espirodienona β-1´/α-O-α´.

12
 I. INTRODUCCIÓN

Durante la dimerización se producen casi exclusivamente los enlaces β-O-4´, β-

β´ y β-5, según Lu y Ralph (2010b), y la proporción de estos enlaces depende de la
proporción de los monolignoles (MH:MG:MS) que se acoplan entre sí (Figura 6b).
Los enlaces 4-O-5´, β-1´ y 5-5´ se formarían entre unidades de diferentes
oligómeros en crecimiento, creando puntos de entrecruzamiento (Figura 8).

OH HO OH HO
OH HO
HO Ar Ar OH HO Ar
β
O HO Ar O O
O
O
Ar
5
1
G G
MeO O OMe 4 MeO OMe
OMe
OH OH OH
OH G-5-5-G
G-(4-O-5)-G
S-(4-O-5)-G Unidad G / S OH HO

Ar OH HO Ar
O O
HO
HO
β Radical de
β HO Ar G G
. O
Radical de G-5-5-G MeO
Radical de . OMe
G/S HO .
MG / M S O 1
β
O OH
MeO
OMe OMe
. Acoplamiento
O
O
O OH HO

Acoplamiento Ar OH HO Ar
MeO
Radical de O O
α OH O Ar
monolignol
β

HO 1
β
MG / M S G G
MeO OMe

OMe O OH
α
O HO HO β
OH OMe

OH HO
Ar
O O Ar O

G G
MeO
α
O Ar MeO O O OMe
β

1 OH OH
HO
MeO
OMe
O O

Espirodienona Dibenzodioxocina

Figura 8. La formación de enlaces éter 4-O-5´, β-1´ y 5-5´ entre unidades de diferentes
oligómeros de lignina da lugar a la formación de entrecruzamientos. Adaptado de Lu y
Ralph (2010b).

La polimerización es de tipo ´endwise´, es decir, consiste principalmente en el

acoplamiento continuo de monolignoles a los oligómeros en crecimiento. La

13
I. INTRODUCCIÓN

abundancia de grupos terminales es muy baja en ligninas nativas, ya que sólo una
proporción muy pequeña de dímeros se acoplan entre sí (Adler, 1977; Chen et al.,
1998; Lu y Ralph, 2010b). Los grupos terminales (Figura 6b) pueden ser
estructuras cinamílicas (a partir de la dimerización tipo β-O-4´ y β-5´) o estructuras
resinol (pinoresinol y siringaresinol, a partir de la dimerización tipo β-β´). Durante
la polimerización tipo ´end-wise´, a medida que crece el polímero, la proporción de
los enlaces β-O-4´ aumenta. Además, la proporción aumenta más todavía cuando
los monolignoles se acoplan preferiblemente a unidades S, como se muestra en la
Figura 9, y es una de los razones por las que la lignina enriquecida en unidades S,
se asocia con una proporción más alta de enlaces β-O-4´.

S G
MeO OMe OMe
OH OH
OH OH OH OH

+ +
MG MS MG MS
OMe MeO OMe OMe MeO OMe
OH OH OH OH

MeO MeO
HO HO
HO HO MeO
S S
HO
O
HO
HO
G HO
G
O
O O
OMe OMe OMe
OMe
G S
OMe
G S
MeO OMe
OMe MeO OMe
OH OH
OH OH
β-O-4´ β-O-4´ β-O-4´ β-O-4´
+ +

G G
HO OMe
HO OMe
O O

G MeO S
HO OMe HO OMe
β-5´ β-5´

Figura 9. Acoplamiento de los alcoholes coniferílico (MG) y sinapílico (MS) a unidades

fenólicas S, dando lugar a la formación de enlaces β-O-4´, y a unidades fenólicas G, dando
lugar a la formación de enlaces β-O-4´ y β-5´ en la estructura de lignina en crecimiento.

14
 I. INTRODUCCIÓN

La proporción relativa con que se incorporan los monolignoles durante la

lignificación puede alterarse en los diferentes estadíos de la vida de una planta
(Freudenberg y Lehmann, 1963; Rencoret et al., 2011). Además, la estructura de la
lignina puede variar en función del lugar de la pared celular donde se sintetice
(Christierini et al., 2005; Fergus y Goering, 1970; Fukushima y Terashima, 1991),
del tipo de célula (Fergus y Goring, 1970; Hardell et al., 1980a, 1980b), y de la
parte morfológica de la planta (Bland, 1966). Es de resaltar que el proceso de
polimerización está siendo motivo de controversia entre científicos que defienden
que la formación del polímero de lignina es un proceso controlado por proteínas
(denominadas proteínas dirigentes) (Davin y Lewis, 2000; Lewis, 1999), y otros
que defienden que se trata de un proceso flexible, bajo un control exclusivamente
químico (Ralph et al., 2004b, 2007, 2008b; Sederoff et al., 1999).

Durante la última década, varios estudios realizados en plantas transgénicas han

revelado que otros monómeros también pueden acoplarse durante la lignificación,
como los alcoholes dihidroxicinamílicos (5-hidroxiconiferol) y los aldehídos
cinamílicos (coniferaldehído) (Morreel et al., 2004; Ralph et al., 1997; Sederoff et
al., 1999). Por otro lado, se ha encontrado que los monolignoles acilados (con
acetatos, p-cumaratos y/o p-hidroxibenzoatos esterificados en posición γ) también
actúan como precursores en la biosíntesis de lignina en diversas plantas
angiospermas (Figura 10), como abacá, sisal, kenaf y palma (del Río et al., 2004,
2007, 2008; Lu y Ralph, 2002, 2008; Martínez et al., 2008; Ralph et al, 1994a;
Rencoret et al., 2013), y también en cultivos transgénicos de álamo (Meyermans et
al., 2000; Stewart et al., 2009). En la lignina de estas especies, los grupos de
acilación están presentes mayoritariamente en unidades S, es decir, el alcohol
sinapílico acilado se acopla más con respecto al alcohol coniferílico acilado. La
acilación de los monolignoles es una línea de investigación actual, y se cree que la
acilación de monolignoles ocurre a través de transferasas (Hatfield et al., 2009).

O O O
γ γ γ

β β β
α O α O α O

1 1 1
6 2 6 2 6 2
5 4 3 5 4 3 5 4 3
(MeO ) (OMe) (MeO ) (OMe) (MeO ) (OMe)
OH OH OH

(I) (II) (III)

Figura 10. Estructura de los monolignoles acilados con grupos acetatos (I), p-cumaratos
(II) y p-hidroxibenzoatos (III).

15
I. INTRODUCCIÓN

En los últimos años se está explorando en qué modo el contenido y la estructura

de la lignina se puede alterar mediante la mutación y regulación de expresión de los
genes que intervienen en la biosíntesis para ´diseñar´ cultivos con un contenido
reducido en lignina y/o con una estructura más fácil de degradar, sin afectar mucho
a las funciones vitales de la planta (Chen y Dixon, 2007; Hu et al., 1999; Huntley et
al., 2003; Vanholme et al., 2008).

Complejos lignina-carbohidrato
Aparte de interacciones no covalentes, la celulosa, las hemicelulosas y la lignina,
también se asocian entre sí mediante enlaces covalentes, presentes en complejos
macromoleculares de lignina-carbohidrato, denominados ´LCCs (lignin-
carbohydrate complexes)´´. Consecuentemente, los materiales lignocelulósicos no
se pueden tratar como un conjunto de fracciones separadas, sino como una matriz
lignocelulósica que en muchos casos dificulta su aprovechamiento industrial.
Todos los métodos de extracción de lignina conllevan inevitablemente un
contenido residual de carbohidratos. Los enlaces encontrados en LCCs de maderas,
son de tipo éter- y éster bencílico, éter glicosídico y acetal (Figura 11).

OH

OH
OH
O
O

O OH O
O

AcO O
( MeO ) ( OMe )
( MeO ) ( OMe ) O

1 3

OH OH
O O OH

O MeO O O O

HO O
OH OH
( MeO ) ( OMe ) O ( MeO ) ( OMe ) O

O O
2 4

Figura 11. Enlaces entre lignina y carbohidrato en LCCs de maderas: (1) éter bencílico,
(2) éster bencílico, (3) éter glicosídico y (4) acetal (adaptado de Watanabe, 2003).

Para entender mejor el origen de la formación de estos enlaces entre la lignina y

los carbohidratos en la pared celular, Barakat et al. (2007) sintetizaron LCCs

16
 I. INTRODUCCIÓN

mediante polimerización del alcohol coniferílico (MG) y glucuronoarabinoxilanos

(GAX) con peroxidasas, y estudiaron su comportamiento en experimentos de
adsorción (Figura 12). Los autores sugirieron que la formación de estos enlaces
está controlada por interacciones no covalentes en dominios hidrofóbicos, que
limitan la presencia de agua para favorecer el acoplamiento de GAX a los
compuestos intermediarios de metiluro de quinona de oligómeros de lignina en
crecimiento.

1 2 3
GAX Peroxidasa Oligómeros

Polímeros LCC

1: Difusión de monolignoles en la pared celular

MH/G/S
2: Interacciones no covalentes en dominios hidrofóbicos
3: Reducción de la concentración local de agua
aumentando la adición de GAX a intermediarios de
Dominio hidrófobo metiluro de quinona
OH OH OH OH OH
I II III IV

Peroxidasa

R1 R2 H2O2 R1 R2 R1 R2 R1 R2 R1 R2
OH O O O O

Monolignol MH / MG / MS
Lignina Acoplamiento Lignina
HO HO

+ O O

R1 R2 GAX H2O R1 R2
O- OH

β-O-4´ metiluro de quinona

Lignina Lignina
GAXO
HO HO HO

O HO
OH O O

R1 R2 R1 R2
OH OH

LCC β-O-4´

Figura 12. Esquema de biosíntesis de LCCs, iniciado con el acoplamiento de los

monolignoles, seguido por formación de enlaces con el residuo arabinofuranosilo (Araf) de
GAX, en competición con la adición de H2O (adaptado de Barakat et al., 2007).

17
I. INTRODUCCIÓN

Al contrario que en las plantas madereras, la unión de lignina y carbohidratos en

los LCCs de las plantas herbáceas no consiste principalmente en enlaces directos
entre ambos, sino en ´puentes´ de ácidos hidroxicinámicos (Himmelsbach, 1993;
Ishii, 1997; Lapierre et al., 1989). Los ácidos hidroxicinámicos, ácido ferúlico y p-
cumárico, contienen un grupo carboxílico en la parte alifática y un grupo hidroxilo
en la parte aromática (grupo fenólico libre). Esto implica que pueden unirse con la
lignina y con los carbohidratos mediante enlaces éster y éter, que se pueden
cuantificar por rotura en medio alcalino y por metilación, respectivamente (Wallace
et al., 1995). Sin embargo, después de varias extracciones sucesivas en medio
alcalino y ácido, una fracción residual de la lignina de plantas herbáceas se
mantiene unida a los carbohidratos. Los oligosacáridos feruloilados pueden
acoplarse mediante peroxidasas y H2O2, formando dehidrodiferulatos, que crean
puntos de entrecruzamiento entre los carbohidratos (Ralph et al., 1994b). Se
descubrió que tanto ácido ferúlico, ferulatos como dehidrodiferulatos, se pueden
acoplar entre ellos y con los monolignoles formando complejos de lignina-
hidroxicinamatos-carbohidratos (Hatfield et al., 1999; Ralph y Helm, 1993, Ralph
et al., 1992, 1994b, 2008a; Zhang et al., 2009). Sin embargo, los mecanismos de
formación de los enlaces entre la lignina y los ferulatos todavía están sujetos a
discusión (Lam et al., 2001; Lu y Ralph, 2010b; Scalbert et al., 1986). La alta
variedad de enlaces entre lignina y ferulatos sólo podría producirse por un
mecanismo activo de acoplamiento radicalario (Ralph et al. 2008a), dando lugar a
un complejo con diversos posibles enlaces entre ambos, tales como enlaces éter β-
O-4´ (Figura 13), pero también enlaces carbono-carbono (β-5´, 8-5´, 5-5´ y 8-β´).

1
3
1
7 8 5

γ 6
1
β 4
α

1
4

Figura 13. Unión entre la lignina y los carbohidratos mediante ferulatos, esterificados con
el C5 de un residuo de arabinofuranosa de un arabinoxilano, y unidos con la lignina
mediante un enlace éter.

18
 I. INTRODUCCIÓN

Por este motivo, ningún método de extracción convencional es capaz de extraer

todos los ferulatos y separarlos por completo (Grabber et al., 1995; Ishii, 1997;
Ralph et al., 1992, 1995), por lo cual el grado de entrecruzamiento en la pared
celular de plantas herbáceas ha sido subestimado en muchos estudios anteriores.
Hatfield et al. (1999) postularon que los ferulatos podrían actuar como puntos de
nucleación para la biosíntesis de lignina.

I.2.4. Compuestos minoritarios

Los compuestos extraíbles solubles en agua o en disolventes orgánicos

representan una pequeña fracción de la pared celular vegetal. No influyen en la
estructura morfológica de la pared celular, pero desempeñan funciones fisiológicas
de vital importancia para la célula vegetal.

Compuestos extraíbles
Los compuestos extraíbles se clasifican según la solubilidad (Fengel y Wegener,
1984; García Hortal, 2007; Hillis, 1962; Rowe, 1989; Sjöström, 1981) en
compuestos extraíbles en disolventes apolares, que incluyen los extraíbles
lipofílicos (lípidos), y compuestos extraíbles en disolventes polares (extraíbles
hidrofílicos). Son compuestos de bajo peso molecular. La presencia de
determinados compuestos extraíbles en los materiales lignocelulósicos, a pesar de
su bajo contenido, puede llegar a tener un gran impacto durante el procesamiento
industrial de dichos materiales, como en la producción de pasta de papel.

Compuestos polares
Los extraíbles polares incluyen diferentes compuestos fenólicos libres (Figura
14), tales como precursores de lignina y compuestos relacionados con ellos (ácidos
y aldehídos p-hidroxicinámicos, lignanos y ácido p-hidroxibenzoico, vainíllico y
siríngico), aldehídos y cetonas aromáticas (p-hidroxibenzaldehído, vainillina,
siringaldehído y propioguayacona), taninos hidrolizables (ésteres de ácido gálico y
sus dímeros), flavonoides, etc. La presencia de estos compuestos durante la cocción
en la producción de pasta de papel, puede incrementar el consumo de reactivos.
Los taninos, cuando están presentes en cantidades importantes, forman complejos
coloreados con cationes metálicos afectando al color de las pastas de papel y a su
blanqueabilidad (García Hortal, 2007).

19
I. INTRODUCCIÓN

HO O

HO O O H O HO O
(I) (II) (III) (IV) (V) (VI)
O

O O O O O HO OH
OH OH OH OH OH O

Figura 14. Estructuras de compuestos representativos de los compuestos extraíbles

polares: ácido siríngico (I); ácido p-hidroxibenzoico (II); vainillina (III); acetosiringona
(IV); ácido gálico (V); 2-fenilbenzopirona (VI).

Compuestos lipofílicos
Los extraíbles lipofílicos juegan un papel importante en la defensa contra
patógenos (Sarkanen y Ludwig 1971). Ejemplos de compuestos extraíbles alifáticos
lineales y esteroidales se muestran en las Figuras 15 y 16, respectivamente. En los
extraíbles alifáticos lineales se han encontrado típicamente ácidos (I-III), alcanos
(IV), alcoholes (V) y aldehídos (VI) de cadena larga, ceras (ésteres de ácidos
grasos con alcoholes de cadena larga) (VII), y mono (VIII), di (IX) y triglicéridos
(X). En la familia de extraíbles esteroidales se distinguen hidrocarburos
esteroidales (XI), esteroles (XII) y triterpenoides (XIII) libres, cetonas (XIV),
glicósidos (XV), y ésteres (XVI) esteroidales.

Una parte de los problemas originados durante la producción de pasta de

celulosa está relacionada con los extraíbles lipofílicos, presentes en los materiales
lignocelulósicos. La liberación de determinados compuestos lipofílicos durante el
pasteado da lugar a la formación de depósitos de ´´pitch´´, tanto en el producto
final como en la maquinaria. La formación de pitch puede disminuir la calidad de
pasta de papel, provocar paradas en la producción y aumentar el consumo de
reactivos químicos utilizados para su control (Allen, 2000; Hillis, 1989). Los
compuestos lipofílicos más problemáticos son los ácidos grasos libres, ácidos
resínicos, ceras, alcoholes, esteroles tanto libres como esterificados, glicéridos,
cetonas y otros compuestos (Fengel y Wegener, 1984; Gutiérrez et al., 1999, 2004;
Hillis, 1962; Rowe, 1989). Sin embargo, el comportamiento de los extraíbles
lipofílicos durante la producción de pasta de celulosa depende también del proceso
empleado para ello. Los depósitos de ´´pitch´´ procedentes de procesos del
pasteado mecánico, muestran una composición similar a la de los materiales
lignocelulósicos iniciales. Durante los procesos alcalinos los ésteres de glicerol se
saponifican y los ácidos grasos y resínicos se disuelven. Los ésteres esteroidales,
los esteroles libres y las ceras, se saponifican más lentamente que los ésteres de

20
 I. INTRODUCCIÓN

glicerol, y no forman jabones solubles como en el caso de los ácidos grasos libres y
tienen tendencia a depositarse (Gutiérrez et al. 2001).

OH
(I)
O

OH
(II)
O

OH
(III)

(IV)

OH
(V)

O
(VI)

O
(VII)
O

O OH
(VIII) OH
HO
O O

O (IX) O
O

O
O O

O (X) O

Figura 15. Estructura de compuestos representativos de los principales extraíbles

lipofílicos alifáticos lineales en materiales lignocelulósicos: ácido palmítico (I); ácido
oleico (II); ácido linoleico (III); n-nonacosano (IV); n-octacosanol (V); n-octacosanal (VI);
octacosanil hexadecanoato (VII); 1-monopalmitina (VIII); 1,3-oleoilpalmitina (IX); 1,2-
dilinoleoilpalmitina (X).

El estudio de la composición de los compuestos extraíbles lipofílicos de diversos

materiales lignocelulósicos constituye una información valiosa para ayudar a
identificar los compuestos responsables para la formación de ´´pitch´´ y sirve de
ayuda en la búsqueda de soluciones para el control del mismo.

21
I. INTRODUCCIÓN

(XI) (XIV)

OH
O
HO O
HO HO
OH
(XII) (XV)

HO O

(XIII) (XVI)

Figura 16. Estructura de compuestos representativos de las principales extraíbles

lipofílicos esteroidales en materiales lignocelulósicos: estigmasta-3,5-dieno (XI); sitosterol
(XII); β-amirina (XIII); estigmasta-4,22-dien-3-ona (XIV); sitosteril β-D-glucopiranósido
(V); sitosteril linoleato.

Aparte de ser problemática, la presencia de determinados compuestos extraíbles

lipofílicos en materiales lignocelulósicos, en cantidades abundantes, también puede
ser una fuente de compuestos químicos con alto valor añadido. Estos compuestos, o
derivados de ellos, pueden tener multitudes de aplicaciones en productos de dieta,
en la producción de fármacos, en productos cosméticos, en capas y superficies, etc.

Otros compuestos minoritarios

Otros compuestos minoritarios son las pectinas, proteínas, cutinas y suberinas, y
las sales minerales. Su contenido varía con el origen botánico, y en general, es más
elevado en las plantas herbáceas (Lu y Ralph, 2010a). La presencia de
determinados minerales puede tener efectos en procesos industriales, como en la
producción de pasta de papel (Samson y Mehdi, 1998).

22
 I. INTRODUCCIÓN

I.3. Biorrefinería de materiales lignocelulósicos

El concepto de biorrefinería consiste básicamente en la producción de

biomateriales y bioenergía a partir de materiales renovables (Figura 17), como son
los materiales lignocelulósicos.

BIOMASA Materias primas Residuos secos Residuos húmedos

Cultivos lignocelulósicos Residuos lignocelulósicos Residuos industriales líquidos
- madereros - forestales Estiércol
- no madereros - agrícolas Aguas residuales domésticas
Cultivos ricos en azúcar y almidón - agro-industriales
Cultivos oleaginosos Sólidos urbanos
Algas Aceites y grasas usados
Residuos de mataderos

PRETRATAMIENTO

CONVERSIÓN Térmicas Químicas Bioquímica Biológicos

Combustión Hidrólisis Hidrólisis enzimática Fermentación
Gasificación - ácida - aeróbica
Pirólisis - alcalina - anaeróbica
- En ausencia de catalizador Esterificación
- En presencia de catalizador

PRODUCTOS Gaseosos Aceitosos Sólidos Disueltos

INTERMEDIARIOS Gas de síntesis Aceite vegetal Pasta de celulosa Azúcares C6
Biogas Aceite de pirólisis Hemicelulosas Azúcares C5
Lignina precipitada Oligosacáridos
Residuos de conversión Extractivos
enriquecida en: Compuestos
- carbón derivados de lignina
- lignina

PRODUCTOS Bio-energía Biocombustibles Papel Productos químicos primarios Biomateriales

FINALES Calor Bioetanol Carbón
Electricidad Biodiesel Fibras de celulosa
Otros - materiales compuestos
- viscosa y celofán
A partir de carbohidratos A partir de lignina A partir de aceite
- derivados con acetatos,
- Aldehídos: (hidroximetil)furfural - Compuestos oxidados: ácidos - Esteres
carboximetil, ácidos
- Alcoholes: etanol, n-butanol, alifáticos y aromáticos, - Glicerina y derivados
grasos, etc.
1,4- butanodiol, etc. vainillina, siringaldehído, (epiclorohidrina, etc.)
- Ácidos: ácido cítrico, succínico, quinonas, etc. Polímeros
fumárico, málico, aspártico, - Hidrocarburos: benceno, A partir de carbohidratos
glucárico, glutámico, itacónico, tolueno, xileno, ciclohexano, Ácido poli-láctico,
levulínico, ácido-2,5-furan estirenos, etc. polietileno, poliésteres,
dicarboxilico, xilitol, etc. - Compuestos fenólicos: polihidroxialcanoato, etc.
fenoles libres y sustituidos, A partir de lignina
catecoles, cresoles, Derivados de lignina con
resorcinoles, eugenol, acetato, metilo, óxido de
coniferoles, guaiacoles, propileno, epóxido,
siringoles, etc. ácidos grasos, etc.
- Productos de gas de síntesis: Resinas, tableros,
H2, metanol, hidrocarburos dispersantes, adhesivos,
del proceso Fischer-Tropsch, fibras de carbono, etc.
etc.

Figura 17. Plataformas de producción en biorrefinerías.

Los constituyentes principales de los materiales lignocelulósicos (celulosa,

hemicelulosas y lignina) son, en teoría, susceptibles de separación en un esquema
de fraccionamiento integral, similar al de la refinería de petróleo (Werpy y

23
I. INTRODUCCIÓN

Petersen, 2004). El alto grado de entrecruzamiento entre los constituyentes es una

de las principales barreras económicas para el desarrollo de la biorrefinería, tanto
por la resistencia de la matriz lignocelulósica a su degradación (Zhang et al., 2007),
como por su insolubilidad en la mayoría de los disolventes (Zakzeski et al., 2010).
Tanto la eficiencia del fraccionamiento de los materiales lignocelulósicos, como el
valor añadido de los productos obtenidos, o derivados de ellos, juegan un papel
esencial en el éxito industrial de los procesos de biorrefinería. Actualmente, los
principales procesos de biorrefinería de materiales lignocelulósicos son la
producción de celulosa y la producción de bioetanol. Los residuos de lignina de
estos procesos también son susceptibles de aprovechamiento industrial. A
continuación se describen más en detalle la obtención de pasta de celulosa para la
producción de papel, y la obtención de bioetanol. Finalmente, se describe
brevemente los posibles productos y aplicaciones de las fracciones de lignina.

I.3.1. Producción de pasta de celulosa

Mucho antes de la introducción del concepto de biorrefinería, los materiales

lignocelulósicos ya se utilizaron para la producción de papel. El papel es un
producto de biorrefinería, en el sentido de que su producción consiste básicamente
en la separación de las fibras de celulosa de la matriz lignocelulósica, a través de
procesos mecánicos y/o químicos (el pasteado).

Aparte de papel, también se puede obtener una amplia variedad de otros

productos de alto valor añadido a partir de celulosa como la viscosa, nanocelulosa,
materiales compuestos, etc., o derivados de celulosa como carboximetilcelulosa,
acetatos, ácidos grasos, etc. Además, el concepto de la biorrefinería no sólo tiene
por objeto el fraccionamiento de la matriz lignocelulósica, sino también aumentar
la sostenibilidad tanto a nivel de uso de los recursos naturales y a nivel técnico-
económico, como a nivel medioambiental. Este concepto se está desarrollando en
el sector papelero mediante el desarrollo de procesos que aumentan el valor
añadido de las fracciones de las hemicelulosas y la lignina, y de procesos más
respetuosos con el medioambiente.

Según la CEPI (´´Confederation of European Paper Industries´´), la producción

de papel y de pasta de papel a nivel global fue respectivamente 371 y 178 millones
de toneladas en 2009, lo que representa el 27 y 24% en Europa, respectivamente
(Figura 18). En 2011 el consumo de madera en la zona CEPI, destinado a la
producción de papel, se situó alrededor de 1.49 × 108 de m3 de madera, de los

24
 I. INTRODUCCIÓN

cuales el 72% procedía de coníferas y el 28% de frondosas. En las últimas décadas,

la producción de papel ha aumentado de forma ininterrumpida. Sin embargo, en
2009 la producción de papel y de pasta de papel en la zona CEPI bajó el 7 y el 10%
(respecto a 2008), debido a la crisis económica mundial. En 2010 se recuperó el
crecimiento global de la producción, sin embargo lo hizo con menos fuerza en la
zona CEPI. La CEPI constató que se necesita un esfuerzo para implementar más
tecnologías sostenibles en la industria papelera, y para garantizar el futuro del
mercado europeo, debido al crecimiento de la producción en países emergentes.

Producción global de papel Producción de papel en la zona CEPI Tipo de papel a nivel global
Asia 43% Alemania 24%
Prensa 9%
Europa 27% Suecia 32%
Finlandia 24% Escribir e imprimir 28%
América del
Tejidos y pañuelos 7%
Norte 23% Francia 10%
América Italia 9% Embalaje y envases
Latina 5% (papel ondulado) 35%
España 6%
Oceanía 1% Gran Bretaña 5% Embalaje (cartón) 13%

África 1% Otro 8%
Otro 22%

Producción global Producción de pasta de Tipo de pasta a nivel global

de pasta de papel papel en la zona CEPI
América del Suecia 32%
Norte 37% Finlandia 24%
Europa 24% Alemania 7%
Pasta química 70%
Asia 24% Portugal 7%
América
Pasta mecánica 21%
Francia 5%
Latina 12% Otro 9%
Noruega 5%
Oceanía 2% Austria 4%
África 1% Otro 16%

Figura 18. Producción de pasta de papel y de papel, a nivel global y en la zona CEPI en
2009 (Adaptado de CEPI Sustainability Report, 2011).

I.3.1.1. Procesos de producción de pasta de papel

La fabricación de pasta de papel consiste básicamente en la separación de las
fibras de celulosa de la matriz lignocelulósica, a través de procesos mecánicos y/o
químicos. El contenido, la composición y la estructura química de la lignina son
parámetros importantes en la producción de pasta de celulosa, ya que influyen en el
consumo de reactivos y en la calidad y el rendimiento final de la pasta. La
proporción de unidades S y G (la relación S/G), presentes en la lignina de especies
frondosas es un parámetro determinante, como consecuencia de sus diferentes
reactividades en medio alcalino (Chang y Sarkanen, 1973; Santos et al., 2011;
Tsutsumi et al, 1995), y su importancia ha sido demostrado en pastas alcalinas de

25
I. INTRODUCCIÓN

eucalipto (del Río et al., 2005; González-Vila et al., 1999) y de álamo (Bose et al.,
2009a).

Procesos mecánicos
El pasteado mecánico separa las fibras por fragmentación mecánica, utilizando
molinos y refinadores de discos, lo que supone un considerable gasto energético.
La acción de las máquinas rompe las fibras de celulosa, por lo que la pasta
resultante es más débil que la obtenida químicamente. La lignina no se disuelve,
simplemente se ablanda, permitiendo que las fibras se asienten fuera de la
estructura de la madera. Es un proceso que ofrece un rendimiento muy alto,
obteniéndose pastas que resultan ventajosas para algunos tipos de papel, ya que
confieren rigidez, volumen y opacidad. No obstante, el alto contenido de lignina en
la pasta limita la calidad del papel ya que las fibras son poco flexibles, no están
bien unidas entre sí y el papel es poco resistente y tiende a amarillear.

Procesos químicos
La acción de los procesos químicos se basa en la disolución (parcial) de la
lignina y de las hemicelulosas. En general durante el pasteado químico, los enlaces
éter tipo alquil-aril se empobrecen en la lignina, los enlaces condensados se
enriquecen, y se pueden formar nuevos enlaces entre lignina y carbohidratos
(Froass et al., 1996, 1998; Gellerstedt et al., 1984; Gierer, 1982, 1985).

Proceso al sulfito
El proceso al sulfito dominó la industria papelera desde finales del siglo XIX
hasta casi mediados del XX, pero actualmente representa sólo una pequeña
proporción de la producción global de pasta de papel. El proceso al sulfito emplea
los reactivos H2SO3 o Ca(HSO3)2, producidos a partir de SO2 y CaCO3. La lignina
disuelta se puede recuperar en forma de lignosulfonatos.

Proceso kraft
Este tipo de pasteado permite obtener pastas con una gran resistencia y un alto
grado de deslignificación (hasta 90%), aunque con menor rendimiento (40-60%), a
partir de diferentes tipos de maderas (García Hortal y Colom, 1992; Santos et al.,
1997). En primer lugar, se cargan en el digestor las astillas con el denominado licor
blanco (mezcla de NaOH y NaS2), y se calienta hasta una determinada temperatura
(160-180 ºC), manteniéndose estas condiciones hasta dos horas, en función del tipo
de madera, el grado de deslignificación y el rendimiento de pasta requerido. Una

26
 I. INTRODUCCIÓN

vez terminada la cocción, la mezcla de pasta y astillas no digeridas salen del

digestor y se separan por cernido, y se separa la pasta, que a continuación pasa a
una etapa de lavado. El licor resultante de la cocción, denominado licor negro, se
somete al ciclo de regeneración. En primer lugar se concentra por evaporación
hasta que su contenido en agua es inferior al 40 %, y se pulveriza en la caldera de
recuperación. La parte orgánica se consume como combustible, generando calor
que se recupera en forma de vapor a elevada temperatura, que se utiliza para
calentar la caldera, evaporar disolvente en la etapa de concentración del licor negro
y para producir energía eléctrica. La parte inorgánica no quemada se recoge en el
fondo de la caldera como una mezcla fundida, que se disuelve en una solución
cáustica débil, obteniéndose un “licor verde” que contiene principalmente Na2S
disuelto y Na2CO3. Este licor se bombea a una planta de recaustificación donde se
clarifica y entonces reacciona con cal apagada (Ca(OH)2), formando NaOH y
CaCO3. El CaCO3 se envía a un horno de cal donde se calienta para regenerar cal
viva (CaO). De este modo se regenera el licor blanco que se filtra y se almacena
para ser usado nuevamente. Para el proceso kraft se pueden utilizar todo tipo de
maderas, aunque los mejores resultados se obtienen con maderas de frondosas.

Proceso a la sosa
El proceso a la sosa es parecido al proceso kraft, en el sentido de que extrae y
degrada la lignina en medio alcalino, pero en ausencia de Na2S, lo que puede
aumentar el tiempo de residencia en el digestor, requerido para alcanzar un
determinado contenido de lignina residual. Se puede añadir antraquinona (AQ), que
actúa como catalizador de la deslignificación y estabilizador de los carbohidratos, y
otros agentes químicos que pueden asistir al álcali en el proceso de cocción. El
proceso sosa-AQ se suele utilizar más con materiales lignocelulósicos no
madereros (Bose et al., 2009b; Kanungo et al., 2009).

Proceso organosolv
El proceso organosolv emplea disolventes orgánicos como alcoholes (metanol,
etanol, butanol, 1,3-butanodiol, glicerol, glicol y derivados, etc.), ácidos orgánicos
(ácido fórmico, acético, etc.), cetonas (acetona, γ-lactona, etc.), éteres (1,4-
dioxano) o aminas (etilen y hexametilen diamina). En general, este proceso opera a
alta temperatura y presión, con la adición de agua como co-disolvente. Este proceso
rompe eficazmente los enlaces entre carbohidratos y lignina, y produce pastas de
celulosa con un bajo contenido en lignina residual (Muurinen, 2000). Se puede
emplear ácidos o bases como catalizadores. La aplicación industrial está sujeta al
coste del disolvente, a la recuperación de esta misma y al impacto medioambiental.

27
I. INTRODUCCIÓN

En este sentido, metanol y etanol resultan los más atractivos. Se han desarrollado
diversos procesos a nivel industrial, a base de metanol (Organocell), etanol
(Alcell), ácido fórmico (Formacell y Milox), etc. Sin embargo, para la producción
de pasta de papel a nivel industrial, el proceso organosolv actualmente no ha sido
implementado a gran escala, ya que los reactivos son más costosos y la mayoría de
los actuales productores de pasta de papel disponen de reactores y equipos
diseñados para el proceso kraft.

I.3.2. Producción de bioetanol

Los biocombustibles poseen un gran interés industrial, ya que gran parte de los
recursos energéticos fósiles se destina a los carburantes para transporte.
Actualmente, los biocombustibles líquidos se producen principalmente a partir de
cultivos oleaginosos (biodiesel), y de cultivos ricos en azúcares y almidón
(bioetanol), y representan los biocombustibles de primera generación. En 2011 la
producción de biodiesel representó el 27% de la producción de biocombustibles a
nivel global, un porcentaje bastante inferior a la de bioetanol (Figura 19).

60 30
a Resto del mundo b
Bioetanol 2001
Europa & Asia Bioetanol 2011
Central & Sur de América Biodiesel 2001
Norte de América Biodiesel 2011

40 20

20 10

Norte de Central & Sur Europa Resto del

2001 2003 2005 2007 2009 2011
América de América & Asia mundo

Figura 19: (a) Evolución de la producción conjunta de bioetanol y biodiesel entre 2001 y
2011, y (b) de bioetanol y biodiesel por separado en 2001 y 2011, expresado en millones
de toneladas de equivalentes de petróleo (BP Statistical Review of World Energy, 2012).

28
 I. INTRODUCCIÓN

Sin embargo, los cultivos destinados para estos biocombustibles de primera

generación compiten con los destinados a la producción de productos alimenticios,
tanto a nivel económico como a nivel de los recursos naturales necesarios para su
cultivo. Por este motivo, se están investigando materiales lignocelulósicos, tanto de
origen forestal como agrícola, que no compiten en el mercado alimenticio, para la
producción de biocombustibles de segunda generación (Figura 20).

PRIMERA GENERACIÓN
Caña de azúcar Azúcar

Granos de maíz Almidón

SEGUNDA GENERACIÓN
Residuos agrícolas
Paja de cereales, rastrojo de maíz,
bagazo de caña de azúcar, etc.
Pretratamiento Hidrólisis
Cultivos energéticos
Álamo (Populus), eucalipto, pasto Pentosas Hexosas
Coproductos
varilla (Panicum), Miscanthus, etc.

Valorización Fermentación
Residuos forestales industrial
Serrín, etc.
Residuo rico Bioetanol
en lignina
Residuos urbanos
Residuos sólidos urbanos, residuos
de papel, etc. Energía (vapor & electricidad)

Figura 20. Las principales fuentes de biomasa aptas para la producción de bioetanol de
primera y de segunda generación.

Los residuos lignocelulósicos, en particular los residuos agrícolas, también son

de gran interés para la producción de bioetanol de segunda generación. Entre ellos,
la paja de trigo es uno de los más interesantes por su bajo coste, disponibilidad
geográfica y abundancia durante todo el año. Según Kim y Dale (2004), la paja de
trigo representa casi un 25% de la capacidad de producción de bioetanol a partir de
los residuos agrícolas más abundantes a nivel global (rastrojo de maíz, bagazo de
caña de azúcar, paja de arroz, sorgo y otros cereales). En general, se obtiene 1.3-1.4
kg de paja por kg de grano. En 2011 la producción de paja de trigo se situó
alrededor de 680 millones de toneladas (Sarkar et al. 2012), a partir de cual se
podrían haber producido potencialmente 1.98×108 m3 de bioetanol, según Kim y

29
I. INTRODUCCIÓN

Dale (2004). Esta cantidad podría haber cubierto alrededor de 12% del consumo
global de gasolina y hubiera sido cinco veces más que la producción efectiva de
bioetanol en 2011 (BP Statistical Review of World Energy 2012).

El bioetanol se obtiene por hidrólisis de los carbohidratos, seguida de

fermentación de los azúcares reductores obtenidos. Los factores principales que
afectan el rendimiento de la hidrólisis son el contenido y estructura de la lignina y
de las hemicelulosas, la cristalinidad de la celulosa, y la porosidad y el área
específica de la materia lignocelulósica (McMillan, 1994). La lignina es un
polímero recalcitrante que limita el acceso de reactivos y enzimas usados durante la
hidrólisis, y a partir de cual se pueden formar inhibidores que influyen
negativamente en el rendimiento de la fermentación. Por este motivo, es necesario
aplicar pretratamientos que eliminen parcialmente la lignina o que modifiquen su
estructura.

I.3.2.1. Pretratamientos del material lignocelulósico

La función principal de los pretratamientos utilizados en la biorrefinería es

reducir el tamaño de las partículas de los materiales lignocelulósicos para reducir
los gastos energéticos en los posteriores procesos. Los requisitos de un
pretratamiento idóneo para la obtención de bioetanol fueron recogidos por
Carvalheiro et al. (2008) y Yang y Wyman (2008):

- Aumentar la porosidad y el área específica de la biomasa para facilitar la

accesibilidad de los reactivos o las enzimas que se emplean durante la
hidrólisis.
- Reducir el pretratamiento mecánico (reducir gastos energéticos).
- Generar celulosa activa (reducir la cristalinidad de la celulosa, apta para los
posteriores procesos de hidrólisis y fermentación).
- Desacoplar las hemicelulosas y maximizar el rendimiento de pentosas en
forma no degradada.
- Obtener hidrolizados con un bajo nivel de inhibidores, y minimizar las
pérdidas de nutrientes necesarios para la fermentación.
- Emplear reactores con una alta carga de sólidos por volumen.
- Minimizar los vertidos y contaminantes residuales.

Entre los distintos pretratamientos existentes, se puede distinguir entre

pretratamientos mecánicos, químicos, físico-químicos y biológicos, que se detallan
a continuación.

30
 I. INTRODUCCIÓN

Pretratamientos mecánicos
Los pretratamientos mecánicos, tales como la molienda, la pulverización por
martilleo y la extrusión, buscan reducir el radio del espacio entre las paredes
celulares, que debería acercarse al tamaño molecular de las enzimas hidrolíticas
(Carroll, 2009), que se sitúa alrededor de 20 kDa (2.5 nm radio Stokes). En general,
los pretratamientos mecánicos son efectivos, pero son largos y/o costosos. La
energía necesaria para la reducción del tamaño de las partículas puede alcanzar
hasta una tercera parte de la energía total necesaria para la producción de bioetanol
(Aden et al., 2002; US Department of Energy, 1993).

Pretratamientos químicos

Pretratamientos ácidos
La hidrólisis con ácido concentrado solubiliza la celulosa y las hemicelulosas,
dejando un residuo sólido enriquecido en lignina, y produce un alto nivel de
monosacáridos, por lo cual no requiere un posterior proceso de hidrólisis
enzimática (Sun and Cheng, 2002). El rendimiento de la hidrólisis aumenta con la
concentración del ácido, sin embargo, la corrosión de reactores y equipos se
intensifica también. Además, la recuperación del ácido es un factor crítico debido a
su alto coste. En general, la hidrólisis ácida está asociada con la generación de
inhibidores de las colonias de levadura utilizadas en el paso de fermentación, tales
como furfural (producido por la hidrólisis de pentosas), hidroximetilfurfural
(producido por la hidrólisis de hexosas), compuestos derivados de lignina
(compuestos fenólicos y aromáticos), ácido acético, ácido fórmico, ácido levulínico
y compuestos extraíbles (ácidos resínicos y terpénicos y derivados de tanino)
(Mussatto y Roberto, 2004). La hidrólisis con ácido diluido hidroliza
mayoritariamente las hemicelulosas, produciendo una mezcla de monosacáridos y
oligosacáridos (Chen et al., 2007). Se pueden separar las fases para fermentarlas
por separado, y reciclar el ácido tras neutralización, para reducir la degradación de
los carbohidratos y disminuir el efecto de la inhibición. El uso de ácidos orgánicos,
como ácido maleico y fumárico, puede reducir la degradación de los carbohidratos
y generar menos inhibidores (Kootstra et al., 2009).

Pretratamientos alcalinos
Este pretratamiento aumenta la superficie específica de la biomasa y reduce la
cristalinidad de la celulosa, y disuelve parcialmente las hemicelulosas y la lignina
(Balat et al., 2008). Su mecanismo de acción consiste en la saponificación de los

31
I. INTRODUCCIÓN

enlaces éster intermoleculares que unen las hemicelulosas y la lignina (Sun y

Cheng, 2002). Además, se hidrolizan los acetatos y ácidos urónicos que acilan
determinadas unidades de las hemicelulosas, lo que aumenta la accesibilidad de las
enzimas hidrolíticas (Chang y Holtzapple, 2000). Sin embargo, la hidrólisis de
acetatos en los xilanos durante el tratamiento alcalino puede bajar su solubilidad, y
puede dar lugar a la deposición en la fibra de celulosa. El nivel de degradación de
los carbohidratos es mucho menor que en medio ácido, ya que en medio alcalino la
hidrólisis pasa por otro mecanismo (Krassig y Schurz, 2002; Solomon, 1988). La
hidrólisis alcalina de celulosa requiere bases fuertes y temperaturas altas (> 150
ºC). El pretratamiento alcalino en combinación con oxígeno o aire puede aumentar
la deslignificación (Chang y Holtzapple, 2000).

Los pretratamientos AFEX (´´Ammonia Fiber Explosion´´) y ARP (´´Ammonia

Recycle Percolation´´) también son tratamientos alcalinos, ya que emplean
amoniaco (NH3). La acción de AFEX se basa en la acción química de álcali y en la
acción física de la explosión de vapor, y también se puede considerar un
pretratamiento físico-químico. Este tratamiento da un rendimiento elevado de
hidrólisis y de fermentación con materiales lignocelulósicos de plantas herbáceas
(Carvalheiro et al., 2008; Mosier et al., 2005). El principal inconveniente es el coste
alto de NH3, y las altas presiones requeridas. En este sentido el tratamiento ARP,
en cual se deja colar una disolución diluida de NH3 en una columna con biomasa a
140-170 ºC, es más atractivo, ya que no requiere altas presiones.

Proceso organosolv
El proceso organosolv se ha descirito anteriormente como proceso químico para
la producción de pasta de papel. Como pretratamiento para procesos de
biorrefinería, se suelen utilizar disolventes que permiten operar a una temperatura
moderada y con menor impacto hacia el medioambiente. El alto grado de
fraccionamiento, y la ausencia de sulfuro en los licores de cocción, hacen que el
proceso organosolv sea un pretratamiento muy atractivo para procesos de
biorrefinería, en particular para la producción de bioetanol (Pan et al., 2006; Xu et
al., 2006; Villaverde et al., 2010).

Pretratamientos oxidativos
La oxidación húmeda emplea oxígeno o aire como catalizador, en combinación
con agua a alta temperatura y presión durante un tiempo corto. Sin embargo, los
reactores, que operan a altas presiones, son costosos. Por este motivo también se
emplea Na2CO3 en vez de oxígeno o aire.

32
 I. INTRODUCCIÓN

El peróxido alcalino emplea H2O2 en medio alcalino. El H2O2 es muy inestable

en medio alcalino, en cual forma hidroperóxido (OOH-) que se descompone en
radicales más reactivos, como hidroxilos (HO.) y aniones de superóxido (O2-.)
(Gierer, 1997).

Disolución en líquidos iónicos

Los líquidos iónicos contienen un anión inorgánico y un catión orgánico, que
pueden ser de diversos tipos. El anión debilita las fuerzas intermoleculares entre
grupos hidroxilos de los diferentes constituyentes, que da lugar al hinchamiento de
la pared celular, demostrado de forma visual mediante microscopía de
fluorescencia avanzada (Figura 21).

Figura 21. (A) Visualización de una sección de células de pasto varilla (Panicum
virgatum) mediante microscopía de fluorescencia avanzada antes del tratamiento con
líquidos iónicos, (B) después 20 min, con un enfoque en la desintegración del
esclerénquima, (C) después 30-50 min (dispersión viscosa), y (D) después 150-180 min,
resultando en la disolución completa de celulosa, hemicelulosas y lignina (Singh et al.,
2009).

El grado de solubilización de celulosa, hemicelulosas y lignina depende del tipo

de líquido iónico. La celulosa se puede regenerar mediante la represión de la
disolución con un ´anti-disolvente´, por ejemplo agua, para recuperarla
parcialmente en estado amorfo. En algunos casos se requiere un paso de
eliminación o recuperación de los líquidos iónicos, ya que pueden desactivar las
celulasas.

33
I. INTRODUCCIÓN

Pretratamientos físico-químicos

Los pretratamientos físico-químicos más utilizados son la explosión de vapor y

variantes de la misma. La explosión de vapor consiste básicamente en calentar el
material lignocelulósico en un reactor con vapor a alta presión durante unos
cuantos minutos (20-50 bar, 210-290 ºC). Después, la bajada brusca de la presión
hace expandir el vapor dentro de la biomasa para facilitar la impregnación de agua
en la matriz lignocelulósica. La explosión de vapor en presencia de catalizadores,
impregna la biomasa previamente con ácido, SO2 o CO2. En ausencia de un
catalizador ácido, el medio ácido se genera por la hidrólisis de ácido acético y
urónico presente en el material lignocelulósico (autohidrólisis). Otra variante de la
explosión de vapor es LHW (´´Liquid Hot Water´´), que emplea agua caliente
(200-230 ºC) a presiones más moderadas. El nivel de inhibidores se puede reducir
en cierto modo mediante el control de pH del agua caliente con álcali. La mayor
limitación es el consumo elevado de agua.

Pretratamientos biológicos y bioquímicos

Una gran parte de los pretratamientos biológicos en estudio de materiales

lignocelulósicos emplean diversos hongos de podredumbre blanca, y se basan en la
acción de las enzimas ligninolíticas (Hatakka, 2001). La degradación selectiva
afecta a la lignina y hemicelulosas y la celulosa se mantiene relativamente intacta,
mientras la degradación no selectiva afecta a los tres constituyentes
lignocelulósicos en un grado parecido (Blanchette, 1995). La mayoría de los
hongos de podredumbre blanca (como Phanerochaete chrysosporium,
Ceriporiopsis subvermispora, Phlebia subserialis, Pleurotus ostreatus, Pycnoporus
cinnabarinus, etc.) son capaces de degradar la lignina selectivamente en diversos
materiales lignocelulósicos (Isroi et al., 2011). Sin embargo, la acción de las
enzimas aisladas es más directa y selectiva que la de los hongos, ya que el
tratamiento con hongos requiere tiempos de residencia más largos y puede
conllevar una degradación más severa de los carbohidratos. Las enzimas
lignolíticas se dividen en las lignina peroxidasas (LiP), lacasas (Lac), manganeso
peroxidasas (MnP), peróxidasas versátiles (VP), glioxal oxidasas (GLOX) y aril
alcohol oxidasas (AAO) (Hatakka, 2001). Las lacasas son un grupo de oxidasas con
un gran potencial para diversas aplicaciones biotecnológicas. Tienen un centro
activo de multi-cobre, que en presencia de oxígeno atacan a sustratos ricos en
electrones, como sustratos fenólicos. En presencia de mediadores (compuestos de
bajo peso molecular), determinadas lacasas aumentan su especificad hacia sustratos
no fenólicos (Bourbonnais y Paice, 1990). Un mediador eficaz es el HBT (1-

34
 I. INTRODUCCIÓN

hidroxibenzotriazol), con el cual se obtuvieron obtuvieron 48 y 32% de

deslignificación en madera molida de eucalipto y de hierba elefante,
respectivamente, con tiempos de residencia relativamente cortos (Gutiérrez et al.,
2012).

I.3.2.2. Hidrólisis y fermentación

En la hidrólisis de los carbohidratos, destinados para la producción de bioetanol,

se persigue obtener el mayor rendimiento posible de celobiosa y monosacáridos
(unidades de carbohidrato con un grupo reductor) como la glucosa, xilosa, xilitol,
arabinosa, galactosa, etc. La hidrólisis ácida es actualmente el proceso más
utilizado a nivel industrial, y frecuentemente se lleva a cabo de forma simultánea a
la fermentación para reducir los costes. Sin embargo, la hidrólisis ácida conlleva un
determinado nivel de corrosión en los reactores y está asociada con la generación
de inhibidores que influyen negativamente en el rendimiento de la fermentación.
Por este motivo, se han desarrollado procesos de hidrólisis enzimática, basados en
mezclas de celulasas y β-glucosidasas que, sin embargo, están limitados por su alto
coste y por la resistencia de la matriz lignocelulósica, debido al carácter
recalcitrante de la lignina (Chen y Dixon, 2007). Además, para obtener un alto
rendimiento de hidrólisis a partir de los xilanos, se requiere una mezcla de
diferentes enzimas como endo-β-1,4-xilanasa, β-xilosidasa, y opcionalmente α-L-
arabinofuranosidasa, α-glucuronidasa, acetilxilano esterasa, feruloil esterasa, y p-
cumaroil esterasa.

La conversión a bioetanol se lleva a cabo por fermentación a partir de los

monosacáridos. El bioetanol es uno de los principales productos de biorrefinería
obtenidos por fermentación, mediante el uso de colonias apropiadas de levaduras
(como Saccharomyces cerevisiae y Candida utilis) o bacterias (como Zymomona
mobilis). Después de la fermentación, el etanol se concentra y se purifica mediante
destilación.

I.3.3. Productos a partir de la lignina

La fracción de la lignina resultante de la obtención de la celulosa, también se

puede aprovechar mediante la obtención de lignina aislada para aplicaciones como
biocombustible sólido y como polímero en biomateriales, y mediante la conversión
de lignina a compuestos químicos primarios (´´building blocks´´). Por un lado,
durante la producción de pasta de celulosa, se produce una gran cantidad de lignina

35
I. INTRODUCCIÓN

en los licores negros, que se suelen aprovechar mediante combustión para

cogenerar energía. Por otro lado, el residuo sólido obtenido después la hidrólisis y
fermentación de los carbohidratos también es rico en lignina. Actualmente en la
producción de bioetanol, se utiliza alrededor de 40% de este residuo en la
combustión para recuperar la energía requerida para cubrir las necesidades térmicas
del pretratamiento y de la destilación. Esto implica que en teoría el 60% de este
residuo está disponible para valorización industrial (Sannigrahi et al., 2010). Sin
embargo, la valorización de esta lignina resulta difícil, porque muchos procesos de
conversión requieren un paso de purificación, ya que el residuo de la fermentación
también contiene restos de carbohidratos, enzimas, bacterias, etc. (Zakzeski et al.,
2010). La valorización de la fracción de lignina, más allá del aprovechamiento
energético por combustión (calor y electricidad), puede aumentar la viabilidad
técnica-económica y medioambiental de procesos de biorrefinería, en particular
para productores papeleros y productores de bioetanol. Es de resaltar que el
aprovechamiento de diferentes tipos de materiales lignocelulósicos implica que se
obtienen productos de lignina con diferentes composiciones y propiedades, ya que
la composición y la estructura de lignina varía con el origen del material y
consecuentemente con los procesos aplicados a ello. Esta alta variedad es uno de
los mayores retos para las aplicaciones de lignina, pero también representa un gran
potencial para la obtención de diversos productos.

Aplicaciones de lignina en biomateriales

Aquellos procesos que valorizan la fracción de lignina para su uso como

polímero en biomateriales, requieren pretratamientos que extraen un máximo de
lignina del material lignocelulósico en forma no degradada, ya que durante los
procesos posteriores el nivel de degradación aumenta (Lora y Glasser, 2002). Estos
pretratamientos de deslignificación son las cocciones alcalinas, organosolv y al
sulfito, a partir de cuales se obtienen licores de cocción. Tradicionalmente, los
lignosulfonatos obtenidos de los licores de cocción del proceso al sulfito se han
aprovechado por combustión. Más recientemente, los lignosulfonatos han
encontrado un importante campo de aplicación como dispersantes. Actualmente, la
lignina kraft se produce en grandes cantidades, ya que este proceso es el más
utilizado para la producción de pasta de papel a nivel industrial. La extracción de
lignina de los licores de cocción aumenta el rendimiento en las calderas de
recuperación en las plantas de productores papeleros, lo que permite aumentar la
capacidad de producción y aprovechar el exceso energético de las plantas
productoras en forma de un combustible sólido, ya que la lignina tiene un

36
 I. INTRODUCCIÓN

contenido energético elevado (Wallmo et al., 2009). Un ejemplo es el proceso

LignoBoost, desarrollado a nivel industrial, que utiliza CO2 y H2SO4 en varios
pasos en combinación con filtración y lavado. Este proceso es capaz de obtener
ligninas precipitadas del proceso kraft con una pureza relativamente elevada (<
0.5% de sodio y 1-5% de carbohidratos), lo que es importante para facilitar sus
posibles aplicaciones, ya que un contenido alto en carbohidratos influye
significativamente en su reactividad y sus propiedades (Singh et al. 2005). Aparte
de combustibles sólidos, se están desarrollando otras aplicaciones de ligninas
precipitadas para su uso en materiales con alto valor añadido, como en la
fabricación de resinas fenol-formaldehído, tableros, polímeros de capa fina para
envases en la industria alimenticia y para protección de electrodos de grafito en
baterías, polímeros espumosos, surfactantes poliméricos, emulsionantes,
dispersantes y floculantes (en mezclas de betunes y asfaltos, pinturas, herbicidas y
adhesivos), anti-oxidantes en productos cosméticos, estabilizadores de plásticos,
nanofibras de carbono, etc.

Conversión de lignina a productos químicos primarios

La conversión de lignina en productos químicos primarios se puede llevar a

cabo por gasificación para obtener gas de síntesis (Zakzeski et al., 2010), y por
licuefacción mediante pirólisis y procesos (hidro)térmicos de craqueo, en los cuales
se utilizan soportes de zeolitas u óxidos de aluminio que fragmentan la lignina y
catalizadores organometálicos que rompen los enlaces C-O de forma reductiva
(hidrogenólisis). Se obtiene una mezcla de oligómeros y monómeros fenólicos y
aromáticos (cresoles, catecoles, resorcinoles, quinonas, vainillina, guayacoles, etc.),
y alcanos alifáticos lineales y cíclicos, a partir de cuales se pueden obtener
biocombustibles y una amplia gama de derivados para la industria química y
farmacéutica. Esta conversión requiere una despolimerización, una reducción del
contenido en oxígeno y un aumento del contenido en hidrógeno, aunque también
existen aplicaciones de oxidación con catalizadores específicos, tanto químicos
como enzimáticos, que producen una mezcla de aldehídos, ácidos carboxílicos y
quinonas aromáticas. La temperatura, el tiempo y el tipo del catalizador y del
soporte influyen en las reacciones de degradación y de condensación. La
condensación da lugar a la formación de materia carbonizada y reduce el
rendimiento de la licuefacción. Este residuo sólido se puede utilizar para
aplicaciones de carbón. Los estudios más recientes intentan combinar la
fragmentación y la reducción del contenido en oxígeno en un paso mediante una
pirólisis reductiva a 350-400 ºC con ácido fórmico y etanol, obteniendo un

37
I. INTRODUCCIÓN

rendimiento máximo de licuefacción, tanto a partir de lignina enzimática (Kleinert

et al., 2009), a partir de lignosulfonatos como a partir de lignina de abedul obtenida
por explosión de vapor (Gellerstedt et al., 2008).

38
II. OBJETIVOS
 II. OBJETIVOS

La presente Tesis aborda el estudio de la composición química de los principales

constituyentes de diversos materiales lignocelulósicos, incluyendo madera de
eucalipto, caña de hierba elefante y paja de trigo. Estos estudios permiten evaluar el
potencial de diversos materiales lignocelulósicos, y sirven como punto de partida
para obtener un mejor aprovechamiento industrial de dichos materiales mediante
procesos de biorrefinería.

Los objetivos específicos de esta Tesis son los siguientes:

- Realizar una caracterización química detallada de los diferentes materiales

lignocelulósicos seleccionados (madera de eucalipto, hierba elefante y paja
de trigo), con especial énfasis en la composición química y estructura de la
lignina y de los lípidos.
- Estudiar la modificación estructural de la lignina de eucalipto durante
diferentes métodos de deconstrucción alcalina (kraft, sosa-AQ y sosa-O2).

41
III. MATERIALES Y MÉTODOS
 III. MATERIALES Y MÉTODOS

III.1. Reactivos y patrones

Acetona: C3H6O, 99.5% (GC), Panreac, (RFE, USP, BP, Ph. Eur.) PRS-CODEX.
Ácido acético glacial 100%: C2H4O2, 99.8-100.5% (acidimétrico), Merck.
Ácido propiónico: C3H6O2,  99.0% (GC), purum, Fluka.
Ácido sulfúrico: H2SO4, 96 % ( 95 % acidimétrico), PA-ISO.
Anhídrido acético: C4H6O3, > 99.0%, Panreac (Reag.Ph.Eur, PA-ACS-ISO).
Anhídrido propiónico: C6H10O3,  96% (NT), purum, Fluka.
Bromuro de acetilo: C2H3OBr,  97% (método morfolina) para síntesis, Merck.
Bromuro de propionilo: C3H5BrO, 98%, Sigma Aldrich.
BSTFA: C8H18ONSi2F3, Supelco.
Cloro-4,4,5,5-tetrametil-l,3,2-dioxafosfolano (reactivo II): C6H12O2ClP, Sigma
Aldrich, 95%.
Cloroformo: CHCl3,  99.8% (GC), SupraSolv®, Merck.
Cloroformo-d:CDCl3, 99.96%, Sigma Aldrich.
Clorito sódico: NaClO2, 25% (m/m) en agua para síntesis.
Cromo(III) acetilacetonato: C15H21CrO6, 99.99%, Sigma Aldrich.
Cloruro de amonio: NH4Cl, 99.5%, Panreac (PA-ACS-ISO).
Diclorometano: CH2Cl2, estabilizado con amileno,  99.5% (GC), Panreac (PA-
ACS-ISO).
1,2-dicloro-etano: C2H4Cl2,  99.5% (GC) puris., Merck.
Dimetilformamida: C3H7ON, SupraSolv® Merck.
Dimetilsulfóxido: C2H6OS,  99.9% (GC) H2O  0.1%, puriss., Fluka.
Dimetilsulfóxido-d6: C2D6OS,  99.8% grado de deuterización (RMN), Panreac.
1,4-Dioxano: C4H8O2,  99.0% (GC) puriss., Merck.
Endo-N-hidroxi-5-norborneno-2,3-dicarboximide: C9H9O3N, 97%, Sigma Aldrich.
Etanol: C2H6O, Merck (Reag.Ph.Eur, ACS-ISO).
Éter de petróleo: punto de ebullición 35–60 ºC, Pestipur, Carlo Erba Reactifs-SDS.
Piridina: C5H5N,  99.5% (GC),SeccoSolv®, Merck.
Piridina-d5: C5D5N,  99.95% grado de deuteración, Panreac.
Sulfato de sodio anhidro: Na2SO4, 99.0%, Panreac (PA-ACS-ISO).
Tetrabutilamoniohidróxido: C16H37ON,  40% (m/m) in H2O purum, Fluka.
Tetrametilamoniohidróxido: C4H13ON, 25% (m/m) en metanol, Sigma Aldrich.
Tolueno: C7H8,  99.5% (GC), Panreac (Reag.Ph.Eur,PA-ACS-ISO).
Zinc: en polvo,  99%, Riedel-de Haën.

45
III. MATERIALES Y MÉTODOS

III.2. Materiales lignocelulósicos

En esta Tesis se ha estudiado en detalle la composición química de diversos

materiales lignocelulósicos, incluyendo los procedentes de cultivos forestales
(eucalipto), cultivos energéticos de crecimiento rápido (hierba elefante), y residuos
agrícolas (paja de trigo). Por otro lado, y con vistas a estudiar la evolución de la
lignina durante la deconstrucción química de madera de eucalipto, también se
estudiaron las ligninas residuales de pastas de eucalipto, y sus respectivas ligninas
precipitadas de los licores negros, producidas por los procesos kraft, sosa-AQ y
sosa-O2. Tanto las muestras de pasta como los licores negros fueron suministrados
por la empresa papelera Suzano (Brasil). A continuación se describen brevemente
los diferentes materiales lignocelulósicos analizados en esta Tesis.

III.2.1. Cultivos forestales: Eucalipto

El eucalipto (Eucalyptus) es un género perteneciente a la familia Myrtaceae,

orden Myrtales, clase Magnoliopsida (dicotelidóneas) de la división
Magnoliophyta (angiospermas). Tienen hojas perennes. El eucalipto es un árbol de
crecimiento rápido, utilizado ampliamente para la producción de pasta y papel. En
general, no necesita un suelo con alta fertilidad pero sí con buen drenaje. Existen
muchas especies de eucalipto, siendo las más importantes desde el punto de vista
de la industria papelera las especies de Eucalyptus globulus, E. nitens, E. maidenii,
E. grandis y E. dunnii entre otras. Todas ellas se caracterizan por su bajo contenido
en lignina y compuestos extraíbles, lo que las hace especialmente interesantes para
la producción de pasta y papel. Actualmente E. globulus (Figura 22) es una de las
especies con más interés industrial, porque su madera tiene una densidad alta (550
kg/m3), con fibras largas y un contenido elevado en holocelulosa y pentosas en
comparación con el resto de las especies mencionadas (Sánchez, 2002). Las
características de las maderas de diferentes especies de eucalipto y de sus ligninas
fueron descritas en detalle (Rencoret et al., 2008). También se utilizan híbridos
como es el caso de E. urograndis (E. urophylla × E. grandis), que es la principal
especie de eucalipto cultivada en Brasil. En esta Tesis se estudiaron las maderas de
varias especies híbridas de eucalipto, desarrolladas por el programa brasileño
Genolyptus (Grattaplagia, 2003, 2004). Estas maderas de los eucaliptos híbridos
fueron suministradas por la Universidad de Viçosa (Brasil), y proceden de árboles
de 5 años de los siguientes cruzamientos: E.grandis × E.urophylla (IP), E.urophylla

46
 III. MATERIALES Y MÉTODOS

× E.urophylla (U1×U2), E.grandis × [E.urophylla × E.globulus] (G1×UGL), y

[E.dunni × E.grandis] × E.urophylla (DG×U2).

a b

Figura 22. Ilustración de (a) árboles de la especie Eucalyptus globulus, y (b) astillas de
madera del eucalipto híbrido G1×UGL.

III.2.2. Cultivos agrícolas

III.2.2.1. Cultivos energéticos: Hierba elefante

La hierba elefante (Pennisetum purpureum) es una planta perteneciente a la

familia Poaceae, orden Cyperales, clase Liliopsida (monocotelidóneas) de la
división Magnoliophyta (Figura 23). No se debe confundir con la especie
Miscanthus giganteus, también llamada a veces hierba elefante. Este cultivo, con
origen en África, ha sido introducido en muchas zonas tropicales y subtropicales.
La hierba elefante crece rápidamente, su altura puede alcanzar hasta tres metros, y
llegan a producir hasta 45 toneladas por hectárea anualmente (Woodard y Prine,
1993). Por este motivo, la hierba elefante representa un recurso lignocelulósico
atractivo, tanto para la producción de bioenergía y biomateriales, como para la
producción de papel.

47
III. MATERIALES Y MÉTODOS

a b

Figura 23. Ilustración de (a) la planta, y (b) la caña de hierba elefante (Pennisetum
purpureum).

La caña de la hierba elefante presenta dos partes bien diferenciadas, la corteza y

la médula (Figura 24), que se separaron y se analizaron independientemente. La
corteza representó el 84% de la materia seca inicial, y la médula 16%.

Médula

Corteza

Figura 24. Separación mecánica de la corteza y médula de Pennisetum purpureum

Schumacher.

48
 III. MATERIALES Y MÉTODOS

III.2.2.2. Residuos agrícolas: Paja de trigo

El trigo (Figura 25) es una planta perteneciente a la familia Poaceae, orden

Poales, clase Liliopsida de la división Magnoliophyta. La paja de trigo es un
residuo agrícola producido en grandes cantidades, entre 1.3-1.4 kg de paja por kg
de grano, con una producción mundial en 2011 de alrededor de 680 millones de
toneladas (Sarkar et al., 2012). Existen muchas variedades de trigo. Las especies
que más se cultivan son Triticum aestivum, T. durum y T. compactum (especies con
paja corta y de alto rendimiento). La paja de trigo contiene celulosa (35-45%),
hemicelulosas (20-30%) y lignina (alrededor de 15%), lo que la convierte en una
materia prima atractiva para la producción de bioetanol de segunda generación así
como otros productos de alto valor añadido. Las características de las fibras
dependen mucho de la forma de cosecha, porque la fracción superior del cultivo
contiene más minerales y tiene fibras más cortas. Además, la sección de entrenudos
está más enriquecida en fibras largas en comparación con los nudos y las hojas.
Estas fracciones también se diferencian en su composición química (Jacobs et al.,
2000; Lu y Ralph, 2010a). La muestra de paja de trigo estudiado en esta Tesis,
proviene de un lote cosechado en Junio 2009 y pertenece a la especie T. durum
(variedad Carioca).

Hoja

Entrenudo
Nudo

Paja

Figura 25. Ilustración del cultivo de trigo y la paja procedente del mismo.

49
III. MATERIALES Y MÉTODOS

III.2.3. Pastas de eucalipto

Se estudiaron las ligninas residuales de una serie de pastas (Figura 26),

procedentes de la deconstrucción química en medio alcalino de madera del híbrido
de eucalipto E. grandis × [E. urophylla × E. globulus] (G1×UGL), proporcionada
por la empresa papelera Suzano (Brasil). Las astillas de madera fueron sometidas a
las cocciones kraft, sosa-AQ y sosa-O2 (Tabla 1) en digestores, equipado con un
sistema de reciclaje del licor y de control de temperatura y presión.

Pasta kraft Pasta sosa-AQ Pasta sosa-O2

Figura 26. Pastas de eucalipto G1×UGL con número de kappa 15, obtenidas mediante las
cocciones alcalinas kraft, sosa-AQ y sosa-O2.

Tabla 1. Condiciones experimentales de las cocciones kraft, sosa-AQ y sosa-O2.

Parámetro Condiciones
Madera (kg) 1.0
Concentración activa de álcali, en % NaOH de astilla seca (m/m) variable
Sulfidez (kraft), en % NaOH sobre astilla seca (m/m) 26
Contenido de antraquinona (sosa-AQ), en % de astilla seca (m/m) 0.05
Carga de oxígeno (sosa-O2), en % de astilla seca (m/m) 6
Relación licor:astilla (L:kg) 4:1
Temperatura máxima (ºC) 170
Tiempo transcurrido hasta alcanzar la temperatura máxima (min) 90

50
 III. MATERIALES Y MÉTODOS

Se realizaron dos series de cocciones, una destinada a la producción de pasta de

papel (kraft y sosa-AQ), de la cual se estudiaron pastas con números de kappa 20 y
15. La otra serie se destinó a la producción de bioetanol y biogas (sosa-AQ y sosa-
O2), de la cual se estudiaron pastas con números de kappa 50, 35 y 15. El digestor
(tipo CRS, modelo CPS 010) tenía 2 reactores de cada uno 10 L, acoplados a un
sistema de enfriamiento para terminar la cocción eficazmente. De cada tipo de
cocción se llevaron a cabo 8 series idénticas, pero con diferentes cargas de NaOH,
para determinar la curva de deslignificación mediante análisis de regresión. El
número de kappa de las pastas se determinó conforme la norma TAPPI T236 cm-
85. En las cocciones sosa-O2 se dosificó el oxígeno en tres pasos (a 50, 70 y 110
min desde el inicio de la cocción) para respetar los límites de la presión del digestor
(20 kgf/cm²). Al final de la cocción quedó una pequeña presión de oxígeno en el
reactor. Se enfrió el digestor, y se recuperó el licor (55 ºC) y la pasta, que
posteriormente se lavó, primero con agua caliente (70 ºC) y después con agua a
temperatura ambiente. Las fibras de pasta se prepararon mediante un equipo
separador Hidrapulper, y se sometieron a una selección mediante el paso por platos
perforados de tamaño 0.2 mm de diámetro (Voith laboratory cleaner). El material
retenido se secó, pesó y juntó con la pasta para la serie de cocciones destinadas a la
producción de bioetanol y biogas. Antes de guardar la pasta obtenida en bolsas de
polietileno, se centrifugó para eliminar el agua residual y se secó hasta obtener una
consistencia del 30%.

III.2.4. Ligninas precipitadas de licores de cocción

Además de las ligninas residuales de las pastas, también se estudiaron las

ligninas disueltas en los correspondientes licores de cocción (licores negros). Se
utilizaron los licores negros de las cocciones kraft y sosa-AQ, correspondiendo con
los números de kappa 20 y 15 de sus respectivas pastas, cedidos por la empresa
Suzano. Las ligninas se precipitaron en medio ácido como se describe más
adelante.

III.3. Métodos

III.3.1. Preparación de las muestras lignocelulósicas

Las fibras largas y las astillas grandes se cortaron apropiadamente, y se procedió

a la homogenización del material en bandejas. Se secó el material durante una

51
III. MATERIALES Y MÉTODOS

noche (37-40 ºC). Luego se llevó a cabo la molienda de cuchillas (molino Ika,
modelo MF10) (Figura 27). Primero se cortaron los fragmentos más grandes
durante el paso por el molino sin tamiz. Después el material se pasó por el molino
con un tamiz de 2 mm y posteriormente de 1 mm.

Eucalipto Hierba elefante

G1 UGL corteza

Paja de trigo Hierba elefante

médula

Figura 27. Reducción de tamaño del material lignocelulósico por molienda de cuchillas.

III.3.2. Composición química

A continuación se describen los diferentes métodos analíticos utilizados para

determinar la composición química de los materiales lignocelulósicos
seleccionados en esta Tesis. Se secaron las muestras molidas a temperatura
ambiente hasta obtener una humedad estable. El contenido de cada constituyente
químico se expresó en base a materia seca. Para ello se determinó la humedad (H)
de la muestra en cada ensayo (Figura 28).

52
 III. MATERIALES Y MÉTODOS

f f
m m

Humedad (H, %) = (m – m´) / m × 100

Figura 28. Determinación de la humedad.

III.3.2.1. Determinación del contenido en cenizas

El contenido en cenizas se determinó según la norma Tappi 211 om-85. Para

ello se tararon cresoles de porcelana previamente limpiados con HCl concentrado y
calentados durante 1 h a 575 °C en una mufla (Heraeus, modelo M110). Se pesaron
alrededor de 200 mg de materia seca (m) en los cresoles tarados que se calentaron
durante 6 h a 575 ºC en una mufla (Figura 30). Se determinó el residuo (c) por
gravimetría y se calculó el contenido de cenizas (C) como % de materia seca.

f f
Material seco m c
Pesar m y f Pesar (c + f)

Cenizas (C, %) = c / m × 100

Figura 29. Determinación del contenido en cenizas.

III.3.2.2. Análisis de metales y otros elementos

La materia molida y seca (0.5 g) se sometió a una hidrólisis ácida con 4 mL de

ácido nítrico concentrado durante 15 min en un horno de microondas (Jones y
Case, 1990). Se filtraron los hidrolizados (filtro Whatman n° 2) y se diluyeron
hasta 50 mL de volumen final (Figura 30). Se determinaron las concentraciones
de metales por espectrometría de emisión por plasma (ICP-OES) (espectrómetro
Termo Jarrel Ash, modelo IRIS Advantages).

53
III. MATERIALES Y MÉTODOS

HNO3

+ H2O

Filtrar ICP-OES

Figura 30. Análisis del contenido en metales por ICP-OES.

La Tabla 2 resume los resultados obtenidos de los materiales lignocelulósicos

analizados en esta Tesis.

Tabla 2. Contenido (mg/kg materia seca incial) de metales y elementos en diversos

materiales lignocelulósicos.
Eucalipto Hierba elefante Paja de
Elemento / metal
trigo
IP U1×U2 G1×UGL DG × U2 Corteza Médula
Azufre (S) 60 60 70 60 60 1460 2460
Arsénico (As) 0.18 0.32 0.30 0.35 < 0.01 < 0.01 < 0.01

Bario (Ba) 2.6 1.3 3.1 2.6 1.5 5.7 5.6

Cadmio (Cd) 0.01 0.01 0.00 0.02 < 0.5 1.0 < 0.5
Calcio (Ca) 70 10 130 260 110 720 2110
Cobalto (Co) 0.0 0.1 0.0 0.1 <1 <1 <1
Cobre (Cu) 1.2 1.0 1.6 2.0 3.4 7.8 1.9
Cromo (Cr) 1.5 0.2 0.2 0.2 <1 1.0 1.9
Fósforo (P) 34 34 83 30 1800 2920 220

Hierro (Fe) 14 4 6 4 11 42 113

Litio (Li) 0.2 0.3 0.3 0.3 <1 <1 <1
Magnesio (Mg) 80 60 80 90 250 710 740
Manganeso (Mn) 20 11 18 14 12 24 24
Níquel (Ni) 5.0 0.3 0.4 0.4 <1 <1 1.0

Plomo (Pb) 0.3 0.2 0.1 0.0 <1 <1 <1

Potasio (K) 460 430 370 390 18200 35000 12000
Sodio (Na) 130 110 150 130 100 120 1820
Zinc (Zn) 0.0 0.0 0.3 0.4 12.4 36.8 5.5

54
 III. MATERIALES Y MÉTODOS

III.3.2.3. Determinación del contenido y de la composición química de

los extraíbles lipofílicos

Determinación del contenido en extraíbles lipofílicos

El material lignocelulósico (2 g) con humedad H se sometió a una extracción
continua durante 8 h con acetona en extractores tipo Soxhlet (Figura 31). A
continuación se evaporó el disolvente a sequedad y la cantidad de extracto seco se
determinó por gravimetría (lac). Los extractos lipofílicos se redisolvieron en CHCl3,
y la fracción soluble se traspasó a un vial nuevo previamente tarado. Se secó el
extracto y se determinó el extracto lipofílico soluble en CHCl3 (lcl). La fracción
insoluble representa la fracción más polar del extracto.

Fracción GC & GC/MS

Acetona polar
Cloroformo
+
m, H Extracto
Extracto seco (lcl)
seco (lac)

Extracto lipofílico (Lac/cl, %) = lac/cl / [m × (100-H)/100] × 100

Figura 31. Determinación del contenido y de la composición química de extraíbles

lipofílicos.

Análisis de la composición química del extracto lipofílico

Los compuestos extraíbles lipofílicos se analizaron por GC y GC/MS. Las
columnas capilares seleccionadas para el análisis de lípidos por GC fueron de
longitud corta (5 m), ya que proporcionan una elución y separación adecuada de los
lípidos de alto peso molecular en un corto período de tiempo (20 min) (Gutiérrez et
al., 1998, 2004). Columnas menores de 5 m no son convenientes ya que no
proporcionan la resolución necesaria para análisis cuantitativos. En el caso de los
análisis por GC/MS, los cromatogramas obtenidos tienen que ser reproducibles con
los obtenidos por GC usando columnas capilares de 5 m. No obstante, en el sistema
GC/MS, debido a las condiciones de alto vacío a las que opera, no se pueden usar
columnas tan cortas, por lo que se usaron columnas de 10-15 m. Esta longitud de

55
III. MATERIALES Y MÉTODOS

columna es apropiada para el análisis de lípidos de alto peso molecular por GC/MS
proporcionando resultados en un período de tiempo corto (30 min).

Los análisis cromatográficos de los extractos lipofílicos se llevaron a cabo en un

cromatógrafo de gases Agilent 6890N equipado con un detector de ionización de
llama FID y una columna capilar corta de sílice fundida DB-5HT (J&W, longitud 5
m, diámetro interno 0.25 mm y espesor de película 0.1 µm). El programa de
calentamiento del horno comenzó a 100°C (1 min), seguido de un incremento de
temperatura hasta 350°C (3 min) a 15ºC/min. Las temperaturas del inyector y del
detector se mantuvieron a 300°C y 350°C, respectivamente. El gas portador que se
utilizó fue Helio y la inyección se realizó en modo splitless.

Para el análisis mediante GC/MS se disolvió una fracción del extracto en CHCl3.
El análisis mediante GC/MS, tanto de muestras silanizadas como no silanizadas se
llevó a cabo en un cromatógrafo de gases Varian 3800 acoplado a un detector de
trampa de iones ITD (Varian 4000), usando una columna capilar de sílice fundida
DB-5HT (J&W, longitud 12 m, diámetro interno 0.25 mm y espesor de película 0.1
µm). El horno se calentó de 120°C (1 min) a 380°C (5 min) a 10ºC/min. La línea de
transferencia se mantuvo a 300ºC. La temperatura del inyector se programó de
120ºC (0.1 min) a 380°C con una rampa de 200ºC/min y manteniéndose hasta el
final del análisis. El gas portador utilizado fue Helio. La identidad de cada
componente se determinó por comparación de sus espectros de masas con los
espectros existentes en las librerías (Wiley y NIST) y con espectros publicados
anteriormente, por sus fragmentaciones y, cuando fue posible, por comparación con
patrones (octadecano, ácido palmítico, colest-4-en-3-ona, sitosterol, colesteril
estearato, sitosteril 3β-D-glucopiranósido y glicerolheptadecanoato). Se utilizó una
recta de calibrado, realizada con los patrones anteriormente mencionados. Los
picos cromatográficos se cuantificaron a partir de sus áreas en los cromatogramas y
el factor de dilución utilizada para preparar la disolución en el vial de análisis.

Para el análisis por GC y GC/MS es esencial que los compuestos existentes en la

muestra sean suficientemente volátiles, por lo que es necesario recurrir a métodos
de derivatización, como la silanización, cuando los compuestos a analizar no son
volátiles. La silanización de los grupos hidroxilo de alcoholes, esteroles, etc., se
realizó con BSTFA. Para ello, una vez seca la muestra, se añadió 0.2 mL de
BSTFA y 0.1 mL de piridina. A continuación se calentó a 70ºC durante 2 horas y
se secó con nitrógeno. Posteriormente, se redisolvió en CHCl3 para analizarla por
GC/MS.

56
 III. MATERIALES Y MÉTODOS

III.3.2.4. Determinación del contenido en extraíbles hidrosolubles

Para determinar el contenido en compuestos hidrosolubles se sometió

aproximadamente 2 g de material lignocelulósico (m) previamente extraído con
acetona para eliminar extraíbles lipofílicos a una extracción con agua destilada a
100 °C durante 3 h conforme la norma Tappi T 207 om-88 (Figura 32).

Acetona

100 ºC

m, H
Extracto seco (hs)

Extraíbles hidrosolubles (HS, %) = hs / [m × (100-H)/100 × (100 + Lac)/100] × 100

Figura 32. Determinación del contenido en extraíbles hidrosolubles.

El extracto se concentró mediante rotaevaporación a 95 ºC bajo presión

reducida. El concentrado se traspasó a cresoles de porcelana que se secaron a 100
ºC hasta peso constante (hs). Se calculó el contenido por determinación
gravimétrica como % sobre materia seca inicial (HS). Para ello se utilizó la
humedad (H) en el momento de pesar la muestra (extraída con acetona) y se
recalculó el peso de la muestra con el contenido de extraíbles lipofílicos
determinado previamente (Lac).

III.3.2.5. Determinación del contenido en proteínas

El contenido en proteínas se determinó a partir del contenido en nitrógeno

usando un factor de 6.25, de acuerdo con (Darwill et al., 1980) según:

% proteínas = % nitrógeno × N
N: factor de conversión nitrógeno-proteínas

Se determinó el contenido de nitrógeno en el hidrolizado obtenido de la

digestión ácida (con H2SO4), según el método de Kjehldal (Method No. G-188-97
Rev. 1 (multitest MT7/MT8), BRAN + LUEBBE, Alemania).

57
III. MATERIALES Y MÉTODOS

III.3.2.6. Determinación del contenido en carbohidratos

III.3.2.6.1. Determinación del contenido en holocelulosa

El contenido en holocelulosa se determinó según el método de Browning

(1967). Se dispersó 5 g de fibra, previamente extraída con acetona y agua a 100 ºC
para eliminar extraíbles (Figura 33), en 160 mL de agua destilada en un baño de
75-80 °C en agitación.

Acetona
100 ºC
NaClO2 NaOH
Holocelulosa (hc)
CH3COOH
H2O
m, H α-celulosa (ac)

Holocelulosa (HC, %) = hc / [m × (100-H)/100 × (100 + L + HS)/100] × 100

ac
α-celulosa (AC, %) = ac / [ m × (100-H)/100 × (100 + L + HS)/100] × 100
ac

Figura 33. Determinación del contenido en holocelulosa y α-celulosa.

Se añadió 1.5 g de clorito sódico y 10 gotas de ácido acético glacial. Cada hora se
añadió otra dosis equivalente de clorito sódico y ácido acético y se agitó fuerte
periódicamente. Se añadieron 3 dosis para fibras madereras y 4 dosis para fibras no
madereras. Se filtró la suspensión y se lavó la holocelulosa con acetona y con agua.
Se determinó el residuo seco por gravimetría (hc) y se determinó el contenido
como % de la materia seca inicial.

III.3.2.6.2. Determinación del contenido en α-celulosa

Se determinó el contenido en α-celulosa conforme a la norma Tappi T 203 OS-

61. Para ello se eliminaron las hemicelulosas de la holocelulosa por tratamiento con
álcali. Primero, se añadieron 75 mL de una solución de álcali (17.5 % m/m NaOH)
en varios pasos sucesivos a 3 g de holocelulosa en un baño a 20 °C, manteniendo
una agitación intensa. Primero se agregaron 15 mL (1 min), después 10 mL (0.75
min) y finalmente 10 mL (0.25 min). Se dejó reposar la suspensión durante 1 min,
y enseguida se añadieron 4 dosis más de 10 mL (2.5 min). Después de reposar

58
 III. MATERIALES Y MÉTODOS

durante 30 min, se agregaron 100 mL de agua destilada a la suspensión y se agitó

vigorosamente. Se dejó reposar de nuevo durante 30 min, y se filtró la suspensión
con un filtro de poro n° 2. Se lavó con 25 mL de 8.3 % NaOH y a continuación se
lavó 5 veces más con 50 mL de agua destilada, y finalmente con 400 mL. Después
se lavó el residuo en el filtro mismo con ácido acético (2 N). Finalmente se lavó
con agua destilada hasta obtener un filtrado con pH neutro. Se secaron los filtros y
se determinó el residuo de α-celulosa por gravimetría (ac).

III.3.2.7. Determinación del contenido en lignina total

El contenido en lignina es la suma del contenido de lignina ácido-soluble y

lignina ácido insoluble (lignina Klason), representado esquemáticamente en la
Figura 34.

H2O
Acetona

30 ºC, 1h
100 ºC
72 % H2SO4
4%
m, H H2SO4 120 ºC, 1h

Lavar Insoluble (LIK)

Cenizas (Clik, %)

Nitrógeno Kjehldal (Nlik, %) lik

4h, 100 ºC

Diluir (f) Soluble (LAS)

A, ξ
Espectrofotómetro UV V

LIK (%) = [lik × (100 – Clik – (Nlik× 6.25))/100] / [m × (100-H)/100 × (100+Lac+HS)/100] × 100

Figura 34. Determinación del contenido en lignina Klason y lignina ácido-soluble.

59
III. MATERIALES Y MÉTODOS

III.3.2.7.1. Determinación del contenido en lignina Klason

El contenido de lignina ácido insoluble se determinó por el método Klason,

conforme a la norma Tappi T222 om-88, con algunas modificaciones. Primero se
eliminaron los extraíbles lipofílicos e hidrosolubles del material lignocelulósico.
Luego se determinó el peso (m, mínimo 300 mg) y la humedad (H) de la muestra
que se sometió a una hidrólisis ácida con 72 % H2SO4 (m/m) a 30 °C durante 1 h
(0.01 mL/mg muestra). Después se diluyó la suspensión con agua destilada hasta
obtener una concentración de 4 % H2SO4 y se trató a 120 ºC durante 1 h en un
autoclave. Tras enfriar y dejar reposar, se filtró la suspensión en un filtro de
kitasato de poro nº 3 previamente tarado. El hidrolizado se guardó para el análisis
de la lignina ácido-soluble. Se lavó el filtro que contiene la lignina ácido insoluble
con agua destilada hasta obtener un filtrado de pH neutro. Luego el filtro se secó
durante un mínimo de 4 h a 100 °C para su determinación gravimétrica (lik). El
contenido final en lignina Klason se determinó restando el contenido en cenizas y
proteínas presentes en el residuo insoluble.

III.3.2.7.2. Determinación del contenido en lignina ácido-soluble

Para la determinación del contenido en lignina ácido-soluble (LAS) se midió la

absorbancia del hidrolizado obtenido de la hidrólisis ácida en el análisis de la
lignina Klason en cubetas de cuarzo a 205 nm. Para el blanco se preparó una
solución de 4 % H2SO4. En caso de absorbancias mayores que 1.0 AU se diluyeron
los hidrolizados y el blanco apropiadamente. Se determinó la media de 3
mediciones puntuales de la absorbancia y se calculó el contenido según:

LAS (%) = (A × V × f) / [ξ × (m × (100-H)/100 × (100 + Lac+ HS)/100)] × 100

A (AU/cm) absorbancia a 205 nm m (g) peso de la muestra
V (L) volumen final del hidrolizado H (%) humedad de la muestra
f factor de dilución Lac (%) contenido de extraíbles lipofílicos
ξ (AU/cm g) coeficiente de extinción (110) HS (%) contenido de extraíbles hidrosolubles

III.3.3. Aislamiento de las ligninas de materiales lignocelulósicos

Para un estudio estructural más detallado, se extrajeron las ligninas, tanto de las
fibras madereras como no madereras seleccionadas en esta Tesis, así como de las
pastas y de los licores de cocción procedentes de la deconstrucción alcalina.

60
 III. MATERIALES Y MÉTODOS

III.3.3.1. Aislamiento de lignina de maderas y fibras

Existen varios métodos para aislar la lignina nativa de materiales

lignocelulósicos. El método convencional para aislar y purificar ligninas nativas,
desarrollado por Björkman (1956), es la extracción de madera molida (´´Milled
Wood Lignin´´, MWL) con una mezcla de 1,4-dioxano y agua (disolventes
neutros). Este método de extracción induce muy pocas alteraciones estructurales
(Ikeda et al., 2002), y se considera que la estructura de MWL es representativa de
lignina de madera o fibra. Previamente al aislamiento, se extrajo el material
lignocelulósico (molido por cuchillas) con acetona y agua. Después del secado, se
sometió la madera o fibra (25-30 g) a una molienda rotativa (Figura 35) en un
molino de bolas (Retsch modelos S100 y PM 100) con jarra de ágata. El modelo
S100 se programó a 400 rpm con intervalos de reposo para reducir posibles
sobrecalentamientos del molino y del material lignocelulósico (ciclos de 25 min de
molienda y 20 min de reposo). El modelo PM 100 se programó a 300 rpm con
ciclos de 10 min y 15 min de reposo. Se molieron durante aproximadamente 84 h
totales que corresponde con 47 h efectivas de molienda. La molienda se llevó a
cabo en tolueno para disipar el calor y reducir posibles alteraciones en la estructura
de la lignina. Después de la molienda el material se recuperó por centrifugación (10
min, 4 ºC, 5000 rpm) y se secó apropiadamente para eliminar restos de tolueno.

Eucalipto
Hierba elefante
G1 UGL
corteza

Paja de trigo Hierba elefante

médula

Figura 35. Reducción de tamaño de partículas mediante molienda con bolas.

61
III. MATERIALES Y MÉTODOS

La lignina se extrajo de 40-50 g de este material finamente molido (Figura 36)

con una mezcla dioxano:agua (9:1 v/v) durante 12 h (25 ml/g). El material
insoluble se recuperó por centrifugación (25 min, 4 °C, 11000 rpm) y este proceso
se repitió 2 veces más. El extracto crudo de lignina se concentró y se rotaevaporó a
presión reducida (40 °C). El residuo (lignina cruda) se resuspendió en una mezcla
de ácido acético:agua (9:1 v/v) (20 ml/g). Esta disolución de lignina se precipitó
gota a gota en agua destilada fría en agitación continua (225 ml/g). Se recuperó la
lignina precipitada por centrifugación. Se secó la lignina a temperatura ambiente en
un desecador acoplado a una bomba, y se trituró en un mortero de ágata para
facilitar su disolución en una mezcla de 1,2-dicloroetano:etanol (2:1 v/v) (25 ml/g).
La fracción insoluble contenía carbohidratos residuales, y el sobrenadante, que
contiene la lignina en disolución, se recuperó por centrifugación (5 min, 4 °C, 5000
rpm) y seguidamente se precipitó gota a gota en éter dietílico frío (225 ml/g). La
lignina precipitada se recuperó por centrifugación (10 min). Luego se suspendió en
25 ml de éter dietílico fresco durante 12 h, y se recuperó por centrifugación. Se
repitió el mismo proceso con éter de petróleo. La lignina purificada se aisló por
centrifugación y se secó con nitrógeno. Una vez seco, se conservó en ausencia de
luz y tapado del aire.

Vacío
CH3COOH:agua
(9:1)

Extracto de Mineral
Materia Centrifugar desecador
Extracción lignina cruda
finamente
1,4-dioxano:agua (9:1)
molida
Precipitación
en agua fría dicloroetano: etanol
(2:1)

Precipitación
en éter dietílico
Lignina
Carbohidratos
MWL residuales
Éter de petróleo

Figura 36. Esquema del protocolo del aislamiento de MWL.

62
 III. MATERIALES Y MÉTODOS

III.3.3.2. Aislamiento de lignina residual de pastas alcalinas

La lignina residual de las pastas procedentes de las cocciones kraft, sosa-AQ y

sosa-O2, se aisló mediante acidólisis (Evtuguin et al., 2001). Para ello primero se
secaron las pastas (37-40 ºC). Después de homogenizar bien la pasta, se extrajo con
acetona durante 9 h en extractores tipo Soxhlet. Después del secado, se eliminaron
los compuestos hidrosolubles mediante extracción en agua a 100 ºC. Se extrajo
aproximadamente 15 g de pasta seca, libre de extraíbles, en 2 pasos consecutivos
(13.3 y 10.0 ml/g), con 0.1 M HCl en 1,4-dioxano:agua (82:18 v/v) durante 40 min
a 88-92 ºC en ambos pasos (Figura 37).

Acidólisis (0.1 M HCl en dioxano:agua (82:18 v/v))

Acetona Refrigerante
Argon

Secar 88-92 ºC
100 ºC
Secar Baño

Pasta residual Filtrar y rotaevaporar

2 40 min. + 1 lavar

Centrifugar

15 h, 4 ºC

Lavar con
agua acificada
Liofilizar
n-pentano Precipitar en agua fría
Figura 37. Esquema del aislamiento de lignina residual de pastas alcalinas por acidólisis.

Previo a la extracción se purgó la dispersión de pasta con Argón durante 10 min

para reducir la concentración de aire disuelto y se estableció una atmósfera inerte
en el matraz acoplado herméticamente a un condensador. Después de la extracción
se enfrió hasta 50 ºC y se lavó la pasta con la misma mezcla de 1,4-dioxano:agua
pero sin HCl. El extracto de lavado se dividió entre ambos extractos que se

63
III. MATERIALES Y MÉTODOS

rotaevaporaron por separado a 40 ºC hasta alcanzar una reducción de

aproximadamente 70 % del volumen para evitar medios muy ácidos. Se añadieron
ambos extractos concentrados a 1.5 L de agua destilada a 4 ºC con agitación fuerte.
La lignina se precipitó a 4 ºC en ausencia de luz y después de 12 h se recuperó por
centrifugación (25 min, 9000 rpm, 4 ºC). Antes de liofilizar, se lavó la muestra con
agua destilada acidificada (pH 2.5) y se recuperó por centrifugación. Se pesó la
muestra seca y se calculó el rendimiento de extracción en base al contenido teórico
de lignina en pastas (% lignina ~ 0.15 × número kappa). El rendimiento de
extracción varió entre 26 y 82 % y bajó drásticamente con el número kappa. La
muestra se extrajo con n-pentano durante 8 h en un extractor tipo Soxhlet. Después
de secar con nitrógeno, se guardó la muestra en ausencia de luz en un desecador.

III.3.3.3. Aislamiento de lignina de licores negros

La lignina presente en los licores de cocción se aisló mediante precipitación en

medio ácido a pH 4 (llevado a cabo por la empresa Suzano). Esta lignina
precipitada se separó y se secó a 40 ºC durante 2-4 días, y posteriormente a
temperatura ambiente en un desecador (Figura 38). Una vez alcanzada una
humedad estable, se procedió a su análisis.

Kraft Sosa-AQ

Figura 38. Ligninas precipitadas de los licores negros de las cocciones kraft y sosa-AQ.

Adicionalmente, las ligninas se purificaron mediante reprecipitación. Se disolvió

la lignina disuelta en 0.1 M NaOH durante 30-40 min (pH 12.8-12.9), y se filtraron
las muestras con papel de filtro Whatman nº 4 y se precipitó la lignina con una

64
 III. MATERIALES Y MÉTODOS

solución analítica de HCl concentrado (1 M) con agitación fuerte hasta alcanzar pH

2.5. Las ligninas se recuperaron por centrifugación (25 min, 9000 rpm, 4 ºC), y se
lavaron dos veces más con agua acidificada a pH 2.5 (preparado con un par de
gotas de disolución analítica de HCl concentrado en agua milli-Q). Las ligninas se
recuperaron por centrifugación y se liofilizaron. De esta forma se recuperaron 40-
48 % y 61-81 % de las ligninas kraft y sosa-AQ no purificadas, respectivamente.
Posteriormente se extrajo la lignina con n-pentano durante 8 h en un extractor tipo
Soxhlet.

III.3.4. Técnicas analíticas para el análisis de lignina

A continuación se describen las diferentes técnicas utilizadas en esta Tesis para

caracterizar las ligninas.

III.3.4.1. Py-GC/MS

La Py-GC/MS (´´Pyrolysis-Gas Chromatography/Mass Spectrometry´´) es un

método degradativo, que transforma compuestos complejos no volátiles en una
mezcla de fragmentos volátiles por descomposición térmica en ausencia de oxígeno
(Fullerton y Franich, 1983; Meier y Faix, 1992). La pirólisis se lleva a cabo
habitualmente a temperaturas de 400-800ºC. Se analizan los productos resultantes
de la pirólisis de lignina mediante un cromatógrafo de gases acoplado a un
espectrómetro de masas. La energía térmica presente da lugar a la disociación de
forma controlada de los enlaces éter entre las unidades de lignina, y en cierta
proporción también de los enlaces carbono-carbono, obteniendo una mezcla de
compuestos fenólicos volátiles, en los cuales los grupos metoxilos se mantienen
unidos. Esto permite deducir de qué tipo de unidad de lignina se produce un
determinado compuesto de pirólisis, y determinar la proporción de las unidades H,
G y S. La Py-GC/MS presenta diversas ventajas frente a otros métodos
degradativos. Es una técnica analítica rápida, que necesita poca cantidad de
muestra y una simple preparación de la misma. También presenta ventajas frente a
los métodos clásicos de análisis de la lignina, pues no es necesario aislar la lignina
de la muestra, permitiendo su análisis in situ.

Para el análisis de lignina por Py-GC/MS se utilizaron aproximadamente 100 µg

de muestra (Figura 39). La pirólisis se llevó a cabo en un pirolizador EGA/Py-
3030 D microfurnace pyrolyzer (Frontier Laboratories Ltd.), que operó a 500 ºC

65
III. MATERIALES Y MÉTODOS

durante 10 s, conectado a un equipo de GC/MS (Agilent 7820A) con una columna

capilar DB-1701 (longitud 60 m, diámetro interno 0.25 mm y espesor de película
0.25 µm). La temperatura del horno de GC se programó de 45 °C (4 min), y
aumentó hasta 280 °C (10 min) a 4 °C/min. La columna capilar se acopló a un
analizador de masas Agilent mass selector 5975 N con la ionización de electrones
programado a 70 eV. El flujo del gas portador (He) fue 1 ml/min. Los compuestos
de pirólisis se identificaron con la ayuda de los espectros publicados (Faix et al.
1990; Ralph y Hatfield 1991). Se calcularon las áreas molares, se normalizó a
100%, y se hizo una media entre las repeticiones de la pirólisis. La desviación
estándar fue inferior al 5%.

Madera
Py

In situ

2 4 6 8 10 12 14 16 18 20 22 (min)
GC/MS
MWL MWL

2 4 6 8 10 12 14 16 18 20 22 (min)

carbohidratos lignina

Figura 39. Análisis de MWL y madera por Py-GC/MS.

Los compuestos de pirólisis vinilfenol, vinilguayacacol y vinilsiringol, pueden

producirse tanto de unidades H, G y S de lignina, respectivamente, como de los
ácidos cinámicos p-cumárico, ferúlico y sinapílico (por descarboxilación),
respectivamente. Por este motivo, la presencia de ácidos hidroxicinámicos,
abundantes en plantas herbáceas, influye en la determinación de la composición
H:G:S mediante Py-GC/MS. Para estimar el contenido en ácidos hidroxicinámicos,
se pirolizan muestras metiladas con hidróxido de tetrametilamonio (Py-TMAH). La

66
 III. MATERIALES Y MÉTODOS

metilación de grupos carboxílicos protege estos grupos durante la pirólisis (Figura

40). Para el análisis en presencia de TMAH se añadieron 0.5 µL de TMAH a 100
µg de muestra, y la pirólisis se llevó a cabo según las condiciones anteriormente
descritas.

COOH COOH OMe

OH OH

4-vinilfenol 4-vinilguayacol
COOMe OMe COOMe
OH OH

ácido p-cumárico ácido ferúlico

OMe
OMe OMe
cis/trans 3-(4-metoxifenil)- cis/trans 3-(3,4-dimetoxifenil)-
ácido-3-propenóico metil ester ácido-3-propenóico metil ester

Figura 40. Productos de pirólisis de ácido p-cumárico y ácido ferúlico en ausencia (Py-
GC/MS) y presencia de TMAH (Py-TMAH).

III.3.4.2. 2D-NMR

En el análisis de lignina mediante 1H NMR muchas señales se solapan, que

dificultan el análisis cuantitativo. En los experimentos de 13C NMR la ventana
espectral es más ancha, y no hay acoplamientos espín-espín cuando se desacopla el
protón, dando lugar a señales bien definidas. Sin embargo, algunas señales siguen
solapadas, y se necesita más tiempo de adquisición y muestras en alta
concentración, debido a la baja abundancia natural del núcleo 13C. En la última
década, se ha desarrollado la NMR bidimensional (2D-NMR) y tridimensional
(3D-NMR), en las que se registran correlaciones 1H-13C, entre otras, que resuelven
señales que aparecen solapadas en los espectros unidimensionales (Capanema et
al., 2001; Liitiä et al., 2003; Ralph et al., 2001). La 2D-NMR se considera en la
actualidad la técnica más potente para el análisis de la estructura de la lignina, y a
través de ella se han podido identificar nuevas subestructuras, tales como las
dibenzodioxocinas (Karhunen et al., 1995) o las espirodienonas (Zhang y
Gellerstedt, 2001, 2007).

El análisis de lignina por 2D-NMR mediante experimentos HSQC registra

correlaciones a través de acoplamientos escalares a un enlace entre un protón y el

67
III. MATERIALES Y MÉTODOS

heteronúcleo al que está directamente unido. Los espectros HSQC de la lignina

presentan tres regiones bien diferenciadas: la región alifática, la región alifática
oxigenada y la región aromática. La región alifática oxigenada (Figura 41a) es la
más importante para el estudio de la estructura de la lignina, ya que en esta zona se
encuentran las correlaciones de la mayoría de los enlaces entre las unidades de
lignina. La región aromática (Figura 42b) es la más importante desde el punto de
vista de la composición de la lignina, ya que en esta zona aparecen las
correlaciones de las distintas unidades H, G y S, así como los p-cumaratos y
ferulatos.

a -OMe g
R O HO
g
MeO
5’
6’
1’

b O
g
MeO
6’
HO a
b 4’
3’
2’
1’ O
g-OH
5’
g-OAc HO a
b 4’
3’
2’
OMe
O 1
6 2
5 3
1
OMe 4
6 2 MeO OMe
5 3
a MeO
4
OMe
O

O OMe
b 3’
O
2’ 4’ OMe
a O 1’
6’
5’ O 1

b (g-Oac)
6 2
g a’ OMe 4
3
a’ a b b’ 5
2
1 MeO
5
4
3

OMe
b (g-OH)
MeO 6
a a
MeO g’ O O
6
1 O g b a b
5
4 2 OH
3 HO 1 g
O 6 2
5 3
OMe 4
MeO OMe

b S’2,6 S’’2,6 S2,6 O

HO O

a a
G2 2’ 1

G5 6
1
2
3
S 6
5
2
3
S’
5 4
4
G6 6' MeO OMe MeO OMe

O OH

HO O OH
a a

6
1
2
3
G 6
5
1
2
3
S’’
5
4 4
OMe MeO OMe
O O

Figura 41. Correlaciones 13C-1H en el espectro HSQC de MWL de sisal y sus

correspondientes subestructuras en (a) la región alifática oxigenada, y (b) la región
aromática (del Río et al., 2008).

En experimentos HMBC, el protón puede acoplarse con otro núcleo a distancia

más larga (1-4 enlaces). Al contrario del HSQC, el análisis HMBC da información
sobre carbonos cuaternarios, siendo principalmente carbonos aromáticos y

68
 III. MATERIALES Y MÉTODOS

carbonilos presentes en la estructura de lignina. De esta forma, un espectro HMBC

puede revelar más información sobre otros enlaces entre unidades, y sobre los
diferentes grupos de acilación en posición γ de la cadena lateral.

Aparte de la lignina aislada, también se ha estudiado la lignina in situ por 2D-

NMR. Esta técnica es atractiva, porque permite estudiar el contenido total de
lignina en el material lignocelulósico. En general, se obtienen espectros complejos,
debido a la presencia de las señales de carbohidratos, y la intensidad de las señales
de lignina está limitada a su contenido en el material lignocelulósico. El uso de
espectrómetros de NMR cada vez más potentes, y el uso de criosondas acopladas a
ellos (Figura 42), permite obtener resoluciones más altas y sensibilidades mejores.
Las muestras se analizaron en estado de gel en DMSO-d6. En muestras de geles,
resulta difícil obtener un espectro de HMBC útil, porque se pierde la magnetización
muy rápidamente debido a la atenuación rápida de la relajación en un medio de tal
viscosidad.

Lignina aislada

DMSO-d6

40 mg
Disolución

Material lignocelulósico
(in situ)
DMSO-d6

Criosonda
100 mg Gel

Figura 42. Análisis de lignina aislada y de lignina in situ por 2D-NMR.

Para el análisis de lignina aislada, se disolvieron 40 mg en 750 µL de disolvente

deuterado (DMSO-d6) en tubos de NMR. Para el análisis de lignina in situ se
dispersaron 100 mg material lignocelulósico (previamente extraído y molido) en
750 µL de DMSO-d6. Los espectros de 2D-NMR de ligninas se registraron a 25 ºC
en un espectrómetro Bruker AVANCE III 500 MHz equipado con una criosonda

69
III. MATERIALES Y MÉTODOS

TCI de 5 mm diámetro de geometría inversa con gradientes en el eje z. Los

experimentos 1H-13C HSQC se realizaron con un programa de secuencias de pulsos
estándar de Bruker del tipo adiabático (“hsqcetgpsisp2.2”). Las ventanas
espectrales fueron 5000 Hz y 20843 Hz para las dimensiones 1H y 13C,
respectivamente. Se registraron 2048 puntos complejos con un tiempo de repetición
de 1.5 segundos. El número de registros fue 64 con 256 incrementos en la
dimensión 13C. La constante de acoplamiento 1JCH fue 145 Hz. El procesado de los
puntos en las dimensiones 1H y 13C se realizaron con apodizaciones de tipo
´´matched Gaussian´´ y ´´squared cosine-bell´´, respectivamente. Antes de generar
los espectros en el dominio de frecuencia mediante la transformada de Fourier, las
matrices de datos se completaron con ceros hasta obtener 1024 puntos en la
dimensión 13C. En los experimentos HMBC se utilizaron tiempos de evolución
entre 66 y 80 milisegundos.

Las señales de correlación se calibraron respecto a la señal del disolvente (δC

39.5; δH 2.49), y se asignaron por comparación con la literatura (Balakshin et al.,
2003; Capanema et al., 2001, 2004, 2005; del Río et al., 2008, 2009, 2011; Ibarra et
al., 2007; Liitiä et al., 2003; Martínez et al., 2008; Ralph et al., 1999, 2004, Ralph y
Landucci, 2010; Rencoret et al., 2008, 2009, 2011). La intensidad de las señales en
los espectros HSQC depende del valor de 1JCH, así como del tiempo de relajación
T2 (Zhang y Gellerstedt, 2007). Por ello la integración de las señales se realizó por
separado en cada uno de las regiones del espectro, correspondiente a correlaciones
13 1
C- H químicamente análogas que tienen constantes de acoplamiento 1JCH
similares. En la región alifática oxigenada se determinó la abundancia relativa de
las diferentes enlaces entre las unidades de lignina, presentes en diferentes
subestructuras. Se utilizaron las integrales de las correlaciones Cα−Hα para evitar
posibles interferencias de acoplamientos homonucleares 1H-1H, excepto para las
subestructura Aox (subestructura β-O-4´ con un grupo carbonilo o carboxilo en
posición Cα) y la subestructura I (grupo terminal cinamílico), para las cuales se
usaron las correlaciones Cβ−Hβ y Cγ−Hγ, respectivamente. Para determinar las
abundancias relativas de unidades H, G y S, y de p-cumaratos y ferulatos, se
utilizaron las integrales de las correlaciones C2-H2 en la región aromática.

31
III.3.4.3. P-NMR

La espectroscopía de 31P-NMR permite cuantificar el contenido de grupos

hidroxilos (Jiang, 1997; Granata y Argyropoulus, 1995), mediante derivatización
con un reactivo fosforilante. Durante la fosforilación se produce HCl(g) que forma

70
 III. MATERIALES Y MÉTODOS

una sal con piridina. Esta sal se disuelve en el cloroformo deuterado. De esta forma
se asegura una derivatización cuantitativa de la muestra. Como reactivo fosforilante
se utiliza 2-cloro-l,3,2-dioxafosfolano (reactivo I) o cloro-4,4,5,5-tetrametil-l,3,2-
dioxafosfolano (reactivo II) (Figura 43a). Con el reactivo I se pueden cuantificar
grupos hidroxilos alifáticos primarios y secundarios, e incluso en las
configuraciones eritro y treo, lo que permite estudiar la cadena lateral de la lignina.
En los espectros obtenidos con el reactivo II, se pueden cuantificar el contenido
total de grupos alifáticos de la cadena lateral, grupos fenólicos y grupos carboxilos
(Figura 43b). En general, el contenido se expresa en mmol (milimoles) de grupos
hidroxilos por 100 unidades de fenilpropano. Sin embargo en ligninas degradadas,
no todas las cadenas laterales son propenílicas, por lo cual el resultado se expresa
mejor en mmol de grupos hidroxilos por gramo de lignina pura.

La fosforilación de ligninas se llevó a cabo según el método desarrollado por

Granata y Argyropoulos (1995). Para ello se pesaron aproximadamente 40 mg de
lignina seca con exactitud, y se disolvieron en una mezcla de
dimetilformamida:piridina:cloroformo-d (1:1.70:2.63 v/v). Después se agregaron
0.010 mmol de endo-N-hidroxi-5-norborneno-2,3-dicarboximida (97% pureza)
como estándar interno, junto con 0.032 mmol de cromo(III) de acetilacetonato, que
facilitó la relajación espín-red del núcleo 31P, permitiendo tiempos de adquisición
más cortos. Después la homogenización de la muestra, se añadieron 200 µL de
reactivo II. Las señales de 31P-NMR se registraron con un espectrómetro Bruker
Avance III 500 MHz a 23 ºC, utilizando una secuencia con pulsos de 90º para el
desacoplamiento del núcleo 1H (´´inverse gated decoupled sequence´´), y se
acumularon 512 registros con un tiempo de repetición de 5 segundos entre pulsos
sucesivos. Los desplazamientos químicos se calibraron frente a la señal del
producto de reacción entre agua y el reactivo II (132.2 ppm). El ´´line-broadening
factor´´ fue 2 Hz. Después se aplicó la función de ventana de multiplicación
exponencial y se obtuvo la señal en el dominio de frecuencia mediante la
transformada de Fourier, seguido por corrección automática o manual de la fase y
corrección automática de la línea base. El área de los integrales de las diferentes
señales fue calculada respecto al área del estándar interno que se fijó en 1.

Para separar las señales de grupos hidroxilos en unidades S fenólicas (A) y en

unidades fenólicas G con sustitución en C5 (B), que solapan en la región 144.2 –
141.0 ppm, se calculó el resultado de deconvolución de esta región mediante
software Bruker topspin. Para obtener un buen resultado de deconvolución, la suma
de todas las señales individuales debe aproximarse lo más cerca posible al espectro
experimental. Para la deconvolución se usaron espectros con un ´´line-broadening

71
III. MATERIALES Y MÉTODOS

factor´´ de 8 Hz, se seleccionaron 12-20 señales en la región analizada con una

anchura parecida entre diferentes muestras. Se observaron señales típicamente más
anchas para grupos hidroxilos en estructuras B que en estructuras A. El resultado
de la deconvolución mejoró a medida que se aumentó la cantidad de señales
seleccionadas, pero al mismo tiempo aumentó el tiempo de cálculo. Por eso se
determinó una cantidad óptima de señales a seleccionar en función del tiempo de
cálculo y el cambio en el resultado. Se calculó una desviación estándar mediante
análisis independiente (muestreo, adquisición, procesado y cálculo) de 3 réplicas de
la misma muestra.

O R
R
OH
a g R O g
O P
R
O
HO b R b R
P O
a O R O a O
R R
O
R
Piridina / CDCl3
+ Cl P
R
R R O R R
R
OH O
R = H: reactivo I
Lignina
R = H, OCH3 R = CH3: reactivo II

b Estándar interno
OH fenólico
OH alifático
B

A C OH carboxílico
D

P (ppm)
R R R R R R

A B C C C D
CH3O OCH3 R OCH3 OH OCH3 OCH3
OH OH OH OH OH OH

Figura 43. (a) Fosforilación de grupos hidroxilos en lignina, y (b) ejemplo de un espectro
31
P-NMR con señales de grupos hidroxilos alifáticos, fenólicos y carboxílicos.

72
 III. MATERIALES Y MÉTODOS

III.3.4.4. DFRC

El método de DFRC (´´Derivatization Followed by Reductive Cleavage´´), es un

método degradativo que rompe selectivamente los enlaces éter, y que produce
mayoritariamente monómeros de lignina, y en pequeña proporción también
oligómeros. El método consiste en dos pasos principales (Figura 44a): la
solubilización de la lignina por bromación y acetilación con bromuro de acetilo (i),
y la fragmentación reductora de los enlaces aril-éter en la lignina con polvo de zinc
(ii). El método de DFRC da información cuantitativa de la composición de las
unidades H, G y S, presentes en subestructuras eterificadas (Lu y Ralph, 1997a,
1997b). La inclusión de un patrón interno en el análisis de GC/MS, permite
cuantificar la proporción de unidades eterificadas.

Una de las ventajas de este método degradativo es que durante la reacción, los
grupos de acilación que esterifican el átomo Cγ de la cadena alifática, permanecen
intactos (Ralph y Lu, 1998). Esto permite determinar el grado de acilación en esta
posición, e identificar estos grupos de acilación (acetatos, p-cumaratos y/o p-
hidroxibenzoatos), y cuantificar la abundancia de cada uno. Sin embargo, en el
análisis DFRC los monómeros obtenidos se acetilan (para mejorar su separación
por GC), por lo cual resulta imposible separar aquellas unidades que están
acetiladas de forma natural. Por este motivo, se desarrolló un método modificado
(DFRC´), que emplea reactivos propionilados en vez de acetilados (Figura 44b)
(Ralph y Lu, 1998). De tal forma, se pueden identificar las unidades acetiladas de
forma natural. En comparación con el método de la tioacidólisis, DFRC tiene la
ventaja de usar condiciones más suaves, lo que facilita el análisis en sí, pero
también permite distinguir unidades aciladas. Por este motivo, la DFRC resulta ser
un método muy útil para analizar la lignina de plantas herbáceas (del Río et al.,
2008; Lu y Ralph, 1999).

Los análisis de DFRC se realizaron conforme al método descrito por Lu y Ralph

(1997a, 1997b, 1998, 1999). Primero se diluyó la lignina aislada (10 mg) en una
mezcla de bromuro de acetilo (CH3COBr) y ácido acético (CH3COOH) en una
proporción de 8:92 (v/v) durante 2 h a 50 °C. Se eliminó el disolvente por
rotaevaporación a 50 °C. El residuo seco se disolvió en una mezcla de 1,4-dioxano:
ácido acético:agua (5:4:1 v/v) y se agregó el zinc (50 mg). La suspensión se agitó
adecuadamente durante 40 min a temperatura ambiente. Tras dejar depositar el
zinc, la muestra se traspasó de forma cuantitativa a un embudo de extracción
añadiendo 10 ml de una solución saturada de cloruro amónico (NH4Cl). Se agitó la
mezcla y se ajustó el pH de la fase acuosa a pH < 3 con una solución de 3 % HCl

73
III. MATERIALES Y MÉTODOS

(m/m). Se añadieron 10 ml de diclorometano para la extracción de los compuestos

de la fase acuosa a la fase orgánica. Se agitó adecuadamente y se separó la fase
orgánica. Se extrajo la fase acuosa 2 veces más con 5 ml de diclorometano. La
adición adecuada de sulfato de sodio anhidro (Na2SO4) a la fase orgánica aseguró
la ausencia de humedad. Se decantó la disolución a un matraz para su posterior
rotaevaporación hasta obtener el producto seco.

HO AcO OAc
a HO Br
O O

AcBr Zn

CH3O OCH3
Ac2O OCH3
CH3O OCH3 CH3O
O piridina O
OAc
O
i ii

RO RO OR

HO Br
O O

AcBr Zn

CH3O OCH3 CH3O

Ac2 O
OCH3 CH3O OCH3
O O piridina OAc

i ii

b HO PropO OProp

HO Br
O O

PropBr Zn
Prop2O
CH3O OCH3 CH3O OCH3 CH3O OCH3
O O
piridina OProp
i ii

Ac O Ac O OAc

HO Br
O O

PropBr Zn
Prop2O
CH3O OCH3 CH3O OCH3 CH 3O OCH3
O O piridina OProp
i ii

Figura 44. Análisis de las unidades de lignina unidas por enlaces éter mediante (a) DFRC,
y (b) DFRC modificado (DFRC’) para el análisis de unidades naturalmente acetiladas.

74
 III. MATERIALES Y MÉTODOS

Se acetiló la muestra con 0.2 ml de anhídrido acético y 0.2 ml de piridina en 1.1

ml de diclorometano durante 1 h a temperatura ambiente. Se eliminó el disolvente y
la piridina por co-evaporación con 10 ml de etanol (4 veces consecutivas). El
residuo seco se recogió con diclorometano, y se procedió a su análisis por GC/MS.
En el caso de la DFRC´, el protocolo aplicado fue el mismo pero sustituyendo los
reactivos acetilantes por los respectivos propionilantes según el método descrito
(del Río et al., 2007, 2008; Ralph y Lu, 1998). Los análisis mediante GC/MS se
llevaron a cabo en un equipo GCMS-QP2010plus instrument (Shimadzu Co.,
Japan), con una columna capilar SHR5XLB de 30 m longitud × 0.25 mm diámetro
interno y 0.25 μm espesor de película. Se utilizó 4,4′-etilenobisfenol como patrón
interno. El programa de temperatura empezó a 140 °C (1 min) incrementándose a 3
°C/min hasta 250 °C, después se elevó hasta 280 °C (10 °C/min) y finalmente a
300 ºC (20 ºC/min) que se mantuvo durante 18 min. Las temperaturas del inyector
y de la línea de transferencia se establecieron a 250 ºC y 310 ºC, respectivamente.
El gas portador fue He (1 ml/min.).

III.3.4.5. SEC

La SEC (´´Size Exclusion Chromatography´´) es un método de cromatografía

que utiliza columnas fabricadas de materiales con una determinada distribución de
tamaño de poro para analizar la distribución del peso molecular de polímeros, que
muestran una determinada correlación con sus respectivos tamaños moleculares, en
soluciones polidispersas (Figura 45). A partir de esta distribución, se pueden
calcular varios pesos moleculares, siendo Mw (promedio en peso) y Mn (promedio
en número) los más utilizados para lignina. Estos pesos moleculares son relativos,
respecto a patrones de polímeros sintéticos. Hortling et al. (1990, 1992) y Robert et
al. (1984) utilizaron la SEC para estudiar procesos de deslignificación durante el
pasteado alcalino. La SEC es una técnica interesante para combinar con otras
técnicas analíticas para analizar procesos de condensación. Rahikainen et al. (2013)
trataron paja de trigo y madera de Picea mediante explosión de vapor, y aislaron
sus ligninas. Estos observaron un incremento en los pesos moleculares, y tras
combinar la SEC con 2D-NMR y 31P-NMR, concluyeron que este incremento fue
el resultado de procesos de condensación, que tuvieron lugar simultáneamente con
la despolimerización de la lignina.

Para el análisis de los pesos moleculares de MWLs, se diluyó la muestra (1

mg/mL) en dimetilformamida con 0.1 M LiBr y se analizaron con un equipo HPLC
LC-20A LC system (Shimadzu) equipado con un detector de PDA (SPD-M20A,

75
III. MATERIALES Y MÉTODOS

Shimadzu) operando a 280 nm. Se inyectaron simultáneamente en dos columnas de

gel acopladas en serie (TSK α-M + α-2500) con un flujo de 0.5 mL/min a 40 ºC.
Para la adquisición y el cálculo de los pesos moleculares se empleó un software de
LCsolution, versión 1.25 (Shimadzu). Los pesos moleculares se calcularon
mediante una calibración previa de una serie de estándares de poliestireno.

Las ligninas residuales y las ligninas precipitadas de las cocciones alcalinas se

analizaron en disolución alcalina. Se diluyeron 4 mg de lignina en una solución
analítica de 0.1 M NaOH (1 mg/ml). Después de 12 h se filtró la disolución sobre
una membrana de jeringa de PTFE (0.45 µm, VWR). Las muestras se inyectaron en
un equipo de Waters HPLC system (USA) equipado con un detector de UV
operando a 280 nm y con dos columnas con un gel de poliestireno con diferentes
porosidades (PSS MCX 1000 y 100 000 Å) acoplado en serie (25 ºC). De esta
forma también se pueden detectar fragmentos de alto peso molecular. El flujo del
eluyente fue 0.5 mL/min. Se utilizó el software Empower 2 para la adquisición de
datos y el cálculo de los pesos moleculares. Los pesos moleculares se calcularon
mediante calibración previa de una serie de estándares de sodio de poliestireno
sulfonado (Na-PSS) con un peso molecular de 3420-148500 g/mol.

Retención de
polímeros de
Inyección Separación pequeño tamaño

Flujo disolvente

Polidispersión

Cálculo de peso
Empaquetado molecular
poroso

Concentración Curva cumulativa

detector Curva de distribución

Tiempo de retención

Figura 45. Análisis de la distribución del peso molecular mediante SEC.

76
IV. REFERENCIAS
 IV. REFERENCIAS

Aden, A. Ruth, M., Ibsen, K., Jechura, J., Neeves, K., Sheehan J. y Wallance, B.
(2002). Lignocellulosic biomass to ethanol process design and economics utilizing
concurrent dilute acid prehydrolysis and enzymatic hydrolysis for corn stover,
NREL/TP-510-32438, National Renewable Energy Laboratory, USA.

Adler E. (1977). Lignin Chemistry – Past, Present and Future. Wood Science
Technology, 11, 169-218.

Aitken, I., Cadel, F. y Voillot, C. (1988). Constituants fibreux des pates papiers
et cartons pratique de l'analyse, 1st edition. Pratique de l´analyse, Centre Technique
du Papier, Grenoble.

Allen, L. H. (2000). Pitch control in paper mills. En: Back, E.L. y Allen, L.H.
(Eds.), Pitch Control, Wood resin and Deresination, TAPPI Press, Atlanta, pp.
307-328.

Balakshin, M.Y., Capanema, E.A., Chen, C.-L. and Gracz, H.S. (2003).
Elucidation of the structures of residual and dissolved pine kraft lignins using an
HMQC NMR technique. Journal of Agricultural and Food Chemistry, 51, 6116–
6127.

Barakat, A., Winter, H., Rondeau-Mouro, C., Saake, B., Chabbert, B. y Cathala,
B. (2007). Studies of xylan interactions and cross-linking to synthetic lignins
formed by bulk and end-wise polymerization: a model study of lignin carbohydrate
complex formation. Planta, 226, 267–281.

Baucher, M., Monties, B., van Montagu, M. y Boerjan, W. (1998). Biosynthesis

and genetic engineering of lignin. Critical Reviews in Plant Sciences, 17, 125-197.

Bidlack, J., Malonge, M. y Benson, R. (1992). Molecular structure and

component integration of secondary cell walls in plants. Proceedings of the
Oklahoma Academy of Science, 72, 51-56.

Björkman, A. (1956). Studies on finely divided wood. Part I. Extraction of

lignin with neutral solvents. Svensk Papperstindning, 13, 477-485.

Blanchette, R.A. (1995). Degradation of the lignocellulosic complex in wood.

Canadian Journal of Botany, 73, 999-1010.

79
IV. REFERENCIAS

Bland, D. E. (1966) Colorimetric and chemical identification of lignins in

different parts of Eucalyptus botryoides and their relation to lignification.
Holzforschung, 20, 12-16.

Boerjan, W., Ralph, J. y Baucher, M. (2003). Lignin biosynthesis. Annual

Review of Plant Biology, 54, 519-546.

Bose, S.K., Francis, R.C., Govender, M., Bush, T. y Spark, A. (2009a). Lignin
content versus syringyl to guaiacyl ratio amongst poplars. Bioresource Technology,
100, 1628-1633.

Bose, S. K., Omori, S., Kanungo, D., Francis, R. D., Shin, N.-H. (2009b).
Mechanistic differences between kraft and soda/AQ pulping. Part 1: Results from
wood chips and pulps. Journal of Wood Chemistry and Technology, 29, 214−226.

Bourbonnais, R. y Paice, M.G. (1990). Oxidation of non-phenolic substrates. An

expanded role for laccase in lignin biodegradation. FEBS Letters, 267, 99–102.

BP Statistical Review of World Energy. (2012). www.bp.com/statisticalreview.

Browning, B.L. (1967). Methods of Wood Chemistry, vol. II., Wiley-

Interscience Publishers, New York.

Capanema, E.A., Balakshin, M.Y., Chen, C.-L., Gratzl, J.S. y Gracz, H. (2001).
Structural analysis of residual and technical lignins by 1H–13C correlation 2D NMR
spectroscopy. Holzforschung, 55, 302–308.

Capanema, E.A., Balakshin, M.Y. y Kadla, J.F. (2004). A comprehensive

approach for quantitative lignin characterization by NMR spectroscopy. Journal of
Agricultural and Food Chemistry, 52, 1850–1860.

Capanema, E.A., Balakshin, M.Y. and Kadla, J.F. (2005). Quantitative

characterization of a hardwood milled wood lignin by nuclear magnetic resonance
spectroscopy. Journal of Agricultural and Food Chemistry, 53, 9639–9649.

Carroll, A. y Somerville, C. (2009). Cellulosic biofuels. Annual Review of Plant

Biology, 60, 165-182.

80
 IV. REFERENCIAS

Carvalheiro, F., Duarte, L.C. y Gírio, F.M. (2008). Hemicellulose biorefineries:

a review on biomass pretreatments. Journal of Scientific & Industrial Research, 67,
849-864.

Chang, H. M. y Sarkanen, K. V. (1973). Species variation in lignin. Effect of

species on the rate of kraft delignification. Tappi, 56, 132-134.

Chang, V. S. y Holtzapple , M.T. (2000). Fundamental factors affecting biomass

enzymatic reactivity. Applied Biochemistry and Biotechnology - Part A Enzyme
Engineering and Biotechnology, 84-86, 5-37.

Chen, F. y Dixon, R.A. (2007). Lignin modification improves fermentable sugar

yields for biofuel production. Nature Biotechnology, 25, 759–761.

Chen, F., Fukushima, K. y Yasuda, S. (1998). New pathway on monolignol

biosynthesis. The 48th annual meeting of the Japan wood research society,
Shizuoka, Japan, 397.

Chen, Y., Sharma-Shivappa R.R., et al. (2007). Potential of agricultural residues

and hay for bioethanol production. Applied Biochemistry and Biotechnology, 142
(3), 276-290.

Christierini, M., Ohlsson, A. B., Berglund, T. y Henriksson, G. (2005). Lignin

isolated from primary walls of hybrid aspen cell cultures indicates significant
differences in lignin structure between primary and secondary cell wall. Plant
Physiology and Biochemistry, 43, 777-785.

Confederation of European Paper Industries. CEPI Sustainability Report 2011.

www.cepi.org.

Côté, W. A. (1967). Wood Ultrastructure. University of Washington Press,

Seattle.

Darwill, A., McNeil, M., Albersheim P. y Delmer D. (1980). The primary cell
walls of flowering plants. En: Tolbert, N. (Ed.), The Biochemistry of Plants,
Academic Press, New York, pp. 91-162.

81
IV. REFERENCIAS

Davin, L. B. y Lewis, N. G. (2000). Dirigent proteins and dirigent sites explain

the mystery of specificity of radical precursor coupling in lignan and lignin
biosynthesis. Plant Physiology, 123, 453-461.

del Río, J. C., Gutiérrez, A. y Martínez, A. T. (2004). Identifying acetylated

lignin units in non-wood fibers using pyrolysis-gas chromatography/mass
spectrometry. Rapid Communications in Mass Spectrometry, 18, 1181-1185.

del Río, J.C., Gutiérrez, A., Hernando, M., Landi, P., Romero, J. y Martinez,
A.T. (2005). Determining the influence of eucalypt lignin composition in paper
pulp yielding using Py-GC/MS. Journal of Analytical and Applied Pyrolisis, 74,
10-115.

del Río, J.C., Marques, G., Rencoret, J., Martinez, A.T. y Gutiérrez, A. (2007).
Occurence of natural acetylated lignin units. Journal of Agricultural and Food
Chemstry, 55, 5461-5468.

del Río J.C., Rencoret, J., Marques, G., Gutiérrez, A., Ibarra, D., Santos, J.I.,
Jiménez-Barbero, J., Zhang, L. y Martínez, A.T. (2008). Highly acylated
(acetylated and/or p-coumaroylated) native lignins from diverse herbaceous plants.
Journal of Agricultural and Food Chemistry, 56, 9525–9534.

del Río, J.C., Rencoret, J., Marques, G., Gutiérrez, A., Ibarra, D., Santos, J.I.,
Jiménez-Barbero, J., Martínez, A.T. y Gutiérrez, A. (2009). Structural
characterization of the lignin from jute (Corchorus capsularis) fibers. Journal of
Agricultural and Food Chemistry, 57, 10271–10281.

, . ., , ., , ., Nieto, L., -Barbero, J. y

Martínez, A. T. (2011). Structural characterization of guaiacyl-rich lignins in flax
(Linum usitatissimum) fibers and shives. Journal of Agricultural and Food
Chemistry, 59, 11088−11099.

Ding, S.-Y. y Himmel, M.E. (2006). The maize primary cell wall microfibril: a
new model derived from direct visualization. Journal of Agricultural and Food
Chemistry, 54, 597-606.

82
 IV. REFERENCIAS

Evtuguin, D.V., Neto, C.P., Silva, A.M.S., Domingues, P.M., Amado, F.M.L.,
Robert, D. y Faix, O. (2001). Comprehensive study on the chemical structure of
dioxane lignin from plantation Eucalyptus globulus wood. Journal of Agricultural
and Food Chemistry, 49, 4252-4261.

Fengel, D. (1989). Chemistry and morphology of quebracho colorado wood.

International Symposium on Wood and Pulping Chemistry, Raleigh, 1-3.

Fengel, D. y Wegener, G. (1984). Wood: Chemistry, Ultrastructure, Reactions.

De Gruyter, Berlin.

Fergus, B. J. y Goring, D. A. I. (1970). The location of guaiacyl and syringyl

lignins in birch xylem tissue. Holzforschung, 24, 113-117.

Freudenberg, K. y Lehmann, B. (1963). Radioactive isotopes and lignin. X.

Investigation of a lignin preparation labeled with 14C. Chemische Berichte, 96,
1850-1854.

Freudenberg, K. y Neish, A. C. (1968). Constitution and biosynthesis of lignin.

Springer-Verlag, New York.

Froass, P. M., Ragauskas, A. J.. Jiang, J.-E. (1996). Chemical structure of

residual lignin from kraft pulp. Journal of Wood Chemistry and Technology, 16,
347−365.

Froass, P. M.; Ragauskas, A. J.; Jiang, J.-E. (1998). Nuclear magnetic resonance
studies. 4. Analysis of residual lignin after kraft pulping. Industrial Engineering
and Chemistry Research, 37, 3388−3394.

Fullerton, T.J. y Franich, R.A. (1983). Lignin analysis by pyrolysis-GC-MS.

Holzforschung, 37, 267-269.

Fukushima, K. y Terashima, N. (1991). Heterogeneity in formation of lignin.

XIV. Formation and structure of lignin in differentiating xylem of Ginkgo biloba.
Holzforschung, 45, 87-94.

García Hortal. (2007). Fibras Papeleras. Ediciones UPC (Universitat Politècnica

de Catalunya), Terrassa (Spain).

83
IV. REFERENCIAS

García Hortal, J. A. y Colom, J. F. (1992). El proceso al sulfato. Vol. I.

Universitat Politècnica de Catalunya, Terrassa (Spain).

Gellerstedt, G., Lindfors, E.L., Lapierre, C. y B. Monties, B. (1984). Structural

changes in lignin during kraft pulping. Part 2. Characterization by acidolysis.
Svensk Papperstidn, 9, 61-67.

Gellerstedt, G., Li, J., Eide, I., Kleinert, M. y Barth, T. (2008). Chemical
structures present in biofuel obtained from lignin. Energy Fuels, 22, 4240-4244.

Gierer, J. (1982). The chemistry of delignification. A general concept.

Holzforschung, 36, 43−51.

Gierer, J. (1985). Chemistry of delignification. I. General concept and reactions

during pulping. Wood Science Technology, 19, 289−312.

Gierer, J. (1997). Formation and involvement of superoxide (O2.-/HO2.) and

hydroxyl (OH.) radicals in TCF bleaching processes. A review. Holzforschung, 51
(1), 34.

Gómez, L. D., Steele-King C. G. y McQueen-Mason S. J. (2008). Sustainable

liquid biofuels from biomass: the writing's on the walls. New Phytologist, 178, 473-
485.

González-Vila, F. J., Almendros, G., del Río, J. C., Martín, F., Gutiérrez, A. y
Romero, J. (1999). Ease of delignification assessment of different Eucalyptus wood
species by pyrolysis (TMAH)-GC/MS and CP/MAS 13C-NMR spectrometry.
Journal of Analytical and Applied Pyrolysis, 49, 295-305.

Grabber, J.H., Hatfield R.D., Ralph J., Zon J. y Amrheins, N. (1995). Ferulate
crosslinking in cell walls isolated from maize cell suspensions. Phytochemistry, 40,
1077-1082.

Granata, A. y Argyropoulus, D.S. (1995). 2-Chloro-4,4,5,5-tetramethyl-1,3,2-

dioxaphospholane, a reagent for the accurate determination of the uncondensed
and condensed phenolic moieties in lignins. Journal of Agricultural and Food
Chemistry, 43, 1538-1544.

84
 IV. REFERENCIAS

Grattaplagia, D. (2003). Genolyptus. En: Borém, A., Gindice, M. y Sediyama,

T. (Eds.), Mehoramento Genômico, Editora de UFV, Viçosa, MG, Brazil, pp. 51–
72.

Grattaplagia, D. (2004). Integrating genomics into Eucalyptus breeding.

Genetics and Molecular Research, 3, 369–379.

Gutiérrez, A., del Río, J. C., González-Vila, F. J. y Martín, F. (1998). Analysis

of lipophilic extractives from wood and pitch deposits by solid-phase extraction
and gas chromatography. Journal of Chromatography A, 823, 449-455.

Gutiérrez, A., del Río, J. C., González-Vila, F. J. y Martín, F. (1999). Chemical

composition of lipophilic extractives from Eucalyptus globulus Labill. wood.
Holzforschung, 53, 481-486.

Gutiérrez, A., del Río, J. C., Martínez, M. J. y Martínez, A. T. (2001). The

biotechnological control of pitch in paper pulp manufacturing. Trends in
Biotechnology, 19, 340-348.

Gutiérrez, A., del Río, J. C. y Martínez, A. T. (2004). Chemical analysis and

biological removal of wood lipids forming pitch deposits in paper pulp
manufacturing. En: Spencer, J. F. T. (Ed.), Protocols in Environmental
Microbiology, Humana Press, Totowa, USA.

Gutiérrez, A., Rencoret, J., Cadena, E.M., Rico, Barth, D., del Río, J.C.,
Martínez, A.T. (2012). Demonstration of laccase-based removal of lignin from
wood and non-wood plant feedstocks. Bioresource Technology, 119, 114–122.

Hardell, H. L., Leary, G. J., Stoll, M. y Westermark, U. (1980a). Variations in

lignin structure in defined morphological parts of birch. Svensk Papperstidn, 83,
71-74.

Hardell, H. L., Leary, G. J., Stoll, M. y Westermark, U. (1980b). Variations in

lignin structure in defined morphological parts of spruce. Svensk Papperstidn, 83,
44-49.

Hatakka, A. (2001). Biodegradation of lignin. En: Hofrichter, M. y Steinbüchel,

A. (Eds.), Biopolymer, Biology, Chemistry, Biotechnology, Applications. Vol. I.
Lignin, Humic Substances and Coal, Wiley-WCH, 129-180.

85
IV. REFERENCIAS

Hatfield, R.D., Ralph, J. y Grabber, J.H. (1999). Review cell wall cross-linking
by ferulates and diferulates in grasses. Journal of the Science of Food and
Agriculture, 79, 403-407.

Hatfield, R., Marita, J.M., Frost, K., Grabber, J.H., Ralph, J., Lu, F.C. et al.
(2009). Grass lignin acylation: p-coumaroyl transferase activity and cell
characteristerics of C3 and C4 grasses. Planta, 229, 1253–1267.

Hillis, W.E. (1962). Wood Extractives. Academic Press, London.

Hillis, W. E. y Sumimoto, M. (1989). Effect of extractives on pulping. En:

Rowe, J.W. (Ed.), Natural Products of Woody Plants. II. Springer-Verlag, Berlin,
pp. 880-920.

Himmelsbach, D.S. (1993). Structure of forage cell walls (session synopsis). En:
Jung, H.G., Buxton, D.R., Hartfield, R.D. y Ralph, J. (Eds.), Forage Cell Wall
Structure and Digestibility, American Society of Agronomy Inc., Madison, WI, 271-
283.

Hortling, B., Ranua M. y Sundquist, J. (1990). Investigation of residual lignin in

chemical pulps. Part I. Enzymatic hydrolysis of the pulps and fractionation of the
products. Nordic Pulp Paper Research Journal, 5 (1), 33.

Hortling, B., Turunen, E.. y Sundquist, J. (1992). Investigation of residual lignin

in chemical pulps. Part II. Purification and characterization of residual lignin after
enzymatic hydrolysis of pulps. Nordic Pulp Paper Research Journal, 7 (3), 144.

Hu, W.J., Harding, S.A., Lung, J., Popko, J.L. y Ralph, J. (1999). Repression of
lignin biosynthesis promotes cellulose accumulation and growth in transgenic trees.
Nature Biotechnology, 17, 808–812.

Huntley, S.K., Ellis, D., Gilbert, M., Chapple, C. y Mansfield, S.D. (2003).
Significant increases in pulping efficiency in C4H-F5H-transformed poplars:
improved chemical savings and reduced environmental toxins. Journal of
Agricultural and Food Chemstry, 51, 6178-6183.

Hückelhoven, R. (2007). Cell wall–associated mechanisms of disease resistance

and susceptibility. Annual Review Phytopathology, 45, 101-127.

86
 IV. REFERENCIAS

Ibarra, D., Chávez, M.I., Rencoret, J., del Río, J.C., Gutiérrez, A., Romero, J.,
Camarero, S., Martinez, M.J., Jiménez-Barbero, J. y Martínez, A.T. (2007). Lignin
modification during Eucalyptus globulus kraft pulping followed by totally chlorine
free bleaching: a two-dimensional nuclear magnetic resonance Fourier transform
infrared, and pyrolysis-gas chromatography/mass spectrometry study. Journal of
Agriculture and Food Chemistry, 55, 3477–3499.

Ikeda, T., Holtman, K., Kadla, J.F., Chang, H. y Jameel, H. (2002). Studies on
the effect of ball milling on lignin structure using a modified DFRC method.
Journal of Agricultural and Food Chemistry, 50, 129–135.

Isroi, R.M., Syamsiak, S., Niklasson, C., Cahyanto, M.N., Lundquist, K. y

Taherzadeh, M.J. (2011). Biological pretreatment of lignoceluloses with white-rot
fungi and its applications: a review. Bioresources, 6 (4), 1-36.

Jacobs, R.S., Pan, W.L., Fuller, W.S. y McKean, W.T. (2000). Wheat straw:
within-plant variation in chemical composition and fiber properties. En: TAPPI
pulping/process and product quality conference 2000, 528-539.

Jiang, Z.H. (1997). Advances and applications of quantitative 31P-NMR for the
structural eludication of lignin. PhD tesis, Department of Chemistry, McGill
University, Montréal, Québec, Canada.

Jones, J. B. y Case, V. W. (1990). Sampling, handling, and analyzing plant

tissue samples. Soil Testing and Plant Analysis, 389-427.

Kanungo, D., Francis, R. C. y Shin, N.-H. (2009). Mechanistic differences

between kraft and soda/AQ pulping. Part 2: Results from lignin model compounds.
Journal of Wood Chemistry Technology, 29, 227−240.

Karhunen, P., Rummakko, P., Sipila, J., Brunow, G. y Kilpeläinen, I. (1995).

Dibenzodioxocins - a novel type of linkage in softwood lignins. Tetrahedron
Letters, 36, 169-170.

Kim, S. y Dale, B.E. (2004). Global potential bioethanol production from

wasted crops and crop residues. Biomass and Bioenergy, 26, 361-375.

87
IV. REFERENCIAS

Kleinert, M., Gasson JR y Barth, T. (2009). Optimizing solvolysis conditions for

integrated depolymerisation and hydrodeoxygenation of lignin to produce liquid
biofuel. Journal of Analytical and Applied Pyrolysis, 85, 108-117.

Klemm, D., Heublein, B., Fink, H.P. y Bohn, A. (2005). Cellulose: fascinating
biopolymer and sustainable raw material. Angewandte Chemie International
Edition, 44, 3358–3393.

Knauf, M.y Moniruzzaman, M. (2004). Lignocellulosic biomass processing: a

perspective. International Sugar Journal, 106, 147–150.

Kootstra, A.M.J, Beeftink, H.H, Scott, E.L. y Sanders, J.P.M. (2009).

Comparison of dilute mineral and organic acid pretreatment for enzymatic
hydrolysis of wheat straw. Biochemical Engineering Journal, 46 (2), 126 - 131.

Koshijima, T.y Watanabe, T. (2003). Association between lignin and

carbohydrates in wood and other plant tissues. Wood Science and Technology, 37
(5), 451.

Krassig, H. y Schurz, J. (2002). Ullmann's Encyclopedia of Industrial

Chemistry, Sixth edition, Wiley-VCH, Weinheim, Alemania,

Lam, T.B.T., Kadoya K. y Iiyama K. (2001). Bonding of hydroxycinnamic acids

to lignin: ferulic and p-coumaric acids are predominantly linked at the benzyl
position of lignin, not the beta-position, in grass cell walls. Phytochemistry, 57,
987–992.

Lapierre, C., Jouin, D. y Monties, B. (1989). On the molecular origin of the

alkali solubility of Gramineae lignins. Phytochemistry, 28, 1401–1402.

Lewis, N. G. (1999). A 20th century roller coaster ride: A short account of

lignification. Current Opinion in Plant Biology, 2, 153-162.

Liitiä, T.M., Maunu, S.L., Hortling, B., Toikka, M. y Kilpeläinen, I. (2003).

Analysis of technical lignins by two- and three-dimensional NMR spectroscopy.
Journal of Agricultural and Food Chemistry, 51, 2136–2143.

88
 IV. REFERENCIAS

Lora, J.H. y Glasser, W.G. (2002). Recent industrial applications of lignin: a

sustainable alternative to nonrenewable materials. Journal of Polymer
Environmental, 10, 39-48.

Lu, F.C. y Ralph, J. (1997a). Derivatization followed by reductive cleavage

(DFRC method), a new method for lignin analysis: protocol for analysis of DFRC
monomers. Journal of Agricultural and Food Chemistry, 45, 2590-2592.

Lu, F.C. y Ralph, J. (1997b). DFRC method for lignin analysis. 1. New method
for beta aryl ether cleavage : lignin model studies. Journal of Agricultural and
Food Chemistry, 45, 4655-4660.

Lu, F.C. y Ralph, J. (1998). The DFRC method for lignin analysis. 2.
Monomers from isolated lignin. J. Agric. Food Chem., 46, 547−552.

Lu, F.C. y Ralph, J. (1999). Detection and determination of p-coumaroylated

units in lignins. Journal of Agricultural and Food Chemistry, 47, 1988-1992.

Lu, F.C. y Ralph, J. (2002). Preliminary evidence for sinapyl acetate as a lignin
monomer in kenaf. Chemical Communications, 90-91.

Lu, F.C. y Ralph, J. (2008). N v ahy f a s s v f β–β-

coupling reactions involving sinapyl acetate in kenaf lignins. Organic and
Biomolecular Chemistry, 6, 3681–3694.

Lu, F.C. y Ralph, J. (2010a). Chapter 2 – Structure, Ultrastructure, and

Chemical Composition, En: Cereal Straw as a Resource for Sustainable
Biomaterials and Biofuels, Elsevier, Amsterdam, pp. 169–207.

Lu F.C. y Ralph, J. (2010b). Chapter 6 – Lignin, En: Cereal Straw as a

Resource for Sustainable Biomaterials and Biofuels, Elsevier, Amsterdam, pp.
169–207

Ludwig, C.H. (1971). Magnetic resonance spectra. En: Sarkanen, K.V. y

Ludwig, C.H., Lignins: Occurrence, Formation, Structure and Reactions, Wiley
Interscience, New York, 299-344.

89
IV. REFERENCIAS

Martínez, A.T., Rencoret, J., Margues, G., Gutiérrez, A., Ibarra, D., Jiménez-
Barbero, J., et al. (2008). Monolignol acylation and lignin structure in some
nonwoody plants: a 2D NMR study. Phytochemistry, 69, 2831-2843.

McMillan, J. D. (1994). Pretreatment of lignocellulosic biomass. En: Himmel,

M. E., Baker, J. O. y Overend, R. P. (Eds.), Enzymatic Conversion of Biomass for
Fuels Production, American Chemical Society, Washington DC, USA.

Meier, D. y Faix, O. (1992). Pyrolisis-gas chromatography-mass spectrometry.

En: Lin, S.Y. y Dence, C.W. (Eds.), Methods in Lignin Chemistry, Springer-
Verlag, Berlin, 177-199.

Meyermans, H., Morreel, K., Lapierre, C., Pollet, B., De Bruyn, A., Busson, R.,
et al. (2000). Modifications in lignin and accumulation of phenolic glucosides in
poplar xylem upon down-regulation of caffeoyl-coenzyme A O-methyltransferase,
an enzyme involved in lignin biosynthesis. Journal of Biological Chemistry, 275,
36899–36909.

Morreel, K., Ralph, J., Lu, F., Goeminne, G., Busson, R., Herdewijn, P.,
Goeman, J.L., Van der Eycken, J., Boerjan, W. y Messens, E. (2004). Phenolic
profiling of caffeic acid O-methyltransferase-deficient poplars reveals novel
benzodioxane oligolignols. Plant Physiology, 136, 4023-4036.

Mosier, N., Wyman, C., Dale, B., Elander, R., Lee, Y. Y., Holtzapple, M., et al.
(2005). Features of promising technologies for pretreatment of lignocellulosic
biomass. Bioresource Technology, 96, 673–686.

Mussatto, S.I. y Roberto, I.C. (2004). Alternatives for detoxification of diluted-

acid lignocellulosic hydrolyzates for use in fermentative processes: a review.
Bioresource Technology, 93, 1–10.

Muurinen, E. (2000). Organosolv pulping. A review and distillation study

related to peroxiacid pulping, Doctoral thesis, University of Oulu, Finlandia.

Pan, X. J., Gilkes, N., Kadla, J., Pye, K., Saka, S., Ehara, K., et al. (2006).
Bioconversion of hybrid poplar to ethanol and co-products using an organsolv
fractionation process: Optimization of process yields. Biotechnology and
Bioengineering, 94, 851-861.

90
 IV. REFERENCIAS

Rahikainen, J.L., Martin-Sampedro, R., Heikkinen, H., Rovio, S., Marjamaa, K.,
Tamminen, T., Rojas, O.J. y Kruus, K. (2013). Inhibitory effect of lignin during
cellulose bioconversion: the effect of lignin chemistry on non-productive enzyme
adsorption. Bioresource Technology, 133, 270–278.

Ragauskas, A. J., Williams, C. K., Davison, B. H., Britovsek, G., Cairney, J.,
Eckert, C. A., Frederick, W. J., Hallett, J. P., Leak, D. J.; Liotta, C. L., Mielenz,
J.R., Murphy, R., Templer, R., Tschaplinski., T. (2006). The path forward for
biofuels and biomaterials. Science, 311, 484−489.

Ralph, J. y Hatfield, R. D. (1991). Pyrolysis-GC-MS characterization of forage

materials. Journal of Agricultural and Food Chemistry, 39, 1426-1437.

Ralph, J., Helm R.F., Quideau S. y Hatfield, R.D. (1992). Lignin-feruloyl ester
crosslinks in grasses. Part 1. Incorporation of feruloyl esters into coniferyl alcohol
dehydrogenation polymers. Journal of the Chemical Society, Perkin Transactions,
1, 2961-2969.

Ralph J. y Helm R.F. (1993). Lignin/hydroxycinnamic acid/polysaccharide

complexes: synthetic models for regiochemical characterization. En: Jung H.G.,
Buxton D.R., Hatfield R.D. y Ralph J. (Eds.), Forage Cell Wall Structure and
Digestibility, ASA-ACSSA-SSSA, Madison, pp. 201–246.

Ralph, J., Hatfield, R.D., Quideau, S., Helm, R.F., Grabber, J.H. y Jung, H.J.G.
(1994a). Pathway of p-coumaric acid incorporation into maize lignin as revealed by
NMR. Journal of American Chemistry Society, 116, 9448–9456.

Ralph, J., Quideau, S., Grabber, J.H. y Hatfield, R.D. (1994b). Identification and
synthesis of new ferulic acid dehydrodimers present in grass cell walls. Journal of
the Chemical Society, Perkin Transactions, 1, 3485-3498.

Ralph, J., Grabber, J.H. y Hatfield, R.D. (1995). Lignin-ferulate crosslinks in

grasses: active incorporation of ferulate polysaccharide esters into ryegrass lignins.
Carbohydrate Research, 275, 167-178.

Ralph, J., MacKay, J. J., Hatfield, R. D., Omalley, D. M., Whetten, R. W. y

Sederoff, R. R. (1997). Abnormal lignin in a loblolly pine mutant. Science, 277,
235-239.

91
IV. REFERENCIAS

Ralph, J. y Lu, F.C. (1998). The DFRC method for lignin analysis. A simple
modification for identifying natural acetates on lignins. Journal of Agricultural and
Food Chemistry, 46, 4616–4619.

Ralph, J., Marita, J. M., Ralph, S. A., Hatfield, R. D., Lu, F., Ede, R. M., Peng,
J., Quideau, S., Helm, R. F., Grabber, J. H., Kim, H., Jimenez-Monteon, G., Zhang,
Y., Jung, H. -J. G., Landucci, L. L., MacKay, J. J., Sederoff, R. R., Chapple, C. y
Boudet, A. M. (1999). Solution-state NMR of lignin. En: Argyropoulos, D.S. (Ed.),
Advances in Lignocellulosics Characterization, Tappi Press, a a, pp 55−108.

Ralph, J., Lapierre, C., Lu, F. C., Marita, J. M., Pilate, G., van Doorsselaere, J.,
Boerjan, W. y Jouanin, L. (2001). NMR evidence for benzodioxane structures
resulting from incorporation of 5-hydroxyconiferyl alcohol into lignins of O-
methyltransferase- deficient poplars. Journal of Agricultural and Food Chemistry,
49, 86-91.

Ralph, S. A., Ralph, J. y Landucci, L. (2004). NMR database of lignin and cell
wall model compounds, US Forest Prod. Lab., One Gifford Pinchot Dr., Madison,
Wisconsin, USA.

Ralph, J., Lundquist, K., Brunow, G., Lu, F.C., Kim, H., Schatz, P.F. et al.
(2004b). Lignins: natural polymers from oxidative coupling of 4-
hydroxyphenylpropanoids. Phytochemistry Review, 3, 29-60.

Ralph, J., Brunow, G. y Boerjan, W. (2007). Lignins. En: Rose, F., Osborne, K.
(Eds.), Encyclopedia of Life Sciences, John Wiley & Sons, Ltd., Chichester, UK,
pp. 1-10.

Ralph, J., Hoon, K., Fachuang, L., Grabber, J.H., Leplé J.-C., Berrio-Sierra, J.,
Derikvand, M.M., Jouanin, L., Boerjan, W. y Lapierre, C. (2008a). Identification of
the structure and origin of a thioacidolysis marker compound for ferulic acid
incorporation into angiosperm lignins (and an indicator for cinnamoyl CoA
reductase deficiency). The Plant Journal, 53, 368–379.

92
 IV. REFERENCIAS

Ralph J., Brunow G., Harris P.J., Dixon R.A., Schatz P.F., Boerjan W. (2008b).
Lignification: are lignins biosynthesized via simple combinatorial chemistry or via
proteinaceous control and template replication?. En: Daayf, F., El Hadrami, A.,
Adam, L. y Balance, G.M. (Eds.), Recent Advances in Polyphenol Research, Wiley-
Blackwell Publishing, Oxford, pp. 36–66.

Ralph, J. y Landucci, L.L. (2010). NMR of lignins. En: Heitner, C., Dimmel,
D.R., Schmidt, J.A. (Eds.), Lignin and Lignans: Advances in Chemistry. CRC
Press (Taylor & Francis Group), Boca Raton, pp. 137–234.

Rencoret, J., Marques, G., Gutiérrez, A., Ibarra, D., Li, J., Gellerstedt, G.,
Santos, J.I., Jiménez-Barbero, J., Martínez, A.T. y del Río, J.C. (2008). Structural
characterization of milled wood lignin from different eucalypt species.
Holzforschung, 62, 514–526.

, ., a s, ., , ., N , ., -Barbero, J.,
Martínez, A. T. y del Río, J. C. (2009). Isolation and structural characterization of
the milled wood lignin from Paulownia fortunei wood. Industrial Crops and
Products, 30, 137−143.

Rencoret, J., Gutiérrez, A., Nieto, L., Jiménez-Barbero, J., Faulds, C.B., Kim,
H., Ralph, J., Martínez, A.T. y del Río, J.C. (2011). Lignin composition and
structure in young versus adult Eucalyptus globulus plants. Plant Physiology, 155,
667–682.

Re , ., a ph ., a s, ., , ., a , . .y , . .
(2013). Structural characterization of lignin isolated from coconut (Cocos nucifera)
coir fibers. Journal of Agricultural and Food Chemistry, 61, 2434−2445.

Robert, D.R., Bardet, M., Gellerstedt, G.y Lindfors, E.-L. (1984). Structural
changes in lignin during kraft cooking. Part 3. On the structure of dissolved lignins,
Journal of Wood Chemistry and Technology, 4, 239–263.

Rowe, J.W. (1989). Natural products of woody plants. I and II. Chemicals
extraneous to the lignocellulosic cell wall, Springer-Verlag, Berlin, 250-273.

93
IV. REFERENCIAS

Saka, S. y Goring, D.A.I. (1985). Localization of lignins in wood cell walls. En:
Higuchi, T. (Ed.), Biosynthesis and Biodegradation of Wood Components,
Academic Press, Orlando, 51–62.

Sánchez, J.R. (2002). Panel de tecnologia, pasta y papel. En: Grupo de

Investigacion AF-4 – Universidad de Vigo Vol.1, Socioeconomia, Patologia,
Tecnologia y Sostenibilidad del Eucalipto, Universidad de Vigo, Graficas
Diumaró, Vigo, España, pp. 187-192.

Sannigrahi, P., Pu, Y. y Ragauskas, A. (2010). Cellulosic biorefineries -

unleashing lignin opportunities. Current Opinion in Environmental Sustainability,
2, 383–393.

Santos, A., Rodríguez, F., Gilarranz, M. A., Moreno, D. y García-Ochoa, F.

(1997). Kinetic modeling of kraft delignification of Eucalyptus globulus. Industrial
& Engineering Chemistry Research, 36, 4114-4125.

Santos, R.B., Capanema, E.A., Balakshin, M.Y., Chang, H.M., y Jameel, H.

(2011). Effect of hardwoods characteristics on kraft pulping process: emphasis on
lignin structure. Bioresources, 6 (4), 3623–3627.

Sarkanen, K. V. y Ludwig, C. H. (1971). Lignins: Ocurrence, Formation,

Structure and Reactions. Wiley Science, New York, NY, USA.

Sarkar, N., Ghosh, S. K., Bannerjee, S. y Aikat, K. (2012). Bioethanol

production from agricultural wastes: an overview. Renewable Energy, 37, 19−27.

Scalbert, A., Monties, B., Rolando, C. y Sierraescudero, A. (1986). Formation of

ether linkage between phenolic-acids and Gramineae lignin: a possible mechanism
involving quinone methides. Holzforschung, 40, 191–195.

Scheller, H.V. y Ulvskov, P. (2010). Hemicelluloses. Annual Review Plant

Biology, 61, 263-289.

Sederoff, R. R., MacKay, J. J., Ralph, J. y Hatfield, R. D. (1999). Unexpected

variation in lignin. Current Opinion in Plant Biology, 2, 145-152.

94
 IV. REFERENCIAS

Singh, R., Singh, S., Trimukhe, K.D., Pandare, K.V., Bastawade, K.B., Gokhale,
D.V. y Varma, A.J. (2005). Lignin-carbohydrate complexes from sugarcane
baggase: preparation, purification and characterization. Carbohydrate Polymers, 62
(1), 57-66.

Singh, S., Blake, A.S. y Vogel, K.P. (2009). Visualization of biomass

solubilization and cellulose regeneration during ionic liquid pretreatment of
switchgrass. Biotechnology and Bioengineering, 104 (1), 68-75.

Sjöström, E. (1981). Wood Chemistry: Fundamentals and Applications.

Academic Press, Orlando, pp. 68-82.

Sjöström, E. (1993). Wood Chemistry. Fundamentals and Applications.

Academic Press, San Diego.

Sjöström, E. y Westermark, U. (1999). Chemical Composition of Wood and

Pulps: Basic constituents and Their Distribution. En: Sjöström, E. y Alén, R.
(Eds.), Analytical Methods in Wood Chemistry, Pulping and Papermaking,
Springer-Verlag, Berlin, Germany, Chap. 1, pp. 1-19.

Shimizu, K. (2001). Wood and Cellulosic Chemistry, 2nd Ed., 177-214.

Solomon, T. W. G. (1988). Organic chemistry, 4th edition, John Wiley & Sons.

Somerville, C., Bauer, S., Brininstool, G., Facette, M., Hamann, T., Milne, J.,
Osborne, E., Paredez, A., Person, S., Raab, T., Vormerk, S. y Youngs, H. (2004).
Toward a systems approach to understanding plant cell walls. Science, 306, 2206-
2211.

Somerville, C., Youngs, H., Taylor, C., Davis, S. C., Long, S. P. (2010).
Feedstocks for lignocellulosic biofuels. Science, 329, 790−792.

Stewart, J. J., Akiyama, T., Chapple, C., Ralph, J. y Mansfield, S. D. (2009).

The effects on lignin structure of overexpression of ferulate 5-hydroxylase in
hybrid poplar. Plant Physiology, 150, 621–635.

Sun, Y. y Cheng, J. (2002). Hydrolysis of lignocellulosic materials for ethanol

production: A review. Bioresource Technology, 83 (1), 1-11.

95
IV. REFERENCIAS

Terashima, N., Fukushima, K., He, L.-F. y Takabe, K. (1993). Comprehensive

model of the lignified plant cell wall. En: Jung, H.G., Buxton, D.R., Hatfield, R.D.
y Ralph, J. (Eds.), Forage Cell Wall Structure and Digestibility, ASA-CSSA-
SSSA, Madison, WI, USA.

Tsutsumi, Y., Kondo, R., Sakai, K. y Imamura, H. (1995). The difference of

reactivity between syringyl lignin and guaiacyl lignin in alkaline systems.
Holzforschung, 49, 423-428.

US Department of Energy. (1993). Assessment of costs and benefits of flexible

and alternative fuel use in the US transportation sector. En: Evaluation of a Wood-
to-Ethanol Process, Technical Report No. 11, DOE/EP-0004, US Department of
Energy,Washington, DC, USA.

Vanholme, R., Morreel, K., Ralph, J. y Boerjan, W. (2008). Lignin engineering.

Current Opinion in Plant Biology, 11, 278–285.

Villaverde, J.J., Ligero, P. y de Vega1, A. (2010). Miscanthus x giganteus as a

source of biobased products through Organosolv fractionation: a mini review. The
Open Agriculture Journal, 4, 102-110.

Wallace, G., Russell, W.R., Lomax, J.A., Jarvis, M.C., Lapierre, C. y Chesson,
A. (1995). Extraction of phenolic-carbohydrate complexes from graminaceous cell
walls. Carbohydrate Research, 272, 41-53

Wallmo, H., Wimby, M. y Larsson, A. (2009). Increased Production in Your

Recovery Boiler with LignoBoost – Extract Valuable Lignin for Biorefinery
Production and Replacement of Fossil Fuels. TAPPI Engineering, Pulping &
Environmental Conference, October 11-14th, Memphis, Tennessee, USA.

Watanabe, T. (2003). Analysis of native bonds between lignin and carbohydrate

by specific chemical reactions. En: Koshijima, T. y Watanabe, T (Eds.),
Association between lignin and carbohydrates in wood and other plant tissues,
Springer series in wood science, pp. 91-130.

Wende, G. y Fry, S.C. (1997). 2-O-β-D-xylopyranosyl-(5-O-feruloyl)-L-

arabinose, a widespread component of grass cell walls. Phytochemistry, 44, 1019–
1030.

96
 IV. REFERENCIAS

Werpy, T. y Petersen, G. (2004). Top Value Added Chemicals from Biomass.

Volume I - Results of Screening for Potential Candidates from Sugars and
Synthesis Gas, Technical Report, Pacific Northwest National Laboratory (PNNL)
and National Renewable Energy Laboratory (NREL).

Woodard, K. R. y Prine, G. M. (1993). Dry matter accumulation of elephant

grass, energy cane and elephant millet in a subtropical climate. Crop Science, 33,
818−824.

Xu, F., Sun, J.X., Sun, R.C., Fowler, P. y Baird, M.S. (2006). Comparative
study of organosolv lignins from wheat straw. Industrial Crops and Products, 23
(2), 180–193.

Yang B. y Wyman C.E. (2008). Pretreatment: the key to unlocking low-cost

cellulosic ethanol. Biofuels, Bioproduction and Biorefinery, 2, 26–40.

Yu, H., Liu, R.G., Shen, D.W., Jiang, Y. y Huang, Y. (2005). Study on
morphology and orientation of cellulose in the vascular bundle of wheat straw.
Polymer, 46, 5689–5694.

Zakzeski, J., Bruijnincx P.C.A., Jongerius, A.L. y Weckhuysen, B.M. (2010).

The catalytic valorization of lignin for the production of renewable chemicals.
Chemical Reviews, 110, 3552-3599.

Zhang, L. y Gellerstedt, G. (2001). NMR observation of a new lignin structure, a

spiro-dienone. Chemcial Communications, 2744-2745.

Zhang, L. y Gellerstedt, G. (2007). Quantitative 2D HSQC NMR determination

of polymer structures by selecting suitable internal standard references. Magnetic
Resonance in Chemistry, 45, 37-45.

Zhang, Y.P., Ding, S., Mielenz, J.R., Cui, J., Elander, R.T., Laser, M., Himmerl,
M.E., McMillan, J.R. y Lynd, L.R. (2007). Fractionating Recalcitrant
Lignocellulose at Modest Reactions Conditions. Biotechnology and
Bioengineering, 97 (2), 214-223.

97
IV. REFERENCIAS

Zhang, A.P., Lu F.C., Sun R.C. y Ralph J. (2009). Ferulate-coniferyl alcohol

cross-coupled products formed by radical coupling reactions. Planta, 229, 1099-
1108.

98
V . RESULTADOS Y DISCUSIÓN
 V. RESULTADOS Y DISCUSIÓN

Los resultados obtenidos durante la Tesis y la discusión de los mismos se

muestran a continuación en forma de publicaciones.

Publicación I. Prinsen P., Gutiérrez A., Rencoret J., Nieto L., Jiménez-Barbero
J., Burnet A., Petit-Conil M., Colodette J.L., Martínez A.T. and del Río J.C. (2012).
Morphological characteristics and composition of lipophilic extractives and lignin
in Brazilian woods from different eucalypt hybrids. Industrial Crops and Products,
36, 572–583.

Publicación II. Prinsen P., Gutiérrez A. and del Río J.C. (2012). Lipophilic
extractives from the cortex and pith of elephant grass (Pennisetum purpureum
Schumach.) stems. Journal of Agricultural and Food Chemistry, 60, 6408−6417.

Publicación III. del Río J.C., Prinsen P., Rencoret J., Nieto L., Jiménez-Barbero
J., Ralph J., Martínez A.T. and Gutiérrez A. (2012). Structural characterization of
the lignin in the cortex and pith of elephant grass (Pennisetum purpureum) stems.
Journal of Agricultural and Food Chemistry, 60, 3619−3634.

Publicación IV. del Río J.C., Prinsen P. and Gutiérrez A. (2013). A

comprehensive characterization of lipids in wheat straw. Journal of Agricultural
and Food Chemistry, 61, 1904-1913.

Publicación V: del Río J.C., Rencoret J., Prinsen P., Martínez A.T., Ralph J.
and Gutiérrez A. (2012). Structural characterization of wheat straw lignin as
revealed by analytical pyrolysis, 2D-NMR, and reductive cleavage methods.
Journal of Agricultural Food and Chemistry, 60, 5922–5935.

Publicación VI: Prinsen P., Rencoret J., Gutiérrez A., Liitiä T., Tamminen T.,
Colodette J.L., Berbis M.A., Martínez A.T. and del Río J.C. (2013). Modification
of the lignin structure during chemical deconstruction of eucalypt wood by kraft-,
soda-AQ and soda-O2 cooking. Industrial & Engineering Chemistry Research
(enviado).

101
 V. RESULTADOS Y DISCUSIÓN

Publicación I:

Prinsen P., Gutiérrez A., Rencoret J., Nieto L., Jiménez-Barbero J., Burnet A.,
Petit-Conil M., Colodette J.L., Martínez A.T. and del Río J.C. (2012).
Morphological characteristics and composition of lipophilic extractives and lignin
in Brazilian woods from different eucalypt hybrids. Industrial Crops and Products,
36, 572–583.

103
 V. RESULTADOS Y DISCUSIÓN

Morphological characteristics and composition of lipophilic

extractives and lignin in Brazilian woods from different eucalypt
hybrids

Pepijn Prinsen1, Ana Gutiérrez1, Jorge Rencoret1, Lidia Nieto2, Jesús Jiménez-
Barbero2, Auphélia Burnet3, Michel Petit-Conil3, Jorge L. Colodette4, Ángel T.
Martínez2, José C. del Río1.

1
Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS), CSIC, PO
Box 1052, E-41080 Seville, Spain

2
Centro de Investigaciones Biológicas (CIB), CSIC, Ramiro de Maeztu 9, E-28040
Madrid, Spain

3
Centre Technique du Papier, BP251, F-38044 Grenoble cedex 9, France

4
Laboratório de Celulose e Papel, Departamento de Engenharia Florestal,
Universidade Federal de Viçosa, 36.570-000 - Viçosa, MG, Brasil

ABSTRACT

The morphological and chemical characteristics of the woods from several eucalypt
hybrids from the Brazilian Genolyptus program were studied. The hybrids selected
for this study were E. grandis × E. urophylla (IP), E. urophylla × E. urophylla
(U1×U2), E. grandis × [E. urophylla × E. globulus] (G1×UGL), and [E. dunnii ×
E. grandis] × E. urophylla (DG×U2). The analyses of the lipophilic extractives
indicated a similar composition in all eucalypt hybrids, which were dominated by
sitosterol, sitosterol esters and sitosteryl 3β-D-glucopyranoside. These compounds
are responsible for pitch deposition during kraft pulping of eucalypt wood. Some
quantitative differences were found in the abundances of different lipid classes, the
wood from U1×U2 having the lowest amounts of these pitch-forming compounds.
The chemical composition and structure of lignins were characterized by Py-
GC/MS and 2D-NMR that confirmed the predominance of syringyl over guaiacyl

105
V. RESULTADOS Y DISCUSIÓN

units and only showed traces of p-hydroxyphenyl units in all the woods, with the
highest S/G ratio for G1×UGL. The 2D-NMR spectra gave additional information
about the inter-unit linkages in the lignin polymer. All the lignins showed a
predominance of β-O-4' ether linkages (75-79% of total side-chains), followed by
β-β' resinol-type linkages (9-11%) and lower amounts of β-5' phenylcoumaran-
type, β-1' spirodienone-type linkages or β-1' open substructures. The lignin from
the hybrid G1×UGL presented also the highest proportion of β-O-4' linkages, and
therefore, it is foreseen that the wood from this hybrid will be more easily
delignifiable than the other selected Brazilian eucalypt hybrids. In complement to
these chemical analyses, the morphological characterization of fibers, vessels and
fines revealed that hybrid eucalypt clone DG×U2 presented the most interesting
properties for the manufacture of paper pulps and biofuels.

KEYWORDS: eucalypt; wood; hybrids; E. grandis × E. urophylla; E. urophylla ×

E. urophylla; E. grandis × [E. urophylla × E. globulus]; [E. dunnii × E. grandis] ×
E. urophylla; lipids; sterols; pitch; lignin; fibers; vessels.

INTRODUCTION

Eucalypt is a fast growing tree whose wood is the main raw material for paper pulp
production in Southwest Europe, Brazil, South Africa, and other countries.
Eucalypt is the largest single global source of market pulp and its use for pulp
production has greatly increased during last decades, the world production attaining
nearly 20 million tons per year, which is about 60% of the total hardwood pulp
produced (Trabado and Wilstermann, 2008). An additional capacity over 10 million
tons/yr is expected in the next 5 years. Different eucalypt species are used for pulp
and papermaking, including Eucalyptus globulus, E. nitens, E. maidenii, E. dunni,
E. grandis, E. urophylla and E. saligna, and single, double and triple crossings
hybrids among these species. The major interest in eucalypt wood comes from its
low production cost in certain regions, due mainly to high forest productivity and
high pulp yield, and the outstanding quality of their fibers.

Biomass production costs are low in Brazil compared to other parts of the world,
due to proper climate, large available areas for cultivation, advanced forest and
agricultural technologies and excellent adaptation of certain crops in the tropical
climate. Thus, Brazil presents great potential for the growing of eucalypt
plantations that are highly productive, and reaching up to 60 m3/ha/year in some
cases (SBS, 2007). Eucalypt plantations in Brazil represent over 3.8 million ha.

106
 V. RESULTADOS Y DISCUSIÓN

Among the planted eucalypts, 55% corresponds to E. grandis, 17% to E. saligna,

11% to hybrids, 9% to E. urophylla, 2% to E. viminalis and 6% to other species
(STCP, 2007).

The interest in the use of eucalypt wood for paper pulp production has promoted
the research of the chemical characteristics of different species (del Río et al.,
2005; Evtuguin et al., 2001; González-Vila et al., 1999; Gutiérrez et al., 1999;
Capanema et al., 2005; Ibarra et al., 2007; Rencoret et al., 2007, 2008, 2011)
aiming to improve the industrial use of this interesting raw material for pulp and
paper manufacturing. However, only limited studies have been published regarding
the chemical characteristics of eucalypt woods grown in tropical areas, such as the
so-called E. urograndis, a hybrid derived from the single crossing between E.
grandis and E. urophylla, that is one of the main eucalypt hybrids used for pulping
in Brazil (Freire et al., 2006; Silvério et al., 2007).

In this paper, we report the morphological and chemical characteristics of the

woods from different eucalypt hybrids grown in Brazil and coming from the
Brazilian Network of Eucalyptus Genome Research, the so-called Genolyptus
program (Grattaplagia, 2003, 2004), a nationwide initiative which involved
integrated advances in genomic resources, molecular breeding and wood
phenotyping technologies. Double and triple crossings were selected for this study.
A special emphasis will be put in the lipid and lignin composition since these two
fractions play an important role during pulping, bleaching and papermaking. The
content and chemical structure of wood components, in particular the lignin content
and its composition in terms of its p-hydroxyphenyl (H), guaiacyl (G) and syringyl
(S) moieties and the different inter-unit linkages are important parameters in pulp
production in view of delignification rates, chemical consumption and pulp yields,
as well as in subsequent bleaching. Higher S/G ratios imply higher delignification
rates, less alkali consumption and therefore higher pulp yield (González-Vila et al.,
1999; del Río et al., 2005). On the other hand, extractives, especially the lipophilic
compounds, are also important for pulp and paper production (Back and Allen,
2000). Extractives are often released from the fibers during pulping and can form
colloidal pitch and cause production troubles as deposits. In the manufacture of
alkaline pulps, a large part of the lipids is removed from the wood during cooking.
However, some substances survive this process and can be found as pulp
extractives. If they form the so-called pitch deposits, the consequences are serious,
including reduced product quality and higher operating costs due to production
stops for cleaning the equipment (del Río et al., 1998, 2000; Gutiérrez et al., 2001a,

107
V. RESULTADOS Y DISCUSIÓN

2001b; Silvestre et al., 1999). The increasing trend in recirculating water in pulp
mills aggravates these problems.

Finally, the morphological characteristics of the wood components, which

strongly influence the quality of the pulp, and of the final paper sheet, were also
addressed. Longer, flexible and fibrillated fibers are preferred for reaching the
required paper strengths, but shorter fibers and fines improve the optical properties.
For hardwoods, and therefore for eucalypts, the vessels are wood components that
can generate lack of resistance into the sheet, dust generation during sheet
converting and printing defaults (picking and speckles). Lower vessel contents into
the pulp would be preferable, and more particularly the early wood vessels.

MATERIALS AND METHODS

Samples

The eucalypt hybrid woods selected for this study were supplied by the
University of Viçosa and consist of five-year-old trees of the following double or
triple crossings: (i) E. grandis × E. urophylla (IP); (ii) E. urophylla × E. urophylla
(U1 × U2); (iii) E. grandis × [E. urophylla × E. globulus] (G1 × UGL); and (iv) [E.
dunnii × E. grandis] × E. urophylla (DG × U2).

The wood samples were air dried. The dried samples were milled using a knife
mill (Janke and Kunkel, Analysenmühle), and successively extracted with acetone
in a Soxhlet apparatus for 8 h and with hot water (3 h at 100 ºC). The acetone
extracts were evaporated to dryness, and resuspended in chloroform for
chromatographic analysis of the lipophilic fraction. The water soluble material was
also evaporated to dryness and weighted. Two replicates were used for each
sample. Klason lignin was estimated according to T222 om-88 (Tappi, 2004).
Milled-wood lignin (MWL) was extracted from finely ball-milled (15 h) plant
material, free of extractives and hot water soluble material, using dioxane-water
(9:1, v/v), followed by evaporation of the solvent, and purified as described
(Björkman, 1956). The final yields ranged from 15% to 20% of the original Klason
lignin content.

GC and GC/MS analyses

The GC analyses of lipids were performed in an Agilent 6890N Network GC

system equipped with a split–splitless injector and a flame ionization detector
(FID). The injector and the detector temperatures were set at 300 ºC and 350 ºC

108
 V. RESULTADOS Y DISCUSIÓN

respectively. Samples were injected in the splitless mode. Helium was used as the
carrier gas. The capillary column used was a high temperature, polyimide coated
fused silica tubing DB5-HT (5 m × 0.25 mm I.D., 0.1 µm film thickness; J&W
Scientific). The oven was temperature programmed from 100 ºC (1 min) to 350 ºC
(3 min) at 15 ºC min−1. Peaks were quantified by area, and a mixture of standards
(octadecane, palmitic acid, sitosterol, cholesteryl oleate, and sitosteryl-3β-D-
glucopyranoside) was used to elaborate calibration curves. The data from the two
replicates were averaged.

The GC/MS analysis were performed on a Varian 3800 gas chromatograph

coupled with an ion-trap detector (Varian 4000) equipped with a high-temperature
capillary column (DB5-HT, 15 m × 0.25 mm i.d., 0.1 µm film thickness; J&W
Scientific). Helium was used as carrier gas at a rate of 2 mL min−1. The oven was
heated from 120 ºC (1 min) to 380 ºC (5 min) at 10 ºC min−1. The temperature of
the injector during the injection was 120 ºC, and 0.1 min after injection was
programmed to 380 ºC at a rate of 200 ºC min−1 and held for 10 min. The
temperature of the transfer line was set at 300 ºC. The lipophilic extractives were
analyzed both underivatized and as trimethylsilyl derivatives formed by reaction
with bis(trimethylsilyl)trifluoroacetamide (BSTFA). Compounds were identified by
comparing their mass spectra wit mass spectra in the Wiley and NIST libraries, by
mass fragmentography and, when possible, by comparison with authentic
standards.

Gel permeation chromatography

GPC was performed on a Shimadzu LC-20A LC system (Shimadzu, Kyoto,

Japan) equipped with a photodiode array (PDA) detector (SPD-M20A; Shimadzu)
g f w g : SK g α- + α-2500 (Tosoh, Tokyo,
Japan); eluent, 0.1 M LiBr in dimethylformamide (DMF); flow rate, 0.5 mL min−1;
temperature, 40 ºC; sample detection, PDA response at 280 nm. The data
acquisition and computation used LCsolution version 1.25 software (Shimadzu).
The molecular weight calibration was via polystyrene standards.

Py-GC/MS

Pyrolysis of MWL (approximately 100 µg) was performed with a 2020 micro-
furnace pyrolyzer (Frontier Laboratories Ltd.) connected to an Agilent 6890
GC/MS system equipped with a HP 5MS fused-silica capillary column (30 m ×
0.25 mm i.d., 0.25 µm film thickness) and an Agilent 5973 mass selective detector
(EI at 70 eV). The pyrolysis was performed at 500 ºC. The oven temperature was

109
V. RESULTADOS Y DISCUSIÓN

programmed from 50 ºC (1 min) to 100 ºC at 30 ºC min−1 and then to 300 ºC (10

min) at 10 ºC min−1. He was the carrier gas (1 mL min−1). The compounds were
identified by comparing their mass spectra with those of the Wiley and NIST
libraries and reported in the literature (Faix et al., 1990; Ralph and Hatfield, 1991).
Peak molar areas were calculated for the lignin-degradation products, the summed
areas were normalized, and the data for two repetitive analyses were averaged and
expressed as percentages.

NMR spectroscopy

2D-NMR spectra were recorded at 25 ºC in a Bruker AVANCE 600 MHz,

equipped with a cryogenically-cooled z-gradient triple resonance probe. Forty
milligrams of MWL were dissolved in 0.75 mL of dimethylsulfoxide (DMSO)-d6,
and HSQC (heteronuclear single quantum correlation) spectra were recorded. The
spectral widths were 5000 and 13,200 Hz for the 1H and 13C dimensions,
respectively. The number of collected complex points was 2048 for 1H dimension,
with a recycle delay of 5 s. The number of transients was 64, and 256 time
increments were always recorded in 13C dimension. The 1JCH used was 140 Hz. The
J-coupling evolution delay was set to 3.2 ms. Squared cosine-bell apodization
function was applied in both dimensions. Prior to Fourier transformation, the data
matrixes were zero filled up to 1024 points in the 13C dimension. The central
v kw f δC 39 5; δH 2.49). HSQC cross-
signals were assigned by comparing with the literature. The cross-signal intensity
depends on the particular 1JCH value, as well on the T2 relaxation time; therefore, a
direct intensity analysis of the different signals is impossible. Thus, integration was
performed separately for the different regions of the spectra, which contain signals
corresponding to chemically-analogous C–H pairs, with similar 1JCH-coupling
values. In the aliphatic oxygenated region, inter-unit linkages were estimated from
Cα–Hα correlations, except for structures E and F described below where Cβ–Hβ
and Cγ–Hγ correlations were used, respectively, and the relative abundance of side-
chains involved in different substructures and terminal structures were calculated
(with respect to total side-chains), as well as the percentage of each inter-unit
linkage type with respect to the linkage total. In the aromatic region, C–H
correlations from S and G units were used to estimate the S/G ratio.

Morphological characterization of the fibers and vessels

The wood samples were treated with a solution of 3.4% sodium chlorite in
acetic acid buffer (pH 4.9) at 100 ºC under stirring until the separation of the wood

110
 V. RESULTADOS Y DISCUSIÓN

components from their matrix. The obtained fibers and vessels suspension was
washed with tap water and filtered before analysis with the MorFi and
CyberMetrics analyzers. The MorFi analyzer was designed and developed by CTP
for the morphological characterization of fibers, vessels and fines (Eymin-Petot-
Tourtollet et al., 2003). The analysis was carried out on a pulp suspension passing
through a specific cell illuminated by a laser beam and connected to a high-
resolution CCD camera. This analysis allowed reliable statistical measurement of
thousands of fibers, vessels and fines to determine the main morphological and
dimension characteristics of the pulp components. The CyberBond analyzer
allowed to measure the relative bonded area index (RBA) of the fibers, an
important characteristics controlling the inter-fiber bonding potential (Das et al.,
2003). After deposition on a glass slide, the fibers were analyzed by the way of a
microscope lens coupled to a CCD camera. The RBA index measurement was
k ’ g y w g
automatically the index (Das et al., 2003). Cross sections of the eucalypt wood
samples were directly examined with a light microscope to observe the distribution
of fibers and vessels.

RESULTS AND DISCUSSION

Chemical characteristics of the Brazilian woods from different eucalypt

hybrids

The content of acetone extractives, water soluble material, Klason lignin and
acid soluble lignin in the Brazilian woods from the different eucalypt hybrids is
listed in Table 1.

Table 1. Content (%) of different components of the woods of the different eucalypt
hybrids. (as dry, ash-free basis).

Component IP U1  U2 G1  UGL DG  U2

Acetone extractives (lipophilics) 0.6 (0.2) 2.1 (0.2) 2.2 (0.3) 0.9 (0.2)

Water soluble material 1.4 1.7 1.8 1.5

Klason lignin* 24.2 24.3 24.1 24.5

Acid-soluble lignin 3.0 3.4 3.4 3.1

* corrected for proteins and ash

111
V. RESULTADOS Y DISCUSIÓN

The Klason lignin content was very similar for all of them, in the range from
24.1 to 24.5%. The total acetone extractives content of the woods from the different
eucalypt hybrids ranges from 0.6 to 2.2% of dry material. However, the content of
lipophilic extractives, estimated as the fraction of the acetone extracts that can be
redissolved in chloroform, is much lower and very similar for all the woods,
ranging from 0.2 to 0.3% of the total dry material. This low lipophilic content is
similar to that found in woods from different eucalypt species (Rencoret et al.,
2007).

Lipid composition of the Brazilian woods from different eucalypt hybrids

Although the lipid content is low and similar in all the selected hybrid eucalypt
woods, it is not only the content but the composition that strongly affects the pitch
deposition during pulping and papermaking (Back and Allen, 2000). Therefore, the
detailed chemical composition of the lipophilic extractives in the different woods
was investigated. The underivatized and silylated lipophilic extracts from the
hybrid eucalypt woods were analyzed by GC and GC/MS using short- and
medium-length high temperature capillary columns, respectively, with thin films,
according to the method previously described (Gutiérrez et al., 1998). The GC/MS
chromatogram of the lipid extracts (as trimethylsilyl ether derivatives) from a
selected eucalypt hybrid (IP) is shown in Fig. 1. The identities and abundances of
the main lipophilic compounds identified in the selected Brazilian eucalypt hybrids
are detailed in Table 2, and their structures are depicted in Fig. 2. The most
predominant lipophilic compounds present in all the eucalypt woods selected for
this study were steroids, including sterols, sterol glycosides and sterol esters with
lower amounts of steroid ketones and steroid hydrocarbons. Other important
lipophilic compounds found were series of fatty acids, glycerides (including mono-,
di- and triglycerides) and minor amounts of squalene, tocopherols (in free and
esterified form) and a series of alkyl ferulates. This composition is similar to that
reported in the woods of other eucalypt species (Gutiérrez et al., 1999; Rencoret et
al., 2007).

112
 V. RESULTADOS Y DISCUSIÓN

Detector response sitosterol

fucosterol
stigmastanol
sterol esters

sitosterol 3β-D-glucopyranose

fatty acids steroid ketones

C18:1
+ C26
C18:2 C24
C16 C22 C28
C18 C20

5 10 15 20
Retention time (min)

Figure 1. Chromatogram of the lipophilic extractives (as TMSi derivatives) from a

selected eucalypt hybrid (IP) wood.

Free sterols were among the major compound class in the extracts of all eucalypt
woods (ranging from 412 to 902 mg/kg wood), sitosterol (I; Fig. 2) and
stigmastanol (II) being the main sterols present in all eucalypt woods, with lower
amounts of campesterol (III), stigmasterol (IV), fucosterol (V), cycloartenol (VI),
24-methylenecycloartanol (VII) and 7-oxositosterol (VIII). The wood from U1 ×
U2 presents the lowest content of free sterols, which are among the main
compounds responsible for pitch deposition during kraft cooking of eucalypt wood
(del Río et al., 1998, 2000; Silvestre et al., 1999), while the rest of the eucalypt
hybrids selected for this study present higher amounts of free sterols, and therefore
will be more prone to have pitch deposition problems.

113
V. RESULTADOS Y DISCUSIÓN

Table 2. Composition of lipophilic extractives (mg/kg wood) from the woods of the
different eucalypt hybrids.

Compound IP U1  U2 G1  UGL DG  U2

Fatty Acids 289.7 139.8 310.2 208.0

n-pentadecanoic acid 2.0 1.1 1.9 1.1
n-hexadecanoic acid 43.9 35.8 54.9 40.9
n-heptadecanoic acid 4.2 2.5 4.1 4.3
9,12-octadecadienoic acid 49.8 28.6 54.1 40.1
9-octadecenoic acid 27.0 15.5 50.0 25.7
n-octadecanoic acid 19.9 10.8 16.5 15.9
n-nonadecanoic acid 1.7 0.8 1.4 1.1
n-eicosanoic acid 6.8 3.7 8.9 6.2
n-heneicosanoic acid 6.3 1.6 5.4 5.1
n-docosanoic acid 16.5 4.2 13.6 8.9
n-tricosanoic acid 4.2 3.8 5.8 9.6
n-tetracosanoic acid 38.0 11.6 35.7 22.1
n-pentacosanoic acid 13.0 3.6 8.1 9.6
n-hexacosanoic acid 40.0 11.0 23.8 13.2
n-heptacosanoic acid 3.6 1.3 3.7 0.9
n-octacosanoic acid 12.8 3.9 22.3 3.3

Steroid hydrocarbons 35.7 16.4 19.3 27.8

stigmasta-4,22-diene 3.1 1.8 2.0 2.4
stigmasta-3,5-diene 25.0 11.2 13.5 24.3
stigmasta-3,5,7-triene 7.6 3.4 3.8 1.1

Other hydrocarbons 23.3 4.8 13.6 12.8

squalene 23.3 4.8 13.6 12.8

Sterols 901.5 411.8 731.1 895.4

campesterol 10.7 4.3 6.5 8.9
stigmasterol 15.3 2.5 3.8 2.4
sitosterol 640.9 300.9 520.1 684.6
stigmastanol 199.7 87.4 169.9 161.6
fucosterol 23.5 14.3 14.6 30.9
cycloartenol 4.1 1.2 13.6 3.8
24-methylenecycloartanol 3.4 0.3 1.4 1.9
7-oxositosterol 3.9 0.9 1.2 1.3

Tocopherols 2.1 1.2 6.9 3.7

-tocopherol 1.7 0.9 5.4 3.0
-tocopherol 0.4 0.3 1.5 0.7

Steroid ketones 44.7 19.8 60.7 26.5

stigmasta-3,5-dien-7-one 1.7 1.2 5.8 3.2
stigmast-4-en-3-one 33.8 13.5 40.0 16.4
stigmastan-3-one 0.9 0.6 3.4 1.2
stigmast-4-en-3,6-dione 1.2 0.6 2.8 0.8
stigmastan-3,6-dione 7.1 3.9 8.7 4.9

114
 V. RESULTADOS Y DISCUSIÓN

Monoglycerides 157.4 130.5 201.9 123.0

2,3-dihydroxypropyl tetradecanoate 0.1 0.1 0.1 0.1
2,3-dihydroxypropyl hexadecanoate 2.6 2.8 2.5 2.1
2,3-dihydroxypropyl octadecanoate 7.3 6.7 8.3 7.6
2,3-dihydroxypropyl eicosanoate 0.2 0.3 1.5 0.5
2,3-dihydroxypropyl docosanoate 5.1 7.8 12.4 12.7
2,3-dihydroxypropyl tricosanoate 0.2 0.3 0.0 0.1
2,3-dihydroxypropyl tetracosanoate 16.9 13.1 22.4 18.1
2,3-dihydroxypropyl pentacosanoate 3.7 2.2 2.9 2.1
2,3-dihydroxypropyl hexacosanoate 48.6 30.6 29.2 39.0
2,3-dihydroxypropyl heptadecanoate 7.4 5.3 6.5 4.1
2,3-dihydroxypropyl octacosanoate 59.7 56.3 102.8 31.8
2,3-dihydroxypropyl nonacosanoate 3.2 2.8 5.7 2.8
2,3-dihydroxypropyl triacontanoate 2.4 2.2 7.6 2.0

Diglycerides 23.4 10.4 14.2 15.7

dipalmitin 8.5 1.9 2.5 1.9
palmitoylstearin 12.8 6.6 8.5 11.4
distearin 2.1 1.9 3.2 2.4

n-Alkylferulates 17.0 9.3 51.6 11.8

trans-docosanylferulate 0.9 0.7 9.0 1.6
trans-tricosanylferulate 0.2 0.1 0.5 0.2
trans-tetracosanylferulate 3.7 2.0 18.1 3.5
trans-pentacosanylferulate 0.5 0.2 1.1 0.3
trans-hexacosanylferulate 5.5 3.5 18.9 3.2
trans-heptacosanylferulate 0.7 0.1 0.6 0.4
trans-octacosanylferulate 5.5 2.7 3.4 2.6

Sterol glycosides 130.0 96.5 128.9 177.7

sitosteryl-3β-D-glucopyranoside 130.0 96.5 128.9 177.7

Tocopherol esters 3.3 4.3 9.2 9.0

-tocopherol ester 0.8 0.6 2.4 3.9
α-tocopherol esters 2.5 3.7 6.8 5.1

Sterol esters 384.6 432.6 375.9 346.4

sitosteryl esters 238.2 274.7 263.8 274.6
stigmastanol esters 40.1 46.3 48.9 33.2
other sterol esters 106.3 111.6 63.2 38.6

Triglycerides 11.1 4.8 10.3 9.5

115
V. RESULTADOS Y DISCUSIÓN

HO HO HO HO
I II III IV

HO HO HO HO O
V VI VII VIII

CH2OH
O
O
OH O
HO
O
OH IX X

O O O O
XI XII XIII O XIV

O O O

OH OH OH
XV XVI XVII

CO-O-CH2 CO-O-CH2

HO-CH CO-O-CH

HO-CH2 CO-O-CH2

XVIII XIX

O O

XX XXI
OCH3
OH

HO HO

O O
XXII XXIII

Figure 2. Structures of the main compounds identified in the hybrid eucalypt woods and
referred in the text. (I) sitosterol, (II) stigmastanol, (III) campesterol, (IV) stigmasterol, (V)
fucosterol, (VI) cycloartenol, (VII) 24-methylenecycloartanol, (VIII) 7-oxositosterol, (IX)
sitosteryl 3β-D-glucopyranoside, (X) sitosteryl linoleate, (XI) stigmastan-3-one, (XII)
stigmasta-3,5-dien-7-one, (XIII) stigmast-4-en-3-one, (XIV) stigmasta-3,6-dione, (XV)
palmitic acid, (XVI) oleic acid, (XVII) linoleic acid, (XVIII) docosanoic acid, 2,3-
dihydroxypropyl ester, (XIX) tripalmitin, (XX) trans-docosanyl ferulate, (XXI) squalene,
(XXII) α-tocopherol, (XXIII) β-tocopherol.

116
 V. RESULTADOS Y DISCUSIÓN

Free fatty acids were also important constituents of the different eucalypt wood
extractives. The series of free fatty acids accounted for 140–290 mg/kg wood, and
ranged from C15 to C28 with a strong even-over-odd carbon atom number
predominance and the dominant component being the saturated palmitic (XV) and
stearic acids together with the unsaturated oleic (XVI) and linoleic (XVII) acids.
Fatty acids were also found as glycerides, including mono-, di- and triglycerides.
Monoglycerides (XVIII) were present in all eucalypt hybrids in relatively high
amounts (from 123 to 202 mg/kg wood), in the range C14–C30, and with a
predominance of the even carbon atom number homologs (C24, C26 and C28 were
the most abundant). The series of diglycerides (10–23 mg/kg) and triglycerides
(XIX; 5–11 mg/kg wood) were present in lower amounts. A series of n-alkyl
ferulates (XX) was also found among the lipophilic extracts in relatively low
amounts (9–52 mg/kg wood). Characterization of intact individual compounds was
achieved based on the mass spectra obtained by GC/MS of the underivatized and
their TMS ether derivatives already published (del Río et al., 2004). The series of
n-alkyl trans-ferulates occurred in the range from C22 to C28, with a predominance
of the even carbon atom number homologs. Alkyl ferulates were also previously
reported in the woods of different eucalypt species (Freire et al., 2002; Rencoret et
al., 2007). Finally, minor amounts of other compounds such as the isoprenoid
hydrocarbon squalene (XXI; 5–23 mg/kg wood), tocopherols (1–7 mg/kg wood)
and tocopherol esters (3–9 mg/kg wood), were also found in all eucalypt extracts.
Free and esterified tocopherols included both α-tocopherol (XXII) and β-
tocopherol (XXIII), with predominance of α-tocopherol.

The different lipids classes have different behavior during cooking and
bleaching (Gutiérrez et al., 2001a, 2001b; Freire et al., 2005, 2006a, 2006b;
Marques et al., 2010). The glycerides (including mono-, di- and triglycerides), are
completely hydrolyzed during alkaline cooking and the fatty acids dissolved. Sterol
esters, however, largely survive alkaline cooking (del Río et al., 1998, 2000;
Gutiérrez et al., 2001a, 2001b). The high amounts of these neutral compounds in
most of these woods, and particularly the high abundances of free and conjugated
sterols, which have a high propensity to form pitch deposits (del Río et al., 1998,
2000; Silvestre et al., 1999; Gutiérrez and del Río, 2001; Gutiérrez et al., 2001a,
2001b) would point to a pitch deposition tendency of the lipophilics from these
woods. Among them, the wood from U1 × U2 has the lowest content of these
detrimental compounds and therefore will have less pitch problems, while the
woods of IP and DG × U2 have the highest content of them, and therefore it is
foreseen that they will have more pitch problems than the wood from the hybrid U1
× U2.

117
V. RESULTADOS Y DISCUSIÓN

Composition and structure of the lignins from the different eucalypt

hybrids

The lignin content of the different eucalypt hybrids selected for this study,
estimated as Klason lignin, is similar in all cases (ca. 24–25%) and slightly higher
in comparison with other eucalypt woods, such as E. globulus (Rencoret et al.,
2007, 2008). However, the delignification reactions and, therefore, the pulping
efficiency are not only affected by the lignin content but are also greatly influenced
by the lignin composition and structure. Therefore, we have thoroughly studied the
lignin composition and structure of the Brazilian woods from the selected eucalypt
hybrids. For this, the MWL, which is considered to be representative of the whole
native lignin in the plant, was isolated by aqueous dioxane extraction from finely
ball-milled wood according to the classical lignin isolation procedure (Björkman,
1956).

The values of the weight-average (Mw) and number-average (Mn) molecular

weights, estimated from the GPC curves (relative values related to polystyrene),
and the polydispersity (Mw/Mn) of the MWL from the selected eucalypt hybrids, are
indicated in Table 3. The MWLs exhibited similar molecular weight distributions,
in the range 11,300–15,000 g mol−1, being slightly higher in the case of the MWL
from IP and lower for the MWL from DG × U2. In addition, all the MWL exhibited
relatively narrow molecular weight distributions, with Mw/Mn < 4. Those values are
comparable to literature values for various isolated lignins (Baumberger et al.,
2007).

Table 3. Weight-average (Mw) and number-average (Mn) molecular weights (g mol-1), and
polydispersity (Mw/Mn) of the MWL from the woods of the different eucalypt hybrids
selected in this study.

IP U1  U2 G1  UGL DG  U2

Mw 15000 12900 13300 11300

Mn 4300 3900 3500 3000
Mw/Mn 3.5 3.3 3.8 3.8

The composition of the MWLs were analyzed by Py-GC/MS. All the eucalypt
hybrid lignins yielded similar Py-GC/MS products, and a representative pyrogram
is shown in Fig. 3. The identities and relative molar abundances of the released

118
 V. RESULTADOS Y DISCUSIÓN

lignin compounds are listed in Table 4. Among them, guaiacyl (G) and syringyl-
type (S) phenols, were identified. Only minor amounts (ca. 1%) of phenol-type
compounds from p-hydroxycinnamyl (H) units could be detected. The most
important compounds identified were guaiacol (3), 4-methylguaiacol (6), 4-
vinylguaiacol (8), syringol (10), 4-methylsyringol (14), 4-vinylsyringol (20),
syringaldehyde (25), trans-4-propenylsyringol (26), homosyringaldehyde (27),
acetosyringone (28), syringylacetone (29), propiosyringone (31) and trans-
sinapaldehyde (32). The molar S/G ratios obtained from the molar areas of all the
lignin-derived compounds are shown in Table 4 and ranged from 2.6 to 3.1. The
lignin from the eucalypt hybrids IP, U1 × U2 and DG × U2 present similar lignin
composition and S/G ratio, while the lignin from the hybrid G1 × UGL presents the
highest content on S lignin and the lowest content on G-lignin (S/G ratio of 3.1).
This composition will make the wood from the hybrid G1 × UGL easier to be
delignified under kraft cooking than the other eucalypt hybrids due to the higher
reactivity of the S lignin in alkaline systems (Chang and Sarkanen, 1973; Tsutsumi
et al., 1995). It has already been shown for eucalypt woods that higher S/G ratios
imply higher delignification rates, less alkali consumption and therefore higher
pulp yield (González-Vila et al., 1999; del Río et al., 2005).

14

10

25
26
Detector response

20
28
18
6

27 29
3 8 31
32
7 15
24
21
16 22
13 17 19 30
4 12 23
2 5
1

4 6 8 10 12 14 16 18 20 22
Retention time (min)
Figure 3. Py-GC/MS chromatogram of a representative MWL isolated from a selected
eucalypt hybrid (DG × U2) wood. The numbers refer to the compounds listed in Table 4.

119
V. RESULTADOS Y DISCUSIÓN

Table 4. Identification and relative molar abundance of the compounds identified in the
Py-GC/MS of MWL from wood of the different eucalypt hybrids selected in this study.

Label Compound IP U1  U2 G1  UGL DG  U2

1 phenol 0.5 0.4 0.4 0.4
2 methylphenol 0.2 0.2 0.2 0.4
3 guaiacol 5.7 5.4 4.0 5.9
4 methylphenol 0.4 0.4 0.1 0.5
5 ethylphenol 0.0 0.0 0.1 0.1
6 4-methylguaiacol 6.5 7.9 6.0 7.1
7 4-ethylguaiacol 1.7 1.9 1.1 2.7
8 4-vinylguaiacol 3.7 2.4 2.7 2.0
9 eugenol 0.5 0.3 0.1 0.1
10 syringol 10.5 9.5 8.1 12.5
11 4-propylguaiacol 0.0 0.0 0.0 0.1
12 cis-isoeugenol 0.5 0.3 0.3 0.3
13 vanillin 1.8 1.8 1.6 1.6
14 4-methylsyringol 11.4 15.5 13.6 13.1
15 trans-isoeugenol 1.9 0.7 1.0 1.0
16 homovanilline 0.7 0.9 1.3 1.3
17 acetoguaiacone 1.0 1.6 1.7 1.4
18 4-ethylsyringol 2.9 2.9 3.4 3.4
19 guaiacylacetone 0.8 1.3 0.9 1.2
20 4-vinylsyringol 7.1 5.0 6.5 4.9
21 4-allylsyringol 1.9 1.4 0.9 1.0
22 4-propylsyringol 0.4 0.5 0.3 0.7
23 propiovanillone 0.1 0.4 0.4 0.2
24 cis-4-propenylsyringol 1.6 1.1 0.9 1.5
25 syringaldehyde 10.5 11.3 10.9 8.9
26 trans-4-propenylsyringol 7.7 3.2 5.1 5.2
27 homosyringaldehyde 2.5 2.6 5.1 3.5
28 acetosyringone 4.1 6.1 5.2 5.1
29 syringylacetone 3.8 2.8 2.9 2.9
30 trans-coniferaldehyde 1.8 2.9 2.9 2.7
31 propiosyringone 2.0 2.2 2.5 2.0
32 trans-sinapaldehyde 5.6 7.1 10.0 6.2

H 1 1 1 1
G 27 28 24 28
S 72 71 75 71
S/G = 2.7 2.6 3.1 2.6

The structure of the isolated lignins was also analyzed by 2D-NMR, that
provides information of the structure of the whole macromolecule and is a powerful
tool for lignin structural characterization. The side-chain and aromatic regions of

120
 V. RESULTADOS Y DISCUSIÓN

the HSQC spectrum of a representative eucalypt MWL (from DG × U2) are shown
in Fig. 4, together with the main lignin substructures present. Cross-signals were
assigned by comparing with the literature (Balakshin et al., 2003; Capanema et al.,
2001, 2004, 2005; del Río et al., 2008, 2009; Ibarra et al., 2007; Liitiä et al., 2003;
Martínez et al., 2008; Ralph et al., 1999; Ralph and Landucci, 2010; Rencoret et
al., 2008, 2011) and the main assignments are listed in Table 5.

Table 5. Assignments of the lignin 13C-1H correlation signals observed in the HSQC
spectra of the MWL from the eucalypt hybrids.
Labels δC/δH (ppm) Assignment
Bβ 53.4/3.06 Cβ-Hβ β-β' B)
Cβ 53.4/3.45 Cβ-Hβ β-5' (phenylcoumaran) substructures (C)
Fβ 54.8/2.75 Cβ-Hβ β-1' substructures (erythro forms) (F)
OMe 55.6/3.73 C-H in methoxyls
Aγ 59.5/3.38-3.71 Cγ-Hγ β-O-4' substructures (A) and others
Dβ 59.8/2.75 Cβ-Hβ β-1' (spirodienone) substructures (D)
Jγ 61.4/4.10 Cγ-Hγ in cinnamyl alcohol end-groups (J)
Cγ 62.0/3.75 Cγ-Hγ β-5' (phenylcoumaran) substructures (C)
Bγ 71.1/3.82 and 4.18 Cγ-Hγ β-β' B)
Aα S 71.8/4.87 Cα-Hα β-O-4' substructures linked to a S unit (A)
Aα 71.3/4.77 Cα-Hα β-O-4' substructures linked to a G unit (A)
Dβ' 79.3/4.11 C β'-H β' β-1' (spirodienone) substructures (D)
Dα 81.1/5.10 Cα-Hα β-1' (spirodienone) substructures (D)
Eβ 82.9/5.22 Cβ-Hβ in Cα- x β-O-4' substructures (E)
Aβ 83.5/4.29 Cβ-Hβ β-O-4' substructures linked to a G unit (A)
Bα 84.6/4.66 Cα-Hα β-β' B)
Aβ S 86.0/4.11 Cβ-Hβ β-O-4' substructures linked to a S unit (A)
Dα' 84.7/4.76 Cα'-Hα' β-1' (spirodienone) substructures (D)
Cα 86.8/5.42 Cα-Hα β-5' (phenylcoumaran) substructures (C)
S2,6 103.9/6.68 C2-H2 and C6-H6 in syringyl units (S)
S'2,6 S''2,6 106.3/7.32 and 7.20 C2-H2 and C6-H6 in Cα-oxidized syringyl units (S' and S'')
G2 110.8/6.96 C2-H2 in guaiacyl units (G)
D2' 113.5/6.26 C2'-H2' β-1' (spirodienone) substructures (D)
G5 114.9/6.70 and 6.94 C5-H5 in guaiacyl units (G)
G6 118.9/6.76 C6-H6 in guaiacyl units (G)
D6' 118.9/6.08 C6'-H6' β-1' (spirodienone) substructures (D)

121
V. RESULTADOS Y DISCUSIÓN

Figure 4. HSQ δC/δH 50-160/2.5-8.0) of a representative MWL (from DG

× U2). See Table 5 for signal assignment. Main structures present in the lignin from the
Brazilian eucalypt hybrid woods: (A β-O-4' substructures; (B) resinol substructures
f y β-β'/α-O-γ'/γ-O-α' k g ; C y f y β-
5'/α-O-4' linkages; (D f y β-1'/α-O-4' linkages; (E) Cα-
x β-O-4' substructures; (F v β-1' structures; (I) p-
hydroxycinnamyl alcohol end-groups; (G) guaiacyl units; (S) syringyl units; (S') oxidized
syringyl units bearing a carbonyl group at Cα (phenolic); (S'') oxidized syringyl units
bearing a carboxyl group at Cα.

122
 V. RESULTADOS Y DISCUSIÓN

The side-chain region of the spectra gave useful information about the different
inter-unit linkages present in the eucalypts lignins. All the spectra showed
g g β-O-4´ alkyl aryl ether linkages (substructure
A I β-O-4´ substructures, other linkages were also observed. Thus,
g g f β–β´/α-O-γ´/γ-O-α´ B y β-5´/α-O-
4´) substructures (C) and small signals corresponding to spirod β-1´/α-O-α´
substructures (D) were observed in all spectra. Other small signals in the side-chain
region of the HSQC spectra corresponded to Cα–Hα f β-O-4´
substructures bearing a Cα carbonyl group (E), the Cβ–Hβ f β-1´ open
substructures (F) and Cγ–Hγ correlations of p-hydroxycinnamyl (I) end-groups. The
main cross-signals observed in the aromatic region of the HSQC spectra
corresponded to the benzenic rings of lignin units. Signals from syringyl (S) and
guaiacyl (G) units could be observed in all the spectra. However, signals of H units
were not detected in any of the HSQC spectra as corresponds to their very low
abundances observed by Py-GC/MS. The percentage of lignin side-chains involved
in the main substructures and terminal structures found in the different eucalypt
lignins (referred to total side-chains) are indicated in Table 6.

Table 6. Percentage of lignin side-chains forming different inter-unit linkages (A-E) and
terminal structures (F) from integration of 13C-1H correlation signals in the HSQC spectra
of MWL from wood of the different eucalypt hybrids analyzed (referred to total side-
chains).

IP U1 × U2 G1 × UGL DG × U2
β-O-4' alkyl aryl ethers (A) 75 74 77 74

Resinols (B) 11 11 9 11

Phenylcoumarans (C) 5 5 5 5

Spirodienones (D) 4 4 3 4

β-O-4' (Cα=O) (E) 1 1 2 1

β-1' substructures (F) 1 2 1 2

p-Hydroxycinnamyl alcohol 3 3 3 3
end-groups (I)
S/G ratio 2.2 2.2 2.8 2.2

123
V. RESULTADOS Y DISCUSIÓN

In all cases, the main lignin substructure w β-O-4´ alkyl aryl ether
(A) that amounted to 75–79% of all side- g x β-O-4´
ones, E). The second most abundant linkage in the eucalypt lignin corresponded to
the resinol substructure (B) that involved around 9-11 % of all side-chains. The
other linkages, such as phenylcoumaran (C), spirodienone (D β-1´ open
substructures (F), were present in lower proportions (1–5% of all side-chains).A
NMR estimation of the molar S/G ratios in the lignins from the different Brazilian
eucalypt hybrids is included in Table 6, and ranges from 2.2 to 2.8. The highest
S/G value corresponded to the lignin from the hybrid G1 × UGL, as similarly
observed by Py-GC/MS. The higher S/G ratio observed in this lignin is related to
g f β -O-4´ linkages present in this lignin. Ether linkages are
v g k k g w k g β–β´ β-5´ and
β-1´) resist cooking conditions (Gierer, 1985; Gierer and Norén, 1980; Ibarra et al.,
2007). Therefore, it is foreseen that the wood from the hybrid G1 × UGL will be
more easily delignifiable than the other selected Brazilian eucalypt hybrids.

Morphological characteristics of the wood components in the different

eucalypt hybrids

According to the MorFi analysis (Fig. 5), the fiber length and width
distributions for the different eucalypt hybrids were quite equivalent, except for the
IP hybrid. For all the eucalypts, fiber length varied between 350 and 3000 µm and
the fiber width between 12 and 23 µm. The mean area-weighted length versus
vessels content for the hybrid eucalypt woods is depicted in Fig. 6. Eucalypt
hybrids U1 × U2 and G1 × UGL presented the highest vessel content, while DG ×
U2 present the best compromise between fiber length and vessels content, with
longer fibers and the lower vessel content. The difference in vessel content among
the selected eucalypt hybrids was not correlated to the wood growth, but most
probably to the growing conditions and genetic variability. Microscopic
examination of cross sections of wood samples confirmed that hybrid DG × U2
contained fewer vessels with a less marked growing ring, while U1 × U2 contained
more vessels (Fig. 7). Eucalypt hybrid DG × U2 presented two families of vessels:
early wood vessels with a square form and latewood vessels with a high
length/width ratio (Fig. 8). On the contrary, hybrid U1 × U2 contained more early
wood vessels. The early wood vessels are more detrimental to papermaking than
latewood vessels because they are not degraded during pulping, bleaching, refining
and papermaking, whereas the latewood ones are reduced into fragments that can

124
 V. RESULTADOS Y DISCUSIÓN

be easily retained in the sheet thickness. This confirms the interest for the DG × U2
eucalypt hybrid for paper pulp manufacture.

35 35

30 30

25 25
Fibre content,%

Fibre content,%
20
20

15
15

10
10

5
5

0
0
]
0]

0]

0]

0]
50 ]
[5 00]

00 ]
00 00]

->
50

[7 7 5 0

0
[2 200

[3 300

[5 500

00 00

00

[1 ]

[2 ]

[3 ]

[4 ]

[5 ]

]
>]
7]

9]

[1 ]
00

17

23

30

41

56

75
12
-3

-5

-
7

5-
10
-

-1
00

00

[5

[7

2-

7-

3-

0-

1-

6-
0-

0-

0-

0-

-
50

[9

[7
0-
00

00

00
[2

[3

[1
[1

[7

Fibre length, µm Fibre width, µm

IP DG x U2 G1 x UGL U1 x U2 IP DG x U2 G1 x UGL U1 x U2

Figure 5. Distribution of the fiber length and fiber width for the hybrid eucalypt woods.

820
Mean area-weighted length, µm

800

780

760

740

720

700
0 2500 5000 7500 10000 12500 15000

Vessel content, nb/g of pulp

IP DG x U2 G1 x UGL U1 x U2

Figure 6. Mean area-weighted length versus vessels content for the hybrid eucalypt woods.

125
V. RESULTADOS Y DISCUSIÓN

Sample: DGxU2 Sample: U1xU2

Figure 7. Microscopy examination of the cross sections of DG × U2 and U1 × U2 eucalypt

wood chips.

0.50-0.75 0.50-0.75
0.45-0,50 0.45-0,50 Sample: U1xU2
Sample: DGxU2
0.40-0.45 0.40-0.45
Length, mm

Length, mm

0.35-0.40 0.35-0.40
0.30-0.35 0.30-0.35
0.25-0.30 0.25-0.30

0.20-0.25 0.20-0.25
141-174 174-215 215-266
114-141 174-215
Width, µm
Width, µm

Figure 8. Vessel distribution for DG × U2 and U1 × U2 eucalypt wood chips.

Fiber flexibility is another important parameter for papermaking. The curl index,
as measured by the MorFi analyzer, is an indirect approach of the fiber flexibility,
the lower the curl index, the higher the fiber flexibility. DG × U2 and U1 × U2
eucalypt hybrids produce more flexible fibers than the other hybrids (Fig. 9). As
the lignin structure is more condensed for DG × U2 and U1 × U2 hybrids, as seen
above, the delignification affected more severely the fiber wall structure, rendering
the fibers more flexible when the same delignification occurred. The higher the
fiber flexibility, the higher the pulp strengths.

The hybrid IP has lower content of lignin and acetone extractives compared to
the other hybrids and thus, similar cooking conditions will produce higher
delignification rate and higher carbohydrate degradation and hence will generates
fibers more sensitive to mechanical action. This was illustrated by the higher

126
 V. RESULTADOS Y DISCUSIÓN

broken fibers and fines content for this hybrid (Fig. 10). This indicates that the
eucalypt hybrid IP would have a slightly different behavior during pulping and the
fibers would be more sensitive to cutting into the pumps and the mixers.

Finally, the bonding potential of the fibers for papermaking was evaluated by
the way of the relative bonded area index (Fig. 11). Eucalypt hybrids DG × U2 and
IP fibers presented the highest index, confirming that the hybrid DG × U2 was the
most promising for paper pulp manufacture based on morphological properties.

5
4.8
Mean fibre curl index,%

4.6
4.4
4.2
4
3.8
3.6
3.4
3.2
3
IP DG x U2 G1 x UGL U1 x U2

Figure 9. Curl index of the fibers of the different eucalypt hybrid clones, allowing to give
information about the fiber flexibility.

19 14
18
12
Broken fibre content,%

17
Fine content,% in area

10
16
15 8

14 6
13
4
12
2
11

10 0
IP DG x U2 G1 x UGL U1 x U2 IP DG x U2 G1 x UGL U1 x U2

Figure 10. Broken fibers and fines content of the different eucalypt hybrid woods.

127
V. RESULTADOS Y DISCUSIÓN

95
90

Relative bnd area index

85
80
75
70
65
60
55
50
IP DG x U2 G1 x UGL U1 x U2

Figure 11. Relative bonded area index of the fibers from the different eucalypt hybrid
woods.

CONCLUSIONS

The morphological and chemical characteristics of the woods from several

eucalypt hybrids grown in Brazil were studied. The lipid and lignin content of these
Brazilian woods were very similar, but a thorough analytical study indicated some
differences in their composition. The detailed analysis of the composition of the
lipophilic extractives indicated the presence of comparatively high amounts of
neutral compounds in these woods, and particularly, the high abundances of free
and conjugated sterols, which have a high propensity to form pitch deposits, would
point to a high pitch deposition tendency of the lipophilics from these woods.
Among them, the wood from U1 × U2 had the lowest content of these detrimental
compounds and therefore is less prone to pitch problems, while the woods IP and
DG × U2 have the highest content of them, with greater pitch potential than the
hybrid U1 × U2. The lignin content of the different eucalypt hybrids, estimated as
Klason lignin, is similar in all cases (ca. 24–25%). However, some differences were
found in the lignin composition, with the lignin from the hybrids IP, U1 × U2 and
DG × U2 having similar and lower S/G ratios and higher abundance of condensed
linkages, and the lignin from the hybrid G1 × UGL presenting the highest S/G ratio
and the hig f β-O-4´ linkages. This composition
makes the wood from the hybrid G1 × UGL more easily delignifiable under kraft
cooking than the other eucalypt hybrids. However, based on dimensional and
morphological characterization of the fibers and the vessels, eucalypt hybrid DG ×
U2 seemed to be the most interesting raw material for pulp manufacture. It

128
 V. RESULTADOS Y DISCUSIÓN

presented the highest forest productivity, longer and flexible fibers with a high
bonding potential, and the lowest vessels and fines contents.

ACKNOWLEDGEMENTS

This study has been funded by the EU-project LIGNODECO (KBBE-244362),

the Spanish project AGL2008-00709, and the CSIC project 201040E075. Pepijn
Prinsen thanks the Spanish Ministry of Science for a FPI fellowship.

REFERENCES

Back, E.L., Allen, L.H., 2000. Pitch Control, Wood Resin and Deresination. Tappi
Press, Atlanta.

Balakshin, M.Y., Capanema, E.A., Chen, C.-L., Gracz, H.S., 2003. Elucidation of
the structures of residual and dissolved pine kraft lignins using an HMQC NMR
technique. J. Agric. Food Chem. 51, 6116–6127.

Baumberger, S., Fasching, M., Gellerstedt, G., Gosselink, R., Hortling, B., Li, J.,
Saake, B., de Jong, E., 2007. Molar mass determination of lignins by size-
exclusion chromatography: towards standardisation of the method.
Holzforschung 61, 459–468.

Björkman, A., 1956. Studies on finely divided wood. Part I. Extraction of lignin
with neutral solvents. Sven. Papperstidn. 13, 477–485.

Capanema, E.A., Balakshin, M.Y., Chen, C.-L., Gratzl, J.S., Gracz, H., 2001.
Structural analysis of residual and technical lignins by 1H–13C correlation 2D-
NMR spectroscopy. Holzforschung 55, 302–308.

Capanema, E.A., Balakshin, M.Y., Kadla, J.F., 2004. A comprehensive approach

for quantitative lignin characterization by NMR spectroscopy. J. Agric. Food
Chem. 52, 1850–1860.

Capanema, E.A., Balakshin, M.Y., Kadla, J.F., 2005. Quantitative characterization

of a hardwood milled wood lignin by nuclear magnetic resonance spectroscopy.
J. Agric. Food Chem. 53, 9639–9649.

129
V. RESULTADOS Y DISCUSIÓN

Chang, H.-M., Sarkanen, K.V., 1973. Species variation in lignin. Effect of species
on the rate of kraft delignification. Tappi 56, 132–134.

Das, S., Cresson, T., Couture, R., 2003. New pulp characterization from drainage,
fiber flexibility & RBA. In: 85th Annual Meeting, CPPA, Montreal, Quebec,
Canada, pp. 345–347, Pre-prints, Vol. A.

del Río, J.C., Gutiérrez, A., González-Vila, F.C., Martín, F., Romero, J., 1998.
Characterization of organic deposits produced in kraft pulping of Eucalyptus
globulus wood. J. Chromatogr. A 823, 457–465.

del Río, J.C., Romero, J., Gutiérrez, A., 2000. Analysis of pitch deposits produced
in kraft pulp mills using a totally chlorine free bleaching sequence. J.
Chromatogr. A 874, 235–245.

del Río, J.C., Rodríguez, I.M., Gutiérrez, A., 2004. Identification of intact
longchain p-hydroxycinnamate esters in leaf fibers of abaca (Musa textilis)
using gas chromatography/mass spectrometry. Rapid Commun. Mass Spectrom.
18, 2691–2696.

del Río, J.C., Gutiérrez, A., Hernando, M., Landín, P., Romero, J., Martínez, A.T.,
2005. Determining the influence of eucalypt lignin composition in paper pulp
yield using Py-GC/MS. J. Anal. Appl. Pyrolysis 74, 110–115.

del Río, J.C., Rencoret, J., Marques, G., Gutiérrez, A., Ibarra, D., Santos, J.I.,
Jiménez-Barbero, J., Zhang, L., Martínez, A.T., 2008. Highly acylated
(acetylated and/or p-coumaroylated) native lignins from diverse herbaceous
plants. J. Agric. Food Chem. 56, 9525–9534.

del Río, J.C., Rencoret, J., Marques, G., Li, J., Gellerstedt, G., Jiménez-Barbero, J.,
Martínez, A.T., Gutiérrez, A., 2009. Structural characterization of the lignin
from jute (Corchorus capsularis) fibers. J. Agric. Food Chem. 57, 10271–
10281.

Evtuguin, D.V., Neto, C.P., Silva, A.M.S., Domingues, P.M., Amado, F.M.L.,
Robert, D., Faix, O., 2001. Comprehensive study on the chemical structure of
dioxane lignin from plantation Eucalyptus globulus wood. J. Agric. Food Chem.
49, 4252–4261.

Eymin-Petot-Tourtollet, G., Cottin, F., Cochaux, A., Petit-Conil, M., 2003. The use
of MorFi analyser to characterise mechanical pulps. In: International

130
 V. RESULTADOS Y DISCUSIÓN

Mechanical Pulping Conference, Quebec city, Quebec, Canada, Proceedings,

pp. 225–232.

Faix, O., Meier, D., Fortmann, I., 1990. Thermal degradation products of wood. A
collection of electron-impact (EI) mass spectra of monomeric lignin derived
products. Holz Roh-Werkstoff 48, 351–354.

Freire, C.S.R., Silvestre, A.J.D., Neto, C.P., 2002. Identification of new hydroxy
fatty acids and ferulic acid esters in the wood of Eucalyptus globulus.
Holzforschung 56, 143–149.

Freire, C.S.R., Silvestre, A.J.D., Neto, C.P., 2005. Lipophilic extractives in

Eucalyptus globulus kraft pulps. Behaviour during ECF bleaching. J. Wood
Chem. Technol. 25, 67–80.

Freire, C.S.R., Silvestre, A.J.D., Neto, C.P., Evtuguin, D.V., 2006a. Effect of
oxygen, ozone and hydrogen peroxide bleaching stages on the contents and
composition of extractives of Eucalyptus globulus kraft pulps. Bioresour.
Technol. 97, 420–428.

Freire, C.S.R., Pinto, P.C.R., Santiago, A.S., Silvestre, A.J.D., Evtuguin, D.V.,
Neto, C.P., 2006b. Comparative study of lipophilic extractives of hardwoods
and corresponding ECF bleached kraft pulps. Bioresources 1, 3–17.

Gierer, J., 1985. Chemistry of delignification. Part I: general concept and reactions
during pulping. Wood Sci. Technol. 19, 289–312.

Gierer, J., Norén, I., 1980. On the course of delignification during Kraft pulping.
Holzforschung 34, 197–200.

González-Vila, F.J., Almendros, G., del Río, J.C., Martín, F., Gutiérrez, A.,
Romero, J., 1999. Ease of delignification assessment of different Eucalyptus
wood species by pyrolysis (TMAH)-GC/MS and CP/MAS 13C-NMR
spectrometry. J. Anal. Appl. Pyrolysis 49, 295–305.

Grattaplagia, D., 2003. Genolyptus. In: Borém, A., Gindice, M., Sediyama, T.
(Eds.), Mehoramento Genômico. Editora de UFV, Viçosa, MG, Brazil, pp. 51–
72.

Grattaplagia, D., 2004. Integrating genomics into Eucalyptus breeding. Genet. Mol.
Res. 3, 369–379.

131
V. RESULTADOS Y DISCUSIÓN

Gutiérrez, A., del Río, J.C., 2001. Gas chromatography/mass spectrometry

demonstration of steryl glycosides in eucalypt wood, kraft pulp and process
liquids. Rapid Commun. Mass Spectrom. 15, 2515–2520.

Gutiérrez, A., del Río, J.C., González-Vila, F.J., Martín, F., 1998. Analysis of
lipophilic extractives from wood and pitch deposits by solid-phase extraction
and gas chromatography. J. Chromatogr. A 823, 449–455.

Gutiérrez, A., del Río, J.C., González-Vila, F.J., Martín, F., 1999. Chemical
composition of lipophilic extractives from Eucalyptus globulus Labill. wood.
Holzforschung 53, 481–486.

Gutiérrez, A., Romero, J., del Río, J.C., 2001a. Lipophilic extractives in process
waters during manufacturing of totally chlorine free Kraft pulp from eucalypt
wood. Chemosphere 44, 1237–1242.

Gutiérrez, A., Romero, J., del Río, J.C., 2001b. Lipophilic extractives from
Eucalyptus globulus pulp during kraft cooking followed by TCF and ECF
bleaching. Holzforschung 55, 260–264.

Ibarra, D., Chávez, M.I., Rencoret, J., del Río, J.C., Gutiérrez, A., Romero, J.,
Camarero, S., Martinez, M.J., Jiménez-Barbero, J., Martínez, A.T., 2007. Lignin
modification during Eucalyptus globulus kraft pulping followed by totally
chlorine free bleaching: a two-dimensional nuclear magnetic resonance, Fourier
transform infrared, and pyrolysis-gas chromatography/mass spectrometry study.
J. Agric. Food Chem. 55, 3477–3499.

Liitiä, T.M., Maunu, S.L., Hortling, B., Toikka, M., Kilpeläinen, I., 2003. Analysis
of technical lignins by two- and three-dimensional NMR spectroscopy. J. Agric.
Food Chem. 51, 2136–2143.

Marques, G., Rencoret, J., Gutiérrez, A., del Río, J.C., 2010. Evaluation of the
chemical composition of different non-woody plant fibers used for pulp and
paper manufacturing. Open Agric. J. 4, 93–101.

Martínez, A.T., Rencoret, J., Marques, G., Gutiérrez, A., Ibarra, D., Jiménez-
Barbero, J., del Río, J.C., 2008. Monolignol acylation and lignin structure in
some nonwoody plants: a 2D-NMR study. Phytochemistry 69, 2831–2843.

Ralph, J., Hatfield, R.D., 1991. Pyrolysis-GC/MS characterization of forage

materials. J. Agric. Food Chem. 39, 1426–1437.

132
 V. RESULTADOS Y DISCUSIÓN

Ralph, J., Landucci, L.L., 2010. NMR of lignins. In: Heitner, C., Dimmel, D.R.,
Schmidt, J.A. (Eds.), Lignin and Lignans; Advances in Chemistry. CRC Press
(Taylor & Francis Group), Boca Raton, pp. 137–234.

Ralph, J., Marita, J.M., Ralph, S.A., Hatfield, R.D., Lu, F., Ede, R.M., Peng, J.,
Quideau, S., Helm, R.F., Grabber, J.H., Kim, H., Jimenez-Monteon, G., Zhang,
Y., Jung, H.-J.G., Landucci, L.L., MacKay, J.J., Sederoff, R.R., Chapple, C.,
Boudet, A.M., 1999. Solution-state NMR of lignin. In: Argyropoulos, D.S.
(Ed.), Advances in Lignocellulosics Characterization. Tappi Press, Atlanta, pp.
55–108.

Rencoret, J., Gutiérrez, A., del Río, J.C., 2007. Lipid and lignin composition of
woods from different eucalypt species. Holzforschung 61, 165–174.

Rencoret, J., Marques, G., Gutiérrez, A., Ibarra, D., Li, J., Gellerstedt, G., Santos,
J.I., Jiménez-Barbero, J., Martínez, A.T., del Río, J.C., 2008. Structural
characterization of milled wood lignin from different eucalypt species.
Holzforschung 62, 514–526.

Rencoret, J., Gutiérrez, A., Nieto, L., Jiménez-Barbero, J., Faulds, C.B., Kim, H.,
Ralph, J., Martínez, A.T., del Río, J.C., 2011. Lignin composition and structure
in young versus adult Eucalyptus globulus plants. Plant Physiol. 155, 667–682.

SBS. Sociedade Brasileira de Silvicultura. 2007. www.sbs.org.br.

Silvério, F.O., Barbosa, L.C.A., Maltha, C.R.A., Silvestre, A.J.D., Pilo-Veloso, D.,
Gomide, J.L., 2007. Characterization of lipophilic wood extractives from clones
of Eucalyptus urograndis cultivated in Brazil. Bioresources 2, 157–168.

Silvestre, A.J.D., Pereira, C.C.L., Pascoal Neto, C., Evtuguin, D.V., Duarte, A.C.,
Cavaleiro, J.A.S., Furtado, F.P., 1999. Chemical composition of pitch deposits
from an ECF Eucalyptus globulus bleached kraft pulp mill: its relationship with
wood extractives and additives in process streams. Appita J. 52, 375–382.

STCP. Engenharia de Projetos Ltda. 2007. www.stcp.com.br.

Tappi Test Methods 2004–2005, Tappi Press, Norcoss, USA.

Trabado, G.I., Wilstermann, D., 2008. Eucalyptus universalis. Global Cultivated

Eucalypt Forest Map 2008. Available from: http://www.git-forestry.com/
(viewed May 2011).

133
V. RESULTADOS Y DISCUSIÓN

Tsutsumi, Y., Kondo, R., Sakai, K., Imamura, H., 1995. The difference of
reactivity between syringyl lignin and guaiacyl lignin in alkaline systems.
Holzforschung 49, 423–428.

134
 V. RESULTADOS Y DISCUSIÓN

Publicación II:

Prinsen P., Gutiérrez A. and del Río J.C. (2012). Lipophilic extractives from the
cortex and pith of elephant grass (Pennisetum purpureum Schumach.) stems.
Journal of Agricultural and Food Chemistry, 60, 6408−6417.

135
 V. RESULTADOS Y DISCUSIÓN

Lipophilic extractives from the cortex and pith of elephant grass

(Pennisetum purpureum Schumach.) stems

Pepijn Prinsen, Ana Gutiérrez and José C. del Río

Instituto de Recursos Naturales y Agrobiología de Sevilla, CSIC, P.O. Box 1052,

E- 41080 Seville, Spain

ABSTRACT

The composition of lipophilic extractives in the cortex and pith of elephant grass
(Pennisetum purpureum Schumach.) stems was thoroughly studied by gas
g y− y w f y
followed by sterols (in free and conjugated forms as esters and glycosides). Other
steroid compounds, as steroid hydrocarbons and ketones, were also present.
Additionally, important amounts of mono-, di-, and triglycerides were identified.
Other aliphatic series such as n-alkanes, n-fatty alcohols, and n-alkyl ferulates,
together with tocopherols and a series of high molecular weight esters, were also
found, although in minor amounts. The analyses also revealed the presence of a β-
diketone (12,14-tritriacontanedione), which was particularly abundant in the cortex.
Finally, two lignans, matairesinol and syringaresinol were also detected. In general
terms, the abundances of the different classes of compounds were higher in the
pith, except for the series of n-fatty alcohols, n-alkyl ferulates, β-diketones, and
lignans, which were more prominent in the cortex.

KEYWORDS: elephant grass, Pennisetum purpureum, lipids, fatty acids, sterols,

β-diketones, lignans.

INTRODUCTION

Elephant grass (Pennisetum purpureum Schumach.), also called Napier grass, is a

species from the Poaceae (alt. Gramineae), native to the tropical grasslands of
Africa and now introduced into most tropical and subtropical countries. It should
not be confused with the species Miscanthus × giganteus, which is also sometimes

137
V. RESULTADOS Y DISCUSIÓN

called elephant grass. The species is a robust grass with perennial stems, reaching
over 3 m high, and is widely recognized as having the highest biomass productivity
among herbaceous plants, attaining up to 45 t per ha per year.1 Elephant grass is
therefore considered to be an excellent feedstock to provide abundant and
sustainable resources of lignocellulosic biomass for the production of energy,
industrial chemicals, and/or pulp and paper.2,3 Detailed studies on the chemical
composition of elephant grass are important for optimizing the industrial use of this
raw material. Hence, some preliminary studies on Napier grass fibers have been
reported in the literature,4,5 although a complete and thorough characterization of
the chemical composition of this crop has not been performed so far. A previous
paper from our group has already described in detail the characteristic of the lignin
polymer of elephant grass;6 however, to our knowledge, no previous studies have
been published dealing with the composition of the lipophilic extractives in this
species. Previous papers have, however, described the lipophilic extractives in
related species from the Poaceae, M. × giganteus7 and Arundo donax.8

Extractives are a heterogeneous group of compounds dissolving in organic

solvents (such as acetone or dichloromethane). Lipophilic extractives from
lignocellulosic materials are composed mainly of fatty and resin acids, fatty acid
esters (e.g., steryl esters, waxes, triglycerides), and neutral compounds such as fatty
alcohols, sterols, and sterol glycosides.9 The amount and composition of extractives
vary markedly with the plant species. Plant extractives present a special relevance,
because the low degradability of most of these compounds causes problems during
the industrial processing of lignocellulosic materials, that is, pitch deposition
during paper pulp manufacturing.9−14 On the other hand, in the so-called
biorefinery concept, all organic fractions derived from lignocellulosic materials can
potentially be used, including the lipophilic fractions, which can be also a highly
valuable source of phytochemicals.7

In this paper, we have carried out a thorough analytical study of the composition
of the lipophilic fraction of elephant grass, which has not yet been addressed. For
this purpose, the elephant grass stems were separated into the cortex and the pith,
and these two fractions were analyzed independently. The analyses were carried
out by gas chromatography (GC) and g g y− y
(GC-MS) using short and medium-length high-temperature capillary columns,
respectively, with thin films, according to the method previously described that
enables the elution and analysis of a wide range of compounds, from fatty acids to
intact high molecular weight lipids.15 The detailed identification of the lipophilic
components of elephant grass is therefore an important step in the search for new

138
 V. RESULTADOS Y DISCUSIÓN

solutions to control pitch deposition as well as to evaluate their potential use as a

source of high added value phytochemicals.

MATERIALS AND METHODS

Samples

Elephant grass (P. purpureum Schumach.), cultivar ’ w y

v y f g w -
y x x
w g k f k K k y and
subsequently extracted with acetone in a Soxhlet apparatus for 8 h. The acetone
extracts were evaporated to dryness and redissolved in chloroform for
chromatographic analysis of the lipophilic fraction. Duplicates were used for each
sample.

GC and GC-MS Analyses

An HP 5890 gas chromatograph (Hewlett-Packard, Hoofddorp, The

Netherlands) equipped with a − j f
(FID) was used for GC analyses. The injector and detector temperatures were set at
300 and 350 °C, respectively. Samples were injected in the splitless mode. Helium
was used as the carrier gas. The capillary column used was a high-temperature,
polyimide-coated fused silica tubing DB5-H 5 ×0 5 01 μ f
thickness; J&W Scientific). The oven was temperature-programmed from 100 °C
(held for 1 min) to 350 °C (held for 3 min) at 15 °C min−1. Peaks were quantified
by area, and a mixture of standards (tetradecane, palmitic acid, cholest-4-en-3-one,
sitosterol, cholesteryl stearate, sitosteryl 3β-D-glucopyranoside, and
glyceroltriheptadecanoate) was used to elaborate calibration curves. The data from
the two replicates were averaged. In all cases, the standard deviations from
replicates were below 10% of the mean values.

The GC-MS analyses were performed on a Varian Star 3400 gaschromatograph

(Varian, Walnut Creek, CA, USA) coupled with an ion-trap detector (Varian
Saturn) equipped with a high-temperature capillary column (DB-5HT, 15 m × 0.25
01μ f k ; &W S f H w g
a rate of 2 mL/min. The oven was heated from 120 °C (held for 1 min) to 380 °C
(held for 5 min) at 10 °C min−1. The temperature of the injector during the injection
was 120 °C and 0.1 min after injection was programmed to 380 °C at a rate of 200
°C min−1 and held for 10 min. The temperature of the transfer line was set at 300

139
V. RESULTADOS Y DISCUSIÓN

°C. Bis(trimethylsilyl)trifluoroacetamide (BSTFA) silylation was used when

required to form the TMS ether derivatives. Compounds were identified by
comparing their mass spectra with mass spectra in the Wiley and NIST libraries, by
mass fragmentography, and, when possible, by comparison with authentic
standards.

RESULTS AND DISCUSSION

The elephant grass stem consisted of two differentiated parts, a cortex and a
pith, that accounted for 84 and 16% (w/w) of the whole stem, respectively. The
chemical composition of the lipophilic extractives was studied separately in both
parts of the stem. The abundance of the main constituents (water-solubles, acetone
extractives, Klason lignin, acid soluble lignin, holocellulose, α-cellulose, and ash)
in the cortex and pith of elephant grass stem has been detailed in a previous paper.6
The total acetone extractives in the cortex and pith of elephant grass account for 1.7
and 2.3% of dry material, respectively. However, the lipophilic content, estimated
as the chloroform-soluble fraction, is lower and accounts for only 0.8 and 1.6%,
respectively. This content is similar to that found in the related species from the
Poaceae giant reed,8 and M. × giganteus.7 Interestingly, in M. × giganteus, a
higher abundance of lipophilic extractives was also found in the pith compared
with the cortex.7 The content is also similar to that found in other nonwoody fibers,
such as flax,16,17 hemp,18 kenaf,19 sisal,20 abaca,21 or jute.22

The composition of the lipophilic extracts in the cortex and pith of elephant
grass was analyzed in detail by GC and GC-MS according to previously developed
protocols.15 The GC-MS chromatograms of the TMS ether derivatives of the
extracts from the cortex and pith of elephant grass are shown in Figure 1. The
identities and abundances of the main compounds identified are detailed in Table
1.

The predominant lipophilic compounds in the cortex and pith of elephant grass
were a series of free fatty acids that accounted for nearly 50% of all identified
compounds in the cortex and up to 33% of the compounds in the pith. Sterols (in
free and conjugated forms as esters and glycosides) were also prominent and
accounted for 25% of all compounds in the cortex and 38% of the compounds in
the pith. Other steroid compounds present were steroid hydrocarbons and ketones.
Additionally, significant amounts of glycerides (mono-, di-, and triglycerides) and
a β-diketone were also identified among the extracts. Other aliphatic series such as

140
 V. RESULTADOS Y DISCUSIÓN

n-alkanes, n-fatty alcohols, n-alkyl ferulates, and high molecular weight wax esters,
together with tocopherols, occurred in minor amounts.

2
100% 3
(a)

1
Relative abundance

8
F18:1
+ 6
F18:2 F30
F32 7
M20 4 Fe20
F22 F24
M16 F26 Fe22
F16 F20 M18 SE
F18 Trigl

5 10 15 20 25
Retention time (min)

3
100%
(b)

2
Relative abundance

1
8
F18:1
+
F18:2 6
4
5
M20 SE
F24 7
M16 F30
F22 M18 F26
F16 F18 F20 Fe22 Trigl
Fe20

5 10 15 20 25
Retention time (min)

Figure 1. GC-MS chromatograms of the TMS-ether derivatives of the lipid extracts from
(a) the cortex, and (b) the pith of elephant grass (P. purpureum Schumach.) stems. F(n): n-
fatty acids; M(n): monoglycerides; Fe(n): trans n-alkylferulates; n indicates the carbon
atom number. SE: sterol esters; Trigl: triglycerides. Other compounds reflected in the
chromatograms are: (1) campesterol; (2) stigmasterol; (3) sitosterol; (4) stigmasta-4-en-3-
one; (5) 12,14-tritriacontanone; (6) campesteryl 3β-D-glucopyranoside; (7) stigmasteryl 3β-
D-glucopyranoside; (8) sitosteryl 3β-D-glucopyranoside.

141
V. RESULTADOS Y DISCUSIÓN

Table 1. Composition and Abundance (mg/Kg d.a.f.) of Main Lipophilic

Compounds Identified in the Extracts of Cortex and Pith of Elephant Grass (P.
purpureum Schumach.) Stems.

Compound Cortex Pith

n-Fatty acids 991 1497
n-tetradecanoic acid 18 35
n-pentadecanoic acid 10 20
n-hexadecanoic acid 176 231
n-heptadecanoic acid 19 20
n-octadeca-9,12-dienoic acid 196 514
n-octadec-9-enoic acid 61 116
n-octadecanoic acid 89 100
n-nonadecanoic acid 12 17
n-eicosanoic acid 50 70
n-heneicosanoic acid 10 26
n-docosanoic acid 43 68
n-tricosanoic acid 25 40
n-tetracosanoic acid 61 88
n-pentacosanoic acid 25 31
n-hexacosanoic acid 35 25
n-heptacosanoic acid 5 2
n-octacosanoic acid 40 15
n-nonacosanoic acid 4 9
n-triacontanoic acid 58 35
n-hentriacontanoic acid 5 3
n-docotriacontanoic acid 38 27
n-tritriacontanoic acid 2 1
n-tetratriacontanoic acid 9 4

n-Fatty alcohols 14 4
n-tetracosanol 1 1
n-hexacosanol 4 1
n-octacosanol 9 2

n-Alkanes 14 21
n-tricosane 1 1
n-tetracosane 0 tr.
n-pentacosane 2 3
n-hexacosane tr. tr.
n-heptacosane 1 2
n-octacosane tr. tr.
n-nonacosane 5 7
n-triacontane tr. 1
n-hentriacontane 4 6
n-dotriacontane tr. 1
n-tritriacontane 1 tr.

Tocopherols 10 9
α-tocopherol 5 5
β-tocopherol 5 4

142
 V. RESULTADOS Y DISCUSIÓN

n-Alkylferulates 23 10
trans-eicosanylferulate 13 7
trans-docosanylferulate 10 3

High molecular weight esters a 62 88

esters C38 2 4
esters C40 13 49
esters C42 11 8
esters C44 17 15
esters C46 16 8
etsres C48 3 4

Monoglycerides 150 343

2,3-dihydroxypropyl tetradecanoate 2 2
2,3-dihydroxypropyl hexadecanoate 50 98
2,3-dihydroxypropyl octadecadienoate 21 60
2,3-dihydroxypropyl octadecenoate 6 19
2,3-dihydroxypropyl octadecanoate 15 21
2,3-dihydroxypropyl eicosanoate 15 23
2,3-dihydroxypropyl docosanoate 11 32
2,3-dihydroxypropyl tricosanoate 3 20
2,3-dihydroxypropyl tetracosanoate 13 48
2,3-dihydroxypropyl pentacosanoate 3 7
2,3-dihydroxypropyl hexacosanoate 5 8
2,3-dihydroxypropyl heptacosanoate 1 1
2,3-dihydroxypropyl octacosanoate 2 2
2,3-dihydroxypropyl triacontanoate 3 0

Diglicerides 50 179
1,2-dipalmitin 4 1
1,3-dipalmitin 8 11
1,2-palmitoyllinolein 4 30
1,2-palmitoylolein 4 10
1,2-palmitoylstearin 3 2
1,3-palmitoyllinolein 6 43
1,3-palmitoylolein 5 17
1,3-palmitoylstearin 11 9
1,2-dilinolein 1 22
1,3-dilinolein 2 21
1,2-distearin 0 5
1,3-distearin 2 8

Triglycerides 26 135
dipalmitoyllinolein 6 30
palmitoyldilinolein 13 69
trilinolein 7 36

Steroid hydrocarbons 10 21
ergosta-3,5-diene 4 8
stigmasta-3,5,7-triene 2 4
stigmasta-3,5-diene 4 9

Steroid ketones 89 381

143
V. RESULTADOS Y DISCUSIÓN

ergost-4-en-3-one 10 33
stigmasta-4,22-dien-3-one 17 47
stigmasta-3,5-dien-7-one 2 9
stigmast-4-en-3-one 25 117
ergost-4-ene-3,6-dione 4 22
ergostane-3,6-dione 4 15
stigmasta-4,22-diene-3,6-dione 8 22
stigmast-22-ene-3,6-dione 5 17
stigmast-4-ene-3,6-dione 7 64
stigmastane-3,6-dione 7 35

Sterols 327 1096

campesterol 74 198
stigmasterol 113 291
sitosterol 139 585
7-oxositosterol 1 22

Sterol glycosides 121 435

campesteryl-β-D-glucopyranoside 50 147
stigmasteryl-β-D-glucopyranoside 34 49
sitosteryl-β-D-glucopyranoside 67 239

Sterol esters 54 181

campesteryl esters 11 33
stigmasteryl esters 11 26
sitosteryl esters 32 122

β-Diketones 156 75
12,14-tritriacontanedione 156 75

Lignans 9 0
matairesinol 3 0
syringaresinol 6 0

The structures of the main classes of lipophilic compounds identified in the

cortex and pith of elephant grass are depicted in Figures 2 and 3, and the
distributions of the main aliphatic series are represented in the histograms of
Figure 4. In general terms, the abundances of the different classes of lipophilic
compounds were higher in the pith than in the cortex, as also observed in the
related species M. × giganteus,7 except for the series of n-fatty alcohols, n-alkyl
ferulates, β-diketones, and lignans, which were more abundant in the cortex.

Aliphatic Series

As said above, the most important class of lipophilic compounds observed in the
cortex and pith of elephant grass comprised free fatty acids, which accounted for

144
 V. RESULTADOS Y DISCUSIÓN

991 mg/kg in the cortex and 1497 mg/kg in the pith. Free fatty acids were identified
in the range from tetradecanoic (C14) to tetratriacontanoic (C34) acid, with a strong
even-over-odd carbon atom predominance, hexadecanoic acid (palmitic acid, I)
being the most abundant saturated fatty acid.

OH
(I)
O O

OH OH
(II) (III)

OH
(IV)

(V)
O
O
O
(VI)
HO
O

O OH
(VII) OH
HO
O
O
O
(VIII)O

O O

O O
OH (IX)

O
O O

O (X) O

O
(XI)
O O

(XII)

Figure 2. Structures of compounds representative of the main aliphatic series identified

among the lipophilic extractives in elephant grass (P. pennisetum Schumach.) and referred
in the text. (I) n-hexadecanoic (palmitic) acid; (II) 9-octadecenoic (oleic) acid; (III) 9,12-
octadecadienoic (linoleic) acid; (IV) n-octacosanol; (V) n-hentriacontane; (VI) trans-
eicosyl ferulate; (VII) 2,3-dihydroxypropyl hexadecanoate (1-monopalmitin); (VIII) 1,2-
palmitoyllinolein; (IX) 1,3-palmitoyllinolein; (X) trilinolein; (XI) eicosanoic acid, eicosyl
ester. (XII) 12,14-tritriacontanedione.

145
V. RESULTADOS Y DISCUSIÓN

HO HO HO HO O

(XIII) (XIV) (XV) (XVI)

O OH
O
O HO O
HO
OH
(XXVII) (XVIII)

R R R

O O O O
O O
(XIX) (XX) (XXI) (XXII)

O O O
O O
(XXIII) (XXIV) (XXV) (XXVI)

Figure 3. Structures of the main steroid compounds identified among the lipophilic
extractives in elephant grass (P. pennisetum Schumach.) and referred in the text. (XIII)
sitosterol; (XIV) campesterol; (XV) stigmasterol; (XVI) 7-oxositosterol; (XVII) sitosteryl
linoleate; (XVIII) sitosteryl 3β-D-glucopyranoside; (XIX) ergost-4-en-3-one (R=H),
stigmast-4-en-3-one (R=CH3); (XX) ergost-4-ene-3,6-dione (R=H), stigmast-4-ene-3,6-
dione; (XXI) ergostane-3,6-dione (R=H), stigmastane-3,6-dione (R=CH3); (XXII)
stigmasta-4,22-dien-3-one; (XXIII) stigmasta-3,5-dien-7-one; (XXIV) stigmasta-4,22-
diene-3,6-dione; (XXV) stigmast-22-ene-3,6-dione; (XXVI) stigmasta-3,5-diene.

The unsaturated 9-octadecenoic (oleic acid, II) and 9,12-octadecadienoic

(linoleic acid, III) acids were also present, the latter being the most abundant free
fatty acid in both the cortex and pith. A series of n-fatty alcohols ranging from
tetracosanol (C24) to octacosanol (C28) were present in elephant grass extracts,
albeit in low amounts (14 mg/kg in the cortex and 4 mg/kg in the pith) with the

146
 V. RESULTADOS Y DISCUSIÓN

presence of only the even carbon atom number, octacosanol (IV) being the most
abundant. Fatty alcohols also occurred in higher amounts in the cortex than in the
pith in the related species M. × giganteus.7 A series of n-alkanes ranging from
tricosane (C23) to tritriacontane (C33) also occurred in elephant grass in low
amounts (14 mg/kg in the cortex and 21 mg/kg in the pith) with a strong odd-over-
even carbon atom number predominance, nonacosane (V) being the most abundant
followed by hentriacontane.

200
500
(a) (b)
Abundance (mg/Kg)

Abundance (mg/Kg)

14 16 18 20 22 24 26 28 30 32 34 14 16 18 20 22 24 26 28 30 32 34
18:2
18:1

18:2
18:1
C atom number C atom number

5 7
(c) (d)
Abundance (mg/Kg)
Abundance (mg/Kg)

23 24 25 26 27 28 29 30 31 32 33 23 24 25 26 27 28 29 30 31 32 33
C atom number C atom number

50 100
(e) (f)
Abundance (mg/Kg)

Abundance (mg/Kg)

14 16 18 20 22 24 26 28 30 14 16 18 20 22 24 26 28 30
18:2
18:1

18:2
18:1

C atom number C atom number

Figure 4. Distribution of the main aliphatic series identified in the extracts of elephant
grass (P.pennisetum Schumach.). Series of n-fatty acids in the cortex (a) and in the pith (b);
Series of n-alkanes in the cortex (c) and in the pith (d); Series of monoglycerides in the
cortex (e) and in the pith (f). The histograms are scaled up to the abundance of the major
compound in the series.

147
V. RESULTADOS Y DISCUSIÓN

A series of ferulic acid esterified to long-chain fatty alcohols were also

identified among the extracts of elephant grass and accounted for 25 mg/kg in the
cortex and 10 mg/kg in the pith. Their characterization was achieved on the basis of
the mass spectra obtained by GC-MS of the underivatized and their TMS ether
derivatives.23 The series of n-alkyl ferulates occurred in the range from C20 to C22,
with the presence of only the even carbon atom number homologues, trans-
eicosylferulate (VI) being the most abundant. This series of n-alkyl ferulates
occurred mostly in the trans form, although minor amounts of the cis isomers were
also observed at lower retention times. Alkyl ferulates have been widely reported in
different plant families, 0 1 3− 6 although they were not found in the related species
giant reed8 and M. × giganteus.7

Glycerides (mono-, di-, and triglycerides) were another class of compounds

present in significant amounts in the cortex and pith of elephant grass. Glycerides,
and especially mono- and diglycerides, were reported in the related species from
the Poacea giant reed8 but were not reported among the lipophilic extractives in M.
× giganteus,7 despite the fact that mono- and diglycerides are very common
constituents in many plants.21,27,28 Monoglycerides were the most abundant
glycerides in elephant grass, with 150 mg/kg in the cortex and 343 mg/kg in the
pith. The series of monoglycerides were found in the range from C14 to C30, with
strong even-overodd carbon atom number predominance, the C16 homologue (1-
monopalmitin, VII) being the most abundant, and with the exclusive presence of
the isomer in position 1. The unsaturated monoglycerides 1-monoolein and 1-
monolinolein were also found in significant amounts. Diglycerides were found in
elephant grass in higher abundance in the pith (179 mg/kg) than in the cortex (50
mg/kg), the most abundant being 1,2-palmitoyllinolein (VIII) and 1,3-
palmitoyllinolein (IX). Triglycerides (dipalmitoyllinolein, dilinoleylpalmitin, and
trilinolein, X) were also present in elephant grass, in contrast to other nonwood
fibers such as flax, hemp, and kenaf, which lack triglycerides,16−19 and are present
in higher abundance in the pith (135 mg/kg) than in the cortex (26 mg/kg).

A series of high molecular weight wax esters also occurred in elephant grass
extract, being more abundant in the pith (79 mg/kg) than in the cortex (59 mg/kg).
High molecular weight wax esters were found in the range from C38 to C48 with the
presence of only the even carbon atom number homologues, the C44 and C46
analogues being the most abundant in the cortex, whereas the homologues C40 were
the most abundant in the pith. A close examination of each chromatographic peak
indicated that they consisted of different long-chain fatty acids esterified to
different long-chain fatty alcohols. The identification and quantitation of the

148
 V. RESULTADOS Y DISCUSIÓN

individual long-chain esters in each chromatographic peak were resolved on the

basis of the mass spectra of the peaks. The mass spectra of long-chain esters are
characterized by a base peak produced by a rearrangement process involving the
transfer of two hydrogen atoms from the alcohol chain to the acid chain, giving a
protonated acid ion.29,30 Therefore, the base peak gives information about the
number of carbon atoms in the acid moiety, whereas the molecular ion provides
information about the total number of carbon atoms in the ester. It is possible then
to determine the individual contribution of esters to every chromatographic peak by
mass spectrometric determination of the molecular ion and the base peak.
Quantitation of individual esters was accomplished by integrating the areas in the
chromatographic profiles of the ions characteristic for the acidic moiety. The
detailed structural composition of the different high molecular weight wax esters
identified in elephant grass is shown in Table 2. The esterified fatty acids ranged
from C14 to C22 and the esterified fatty alcohols from C18 to C30. The acyl moiety of
the high molecular weight wax esters was exclusively constituted by saturated fatty
acids with even carbon number. High molecular weight wax esters with unsaturated
fatty acids could be detected only in minor amounts despite the high amounts of
oleic and linoleic acids being present in free form. The predominant high molecular
weight ester in the cortex of elephant grass was C44, which is mostly constituted by
hexadecanoic acid, octacosyl ester, followed by ester C46, mostly constituted by
octadecanoic acid, octacosyl ester. In the pith, the predominant high molecular
weight ester was C40, mostly constituted by eicosanoic acid, eicosyl ester (XI).
Additionally, two tocopherols, namely, α- and β-tocopherols, were found among
the extracts of the cortex and pith of elephant grass, although in minor amounts, ca.
10 mg/kg.

β-Diketones

The analysis of the lipophilic extractives of the cortex and pith of elephant grass
revealed the presence of significant amounts of a compound with a β-diketone
structure. The identification of this compound was achieved on the basis of its mass
spectrum that is depicted in Figure 5. The molecular ion at m/z 492 indicates that
this is a tritriacontanedienone, and the fragments at m/z 222 and 334, arising from
the McLafferty rearrangement at both sides of the diketone group followed by loss
of water,31 clearly indicate that the structure of this compound is 12,14-
tritriacontanedione (XII). β-Diketones are relatively common constituents of plant
waxes and have been identified in the leaves of different grasses.3 −36 In elephant
grass, 12,14-tritriacontanedione was present in higher abundance in the cortex (156
mg/kg) than in the pith (75 mg/kg), in contrast to most of the lipophilic compounds.

149
V. RESULTADOS Y DISCUSIÓN

Table 2. Composition and Abundance (mg/Kg, d.a.f.) of the Different Individual High
Molecular Weight Esters Identified in the Extracts of the Cortex and Pith of Elephant
Grass (P. purpureum Schumach.) Stems.

Compound Fatty acid:Fatty alcohol Cortex Pith

esters C38 2 4
eicosanoic acid, octadecyl ester C20:C18 2 4

esters C40 13 49
hexadecanoic acid, tetracosyl ester C16:C24 1 1
octadec-9-enoic acid, docosyl ester C18:1:C22 3 8
octadeca-9,12-dienoic acid, docosyl ester C18:2:C22 1 2
eicosanoic acid, eicosyl ester C20:C20 6 37
docosanoic acid, octadecyl ester C22:C18 2 1

esters C42 11 8
tetradecanoic acid, octacosyl ester C14:C28 1 2
hexadecanoic acid, hexacosyl ester C16:C26 3 2
octadecanoic acid, tetracosyl ester C18:C24 1 1
eicosanoic acid, docosyl ester C20:C22 1 1
docosanoic acid, eicosyl ester C22:C20 5 2

esters C44 17 15
hexadecanoic acid, octacosyl ester C16:C28 12 12
octadecanoic acid, hexacosyl ester C18:C26 3 1
eicosanoic acid, tetracosyl ester C20:C24 1 1
docosanoic acid, docosyl ester C22:C22 1 1

esters C46 16 8
hexadecanoic acid, triacontyl ester C16:C30 3 2
octadecanoic acid, octacosyl ester C18:C28 11 4
eicosanoic acid, hexacosyl ester C20:C26 1 1
docosanoic acid, tetracosyl ester C22:C24 1 1

esters C48 3 4
octadecanoic acid, triacontyl ester C18:C30 1 1
eicosanoic acid, octacosyl ester C20:C28 1 2
docosanoic acid, hexacosyl ester C22:C26 1 1

Steroid Compounds

Different classes of steroid compounds were present in the extracts of the cortex
and pith of elephant grass, namely, steroid hydrocarbons, steroid ketones, sterols,
sterol glycosides, and sterol esters. Sterols were the most abundant steroid
compounds in elephant grass, accounting for 327 mg/kg in the cortex and 1096
mg/kg in the pith. The particularly higher abundance of free sterols in the pith was
also observed in the related species M. × giganteus by Villaverde et al.,7 which

150
 V. RESULTADOS Y DISCUSIÓN

indicated this as a potential source of highly valuable sterols. Sitosterol (XIII) was
the most important sterol identified in elephant grass extracts, together with
campesterol (XIV), stigmasterol (XV), and minor amounts of 7-oxositosterol
(XVI). Sterols were also found esterified with long-chain fatty acids and were also
more abundant in the pith (181 mg/kg) than in the cortex (54 mg/kg). The
structures of the sterol esters were determined by their mass spectra. However,
sterol esters generally show fragments arising from the sterol moiety by electron
impact MS and rarely give detectable molecular ions.37,38 By monitoring the ions
corresponding to the different sterol moieties, it was possible to identify series of
campesterol, stigmasterol, and sitosterol esters. Figure 6 shows the distribution of
the esterified sterols by monitoring the characteristic fragments for the different
sterol moieties in their mass spectra, m/z 382 for campesterol esters, m/z 394 for
stigmasterol esters, and m/z 396 for sitosterol esters. All of the esterified sterol
ester series showed two major peaks for the esterification with C16 and C18 fatty
acids, including the unsaturated oleic and linoleic acids, the most predominant
sterol ester being sitosterol linoleate (XVII). Sterol glycosides were also identified
in elephant grass in important amounts (121 mg/kg in the cortex and 435 mg/kg in
the pith). Sitosteryl β-D-glucopyranoside (XVIII) was the most predominant with
lower amounts of campesteryl and stigmasteryl β-D-glucopyranosides. The
identification of sterol glycosides was accomplished (after BSTFA derivatization of
the lipid extract) by comparison with the mass spectra and relative retention times
of authentic standards.39

138
100%

492
Relative abundance

95
222

109
474
81 100 334
57 124 164
151 275

100 200 300 400 500 m/z

Figure 5. Mass spectrum of 12,14-tritriacontanedione (VI) identified among elephant grass

(P. purpureum Schumach.) extracts.

151
V. RESULTADOS Y DISCUSIÓN

100%

TIC

30%

m/z 382

50%

m/z 394

100%

m/z 396

19 20 21 22 23 24
Retention time (min)

Figure 6. Single ion chromatograms showing the distribution of the different sterol esters
in the pith of elephant grass (P. pennisetum Schumach.). m/z 382: campesterol ester series;
m/z 394: stigmasterol ester series; m/z 396: sitosterol ester series.

Steroid ketones were identified in relatively high amounts among the lipophilic
extractives of elephant grass, being more abundant in the pith (381 mg/kg) than in
the cortex (89 mg/kg), as also occurred in the related species M. × giganteus.7 The

152
 V. RESULTADOS Y DISCUSIÓN

main steroid ketones identified were ergost-4-en-3-one (XIX, R = H), stigmast-4-

en-3-one (XIX, R = CH3), ergost-4-ene-3,6-dione (XX, R = H), stigmast-4-ene-3,6-
dione (XX, R = CH3), ergostane-3,6-dione (XXI, R = H), stigmastane-3,6-dione
(XXI, R = CH3), stigmasta-4,22-dien-3-one (XXII), stigmasta-3,5-dien-7-one
(XXIII), stigmasta-4,22-diene-3,6-dione (XXIV), and stigmast-22-ene-3,6-dione
(XXV). Finally, several steroid hydrocarbons, such as ergosta-3,5-diene, stigmasta-
3,5,7-triene, and stigmasta-3,5-diene (XXVI), were also found in minor amounts
among the elephant grass extractives, being more abundant in the pith (23 mg/kg)
than in the cortex (10 mg/kg).

Lignans

Finally, two lignans, namely matairesinol and syringaresinol (Figure 7), could
be identified among the lipophilic extractives of elephant grass. Interestingly, these
lignans could be observed only in the cortex, although in low amounts (9 mg/kg),
and were completely absent in the pith. Lignans have been described as powerful
antioxidants with metal-chelating properties and therefore have been attributed a
protective effect against fungal attack.40

OCH3

O OH
H
CH3O
O
O OCH3
HO H H
H
CH3O
O

OCH3 HO
OH OCH3

matairesinol syringaresinol

Figure 7. Structure of the lignans matairesinol and syringaresinol identified in the extracts
of elephant grass (P. purpureum Schumach.).

In conclusion, this study provides for the first time a comprehensive and detailed
chemical characterization of the lipophilic extractives in elephant grass, which is
highly valuable information for improving the industrial uses of this interesting
fast-growing crop

153
V. RESULTADOS Y DISCUSIÓN

FUNDING

This study has been funded by the Spanish project AGL2011-25379, the CSIC
project 201040E075, and the EU project LIGNODECO (KBBE-244362). P.P.
thanks the Spanish Ministry of Science and Innovation for a FPI fellowship.

ACKNOWLEDGMENTS

We thank Prof. Jorge L. Colodette and Prof. Jose L. Gomide (University of

Viçosa, Brazil) for providing the elephant grass.

REFERENCES

(1) Woodard, K. R.; Prine, G. M. Dry matter accumulation of elephant grass,

energy cane and elephant millet in a subtropical climate. Crop Sci. 1993, 33,
818−8 4

(2) Somerville, C.; Youngs, H.; Taylor, C.; Davis, S. C.; Long, S. P. Feedstocks
for lignocellulosic biofuels. Science 2010 3 9 790−79

(3) Madakadze, I. C.; Masamvu, T. M.; Radiotis, T.; Li, J.; Smith, D.L.
Evaluation of pulp and paper making characteristics of elephant grass (Pennisetum
purpureum Schum) and switchgrass (Panicum virgatum L.). Afr. J. Environ. Sci.
Technol. 2010 4 465−467

(4) Reddy, K. O.; Maheswari, C. M.; Reddy, D. J. P.; Rajulu, A. V. Thermal

properties of Napier grass fibers. Mater. Lett. 2009 63 390− 39

(5) Reddy, K. O.; Maheswari, C. U.; Shukla, M.; Rajulu, A. V. Chemical

composition and structural characterization of Napier grass fibers. Mater. Lett.
2012 67 35−38

6 ; ; ; ; - Barbero, J.;
Ralph, J.; Martínez ; S characterization of the lignin in
the cortex and pith of elephant grass (Pennisetum purpureum) stems. J. Agric.
Food Chem. 2012 60 3619−3634

(7) Villaverde, J. J.; Domingues, R. M. A.; Freire, C. S. R.; Silvestre, A. J. D.;

Pascoal Neto, C.; Ligero, P.; Vega, A. Miscanthus × giganteus extractives: a
source of valuable phenolic compounds and sterols. J.Agric. Food Chem. 2009, 57,
36 6−3631

154
 V. RESULTADOS Y DISCUSIÓN

8 ; ; ;S v ; Río, J. C.
Chemical characterization of the lipophilic fraction of Giant reed (Arundo donax)
fibres used for pulp and paper manufacturing. Ind. Crops Prod. 2007 6 9− 36

(9) Back, E. L.; Allen, L. H. Pitch Control, Wood Resin and Deresination;
TAPPI Press: Atlanta, GA, 2000.

10 ; ; -Vila, F. J.; Martín, F.; Romero, J.

Characterization of organic deposits produced in the Kraft pulping of Eucalyptus
globulus wood. J. Chromatogr., A 1998 8 3 457−465

(11) del Río, J. C.; Rome ; y f deposits

produced in kraft pulp mills using totally chlorine free bleaching sequences. J.
Chromatogr., A 2000 874 35− 45

1 ; f deposits
produced in the manufacturing of high-quality paper pulps from hemp fibers.
Bioresour. Technol. 2005 96 1445−1450

(13) Silvestre, A. J. D.; Pereira, C. C. L.; Neto, C. P.; Evtuguin, D. V.; Duarte,
A. C.; Cavaleiro, J. A. S.; Furtado, F. P. Chemical composition of pitch deposits
from ECF Eucalyptus globulus bleached kraft pulp mill: its relationship with wood
extractives and additives in process streams. Appita J. 1999 5 375−38

14 ; ; x v from
several nonwoody lignocellulosic crops (flax, hemp, sisal, abaca) and their fate
during alkaline pulping and TCF/ECF bleaching. Bioresour. Technol. 2010, 101,
60− 67

15 ; ; -Vila, F. J.; Martín, F. Analysis of

lipophilic extractives from wood and pitch deposits by solid-phase extraction and
gas chromatography. J. Chromatogr., A 1998 8 3 449−455

16 ; f f xf f in alkaline
pulping. J. Agric. Food Chem. 2003, 51, 4965−4971

17 ; f f xf f in alkaline
pulping (addition/correction). J. Agric. Food Chem. 2003, 51 6911−6914

155
V. RESULTADOS Y DISCUSIÓN

18 ; g I ; characterization of
lignin and lipid fractions in industrial hemp bast fibers used for manufacturing
high-quality paper pulps. J. Agric. Food Chem. 2006 54 138− 144

19 ; g I ; f
lignin and lipid fractions in kenaf bast fibers used for manufacturing high-quality
papers. J. Agric. Food Chem. 2004 5 4764−4773

0 ; g I; f
lipophilic extractives from sisal (Agave sisalana) fibers. Ind. Crops Prod. 2008, 28,
81−87

1 ; f Musa textilis)
leaf fibers used for manufacturing of high quality paper pulps. J. Agric. Food
Chem. 2006 54 4600−4610

(22) del Río, J. C.; Marques, G.; Rodríguez, I ;

composition of lipophilic extractives from jute (Corchorus capsularis) fibers used
for manufacturing of high-quality paper pulps. Ind. Crops Prod. 2009, 30,
41− 49

3 ; g ; I ification of intact long-

chain p-hydroxycinnamate esters in leaf fibers of abaca (Musa textilis) using gas
chromatography/mass spectrometry. Rapid Commun. Mass Spectrom. 2004, 18,
691− 696

4 ; ; g n composition of
woods from different eucalypt species. Holzforschung 2007 61 165−174

5 K k y ; K y y f f
suberin and associated waxes. In Natural Products of Woody Plants; Rowe, J. W.,
Ed.; Springer- g: y 1989; 1 304−367

6 ; y ; ;W y ; k
acid esters from stem bark of Pavetta owariensis. Phytochemistry 1991, 30,
10 4−10 6

7 ; , A.; del Río, J. C. Chemical characterization of

lignin and lipophilic fractions from leaf fibers of curaua (Ananas erectifolius). J.
Agric. Food Chem. 2007 55 13 7−1336

156
 V. RESULTADOS Y DISCUSIÓN

8 ; ; ; v f
chemical composition of different non-woody plant fibers used for pulp and paper
manufacturing. Open Agric. J. 2010 4 93−101

(29) Moldovan, Z.; Jover, E.; Bayona, J. M. Systematic characterization of long-

chain aliphatic esters of wool wax by gas chromatography-electron impact
ionisation mass spectrometry. J. Chromatogr., A 2002 95 193− 04

(30) Sharkey, A. G., Jr.; Shultz, J. L.; Friedel, R. A. Mass spectra of esters.
Formation of rearrangement ions. Anal. Chem. 1959 31 87−94

(31) Evans, D.; Knights, B. A.; Math, V. B.; Ritchie, A. L. β-Diketones in

Rhododendron waxes. Phytochemistry 1975 14 447− 451

(32) Bianchi, A.; Bianchi, G. Surface lipid composition of C3 and C4 plants.

Biochem. Syst. Ecol. 1990 18 533−537

(33) Bianchi, G.; Figini, M. L. Epicuticular waxes of glaucous and nonglaucous

durum wheat lines. J. Agric. Food Chem. 1986 34 4 9−433

(34) Tulloch, A. P.; Hoffman, L. L. Epicuticular waxes of Secale cereale and

Triticale hexaploide leaves. Phytochemistry 1974 13 535− 540

(35) Tulloch, A. P.; Hoffman, L. L. Epicuticular wax of Apropyron

intermedium. Phytochemistry 1976 15 1145−1151

(36) Tulloch, A. P. Carbon-13 NMR spectra of β-diketones from wax of the

gramineae. Phytochemistry 1985 4 131−137

(37) Lusby, W. R.; Thompson, M. J.; Kochansky, J. Analysis of sterol esters by

capillary gas chromatography electron impact and chemical ionization-mass
spectrometry. Lipids 1984 19 888−901

(38) Evershed, R. P.; Prescott, M. C.; Spooner, N.; Goad, L. J. Negative ion
ammonia chemical ionization and electron impact ionization mass spectrometric
analysis of steryl fatty acyl esters. Steroids 1989 53 85−309

39 ; g y/ y
demonstration of steryl glycosides in eucalypt wood, kraft pulp and process liquids.
Rapid Commun. Mass Spectrom. 2001 15 515− 5 0

157
V. RESULTADOS Y DISCUSIÓN

(40) Donoso-Fierro, C.; Becerra, J.; Bustos-Concha, E; Silva, M. Chelating and

antioxidant activity of lignans from Chilean woods (Cupressaceae). Holzforschung
2009 63 559−563.

158
 V. RESULTADOS Y DISCUSIÓN

Publicación III:

del Río J.C., Prinsen P -

01 S f g
the cortex and pith of elephant grass (Pennisetum purpureum) stems. Journal of
Agricultural and Food Chemistry, 60, 3619−3634.

159
 V. RESULTADOS Y DISCUSIÓN

Structural Characterization of the Lignins Isolated from the

Cortex and Pith of Elephant Grass (Pennisetum purpureum)

José C. del Río†, Pepijn Prinsen†, Jorge Rencoret†‡, Lidia Nieto§, Jesus Jiménez-
Barbero§, John Ralph‡, Angel T. Martínez§ and Ana Gutiérrez†

†
Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS), CSIC, PO
Box 1052, E-41080 Seville, Spain

‡
Departments of Biochemistry and Biological Systems Engineering, the Wisconsin
Bioenergy Initiative, and the DOE Great Lakes Bioenergy Research Center,
University of Wisconsin, Madison, Wisconsin 53706, United States

§
Centro de Investigaciones Biológicas (CIB), CSIC, Ramiro de Maeztu 9, E-28040
Madrid, Spain

ABSTRACT

The structure of the lignin in the cortex and pith of elephant grass (Pennisetum
purpureum w “w ” g y
several analytical methods. The presence of p-coumarate and ferulate in the cortex
and pith, as well as in their isolated lignins, was revealed by pyrolysis in the
presence of tetramethylammonium hydroxide, and by 2D NMR, and indicated that
ferulate acylates the carbohydrates while p-coumarate acylates the lignin polymer.
2D NMR showed a predominance of alkyl aryl ether (β−O−4′ k g 8 % f
k g w w f“ ”
resinols (β−β′ y β−5′ β−1′ v
NMR also indicated that these lignins are extensively acylated at the γ-carbon of
the side chain. DFRC analyses confirmed that p-coumarate groups acylate the γ-
OHs of these lignins, and predominantly on syringyl units.

KEYWORDS: elephant grass, Pennisetum purpureum x y− / S

TMAH, HSQC, DFRC, milled wood lignin, p-coumarate, ferulate, syringyl,
guaiacyl.

161
V. RESULTADOS Y DISCUSIÓN

INTRODUCTION

Renewable sources of energy and consumer products are required for sustainable
development of modern society. Plant biomass is the main source of renewable
materials on Earth and represents a potential source of renewable energy and
biobased products. The substitution of fossil fuels by biomass is an important
contribution to reduce anthropogenic net CO2 emissions. Biomass is available in
high amounts (as forest, agricultural or industrial lignocellulosic wastes and crops)
at relatively low cost and could be a widely available and inexpensive source for
biofuels and bioproducts in the near future. In this sense, there is a growing need to
consider alternative agricultural strategies that move an agricultural industry
focused on food production to one that also supplies the needs of other industrial
sectors, such as pulp and paper, textiles, biofuels, or value-added chemicals, in the
context of the so-called lignocellulose biorefinery. Biorefineries use renewable raw
materials to produce energy together with a wide range of everyday commodities in
an economic and sustainable manner.1,2 Therefore, significant efforts are being put
into the search for highly productive biomass crops, including elephant grass.

Elephant grass (Pennisetum purpureum), also called Napier grass, is a species

from the Poaceae native to the tropical grasslands of Africa and now introduced
into most tropical and subtropical countries. It should not be confused with the
species Miscanthus giganteus, also sometimes called elephant grass. The species is
a robust grass with perennial stems, reaching over 3 m high, and is widely
recognized as having the highest biomass productivity among herbaceous plants,
attaining up to 45 Mg ha−1 y−1,3,4 and therefore has been considered an excellent
alternative feedstock to provide abundant and sustainable resources of
lignocellulosic biomass for the production of biofuels.5

Cell wall polysaccharides can be used as feedstocks for the fermentative

production of bioethanol or other biofuels after being broken down into simple
sugars. However, in addition to cellulose, plant cell walls also contain more
complex hemicellulosic polysaccharides, and an aromatic polymer, lignin, that
hinders the degradation of cell wall polysaccharides to simple sugars. Therefore,
the utilization of a specific energy crop for bioethanol production is strongly
limited by its lignin content, composition, and structure. Biomass pretreatment is an
essential step required to remove or modify the lignin to allow the access to, or
isolation of, the plant cell wall polysaccharides for saccharification and eventual
fermentation. Lignin is a heterogeneous and complex polymer synthesized mainly
from three p-hydroxycinnamyl alcohols differing in their degree of methoxylation:

162
 V. RESULTADOS Y DISCUSIÓN

p-coumaryl, coniferyl and sinapyl alcohols. Each of these monolignols gives rise to
a different type of lignin unit called p-hydroxyphenyl (H), guaiacyl (G) and
syringyl (S) units, respectively, generating a variety of structures and linkages
within the polymer.6−8 The lignin content, composition and structure vary widely
among different plant species, among individuals, and even in different tissues of
the same individual. The lignin composition greatly influences delignification
reactions, and, therefore, structural characterization is an essential step to develop
appropriate methods to design effective lignin depolymerization strategies.

However, there is a lack of studies regarding the chemistry of the lignin of

elephant grass. Previous papers have only reported the lignin contents of different
cultivars of elephant grass and their variation with growth development,9 but did
not provide any information about its composition and structure. In this paper, we
therefore report the detailed chemical composition and structural characteristics of
the lignin in elephant grass. For this purpose, the cortex and the pith of elephant
grass stems were separated manually and analyzed independently by an array of
analytical t g w y y −gas chromatography-mass
y y− / S g y v f
10−13
characterizing the chemical composition of lignin. However, the presence of p-
hydroxycinnamates (p-coumarate and ferulate), which are abundant in grasses,
constitutes a complication for lignin analysis by analytical pyrolysis since they
yield products similar to those of corresponding lignin units due to decarboxylation
reactions. This problem can however be partially solved by using pyrolysis in the
presence of tetramethylammonium hydroxide (TMAH), that avoids
decarboxylation and releases intact (phenol- and acid-) methylated p-
hydroxycinnamates.13,14 Additional information regarding the different units and
interunit linkages present in the lignin polymer was provided by 2D NMR
spectroscopy, which provides information of the structure of the whole
macromolecule and is a powerful tool for lignin structural characterization,15− 9 and
DFRC (derivatization followed by reductive cleavage), which provides a measure
of the monomer composition of lignins and gives information on the nature and
x f γ-acylation of the lignin side chain. 6 30−35 The knowledge of the
composition and structure of the lignin of elephant grass will help to maximize the
exploitation of this interesting crop for biomass and biofuel production.

163
V. RESULTADOS Y DISCUSIÓN

MATERIALS AND METHODS

Samples

Elephant grass (P. purpureum), cultivar “Paraiso”, was collected at an age of

150 days old, from the experimental station of the University of Viçosa (Brazil).
Elephant grass stems were air-dried and subsequently separated into the cortex and
pith fractions. The dried samples were milled using a knife mill and successively
extracted with acetone in a Soxhlet apparatus for 8 h, and with hot water (3 h at 100
°C). The water-soluble material was lyophilized and then weighed to determine its
content, while the total acetone extractives once weighed were redissolved in
chloroform to separate the lipophilics from the polars, which were also weighed.
Klason lignin content was estimated as the residue after sulfuric acid hydrolysis of
the preextracted material according to Tappi test method T222 om-88.36 The
Klason lignin content was then corrected for proteins, determined from the N
content by the Kjeldahl method using a 6.25 factor,37 and ash (determined as
indicated below for the whole samples). The acid-soluble lignin was determined,
after the insoluble lignin was filtered off, by UV-spectroscopic determination at
205 nm wavelength using 110 L cm−1 g−1 as the extinction coefficient.
Holocellulose was isolated from the pre-extracted fibers by delignification for 4 h
using the acid chlorite method.38 The α-cellulose content was determined by
removing the hemicelluloses from the holocellulose by alkali extraction.38 Ash
content was estimated as the residue after 6 h of heating at 575 °C. Three replicates
were used for each sample.

“Milled-Wood Lignin” (MWL) Isolation

The MWLs were obtained according to the classical procedure.39 Extractive-

free ground cortex and pith samples (prepared as above) were finely ball-milled in
a Retsch PM100 planetary mill (40 h at 400 rpm for 25 g of wood) using a 500 mL
agate jar and agate ball bearings (20 × 20 mm), and toluene as coolant. The milled
samples were submitted to an extraction (3 × 12 h) with dioxane:water (9:1, v/v)
(20 mL solvent/g milled sample). The suspension was centrifuged and the
supernatant evaporated at 40 °C under reduced pressure. The residue obtained (raw
MWL, 1.765 g) was redissolved in acetic acid/ water 9:1 (v/v) (25 mL solvent/g
raw MWL). The solution was then precipitated into stirred cold water, and the
residue was separated by centrifugation, milled in an agate mortar and dissolved in
1,2-dichloroethane:ethanol (2:1, v/v). The mixture was then centrifuged to
eliminate the insoluble material. The resulting supernatant was precipitated into

164
 V. RESULTADOS Y DISCUSIÓN

cold diethyl ether, centrifuged, and subsequently resuspended in 30 mL of

petroleum ether and centrifuged again to obtain the purified MWL, which was
dried under a current of N2. The final yields ranged from 15 to 20% based on the
Klason lignin content.

Gel Permeation Chromatography (GPC)

GPC analyses of the isolated MWLs were performed on a Shimadzu LC-20A

LC system (Shimadzu, Kyoto, Japan) equipped with a photodiode array (PDA)
detector (SPD- 0 ; S g f w g : SK g α-M +
α-2500 (Tosoh, Tokyo, Japan) column; 0.1 M LiBr in dimethylformamide (DMF)
as eluent; 0.5 mL min−1 flow rate; 40 °C oven temperature; PDA detection at 280
nm. The data acquisition and computation used LCsolution version 1.25 software
(Shimadzu). The molecular weight calibration was via polystyrene standards.

Analytical Pyrolysis

Pyrolysis of the elephant grass stem cortex and pith fractions and their isolated
MWL samples (approximately 100 μg w f w 0 0 f
pyrolyzer (Frontier Laboratories Ltd.) connected to an Agilent 6890 GC/MS
system equipped with a DB-1701 fused-silica capillary column (30 m × 0.25 mm
0 5μ f k g 5973 lective detector (EI at 70
eV). The pyrolysis was performed at 500 °C. The GC oven temperature was
programmed from 50 °C (1 min) to 100 at 30 °C min−1 and then to 290 °C (10 min)
at 6 °C min−1. Helium was the carrier gas (1 mL min−1 y/ H 100 μg f
w x w x y05μ f H 5% w/w
and the pyrolysis was carried out as described above. The compounds were
identified by comparing their mass spectra with those of the Wiley and NIST
libraries and those reported in the literature.10,11 Peak molar areas were calculated
for the released pyrolysis products, the summed areas were normalized, and the
data for two repetitive analyses were averaged and expressed as percentages.

NMR Spectroscopy

For the NMR of the whole cell walls, around 100 mg of finely divided (ball-
milled) extractive-free samples was swollen in 0.75 mL of DMSO-d6 according to
the method previously described.21,24 In the case of the isolated MWLs, around 40
mg was dissolved in 0.75 mL of DMSO-d6. NMR spectra were recorded at 25 °C
on a Bruker AVANCE 600 MHz instrument equipped with a cryogenically cooled
z-gradient triple-resonance probe. HSQC (heteronuclear single quantum coherence)

165
V. RESULTADOS Y DISCUSIÓN

experiments used Bruker’s “hsqcetgp” pulse program with spectral widths of 5000
and 13200 Hz for the 1H and 13C dimensions. The number of collected complex
points was 2048 for the 1H dimension with a recycle delay of 1 s. The number of
transients was 64, and 256 time increments were recorded in the 13C dimension.
The 1J CH used was 140 Hz. Processing used typical matched Gaussian apodization
in 1H and a squared cosine-bell in 13C. Prior to Fourier transformation, the data
matrices were zerofilled up to 1024 points in the 13C dimension. The central solvent
peak was used as an internal reference (δ C 39.5; δ H 2.49). HSQC cross-signals
were assigned by comparison with the literature.15−29 A semiquantitative analysis of
the volume integrals of the HSQC cross-correlation signals was performed. As the
volume integral depends on the particular 1J CH value, as well on the T 2 relaxation
time, absolute quantitation is impossible but relative integrals (between spectra)
allow valid comparisons. Thus, the integration of the cross-signals was performed
separately for the different regions of the HSQC spectrum, which contain signals
that correspond to chemically analogous carbon−proton pairs. For these signals, the
1
J CH -coupling value is similar and integrals can be used semiquantitatively to
estimate the relative abundance of the different species. In the aliphatic oxygenated
region, the relative abundances of side chains involved in interunit linkages or
present in terminal units were estimated from the C α −H α correlations to avoid
possible interference from homonuclear 1H−1H couplings, except for substructures
E and I/I′, for which C β −H β and C γ −H γ correlations were used. In the aromatic
region, C 2 −H 2 correlations from H, G and S lignin units and from p-coumarate and
ferulate were used to estimate their relative abundances.

DFRC (Derivatization Followed by Reductive Cleavage)

The DFRC degradation was performed according to the developed protocol.30−33

Lignins (10 mg) were stirred for two hours at 50 °C with acetyl bromide in acetic
acid (8:92). The solvents and excess acetyl bromide were removed by rotary
evaporation at reduced pressure. The products were then dissolved in
dioxane/acetic acid/water (5:4:1, v/v/v), and 50 mg of powdered Zn was added.
After 40 min of stirring at room temperature, the mixture was transferred into a
separatory funnel with dichloromethane and saturated ammonium chloride. The pH
of the aqueous phase was adjusted to less than 3 by adding 3% HCl, the mixture
vigorously mixed and the organic layer separated. The water phase was extracted
twice more with dichloromethane. The combined dichloromethane fractions were
dried over anhydrous Na 2 SO 4 , and the filtrate was evaporated on a rotary
evaporator. The residue was acetylated for 1 h in 1.1 mL of dichloromethane
containing 0.2 mL of acetic anhydride and 0.2 mL of pyridine. The acetylated

166
 V. RESULTADOS Y DISCUSIÓN

lignin degradation products were collected after rotary evaporation of the solvents
and subsequently analyzed by GC/MS using mass spectra and relative retention
times to authenticate the DFRC monomers and their p-coumarate conjugates as
described.30−33 To assess the presence of naturally acetylated lignin units, the
described modification of the standard DFRC method using propionylating instead
f y g g ′ w y 26,34,35

The GC/MS analyses were performed with a GCMS-QP2010plus instrument

(Shimadzu Co.) using a capillary column (SHR5XLB 30 m × 0.25 mm i.d., 0.25
−1
μ f k v w f 140 ° 1 50 3 ° / ,
−1 −1
then ramped at 10 °C/min to 280 °C (1 min) and finally ramped at 20 °C/min to
300 °C, and held for 18 min at the final temperature. The injector was set at 250
°C, and the transfer line was kept at 310 °C. Helium was used as the carrier gas at a
rate of 1 mL/min−1. Quantitation of the released individual monomers was
f g 4 4′-ethylenebisphenol as internal standard. Molar yields were
calculated on the basis of molecular weights of the respective acetylated and/or
propionylated compounds.

RESULTS AND DISCUSSION

In this work, we separated the cortex (84%) and the pith (16%) fractions of the
stem and studied them independently. The abundance of the main constituents
(namely, water-soluble material, acetone extractives, Klason lignin, acid-soluble
g α-cellulose, and ash) of the cortex and pith are shown in
Table 1. The lignin content differs in each fraction of the elephant grass stem, with
the cortex having higher lignin content (18.5% Klason lignin) than the pith
(15.5%). The compositions of the lignin in both parts of the elephant grass were
y y y− / S y f
structural characterization the MWLs were isolated by aqueous dioxane extraction
from finely ball-milled samples according to the classical lignin isolation
procedure39 and su y y y y− / S
the molecular weights estimate by GPC.

167
V. RESULTADOS Y DISCUSIÓN

Table 1. Abundance of the Main Constituents (% dry weight) of Elephant Grass (P.
purpureum) Fractions.

Elephant grass cortex Elephant grass pith

% of whole material 84 16
Water solubles 8.8 ± 0.4 10.4 ± 0.1
Acetone extractives 1.7 ± 0.4 2.3 ± 0.2
Klason lignin 18.5 ± 0.6 15.5 ± 1.0
Acid soluble lignin 1.5 ± 0.1 1.6 ± 0.1
Holocellulose (α-Cellulose) 64.1 ± 1.7 (40.0 ± 1.4) 59.7 ± 0.5 (46.4 ± 3.4)
Ash 5.4 ± 0.2 10.5 ± 1.5

Molecular Weight Distributions

The values of the weight-average (M w ) and number-average (M n ) molecular

weights of the MWL isolated from the cortex and pith fraction of elephant grass
stems, estimated from the GPC curves (relative values related to polystyrene
standards), and the polydispersit (M w /M n ), are indicated in Table 2. The two
lignins exhibited similar molecular weight distributions, in the range 6920−6720 g
mol−1, being slightly higher in the case of the lignin from the cortex. In addition,
both lignins exhibited relatively narrow molecular weight distributions, with
M w /M n < 3. Those values are comparable to literature values for various isolated
lignins.28,29,40

Table 2. Weight-Average (M w ) and Number-Average (M n ) Molecular Weights (g mol−1),

and Polydispersity (M w /M n ) of the MWLs Isolated from the Cortex and Pith of Elephant
Grass (P. purpureum).

MWL cortex MWL pith

Mw 6920 6720

Mn 2390 2490

M w /M n 2.9 2.7

168
 V. RESULTADOS Y DISCUSIÓN

Py−GC/MS

The pyrograms of the cortex and pith of elephant grass, and of their
corresponding MWLs, are shown in Figure 1. The identities and relative molar
abundances of the released compounds are listed in Table 3. Pyrolysis of the whole
cell walls of cortex and pith (Figures 1a and 1b) released compounds from the
carbohydrate and lignin moieties, as well as from p-hydroxycinnamates.

30 27
100 27 100

(a) (c)
Relative abundance (%)

Relative abundance (%)

75 75

2 5
50 21 33 50

7 33

30
25 4 25
44
1 3 12 1618 44
29 21
2526 38 3738 51
37 45 47 51 25
26 29 45 47
16 52 54 56
2 4 6 8 10 12 14 16 18 20 22 2 4 6 8 10 12 14 16 18 20 22
Retention time (min) Retention time (min)

27
100 27 100

(b) (d)
Relative abundance (%)

30
Relative abundance (%)

75 75

5
50 2 50

7 30
33
21 51
4 33 21 37
25 25 38 44
13 1216 26
18 23 44
16 25 29 45 47
24 20 35 48
51 52 55 56
2 4 6 8 10 12 14 16 18 20 22 2 4 6 8 10 12 14 16 18 20 22
Retention time (min) Retention time (min)

Figure 1. Py-GC/MS chromatograms of elephant grass (P. purpureum) cortex (a) and pith
(b), and of the MWL isolated from cortex (c) and pith (d). The identities and relative
abundances of the released numbered compounds are listed in Table 3.

Among the carbohydrate-derived compounds, the main ones were 2-methylfuran

(1), hydroxyacetaldehyde (2), (3H)-furan-2-one (4), propanal (5), furfural (7),
(5H)-furan-2-one (12) and 2-hydroxy-3-methyl-2-cyclopenten-1-one (18). Among

169
V. RESULTADOS Y DISCUSIÓN

the lignin derived phenols, the pyrograms of the cortex and pith showed
compounds derived from p-hydroxyphenyl (H), guaiacyl (G) and syringyl (S)
lignin units as well as from the cinnamic acid esters in the wall. The most
prominent cinnamate or lignin-derived compounds released were 4-vinylphenol
(27) and 4-vinylguaiacol (30), with important amounts of other lignin-derived
compounds such as phenol (16), guaiacol (21), syringol (33) and 4-vinylsyringol
(44). However, the high amounts of 4-vinylphenol released upon pyrolysis from
these samples, as in other grasses, is mostly due to the presence of p-coumarates, as
will be shown below, which decarboxylates efficiently under pyrolytic
conditions.13,14 Similarly, 4-vinylguaiacol, which is present in high abundance
among the pyrolysis products of the whole cell walls, also arises from ferulate after
decarboxylation upon pyrolysis. Therefore, it is obvious that these vinyl
compounds cannot be used for the estimation of the lignin H:G:S composition upon
Py−GC/MS as the major part of them do not arise from the core lignin structural
units but from p-hydroxycinnamates that are either not associated with the lignin
structure (ferulates on arabinoxylans) or only partially (p-coumarates on
arabinoxylans, but also acylating lignin side chains). A rough estimation of the S/G
ratio can be obtained (by using the molar areas of all the G- and S-derived
compounds, except 4-vinylguaiacol, that also arises from ferulates, and its
respective 4-vinylsyringol), being 1.2 in both cortex and pith.

Table 3. Identities and Relative Molar Abundances of the Compounds Released after
Py−GC/MS of Elephant Grass (P. purpureum) Cortex and Pith and Their Isolated MWLs.

Label Origin a Cortex MWL Pith MWL

cortex pith
1 2-methylfuran PS 2.3 0.0 1.2 0.0
2 hydroxyacetaldehyde PS 9.1 0.0 7.9 0.0
3 3-hydroxypropanal PS 0.7 0.0 1.2 0.0
4 (3H)-furan-2-one PS 0.8 0.0 1.2 0.0
5 propanal PS 6.7 0.0 7.8 0.0
6 (2H)-furan-3-one PS 0.8 0.0 0.6 0.0
7 furfural PS 4.9 0.0 4.9 0.0
8 4-methyltetrahydrofuran-3-one PS 0.9 0.0 0.5 0.0
9 1-acetoxypyran-3-one PS 0.4 0.0 0.3 0.0
10 2-hydroxymethylfuran PS 0.4 0.0 0.3 0.0
11 cyclopent-1-ene-3,4-dione PS 1.2 0.0 1.2 0.0
12 (5H)-furan-2-one PS 2.4 0.0 1.9 0.0
13 2,3-dihydro-5-methylfuran-2-one PS 1.8 0.0 2.6 0.0
14 2-acetylfuran PS 0.6 0.0 0.5 0.0
15 2-methyl-2-cyclopenten-1-one PS 1.0 0.0 0.5 0.0

170
 V. RESULTADOS Y DISCUSIÓN

16 phenol LH 3.1 2.3 4.1 3.9

17 4-hydroxy-5,6-dihydro-(2H)-pyran-2-one PS 0.2 0.0 0.3 0.0
18 2-hydroxy-3-methyl-2-cyclopenten-1-one PS 2.3 0.0 2.1 0.0
19 4-hydroxybenzaldehyde PS 0.4 0.0 0.3 0.0
20 methylphenol LH 0.4 1.3 0.3 3.1
21 guaiacol LG 4.4 4.4 3.0 3.0
22 methylphenol LH 1.3 0.0 1.3 0.0
23 anhydrosugar PS 0.8 0.0 1.2 0.0
24 2-hydroxy-3-ethyl-2-cyclopenten-1-one PS 0.5 0.0 0.6 0.0
25 C2-phenol LH 1.0 1.0 0.5 2.4
26 4-methylguaiacol LG 0.6 1.4 0.4 2.7
27 4-vinylphenol LH/PCA 27.2 54.8 33.9 45.3
28 4-allylphenol LH 0.0 0.1 0.0 0.1
29 4-ethylguaiacol LG 0.5 0.5 0.3 1.0
30 4-vinylguaiacol LG/Fer 7.3 5.0 7.3 6.6
31 cis-4-propenylphenol LH 0.0 0.1 0.0 0.1
32 eugenol LG 0.2 0.5 0.2 0.7
33 syringol LS 6.8 9.9 4.6 6.4
34 4-trans-propenylphenol LH 0.0 0.6 0.0 1.1
35 cis-isoeugenol LG 0.2 0.5 0.2 0.7
36 vanillin LG 0.7 1.0 0.7 1.1
37 4-methylsyringol LS 0.6 2.1 0.3 3.5
38 trans-isoeugenol LG 0.7 1.4 0.4 2.6
39 propineguaiacol LG 0.3 0.9 0.3 0.7
40 propineguaiacol LG 0.3 0.8 0.4 0.5
41 acetoguaiacone LG 0.1 0.2 0.1 0.3
42 4-ethylsyringol LS 0.4 0.5 0.2 0.6
43 guaiacylacetone LG 0.2 0.3 0.2 0.3
44 4-vinylsyringol LS 2.5 3.7 2.4 3.1
45 4-allylsyringol LS 0.4 0.6 0.2 1.2
46 4-propylsyringol LS 0.1 0.1 0.1 0.1
47 cis-4-propenylsyringol LS 0.4 0.6 0.3 1.0
48 syringaldehyde LS 0.3 0.8 0.3 1.5
49 propinesyringol LS 0.2 0.3 0.2 0.3
50 propinesyringol LS 0.2 0.2 0.2 0.3
51 trans-propenylsyringol LS 1.2 1.6 0.7 3.5
52 acetosyringone LS 0.1 0.5 0.0 0.8
53 trans-coniferaldehyde LG 0.0 0.3 0.0 0.4
54 syringylacetone LS 0.1 1.0 0.2 0.7
55 propiosyringone LS 0.0 0.1 0.0 0.3
56 trans-sinapaldehyde LS 0.0 0.6 0.0 0.2
S/Gb 1.2 1.5 1.2 1.4
ratio

a
PS: polysaccharide. LH: lignin p-hydroxyphenyl-type. LG: lignin guaiacyl-type. LS: lignin
syringyl-type. PCA: p-coumarate. FA: ferulate.
b
All G- and S-derived peaks were used for the estimation of the S/G ratio, except 4-vinylguaiacol
(arising from ferulates), and the analogous 4-vinylsyringol.

171
V. RESULTADOS Y DISCUSIÓN

Pyrolysis of the MWLs isolated from the cortex and pith of elephant grass stems
(Figures 1c and 1d) released a similar distribution of cinnamate- and lignin-
derived compounds as from their respective whole cell walls, except for the much
lower relative abundance of 4-vinylguaiacol (30). The most prominent compound
in the pyrograms of the MWLs was still 4-vinylphenol (27), derived largely from
the p-coumarate esters acylating lignin side chains (see later), and as also occurred
in the pyrolysis of their whole cell walls. The estimation of the S/G ratios (by
ignoring the respective vinyl compounds) was 1.5 and 1.4 in the MWL of the
cortex and pith, suggesting that syringyl-rich oligomers were slightly preferentially
extracted into the MWL fraction. The similarity with the lignin S/G ratios observed
in the whole cell walls reveals the importance of excluding the vinyl compounds
for calculation, at least in grasses, where p-hydroxycinnamates are important
components.

It is clear then that p-hydroxycinnamic acids, which form linkages with lignin
and/or carbohydrates in plants, and are particularly abundant in grasses,13,41−48
interferes in the estimation of the lignin composition which cannot, therefore, be
evaluated properly by conventional pyrolysis. The presence of p-
hydroxycinnamates in the whole cell walls, as well as in the isolated lignins,
however, could be addressed by pyrolysis in the presence of a methylating agent,
tetramethylammonium hydroxide (TMAH), that efficiently prevents
decarboxylation and results in transesterification (producing methyl esters, as well
as methylating the phenol),13,14 as shown in Figure 2. The identities of the
compounds released and their relative molar abundances are listed in Table 4.
Py/TMAH induces cleavage of alkyl aryl ether bonds in lignin and releases
products similar to those obtained upon CuO alkaline degradation, including
methylated hydroxybenzaldehydes (peaks 6, 12 and 18), hydroxyacetophenones
(peaks 15 and 22) and hydroxybenzoic acids (peaks 10, 16 and 24).13,14,49,50 As seen
in Figure 2, Py/TMAH of the cortex and pith released high amounts (over 45% of
the total peak area) of the fully methylated derivative of p-coumaric acid, i.e.,
trans-3-(4-methoxyphenyl)propenoic acid methyl ester, or methyl trans-4-O-
methyl-p-coumarate (peak 23), as well as lower amounts (nearly 5% of total peak
area) of the fully methylated derivative of ferulic acid, i.e., trans-3-(3,4-
dimethoxyphenyl)propenoic acid methyl ester, or methyl 4-O-methyl-ferulate (peak
29). In addition to the trans-forms of methylated p-hydroxycinnamic acids, minor
amounts of the cis-isomers (peaks 17 and 26) were also identified. These TMAH
transesterification and methylation products clearly establish that the high amounts
of 4-vinylphenol and 4-vinylguaiacol released upon Py−GC/MS of the cell wall
and lignin samples arise mainly from p-coumarate and ferulate esters in the wall,

172
 V. RESULTADOS Y DISCUSIÓN

and not from the core lignin itself, highlighting the fact that these two vinyl
products cannot be used in the estimate of lignin composition from pyrolysis.
Unfortunately, these components have been used in H:G:S determinations in the
past, but this practice must cease: the H:G:S ratio of lignins refers (or should refer!)
strictly to the composition of the core lignin that arises from polymerization of the
H, G, and S monolignols, namely, p-coumaryl, coniferyl, and sinapyl alcohols.

23 23
100 100

(a) (c)
4
Relative abundance (%)

Relative abundance (%)

75 75

50 50

29 22
22 4
25 9 25
16 18
2 3 6 12 +
16 20
+ 17 20 12 + +
8 13 14 18 + 24 +
1 11 21 5 9 11 13 14 17 21 24 25
28
1 2 3 67
8 27 28

4 6 8 10 12 14 16 18 20 22 4 6 8 10 12 14 16 18 20 22
Retention time (min) Retention time (min)

23 23
100
(b) 100
(d)
Relative abundance (%)

Relative abundance (%)

75 75

50 50
4

22 22
25 25 16
10 29 +
3 9 16 17
8 13 + 4 18 29
1
2 5 67 17 20
+
10 12 20
+
11 12 14 18 21 24 + 21 2425
25 1 2 3 5 6 7 8 9 11 13 14 15 19 27 28
4 6 8 10 12 14 16 18 20 22 4 6 8 10 12 14 16 18 20 22
Retention time (min) Retention time (min)

Figure 2. Py−TMAH−GC/MS chromatograms of elephant grass (P. purpureum) cortex (a)

and pith (b), and the MWLs isolated from cortex (c) and pith (d). The identities and relative
abundances of the numbered released compounds are listed in Table 4. [Note that the com-
pound numbers are not the same as those in Figure 1 and Table 3.]

173
V. RESULTADOS Y DISCUSIÓN

The relative abundances of p-hydroxycinnamates (p-coumarate/ferulate ratio)

present in the cortex and pith and in their isolated lignins were estimated by
Py/TMAH (Table 4) and revealed additional features.

Table 4. Identity and Relative Molar Abundances of the Compounds Released after
Py/TMAH of Elephant Grass (P. purpureum) Cortex and Pith and Their Isolated Lignins
(MWLs).
Cortex MWL Pith MWL
cortex pith
1 methoxybenzene 1.7 0.6 2.2 0.6
2 4-methoxytoluene 2.7 1.8 2.3 0.9
3 1,2-dimethoxybenzene 2.9 0.8 4.2 0.4
4 4-methoxystyrene 16.2 10.4 13.8 5.7
5 3,4-dimethoxytoluene 0.7 1.9 1.9 0.7
6 4-methoxybenzaldehyde 3.3 1.3 3.1 1.8
7 trans-4-methoxypropenyl 1.4 1.5 2.2 0.9
8 1,2,3-trimethoxybenzene 1.7 1.2 2.6 0.6
9 3,4-dimethoxystyrene 4.5 2.0 3.7 1.1
10 4-methoxybenzoic acid methyl ester 0.9 0.4 0.3 0.5
11 3,4,5-trimethoxytoluene 0.8 2.0 0.8 1.1
12 3,4-dimethoxybenzaldehyde 1.7 1.9 1.5 1.8
13 1-(3,4-dimethoxyphenyl)-1-propene 0.8 1.2 1.5 0.5
14 3,4,5-trimethoxystyrene 1.0 1.4 1.0 1.0
15 3,4-dimethoxyacetophenone 1.0 0.4 0.4 0.4
16 3,4-dimethoxybenzoic acid methyl ester 0.8 1.1 1.4 1.1
17 cis 3-(4-methoxyphenyl)-3-propenoic acid methyl ester 1.1 1.8 1.5 4.7
18 3,4,5-trimethoxybenzaldehyde 1.3 5.1 1.4 4.1
19 cis 1-(3,4dimethoxyphenyl)-2-methoxyethylene 0.2 0.5 0.4 0.5
20 trans 1-(3,4dimethoxyphenyl)-2-methoxyethylene 0.2 0.6 0.4 0.6
21 1-(3.4.5-trimethoxyphenyl)-1-propene 1.1 1.9 1.2 1.2
22 3,4,5-trimethoxyacetophenone 0.9 1.3 1.1 1.5
23 trans 3-(4-methoxyphenyl)-3-propenoic acid methyl ester 45.6 53.8 43.1 63.7
24 3,4,5-trimethoxybenzoic acid methyl ester 1.0 1.4 1.1 1.8
25 cis 1-(3,4,5-trimethoxyphenyl)-2-methoxyethylene 0.3 1.1 0.7 0.7
26 cis 3-(3,4-dimethoxyphenyl)-3-propenoic acid methyl ester 0.5 0.1 0.4 0.1
27 trans 1-(3,4,5-trimethoxyphenyl)-2-methoxyethylene 0.2 0.7 0.4 0.5
28 trans-1-(3,4-dimethoxy)-2,3-dimethoxyprop-1-ene 0.6 1.0 0.3 0.5
29 trans 3-(3,4-dimethoxyphenyl)-3-propenoic acid methyl ester 5.0 0.8 5.1 1.1
p-Coumaric acid/Ferulic acid ratioa 9.1 67.3 8.5 57.9
a
Relative abundance of p-coumarates (peaks 17 and 23) with respect to ferulates (peaks 26 and 29).

174
 V. RESULTADOS Y DISCUSIÓN

Both p-coumarate and ferulate were found in the whole cell walls of the cortex
and pith, while only p-coumarate and essentially no ferulate (only trace amounts,
∼1%) was found in the isolated lignins. This indicates, as has been known for a
long time in grasses,41−48 that, in the cortex and pith of elephant grass, ferulate is
mostly attached to the carbohydrates while p-coumarate is primarily attached to the
lignin polymer. Studies on different plants, including other grasses, have indicated
that p- y γ-OH of lignin side chains, and predominantly on S
26,33,41,43,47,48,51,52
units. Therefore, the major part of the p-coumarate present in the
lignin of the cortex and pith of elephant grass was hypothesized to also acylate the
γ-OH of the lignin side chain, as will be validated below.

2D NMR

The whole cell walls of the elephant grass cortex and pith fractions were
analyzed in situ by gel-state 2D NMR, according to the method previously
described,21,24 and the spectra were compared with those from the lignins (MWLs)
isolated from the same samples. It is convenient to look at three characteristic
regions of the HSQC spectra corresponding to nonoxygenated aliphatic,
oxygenated aliphatic side chain, and aromatic 13 −1H correlations. The
nonoxygenated aliphatic region, not plotted here, showed signals with little
relevance to the structure of the cell wall polymers, except for a strong cross-signal
δC/δH 20.6/2.00, assigned to methyls in acetate groups attached to xylan moieties,
w w k g g δC/δH 0 6/1 7−1 9 g
53
methyls in acetate groups attached to the lignin polymer.

xyg δC/δH 45−90/ 4−5 6

δC/δH 95−150/5 5−8 0 g ons of the HSQC spectra of the whole cell walls from
cortex and pith, and their isolated MWLs, are shown in Figures 3 and 4. It has
already been shown that HSQC-NMR of DMSO-d6 gels of ball-milled plant
ff f “ ” lysis of the major structural
features of native lignin in plants, without the need of prior isolation.21,24 Xylan
polysaccharide signals (X2, X3, X4, X5) were predominant in the hydroxylated
aliphatic region of the spectra of the whole cell walls, which partially overlapped
with some lignin signals, and also included signals from acetylated xylan moieties
X′2 X′3). On the other hand, the spectra of the MWLs presented mostly lignin
signals that, in general terms, matched those observed in the HSQC spectra of their
respective whole cell walls. The main lignin and carbohydrate cross-signals
assigned in the HSQC spectra are listed in Table 5, and the main lignin
substructures found are depicted in Figure 5.

175
V. RESULTADOS Y DISCUSIÓN

(a) (b)
C’β
50 50
Bβ Cβ
-OMe -OMe

Iγ Aγ 60 Iγ Aγ Dβ 60
A’γ A’γ
I’γ I’γ
X5

Aα/A’α Aα/A’α
70 70
X2 X2
Cγ
X’2 X3 X3
X’3 X4 X4

80 Dα 80
A’β(S)/Aβ(G) Eβ A’β(S)/Aβ(G)
C’α C’α
Cα Aβ(S) Aβ(S)
Cα
Bα Bα
5.5 5.0 4.5 4.0 3.5 3.0 2.5 5.5 5.0 4.5 4.0 3.5 3.0 2.5

S2,6 100 S2,6 100

S’2,6 S’2,6

G2 G2
FA2 110 110
PCA3,5 PCAβ/FAβ PCA3,5 PCAβ/FAβ

G5/G6
{ 120
G5/G6
{ H3,5
120

H2,6
PCA2,6 PCA2,6
130 130

140 140
PCAα/FAα PCAα/FAα

7.5 7.0 6.5 6.0 7.5 7.0 6.5 6.0

Figure 3. Side-chain (δC/δH 45−90/2.40−5.60) and aromatic (δC/δH 95−150/5.50−8.00) regions

in the 2D HSQC NMR spectra of elephant grass (P. purpureum) cortex (a) and its isolated
MWL (b). See Table 5 for signal assignments and Figure 5 for the main lignin structures
identified.

176
 V. RESULTADOS Y DISCUSIÓN

(a) (b) C’β

50 50
Bβ Cβ
-OMe -OMe

Iγ Aγ Iγ Aγ
60 A’γ 60
A’γ
I’γ I’γ
X5 X5

Aα/A’α Aα/A’α
70 70
X’2 X2 X’2 X2
Cγ
X3 X3
X’3 X4 X’3
X4

80 A’β(G) 80
A’β(S)/Aβ(G) Eβ A’β(S)/Aβ(G)
C’α
Aβ(S) Cα Aβ(S)
Bα

5.5 5.0 4.5 4.0 3.5 3.0 2.5 5.5 5.0 4.5 4.0 3.5 3.0 2.5

100 100
S2,6 S2,6
S’2,6 S’2,6

G2 G2
FA2 110 FA2 110
PCA3,5 PCAβ/FAβ PCA3,5 PCAβ/FAβ

G5/G6
{ H3,5
120
G5/G6
{ H3,5
120
FA6
H2,6 H2,6
PCA2,6 PCA2,6
130 130

140 140
PCAα/FAα PCAα/FAα

7.5 7.0 6.5 6.0 7.5 7.0 6.5 6.0

Figure 4. Side-chain (δC/δH 45−90/2.40−5.60) and aromatic (δC/δH 95−150/5.50−8.00)

regions in the HSQC NMR spectra of elephant grass (P. purpureum) pith (a) and its isolated
MWL (b). See Table 5 for signal assignments and Figure 5 for the main lignin structures
identified.

177
V. RESULTADOS Y DISCUSIÓN

Table 5. Assignments of 13C−1H Correlation Signals in the 2D HSQC Spectra of the

Whole Cell Walls of the Cortex and Pith of Elephant Grass (P. purpureum) and Their
Isolated MWLsa.

Label δ C /δ H (ppm) Assignment

Lignin Cross-Peak Signals
C' β 49.8/2.59 C β -H β in γ-acylated β-β' tetrahydrofuran structures
(C')
Bβ 53.5/3.46 C β -H β in phenylcoumaran substructures (B)
Cβ 53.5/3.06 C β -H β in β-β' resinol substructures (C)
-OCH 3 55.6/3.73 C-H in methoxyls
Aγ 59.4 /3.40 and 3.72 C γ H γ in γ-hydroxylated β-O-4' substructures (A)
Dβ 59.6/2.75 C β -H β in spirodienone substructures (D)
Iγ 61.3/4.09 C γ H γ in cinnamyl alcohol end-groups (I)
A' γ 62.7/3.83-4.30 C γ H γ in γ-acylated β-O-4' substructures (A')
I' γ 64.0/4.79 C γ H γ in γ-acylated cinnamyl alcohol end-groups (I')
Cγ 71.0/3.83 and 4.19 C γ H γ in β-β' resinol substructures (C)
A α /A' α 71.7/4.86 C α -H α in β-O-4' substructures (A, A')
A´ β(G) 80.8/4.58 C β -H β in γ-acylated β-O-4' substructures linked to a G
unit (A')
Dα 81.2/5.09 C α -H α in spirodienone substructures (D)
C' α 82.8/5.00 C α -H α in γ-acylated β-β' tetrahydrofuran structures
(C')
Eβ 82.8/5.23 C β -H β in α-oxidized β-O-4' substructures (E)
A β(G) /A' β(S) 83.5/4.28 and C β -H β in β-O-4' substructures linked to a G unit (A)
83.0/4.32 and in γ-acylated β-O-4' substructures linked to a S
unit (A')
Cα 84.8/4.67 C α -H α in β-β' resinol substructures (C)
A β(S) 85.8/4.11 C β -H β in β-O-4' substructures linked to a S unit (A)
Bα 86.8/5.46 C α -H α in phenylcoumaran substructures (B)
S 2,6 103.8/6.69 C 2 -H 2 and C 6 -H 6 in etherified syringyl units (S)
S' 2,6 106.1/7.32 and C 2 -H 2 and C 6 -H 6 in α-oxidized syringyl units (S')
106.4 /7.19
G2 110.9/6.99 C 2 -H 2 in guaiacyl units (G)
FA 2 111.0/7.32 C 2 -H 2 in ferulic acid units (FA)
PCA β and FA β 113.5/6.27 C β -H β in p-coumaric (PCA) and ferulic acids (FA)
G 5 /G 6 114.9/6.72 and 6.94, C 5 -H 5 and C 6 -H 6 in guaiacyl units (G)
118.7/6.77
PCA 3,5 115.5/6.77 C 3 -H 3 and C 5 -H 5 in p-coumarate (PCA)
FA 6 123.3/7.10 C 6 -H 6 in ferulate (FA)
H 2,6 128.0/7.23 C 2,6 -H 2,6 in p-hydroxyphenyl units (H)
PCA 2,6 130.0/7.46 C 2 -H 2 and C 6 -H 6 in p-coumarate (PCA)
PCA α and FA α 144.4/7.41 C α -H α in p-coumarate (PCA) and ferulate (FA)

Polysaccharide Cross-Peak Signals

X5 63.2/3.26 and 3.95 C 5 -H 5 in β-D-xylopyranoside
X2 72.9/3.14 C 2 -H 2 in β-D-xylopyranoside
X′ 2 73.5/4.61 C 2 -H 2 in 2-O-acetyl-β-D-xylopyranoside
X3 74.1/3.32 C 3 -H 3 in β-D-xylopyranoside
X′ 3 74.9/4.91 C 3 -H 3 in 3-O-acetyl-β-D-xylopyranoside
X4 75.6/3.63 C 4 -H 4 in β-D-xylopyranoside

178
 V. RESULTADOS Y DISCUSIÓN
HO

O
HO
γ O γ 5′
4′ γ β
HO 4′ OMe
α β O HO HO
β
α O α O
OMe
OMe

OMe
OMe
O O OMe
O
A A’ B

O OMe
OMe OMe
O OR O
γ′
β′
O α′ γ α
γ α′ O γ′
β HO
β′ RO O β α′ OH
β α γ β′
γ′ 1′
α O
OMe
O O OMe O
OMe
C C’ D

OH HO γ O
HO γ
γ O γ O

4′ β β β
O α α
β O α
α
OMe

OMe OMe
OMe
O O
O OH

E I/I’ PCA FA

OH OH OH R O
α α α α

OMe MeO OMe MeO OMe

O O O O

H G S S′
Figure 5. Main structures present in the lignins of elephant grass (P. purpureum): (A) β−O−4′
structures; (A′) β−O−4′ structures with acylated (by acetates or p-coumarates) γ-OH; (B)
phenylcoumaran structures formed by β−5′ coupling; (C) resinol structures formed by β−β′
coupling; (C′) tetrahydrofuran structures formed by β−β coupling of monolignols acylated at
the γ-carbon; (D) spirodienone structures formed by β−1′ coupling; (E) Cα-oxidized β−O−4′
structures; (I) p-hydroxy-cinnamyl alcohol end-groups; (I′) p-hydroxycinnamyl alcohol end-
groups acylated at the γ-OH; (PCA) p-coumarate units; (FA) ferulate units; (H)
p-hydroxyphenyl units; (G) guaiacyl units; (S) syringyl units; (S′) oxidized syringyl units
bearing a carbonyl (R, lignin side-chain) or carboxyl (R, hydroxyl group) group at Cα.

179
V. RESULTADOS Y DISCUSIÓN

The side-chain region of the spectra gave useful information about the different
interunit linkages present in the lignin. In this region, cross-signals from methoxyls
(δC/δH 55.6/3.73) and side chains in β−O−4′ substructures A were the most
prominent. The spectra of the whole cell walls and of their corresponding MWLs
clearly showed the presence of intense signals in the range δC/δH 62.7/3.83−4.19
corresponding to the γ-C/H of γ-acylated units (including structure A′), together
with the presence of signals from normally hydroxylated γ-carbons in β−O−4′ units
A and other substructures (at δC/δH 60.2/3.30 and 3.70). The HSQC spectra
therefore indicate that these lignins are extensively acylated at the γ-position of the
lignin side chain. An estimation of the percentage of γ-acylation of the lignin side
chain was performed by integration of the signals corresponding to the
hydroxylated vs acylated γ-C/H correlations in the HSQC spectra of the isolated
MWLs, where the signals are better resolved and carbohydrates do not interfere,
and ranged from 39% in the cortex to 55% in the pith lignin (Table 6). The spectra
showed other prominent signals corresponding to β−O−4′ alkyl-aryl ether linkages
A. The Cα−Hα correlations in β−O−4′ substructures were observed at δC/δH
71.7/4.86 (structures A and A′), while the Cβ−Hβ correlations were observed at
δC/δH 85.8/4.11 in normal γ-OH β−O−4′ substructures A linked to a S unit but
shifted to δC/δH 83.0/4.32 in γ-acylated β−O−4′ substructures A′, which overlaps
with the Cβ−Hβ correlations of normal γ-OH β−O−4′ substructures A linked to a G
unit at δC/δH 83.5/4.28. The Cβ−Hβ correlations of γ-acylated β−O−4′ substructures
A′ linked to a G unit shifted to δC/δH 80.8/4.52, and were clearly observed in the
MWL from the pith, indicating an important acylation extent of G-lignin units in
this lignin, as will be shown below. Other substructures were also observed in
lower amounts. Phenylcoumaran (β−5′) substructures B were found, the signals for
their Cα−Hα and Cβ−Hβ correlations being observed at δC/δH 86.8/5.46 and
53.5/3.46, and the Cγ−Hγ correlation overlap with other Cγ−Hγ signals around δC/δH
62/3.8. Small signals for resinol (β−β′) substructures C were also observed in the
spectra, with their Cα−Hα, Cβ−Hβ and the double Cγ−Hγ correlations at δC/δH
84.8/4.67, 53.5/3.06 and 71.0/3.83 and 4.19. Finally, small signals corresponding to
spirodienone (β−1′) substructures (D) could also be observed in the spectrum (at
contour levels lower than those plotted), their Cα−Hα and Cβ−Hβ correlations being
at δC/δH 81.2/5.09 and 59.6/2.75. Other small signals in the side-chain region of the
HSQC spectra corresponded to Cγ−Hγ correlations (at δC/δH 61.3/4.09 and
64.0/4.79) assigned to cinnamyl alcohol end-groups (I) and γ-acylated cinnamyl
alcohol end-groups (I′), and the Cβ−Hβ correlations of α-keto-β−O−4′ substructures
(E).

180
 V. RESULTADOS Y DISCUSIÓN

Table 6. Structural Characteristics (Lignin Interunit Linkages, End-Groups, Percentage of

γ-Acylation, Relative Molar Composition of the Lignin Aromatic Units, S/G Ratio and p-
Coumarate/and Ferulate Content and Ratio) from Integration of 13 −1H Correlation Signals
in the HSQC Spectra of Whole Cell Walls of Cortex and Pith of Elephant Grass (P.
purpureum) and Their Isolated MWLs.

Cortex MWL Pith MWL

cortex Pith
Lignin interunit linkages
β-O-4´ substructures (A/A') - 82 - 82
β-5´ phenylcoumarane substructures (B) - 8 - 7
β-β´ C) - 2 - 1
β-β´ y f C´) - 3 - 5
β-1´ Spirodienones (D) - 2 - 2
α- x β-O-4´ structures (E) - 2 - 3
a
LLignin end groups
Cinnamyl alcohol end-groups (I) - 6 - 7
γ-acylated cinnamyl alcohol end-groups (I') - 3 - 4
g f γ-acylation - 39 - 55
Lignin aromatic unitsb
Percentage of aromatic units
H (%) 0 3 3 3
G (%) 44 40 42 39
S (%) 56 57 55 58
S/G ratio 1.3 1.4 1.3 1.5
p-Hydroxycinnamtesc
p-coumarates (%) 26 29 39 40
ferulates (%) 11 3 16 4
p-coumarates/ferulates ratio 2.4 9.7 2.4 10.0
a
Expressed as a fraction of the total lignin inter-unit linkage types A−E.
b
Molar percentages (H + G + S = 100).
c
p-Coumarate and ferulate levels expressed as a fraction of lignin content (H + G +S).
Because p-coumarate and ferulate groups are terminal, freely rotating, and therefore
have longer relaxation times than internal lignin units, they overquantitate (relative to
the lignin) by HSQC integration.

I gy g f β−β′-linked tetrahydrofuran structure C′ were clearly

seen in the spectra of both MWLs, with the characteristic Cα−Hα and Cβ−Hβ
correlations at 82.8/5.00 and 49.8/2.59, as already described in kenaf, corn, and
palms54,55 and also reported in the lignins of sisal and abaca.22 These structures
f β−β′ -coupling of two γ-acylated sinapyl alcohol monomers
and have been found in the lignins of several plants and with different acylating
groups (acetates, p-coumarates and p-hydroxybenzoates).54,55 Therefore, the

181
V. RESULTADOS Y DISCUSIÓN

presence of this structure C′ is related to the high extent of γ-carbon acylation in

the lignins of elephant grass, as shown above.

The main cross-signals in the aromatic regions of the HSQC spectra

corresponded to the different lignin and p-hydroxycinnamate units. Signals from p-
hydroxycinnamyl (H), guaiacyl (G) and syringyl (S) units were observed in the
spectra of the whole cell walls and in their isolated MWLs. The S-lignin units
showed a prominent signal for the C2,6−H2,6 correlation at δC/δH 103.8/6.69, while
the G-lignin units showed different correlations for C2−H2 (δC/δH 110.9/6.99), and
for C5−H5 and C6−H6 (δC/δH 114.9/6.72 and 6.94, and 118.7/6.77). Signals
corresponding to C2,6−H2,6 correlations in Cα-oxidized S-lignin units (S′) were
observed at δC/δH 106.1/ 7.32 and 106.4/7.19. Signals for C2,6−H2,6 of H lignin units
at δC/δH 128.0/7.23 were also detected in the HSQC spectra, although in lower
amounts. This confirms that the H-unit content in the lignins from these elephant
grass samples, as in other grasses, is quite low (∼3%, Table 6), and that the high
abundance of “H units” observed upon pyrolysis was due to the presence of p-
coumarate, as shown by Py-TMAH (although some overestimation of ester-
forming structures is produced by this technique). It is a commonly accepted
fallacy that grass lignins are high in H units: as discussed above, they are not!
Prominent signals corresponding to p-coumarate structures (PCA) were observed
in the spectra of the whole cell walls and of their isolated MWLs. Cross-signals
corresponding to the C2,6−H2,6 at δC/δH 130.0/7.46 and C3,5−H3,5 at δC/δH 115.5/6.77
correlations of the aromatic ring and signals for the correlations of the unsaturated
Cα−Hα at δC/δH 144.4/7.41 and Cβ−Hβ at 113.5/6.27 of the p-coumarate unit were
observed in this region of the HSQC spectra. Signals corresponding to the C2−H2
and C6−H6 correlations of ferulate moieties (FA) were also observed at δC/δH
111.0/7.32 and 123.3/7.10 in the whole cell wall spectra. The correlations
corresponding to the unsaturated Cα−Hα and Cβ−Hβ overlapped with those of the p-
coumarate.

The molar composition of different lignin units (H, G, S) and p-

hydroxycinnamate (p-coumarate, ferulate) in the cortex and in the pith, as well as
in their isolated MWLs, as estimated from the HSQC data, is reflected in Table 6.
The relative abundances of the main lignin interunit linkages were estimated only
in the HSQC spectra of the MWLs because the high abundances of carbohydrate
signals in the spectra of the whole cell walls interfere with quantitation of some
lignin signals. The data indicated a similar lignin composition in the cortex and
pith, with a low abundance of H-lignin units and with similar S/G ratio around 1.3.
On the other hand, p-coumarate is highly abundant in the whole cell walls of the

182
 V. RESULTADOS Y DISCUSIÓN

cortex and pith, as well as in their isolated MWLs, while the abundance of ferulate
is much lower in the isolated MWL than in the respective whole cell walls, as
v y v y y− / S f f y
entirely attached to the carbohydrates (primarily, as has been established in other
grasses,56 acylating the C5−OH f y
((glucurono)arabinoxylans)), while p-coumarate is predominantly attached to the
g y f g w g x f y f γ-
OH observed in these lignins, seems to indicate that p-coumarate is mostly
y g γ-position of the lignin side chain, as also observed in other
26,33,41,43,47,48,51,52
lignins, although no direct evidence of the nature of the group
y g γ-carbon can be provided by HSQC. Esterification of p-coumarate to
α-carbon can be excluded from the absence of the corresponding cross-signal in
the HSQC spectra, which is at ∼ 6.1/75 ppm.51,57

I W g β−O−4′ ky y
ether, which accounts for up to 82% of all interunit linkages in the cortex and in the
pi w k g β−β′ β−5′ y β−1′
spirodienones) are present in minor amounts. In particular, there is a strikingly low
f β−β′ w f y % f
linkages in the cortex and only 1% in the pith lignin. This low proportion of resinol
g x f γ-acylation of the lignin side chain, as also
observed in other highly acylated lignins.22,26 If γ-OH of a monolignol is
acylated, the formation of the normal resinol structures cannot occur because a free
γ-hydroxyl is needed to rearomatize the intermediate quinone methide (following
the radical dehydrodimerization step). Instead, new tetrahydrofuran structures are
f f β−β′ o-coupling of two acylated monolignols or cross-
coupling of a monolignol with an acylated monolignol.22,26,54,55,58 Interestingly,
tetrahydrofuran structures C′ f y β−β′ - g f w γ-acylated
monolignols are also present in these lignins, being especially abundant in the pith
5% f k g g w g g f γ-
y 55% f f β−β′ g
y f 5−6% f erunit linkages) in both
lignins from cortex and pith, and this value is similar to that noted in other lignins
without monolignol acylation, but with similar S/G levels.20

DFRC (and DFRC′)

The HSQC data shown above indicate that the lignins in the cortex and pith of
g v y y γ-position of the side chain, but cannot

183
V. RESULTADOS Y DISCUSIÓN

provide additional information on the nature of the acylating group. The DFRC
degradation method, which cleaves α- and β-ether linkages in the lignin polymer
leaving γ-esters intact,30−33 seems to be the most appropriate method for the
analysis of γ-acylated lignins.

The chromatograms of the DFRC degradation products of the MWLs isolated

from the cortex and pith of elephant grass are shown in Figure 6.

100 IS
(a)
Relative abundances (%)

tS
tG
75

50
tSpc

25
cGpc tGpccSpc
tH cG cS

15 20 25 30 35 40 45
Retention time (min)

IS
100 tG
(b)
Relative abundances (%)

tS
75

50

tSpc
25
cG cGpc tGpc cSpc
tH cS

15 20 25 30 35 40 45
Retention time (min)

Figure 6. GC−TIC (TIC: total ion current) chromatograms of the DFRC degradation
products from the MWLs isolated from elephant grass (P.purpureum) cortex (a) and pith
(b), showing the presence of syringyl (and minor guaiacyl) units acylated by p-coumarate
moieties. cG, tG, cS and tS are the normal cis- and trans-coniferyl and sinapyl alcohol
(guaiacyl and syringyl) monomers (as their acetate derivatives). cGpc, tGpc, cSpc and tSpc are
the cis- and trans-coniferyl and sinapyl p-coumarates (as their acetate derivatives). IS:
internal standard (4,4′-ethylenebisphenol). Nonlabeled peaks eluting at 40−42 min are
from carbohydrate impurities.

184
 V. RESULTADOS Y DISCUSIÓN

The lignins released the cis- and trans-isomers of p-hydroxyphenyl (tH),

guaiacyl (cG and tG), and syringyl (cS and tS) lignin monomers (as their acetylated
v v gf γ-OH) units in lignin. The presence of important
k g γ-p-coumaroylated syringyl (cSpc and tSpc) and guaiacyl
(cGpc and tGpc) lignin units in the DFRC chromatograms confirmed that p-
g γ-carbon of these lignins, and predominantly
on syringyl units; minor amounts of the guaiacyl p-coumarate (cGpc and tGpc)
DFRC monomer conjugates could also be detected in both MWLs.

As noted above, signals for aceta g γ-attached to lignin moiety were also
observed in the HSQC spectra, including in the MWLs and indicating that acetates
g y γ-OH of these lignins, as widely occurs in many other
12,26,27,34,35,59
lignins. The original DFRC degradation method, however, does not
allow the analysis of natively acetylated lignin because the degradation products
are acetylated during the analytical procedure, but with appropriate modification of
the protocol by substituting acetylating reagents with propionylating reagents (in
the so- ′ f
26,34,35
occurrence of native lignin acetylation. Figure 7 shows the chromatograms
f ′ g f W f om the
cortex and pith of elephant grass. The lignins released the cis- and trans-isomers of
guaiacyl (cG and tG) and syringyl (cS and tS) lignin monomers (as their
y v v gf γ-OH) units in lignin. In addition,
the pre f γ-acetylated guaiacyl (cGac and tGac) and syringyl (cSac and tSac)
lignin units could also be observed in the chromatograms, indicating that native
y γ-OH of the lignin side chain also occurred in these lignins,
although to a low extent. Low levels of lignin acetylation, with a preference for G
units, were also found in other grasses, such as bamboo.35

f ′ y f W f
cortex and pith of elephant grass, namely, the molar yields of the released
monomers (H, G, Gac, Gpc, S, Sac, Spc), as well as the percentages of naturally
acetylated guaiacyl (% Gac) and syringyl (% Sac) and p-coumaroylated guaiacyl (%
Gpc) and syringyl (% Spc) lignin units, are presented in Table 7. The data indicate
g x f γ-acylation occurs in the lignins of both the cortex and pith of
the elephant grass, and that p-coumarate is the main group acylating these lignins,
with lower amounts of acetates, in agreement with the NMR data. While p-
coumarate groups are preferentially attached to syringyl units in both fractions,
acetates are attached preferentially to S units in the cortex and to G units in the pith

185
V. RESULTADOS Y DISCUSIÓN

lignin. We find this intriguing as the high levels of γ-acetylation in various dicots
such as kenaf or jute are heavy on syringyl units.35,58,59

tG
100
(a)
Relative abundances (%)

75

50 tS

25 tGac
tH cG tSac
cS

20 25 30
Retention time (min)

100
tG
(b)
Relative abundances (%)

75

50 tS

tGac
25 tH
cG
cS
tSac

20 25 30
Retention time (min)

Figure 7. Chromatograms of the DFRC′ degradation products from MWLs isolated from
elephant grass (P. purpureum) cortex (a) and pith (b). cG, tG, cS and tS are the normal cis-
and trans-coniferyl and sinapyl alcohol (guaiacyl and syringyl) monomers (as their
propionylated derivatives). cGac, tGac, cSac and tSac are the originally (natively) γ-acetylated
cis- and trans-coniferyl and sinapyl alcohol (guaiacyl and syringyl) monomers (as their
propionylated derivatives).

The monolignol conjugates sinapyl acetate and sinapyl p-coumarate have been
demonstrated to behave as monomers in lignification, participating normally in
coupling and cross-coupling reactions.26,35,54,55,58,60 If the γ-OH of a monolignol is
acylated, however, the formation of the normal resinol structures cannot occur
because a free γ-hydroxyl is needed to rearomatize the quinone methide moiety.
Instead, new tetrahydrofuran structures are formed from the β−β′ homo-coupling

186
 V. RESULTADOS Y DISCUSIÓN

and cross-coupling reactions involving acylated monolignols.22,26,35,54,55,58 Since p-

coumarates are by far the main group acylating the sinapyl alcohol monomer, as
seen above, it is clear that the tetrahydrofuran structure C′, identified in the HSQC
spectra, is formed from the β−β′ coupling of two sinapyl p-coumarate monomers,
and will bear two p-coumarate groups in its structure, as in Figure 8.

Table 7. Abundance (μmol/g of Lignin) of the Monomers Obtained from DFRC and
DFRC′ Degradation of the MWLs Isolated from the Cortex and Pith of Elephant Grass (P.
purpureum) and Relative Percentages of the Different Acylated (Acetylated and p-
Coumaroylated) Lignin Monomers.

H G Gac Gpc S Sac Spc %Gaca %Gpcb %Sacc %Spcd

MWL
4.6 156.6 4.1 3.4 191.2 17.7 86.7 2.5 2.0 6.0 29.3
cortex

MWL
2.8 74.9 8.0 3.5 79.2 2.1 26.0 9.3 4.1 2.0 24.2
pith

a
% Gac is the percentage of acetylated G units (Gac) with respect to the total G units (G, Gac, Gpc).
b
% Gpc is the percentage of p-coumaroylated G units (Gpc) with respect to the total G units (G, Gac, Gpc).
c
% Sac is the percentage of acetylated S units (Sac) with respect to the total S units (S, Sac, Spc).
d
% Spc is the percentage of p-coumaroylated S units (Spc) with respect to the total S units (S, Sac, Spc).

HO

HO O
O O
OMe
O HO OH
O γ γ´
γ α´ O
O O
γ O Peroxidase MeO
β β´
2 α β H2O2
β β´
γ´ OMe H2O
MeO OMe
O
α α O α´
O HO OH
MeO
OMe OMe
O OH
MeO OMe
OH

Figure 8. Pathway for the β−β′ homocoupling of two sinapyl p-coumarate monolignol
conjugates producing the tetrahydrofuran structure C′ observed in the HSQC spectra with
a β−β′ linkage and p-coumarate groups acylating both γ-OHs.

When p-coumaroylated sinapyl alcohol dimerizes, it forms the β−β-coupled bis-

quinone methide intermediate. However, this intermediate cannot be rearomatized
by internal trapping; rearomatization will occur after water attack on one quinone
methide moiety with the resulting α-OH attacking the other quinone methide to
form the tetrahydrofuran structure C′. The p-coumaroyl monolignol transferase

187
V. RESULTADOS Y DISCUSIÓN

involved in the p-coumaroylation of sinapyl alcohol has already been described in

grasses,47 and a candidate gene has now been identified.61 The presence of these
tetrahydrofuran substructures in the lignin polymer here is indicative of the
occurrence of pre-p-coumaroylated monolignols that participate in coupling and
cross-coupling reactions in the lignification of elephant grass, and therefore
implicate the presence of analogous transferases in this plant.

In conclusion, the analyses of the lignins from the cortex and pith of elephant
grass indicate that they have a typical G-S lignin, with low amounts (∼3%) of H
units, and a S/G ratio of 1.3−1.5, depending on the analytical method used. The
analyses also indicate the presence of high amounts of p-coumarate groups on
lignin which acylate the γ-OH of the lignin side chains, and preferentially on
syringyl units. Minor amounts of acetate groups were also found acylating the
lignin. The main interunit linkage present in these lignins is the β−O−4′ alkyl aryl
ether (82% of all interunit linkages), with lower amounts of condensed linkages:
resinols and tetrahydrofurans (β−β′), phenylcoumarans (β−5′), and spirodienones
(β−1′). The presence of a tetrahydrofuran structure formed from the β−β′ homo-
coupling of two γ-acylated monolignols, presumably two γ-p-coumaroylated
sinapyl alcohols, was observed in significant amounts, being especially abundant in
the pith (5% of all interunit linkages), corresponding with its higher degree of γ-
acylation. The presence of these tetrahydrofuran substructures in the lignin polymer
is indicative of the occurrence of p-coumaroylated monolignol conjugates that
participate in coupling and cross-coupling reactions during elephant grass
lignification.

FUNDING
This study has been funded by the Spanish project AGL2011-25379, the CSIC
project 201040E075 and the EU-project LIGNODECO (KBBE-244362). John
Ralph was funded in part by the DOE Great Lakes Bioenergy Research Center
(DOE Office of Science BER DE-FC02-07ER64494). Jorge Rencoret thanks the
CSIC for a JAE-DOC contract of the program “Junta para la Ampliación de
Estudios” cofinanced by Fondo Social Europeo (FSE), and Pepijn Prinsen thanks
the Spanish MICINN for a FPI fellowship.

ACKNOWLEDGMENTS
We thank Prof. Jorge L. Colodette and Prof. Jose L. Gomide (Univ. of Viçosa,
Brazil) for providing the elephant grass. We also thank Dr. Yuki Tobimatsu (Univ.
Wisconsin, Madison) for performing the GPC analyses, Dr. Fachuang Lu for help

188
 V. RESULTADOS Y DISCUSIÓN

v − S f he products, and Hoon Kim for

providing the method and most of the assignments in his gel-state NMR method.

REFERENCES
(1) Himmel, M. E. Biomass Recalcitrance. Deconstructing the Plant Cell Wall
for Bioenergy; Blackwell: Oxford, U.K., 2008.

(2) Zhang, Y. H. P. Reviving the carbohydrate economy via multiproduct

lignocellulose biorefineries. J. Ind. Microbiol. Biotechnol. 2008, 35 367−375

(3) Schank, S. C.; Chynoweth, D. P.; Turick, C. E.; Mendoza, P. E. Napiergrass

genotypes and plant parts for biomass energy. Biomass Bioenergy 1993, 4 1−7

(4) Woodard, K. R.; Prine, G. M. Dry matter accumulation of elephant grass,

energy cane and elephant millet in a subtropical climate. Crop Sci. 1993, 33,
818−8 4

(5) Somerville, C.; Youngs, H.; Taylor, C.; Davis, S. C.; Long, S. P. Feedstocks
for lignocellulosic biofuels. Science 2010, 329 790−79

(6) Higuchi, T. Biochemistry and Molecular Biology of Wood; Springer Verlag:

London, U.K., 1997.

(7) Boerjan, W.; Ralph, J.; Baucher, M. Lignin biosynthesis. Annu. Rev. Plant
Biol. 2003, 54 519−546

(8) Ralph, J.; Lundquist, K.; Brunow, G.; Lu, F.; Kim, H.; Schatz, P. F.; Marita,
J. M.; Hatfield, R. D.; Ralph, S. A.; Christensen, J. H.; Boerjan,W. Lignins: Natural
polymers from oxidative coupling of 4-hydroxyphenylpropanoids. Phytochem. Rev.
2004, 3 9−60

(9) Xie, X. -M.; Zhang, X. -Q.; Dong, Z. -X.; Guo, H. -R. Dynamic changes of
lignin contents of MT-1 elephant grass and its closely related cultivars. Biomass
Bioenergy 2011, 35 173 −1738

(10) Faix, O.; Meier, D.; Fortmann, I. Thermal degradation products of wood. A
collection of electron-impact (EI) mass spectra of monomeric lignin derived
products. Holz Roh- Werkst. 1990, 48 351−354

(11) Ralph, J.; Hatfield, R. D. Pyrolysis-GC/MS characterization of forage

materials. J. Agric. Food Chem. 1991, 39 14 6−1437

189
V. RESULTADOS Y DISCUSIÓN

(12) del Río, J. C.; Gutiérrez, A.; Martínez, A. T. Identifying acetylated lignin
units in non-wood fibers using pyrolysis-gas chromatography/mass spectrometry.
Rapid Commun. Mass Spectrom. 2004, 18, 1181−1185.

(13) del Río, J. C.; Gutiérrez, A.; Rodríguez, I. M.; Ibarra, D.; Martínez, A. T.
Composition of non-woody plant lignins and cinnamic acids by Py-GC/MS,
Py/TMAH and FT-IR. J. Anal. Appl. Pyrolysis 2007, 79, 39−46.

(14) del Río, J. C.; Martín, F.; González-Vila, F. J. Thermally assisted

hydrolysis and alkylation as a novel pyrolytic approach for the structural
characterization of natural biopolymers and geomacromolecules. Trends Anal.
Chem. 1996, 15, 70−79.

(15) Liitiä, T. M.; Maunu, S. L.; Hortling, B.; Toikka, M.; Kilpeläinen, I.
Analysis of technical lignins by two- and three-dimensional NMR spectroscopy. J.
Agric. Food Chem. 2003, 51, 2136−2143.

(16) Capanema, E. A.; Balakshin, M. Y.; Kadla, J. F. A comprehensive

approach for quantitative lignin characterization by NMR spectroscopy. J. Agric.
Food Chem. 2004, 52, 1850−1860.

(17) Capanema, E. A.; Balakshin, M. Y.; Kadla, J. F. Quantitative

characterization of a hardwood milled wood lignin by nuclear magnetic resonance
spectroscopy. J. Agric. Food Chem. 2005, 53, 9639−9649.

(18) Ralph, J.; Marita, J. M.; Ralph, S. A.; Hatfield, R. D.; Lu, F.; Ede, R. M.;
Peng, J.; Quideau, S.; Helm, R. F.; Grabber, J. H.; Kim, H.; Jimenez-Monteon, G.;
Zhang, Y.; Jung, H. -J. G.; Landucci, L. L.; MacKay, J. J.; Sederoff, R. R.;
Chapple, C.; Boudet, A. M. Solutionstate NMR of lignin. In Advances in
lignocellulosics characterization, Argyropoulos, D. S., Ed.; Tappi Press: Atlanta,
1999; pp 55−108.

(19) Ralph, S. A.; Ralph, J.; Landucci, L. NMR database of lignin and cell wall
model compounds; US Forest Prod. Lab.: One Gifford Pinchot Dr., Madison, WI
53705 (http://ars.usda.gov/Services/docs.htm?docid=10491), (accessed: January
2009), 2004.

(20) Ralph, J.; Landucci, L. L. NMR of lignins. In Lignin and Lignans;

Advances in Chemistry; Heitner, C., Dimmel, D. R., Schmidt, J. A., Eds.; CRC
Press (Taylor & Francis Group): Boca Raton, FL, 2010; pp 137−234.

190
 V. RESULTADOS Y DISCUSIÓN

(21) Kim, H.; Ralph, J.; Akiyama, T. Solution-state 2D-NMR of ball-milled

plant cell-wall gels in DMSO-d6. Bioenergy Res. 2008, 1 56−66.

; ; ; ; I ;
-Barbero, J.; del Río, J. C. Monolignol acylation and lignin structure in
some nonwoody plants: A 2D-NMR study. Phytochemistry 2008, 69 831− 843

3 ; ; ;I ; ; ;
S I; -Barbero, J.; Martínez, A. T.; del Río, J. C. Structural
characterization of milled wood lignin from different eucalypt species.
Holzforschung 2008, 62 514−5 6

4 ; ; ; ;S I; -
Barbero, J.; Martínez, A. T.; del Río, J. C. HSQC-NMR analysis of lignin in woody
(Eucalyptus globulus and Picea abies) and non-woody (Agave sisalana) ball-milled
plant materials at the gel state. Holzforschung 2009, 63 691−698

5 ; ; ; ; - Barbero, J.;
Martínez, A. T.; del Río, J. C. Isolation and structural characterization of the milled
wood lignin from Paulownia fortune wood. Ind. Crops Prod. 2009, 30 137−143

6 ; ; ; ;I ;S
I; -Barbero, J.; Zhang, L.; Martínez, A. T. Highly acylated (acetylated
and/or p-coumaroylated) native lignins from diverse herbaceous plants. J. Agric.
Food Chem. 2008, 56 95 5−9534

(27) del Río, J. C.; Rencoret, ; ; ; ; -

; ; S f g
from jute (Corchorus capsularis) fibers. J. Agric. Food Chem. 2009, 57,
10 71−10 81

(28) del Río, J. C.; Rencoret, J ; ; ; - Barbero, J.;

Martínez, A. T. Structural characterization of guaiacyl-rich lignins in flax (Linum
usitatissimum) fibers and shives. J. Agric. Food Chem. 2011, 59 11088−11099

9 ; ; ; -Barbero, J.; Faulds, C. B.;

Kim, H.; Ralph, J.; Martínez, A. T.; del Río, J. C. Lignin composition and structure
in young versus adult Eucalyptus globulus plants. Plant Physiol. 2011, 155,
667−68

191
V. RESULTADOS Y DISCUSIÓN

(30) Lu, F.; Ralph, J. Derivatization followed by reductive cleavage (DFRC

method), a new method for lignin analysis: protocol for analysis of DFRC
monomers. J. Agric. Food Chem. 1997, 45, 2590−2592.

(31) Lu, F.; Ralph, J. The DFRC method for lignin analysis. Part 1. A new
method for β-aryl ether cleavage: lignin model studies. J. Agric. Food Chem. 1997,
45, 4655−4660.

(32) Lu, F.; Ralph, J. The DFRC method for lignin analysis. 2. Monomers from
isolated lignin. J. Agric. Food Chem. 1998, 46, 547−552.

(33) Lu, F.; Ralph, J. Detection and determination of p-coumaroylated units in

lignins. J. Agric. Food Chem. 1999, 47, 1988−1992.

(34) Ralph, J.; Lu, F. The DFRC method for lignin analysis. 6. A simple
modification for identifying natural acetates in lignin. J. Agric. Food Chem. 1998,
46, 4616−4619.

(35) del Río, J. C.; Marques, G.; Rencoret, J.; Martínez, A. T.; Gutiérrez, A.
Occurrence of naturally acetylated lignin units. J. Agric. Food Chem. 2007, 55,
5461−5468.

(36) Tappi Test Methods 2004−2005, Tappi Press, Norcoss, GA 30092, USA,
2004.

(37) Darwill, A.; McNeil, M.; Albersheim, P.; Delmer, D. The primary cell-
walls of flowering plants. In The Biochemistry of Plants; Tolbert, N., Ed.;
Academic Press: New York, 1980; pp 91−162.

(38) Browning, B. L. Methods of Wood Chemistry; Wiley-Interscience

Publishers: New York, 1967; Vol. II.

(39) Björkman, A. Studies on finely divided wood. Part I. Extraction of lignin

with neutral solvents. Sven. Papperstidn. 1956, 59, 477−485.

(40) Baumberger, S.; Fasching, M.; Gellerstedt, G.; Gosselink, R.; Hortling, B.;
Li, J.; Saake, B.; de Jong, E. Molar mass determination of lignins by size-exclusion
chromatography: towards standardisation of the method. Holzforschung 2007, 61,
459−468.

192
 V. RESULTADOS Y DISCUSIÓN

(41) Grabber, J. H.; Quideau, S.; Ralph, J. p-Coumaroylated syringyl units in maize
g : I f β-ether cleavage by thioacidolysis. Phytochemistry 1996,
43 1189−1194

(42) Grabber, J. H.; Ralph, J.; Hatfield, R. D. Cross-linking of maize walls by

ferulate dimerization and incorporation into lignin. J. Agric. Food Chem. 2000, 48,
6106−6113

(43) Grabber, J. H.; Lu, F. Formation of syringyl-rich lignins in maize as

influenced by feruloylated xylans and p-coumaroylated monolignols. Planta 2007,
226 741−751

(44) Lam, T. B. T.; Iiyama, K.; Stone, B. A. Cinnamic acid bridges between cell
wall polymers in wheat and phalaris intemodes. Phytochemistry 1992, 31,
1179−1183

(45) Sun, R. -C.; Sun, X. -F.; Zhang, S. -H. Quantitative determination of

hydroxycinnamic acids in wheat, rice, rye, and barley straws, maize stems, oil palm
frond fiber, and fast-growing poplar wood. J. Agric. Food Chem. 2001, 49,
51 −51 9

(46) Sun, R. -C.; Sun, X. F.; Wang, S. Q.; Zhu, W.; Wang, X. Y. Ester and ether
linkages between hydroxycinnamic acids and lignins from wheat, rice, rye, and
barley straws, maize stems, and fast-growing poplar wood. Ind. Crop. Prod. 2002,
15 179−188

(47) Hatfield, R. D.; Marita, J. M.; Frost, K.; Grabber, J.; Ralph, J.; Lu, F.; Kim, H.
Grass lignin acylation: p-coumaroyl transferase activity and cell wall
characteristics of C3 and C4 grasses. Planta 2009, 229 1 53−1 67

(48) Ralph, J. Hydroxycinnamates in lignification. Phytochem. Rev. 2010, 9,

65−83

49 ; ; -Vila, F. J.; Verdejo, T. Thermally assisted

hydrolysis and alkylation of lignins in the presence of tetra-alkylammonium
hydroxides. J. Anal. Appl. Pyrolysis 1995, 35 1−13

(50) del Río, J. C.; McKinney, D. E.; Knicker, H.; Nanny, M. A.; Minard, R. D.;
Hatcher, P. G. Structural characterization of bio- and geo-macromolecules by off-
line thermochemolysis with tetramethylammonium hydroxide. J. Chromatogr., A
1998, 823 433−448

193
V. RESULTADOS Y DISCUSIÓN

(51) Ralph, J.; Hatfield, R. D.; Quideau, S.; Helm, R. F.; Grabber, J. H.; Jung, H. -
J. G. Pathway of p-coumaric acid incorporation into maize lignin as revealed by
NMR. J. Am. Chem. Soc. 1994, 116, 9448−9456.

(52) Crestini, C.; Argyropoulos, D. S. Structural analysis of wheat straw lignin by

quantitative 31P and 2D-NMR spectroscopy. The occurrence of ester bonds and α-
O-4 substructures. J. Agric. Food Chem. 1997, 45, 1212−1219.

(53) Rencoret, J.; del Río, J. C.; Gutiérrez, A.; Martínez, A. T.; Li, S.; Parkås, J.;
Lundquist, K. Origin of the acetylated structures present in white birch (Betula
pendula Roth) milled wood lignin. Wood Sci. Technol. 2012, 46, 459−471.

(54) Lu, F.; Ralph, J. Novel β−β structures in lignins incorporating acylated
monolignols. Appita 2005, 233−237.

(55) Lu, F.; Ralph, J. Novel tetrahydrofuran structures derived from β−β-coupling
reactions involving sinapyl acetates in kenaf lignins. Org. Biomol. Chem. 2008, 6,
3681−3694.

(56) Fry, S. C.; Willis, S.; Paterson, A. Intraprotoplasmic and wall-localised

formation of arabinoxylan-bound diferulates and larger ferulate coupling-products
in maize cell-suspension cultures. Planta 2000, 211, 679−692.

(57) Helm, R. F.; Ralph, J. Lignin-hydroxycinnamoyl model compounds related to

forage cell wall structure. 2. Ester-linked structures. J. Agric. Food Chem. 1993,
41, 570−576.

(58) Lu, F.; Ralph, J. Preliminary evidence for sinapyl acetate as a lignin monomer
in kenaf. Chem. Commun. 2002, 90−91.

(59) Ralph, J. An unusual lignin from kenaf. J. Nat. Prod. 1996, 59, 341−342.

(60) Ralph, J. What makes a good monolignol substitute? In The Science and Lore
of the Plant Cell Wall Biosynthesis, Structure and Function; Hayashi, T., Ed.;
Universal Publishers (BrownWalker Press): Boca Raton, FL, 2006; pp 285−293.

(61) Withers, S.; Lu, F.; Kim, H.; Zhu, Y.; Ralph, J.; Wilkerson, C. G.
Identification of a grass-specific enzyme that acylates monolignols with p-
coumarate. J. Biol. Chem. 2012, 287, 8347−8355.

194
 V. RESULTADOS Y DISCUSIÓN

Publicación IV:

del Río J.C., Prinsen P. and Gutiérrez A. (2013). A comprehensive characterization

of lipids in wheat straw. Journal of Agricultural and Food Chemistry, 61, 1904-
1913.

195
 V. RESULTADOS Y DISCUSIÓN

A comprehensive characterization of lipids in wheat straw

José C. del Río, Pepijn Prinsen and Ana Gutiérrez

Instituto de Recursos Naturales y Agrobiología de Sevilla, CSIC, P.O. Box 1052,

E- 41080 Seville, Spain

ABSTRACT

The chemical composition of the lipids in wheat straw was studied in detail by gas
chromatography and mass spectrometry. Important discrepancies with the data
reported in previous papers were found. The predominant lipids identified were
series of long-chain free fatty acids (25% of total extract), followed by series of
free fatty alcohols (ca. 20%). High molecular weight esters of long chain fatty acids
esterified to long chain fatty alcohols were also found (11%), together with lower
amounts of other aliphatic series such as n-alkanes, n-aldehydes and glycerides
(mono-, di- and triglycerides). Relatively high amounts of β-diketones (10%),
particularly 14,16-hentriacontanedione, which is the second most abundant single
compound among the lipids in wheat straw, were also identified. Finally, steroid
compounds (steroid hydrocarbons, steroid ketones, free sterols, sterol esters and
sterol glycosides) were also found, with sterols accounting for nearly 14% of all
identified compounds.

KEYWORDS: Wheat straw, lipids, fatty acids, fatty alcohols, sterols, β-diketones.

INTRODUCTION

There is a growing need to consider alternative agricultural strategies that move an

agricultural industry focused on food production to one that also supplies the needs
of other industrial sectors, such as paper, textiles, biofuels or added-value
chemicals, in the context of the so-called lignocellulose biorefinery. The term
“ f y” f g w
renewable raw materials to produce energy together with a wide range of everyday
commodities in an economic and sustainable manner.1–3 Plant biomass is the main

197
V. RESULTADOS Y DISCUSIÓN

source of renewable materials in Earth and represents a potential source of

renewable energy and biobased products. Biomass is available in high amounts at
very low cost (as forest, agricultural or industrial lignocellulosic wastes and
cultures) and could be a widely available and inexpensive source for biofuels and
bioproducts in the near future.

The high abundance, wide availability and very low-cost of some agricultural
wastes, as cereal straws, makes them excellent raw materials for future
biorefineries. Among them, wheat straw has the greatest potential of all agricultural
residues because of its wide availability and low cost.4,5 Wheat straw is an
abundant by-product from wheat production in many countries. The average yield
of wheat straw is 1.3–1.4 kg/kg of wheat grain, with a world production of wheat
estimated to be around 680 million tons in 2011. Wheat straw contains 35–45%
cellulose, 20–30% hemicelluloses, and around 15% lignin, which makes it an
attractive feedstock to be converted to ethanol and other value-added products.6

Wheat straw also contains significant amounts of lipids (ca. 1-2% by weight)
that can be extracted to produce high-value waxes.7 Natural waxes have a wide
range of industrial applications in cosmetics, personal care products, polishes and
coatings. On the other hand, these lipids, even when present in lower amounts in
the raw material, may play an important role during the industrial processing, as in
pulp and paper production, since they are at the origin of the so-called pitch
deposits.8 Lipids include different classes of compounds (i.e. alkanes, fatty
alcohols, fatty acids, free and conjugated sterols, terpenoids and triglycerides),
which have different behavior during pulping and bleaching.8–12 Pitch deposition is
a serious problem in the pulp and paper industry being responsible for reduced
production levels, higher equipment maintenance costs, higher operating costs, and
an increased incidence of defects in the finished products, which reduces quality
and benefits.8

Studies concerning the composition of lipids in wheat straw have been relatively
scarce, although some papers have been published in this regard.7, 13–15 However,
most of these studies are somewhat limited and controversial. Thus, some papers
have reported the occurrence of resin acids (i.e. abietic acid), which are exclusively
restricted to conifers, among the lipophilic extractives in wheat straw.13–15
Moreover, high amounts of ergosterol, a sterol that only occurs in fungi and that is
absent in plants, were also reported in those studies.13–15 Therefore, the presence of
these compounds clearly indicates cross-contamination from other lignocellulosic
materials, as well as fungal degradation of the studied wheat straw sample, which

198
 V. RESULTADOS Y DISCUSIÓN

certainly impairs to obtain accurate information about the authentic composition of

the lipids present in wheat straw. In the present work, a thorough and
comprehensive characterization of the lipophilic extractives in wheat straw has
been performed, and important discrepancies with the data reported in previous
papers have been found. In this paper, the composition of the lipophilic compounds
was carried out by gas chromatography (GC) and gas chromatography mass
spectrometry (GC-MS) using short- and medium-length high temperature capillary
columns, respectively, with thin films, which enables the elution and analysis of a
wide range of compounds from fatty acids to intact high molecular weight lipids
such as sterol esters, sterol glycosides or triglycerides.16 The knowledge of the
precise composition of the lipophilic extractives in wheat straw will help to
maximize the exploitation of this important agricultural waste.

MATERIALS AND METHODS

Samples

Wheat straw (Triticum durum var. Carioca) was harvested from an experimental
field in Seville (South Spain) in June 2009. Wheat straw was air-dried and the dried
samples were milled using a knife mill (Janke and Kunkel, Analysenmühle), and
subsequently extracted with acetone in a Soxhlet apparatus for 8 h. The acetone
extracts were evaporated to dryness, and resuspended in chloroform for
chromatographic analysis of the lipophilic fraction. Two replicates were used for
each sample.

GC and GC-MS analyses

An HP 5890 gas chromatograph (Hewlett Packard, Hoofddorp, Netherlands)

equipped with a split-splitless injector and a flame ionization detector (FID) was
used for GC analyses. The injector and the detector temperatures were set at 300 ºC
and 350 ºC respectively. Samples were injected in the splitless mode. Helium was
used as the carrier gas. The capillary column used was a high temperature,
polyimide coated fused silica tubing DB5-HT (5 m x 0.25 mm I.D., 0.1 µm film
thickness; J&W Scientific). The oven was temperature-programmed from 100 ºC (1
min) to 350 ºC (3 min) at 15 ºC min-1. Peaks were quantified by area, and a mixture
of standards (octadecane, palmitic acid, sitosterol, cholesteryl oleate, and sitosteryl
3β-D-glucopyranoside) with a concentration range between 0.1 and 1 mg/mL, was
used to elaborate calibration curves. The correlation coefficient was higher than
0.99 in all the cases. The data from the two replicates were averaged. In all cases,
the standard deviations from replicates were below 10% of the mean values. The

199
V. RESULTADOS Y DISCUSIÓN

GC-MS analysis were performed on a Varian Star 3400 gas chromatograph

(Varian, Walnut Creek, CA) coupled with an ion-trap detector (Varian Saturn)
equipped with a high-temperature capillary column (DB-5HT, 15 m × 0.25 mm
01 μ f k ; &W S f H w g
rate of 2 mL/min. The samples were injected with an autoinjector (Varian 8200)
directly onto the column using a SPI (septum-equipped programmable injector)
system. The temperature of the injector during the injection was 60 ºC, and 0.1 min
after injection was programmed to 380 ºC at a rate of 200 ºC min-1 and held for 10
min. The oven was heated from 120 ºC (1 min) to 380 ºC (5 min) at 10 ºC min -1.
The temperature of the transfer line was set at 300 ºC.
Bis(trimethylsilyl)trifluoroacetamide (BSTFA) silylation were used when required.
Compounds were identified by comparing their mass spectra with mass spectra in
the Wiley and NIST libraries, by mass fragmentography and, when possible, by
comparison with authentic standards.

RESULTS AND DISCUSSION

The abundance of the main constituents of wheat straw (water solubles, acetone
extractives, Klason g g α-cellulose and ash) is
shown in Table 1.

Table 1. Abundance of the Main Constituents (% dry-weight) of Wheat Straw.

Water solubles 9.6

Total Acetone extractives 2.7

Lipophilics 2.0
Polars 0.7

Klason lignina 16.2

Acid soluble lignin 1.5

Holocellulose 67.2
Cellulose 36.5
Hemicelluloses 30.7

Ash 6.6
a
Corrected for proteins and ash.

200
 V. RESULTADOS Y DISCUSIÓN

The total acetone extractives of wheat straw accounts for 2.7% of dry material.
However, the lipohilic content, estimated as the chloroform solubles is lower and
accounts for 2% while the rest (0.7%) correspond to polar compounds extracted in
acetone. This content is similar to that reported for wheat straw in previous
papers,13–15 and also similar to that found in other nonwoody materials such as
flax,9,10 hemp,17 kenaf,18 sisal,19 abaca,20 jute,21 giant reed,22 or Miscanthus.23

The underivatized and TMS-ether derivatives of the lipophilic extracts from

wheat straw were analyzed by GC and GC-MS using short- and medium-length
high temperature capillary columns, respectively, with thin films, according to the
method previously described.16 The GC-MS chromatograms of the underivatized
and TMS-ether derivatives of the lipid extracts from wheat straw are shown in
Figure 1. The identities and abundances of the main lipid compounds identified are
detailed in Table 2.

The most predominant lipids present in wheat straw were series of fatty acids
that accounted for 25% of all identified compounds, followed by series of free fatty
alcohols (ca. 20%). High molecular weight esters of long-chain fatty acids
esterified to long-chain fatty alcohols were also found in significant amounts
(11%). Additionally, lower amounts of other aliphatic series such as n-alkanes, n-
aldehydes and glycerides (mono-, di- and triglycerides), were also observed.
Important amounts of β-diketones (10% of all identified compounds) were also
found in the extracts of wheat straw. Steroid compounds (hydrocarbons, ketones,
free sterols, sterol esters and sterol glycosides) were also present among the
lipophilic extracts of wheat straw in important amounts, with sterols accounting for
nearly 14% of all identified compounds. The structures of the main lipophilic
compounds present in wheat straw are depicted in Figure 2 and Figure 3. The
distributions of the main aliphatic series are represented in the histograms of
Figure 4. It is important to note that significant differences were observed with the
composition reported in previous papers.13–15 Previous papers also indicated a
predominance of free fatty acids in wheat straw. However, they failed to report the
occurrence of fatty alcohols, which are the second most abundant class of aliphatic
compounds in wheat straw, as well as the presence of series of alkanes and
aldehydes. In addition, they did not report the presence of the important amounts of
β-diketones that were observed in our work. Finally, previously published papers
reported the presence of important amounts of free and esterified sterols, 13–15 but
failed to identify other important steroids such as sterol glycosides, steroid ketones
and steroid hydrocarbons.

201
V. RESULTADOS Y DISCUSIÓN

3
4
Ac28

1
2

Ak31 E44

SE
Ak29
E42 E46
Ad28 E40
E48 E Trigl
50
F18:1
F16 +
F18:2 Ak27 E52
F18 Ad26

5 10 15 20 25 30

4 7
Ac28

2
1
56

Ak29 SE
Ak31 E46
F18:1 Ac26
+ F26 E40 E42 E48 E
F18:2 Ak27 F
50
Trigl
F16 F22 Ac 24
24
F18 F20

5 10 15 20 25 30
Retention time (min)

Figure 1. GC-MS chromatograms of the lipid extracts from wheat straw (a) underivatized,
(b) as TMS-ether derivatives. Fn: n-fatty acid series; Akn: n-alkane series; Acn: n-fatty
alcohol series; Adn: n-aldehyde series; En: high molecular weight ester series; n denotes the
total carbon atom number. SE: sterol esters; Trigl: triglycerides. Other compounds
reflected are: 1: campesterol; 2: stigmasterol; 3: sitosterol; 4: 14,16-hentriacontanedione; 5:
campesteryl 3β-D-glucopyranoside; 6: stigmasteryl 3β-D-glucopyranoside; 7: sitosteryl 3β-
D-glucopyranoside.

202
 V. RESULTADOS Y DISCUSIÓN

Table 2.Composition and Abundance (mg/Kg fiber, d.a.f.) of Main Lipids Identified in
the Extracts of Wheat Straw.

Compound Abundance
n-Fatty acids 2080
n-tetradecanoic acid 24
n-pentadecanoic acid 8
n-hexadecanoic acid 400
n-heptadecanoic acid 9
octadeca-9,12-dienoic acid 164
octadec-9-enoic acid 228
n-octadecanoic acid 112
n-nonadecanoic acid 5
n-eicosanoic acid 53
n-heneicosanoic acid 20
n-docosanoic acid 122
n-tricosanoic acid 66
n-tetracosanoic acid 114
n-pentacosanoic acid 32
n-hexacosanoic acid 104
n-heptacosanoic acid 13
n-octacosanoic acid 213
n-nonacosanoic acid 9
n-triacontanoic acid 104
n-hentriacontanoic acid 5
n-dotriacontanoic acid 69
n-tritriacontanoic acid 2
n-tetratriacontanoic acid 13

n-Fatty alcohols 1615

n-docosanol 14
n-tricosanol 1
n-tetracosanol 49
n-pentacosanol 7
n-hexacosanol 94
n-heptacosanol 30
n-octacosanol 1392
n-nonacosanol 12
n-triacontanol 16

n-Alkanes 371
n-tricosane 1
n-tetracosane 1
n-pentacosane 6
n-hexacosane 3
n-heptacosane 34
n-octacosane 7
n-nonacosane 157
n-triacontane 7
n-hentriacontane 128
n-dotriacontane 0
n-tritriacontane 27

203
V. RESULTADOS Y DISCUSIÓN

n-Aldehydes 99
n-eicosanal 2
n-heneicosanal 0
n-docosanal 2
n-tricosanal 0
n-tetracosanal 3
n-pentacosanal 0
n-hexacosanal 10
n-heptacosanal 4
n-octacosanal 69
n-nonacosanal 1
n-triacosanal 6
n-dotriacosanal 2

High molecular weight esters a 915

esters C38 17
esters C39 3
esters C40 102
esters C41 10
esters C42 113
esters C43 17
esters C44 273
esters C45 10
esters C46 189
esters C47 8
esters C48 85
esters C49 7
esters C50 90
esters C51 3
esters C52 42
esters C54 5

Monoglycerides 127
2,3-dihydroxypropyl tetradecanoate 1
2,3-dihydroxypropyl hexadecanoate 26
2,3-dihydroxypropyl octadecadienoate 8
2,3-dihydroxypropyl octadecenoate 10
2,3-dihydroxypropyl octadecanoate 9
2,3-dihydroxypropyl eicosanoate 2
2,3-dihydroxypropyl docosanoate 5
2,3-dihydroxypropyl tricosanoate 1
2,3-dihydroxypropyl tetracosanoate 5
2,3-dihydroxypropyl pentacosanoate 1
2,3-dihydroxypropyl hexacosanoate 6
2,3-dihydroxypropyl heptacosanoate 1
2,3-dihydroxypropyl octacosanoate 28
2,3-dihydroxypropyl nonacosanoate 2
2,3-dihydroxypropyl triacontanoate 22

Diglicerides 85
1,2-dipalmitin 13
1,3-dipalmitin 14
1,2-palmitoyllinolein 5

204
 V. RESULTADOS Y DISCUSIÓN

1,2-palmitoylolein 5
1,2-palmitoylstearin 2
1,3-palmitoyllinolein 7
1,3-palmitoylolein 10
1,3-palmitoylstearin 12
1,2-diolein 3
1,3-diolein 4
1,2-distearin 1
1,3-distearin 9

Triglycerides 198
dipalmitoylolein 46
dioleoylpalmitin 91
triolein 61

-Diketones 883
14,16-hentriacontanedione 875
12,14-tritriacontanedione 7

Steroid hydrocarbons 16
ergosta-3,5-diene 3
stigmasta-3,5,22-triene 4
stigmasta-4,22-diene 1
stigmasta-3,5,7-triene 2
stigmasta-3,5-diene 6

Steroid ketones 88
stigmasta-4,22-dien-3-one 6
stigmasta-3,5-dien-7-one 23
ergost-4-ene-3,6-dione 4
ergostane-3,6-dione 6
stigmasta-4,22-diene-3,6-dione 1
stigmast-22-ene-3,6-dione 3
stigmast-4-ene-3,6-dione 21
stigmastane-3,6-dione 24

Sterols 1121
campesterol 300
stigmasterol 240
sitosterol 581

Sterol glycosides 680

campesteryl--D-glucopyranoside 164
stigmasteryl--D-glucopyranoside 191
sitosteryl--D-glucopyranoside 325

Sterol esters 70
campesterol esters 12
stigmasterol esters 6
sitosterol esters 53
a
See Table 3 for the detailed description of the individual esters.

205
V. RESULTADOS Y DISCUSIÓN

OH
(I)
O O

OH OH
(II) (III)

OH
(IV)

(V)
O

H
(VI)
O

O OH
(VII) OH

HO
O
O
O
(VIII) O

O O

O O
OH
(IX)
O

O
O O

O (X) O

O
(XI)

O O

(XII)

Figure 2. Structures representative of the main aliphatic lipophilic compounds identified in

wheat straw and referred in the text. (I) hexadecanoic (palmitic) acid; (II) 9-octadecenoic
(oleic) acid; (III) 9,12-octadecadienoic (linoleic) acid; (IV) n-octacosanol; (V) n-
nonacosane; (VI) n-nonacosanal; (VII) 2,3-dihydroxypropyl hexadecanoate (1-
monopalmitin); (VIII) 1,2-dipalmitin; (IX) 1,3-dipalmitin; (X) dioloylpalmitin; (XI)
hexadecanoic acid, octacosyl ester; (XII) 14,16-hentriacontanedione.

206
 V. RESULTADOS Y DISCUSIÓN

HO HO HO

(XIII) (XIV) (XV)

OH O
O
HO O O
HO
OH
(XVI) (XVII)

R R

O O O O
O O
(XVIII) (XIX) (XX) (XXI)

O O
O O
(XXII) (XXIII) (XXIV)

Figure 3. Structures of the main steroid compounds identified in wheat straw and referred
in the text. (XIII) sitosterol; (XIV) campesterol; (XV) stigmasterol; (XVI) sitosteryl 3β-D-
glucopyranoside; (XVII) sitosteryl palmitate; (XVIII) stigmasta-4,22-dien-3-one; (XIX)
stigmasta-3,5-dien-7-one; (XX, R=H) ergost-4-ene-3,6-dione; (XX, R=CH3) stigmast-4-
ene-3,6-dione; (XXI, R=H) ergostane-3,6-dione; (XXI, R=CH3) stigmastane-3,6-dione;
(XXII) stigmasta-4,22-diene-3,6-dione; (XXIII) stigmast-22-ene-3,6-dione; (XXIV)
stigmasta-3,5-diene.

Aliphatic series

Free fatty acids were the most predominant series in the extracts of wheat straw,
accounting for 2080 mg/Kg. The series ranges from tetradecanoic acid (C14) to
tetratriacontanoic acid (C34), with a strong even-over-odd carbon atom number

207
V. RESULTADOS Y DISCUSIÓN

predominance, and hexadecanoic (palmitic) acid (I) being the most predominant.
The unsaturated 9-octadecenoic (oleic, II), and 9,12-octadecadienoic (linoleic, III)
acids, were also found in important amounts, as already.13 However, previous
papers have reported the occurrence of important amounts of abietic acid,13–15 a
compound that is restricted only to conifers and should not be present among the
lipids in wheat straw. Its occurrence could suggest cross-contamination of the lipids
from other lignocellulosic sources. Free fatty alcohols were the second most
abundant class of aliphatic series in the extracts of wheat straw, accounting for
1615 mg/Kg, although their occurrence were not reported before.13–15

400 1400
(a) (b)
Abundance (mg/Kg)

Abundance (mg/Kg)

14 16 18 20 22 24 26 28 30 32 34 22 24 26 28 30
18:2
18:1

160 70
(c) (d)
Abundance (mg/Kg)

Abundance (mg/Kg)

23 25 27 29 31 33 20 22 24 26 28 30 32

30 300
(e) (f)
Abundance (mg/Kg)

Abundance (mg/Kg)

14 16 18 20 22 24 26 28 30 38 40 42 44 46 48 50 52 54
18:2
18:1

Figure 4. Distribution of the main aliphatic series identified in the extracts of wheat straw.
(a) n-Fatty acids; (b) n-fatty alcohols; (c) n-alkanes; (d) n-aldehydes; (e) monoglycerides;
(f) high molecular weight esters. The histograms are scaled up to the abundance of the
major compound in the series.

208
 V. RESULTADOS Y DISCUSIÓN

Free fatty alcohols were found in the range from n-docosanol (C22) to n-
triacontanol (C30), with a strong even-over-odd carbon atom number predominance,
and n-octacosanol (IV) being the most predominant homolog in the series. In fact,
n-octacosanol was the most important single compound among the lipids of wheat
straw. The series of n-alkanes was present in lower amounts (371 mg/Kg) and
ranged from n-tricosane (C23) to n-tritriacontane (C33), with a strong odd-over-even
atom carbon number predominance and nonacosane (V) being the predominant
homolog in the series, followed by hentriacontane. Finally, minor amounts of n-
aldehydes (99 mg/Kg) were identified from n-eicosanal (C20) to n-dotriacosanal
(C32), with a strong even-over-odd atom carbon atom predominance and n-
octacosanal (VI) being the major compound in the series. The distribution of the
aldehyde series parallels that of free alcohols, as usually occurs in the plant
kingdom and observed in other plants,9,10 suggesting that aldehydes are
intermediates in the biosynthesis of alcohols from fatty acids.24,25 Fatty alcohols,
alkanes and aldehydes were not detected in previous papers, 13–15 although alkanes
and aldehydes were already reported in wheat straw by Deswarte et al.7

A series of high molecular weight esters also occurred in wheat straw extracts in
important amounts (915 mg/Kg). High molecular weight esters were found in the
range from C38 to C48 with a strong predominance of the even atom carbon number
homologues, and the C44 and C46 analogs being the most abundant ones. Our results
completely differ from previous papers that only reported the presence of high
molecular weight esters C32 and C34,13–15 which were not detected in our study, but
failed to detect the important presence of esters of higher molecular weight. A close
examination of each chromatographic peak indicated that they consisted of a
mixture of esters of different long-chain fatty acids esterified to different long-
chain fatty alcohols. The identification and quantitation of the individual long-
chain esters in each chromatographic peak was resolved based on the mass spectra
of the peaks. Figure 5 shows the mass spectra of the chromatographic peaks
corresponding to the high molecular weight esters C44, C46, C48 and C50. The mass
spectra of long-chain esters are characterized by a base peak produced by a
rearrangement process involving the transfer of 2H atoms from the alcohol chain to
the acid chain giving a protonated acid ion.26 The fragments at m/z 257, 285, 313,
and 341 therefore correspond to the protonated hexadecanoic, octadecanoic,
eicosanoic, and docosanoic acids, respectively. Hence, the base peak gives
information about the number of carbon atoms in the acid moiety while the
molecular ion provides information about the total number of carbon atoms in the
ester. It is possible then to determine the contribution of individual esters in every
chromatographic peak by mass spectrometric determination of the molecular ion

209
V. RESULTADOS Y DISCUSIÓN

and the base peak. Quantitation of individual esters was accomplished by

integrating the areas in the chromatographic profiles of the ions characteristic for
the acidic moiety. The detailed structural composition of the different high
molecular weight ester waxes identified in wheat straw is shown in Table 3.

257
100%

(a)
Relative abundance

81

57
69 95
111
648
237 437

100 200 300 400 500 600 700 m/z

100% 81 285

(b)
57 95
Relative abundance

69

111

341
264 676
257 313
381 437

100 200 300 400 500 600 700 m/z

313
100%

(c)
Relative abundance

57
81
69
97

111

369
341 437 704

100 200 300 400 500 600 700 m/z

100% 341

(d)
Relative abundance

81
57
97
69

111

437 732

100 200 300 400 500 600 700 m/z

Figure 5. Mass spectra of the chromatographic peaks corresponding to the high molecular
weight esters (a) C44, (b) C46, (c) C48 and (d) C50. Mass fragments at m/z 257, 285, 313 and
341 correspond to the protonated fatty acid moieties (hexadecanoic, octadecanoic,
eicosanoic and docosanoic acids, respectively).

210
 V. RESULTADOS Y DISCUSIÓN

Table 3. Composition and Abundance (mg/Kg, d.a.f.) of the Different Individual Esters
Identified Among the Waxes Identified in the Extracts of Wheat Straw.

Compound Fattyacid:Fatty alcohol Abundance

esters C38 17
tetradecanoic acid, tetracosyl ester C14:C24 7
hexadecanoic acid, docosyl ester C16:C22 8
octadecanoic acid, eicosyl ester C18:C20 1
eicosanoic acid, octadecyl ester C20:C18 1

esters C39 3
hexadecanoic acid, tricosyl ester C16:C23 3

esters C40 102

dodecanoic acid, octacosyl ester C12:C28 3
tetradecanoic acid, hexacosyl ester C14:C26 6
hexadecanoic acid, tetracosyl ester C16:C24 81
docosyl ester C18:2:C22 2
octadec-9-enoic acid, docosyl ester C18:1:C22 2
octadecanoic acid, docosyl ester C18:C22 2
eicosanoic acid, eicosyl ester C20:C20 4
docosanoic acid, octadecyl ester C22:C18 2

esters C41 10
hexadecanoic acid, pentacosyl ester C16:C25 10

esters C42 113

tetradecanoic acid, octacosyl ester C14:C28 46
hexadecanoic acid, hexacosyl ester C16:C26 43
tetracosyl ester C18:2:C24 5
octadec-9-enoic acid, tetracosyl ester C18:1:C24 1
octadecanoic acid, tetracosyl ester C18:C24 10
eicosanoic acid, docosyl ester C20:C22 6
docosanoic acid, eicosyl ester C22:C20 2

esters C43 17
hexadecanoic acid, heptacosyl ester C16:C27 11
octadecanoic acid, pentacosyl ester C18:C25 1
eicosanoic acid, tricosyl ester C20:C23 1
docosanoic acid, heneicosyl ester C22:C21 4

esters C44 273

hexadecanoic acid, octacosyl ester C16:C28 253
hexacosyl ester C18:2:C26 4
octadec-9-enoic acid, hexacosyl ester C18:1:C26 1
octadecanoic acid, hexacosyl ester C18:C26 3
eicosanoic acid, tetracosyl ester C20:C24 7
docosanoic acid, docosyl ester C22:C22 5

esters C45 10
hexadecanoic acid, nonacosyl ester C16:C29 6
octadecanoic acid, heptacosyl ester C18:C27 1

211
V. RESULTADOS Y DISCUSIÓN

eicosanoic acid, pentacosyl ester C20:C25 1

docosanoic acid, tricosyl ester C22:C23 2

esters C46 189

hexadecanoic acid, triacontyl ester C16:C30 22
octacosyl ester C18:2:C28 52
octadec-9-enoic acid, octacosyl ester C18:1:C28 19
octadecanoic acid, octacosyl ester C18:C28 74
eicosanoic acid, hexacosyl ester C20:C26 9
docosanoic acid, tetracosyl ester C22:C24 27
tetracosanoic acid, docosyl ester C24:C22 4

esters C47 8
docosanoic acid, pentacosyl ester C22:C25 8

esters C48 85
hexadecanoic acid, dotriacontyl ester C16:C32 4
octadecanoic acid, triacontyl ester C18:C30 1
eicosanoic acid, octacosyl ester C20:C28 62
docosanoic acid, hexacosyl ester C22:C26 7
tetracosanoic acid, tetracosyl ester C24:C24 10

esters C49 7
docosanoic acid, heptacosyl ester C22:C27 7

esters C50 90
eicosanoic acid, triacontyl ester C20:C30 2
docosanoic acid, octacosyl ester C22:C28 81
tetracosanoic acid, hexacosyl ester C24:C26 2
hexacosanoic acid, tetracosyl ester C26:C24 3
octacosanoic acid, docosyl ester C28:C22 2

esters C51 3
docosanoic acid, nonacosyl ester C22:C29 3

estersC52 42
docosanoic acid, triacontyl ester C22:C30 1
tetracosanoic acid, octacosyl ester C24:C28 27
hexacosanoic acid, hexacosyl ester C26:C26 2
octacosanoic acid, tetracosyl ester C28:C24 12

esters C54 2
hexacosanoic acid, octacosyl ester C26:C28 1
octacosanoic acid, hexacosyl ester C28:C26 1

The esterified fatty acids ranged from dodecanoic acid (C12) to octacosanoic
acid (C28) and the esterified fatty alcohols from octadecanol (C18) to triacontanol
(C30). The acyl moiety of the high molecular weight ester waxes was mostly
constituted by saturated fatty acids with even carbon atom number, although high

212
 V. RESULTADOS Y DISCUSIÓN

molecular weight esters with unsaturated fatty acids (oleic and linoleic acids) could
also be detected. In addition, even atom carbon number esters were also identified,
and they mostly corresponded to odd carbon atom number fatty alcohols.
According to our analyses, the predominant high molecular weight ester in wheat
straw was C44, which was mostly constituted by hexadecanoic acid, octacosyl ester
(XI).

Finally, glycerides (mono-, di- and triglycerides), were also found among the
lipophilic extractives in wheat straw, although in lower amounts. Monoglycerides
accounted for 127 mg/Kg, and ranged from 2,3-dihydroxypropyl tetradecanoate to
2,3-dihydroxypropyl triacontanoate, with a strong even-over odd carbon atom
number predominance, and with 2,3-dihydroxypropyl hexadecanoate (1-
monopalmitin, VII) being the most abundant. The unsaturated monoglycerides 1-
monoolein and 1-monolinolein were also present in minor amounts. Diglycerides
were also found in low amounts (85 mg/Kg), the most abundant being 1,2-
dipalmitin (VIII) and 1,3-dipalmitin (IX). Finally, triglycerides were also identified
among the lipophilic extractives of wheat straw and accounted for 198 mg/Kg,
dioleoylpalmitin (X) being the most abundant.

β-diketones

The analysis of the lipophilic extractives of wheat straw revealed the presence of
important amounts (883 mg/Kg) of a compound with a β-diketone structure. The
identification of this compound was achieved based on its mass spectrum (Figure
6). The molecular ion at m/z 464 indicates that this is a hentriacontanedienone, and
the fragments at m/z 250 and m/z 278 that arise from the McLafferty
rearrangement at both sides of the diketone group followed by loss of water, 27
clearly indicate that the structure of this β-diketone is 14,16-hentriacontanedione
(XII). Despite 14,16-hentriacontanedione was the second most abundant single
compound among the lipophilic extractives in wheat straw, its occurrence was not
reported in previous papers.13–15 Minor amounts of 12,14-tritriacontanedione were
also present among the lipophilic compounds of wheat straw. β-Diketones are
relatively common constituents of plant waxes and have been identified in the leafs
of different grasses, including wheat straw.28–35

Steroid compounds

Different classes of steroid compounds were present in the extracts of wheat

straw, namely steroid hydrocarbons, steroid ketones, sterols, sterol glycosides and
sterol esters. Free sterols were the most abundant steroid compounds, accounting

213
V. RESULTADOS Y DISCUSIÓN

for 1135 mg/Kg. Sitosterol (XIII) was the most important sterol in wheat straw,
together with campesterol (XIV) and stigmasterol (XV). Surprisingly, previous
papers reported the occurrence of ergosterol in wheat straw. 13–15 However,
ergosterol is a characteristic sterol in fungi and does not occur in plant cells;
therefore, its occurrence may be attributable to fungal presence in the wheat straw
sample analyzed in those papers and its probable degradation. Minor amounts of
sterols were found esterified forming sterol esters (70 mg/Kg), sitosteryl palmitate
(XVII) being the most important one. Sterol glycosides were also identified among
the lipophilic extractives of wheat straw in important amounts (680 mg/Kg).
Sitosteryl 3β-D-glucopyranoside (XVI) was the most predominant with lower
amounts of campesteryl and stigmasteryl β-D-glucopyranosides. The identification
of sterol glycosides was accomplished (after BSTFA derivatization of the lipid
extract) by comparison with the mass spectra and relative retention times of
authentic standards.36 Sterol glycosides were not reported previously among the
lipophilic compounds in wheat straw, despite their high abundance.13–15

138
100%
Relative abundance

95 464

278
250
446
100
124
81

151
235 263 281 413
192

100 200 300 400 500

m/z

Figure 6. Mass spectrum of the β-diketone (14,16-hentriacontanedione, XII) identified

among the lipophilic extractives of wheat straw.

Steroid ketones were observed in low amounts (88 mg/Kg) and consisted mainly
of stigmasta-4,22-dien-3-one (XVIII), stigmasta-3,5-dien-7-one (XIX), ergost-4-

214
 V. RESULTADOS Y DISCUSIÓN

ene-3,6-dione (XX, R=H), stigmast-4-ene-3,6-dione (XX, R=CH3), ergostane-3,6-

dione (XXI, R=H), stigmastane-3,6-dione (XXI, R=CH3), stigmasta-4,22-diene-
3,6-dione (XXII), and stigmast-22-ene-3,6-dione (XXIII). Finally, minor amounts
of steroid hydrocarbons (16 mg/Kg) were also identified, stigmasta-3,5-diene
(XXIV) being the most important one, and with lower amounts of ergosta-3,5-
diene, stigmasta-3,5,22-triene, stigmasta-4,22-diene and stigmasta-3,5,7-triene.
Most probably, these steroid hydrocarbons might arise from degradation of free and
conjugated sterols, either within the plant or during the lipids isolation and/or
analysis.

In conclusion, the present paper provides for the first time a detailed and
comprehensive description of the lipophilic compounds in wheat straw, which is a
highly valuable information for a more complete industrial utilization of this
lignocellulosic material that is regarded as a waste.

FUNDING

This study has been funded by the Spanish project AGL2011-25379, the CSIC
project 201040E075 and the EU-project LIGNODECO (KBBE-244362). Pepijn
Prinsen thanks the Spanish Ministry of Science and Innovation for a FPI
fellowship.

ACKNOWLEDGEMENTS

We thank Jorge Rencoret for technical assistance in the GC-MS analyses.

REFERENCES

(1) Ragauskas, A. J.; Williams, C. K.; Davison, B. H.; Britovsek, G.; Cairney,
J.; Eckert, C. A.; Frederick, W. J.; Hallett, J. P.; Leak, D. J.; Liotta, C. L.; Mielenz,
J. R.; Murphy, R.; Templer, R.; Tschaplinski. T. The path forward for biofuels and
biomaterials. Science 2006, 311, 484–489.

(2) Clark, J. H. Green chemistry for the second generation biorefinery -

sustainable chemical manufacturing based on biomass. J. Chem. Technol.
Biotechnol. 2007, 82, 603-609.

(3) Kamm, B.; Kamm, M. The concept of a biorefinery – production of platform

chemicals and final products. Chem. Ing. Tech. 2007, 79, 592–603.

215
V. RESULTADOS Y DISCUSIÓN

(4) Sarkar, N.; Ghosh, S. K.; Bannerjee, S.; Aikat, K. Bioethanol production
from agricultural wastes: An overview. Renewable Energy 2012, 37, 19–27.

(5) Kim, S.; Dale, B. E. Global potential bioethanol production from wasted
crops and crop residues. Biomass Bioenerg. 2004, 4, 361–375.

(6) del Río, J.C.; Rencoret, J.; Prinsen, P.; Martínez, A.T.; Ralph, J.; Gutiérrez,
A. Structural characterization of wheat straw lignin as revealed by analytical
pyrolysis, 2D-NMR, and reductive cleavage methods. J. Agric. Food Chem. 2012,
60, 5922–5935.

(7) Deswarte, F. E. I.; Clark, J. H.; Hardy, J. J. E.; Rose, P. M. The fractionation
of valuable wax products from wheat straw using CO2. Green Chem. 2006, 8, 39–
42.

(8) Back, E. L.; Allen, L. H. Pitch Control, Wood Resin and Deresination,
Tappi press, Atlanta, GA., 2000; pp. 392.

(9) Gutiérrez, A.; del Río, J. C. Lipids from flax fibers and their fate in alkaline
pulping. J. Agric. Food Chem. 2003a, 51, 4965–4971.

(10) Gutiérrez, A.; del Río, J. C. Lipids from flax fibers and their fate in alkaline
pulping. (Addition/Correction). J. Agric. Food Chem. 2003b, 51, 6911–6914.

(11) Freire, C.S.R.; Silvestre, A.J.D.; Pascoal Neto, C.; Evtuguin, D.V. Effect of
oxygen, ozon and hydrogen peroxide bleaching stages on the contents and
composition of extractives of Eucalyptus globulus kraft pulps. Biores. Technol.
2006, 97, 420–428.

(12) Marques, G.; del Río, J. C.; Gutiérrez, A. Lipophilic extractives from
several nonwoody lignocellulosic crops (flax, hemp, sisal, abaca) and their fate
during alkaline pulping and TCF/ECF bleaching. Biores. Technol. 2010, 101, 260–
267.

(13) Sun, R. C.; Sun, X. F. Identification and quantitation of lipophilic

extractives from wheat straw. Ind. Crops & Prod. 2001, 14, 51–64.

(14) Sun, R. C.; Tomkinson, J. Comparative study of organic solvent and water-
soluble lipophilic extractives from wheat straw I: yield and chemical composition.
J. Wood Sci. 2003, 49, 47–52.

216
 V. RESULTADOS Y DISCUSIÓN

(15) Sun, R. C.; Salisbury, D.; Tomkinson, J. Chemical composition of

lipophilic extractives released during the hot water treatment of wheat straw.
Biores. Technol. 2003, 88, 95–101.

(16) Gutiérrez, A.; del Río, J. C.; González-Vila, F. J.; Martín, F. Analysis of
lipophilic extractives from wood and pitch deposits by solid-phase extraction and
gas chromatography. J. Chromatogr. A 1998, 823, 449–455.

(17) Gutiérrez, A.; Rodríguez, I.M.; del Río, J.C. Chemical characterization of
lignin and lipid fractions in industrial hemp bast fibers used for manufacturing
high-quality paper pulps. J. Agric. Food Chem. 2006, 54, 2138–2144.

(18) Gutiérrez, A.; Rodríguez, I. M.; del Río, J. C. Chemical characterization of

lignin and lipid fractions in kenaf bast fibers used for manufacturing high-quality
papers. J. Agric. Food Chem. 2004, 52, 4764–4773.

19 ; g I; f
lipophilic extractives from sisal (Agave sisalana) fibers. Ind. Crops Prod. 2008, 28,
81−87

0 ; f (Musa textilis)
leaf fibers used for manufacturing of high quality paper pulps. J. Agric. Food
Chem. 2006, 54 4600−4610

1 ; ; g I ; Chemical
composition of lipophilic extractives from jute (Corchorus capsularis) fibers used
for manufacturing of high-quality paper pulps. Ind. Crops Prod. 2009, 30,
41− 49

; ; ;S v ;
Chemical characterization of the lipophilic fraction of Giant reed (Arundo donax)
fibres used for pulp and paper manufacturing. Ind. Crops Prod. 2007, 26 9− 36

(23) Villaverde, J. J.; Domingues, R. M. A.; Freire, C. S. R.; Silvestre, A. J. D.;

Pascoal Neto, C.; Ligero, P.; Vega, A. Miscanthus x giganteus extractives: A
source of valuable phenolic compounds and sterols. J. Agric. Food Chem. 2009, 57,
36 6−3631

(24) Tulloch, A. P. Chemistry of waxes of higher plants. In Chemistry and

Biochemistry of Natural Waxes; Kolattukudy, P. E., Ed.; Elsevier: Amsterdam,
1976; 36− 5

217
V. RESULTADOS Y DISCUSIÓN

(25) Bianchi, G. Plant waxes. In Waxes: Chemistry, Molecular Biology and

Functions; Hamilton, R. J., Ed.; The Oily Press: Dundee, Scotland, 1995; pp
175−

(26) Moldovan, Z.; Jover, E.; Bayona, J. M. Systematic characterization of long-

chain aliphatic esters of wool wax by gas g y−
ionisation mass spectrometry. J. Chromatogr. A 2002, 952 193− 04

(27) Evans, D.; Knights, B. A.; Math, V. B.; Ritchie, A. L. β-Diketones in

Rhododendron waxes. Phytochemistry 1975, 14 447− 451

(28) Tulloch, A. P.; Hoffman, L. L. Leaf wax of Triticum aestivum.

Phytochemistry 1973, 12, 2217–2223.

(29) Tulloch, A. P.; Hoffman, L. L. Epicuticular waxes of secale cereal and

tricale hexaploide leaves. Phytochemistry 1974, 13, 2535–2540.

(30) Tulloch, A. P.; Hoffman, L. L. Epicuticular wax of Apropyron

intermedium. Phytochemistry 1976, 15, 1145–1151.

(31) Tulloch, A. P. Carbon-13 NMR spectra of β-diketones from wax of the

gramineae. Phytochemistry 1985, 24, 131–137.

(32) Bianchi, A.; Bianchi, G. Surface lipid composition of C3 and C4 plants.

Biochem. System. Ecol.1990, 18, 533–537.

(33) Bianchi, G.; Figini, M. L. Epicuticular waxes of glaucous and nonglaucous

durum wheat lines. J. Agric. Food Chem. 1986, 34, 429–433

(34) Prinsen, P.; Gutiérrez, A.; del Río, J.C. Lipophilic extractives from the
cortex and pith of elephant grass (Pennisetum purpureum Schumach.) stems. J.
Agric. Food Chem. 2012, 60, 6408–6417.

(35) Athukorala, Y.; Mazza, G. Supercritical carbon dioxide and hexane

extraction of wax from triticale straw: content, composition and thermal properties.
Ind. Crops & Prod. 2010, 31, 550–556.

(36) Gutiérrez, A.; del Río, J. C. Gas chromatography/mass spectrometry

demonstration of steryl glycosides in eucalypt wood, kraft pulp and process liquids.
Rapid Commun. Mass Spectrom. 2001, 15, 2515–2520.

218
 V. RESULTADOS Y DISCUSIÓN

Publicación V:

del Río J.C., Rencoret J., Prinsen P., Martínez A.T., Ralph J. and Gutiérrez A.
(2012). Structural characterization of wheat straw lignin as revealed by analytical
pyrolysis, 2D-NMR, and reductive cleavage methods. Journal of Agricultural Food
and Chemistry, 60, 5922–5935.

219
 V. RESULTADOS Y DISCUSIÓN

Structural Characterization of Wheat Straw Lignin as Revealed by

Analytical Pyrolysis, 2D-NMR, and Reductive Cleavage Methods

†
Jorge Rencoret,† ‡ Pepijn Prinsen,† g §
John
Ralph,‡ †

†
Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS), CSIC, PO
Box 1052, E-41080 Seville, Spain

‡
Departments of Biochemistry and Biological Systems Engineering, the Wisconsin
Bioenergy Initiative, and the DOE Great Lakes Bioenergy Research Center,
University of Wisconsin, Madison, Wisconsin 53706, United States

§
I v g g I SI 9 -28040
Madrid, Spain

ABSTRACT

The structure of the lignin in wheat straw has been investigated by a combination
of analytical pyrolysis, 2D-NMR and derivatization followed by reductive cleavage
(DFRC). It is a p- hydroxyphenyl-guaiacyl-syringyl lignin (with an H:G:S ratio of
6:64:30) associated with p-coumarates and ferulates. 2D-NMR indicated that the
main substructures present are β-O-4′-ethers (∼75%), followed by
phenylcoumarans (∼11%), with lower amounts of other typical units. A major new
finding is that the flavone tricin is apparently incorporated into the lignins. NMR
and DFRC indicated that the lignin is partially acylated (∼10%) at the γ-carbon,
predominantly with acetates that preferentially acylate guaiacyl (12%) rather than
syringyl (1%) units; in dicots, acetylation is predominantly on syringyl units. p-
Coumarate esters were barely detectable (<1%) on monomer conjugates released
by selectively cleaving β-ethers in DFRC, indicating that they might be
preferentially involved in condensed or terminal structures.

221
V. RESULTADOS Y DISCUSIÓN

KEYWORDS: wheat straw, Py-GC/MS, TMAH, HSQC, DFRC, milled wood

lignin, p-coumarate, ferulate, coniferyl acetate, tricin.

INTRODUCTION

Energy consumption has increased gradually over the last decades as the world
population has grown and more countries have become industrialized. Crude oil
has been the major resource used to meet the increased energy demand. However,
concerns about declining of energy resources and the need to mitigate green-house
gas emissions and decrease our dependency on fossil fuel reserves have focused
attention on the use of plant biomass as a source for the production of biofuels
and/or bioproducts.1

The first generation of biofuel feedstocks included sugar cane and cereal grains.
Bioconversion of such crops to biofuels, however, competes with food production
for land and has a considerable effect on food and feed prices. A promising
alternative for second generation biofuels will come from cultivated lignocellulosic
crops or agricultural wastes, which are available in high amounts at relatively low
cost and could be a widely available and relatively inexpensive source for biofuels
and/or bioproducts. Therefore, increasing attention is being paid to the use of
lignocellulosic biomass as a renewable feedstock for the above industrial uses. −4

Common lignocellulosic feedstocks considered for second generation biofuel

production include woods (e.g., poplar or eucalyptus), perennial energy crops (e.g.,
switchgrass or Miscanthus species), and agricultural wastes (e.g., corn stover or
cereal straws). Among them, wheat straw has the greatest potential of all
agricultural residues because of its wide availability and low cost.4,5 Wheat straw is
an abundant byproduct from wheat production in many countries. The average
y fw w 1 3−1.4 kg/kg of wheat grain, with a world production of
wheat estimated to be around 680 million tons in 2011. Wheat straw contains
35−45% 0−30% 15% g w k
an attractive feedstock to be converted to ethanol and other value-added products.

The conversion of lignocellulosic biomass to bioethanol involves

saccharification of carbohydrates to fermentable reducing sugars via hydrolysis and
then fermentation of these free sugars to ethanol. However, the presence of lignin, a
complex and amorphous polymer playing a major structural role in vascular plants,
limits the accessibility of enzymes to cellulose, thus reducing the efficiency of the

222
 V. RESULTADOS Y DISCUSIÓN

hydrolysis.1,6 Pretreatment of lignocellulosic materials to remove or modify the

lignin is therefore needed to enhance the hydrolysis of carbohydrates. 7,8 The
efficiency of pretreatment methods is highly dependent on the lignin structure, and
hence a knowledge of the structure of the lignin polymer in different plant species
is important to develop appropriate pretreatment methods for lignin modification
and/ or removal.

Lignin is a complex macromolecule synthesized by chemical polymerization of

three main precursors, p-coumaryl (4-hydroxycinnamyl), coniferyl (4-hydroxy-3-
methoxycinnamyl), and sinapyl (3,5-dimethoxy-4-hydroxycinnamyl) alcohols, via
enzymatically generated radicals.9 These monolignols produce the p-
hydroxyphenyl (H), guaiacyl (G), and syringyl (S) phenylpropanoid lignin units
when incorporated into the lignin polymer, units which are linked by several types
f − bonds. The lignin composition depends on the botanical origin.
Thus hardwood lignins are composed of S and G units in varying ratios, softwood
lignin is composed of G units and small amounts of H units, and grass lignins
include the three units (with H-units still comparatively minor), making its
structure apparently more complex. Additionally, p-hydroxycinnamates (p-
coumarates and ferulates) also widely occur in the structure of grass lignins, with p-
y y g γ-OH of the lignin side chain, whereas ferulates
and diferulates acylate cell wall polysaccharides and participate in both
y − y g − y -coupling
reactions, in the latter case becoming integrally bound into the lignin polymer.10

The composition and structure of the lignin in wheat straw has been a matter of
study for many years.11−17 In this paper, a more in-depth and complete
characterization of the lignin of wheat straw has been performed by the use of an
array of analytical techniques, including Py-GC/MS (in the absence and in the
presence of tetramethylammonium hydroxide, TMAH, as transesterification and
methylating agent), 2D-NMR, and derivatization followed by reductive cleavage
(DFRC), and important discrepancies with the data reported in previous papers
have been found. In this work, lignin was also isolated from wheat straw according
18
to the classical procedure of j k to complement the analyses performed on
whole lignocellulosic material. A knowledge of the composition and structure of
wheat straw lignin will help to maximize the exploitation of this important
agricultural waste as a feedstock for biofuels and other biorefinery products.

223
V. RESULTADOS Y DISCUSIÓN

MATERIALS AND METHODS

Samples

Wheat straw (Triticum durum var. Carioca) was harvested from an experimental
field in Seville (South Spain) in June 2009. Wheat straw was air-dried, and the
dried samples were milled using a knife mill (1 mm screen) and successively
extracted with acetone (200 mL) in a Soxhlet apparatus for 8 h (at which time the
extracting solvent was clear and extractive-free) and hot water (100 mL, 3 h at 100
°C). Klason lignin content was estimated as the residue after sulfuric acid
hydrolysis of the pre-extracted material, corrected for ash and protein content,
according to the TAPPI method T222 om-88.19 The acid-soluble lignin was
determined, after the insoluble lignin was filtered off, by UV spectroscopic
determination at 205 nm wavelength using 110 L cm−1 g−1 as the extinction
coefficient. Holocellulose was isolated from the preextracted fibers by
delignification for 4 h using the acid chlorite method.20 The α-cellulose content was
determined by removing the hemicelluloses from the holocellulose by alkali
extraction.20 Ash content was estimated as the residue after 6 h of heating at 575
°C. Three replicates were used for each sample.

“Milled Wood Lignin” (MWL) Isolation

The wheat straw MWL was obtained according to the classical method,18 from
extractive-free wheat straw. The experimental procedure has been explained in
detail in previous papers.21 The final yield was ∼20% based on the Klason lignin
content of the original material.

Gel Permeation Chromatography (GPC)

GPC of the isolated MWL was performed on a Shimadzu LC-20A liquid

chromatography system (Shimadzu, Kyoto, Japan) equipped with a photodiode
array detector (SPD-M20A; Shimadzu) using the following conditions: TSK g α-
+ α-2500 (Tosoh, Tokyo, Japan) column; 0.1 M LiBr in dimethylformamide
(DMF) eluent; 0.5 mL min−1 flow rate; 40 °C column oven temperature; and 280
nm sample detection. The data acquisition and computation used LCsolution
version 1.25 software (Shimadzu). The molecular weight calibration was via
polystyrene standards.

224
 V. RESULTADOS Y DISCUSIÓN

Analytical Pyrolysis

Pyrolysis of wheat straw and the isolated W x y 100 μg w

performed with a 2020 microfurnace pyrolyzer (Frontier Laboratories Ltd.)
connected to an Agilent 6890 GC/MS system equipped with a DB-1701 fused-
silica capillary 30 × 0 5 0 5 μ f k
Agilent 5973 mass selective detector (EI at 70 eV). The pyrolysis was performed at
500 °C. The GC oven temperature was programmed from 50 °C (1 min) to 100 at
30 °C min−1 and then to 290 °C (10 min) at 6 °C min−1. Helium was the carrier gas
(1 mL min−1 y/ H 100 μg f sample was mixed with approximately 0.5
μ f H 5% w/w methanol), and the pyrolysis was carried out as
described above. The compounds were identified by comparing their mass spectra
with those of the Wiley and NIST libraries and those reported in the literature.22,23
Peak molar areas were calculated for the lignin-degradation products, the summed
areas were normalized, and the data for two repetitive analyses were averaged and
expressed as percentages.

NMR Spectroscopy

For NMR of the whole cell wall material, around 100 mg of finely divided (ball-
milled) extractive-free samples was swollen in 0.75 mL of DMSO-d6 according to
the method previously described.24,25 In the case of the MWL, around 40 mg was
dissolved in 0.75 mL of DMSO-d6. NMR spectra were recorded at 25 °C on a
Bruker AVANCE III 500 MHz instrument equipped with a cryogenically cooled
5mmTCI gradient probe with inverse geometry (proton coils closest to the sample).
HSQC (heteronuclear single quantum coherence) x k ’
“ g ” g c-pulsed version) with spectral widths of
5000 Hz (from 10 to 0 ppm) and 20,843 Hz (from 165 to 0 ppm) for the 1H- and
13
C dimensions. The number of collected complex points was 2048 for the 1H-
dimension with a recycle delay of 1.5 s. The number of transients was 64, and 256
time increments were always recorded in the 13C dimension. The 1JCH used was 145
Hz. Processing used typical matched Gaussian apodization in the 1H dimension and
squared cosine-bell apodization in the 13C dimension. Prior to Fourier
transformation, the data matrixes were zero-filled up to 1024 points in the 13C-
dimension. v kw f δC 39 5; δH
2.49). Long range J-coupling evolution times of 66 and 80 ms were used in
different heteronuclear multiple bond correlation (HMBC) acquisition experiments.
HSQC correlation peaks were assigned by comparing with the literature. 6−35 A
semiquantitative analysis24,34,35 of the volume integrals (uncorrected) of the HSQC

225
V. RESULTADOS Y DISCUSIÓN

correlation peaks was f g k ’ 1 W w

Topspin 3.1 (Mac) processing software. In the aliphatic oxygenated region, the
relative abundances of side chains involved in the various interunit linkages were
estimated from the Cα−Hα correlations to avoid possible interference from
homonuclear 1H−1H couplings, except for substructures Aox and I, for which Cβ−Hβ
and Cγ−Hγ correlations had to be used. In the aromatic/unsaturated region, C2−H2
correlations from H, G, and S lignin units and from p-coumarate and ferulate were
used to estimate their relative abundances. Note that p-coumarate and ferulate
quantitation relative to the lignin is overestimated due to the longer relaxation times
of these end-units compared to the rapidly relaxing polymer and the more extensive
relaxation the latter experiences during the significant duration of the pulse
experiment itself.

DFRC (Derivatization Followed by Reductive Cleavage)

DFRC degradation was performed according to the developed protocol.36−38

Lignins (10 mg) were stirred for 2 h at 50 °C with acetyl bromide in acetic acid
(8:92, v/v). The solvents and excess acetyl bromide were removed by rotary
evaporation at reduced pressure. The products were then dissolved in
dioxane/acetic acid/water (5:4:1, v/v/v), and 50 mg of powdered Zn was added.
After 40 min stirring at room temperature, the mixture was transferred into a
separatory funnel with dichloromethane and saturated ammonium chloride. The pH
of the aqueous phase was adjusted to less than 3 by adding 3% aqueous HCl, the
mixture vigorously mixed, and the organic layer separated. The water phase was
extracted twice more with dichloromethane. The combined dichloromethane
fractions were dried over anhydrous Na2SO4, and the filtrate was evaporated on a
rotary evaporator. The residue was acetylated for 1 h in 1.1 mL of dichloromethane
containing 0.2 mL of acetic anhydride and 0.2 mL of pyridine. The acetylated
lignin degradation products were collected after rotary evaporation of the solvents,
and subsequently analyzed by GC/MS. To assess the presence of naturally
acetylated lignin units, the described modification of the standard DFRC method
using propionylating instead of acetylating g ′ v y
31,39,40
published, was used.

GC/MS analyses were performed with a GCMS-QP2010 plus instrument

(Shimadzu Co.) using a capillary column (SHR5XLB 30 m × 0.25 mm i.d., 0.25
μ f k v w f 140 ° 1 min) to 250 at 3 °C min−1,
then ramped at 10 °C min−1 to 280 °C (1 min) and finally ramped at 20 °C min−1 to
300 °C, and held for 18 min at the final temperature. The injector was set at 250

226
 V. RESULTADOS Y DISCUSIÓN

°C, and the transfer line was kept at 310 °C. Helium was used as the carrier gas at a
rate of 1 mL min−1. Quantitation of the released individual monomers was
f g 4 4′-ethylenebisphenol as internal standard. Molar yields were
calculated on the basis of molecular weights of the respective acetylated and/or
propionylated compounds.

RESULTS AND DISCUSSION

The relative abundances of the main constituents of wheat straw (water-soluble

material, acetone extractives, Klason lignin, acid soluble lignin, hemicelluloses,
cellulose, and ash) are presented in Table 1. The lignin content (16.2% Klason
lignin), as well as the content of the other main constituents (i.e., cellulose,
hemicelluloses, etc.), agrees well with the data reported in previous papers. 13,14,41,42
In this work, we have analyzed in detail the structural characteristics of the lignin
polymer in wheat straw. For this, we analyzed first the composition of the lignin in
situ by Py-GC/MS (both in the absence and in the presence of TMAH), and 2D-
NMR. Then, for a more detailed structural characterization, the lignin (often termed
“ w g ” W w y x x f
finely ball-milled wheat straw according to the classical lignin isolation
procedure.18

Table 1. Abundance of the Main Constituents of Wheat Straw (% Dry Weight).

Water solubles 9.6

Acetone extractives 2.7

Lipophilics (% of acetone extractives) 73.4

Polars (% of acetone extractives) 26.6

Klason lignina 16.2

Acid-soluble lignin 1.5

Holocellulose 58.9

Cellulose 32.0

Hemicelluloses 26.9

Ash 6.6
a
Corrected for proteins and ash.

227
V. RESULTADOS Y DISCUSIÓN

Molecular Weight Distribution of Wheat Straw MWL

The values of the weight-average (Mw) and number-average (Mn) molecular

weights were estimated from the GPC curves (relative values related to polystyrene
standards). The MWL exhibited a weight-average (Mw) molecular weight of 4210 g
mol−1 and a number-average (Mn) molecular weight of 1850. In addition, the MWL
exhibited relatively narrow polydispersity, with Mw/Mn of 2.27. The Mw value is
comparable to, or even slightly higher than, values previously reported for the
lignin in wheat straw.14,42

Py-GC/MS of Wheat Straw and Its Isolated MWL

The chemical composition of the lignin in wheat straw was analyzed in situ,
without prior isolation, by Py-GC/MS. In addition, the isolated MWL was also
analyzed by Py-GC/MS. The pyrograms of wheat straw and its isolated MWL are
shown in Figure 1. The identities and relative abundances of the lignin-derived
compounds released are listed in Table 2.

Pyrolysis of wheat straw (Figure 1a) released compounds derived from the
carbohydrate, lignin, and p-hydroxycinnamate moieties. p-Hydroxycinnamates, in
addition to being the precursors of monolignols, are widely present as such in
herbaceous plants10 13 43−46 and efficiently produce upon pyrolysis similar
compounds as those derived from lignin, such as 4-vinylphenol (from p-
coumarates) and 4-vinylguaiacol (from ferulates), which will overestimate the
composition of the H- and G-lignin units and affect the calculation of the S/G ratio
unless they are left out of the calculation.21 The important amounts of 4-
vinylphenol (∼17% of all phenolic compounds) and 4-vinylguaiacol (∼28% of all
phenolic compounds) released upon pyrolysis from wheat straw indicate the
presence of p-coumarates and ferulates in this sample, as will be shown below. It is
obvious then that these vinyl compounds cannot be used for the estimation of the
lignin H:G:S composition upon Py-GC/MS, as the major part of them do not arise
from the core lignin structural units but from p-hydroxycinnamates. An estimation
of the S/G ratio of the lignin in wheat straw was, however, performed by ignoring
4-vinylguaiacol (and the analogous 4-vinylsyringol) and revealed a value of 0.5
(Table 2).

Pyrolysis of the MWL isolated from wheat straw (Figure 1b) released a similar
distribution of cinnamate and lignin-derived compounds, except for the relatively
lower abundance of 4-vinylguaiacol (10), which is still the most predominant
compound in the pyrogram of MWL, and the much higher abundance of 4-

228
 V. RESULTADOS Y DISCUSIÓN

methylguaiacol (6). The S/G ratio of the MWL (estimated as above, by ignoring the
4-vinylguaiacol and its analogous 4-vinylsyringol) is similar to that observed in the
whole cell walls (S/G 0.5). These results confirm that the lignin in wheat straw is
an H:G:S lignin, in agreement with other papers,11,13,14,16,39 but in sharp contrast
with the results from Banoub et al.,15 which indicated the complete absence of S-
lignin units in the lignin of wheat straw.

The occurrence of p-hydroxycinnamates in the cell walls of wheat straw, as well

as in the isolated MWL, was assessed by pyrolysis in the presence of TMAH,21,46,47
as shown in Figure 2. The identities of the compounds released and their relative
abundances are listed in Table 3. Py/TMAH induces cleavage of th β-O-4-ether
bonds in the lignin and released products similar to those obtained upon CuO
alkaline degradation,41 including methylated aldehydes (peaks 6, 12, and 19),
ketones (peaks 15 and 24), and acids (peaks 10, 17, and 26).21,46,47 Py/TMAH also
induces transesterification of the p-hydroxycinnamate esters, and breakdown of
ether linkages at C4, with subsequent methylation of the phenolic hydroxyl
groups.21,46,47 As seen in Figure 2, Py/TMAH of wheat straw released important
amounts (over 15% of total peak areas) of the dimethyl derivative of p-coumaric
acid (peak 25), as well as similar amounts (13% of total peak area) of the methyl
derivative of ferulic acid (peak 30). In addition to the trans-forms, minor amounts
of the cis-isomers (peaks 18 and 28) were also identified. An estimation of the S/G
ratio can now be performed with fewer restrictions than in the case of conventional
pyrolysis because the compounds derived from lignin and p-hydroxycinnamates are
now clearly differentiated (Table 3). The value was similar in the cell walls and the
MWL (S/G 0.5) samples, and was also similar to that estimated by Py-GC/MS (by
ignoring 4-vinylguaiacol and 4-vinylsyringol). The relative abundances of p-
hydroxycinnamates (p-coumarate/ferulate ratio) present in wheat straw and the
isolated MWL, estimated by Py/TMAH (Table 3), revealed additional features.
Both p-coumarate and ferulate are present in the whole cell walls of wheat straw in
similar abundances, whereas in the isolated MWL p-coumarate was released in
higher relative abundance (∼22% of the Py/TMAH products analyzed) than
ferulate (∼4%). These data, therefore, indicate that ferulate is mostly attached to
carbohydrates while p-coumarate is predominantly attached to the lignin polymer.
Previous studies have indicated that the bulk of p-coumarate in wheat straw is
esterified to the lignin side chains13 f y y γ-OH of the
lignin side chain,11 as established for other grasses.26 Ferulates, on the other hand,
have previously been supposed to be mostly etherif α- β-carbons
48
g g − y g Recent studies have concluded that

229
V. RESULTADOS Y DISCUSIÓN

ferulates are an intrinsic part of the lignin structure in grasses, participating in

coupling and cross-coupling reactions with other monolignols.10

Table 2. Identities and Relative Abundances of the Lignin-Derived Phenolic Compounds

Released after Py-GC/MS of Wheat Straw and the Isolated MWL.
Label Compound MW Origin Wheat straw Wheat straw
cell-walls MWL
1 phenol 94 H 2.3 2.4
2 2-methylphenol 108 H 0.8 0.5
3 4-methylphenol 108 H 1.3 2.8
4 guaiacol 124 G 11.0 8.4
5 C2-phenol 122 H 0.3 0.3
6 4-methylguaiacol 138 G 3.7 12.0
7 4-vinylphenol 120 H/PCA 16.9 10.8
8 4-allylphenol 134 H 0.0 0.4
9 4-ethylguaiacol 152 G 1.4 2.9
10 4-vinylguaiacol 150 G/FA 27.7 14.8
11 cis-4-propenylphenol 134 H 0.0 0.1
12 eugenol 164 G 0.9 1.4
13 syringol 154 S 6.9 6.9
14 trans-4-propenylphenol 134 H 0.0 0.4
15 cis-isoeugenol 164 G 0.5 1.5
16 vanillin 152 G 2.7 5.3
17 4-methylsyringol 168 S 1.3 4.8
18 trans-isoeugenol 164 G 2.7 5.1
19 4-propinylguaiacol 162 G 1.2 1.2
20 4-allenylguaiacol 162 G 1.3 1.3
21 acetoguaiacone 166 G 0.4 1.7
22 4-ethylsyringol 182 S 0.6 0.8
23 guaiacylacetone 180 G 0.8 0.7
24 4-vinylsyringol 180 S 6.7 3.6
25 4-allylsyringol 194 S 0.6 1.0
26 cis-4-propenylsyringol 194 S 0.5 0.7
27 syringaldehyde 182 S 0.9 2.2
28 4-propinylsyringol 192 S 0.6 0.2
29 4-allenylyringol 192 S 0.6 0.2
30 trans-propenylsyringol 194 S 1.7 1.8
31 acetosyringone 196 S 0.6 2.3
32 trans-coniferaldehyde 178 G 1.8 1.0
33 syringylacetone 210 S 0.4 0.2
34 propiosyringone 210 S 0.0 0.2
35 trans-sinapaldehyde 208 S 0.7 0.3
S/G ratioa 0.5 0.5
a
H: lignin p-hydroxyphenyl-type. G: lignin guaiacyl-type. S: lignin syringyl-type. PCA: p-
coumarate. FA: ferulate. bAll G- and S-derived peaks were used for the estimation of the
S/G ratio, except 4-vinylguaiacol (which also arises from ferulates), and the analogous 4-
vinylsyringol.

230
 V. RESULTADOS Y DISCUSIÓN

10
100
(a)

75
Relative abundance (%)

4
50 carbohydrates

eg
n o
b
m p
25 7 13
24
df h 29
a c q 18 28
i 3 6 9 22 27
jk 1 12 30 32
2 5 1617 19 20 25 +
15 21 23 26 31 33 35

2 4 6 8 10 12 14 16 18 20 22
Retention time (min)

10
100
6
(b)

75 4
Relative abundance (%)

18
50
17
13
7
9 24
25
16 30 31
15 20
22 29
12 19 27 32
3 21 23 25 +
1 8 14 26
2 5 33 34 35

2 4 6 8 10 12 14 16 18 20 22
Retention time (min)

Figure 1. Py-GC/MS chromatograms of wheat straw (a) and of the isolated MWL (b). The
identities and relative abundances of the ligninderived phenolic compounds released are
listed in Table 2. Letters refer to carbohydrate compounds: (a) 2-methylfuran; (b)
hydroxyacetaldehyde; (c) 3-hydroxypropanal; (d) (3H)-furan-2-one; (e) propanal; (f) (2H)-
furan-3-one; (g) furfural; (h) 1-acetoxypyran-3-one; (i) 2-hydroxymethylfuran; (j)
cyclopent-1-ene-3,4-dione; (k) (5H)-furan-2-one; (m) 2,3-dihydro-5-methylfuran-2-one;
(n) 2-acetylfuran; (o) 4-hydroxy-5,6-dihydro-(2H)-pyran-2-one; (p) 2-hydroxy-3-methyl-2-
cyclopenten-1-one; (q) 3-hydroxy-2-methyl-(4H)-pyran-4-one.

231
V. RESULTADOS Y DISCUSIÓN

25
100 30
(a)

75
9
Relative abundance (%)

50 3 Fame 16
5
17
8 12 +
2 18
13
25
1 6 19 26
11
7 14 24 28
23
15
16 20 27
29

2 4 6 8 10 12 14 16 18 20 22
Retention time (min)

100 25

(b)

75
Relative abundance (%)

Fame 16
50 5 12

9 17 30
+ 26
18 24
2 13 19
25 11
14 1516
3 8 20 23
27 29
6 7
1 10

2 4 6 8 10 12 14 16 18 20 22
Retention time (min)

Figure 2. Py-TMAH-GC/MS chromatograms of wheat straw (a) and of the isolated MWL
(b). The identities and relative abundances of the released compounds are listed in Table 3.
Fame 16: hexadecanoate methyl ester.

232
 V. RESULTADOS Y DISCUSIÓN

Table 3. Identity and Relative Abundances of the Compounds Released after Py/TMAH of
Wheat Straw and the Isolated MWL.

MW Wheat Wheat
straw straw
cell-walls MWL
1 methoxybenzene 108 2.4 0.7
2 4-methoxytoluene 122 4.4 2.6
3 1,2-dimethoxybenzene 138 5.9 2.8
4 4-methoxystyrene 134 8.0 6.3
5 3,4-dimethoxytoluene 152 4.7 5.6
6 4-methoxybenzaldehyde 136 2.5 1.1
7 trans-4-methoxypropenylbenzene 148 1.0 1.0
8 1,2,3-trimethoxybenzene 168 3.6 2.4
9 3,4-dimethoxystyrene 164 8.6 7.5
10 4-methoxybenzoic acid methyl ester 166 0.7 0.7
11 3,4,5-trimethoxytoluene 182 1.8 2.6
12 3,4-dimethoxybenzaldehyde 166 5.2 11.2
13 1-(3,4-dimethoxyphenyl)-1-propene 178 3.5 1.9
14 3,4,5-trimethoxystyrene 194 1.2 2.0
15 3,4-dimethoxyacetophenone 180 1.4 2.9
16 1-(3,4-dimethoxyphenyl)-2-propanone 194 1.0 2.4
17 3,4-dimethoxybenzoic acid methyl ester 196 3.5 4.5
18 cis-3-(4-methoxyphenyl)-3-propenoic acid methyl ester 192 0.6 0.8
19 3,4,5-trimethoxybenzaldehyde 196 2.4 3.6
20 3,4-dimethoxybenzeneacetic acid methyl ester 210 0.7 1.6
21 cis-1-(3,4dimethoxyphenyl)-2-methoxyethylene 194 0.8 1.1
22 trans-1-(3,4dimethoxyphenyl)-2-methoxyethylene 194 0.7 0.7
23 1-(3.4.5-trimethoxyphenyl)-1-propene 208 1.0 0.8
24 3,4,5-trimethoxyacetophenone 210 1.6 5.4
25 trans-3-(4-methoxyphenyl)-3-propenoic acid methyl ester 192 15.1 18.8
26 3,4,5-trimethoxybenzoic acid methyl ester 226 2.3 3.6
27 cis-1-(3,4,5-trimethoxyphenyl)-2-methoxyethylene 224 1.1 0.5
28 cis-3-(3,4-dimethoxyphenyl)-3-propenoic acid methyl ester 222 0.9 0.2
29 trans-1-(3,4,5-trimethoxyphenyl)-2-methoxyethylene 224 0.5 0.5
30 trans-3-(3,4-dimethoxyphenyl)-3-propenoic acid methyl ester 222 13.2 4.3

S/G ratio 0.4 0.5

p-coumarate/ferulate ratioa 1.1 4.3
a
Relative abundance of p-coumarates (peaks 18 and 25) with respect to ferulates (peaks 28 and 30).

2D-NMR of Wheat Straw and Its Isolated MWL

In order to obtain additional information on the structure of the lignin, the whole
cell walls of wheat straw were analyzed in situ by gel state 2D-NMR, according to
the method previously described, 24,25 and the spectrum was compared with that of
the isolated MWL. The acetylated MWLwas also analyzed by HSQC (not shown)

233
V. RESULTADOS Y DISCUSIÓN

to confirm the assignments of the correlation peaks. The side- δC/δH

50−90/ 5−5 8 / δC/δH 90−155/6 0−8 0 g f
HSQC NMR spectra of the whole cell walls from wheat straw, and its isolated
MWL, are shown in Figure 3. Polysaccharide signals, dominated by hemicellulose
correlations as cellulose signals are hardly detectable in the gel state,24 were
predominant in the spectrum of the whole cell walls, including xylan correlations in
g δC/δH 60−85/ 5−5 5 f X2, X3, X4, and X5), which partially overlapped
with g g g f v y y xy X′2
X′3). Lignin signals could also be clearly observed in the HSQC spectrum of
the whole cell walls, despite its moderate lignin content (16.2% Klason lignin). On
the other hand, the spectrum of the MWL presented mostly lignin signals that, in
general terms, matched those observed in the HSQC spectrum of the whole cell
walls, but improved the detection of more minor lignin structures. The main lignin
correlation peaks assigned in the HSQC spectra are listed in Table 4, and the main
substructures found are depicted in Figure 4.

Interunit Linkage Characterization. The aliphatic-oxygenated region of the

spectra (Figure 3, top) gave information about the different interunit linkages
present in the lignin. In this region, correlation peaks from methoxyls and side
β-O-4′ A) were the most prominent in the HSQC spectra of
the whole cell walls and the isolated MWL. Other substructures were more clearly
visible in the HSQC spectrum of the MWL, including signals for phenylcoumarans
(B), resinols (C), dibenzodioxocins (D), and spirodienones (F). Minor amounts of
α β-diaryl ether substructures (E) could also be detected in the HSQC of the MWL,
as revealed by the Cα−Hα c δC/δH 79.5/5.50, although they were not
11
detected in previous works. I α β-diaryl ether linkages
are usually either undetectable or present at very low levels, although significant
amounts have been found to occur in tobacco lignin.49 p-Hydroxycinnamates,
yf v f y α
12
of the lignin side chain. H w v α-hydroxycinnamate ethers are shifted from
this position, indicating that only normal l g α- α-
etherification by hydroxycinnamates is insignificant.

234
 V. RESULTADOS Y DISCUSIÓN

Bβ δC Bβ δC
Cβ

Aγ
Iγ Aγ 60 Iγ 60
Fβ
A’γ
Bγ

Aα(G) Aα(G)
70 70
Aα(S) X2 Aα(S) X2
Cγ X3 Cγ X3
X’2
X’3 X4
X4
Eα
80 Fα A’β(G) 80
Aβ(G) Aoxβ Dα Aβ(H)
Aβ(G)
Bα Cα Fα’
Aβ(S) Bα Cα Dβ
Aβ(S)

5.5 5.0 4.5 4.0 3.5 3.0 δH 5.5 5.0 4.5 4.0 3.5 3.0 δH

T8 δC T8 δC
T6 T6
S2,6 100
T2’,6’ S2,6 100
T2’,6’

S’2,6 T3 S’2,6 T3 H3,5

G2 H3,5 G2
FA2 PCAβ/FAβ
110
FA2 PCAβ/FAβ
110

G5/G6
FA6
{ PCA3,5
120
G5/G6

FA6
{ PCA3,5
120

H2,6 H2,6 Jβ
PCA2,6 PCA2,6
Iα Iβ 130 Iα Iβ 130

140 140
PCAα PCAα

150 150
Jα Jα
7.5 7.0 6.5 δH 7.5 7.0 6.5 δH

Figure 3. Side chain (δC/δH 50−90/2.5−5.8) and aromatic/unsaturated (δC/δH 90−155/5.5−8.0) regions
in the 2D HSQC NMR spectra of wheat straw cell walls (left) and of the isolated MWL (right). See
Table 4 for signal assignments and Figure 4 for the main lignin structures identified. See Figure 5 for
the structure and assignments of the signals of the tricin moiety (T).

235
V. RESULTADOS Y DISCUSIÓN
R O

HO O HO γ
γ γ
5′
HO HO O γ
α β O 4′
α β O 4′ β O 4′ OMe
α β
HO
OMe OMe OMe O
α

OMe OMe OMe

O O O
O OMe

A A′ Aox B

OMe
O
5′ 5′′ HO γ
OMe
O α′
4′
γ 4′ 4′′ 4′ O
MeO O O OMe α β O
β′
β α β
γ′ OMe
α O γ OH

OMe
O
OMe O
OMe O

C D E

O OMe

OH H O
γ
γ O O
β β γ
α α α
O γ′ β
HO β α′ α
1′
OH
γ β′

OAr
OMe OMe
OMe
O O
O OH

F I J PCA

O O
γ
β OH OH OH R O
α α α α α

OMe OMe MeO OMe MeO OMe

O O O O O

FA H G S S′
Figure 4. Main structures present in the lignins of wheat straw: (A) β-O-4′ alkyl-aryl ethers;
(A′) β-O-4′ alkyl-aryl ethers with acylated γ-OH; (Aox) Cα oxidized β-O-4′ structures; (B)
phenylcoumarans; (C) resinols; (D) dibenzodioxocins; (E) α,β-diaryl ethers; (F) spirodieno-
nes; (I) cinnamyl alcohol end-groups; (J) cinnamyl aldehyde end-groups; (PCA)
p-coumarates; (FA) ferulates; (H) p-hydroxyphenyl units; (G) guaiacyl units; (S) syringyl
units.

236
 V. RESULTADOS Y DISCUSIÓN

Table 4 g f g 13 −1H k HSQ S f

Wheat Straw and the Isolated MWLa

Label δC/δH (ppm) Assignment

Bβ 53.1/3.43 Cβ–Hβ in phenylcoumaran substructures (B)
Cβ 53.5/3.05 Cβ–Hβ in β–β´ resinol substructures (C)
-OCH3 55.6/3.73 C–H in methoxyls
Aγ 59.4 /3.40 and 3.72 Cγ–Hγ in γ-hydroxylated β–O–4´ substructures (A)
Fβ 59.5/2.75 Cβ–Hβ in spirodienone substructures (F)
Iγ 61.3/4.08 Cγ–Hγ in cinnamyl alcohol end-groups (I)
Bγ 62.6/3.67 Cγ–Hγ in phenylcoumaran substructures (B)
A'γ 63.5/3.83 and 4.30 Cγ–Hγ in γ-acylated β–O–4´ substructures (A')
Aα 70.9/4.71 Cα–Hα in β–O–4´ substructures (A) linked to a G-unit
Cγ 71.0/3.81 and 4.17 Cγ–Hγ in β–β´ resinol substructures (C)
Aα S 71.8/4.83 Cα–Hα in β–O–4´ substructures (A) linked to a S-unit
Eα 79.5/5.50 Cα–Hα α–O–4´ substructures (E)
A'β 80.8/4.52 Cβ–Hβ in γ-acylated β–O–4´ substructures linked to a
G-unit (A')
Fα 81.2/5.01 Cα–Hα in spirodienone substructures (F)
A xβ 82.7/5.12 Cβ–Hβ in α-oxidized β–O–4´ substructures (Aox)
Aβ H 82.9/4.48 Cβ–Hβ in β–O–4´ substructures (A) linked to a H-unit
Dα 83.3/4.81 Cα–Hα in dibenzodioxocin substructures (D)
Aβ 83.4/4.27 Cβ–Hβ in β–O–4´ substructures (A) linked to a G unit
Cα 84.8/4.65 Cα–Hα in β–β´ resinol substructures (C)
Fα´ 84.6/4.75 Cα´–Hα´ in spirodienone substructures (F)
Dβ 85.3/3.85 Cβ–Hβ in dibenzodioxocin substructures (D)
Aβ S 85.9/4.10 Cβ–Hβ in β–O–4´ substructures linked (A) to a S unit
Bα 86.8/5.43 Cα–Hα in phenylcoumaran substructures (B)
S2,6 103.8/6.69 C2–H2 and C6–H6 in etherified syringyl units (S)
G2 110.9/6.99 C2–H2 in guaiacyl units (G)
Fer2 111.0/7.32 C2–H2 in ferulate (FA)
PCAβ and FAβ 113.5/6.27 Cβ–Hβ in p-coumarate (PCA) and ferulate (FA)
G5/G6 114.9/6.72 and 6.94, 118.7/6.77 C5–H5 and C6–H6 in guaiacyl units (G)
PCA3,5 115.5/6.77 C3–H3 and C5–H5 in p-coumarate (PCA)
FA6 123.2/7.15 C6–H6 in ferulate (FA)
Jβ 126.3/6.76 Cβ–Hβ in cinnamyl aldehyde end-groups (J)
H2,6 127.8/7.22 C2,6–H2,6 in p-hydroxyphenyl units (H)
Iβ 128.4/6.23 Cβ–Hβ in cinnamyl alcohol end-groups (I)
Iα 128.4/6.44 Cα–Hα in cinnamyl alcohol end-groups (I)
PCA2,6 130.1/7.45 C2–H2 and C6–H6 in p-coumarate (PCA)
PCAα and FAα 144.7/7.41 Cα–Hα in p-coumarate (PCA) and ferulate (FA)
Jα 153.4/7.61 Cα–Hα in cinnamyl aldehyde end-groups (J)
a 6−35
Signals were assigned by comparison with the literature.

237
V. RESULTADOS Y DISCUSIÓN

Lignin Acylation. The HSQC spectrum of the isolated wheat straw MWL also
readily reveals the presence of characteristic signals corresponding to the C γ−Hγ
f γ- y β-O-4′ A′ g f δC/δH
63.5/3.83 and ∼4.30. Signals for α- y β-O-4′ w
appear at ∼6.1/75 ppm, were not observed in the spectrum. Therefore, it is possible
to conclude that the lignin of wheat straw is partially acylated and that this
y x v y γ-position of the lignin side chain, as already
observed in this and other grasses.11,21,26 In addition, a signal for the Cβ−Hβ
f γ- y β-O-4′ k - ′β ) is clearly
52,21
v δC/δH 80.8/4. indicating an important acylation extent of G-lignin
units in this lignin, as will be discussed below. An estimate of ∼10% for the
g f γ-acylation of lignin side chains was calculated by integration of the
Cγ-Hγ correlation peaks corresponding to the hydroxylated (Aγ y ′γ)
substructures in the HSQC spectrum of the isolated MWL (Table 5).

Lignin Aromatic/Unsaturated Components. The main correlation peaks in the

aromatic/unsaturated region of the HSQC spectra (Figure 3, bottom) corresponded
to the aromatic rings and unsaturated side chains of the different lignin units and
y xy “ w” f w Sg f
p-hydroxyphenyl (H), guaiacyl (G), and syringyl (S) units were observed almost
equivalently in the spectra of the whole cell walls and in the isolated MWL, due to
g ’ g y v f yf f
polysaccharide correlations that can overwhelm other regions of the spectrum). In
addition, as is typical in spectra from grasses, prominent signals corresponding to
p-coumarate (PCA) and ferulate (FA) structures were also observed. HSQC
analysis of the acetylated MWL (not shown) indicated that the p-coumarate phenol
is free, not etherified. Therefore, p-coumarates in wheat straw lignin are solely
ester linked, as already advanced by other authors,13 and as found in other
grasses.21,26 Other signals in this region of the spectrum are from the unsaturated
side chains of cinnamyl alcohol end-groups (I) and cinnamaldehyde end-groups
(J). The total relative content of the cinnamaldehyde end-groups was estimated by
comparison of the intensities of the Cβ−Hβ correlations in cinnamyl alcohols (I) and
aldehydes (J).

Identification of a New Component in Wheat Lignin. We also report here on the

successful structural elucidation of an initially puzzling component giving unusual
correlation peaks in the aromatic regions of HSQC spectra, and provide the first
evidence that a flavone is linked to lignin in wheat and, apparently, other grasses.
Two strong and well-resolved signals from unknown structures, not previously

238
 V. RESULTADOS Y DISCUSIÓN

g w y v δC/δH 94.1/6.56 and 98.8/6.20 in the

HSQC spectra, Figure 3 (bottom). Their appearance in the spectrum of the whole
cell walls indicated that they did not arise from impurities or artifacts formed
during the lignin isolation process, and their retention in the MWL suggested that
they might belong to structures bound into the lignin network. Interestingly, these
signals can also be observed in the published HSQC spectra of another lignin
preparation from wheat straw,14 although they were not assigned in that paper.
Likewise, we have also noticed the occurrence of these two signals in the lignins
from other grasses, such as elephant grass, and they have remained a mystery.
Their prevalence in these wheat spectra made it imperative that we attempt to
identify the component. Further valuable information about the nature of this
structure was obtained by performing long-range 13 −1H correlation (HMBC)
experiments (Figure 5). The correlation peaks between protons and carbons
y −3 v H y
moiety has a flavone-type structure. Flavones are a class of flavonoids that have the
2-phenylchromen-4-one backbone. There is a high diversity of flavones that arise
from the different phenolic hydroxyl group substitutions and include compounds
such as luteolin or apigenin. From the proton and carbon (including quaternary
carbon, not revealed in HSQC spectra) HMBC data, it is possible to conclude that
the structure of the flavone moiety present in the MWL of wheat straw is tricin
5 7 4′-trihydroxy-3′ 5′-dimethoxyflavone, Figure 5); the 1H and 13C shifts match
those published for tricin.50,51

g g HSQ δC/δH 94.1/ 6.56 and 98.8/6.20

(Figure 3) thus correspond to the C8−H8 and C6−H6 correlations, respectively. The
HSQC also shows the C3−H3 δC/δH 104.5/7.04, near the S2,6 signal.
On the other hand, the phenyl moiety linked at C-2 is of syringyltype, the
correlations for C ′-H2′ and C6′-H6′ being also observed in the HSQC spectrum at
δC/δH 103.9/7.31. Tricin has two phenolic hydroxyls at C-5 and C-7 of the
chroman-4-one skeleton, with diagnostic phenolic proton chemical shifts that are
readily apparent in the HMBC; DMSO is a solvent that limits proton transfer,34 so
the H-bonded C5-OH proton signal at 12.86 ppm and the C7-OH proton signal at
10.88 ppm are sharp in this solvent and therefore produce good correlations in
longrange 13 −1H correlation spectra. In addition, the absence of the signals for the
4′-OH of tricin in the HMBC proton indicates that it is not free. Therefore,
incorporation of tricin into the lignin network through 4-O-β k g
occurs with the flavonolignans (see below), is indicated. In fact, a signal for the
f 4′ 139 5 β-position of
a G-unit at 4.28 ppm was clearly observed in the HMBC spectrum (Figure 6),

239
V. RESULTADOS Y DISCUSIÓN

providing evidence for this incorporation. Tricin is widely distributed in grasses,

including wheat, rice, barley, sorghum, and maize,52 and can occur in either free or
conjugated form. Tricin can form flavonolignan derivatives with a tricin skeleton
linked to a phenylpropanoid (p-hydroxyphenyl or guaiacyl) moiety through a β-O-4
bond, such as tricin 4´-O-β-guaiacylglycerol ether, among others.50,51 Flavones (as
all other flavanoids/flavonoids) are metabolic hybrids as they are derived from a
combination of the shikimate-derived phenylpropanoid and the acetate/malonate-
derived polyketide pathways. Since polyphenols are formed in lignified regions by
oxidative coupling, incorporation into the lignin structure is a possibility. In fact,
other related benzene diols and triols, such as the flavanols epicatechin,
epigallocatechin, or epigallocatechin gallate, although they are not known in actual
plant cell walls, have been shown to produce lignin copolymers with normal
monolignols.53

OH(5) OH(7) H2’,6’ H3 H8 H6

C8
δC
C6
C10
100

C2’,6’

120
C1’

C4’ 140

C3’,5’

C9
C5 160
C2
C7

180
C4

13 12 11 10 9 8 7
δH
OMe
3’ O
2’
4’
HO 8 O 1’
7 9 5’
2 6’ OMe

6 3
10 4
5
OH O

Figure 5. Partial HMBC spectrum (δC/δH 90−185/6.0−13.0) of wheat straw MWL showing
the main correlations and the structure of tricin (5,7,4′-trihydroxy-3′,5′-dimethoxyflavone)
units in the lignin.

240
 V. RESULTADOS Y DISCUSIÓN

δC(ppm)
Aoxβ Aβ(H) Aβ(G) 82
84
Aβ(S)
Bα Cα 86
88
7 6 5 4

C4’ 139
140
7 6 5 4 δH(ppm)

O
MeO

OMe
H
3’ O
2’ α
4’ β OH
8 1’ γ
HO 7 9 O
5’ OMe OH
2 6’
6 3
10 4
5
OH O

Figure 6. Section of the HMBC spectrum of wheat straw MWL showing the correlation
for the tricin C4´ carbon (at 139.5 ppm) and the proton at the β-position of a G-unit at 4.28
ppm (bottom). The section of the HSQC spectrum for the Cβ-Hβ correlations of the β–O–4´
alkyl-aryl ethers is also shown (top). The structure illustrates the likely incorporation of
tricin into the lignin polymer through a 4´–O–β ether linkage with a guaiacyl unit.

The presence (or otherwise) in lignins of components from other pathways is of

significant interest. It has never been demonstrated, for example, that lignans,
dimers, and higher oligomers that also arise from radical coupling of monolignols,
become incorporated into lignins.9,54−56 Lignans are produced under proteinaceous
control such that they are always (at least partially) optically active.57 Lignins are
completely racemic;55 no components excised from lignins have ever been shown
to be optically active, so the lignin polymer and the lignan “extractives” are
assumed to be independently produced in time and space. For this reason, this
observation along with the evidence presented here that a flavone, tricin, is
integrally incorporated into lignin, if validated completely, is a new phenomenon
with rather profound implications. It implies that the monomer is exported to the
cell wall where it undergoes radical coupling reactions with monolignols, or at least
with the primary monolignol coniferyl alcohol, and becomes part of the lignin

241
V. RESULTADOS Y DISCUSIÓN

polymer. The occurrence of tricin in the MWL of wheat straw seems to be evidence
for this incorporation. At the very least, then, this observation will require further
recognition of the malleability of lignification and perhaps another addition to the
f “ g ” 58

Quantitation. The relative abundances of the main lignin interunit linkages and
end-g w g f γ-acylation, the molar abundances of the
different lignin units (H, G, and S), p-coumarates, and ferulates, and the molar S/G
ratios of the lignin in wheat straw, estimated from volume integration of contours
in the HSQC spectra,24,34,35 are shown in Table 5.

Table 5. Structural Characteristics (Lignin Interunit Linkages, End- γ-Acylation,

Aromatic Units and S/G Ratio, Cinnamate Content, and p-Coumarate/Ferulate Ratio) from
Integration of 13 −1H Correlation Peaks in the HSQC Spectra of the Wheat Straw and the
Isolated MWL.
Wheat Straw cell-walls Wheat Straw MWL
Lignin inter-unit linkages (%)
β–O–4´ aryl ethers (A/A') - 75
α-oxidized β–O–4´ aryl ethers (Aox) - 2
Phenylcoumarans (B) - 11
Resinols (C) - 4
Dibenzodioxocins (D) - 3
α,β-diaryl ethers (E) - 2
Spirodienones (F) - 3
-----
Total 100
Lignin end-groupsa
Cinnamyl alcohol end-groups (I) - 4
Cinnamaldehyde end-groups (J) - 4
Lignin side-chain γ-acylation (%) - 10
Lignin aromatic unitsb
H (%) 6 6
G (%) 64 64
S (%) 30 30
S/G ratio 0.5 0.5
p-Hydroxycinnamatesc
p-Coumarates (%) 4 4
Ferulates (%) 11 2
p-Coumarates/ferulates ratio 0.4 2.0
a
Expressed as a fraction of the total lignin interunit linkage types A−F.
b
Molar percentages (H + G + S = 100).
c
p-Coumarate and ferulate molar contents as percentages of lignin content (H + G + S).

242
 V. RESULTADOS Y DISCUSIÓN

Similarly as observed by Py-GC/MS, the abundance of ferulate is lower in the

isolated MWL than in the corresponding whole cell wall, confirming that ferulate is
mostly attached to polysaccharides, whereas p-coumarate is predominantly
attached to lignin. In addition, the S/G ratio estimated from the HSQC (S/G 0.5) is
similar to that estimated by analytical pyrolysis. The relative abundances of p-
coumarate and H lignin units are probably more realistic than those provided by
Py-GC/MS and Py-TMAH analyses, as Py-GC/MS does not distinguish between
both types of structures, as already discussed, and Py-TMAH overestimates the
esters, including p-coumarate residues, because of the production of a single vinyl-
phenol from each with high pyrolytic efficiency. However, the end-groups,
estimated as ∼10% of the total lignin linkages (with similar amounts of cinnamyl
alcohols and aldehydes), are overestimated in HSQC spectra because of their
longer relaxation than the bulk polymer (see Materials and Methods). The data
f g y f β-O-4′ k g s
(accounting for 75% of all the interunit linkages), followed by phenylcoumarans
11% w f x α β-diaryl ethers, and
spirodienones. These results sharply contrast with the data reported by Banoub et
al.15 that indicated that wheat straw lignin was made up almost exclusively of
repeating phenylcoumaran units, an occurrence that would be novel indeed.

Resolving the Lignin Acylation Anomaly. The fact that the side chain of the
lignin in wheat straw is partially acylat γ-OH, together with the presence of
significant amounts of p-coumarate moieties (4% with respect to lignin), which are
k w y γ-OH of the lignin side chain in many plants, and particularly
10,21,26,31,59
in grasses, led other authors to conclude that p-coumarates also acylate
γ-OH in the lignin of wheat straw,11 although no direct evidence of this linkage
was provided. The HSQC spectrum only indicates that the lignin in wheat straw is
y y γ-position, but cannot provide information on the nature of
the acyl group. Additional analyses are therefore needed to confirm whether p-
g y γ-OH of the lignin side chain. For this
purpose, we performed HMBC experiments, again correlating protons with carbons
separated by 2 or 3 bonds, that can give important information about the
connectivity of the ester moiety to the lignin skeleton.26 Figure 7 shows the section
of the HMBC spectrum of wheat straw MWL for the correlations of the carbonyl
f ff g y g g γ-OH. Two distinct carbonyl
carbons were observed in this region of the HMBC spectrum, at 166.0 ppm for p-
coumarates and 169.8 ppm for acetates, Figure 7B. The correlations of the
carbonyl carbon at 166 0 w α- β-protons of p-coumarate esters (at
7.41 and 6.27 ppm) confirm that they belong to the p-coumarate esters. The

243
V. RESULTADOS Y DISCUSIÓN

correlations of this carbonyl carbon with several protons in the range 4.0−4.8 ppm
conclusively demonstrate that p-coumarate is acylating the γ-position of the lignin
side chains, as also occurs in other grasses,10,21,26,31,59 although the intensity of these
signals is low as the p-coumarate level is only 4% here (and HMBC spectra are not
quantitative). In addition, the HBMC spectrum also showed the correlations of the
carbonyl carbon at 169.8 for acetate groups attached to the lignin network. As
expected, there was a strong correlation with the acetate methyl group at 1.88 ppm
(not shown). The correlations of this carbonyl carbon with the protons in the range
3.6−4.4 ppm indicate that acetates are also acylating the γ-OH of the lignin side
chain. As with p-coumarates, several correlations were also observed in this region
suggesting, beyond the fact that there are two usually distinct γ-proton shifts, the
involvement of both acetates and p-coumarates at the γ-positions of different lignin
substructures. Acetates have also been previously found acylating the γ-OH in the
lignins of many plants, including grasses.21,31,39,40,60,61 Further details regarding the
lignin acylation in wheat, via p-coumarate and acetate, is provided by additional
experiments below.

PCAα PCAβ γ-OH acylated

A) 60

62

64

66

111
112
113
114
115
116

143
144
145
146

7.0 6.0 5.0 4.0

B) 165
166
167
168
169
170

7.0 6.0 5.0 4.0

Figure 7. Section of the HMBC spectrum (δC/δH 164−171/3.5−7.8) of wheat straw MWL
showing the main correlations for the carbonyl carbons of the groups (p-coumarates and
acetates) acylating the γ-position of the lignin side chain (B). Appropriate sections of the
HSQC spectrum showing the Cγ−Hγ correlations of the acylated lignin γ-carbon (δC 60−
66) and the Cα−Hα and Cβ−Hβ correlations of pcoumarates (δC 111−116 and 142−147,
respectively), are also depicted (A).

244
 V. RESULTADOS Y DISCUSIÓN

Derivatization Followed by Reductive Cleavage (DFRC and DFRC′) of

Wheat Straw MWL.

Additional information regarding the acylation of the γ-OH of the lignin side
chain can be obtained from DFRC, a degradation method that cleaves α- and β-
ether linkages in the lignin polymer leaving γ-esters intact.36−38 The chromatogram
of the DFRC degradation products of the MWL isolated from the wheat straw is
shown in Figure 8 (top). The lignin released the cis- and trans-isomers of p-
hydroxyphenyl (cH and tH), guaiacyl (cG and tG), and syringyl (cS and tS) lignin
monomers (as their acetylated derivatives) arising from normal, non-γ-p-
coumaroylated β-ethers in lignin. Significant amounts of the monolignol p-
coumarate conjugates were anticipated to be released, if p-coumarates acylate the
γ-OH in the lignin, upon DFRC. However, and unexpectedly, only traces of γ-p-
coumaroylated syringyl (cSpc and tSpc) monomers could be detected in the
chromatogram (∼1% of S units, as shown in Table 6), despite the decent amounts
of p-coumarates (4% with respect to the total aromatic units) present in this lignin,
and the fact that p-coumarates are acylating the γ-carbon, as observed by 2D-NMR.

Table 6. Abundance (Molar Yields) of the DFRC and DFRC′ Degradation Monomers of
the MWL Isolated from Wheat Straw and Relative Abundances of the Different Acylated
Lignin Monomers.
Monomers (μmol/g lignin)

H G Gac Gpc S Sac Spc %Gaca %Gpcb %Sacc %Spcd S/G

MWL
67 386 52 0 196 2 2 12 0 1 1 0.5
wheat straw
a
%Gac is the percentage of acetylated G units (Gac) with respect to the total G units (G, Gac, Gpc).
b
%Gpc is the percentage of p-coumaroylated G units (Gpc) with respect to the total G units (G, Gac, Gpc).
c
%Sac is the percentage of acetylated S units (Sac) with respect to the total S units (S, Sac, Spc).
d
%Spc is the percentage of p-coumaroylated S units (Spc) with respect to the total S units (S, Sac, Spc).

A chromatogram of the DFRC degradation products of the MWL isolated from

elephant grass (Figure 8, bottom), that releases significant amounts of sinapyl p-
coumarate, is shown for comparison to emphasize the stark difference with wheat
straw. We have to stress here that both the DFRC degradation and the GC−MS
analysis were repeated several times with increasing amounts of sample for the
DFRC degradation, and increasing product concentrations for the GC−MS

245
V. RESULTADOS Y DISCUSIÓN

analyses, and we always failed to detect higher amounts of the conjugate Spc . No
traces of γ-p-coumaroylated guaiacyl (cGpc and tGpc) lignin units could be found,
despite this lignin’s being enriched in G-units. This finding clearly indicates that, in
wheat straw, p-coumarate groups are attached to the lignin γ-OH in β-ether
structures only to a very low extent. As NMR indicated that some p-coumarates are
indeed attached to the γ-OH, the only conclusion is that they are largely in other
(i.e., non-β-ether) lignin substructures.

tG
100
(a)
Relative abundances (%)

75

50

tS
25 tSpc
tH IS
cH cG
cS
15 20 25 30 35 40 45
Retention time (min)

100 IS
(b) tS
Relative abundances (%)

tG
75

50
tSpc

25
cGpc tGpccSpc
tH cG cS

15 20 25 30 35 40 45
Retention time (min)

Figure 8. Chromatograms (GC-TIC) of the DFRC degradation products from the MWL
isolated from wheat straw (a), and the MWL isolated from elephant grass cortex (b), that is
shown here only for comparison. cG, tG, cS, and tS are the normal cis- and trans-coniferyl
and -sinapyl alcohol (guaiacyl and syringyl) monomers (as their acetate derivatives). Note
the absence of cis- and trans-coniferyl and -sinapyl p-coumarates (cGpc and tGpc, and cSpc
and tSpc) in the MWL from wheat straw, and which are present in elephant grass.

Acetate groups, on the other hand, also widely occur acylating the γ-OH in the
lignin of many plants, including grasses, palms, and (at trace levels) most

246
 V. RESULTADOS Y DISCUSIÓN

hardwoods,31,39,40,60 and their occurrence in the lignin of wheat straw was also
clearly observed in the HMBC spectrum of the MWL, as noted above. The original
DFRC degradation method, however, does not allow the analysis of natively
acetylated lignin because the degradation products are acetylated during the
procedure, but with appropriate modification of the protocol by substituting
acetylating reagents with propionylating ones (in the so-called DFRC′ method) it is
possible to obtain information about the occurrence of native acetates in
lignin.31,39,40

The chromatogram of the DFRC′ degradation products released from the wheat
straw MWL is shown in Figure 9. The lignin released the cis- and trans-isomers of
p-hydroxyphenyl (cH and tH), guaiacyl (cG and tG), and syringyl (cS and tS)
lignin monomers (as their propionylated derivatives) arising from normal,
nonacetylated γ-units in lignin. Interestingly, the presence of γ-acetylated guaiacyl
(cGac and tGac ) and syringyl (tSac ) lignin units could also be clearly observed in the
chromatogram, and confirms the occurrence of native acetylation at the γ-carbon of
the lignin side chain of wheat straw. The results from the DFRC and DFRC′
analyses of the MWL selected for this study, namely, the molar yields of the
released monomers, the percentages of naturally acetylated guaiacyl (% Gac ) and
syringyl (% Sac ) lignin moieties, and the S/G ratio, are presented in Table 6.

tG
100
Relative abundance (%)

75

50

tH tGac
25
cG tS
cH cGac cS tSac

20 25 30
Retention time (min)

Figure 9. Chromatogram (GC-TIC) of the DFRC′ degradation products from the MWL
isolated from wheat straw. cG, tG, cS, and tS are the normal cis- and trans-coniferyl and
sinapyl alcohol (guaiacyl and syringyl) monomers (as their dipropionylated derivatives).
cGac, tGac, cSac, and tSac are the natively γ-acetylated cis- and trans-coniferyl and -sinapyl
alcohol (guaiacyl and syringyl) monomers (as their phenol-propionylated derivatives).

247
V. RESULTADOS Y DISCUSIÓN

The analyses indicated that up to 12% of the releasable G-lignin units are
acetylated, while only 1% of the total S-lignin units are acetylated. It is interesting
to note that in the lignin of most plants in which acetylated lignins have been
k f γ-acetates have always been
31,40,60
preferentially attached to S-lignin units; w v f γ-acetylation
of G-lignin units has been observed in other grasses, such as bamboo and elephant
grass.21,40 Previous papers describing the structure of the lignin in wheat straw
fa v f g y g g γ-OH; the
acylation was previously attributed exclusively to p-coumarates. It is now clear
g fw w x f γ-acylation (∼10% as
observed HSQ y β-ether units (that
f y ′ f g γ-acylation
involves coupling and cross-coupling reactions of previously acylated
monolignols,61 it is evident that, in the lignin of wheat straw, the most important
monolignol conjugate would be coniferyl acetate.

With respect to p- f y γ-carbon on the lignin

in wheat straw, as suggested by 2D-NMR, the question is why are they not better
β-ether units in lignins? To these authors, it seems that the only
reasonable hypothesis is that wheat differs from corn and other grasses in two
respects. First, the acylation is largely on guaiacyl units and therefore derives from
acylated coniferyl alcohol conjugates. We already see this evidence for the
acetylated monolignols, where coniferyl acetate is significantly favored over
y v y ′ Table 6). Second, one has to
contend, and it is a possible consequence of the coniferyl alcohol (vs sinapyl
alcohol) acylation, that these conjugates are present early during lignification. It
has been nicely established, by autoradiographic methods,62 that sinapyl alcohol
incorporation into cell walls occurs later during wall development. It is therefore
reasonable that sinapyl p-coumarate, the major monolignol p-coumarate conjugate
in corn, for example, would also enter the wall later during development. There
have been no studies on the temporal aspects of monolignol conjugates in
lignification, but if coniferyl p-coumarate was sent to the wall early during
lignification, it would be more heavily involved in monolignol (conjugate)
dimerization events (rather than chain extension), events that produce only low
v f β-ether coupling. Thus, the p-coumarate would be involved preferentially
β−β- β−5-coupled products, and also in cinnamyl alcohol end-groups, which
w w γ-C/H correlations in HSQC spectra
remain typical f f y γ-OHs. This kind of coupling would also be
more favored in the so-called bulk vs endwise coupling mode, and, as noted above,

248
 V. RESULTADOS Y DISCUSIÓN

there is evidence in the spectra of lignins from wheat here that such bulk coupling
is occurring. As noted a v α β-diaryl ethers (E) found in the wheat lignin
(Figure 3, top right) are common in synthetic lignins but are rarely seen in natural
lignins where the conditions of slow diffusion of monomers (and radicals) are more
conducive toward endwise coupling. The currently best hypothesis, then, is that
coniferyl p-coumarate (rather than sinapyl p-coumarate) is the major p-coumarate
conjugate destined for wheat lignification (just as coniferyl acetate is) and that its
export to the wall is early during development such that condensed structures rather
β-ethers (that can be quantified via ether-cleaving reactions such as those in
the DFRC method) are acylated by p-coumarate. Establishing the validity of this
hypothesis is therefore nontrivial and multifaceted, but will hopefully be the subject
of further investigations along with a more careful evaluation, as outlined here, of
the exact nature and distribution of lignin acylation in a variety of plant materials.

In conclusion, the lignin from wheat straw has been characterized by different
analytical methods that indicated that it is an H:G:S lignin, with a strong
predominance of G-lignin units (S/G 0.5), and with some amounts of associated p-
coumarates and ferulates. Our data indicated that in wheat straw ferulates are
mostly attached to carbohydrates (although radical coupling into lignins is complex
and difficult to detect), while p-coumarates are predominantly attached to the
lignin. 2D-NMR g k g β-O-4′ lkyl-
aryl ethers, followed by phenylcoumarans and minor amounts of resinols,
x α β-diaryl ethers, together with cinnamyl
alcohol and cinnamaldehyde end-groups. 2D-NMR also indicated that the lignin of
wheat straw is partially acylated (∼10% of all side chains), and exclusively at the
γ-carbon of the side chain, with acetates and p-coumarates. DFRC analyses
f y y γ-OH in guaiacyl (12%) rather than
in syringyl units (1%), as has also been found to occur in other grasses and in
contrast to what occurs in dicots. On the other hand, and despite p- ’
having been found y g γ-OH, they were barely detectable as the
g j g f v y v g β-ethers in lignin in the DFRC
method, which seems to indicate that p-coumarates must be preferentially involved
in structures other β-ethers. Finally, we present the first evidence that the
flavone tricin was found in wheat lignin, etherified by a G-type unit. If it is
ultimately shown to have incorporated, in the cell wall, into the lignin by the
radical coupling reactions that typify lignification (as it appears), the definition of
lignin, and what constitutes a lignin monomer, will need further refinement.

249
V. RESULTADOS Y DISCUSIÓN

FUNDING

y f y S j 011- 5379 SI
j 01040 075 - j I O O K - 4436 g
k SI f - O f g “
” f y S S k
the Spanish Ministry of Science and Innovation for an FPI fellowship. J.R. was
funded in part by the DOE Great Lakes Bioenergy Research Center (DOE Office of
Science BER DE-FC02-807ER64494).

ACKNOWLEDGMENTS

We thank Manuel Angulo (CITIUS, University of Seville) for providing technical

assistance in the NMR analyses and Dr. Yuki Tobimatsu (Univ. Wisconsin,
Madison) for performing the GPC analyses.

REFERENCES

(1) Simmons, B. A.; Loque, D.; Ralph, J. Advances in modifying lignin for
enhanced biofuels production. Curr. Opin. Plant Biol. 2010, 13 313−3 0

(2) Ragauskas, A. J.; Williams, C. K.; Davison, B. H.; Britovsek, G.; Cairney,
J.; Eckert, C. A.; Frederick, W. J.; Hallett, J. P.; Leak, D. J.; Liotta, C. L.; Mielenz,
J. R.; Murphy, R.; Templer, R.; Tschaplinski., T. The path forward for biofuels and
biomaterials. Science 2006, 311 484−489

(3) Somerville, C.; Youngs, H.; Taylor, C.; Davis, S. C.; Long, S. P. Feedstocks
for lignocellulosic biofuels. Science 2010, 329 790−79

(4) Sarkar, N.; Ghosh, S. K.; Bannerjee, S.; Aikat, K. Bioethanol production
from agricultural wastes: An overview. Renewable Energy 2012, 37 19− 7

(5) Kim, S.; Dale, B. E. Global potential bioethanol production from wasted
crops and crop residues. Biomass Bioenergy 2004, 4 361−375

(6) Abramson, M.; Shoseyov, O.; Shani, Z. Plant cell wall reconstruction toward
improved lignocellulosic production and processability. Plant Sci. 2010, 178,
61−7

(7) Hendriks, A. T. W. M.; Zeeman, G. Pretreatments to enhance the

digestibility of lignocellulosic biomass. Bioresour. Technol. 2009, 100 10−18

250
 V. RESULTADOS Y DISCUSIÓN

(8) Agbor, V. B.; Cicek, N.; Sparling, R.; Berlin, A.; Levin, D. B. Biomass
pretreatment: Fundamentals toward application. Biotechnol. Adv. 2011, 29,
675−685

(9) Ralph, J.; Lundquist, K.; Brunow, G.; Lu, F.; Kim, H.; Schatz, P. F.; Marita,
J. M.; Hatfield, R. D.; Ralph, S. A.; Christensen, J. H.; Boerjan, W. Lignins:
Natural polymers from oxidative coupling of 4-hydroxyphenylpropanoids.
Phytochem. Rev. 2004, 3 9−60

(10) Ralph, J. Hydroxycinnamates in lignification. Phytochem. Rev. 2010, 9,

65−83

(11) Crestini, C.; Argyropoulos, D. S. Structural analysis of wheat straw lignin

by quantitative 31P and 2D NMR spectroscopy. The occurrence of ester bonds and
α-O-4 substructures. J. Agric. Food Chem. 1997, 45 1 1 −1 19

(12) Sun, R.; Lawther, J. M.; Banks, W. B. A tentative chemical structure of

wheat straw lignin. Ind. Crops Prod. 1997, 6 1−8

(13) Sun, R. -C.; Sun, X. F.; Wang, S. Q.; Zhu, W.; Wang, X. Y. Ester and ether
linkages between hydroxycinnamic acids and lignins from wheat, rice, rye, and
barley straws, maize stems, and fast-growing poplar wood. Ind. Crops Prod. 2002,
15 179−188

(14) Sun, X.-F.; Sun, R. -C.; Fowler, P.; Baird, M. S. Extraction and
characterization of original lignin and hemicelluloses from wheat straw. J. Agric.
Food Chem. 2005, 53 860−870

(15) Banoub, J. H.; Benjelloun-Mlayah, B.; Ziarelli, F.; Joly, N.; Delmas, M.
Elucidation of the complex molecular structure of wheat straw lignin polymer by
atmospheric pressure photoionization quadrupole time-of-flight tandem mass
spectrometry. Rapid Commun. Mass Spectrom. 2007, 21 867− 888

(16) Zhang, J.; Deng, H.; Lin, L.; Sun, Y.; Pan, C.; Liu, S. Isolation and
characterization of wheat straw lignin from a formic acid process. Bioresour.
Technol. 2010, 101 311− 316

(17) Yang, Q.; Wu, S.; Lou, R.; Gaojin, L. V. Structural characterization of
lignin from wheat straw. Wood Sci. Technol. 2011, 45 419−431

251
V. RESULTADOS Y DISCUSIÓN

18 j k Studies on finely divided wood. Part I. Extraction of lignin

with neutral solvents. Sven. Papperstidn. 1956, 59 477−485

(19) Tappi Test Methods 2004−2005; Tappi Press: Norcoss, GA 30092, USA,
2004.

(20) Browning, B. L. Methods of Wood Chemistry; Wiley-Interscience

Publishers: New York, 1967; Vol. II.

1 ; ; ; ; - ;
; ; S f g
the cortex and pith of elephant grass (Pennisetum purpureum) stems. J. Agric.
Food Chem. 2012, 60 3619−3634

(22) Faix, O.; Meier, D.; Fortmann, I. Thermal degradation products of wood. A
collection of electron-impact (EI) mass spectra of monomeric lignin derived
products. Holz Roh- Werkstoff 1990, 48 351−354

(23) Ralph, J.; Hatfield, R. D. Pyrolysis-GC/MS characterization of forage

materials. J. Agric. Food Chem. 1991, 39 14 6−1437

(24) Kim, H.; Ralph, J.; Akiyama, T. Solution-state 2D NMR of ball-milled

plant cell-wall gels in DMSO-d6. Bioenergy Res. 2008, 1 56−66

5 ; ; ; ;S I; -
Barbero, J.; Martínez, A. T.; del Río, J. C. HSQC-NMR analysis of lignin in woody
(Eucalyptus globulus and Picea abies) and non-woody (Agave sisalana) ball-milled
plant materials at the gel state. Holzforschung 2009, 63 691−698

(26) Ralph, J.; Hatfield, R. D.; Quideau, S.; Helm, R. F.; Grabber, J. H.; Jung,
H. -J. G. Pathway of p-coumaric acid incorporation into maize lignin as revealed by
NMR. J. Am. Chem. Soc. 1994, 116 9448−9456

(27) Ralph, J.; Marita, J. M.; Ralph, S. A.; Hatfield, R. D.; Lu, F.; Ede, R. M.;
Peng, J.; Quideau, S.; Helm, R. F.; Grabber, J. H.; Kim, H.; Jimenez- Monteon, G.;
Zhang, Y.; Jung, H. -J. G.; Landucci, L. L.; MacKay, J. J.; Sederoff, R. R.;
Chapple, C.; Boudet, A. M. Solution-state NMR of lignin. In Advances in
lignocellulosics characterization; Argyropoulos, D.S., Ed.; Tappi Press: Atlanta,
1999; 55−108

252
 V. RESULTADOS Y DISCUSIÓN

(28) Ralph, S. A.; Ralph, J.; Landucci, L. NMR database of lignin and cell wall
model compounds; US Forest Prod. Lab.: One Gifford Pinchot Dr., Madison, WI
53705 (http://ars.usda.gov/Services/docs.htm?docid=10491) (accessed: January
2009), 2004.

9 ; ; ;I ; ; ;
S I; -Barbero, J.; Martínez, A. T.; del Río, J. C. Structural
characterization of milled wood lignin from different eucalypt species.
Holzforschung 2008, 62 514−5 6

30 ; ; ; ; - Barbero, J.;
Martínez, A. T.; del Río, J. C. Isolation and structural characterization of the milled
wood lignin from Paulownia fortunei wood. Ind. Crops Prod. 2009, 30 137−143

31 ; ; ; ;I ;S
I; -Barbero, J.; Zhang, L.; Martínez, A. T. Highly acylated (acetylated
and/or p-coumaroylated) native lignins from diverse herbaceous plants. J. Agric.
Food Chem. 2008, 56 95 5−9534

(32) del Río, J. C.; Rencoret, J.; Gut ; ; -Barbero, J.;

Martínez, A. T. Structural characterization of guaiacyl-rich lignins in flax (Linum
usitatissimum) fibers and shives. J. Agric. Food Chem. 2011, 59 11088−11099

(33) Martínez, A. T.; Rencoret, J.; Marques, G.; ; I ;

-Barbero, J.; del Río, J. C. Monolignol acylation and lignin structure in
some nonwoody plants: A 2D-NMR study. Phytochemistry 2008, 69 831− 843

(34) Ralph, J.; Landucci, L. L. NMR of Lignins. In Lignin and Lignans;

Advances in Chemistry; Heitner, C., Dimmel, D. R., Schmidt, J. A., Eds.; CRC
y & : 010; 137− 34

(35) Kim, H.; Ralph, J. Solution-state 2D NMR of ball-milled plant cell wall
gels in DMSO-d6/pyridine-d5. Org. Biomol. Chem. 2010, 8 576−591

(36) Lu, F.; Ralph, J. Derivatization followed by reductive cleavage (DFRC

method), a new method for lignin analysis: protocol for analysis of DFRC
monomers. J. Agric. Food Chem. 1997, 45 590− 59

(37) Lu, F.; Ralph, J. The DFRC method for lignin analysis. Part 1. A new
f β-aryl ether cleavage: lignin model studies. J. Agric. Food Chem. 1997,
45 4655−4660

253
V. RESULTADOS Y DISCUSIÓN

(38) Lu, F.; Ralph, J. The DFRC method for lignin analysis. 2. Monomers from
isolated lignin. J. Agric. Food Chem. 1998, 46 547−55

(39) Ralph, J.; Lu, F. The DFRC method for lignin analysis. 6. A simple
modification for identifying natural acetates in lignin. J. Agric. Food Chem. 1998,
46 4616−4619

(40) del Río, J. C.; Marques, G.; Rencoret, J.; Martínez, A. ;

Occurrence of naturally acetylated lignin units. J. Agric. Food Chem. 2007, 55,
5461−5468

(41) Valmaseda, V.; Martínez, M. J.; Martínez, A. T. Kinetics of wheat straw

solid-state fermentation with Trametes versicolor and Pleurotus ostreatus - Lignin
and polysaccharide alteration and production of related enzymatic activities. Appl.
Microbiol. Biotechnol. 1991, 35 817−8 3

(42) Kondo, T.; Ohshita, T.; Kyuma, T. Comparison of characteristics of soluble

lignins from untreated and ammonia-treated wheat straw. Anim. Feed Sci. Technol.
1992, 39 53− 63

(43) Scalbert, A.; Monties, B.; Lallemand, J.-Y.; Guittet, E.; Rolando, C. Ether
linkage between phenolic acids and lignin fractions from wheat straw.
Phytochemistry 1985, 24 1359−136

(44) Lam, T. B. T.; Iiyama, K.; Stone, B. A. Cinnamic acid bridges between cell
wall polymers in wheat and phalaris intemodes. Phytochemistry 1992, 31,
1179−1183

(45) Grabber, J. H.; Ralph, J.; Hatfield, R. D. Cross-linking of maize walls by

ferulate dimerization and incorporation into lignin. J. Agric. Food Chem. 2000, 48,
6106−6113

46 ; ; g I ;I ;
Composition of non-woody plant lignins and cinnamic acids by Py-GC/MS,
Py/TMAH and FT-IR. J. Anal. Appl. Pyrol. 2007, 79 39−46

47 ; ; -Vila, F. J. Thermally assisted

hydrolysis and alkylation as a novel pyrolytic approach for the structural
characterization of natural biopolymers and geomacromolecules. Trends Anal.
Chem. 1996, 15 70−79

254
 V. RESULTADOS Y DISCUSIÓN

(48) Ralph, J.; Bunzel, M.; Marita, J. M.; Hatfield, R. D.; Lu, F.; Kim, H.;
Schatz, P. F.; Grabber, J. H.; Steinhart, H. Peroxidase-dependent cross-linking
reactions of p-hydroxycinnamates in plant cell walls. Phytochem. Rev. 2004, 3,
79−96

(49) Ralph, J.; Hatfield, R. D.; Piquemal, J.; Yahiaoui, N.; Pean, M.; Lapierre,
C.; Boudet, A. M. NMR characterization of altered lignins extracted from tobacco
plants down-regulated for lignification enzymes cinnamyl-alcohol dehydrogenase
and cinnamyl-CoA reductase. Proc. Natl. Acad. Sci. U.S.A. 1998, 95,
1 803−1 808

50 W g ;K O; ;S ;S y W; ;
Hiermann, A. Flavonolignans from Avena sativa. J. Nat. Prod. 2005, 68 89− 9

(51) Chang, C. L.; Wang, G. J.; Zhang, L. J.; Tsai, W. J.; Chen, R. Y.; Wu, Y.
C.; Kuo, Y. H. Cardiovascular protective flavolignans and flavonoids from
Calamus quiquesetinervius. Phytochemistry 2010, 71 71− 79

5 Z ; I K − f nctional
nutraceutical. Phytochem. Rev. 2010, 9 413−4 4

(53) Grabber, J. H.; Schatz, P. F.; Kim, H.; Lu, F.; Ralph, J. Identifying new
lignin bioengineering targerts: 1. Monolignol-substitute impacts on lignin
formation and cell wall fermentability. BMC Plant Biol. 2010, 10, 114.

(54) Sederoff, R. R.; MacKay, J. J.; Ralph, J.; Hatfield, R. D. Unexpected

variation in lignin. Curr. Opin. Plant Biol. 1999, 2 145−15

(55) Ralph, J.; Peng, J.; Lu, F.; Hatfield, R. D. Are lignins optically active? J.
Agric. Food Chem. 1999, 47 991− 996

(56) Ralph, J.; Brunow, G.; Harris, P. J.; Dixon, R. A.; Schatz, P. F.; Boerjan,
W. Lignification: Are lignins biosynthesized via simple combinatorial chemistry or
via proteinaceous control and template replication? In Recent Advances in
Polyphenol Research; Daayf, F., El Hadrami, A., Adam, L., Ballance, G. M., Eds.;
Wiley-Blackwell g: Oxf K 008; 1 36−66

(57) Umezawa, T. Diversity in lignan biosynthesis. Phytochem. Rev. 2004, 2,

371−390

255
V. RESULTADOS Y DISCUSIÓN

(58) Boerjan, W.; Ralph, J.; Baucher, M. Lignin biosynthesis. Annu. Rev. Plant
Biol. 2003, 54 519−549

(59) Lu, F.; Ralph, J. Detection and determination of p-coumaroylated units in

lignins. J. Agric. Food Chem. 1999, 47 1988−199

(60) Ralph, J. An unusual lignin from kenaf. J. Nat. Prod. 1996, 59, 341−34

(61) Lu, F.; Ralph, J. Novel tetrahydrofuran structures derived from β−β-
coupling reactions involving sinapyl acetate in Kenaf lignins. Org. Biomol. Chem.
2008, 6 3681−3694

(62) Terashima, N.; Fukushima, K.; He, L.-F.; Takabe, K. Comprehensive

model of the lignified plant cell wall. In Forage Cell Wall Structure and
Digestibility; Jung, H. G., Buxton, D. R., Hatfield, R.D., Ralph, J., Eds.; American
Society of Agronomy, Crop Science Society of America, Soil Science Society of
: 1993; 47− 70

256
 V. RESULTADOS Y DISCUSIÓN

Publicación VI:

Prinsen P., Rencoret J., Gutiérrez A., Liitiä T., Tamminen T., Colodette J.L.,
Berbis M.A., Martínez A.T. and del Río J.C. (2013). Modification of the lignin
structure during chemical deconstruction of eucalypt wood by kraft-, soda-AQ and
soda-O2 cooking. Industrial & Engineering Chemistry Research (enviado).

257
 V. RESULTADOS Y DISCUSIÓN

Modification of the lignin structure during chemical

deconstruction of eucalypt wood by kraft, soda-AQ and soda-O2
cooking

Pepijn Prinsen1, Jorge Rencoret1, Ana Gutiérrez1, Tiina Liitiä2, Tarja Tamminen2,
Jorge L. Colodette3, Manuel A. Berbis4, Jesús Jiménez-Barbero4, Ángel T.
Martínez4 and José C. del Río1

1
Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS), CSIC, PO
Box 1052, E-41080 Seville, Spain

2
VTT Technical Research Centre of Finland, Biologinkuja 7, Espoo, P.O. Box
1000, FI-02044 VTT, Finland

3
Department of Forestry Engineering at Federal University of Viçosa, Viçosa, MG
36.570-000, Brazil

4
Centro de Investigaciones Biológicas (CIB), CSIC, Ramiro de Maeztu 9, E-28040
Madrid, Spain

ABSTRACT

The lignins isolated from alkaline pulping (kraft, soda-AQ and soda-O2) of
eucalypt feedstock were characterized by different analytical techniques (SEC, Py-
GC/MS, 1H-13C 2D-NMR, and 31P-NMR). The characteristics of the lignins were
compared at different pulp kappa levels, and with the native lignin isolated from
the wood. The structural differences between the kraft, soda-AQ, and soda-O2
residual lignins were more significant at earlier pulping stages. At the final stages
all the lignin characteristics were similar with exception of their phenolic content.
Strong differences between lignins from pulps and cooking liquors were observed,
including pulp residual lignin enrichment in guaiacyl units and black liquor lignin
enrichment in syringyl units. The soda-O2 process showed higher lignin

259
V. RESULTADOS Y DISCUSIÓN

degradation efficiency, and provided promising results as pretreatment for the

deconstruction of eucalypt feedstocks for subsequent hydrolysis and fermentation
in the production of bioethanol.

KEYWORDS: Eucalypt wood; Kraft; Soda-AQ; Soda-O2; lignin.

INTRODUCTION

Alkaline deconstructions appear as attractive pretreatments for biorefinery

purposes. Kraft cooking is the alkaline process most widely used for delignification
of hardwoods and softwoods for the production of paper pulp. The key to the kraft
process is the recovery furnace that is quite efficient at recovering the pulping
chemicals NaOH and Na2S. When the cooking liquor is used for further
fractionation into valuable components, the soda- anthraquinone (soda-AQ)
process, in which AQ is used as a pulping additive to decrease the carbohydrate
degradation, is favorable because the absence of sulfur compounds in the liquor
enables recovery of sulfur-free raw lignin (and lignin products) for the production
of biobased materials. On the other hand, the soda-O2 process is especially
promising as pretreatment for the hydrolysis and fermentation, as the oxidative
conditions affect the fiber ultrastructure in a deconstructive manner. In contrast to
paper grade pulps, the fiber strength properties are not important in biofuel
production, as the deconstruction of the lignocellulosic matrix is aimed by
removing and/or modifying the lignin and getting better access to the
carbohydrates.

The chemistry and efficiency of alkaline cooking is influenced by several

process parameters, including cooking time and temperature, active alkali charge,
and the possible presence of other cooking chemicals such as sulfide or
anthraquinone (AQ). In addition, the structural characteristics of the native lignin
also play an important role in the chemistry of the pulping process. Hence, in
hardwoods, the content of syringy w y β-
1
aryl ether linkages, is one of the most determinant characteristics.

Numerous studies have attempted to correlate the chemical structure of residual

lignin in pulp and dissolved lignin in cooking liquors to the pulping efficiency,
contributing to a more comprehensive study of the delignification mechanisms
involved.2–10 The dominating reactions in lignin during alkaline pulping are
v g f β-aryl ether linkages in phenolic units (at initial stage) and in etherified

260
 V. RESULTADOS Y DISCUSIÓN

units (at the bulk phase).11 The attack of an ionized group from the adjacent carbon
β-position of the lignin side chain or from the cooking medium
(hydrogensulfide, hydroxide ion, etc.) cleaves the alkyl-aryl ether linkage
fragmenting the lignin. In this way, the presence of phenolic hydroxyl groups is
believed to play an important role in alkaline pulping.12 Moreover, they assist the
f g g g y g g ’ y k
13
medium. However, native wood lignin contains a low amount of phenolic
hydroxyl groups. During delignification, the cleavage of alkyl-aryl ether linkages
creates new phenolic units in the residual lignin, and also in the lignin solubilized
into the cooking liquor. As a result, the residual and dissolved lignins have a higher
abundance of phenol hydroxyl groups than native wood lignin.12 However,
condensation reactions may also occur via the addition of nucleophiles to quinone
methide intermediates. In alkaline conditions, quinone methide structures are
formed from phenolic units starting with the delocalization of the phenolate group,
which depends on the actual alkalinity of the cooking medium.13 The extent of
condensation also depends on the leaving group capacity of the nucleophile-
quinone methide adduct. The extent and distribution of fragmentation and
condensation reactions determine the structural modification of the lignin. The
factors which influence the extent of condensation are still subject of
discussion.7,9,14,15 In general terms, residual lignins at lower kappa numbers have a
w f β-aryl ether linkages and a higher amount of condensed
3,6,10
structures.

The aim of this study is investigating the behavior and fate of the lignin of an
eucalypt feedstock during chemical deconstruction by kraft, soda-AQ and soda-O2
cooking processes. For this purpose, the residual lignins from the pulps and the
precipitated lignins from the black liquors were isolated and analyzed, and their
structural characteristics compared to that of the milled wood lignin (MWL)
isolated from the initial raw material. The study provides information regarding the
chemistry of the delignification and deconstruction processes necessary to optimize
the processing of eucalypt wood for traditional paper and packaging purposes, as
well as more widely for various biorefinery concepts.

MATERIALS AND METHODS

Pulp and black liquor samples from cooking experiments

The feedstock consisted of wood chips from an eucalypt E. grandis × [E.

urophylla × E. globulus] (G1×UGL) hybrid grown in Brazil. The cooking

261
V. RESULTADOS Y DISCUSIÓN

experiments were carried out at Suzano Papel e Celulose, Brazil. Chips were
classified according to standard SCAN-CM 40:94, which is based on chip thickness
(4-6 mm), and the accepts were stored. The alkaline cooking consisted of three
different processes: kraft, NaOH in the presence of anthraquinone (soda-AQ) and
NaOH in the presence of oxygen (soda-O2). General conditions for the kraft, soda-
AQ and soda-O2 alkaline cooking of the eucalypt G1×UGL hybrid wood are
detailed in Table 1.

Table 1. General conditions for the kraft, soda-AQ and soda-O2 alkaline cooking of wood
of the eucalypt G1UGL hybrid.

Parameter Conditions
Wood (kg) 1.0
Active alkali as NaOH (% on oven dry chips) variable
Sulfidity as NaOH (% on oven dry chips; only in kraft) 26
Anthraquinone charge (% on oven dry chips; only in soda-AQ) 0.05
Oxygen charge (% on oven dry chips; only in soda-O2) 6
Liquor to chips ratio (L/kg) 4
Maximum temperature (ºC) 170
Time to maximum temperature (min) 90
Time at maximum temperature (min) 50

The cooking experiments were carried out in a CRS digester (model CPS 010:
Recycle Digester System) containing two individual reactors of 10 L each,
equipped with a forced liquor circulation system and electrically heated with
temperature and pressure control. The digester was coupled with a cooling system
to ensure the cooling of the liquor after the cooking simulation. Eight cooking
experiments were carried out for every type of process using variable alkali charges
in order to establish the delignification curves for each sample. The kappa number
of the pulp was determined according to the TAPPI T236 cm-85 standard.16 In the
soda-O2 cooking the oxygen dosage was split into three parts and applied after 50,
70 and 110 min reaction time from the beginning of cooking, thereby ensuring that
the pressure limit (20 kgf/cm²) of the digester was not exceeded. The dosage of
oxygen applied was enough to ensure small residual oxygen content at the end of
the cooking. At this point the cooling system was turned on for 30 minutes, and the

262
 V. RESULTADOS Y DISCUSIÓN

temperature of the residual liquor reduced to levels close to 55°C. After cooling
and unloading the digester, the cooked chips were washed with hot water (70 °C)
for 3 hours and then with water at ambient temperature until the complete removal
of residual liquor (about 12 h). Fiber individualization was achieved in a
"hidrapulper" (15 liters capacity), followed by fine screening (Voith laboratory
cleaner, equipped with perforated plates with 0.2 mm openings). The material
retained on the sieve (rejects) was dried and weighted. The accept pulp was
dewatered in centrifuge to a consistency of about 30% and stored for further
analysis. Two series of pulping were completed, one for the production of paper
grade pulp and the other as pretreatment for bioetanol- and biogas production. In
the latter case, no reject removal was performed. Kraft- and soda-AQ processes
were carried out for the paper pulp series. Pulp and black liquor samples were
collected from the digester at kappa numbers 20 and 15. For the bioethanol- and
biogas pulping series the eucalypt feedstock was submitted to soda-AQ- and soda-
O2 cooking and pulp samples were collected at kappa numbers 50, 35 and 15.
Consequently, two soda-AQ pulps (from both pulp series) were available at kappa
number 15. The reject content in this case was so low that the pulps from both
series were practically identical. In fact, similar results of the soda-AQ residual
lignins at kappa 15 were obtained in both series and the mean values are given for
the results of this residual lignin.

MWL isolation from wood

The isolation of MWL from wood of the eucalypt G1×UGL hybrid was
previously described.17 MWL was extracted from finely ball-milled (15 h) wood,
free of extractives and hot water soluble material, using dioxane-water (9:1, v/v),
followed by evaporation of the solvent, and purified as described.18 The final yield
ranged from 15-20% of the original Klason lignin content.

Residual lignin isolation from pulp

For the isolation of residual lignin, first all pulp samples were air dried at 37-40
ºC. Extractives were eliminated by Soxhlet-extraction during 9 hours with acetone
and subsequent extraction with water at 100 ºC (3 steps). Then 15 grams of dry
pulp was submitted to acidolysis according to the method previously described,19
using a two-step extraction with 0.1 M HCl in a 1,4-dioxane:water mixture (82:18
v/v) under an inert (Argon) atmosphere at 88-92 ºC. The solid:liquid ratio for the
first and second extraction step was respectively 13.3 and 10.0 mL per gram of dry
pulp. After the extractions the pulp was washed with the same 1,4-dioxane:water

263
V. RESULTADOS Y DISCUSIÓN

mixture without containing HCl. In order to avoid high acid concentration, the
washing liquor was added equally to both extracts which were evaporated
separately under reduced pressure until about 70% volume reduction. Both
concentrated extracts were added together in 1.5 L of cold distilled water under
strong stirring. The lignin was precipitated overnight at 4 ºC and then centrifugated
(25 minutes, 9000 rpm, 4 ºC). A washing step with cold distilled water was
included, and after recovery (30 minutes, 9000 rpm, 4 ºC) the samples were freeze-
dried. The extraction yields of the residual lignins were between 26 and 82%, based
on the theoretical lignin content of pulp (% lignin = 0.15 × kappa number). The
yield drastically lowered at kappa number 20 and 15. The residual lignins were
submitted to Soxhlet extraction with n-pentane during 8 hours. After drying with
nitrogen the residual lignin was ready for analysis.

Lignin isolation from black liquors

The lignins from the collected black liquors were precipitated at pH 4.0 and the
obtained slurries were air-dried at 40 ºC. The homogenized samples were stored in
a desiccator until stable humidity. All analytical techniques (SEC, Py-GC/MS, 1H-
13
C 2D-NMR and 31P-NMR) were performed on these samples.

Pyrolysis-Gas Chromatography-Mass Spectrometry (Py-GC/MS)

y y f g x y 100 μg w f w / Y-
3030D micro-furnace pyrolyzer (Frontier Laboratories Ltd.) connected to an
Agilent 7820A gas chromatograph using a DB-1701 fused-silica capillary column
60 x 0 5 0 5μ f k g 5975 v
detector (EI at 70 eV). The pyrolysis was performed at 500 ºC. The oven
temperature was programmed from 45 ºC (4 min) to 280 ºC (10 min) at 4 ºC min -1.
Helium was the carrier gas (1 mL min-1). The compounds were identified by
comparing their mass spectra with those of the Wiley and NIST libraries and those
reported in the literature.20 Peak molar areas were calculated for the lignin-
degradation products, the summed areas were normalized, and the data for two
repetitive analyses were averaged and expressed as percentages.

Two-dimensional Nuclear Magnetic Resonance (2D-NMR)

2D-NMR spectra were recorded at 25ºC in a Bruker AVANCE 600 MHz

spectrometer, equipped with a cryogenically-cooled z-gradient triple resonance
probe. Forty mg of lignin sample were dissolved in 0.75 mL of dimethylsulfoxide
(DMSO)-d6, and 1H-13C HSQC (heteronuclear single quantum correlation) spectra

264
 V. RESULTADOS Y DISCUSIÓN

were recorded. The spectral widths were 5000 and 13200 Hz for the 1H and 13C
dimensions, respectively. The number of collected complex points was 2048 for the
1
H dimension, with a recycle delay of 1 s. The number of transients was 64, and
256 time increments were recorded in the 13C dimension. The 1JCH constant was set
to 140 Hz. The J-coupling evolution delay was set to 3.2 ms. Squared cosine-bell
apodization function was applied in both dimensions. Prior to Fourier
transformation, the data matrices were zero filled up to 1024 points in the 13C
v k w f δC 39.5
; δH 2.49 ppm). HSQC cross-signals were assigned by comparing with the
literature.6,21–26 In the aromatic region, C2,6-H2,6 correlations from S units and C2-H2
correlations from G units were used to estimate the S/G lignin ratios. The
abundances of the different inter-unit linkages were referred to the number of
aromatic units (as per 100 aromatic units), to obtain a comparative estimation of
their removal during pulping and bleaching.
31
P-NMR spectroscopy

Phosphitylation of lignin samples was performed according to the literature.27

About 40 mg of pure lignin was weighted and dissolved in 1.6 mL of a solvent
mixture which consisted of dimethylformamide:pyridine:deuterated chloroform
(1:1.70:2.63 v/v) to which 0.01 mmol of endo-N-hydroxy-5-norbornene-2,3-
dicarboximide was added as internal standard, together with 0.032 mmol
chromium(III)acetylacetonate as relaxation agent. Then 200 µL of 2-chloro-
4,4,5,5-tetramethyl-1,3,2-dioxaphospholane (phosphitylating reagent II) was added.
The 31P-NMR spectra were recorded on a Bruker Avance III 500 MHz
spectrometer at 23 ºC using a 90º pulse width. For acquisition an inversegated
decoupling sequence was used with deuterated chloroform as the locking solvent
and 512 scans were recorded with a delay time of 5 s between successive pulses.
For processing data, the chemical shifts were calibrated to the sharp signal (132.2
ppm) of the reaction product between water and phosphitylation reagent II and the
line broadening was set to 2 Hz, followed by exponential window multiplication,
Fourier transformation, automatic/manual phase correction and automatic baseline
correction. The integration areas were expressed relative to the area of the internal
standard (IS) signal which was set to 1. Additionally the peak width at half height
of the IS signal was checked to guarantee a similar resolution in all samples and
therefore an exact quantification. The deconvolutions of the region containing
syringyl and condensed hydroxyl group signals were processed on spectra with a
line broadening factor of 8 Hz. In order to obtain a satisfactory deconvolution result
(as close as possible to the experimental 31P-NMR spectrum), 12-20 signals were

265
V. RESULTADOS Y DISCUSIÓN

selected in the 144.2 – 141.0 ppm region. For the calculation of the hydroxyl group
content the impurity of the IS (97%) was taken into account. The results were
expressed as mmol of hydroxyl groups per g of pure lignin. For the black liquor
samples the purity was based on the sum of Klason and acid soluble lignin content.
The purity of the MWL and residual lignin samples was set to 100 %, as their
isolation included purifications steps. Standard deviations were calculated from the
results of 3 independent samples taken from the soda-O2 residual lignin bulk
sample at kappa number 50. Their 31P-NMR spectra were recorded and processed
independently and the 144.2 – 141.0 ppm region was deconvoluted with selected
signals having similar chemical shift and width.

Size Exclusion Chromatography (SEC)

The SEC analysis were performed on a Waters HPLC system equipped with a
UV detector with response set at 280 nm using the following conditions: columns,
SS’ X 1000 100 000 Å; 01 OH; f w 05 /
temperature, 25 ºC. The data acquisition and computation was performed with
Empower 2 software. Before analysis about 4 mg lignin was dissolved overnight in
4 y OH 0 1 f w 0 45 μ y g
filters. The molecular masses were calculated relative to the sulphonated Na-
polystyrene (Na-PSS, 3420-148500 g/mol) standards. The weight-average
molecular weights (Mw), number average molecular weights (Mn), and the
polydispersities (Mw/Mn) were calculated.

RESULTS AND DISCUSSION

The effects of different chemical alkaline delignification processes (kraft, soda-

AQ and soda-O2 processes) of a eucalypt feedstock on the structures of the residual
and dissolved lignins were evaluated. Two different sets of pulps were produced
from wood of the eucalypt G1×UGL hybrid: i) Pulps intended for paper production
by the kraft and soda-AQ processes at kappa 20 and 15 and ii) Pulps intended for
bioethanol and biogas production by the soda-AQ and soda-O2 processes at kappa
50, 35 and 15 without rejects removal after cooking. The residual lignins (isolated
from the pulps by acidolysis) and the lignins precipitated from the black liquors
were analyzed by different analytical techniques, including SEC, Py-GC/MS, 1H-
13
C 2D-NMR and 31P-NMR, and their structural characteristics were compared to
those of the MWL isolated from the initial raw material.

266
 V. RESULTADOS Y DISCUSIÓN

SEC

The comparison of the molecular weights of the residual lignins, and the lignins
precipitated from the black liquors, for the different cooking processes informed on
the extent of the structural changes in the lignin polymer during cooking. The
values of the weight-average (Mw) and number-average (Mn) molecular weights
and the polydispersity (Mw/Mn) of the analyzed samples are listed in Table 2.

Table 2. Weight-average (Mw) and number-average (Mn) molecular weights (Da), and
polydispersity (Mw/Mn) of the MWL from wood of the eucalypt G1×UGL hybrid, and
residual and black liquor lignins from kraft, soda-AQ and soda-O2 pulping at different
kappa numbers.

MWL Residual lignins Black liquor lignins

Kraft Soda-AQ Soda-O2 Kraft Soda-AQ
Kappa K20 K15 K50 K35 K20 K15 K50 K35 K15 K20 K15 K20 K15
number
Mn 3500 2440 2370 2480 2510 2710 2530 2380 2360 2300 1680 1640 1660 1580

Mw 13300 4340 3490 10550 8300 4770 3870 9380 6080 3640 2390 2280 2540 2350

Mw/Mn 3.8 1.7 1.5 4.3 3.3 1.8 1.5 3.9 2.6 1.6 1.4 1.4 1.5 1.5

The molecular weights of the residual lignins decreased concomitantly with the
lignin content in the pulp (with decreasing kappa number). At kappa number 50
and 35 the molecular weights (Mw) of the residual lignins from soda-O2 were lower
than the residual lignins from soda-AQ, in agreement with the results from Py-
GC/MS and 2D-NMR that indicated a poorer delignification efficiency of the soda-
O2 cooking process. However, at kappa number 15 the molecular weights of the
residual lignins were rather similar, with the residual lignins from kraft and soda-
O2 processes having the lowest values. During the chemical delignification process,
fractionation of the lignin polymer occurred resulting in residual lignin fragments
with smaller molecular size, which were more uniformly distributed in size as seen
from the decreasing polydispersity. Residual lignins at kappa 50 had a slightly
higher polydispersity than native lignin from the wood, mainly due to the presence
of fractions with higher molecular weight as seen from the SEC chromatographs.
No significant condensation of the residual lignin was observed from the SEC

267
V. RESULTADOS Y DISCUSIÓN

chromatographs in the alkaline cooking processes. On the other hand, the

molecular weights of the lignins precipitated from the black liquors from kraft and
soda-AQ processes were lower than their respective residual lignins. These results
show the high degradation degree of the lignin fragments dissolved in the alkaline
cooking media, in accordance with the results from Py-GC/MS and 2D-NMR
shown below.

Py-GC/MS

All the lignin samples were analysed by Py-GC/MS. The pyrograms of a

selected pulp residual lignin from eucalypt G1xUGL hybrid (from kraft pulp at
kappa 20) and the lignin precipitated from the respective black liquor are shown in
Figure 1. The Py-GC/MS of the MWL from eucalypt G1xUGL hybrid is also
shown for comparison. The identities and relative molar abundances of the
identified lignin compounds are shown in Table 3. The main lignin structural
characteristics obtained from the Py-GC/MS data (percentage of H, G and S units,
and S/G ratio) of the residual lignins and black liquors from the kraft, soda-AQ and
soda-O2 processes at different kappa numbers are shown in Table 3.

Interestingly, there is a similarity between the residual lignins isolated from

kraft and soda-AQ pulps and the native lignin (as reflected by the MWL).
However, the residual lignins are comparatively depleted in S-lignin units and
enriched in G- and H-lignin units, with decreasing kappa number (with increasing
the extent of delignification), resulting in a decrease of the S/G ratio, as a
consequence of the preferential removal of S-lignin during alkaline delignification.
A small increase in the relative amounts of pyrolysis compounds with shorter chain
(as indicated by the ratio of lignin phenolic markers bearing side-chains of 0–2
carbon atoms respect to lignin phenolic markers bearing side-chains of 3 carbon
atoms, Ph-C0-2/Ph-C3) indicates the side chain degradation of the residual lignins,
which is more evident at lower kappa numbers.

On the other hand, the Py-GC/MS data indicate that the lignins precipitated from
the black liquors are completely different from the native lignin and from the
residual lignins in the pulps. The lignins in black liquors are highly enriched in S-
lignin units (due to the preferential solubilisation of S-lignin units) and depleted in
G-units. In addition, the higher amounts of pyrolysis compounds with shorter chain
and/or oxidized, indicates that this lignin is heavily degraded and oxidized. Similar
results have been previously observed during kraft cooking of eucalypt wood.5,6

268
 V. RESULTADOS Y DISCUSIÓN

15
100%
(a)
31
23
30
Relative abundance

12

6 9 33
14

25
+ 38
2 16 26
20 27 35
18 32
17 22 28
4 34
+ 8 10 13 19 29 36
5 11 24 37
1 3 7

25 30 35 40 45 50 55
12
100% 15
(b)
Relative abundance

23

30

2 20
9
25 31 33
+ 35
14 26
4 8 21 27
+ 36
16 22 24 29
5 13 18 34 38
3 7 10 11 17 28 37
1

25 30 35 40 45 50 55
12
100%
(c)
Relative abundance

23
2 33

15 25
9 10
4 20 +
+ + 31
5 6 8 11 14 16 19 22 24 26 29 30 35
1 3 13 27 32 36

25 30 35 40 45 50 55
Retention time (min)

Figure 1. Py-GC/MS chromatograms of: (a) MWL isolated from wood of the eucalypt
G1×UGL hybrid; (b) kraft residual lignin at kappa number 20, and (c) lignin from the
corresponding black liquor. The numbers refer to the compounds listed in Table 3.

269
V. RESULTADOS Y DISCUSIÓN

Table 3. Identification and relative molar abundance (%) of the compounds identified in
the Py-GC/MS of MWL from wood of the eucalypt G1×UGL hybrid, and residual and
black liquor lignins from kraft, soda-AQ and soda-O2 pulping at different kappa numbers.

MWL Residual lignins Black liquor lignins

Kraft Soda-AQ Soda-O2 Kraft Soda-AQ
Label Compound K20 K15 K50 K35 K20 K15 K50 K35 K15 K20 K15 K20 K15
1 phenol 0.3 0.5 0.7 0.2 0.3 0.4 0.9 0.2 0.3 0.5 0.3 0.6 0.5 0.6
2 guaiacol 3.5 6.1 8.0 4.0 4.4 6.0 7.3 4.8 6.4 8.4 12.0 12.5 11.5 11.3
3 2-methylphenol 0.4 0.7 0.9 0.5 0.5 0.7 1.0 0.5 0.6 0.8 0.4 0.6 0.6 0.7
4 4-methylphenol 0.3 0.6 0.8 0.3 0.3 0.5 0.9 0.3 0.4 0.6 0.2 0.3 0.4 0.5
5 3-methylphenol 0.3 0.3 0.4 0.1 0.2 0.2 0.5 0.2 0.2 0.3 0.1 0.2 0.2 0.2
6 4-methylguaiacol 5.8 7.1 8.6 6.2 6.1 8.8 9.0 7.5 8.5 9.4 1.2 1.3 1.3 1.5
7 4-ethylphenol 0.4 0.7 0.8 0.5 0.6 0.7 0.8 0.6 0.7 0.7 0.1 0.2 0.3 0.4
8 4-ethylguaiacol 1.1 2.1 3.1 1.7 1.7 2.6 2.5 1.8 2.1 1.9 1.0 1.1 1.5 1.6
9 4-vinylguaiacol 5.2 4.8 4.4 3.8 4.3 4.6 6.1 4.8 5.5 6.5 4.2 4.5 5.9 5.8
10 eugenol 1.0 0.6 0.1 0.8 0.7 0.5 0.4 0.8 0.7 0.0 0.2 0.2 0.2 0.2
11 4-propylguiacol 0.2 0.4 0.4 0.3 0.3 0.4 0.3 0.3 0.3 0.2 0.1 0.1 0.1 0.2
12 syringol 7.0 15.7 20.8 12.5 14.2 19.0 15.3 11.9 14.2 16.8 45.5 42.5 40.4 38.9
13 cis-isoeugenol 0.9 0.8 0.7 0.9 0.8 0.6 0.6 1.5 1.3 0.5 0.2 0.2 0.2 0.2
14 trans-isoeugenol 4.7 2.8 2.3 3.7 3.6 2.4 1.9 4.2 3.6 1.3 0.8 0.7 0.9 0.9
15 4-methylsyringol 10.2 14.6 14.6 14.5 14.9 15.9 14.7 14.2 14.2 13.9 4.4 4.4 4.7 5.3
16 vanillin 2.9 1.4 1.6 2.2 1.8 1.7 2.1 2.7 2.4 2.8 0.8 1.4 1.7 1.7
17 4-propinylguaiacol 1.4 0.3 0.1 0.2 0.6 0.2 0.3 0.5 0.4 0.2 0.1 0.0 0.1 0.1
18 4-allenylguaiacol 1.5 0.3 0.2 0.6 0.8 0.2 0.3 0.8 0.4 0.3 0.1 0.1 0.1 0.0
19 homovanillin 1.0 0.0 0.0 0.0 0.0 0.0 0.8 0.0 0.0 0.1 0.3 0.2 0.5 0.5
20 4-ethylsyringol 1.8 4.1 4.9 3.5 3.6 4.5 3.7 2.9 3.2 2.7 2.6 2.5 3.6 4.0
21 guaiacyl vinyl ketone 0.3 0.3 0.3 0.3 0.4 0.2 0.3 0.3 0.4 0.5 0.1 0.0 0.1 0.2
22 acetoguaiacone 1.7 1.2 2.1 1.3 1.1 1.3 1.8 1.5 1.5 1.9 1.1 1.7 0.9 1.0
23 4-vinylsyringol 8.4 9.5 5.9 8.2 9.9 7.4 7.6 8.5 8.0 8.7 8.8 6.8 10.0 9.0
24 guaiacylacetone 0.7 1.4 1.6 1.0 1.1 1.6 1.6 1.5 1.6 2.0 0.3 0.5 0.7 0.6
25 4-allylsyringol 2.0 2.5 1.2 2.6 2.3 1.3 0.9 2.3 1.5 1.1 0.4 0.3 0.4 0.5
26 4-propylsyringol 0.2 0.6 0.6 0.5 0.4 0.5 0.4 0.4 0.3 0.3 0.2 0.1 0.2 0.2
cis-4-
27 1.6 1.5 0.9 2.1 1.9 1.0 1.6 1.6 1.2 0.6 0.4 0.3 0.4 0.4
propenylsyringol
28 4-propinylsyringol 1.2 0.6 0.2 0.5 0.5 0.5 0.6 0.5 0.3 0.4 0.1 0.1 0.0 0.0
29 4-allenylsyringol 0.9 0.8 0.2 0.3 0.4 0.3 0.4 0.3 0.2 0.4 0.0 0.1 0.0 0.0
4-trans-
30 7.3 6.3 3.1 8.6 7.9 4.1 2.4 7.1 5.6 3.1 1.5 1.1 1.5 1.5
propenylsyringol
31 syringaldehyde 9.6 3.5 2.8 6.8 5.6 4.4 4.6 5.9 4.8 5.0 2.3 4.5 2.5 3.5
32 homosyringaldehyde 1.7 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.6 0.4 1.2 1.1
33 acetosyringone 4.2 3.1 4.4 3.8 3.2 2.8 3.8 3.5 3.2 3.5 7.6 9.0 5.0 5.4
34 trans-coniferaldehyde 1.0 0.7 0.1 0.4 0.2 0.1 0.1 0.2 0.1 0.1 0.0 0.0 0.0 0.1
35 syringylacetone 1.4 2.6 1.9 2.4 2.8 3.0 2.7 3.4 3.4 3.0 1.2 0.9 1.9 1.7
36 propiosyringone 0.7 1.1 1.1 1.4 1.1 1.1 0.8 1.3 1.3 0.9 0.3 0.4 0.3 0.3
37 syringyl vinyl ketone 0.7 0.2 0.0 0.4 0.2 0.1 0.2 0.2 0.3 0.4 0.1 0.1 0.1 0.1
38 trans-sinapaldehyde 6.3 0.4 0.1 2.7 1.5 0.2 0.5 0.8 0.7 0.5 0.1 0.1 0.1 0.1
%H 1.7 2.7 3.6 1.6 1.8 2.6 4.1 1.8 2.2 2.8 1.1 1.9 2.0 2.3
%G 33.1 30.3 33.5 27.5 27.6 31.3 35.6 33.3 35.3 36.2 22.6 24.4 25.7 25.8
%S 65.2 67.0 62.8 70.9 70.5 66.1 60.4 64.9 62.5 61.0 76.3 73.7 72.4 71.9
S/G 2.0 2.1 1.9 2.6 2.5 2.1 1.7 2.0 1.8 1.7 3.4 3.1 2.8 2.8
a
Ph-C0-2/Ph-C3 1.7 3.2 5.7 2.3 2.6 4.6 5.3 2.6 3.2 5.3 15 19 13 13
b
% Ph-C=O 18 9 11 14 12 10 12 14 12 13 12 16 10 12
a
Ratio of lignin phenolic markers bearing side-chains of 0–2 carbon atoms respect to lignin
phenolic markers bearing side-chains of 3 carbon atoms.
b
Percentage of lignin phenolic markers bearing a carboxyl/carbonyl group at the α-carbon.

270
 V. RESULTADOS Y DISCUSIÓN

Interestingly, at earlier stages of delignification (kappa numbers 50 and 35), the

S/G ratios of the residual lignins from the soda-O2 process were significantly lower
than those of the soda-AQ residual lignins. Therefore, soda-O2 process seems to be
a less efficient delignification process than soda-AQ, at least at higher kappa
numbers, requiring removal of the more easily delignified S lignin to a higher
extent in order to reach the same kappa number level. However, at low kappa
numbers (kappa 20 and 15) the S/G ratios of the kraft, soda-AQ and soda-O2
residual lignins were rather similar suggesting that oxidation under the alkaline
oxygen conditions is the prevalent reaction under conditions where free phenolic
hydroxyl groups are present in the residual lignin in reasonable amounts.

No major structural oxidation was observed in the residual lignins isolated from
the kraft, soda-AQ and soda-O2 processes, compared to the MWL, as shown by the
relatively low amounts of Cα-oxidized lignin compounds (i.e. peaks 16: vanillin;
22: acetoguaiacone; 31: syringaldehyde; 33: acetosyringone; 37: propiosyringone)
in their pyrograms. The content of carbonyl compounds decreased in the residual
lignin until kappa number 35-20 with respect to the MWL (Table 3), indicating
that native oxidized units were removed during delignification and that new
oxidized units were cleaved rapidly. Then, at kappa number 15, the content was
higher again in all residual lignins. This could be the result of a slower removal
degree during the terminal phase. The slower delignification at the final stage has
been explained to be due to the recalcitrance of the residual lignin structure and/or
its binding with carbohydrates.28 At kappa number 15 the side chain of the kraft
residual lignin was 8% more degraded than the soda-AQ residual lignin and the
content of oxidized pyrolysis compounds was 18% lower. The content of oxidized
pyrolysis compounds was 22% higher in the corresponding kraft black liquor lignin
with respect to the soda-AQ.

2D-NMR

Additional information regarding the structure of these lignins was obtained

from 2D-NMR. The HSQC spectra of a representative residual lignin (isolated
from the eucalypt G1xUGL hybrid kraft pulp at kappa 20), and the precipitated
lignin from its respective black liquor, are shown in Figure 2. The spectrum of the
MWL isolated from G1xUGL is also shown for comparison. The main
substructures found are also depicted in Figure 2.

The detailed structural characteristics of the MWL of eucalypt G1×UGL from

HSQC data has been described previously.17 The main inter-unit linkages observed

271
V. RESULTADOS Y DISCUSIÓN

in the residual lignins from the kraft, soda-AQ and soda-O2 w β–O–4'
alkyl- y β–β β–5 phenylcoumaran structures. Interestingly, no
oxidized lignin side chains were observed by 2D-NMR in the residual lignins, as
already seen by Py-GC/MS. The distributions of the different inter-unit linkages in
the pulp residual lignins are similar to those observed in the native lignin in wood,
w f β–O–4' alkyl-aryl ether linkages, followed by lower
amounts of resinols and phenylcoumarans, but with a significant reduction in their
content, especially at low kappa numbers. However, the structure of the lignins
f k y ff g β–β
resinol substructures, while the other linkage β–O–4' alkyl- y β–5
phenylcoumarans), if present, are in much lower abundance. In adition, and as
already observed by Py-GC/MS, the lignins from the black liquors show a S/G ratio
higher than the residual lignins, which indicates that S-lignin units, which are
y f g β–O–4 alkyl-aryl ether structures, are preferentially
removed from the eucalypt during pulping and are being enriched in the black
liquors. It is known that eucalypt lignin resinol structure is made almost exclusively
by syringaresinol.26,29

Quantitation of the abundance of the main lignin inter-unit linkages, as well as

the abundance of the G and S lignin units, present in the different residual lignins
from the eucalypt pulps produced from the kraft, soda-AQ and soda-O2 processes,
was performed by integration of the volume contours of their cross-signals and was
referred to as per 100 aromatic units (Table 4). Interestingly, the pulps with higher
k k 50 35 g f β–O–4' alkyl-aryl
ether linkages, which were drastically reduced in the pulps with low kappa
numbers due to the increasing extent of delignification. The kraft and soda-AQ
residual lignins showed similar abundance of linkages while the soda-O2 residual
lignins showed significant differences. At kappa numbers 50 and 35, the abundance
f β-O-4´ ether linkages (A) in the residual lignin from soda-O2 pulps was
significantly lower than in the corresponding soda-AQ residual lignins. At kappa
number 15, the amounts of β-O-4´ ether linkages was similar in all residual lignins.
Therefore, it is concluded that soda-O2 is a less efficient delignification process
than kraft and soda-AQ processes, at least at high kappa number levels, requiring
higher extent of cleavage of th β-O-4' ether linkages to reach the same kappa
number level. This is in agreement with the Py-GC/MS results.

272
 V. RESULTADOS Y DISCUSIÓN

Figure 2. HSQC NMR spectra of: (a) MWL isolated from wood of eucalypt G1×UGL
hybrid; (b) residual lignin isolated from kraft pulp at kappa number 20; and (c) the
g k g w g xyg g δ C/δH
50-90/2.5-5 7 w g δC/δH 100-123/5.5-8.0). The main lignin
substructures identified are: (A β-O-4’ ; B) phenylcoumarane substructures
f y β-5’/α-O-4’ k g ; C f y β-β’/α-O-γ’/γ-O-α’
linkages; (D f y β-1’/α-O-4’ k g ; G) guaiacyl
units; (S) syringyl units; (S’) oxidized syringyl units bearing a carbonyl group at Cα.

273
V. RESULTADOS Y DISCUSIÓN

Table 4. Abundance of lignin inter-unit linkages (expressed as linkages per 100 aromatic
units) and S/G ratio in MWL from wood of the eucalypt G1×UGL hybrid, and in residual
and black liquor lignin from kraft, soda-AQ and soda-O2 pulping at different kappa
numbers.

MWL Residual lignins Black liquor lignins

Kraft Soda-AQ Soda-O2 Kraft Soda-AQ

Kappa number K20 K15 K50 K35 K20 K15 K50 K35 K15 K20 K15 K20 K15

-O-4' alkyl-aryl ether 90 16 14 56 47 18 12 41 29 12 1 0 0 0

-´ resinol 11 7 7 11 11 7 7 11 10 8 10 9 9 8
-5´ phenylcoumaran 5 0 0 1 1 0 0 1 1 1 0 0 0 0
β-1´ spirodienone 1 0 0 0 0 0 0 0 0 0 0 0 0 0
S/G ratio 2.8 5.1 4.4 5.9 5.9 5.2 4.8 4.6 4.5 4.6 8.6 8.2 7.9 8.5
 

31
P-NMR spectroscopy

Additional information about the occurrence of hydroxyl groups was obtained

from 31P-NMR spectroscopy. Figure 3 shows the 31P-NMR spectra of a
representative residual lignin (isolated from kraft pulp at kappa 20), and the
corresponding black liquor lignin. The 31P-NMR of the MWL isolated from wood
of the eucalypt G1×UGL hybrid is also shown for comparison. The contents of
aliphatic hydroxyl groups were calculated from the integration area in the 149.4 -
145.0 ppm region and are shown in Table 5. At kappa number 15 the removal of
aliphatic hydroxyl groups in the residual lignin side chain is similar in kraft, soda-
AQ and soda-O2 (62-63%). At earlier stages (kappa 50) the soda-O2 residual lignin
was 11% more depleted in aliphatic hydroxyl groups than the soda-AQ residual
lignin. In the black liquor lignins the removal was only different at kappa 15 (80%
removal in kraft lignin while 70% in soda-AQ lignin). This difference may be
related to the higher α-carbonyl content in kraft lignin as observed by Py-GC/MS,
and can influence in the chemical reactivity and physical characteristics of lignin
for further industrial applications. The use of phosphitylating reagent I (2-chloro-
1,3,2-dioxaphospholane) could give more detailed information about the side chain
oxidation pattern.30

274
 V. RESULTADOS Y DISCUSIÓN

Internal Aliphatic OH S-OH and condensed

standard OH
5-5´ biphenol (signal 13)
Carboxyl
Residual
carbohydrates G-OH Catechol

P (ppm)

Figure 3. 31P-NMR spectra of: (a) MWL isolated from wood of the eucalypt G1×UGL
hybrid; (b) kraft residual lignin at kappa number 20, and (c) the corresponding black liquor
lignin, processed with a line-broadening factor of 8 Hz.

During alkaline pulping, the content of carboxyl groups in the lignins increases
(Table 5). At kappa number 50 the formation of carboxyl groups in soda-O2
residual lignin was 60% higher than in soda-AQ residual lignin due to the oxidative
conditions prevailing in soda-O2. At kappa number 15 the carboxyl content was
very similar in kraft, soda-AQ and soda-O2 residual lignin. The black liquor lignins
recuperated from the kraft and soda-AQ cookings contained 200-400 % more
carboxyl groups with respect to their respective residual lignins. The active alkali
demand was higher in the soda-O2 cookings than in the kraft and soda-AQ
cookings (data from Suzano Papel e Celulose), which supports that more carboxyl
acid groups were formed during soda-O2 cooking.

During the chemical delignification of the eucalypt G1×UGL, the residual lignin
was enriched with phenolic groups (Table 5). In hardwoods, the phenolic groups in
C5-substituted guaiacyl units have similar chemical environments as the phenolic
groups in syringyl units, and their corresponding signals partially overlap. All
previous works reporting the phenolic groups of hardwoods by 31P-NMR expressed

275
V. RESULTADOS Y DISCUSIÓN

Table 5. Aliphatic -, phenol - (syringyl + guaiacyl + total condensed + catechol) and

carboxyl hydroxyl group content (mmol OH/g pure lignin) determined by quantitative 31P-
NMR in MWL isolated from wood of the eucalypt G1×UGL hybrid, residual and black
liquor lignins isolated from alkaline pulping at different kappa numbers. The purity of
MWL and residual lignins was set to 100 % and for the black liquor lignins it was
estimated from the sum of the Klason lignin and acid-soluble lignin content.

MWL Residual lignin Black liquor lignin

Kraft Soda-AQ Soda-O2 Kraft Soda-AQ
Kappa number K20 K15 K50 K35 K20 K15 K50 K35 K15 K20 K15 K20 K15
Aliphatic OH
4.69 2.20 1.69 3.33 2.80 2.20 1.73 2.81 2.55 1.75 1.67 0.92 1.66 1.43
( 0.026 )

Total phenolic OH
0.67 1.72 2.05 1.47 1.56 2.13 2.25 1.64 1.71 1.82 3.91 4.32 3.99 4.21
( 0.01)

Syringyl
0.21 0.75 0.85 0.86 0.69 1.01 1.13 0.87 0.79 0.79 2.38 2.67 2.35 2.39
( 0.03)

Guaiacyl
0.38 0.27 0.36 0.25 0.21 0.35 0.39 0.32 0.33 0.37 0.71 0.87 0.71 0.74
( 0.01)

Total condensed
0.07 0.64 0.78 0.34 0.62 0.72 0.69 0.43 0.57 0.64 0.82 0.78 0.93 1.08
( 0.04)

Total 5-5´ biphenol

0.04 0.20 0.28 0.10 0.14 0.20 0.19 0.12 0.18 0.20 0.26 0.41 0.28 0.25
(± 0.01)

5-5´ biphenol
0 0.10 0.10 0.05 0.09 0.07 0.07 0.05 0.04 0.03 0 0 0 0
signal 13 (± 0.01)

Catechol
0 0.06 0.07 0.02 0.03 0.04 0.04 0.02 0.02 0.02 0 0 0 0
( 0.01)

Carboxyl OH
0.005 0.25 0.27 0.09 0.09 0.20 0.26 0.14 0.18 0.24 1.04 0.83 1.02 0.93
( 0.01)

Total OH
5.37 4.17 4.01 4.89 4.44 4.40 4.24 4.59 4.44 3.81 6.62 6.07 6.67 6.57
( 0.03)

S/G
0.6 2.7 2.4 3.5 3.2 2.9 2.9 2.7 2.4 2.1 3.4 3.1 3.3 3.2
( 0.1)

the syringyl and condensed phenol groupsas their sum, except for Rovio et al.9 In
our work, the separation of both signals was achieved with an appropriate
deconvolution technique. Figure 4 shows the deconvolution result of the residual
kraft lignin sample from Figure 3. The deconvoluted signals 1-4 represent syringyl
phenolic groups,30 and the signals 5-7 and 9-15 correspond to C5-condensed

276
 V. RESULTADOS Y DISCUSIÓN

phenolic groups.9,27,30,31 Signal 8 might represent both types of phenolic groups. At

the final stage (kappa number 15), the aliphatic hydroxyl groups were removed in
almost equal extent in all alkaline cookings, but the enrichment of phenolic groups
was different (Table 5). The content of phenolic groups in residual kraft lignin at
kappa number 20 was 157 % higher respect to the MWL. The black liquor lignins
are even more enriched in phenolic groups. Between kappa number 50 and 15,
syringyl phenolic groups were still enriched in the kraft and soda-AQ residual
lignins, but in soda-O2 residual lignins they were slightly depleted (Table 5). The
further enrichment of syringyl phenolic groups in soda-O2 residual lignin is masked
by the reactivity towards active oxygen species, as an increasing in the content of
g v g f β-aryl ether linkages in the residual lignin
3,8
is still going on.

Deconvolution
result

1
3
13

8 2
15
7 9
4 14 12
5 6 10 11

P (ppm)

Figure 4. Deconvolution results of experimental 31P-NMR spectrum of residual kraft

lignin at kappa number 20, as the sum of individual deconvoluted signals. Signals 1-4 are
syringyl phenolic groups; signals 5-7 are condensed phenolic groups in Cα-C5´ aryl and C5-
C5´ diarylmethane structures; signal 8 are condensed phenolic groups in Cβ-C5´
dihydroxystilbene structures and/or syringyl phenolic groups in Cβ-C1´ dihydroxystilbene
structures; signals 9-13 are condensed phenolic groups in 5-5´ biphenol structures; signals
14-15 are condensed phenolic groups in 4-O-5´ structures.

277
V. RESULTADOS Y DISCUSIÓN

In contrast to residual kraft lignin from pine wood,3,32 no enrichment of non-

condensed guaiacyl phenolic groups was observed in eucalypt residual lignin. They
were depleted until kappa number 50-35 and then reached again the same content
as in native lignin at kappa number 15. By comparing the content of total and
condensed phenolic groups in residual kraft lignin from softwood,3,8 and our
eucalypt wood, it comes clear that the enrichment of condensed phenolic groups in
hardwoods residual lignins is drastically reduced as a consequence of the
preferential cleavage of syringyl units, which lowers the abundance of non-
condensed guaiacyl phenolic groups (the MWL has a phenolic S/G ratio of 0.6!),
preventing condensation reactions to occur. Condensed phenolic groups show a
strong recalcitrance towards oxygen bleaching.2,14,33 One of the main condensed
phenolic groups found in kraft residual lignins of softwood are diarylmethane and
o,p´-dihydroxystilbene structures.11 Some signals were observed from Cα-C5´ aryl
and C5-C5´ diarylmethane structures (signals 5-7), and Cβ-C5´ dihydroxystilbene
structures (signal 8, Figure 4 α-C5´ aryl structures result from the attack of C5
carban ions to the Cα position of quinone methide intermediates. C5-C5´
diarylmethane structures are formed by condensation between a non-condensed
guaiacyl phenolic unit and formaldehyde released from quinone methide
intermediates,11,30,31 however we concluded that the abundance of diarylmethane
structures in the residual lignins was relatively low as their corresponding CH2
cross signals were not observed in any HSQC spectrum, in agreement with
previous works.32 β-C5´ dihydroxystilbene structures result from formaldehyde
β-5´ phenylcoumarane structures. However, signal 8 might also
y gy g β-C1´ dihydroxystilbene structures. The signals
14 and 15 most probably represented 4-O-5´ structures.31 Other condensed phenolic
structures are 5-5´ biphenols which represent an important fraction of the
recalcitrant structures to posterior bleaching stages.34 These 5-5´ biphenol
structures are mainly formed from alkaline cleavage of dibenzodioxocin
structures.21,35 Figure 3 and 4 (signals 9-13) clearly show the presence of these
structures in residual and black liquor lignin from the kraft cooking.27,30 In all
residual lignins, two sharp signals were present at 141.61 and 141.66 ppm.
However, they were observable as only one signal when the line-broadening factor
was increased (signal 13, Figure 4). We suggest that these signals might
correspond to a biphenyl catechol structure based on the experimental data and
correlations from previous works,30 and indirect evidence from our experimental
data. The signals disappeared completely in the black liquor lignins.

278
 V. RESULTADOS Y DISCUSIÓN

CONCLUSIONS

The structural characteristics of MWL isolated from wood of the eucalypt

G1×UGL hybrid, residual lignins isolated from kraft, soda-AQ and soda-O2 pulps
at different kappa numbers, and lignins of the corresponding black liquors were
investigated. At kappa numbers 50 and 35 the soda-O2 residual lignins had
significantly w S/ w β-O-4´ ether linkages, more phenolic and
carboxylic hydroxyl groups than the kraft and soda-AQ residual lignins. These
results show that the soda-O2 process has more deconstructive character compared
to kraft and soda-AQ processes. It is thus a potential alkaline pretreatment process
for subsequent enzymatic hydrolysis for the sugar platform.

ACKNOWLEDGEMENTS

This study has been funded by the EU-project LIGNODECO (KBBE-244362),

the Spanish project AGL2008-00709, and the CSIC project 201040E075. Dr. Jorge
Rencoret thanks the CSIC for a JAE- O f g “
” -financed by Fondo Social Europeo (FSE). Finally, we
thank Suzano Papel e Celulose (Brazil) for providing the pulp and black liquor
samples from the alkaline cooking experiments and the corresponding additional
data.

LITERATURE CITED

(1) Santos, R. B.; Capanema, E. A.; Balakshin, M. Y.; Chang, H. M.; Jameel,
H. Effect of hardwoods characteristics on kraft pulping process: emphasis on lignin
structure. Bioresources 2011, 6, 3623–3627.

(2) Argyropoulos, D. S. Salient reactions in lignin during pulping and oxygen

bleaching: an overview. J. Pulp Pap. Sci. 2003, 29, 308–313.

(3) Froass, P. M.; Ragauskas A. J.; Jiang J. E. Nuclear Magnetic Resonance

Studies. 4. Analysis of residual lignin after kraft pulping. Ind. Eng. Chem. Res.
1998, 37, 3388–3394.

(4) Gustafsson, C.; Sjöström, K.; Wafa Al-Dajani, W. The influence of cooking
conditions on the bleachability and chemical structure of kraft pulps. Nordic Pulp
Pap. Res. J. 1999, 14, 71–81.

279
V. RESULTADOS Y DISCUSIÓN

(5) Ibarra, D.; del Río, J. C.; Gutiérrez, A.; Rodríguez, I. M.; Romero, J.;
Martínez, M. J.; Martínez, A. T. Chemical characterization of residual lignins from
eucalypt paper pulps. J. Anal. Appl. Pyrolysis 2005, 74, 116–122.

(6) Ibarra, D.; Chávez, M. I.; Rencoret, J.; del Río, J. C.; Gutiérrez, A.;
Romero, J.; Camarero, S.; Martinez, M. J.; Jiménez-Barbero, J.; Martínez, A. T.
Lignin modification during Eucalyptus globulus kraft pulping followed by totally
chlorine free bleaching: a two-dimensional nuclear magnetic resonance, Fourier
transform infrared, and pyrolysis-gas chromatography/mass spectrometry study. J.
Agric. Food Chem. 2007, 55, 3477–3499.

(7) Kuitunen, S.; Kalliola, A.; Tarvo, V.; Tamminen, T.; Rovio, S.; Liitiä, T.;
Ohra-aho, T.; Lehtimaa, T.; Vuorinen, T.; Alopaeus, V.. Lignin oxidation
mechanisms under oxygen delignification conditions. Part 3. Reaction pathways
and modelling. Holzforschung 2011, 65, 587–599.

(8) Moe, S. T.; Ragauskas, A. J. Oxygen delignification of high-yield kraft

pulp. Part 1: structural properties of residual lignins. Holzforschung 1999, 53, 416–
422.

(9) Rovio, S.; Kuitunen, S.; Ohra-aho, T.; Alakurtti, S.; Kalliola, A.;
Tamminen, T. Lignin oxidation mechanisms under oxygen delignification
conditions. Part 2: Advanced methods for the detailed characterization of lignin
oxidation mechanisms. Holzforschung 2011, 65, 575–585.

(10) Rencoret, J.; Marques, G.; Gutiérrez, A.; Jiménez-Barbero, J.; Martínez, A.
T.; del Río, J. C. Structural modifications of residual lignins from sisal and flax
pulps during soda/AQ pulping and TCF/ECF-bleaching. Ind. & Eng. Chem. Res.
2013, 52, 4695–4703.

(11) Gierer, J. Chemistry of delignification. Part 1. General concept and

reactions during pulping. Wood Sci. Technol. 1985, 19, 289–312.

(12) Gierer, J. Chemical aspects of kraft pulping. Wood Sci. Technol. 1980, 74,
241–266.

(13) Gellerstedt, G.; Gustafsson, K.; Northey, R. A. Structural changes in lignin

during kraft pulping. Part 8. Birch Lignins. Nordic Pulp Paper Res. J. 1988, 2, 87–
94.

280
 V. RESULTADOS Y DISCUSIÓN

(14) Argyropoulos, D. S.; Liu, Y. J. The role and fate of lignin´s condensed
structures during oxygen delignification. Pulp Paper Sci. 2000, 26, 107–113.

(15) Gellerstedt, G.; Majtnerova, A.; Zhang, L. Towards a new concept of lignin
condensation in kraft pulping. Initial results. C.R. Biologies 2004, 327, 817–826.

(16) TAPPI. TAPPI Test Methods, 1996-1997, Tappi Press, Atlanta, GA, 1996.

(17) Prinsen, P.; Gutiérrez, A.; Rencoret, J.; Nieto, L.; Jiménez-Barbero, J.;
Burnet, A.; Petit-Conil, M.; Colodette, J. L.; Martínez, A. T.; del Río, J. C.
Morphological characteristics and composition of lipophilic extractives and lignin
in Brazilian woods from different eucalypt hybrids. Ind. Crop. & Prod. 2012, 36,
572–583.

(18) Björkman, A. Studies on finely divided wood. Part I. Extraction of lignin

with neutral solvents. Sven. Papperstidn. 1956, 13, 477–485.

(19) Evtuguin, D. V.; Neto, C. P.; Silva, A. M. S.; Domingues, P. M.; Amado,
F. M. L.; Robert, D.; Faix, O. Comprehensive study on the chemical structure of
dioxane lignin from plantation Eucalyptus globulus wood. J. Agric. Food Chem.
2001, 49, 4252–4261.

(20) Ralph, J.; Hatfield, R. D. Pyrolysis-GC/MS characterization of forage

materials. J. Agric. Food Chem. 1991, 39, 1426–1437.

(21) Capanema, E. A.; Balakshin, M. Y.; Kadla, J. F. A comprehensive

approach for quantitative lignin characterization by NMR spectroscopy. J. Agric.
Food Chem. 2004, 52, 1850–1860.

(22) del Río, J. C.; Rencoret, J.; Marques, G.; Gutiérrez, A.; Ibarra, D.; Santos,
J. I.; Jiménez-Barbero, J.; Martínez, A. T.; Gutiérrez, A. Structural characterization
of the lignin from jute (Corchorus capsularis) fibers. J. Agric. Food Chem. 2009,
57, 10271–10281.

(23) Martínez, A. T.; Rencoret, J.; Marques, G.; Gutiérrez, A.; Ibarra, D.;
Jiménez-Barbero, J.; del Río, J. C. Monolignol acylation and lignin structure in
some nonwoody plants: a 2D NMR study. Phytochemistry 2008, 69, 2831–2843.

(24) Ralph, J.; Marita, J. M.; Ralph, S. A.; Hatfield, R. D.; Lu, F.; Ede, R. M.;
Peng, J.; Quideau, S.; Helm, R. F.; Grabber, J. H.; Kim, H.; Jimenez-Monteon, G.;
Zhang, Y.; Jung, H. -J. G.; Landucci, L. L.; MacKay, J. J.; Sederoff, R. R.;

281
V. RESULTADOS Y DISCUSIÓN

Chapple, C.; Boudet, A. M. Solution-state NMR of lignin. In Advances in

Lignocellulosics Characterization; Argyropoulos, D.S., Ed.; Tappi Press: Atlanta,
GA, 1999; pp 55–108.

(25) Ralph, J.; Landucci, L. L. NMR of lignins. In Lignin and Lignans; Heitner,
C., Dimmel, D. R., Schmidt, J. A., Eds.; CRC Press (Taylor & Francis Group):
Boca Raton, FL, 2010; pp 137–234.

(26) Rencoret, J. ; Gutiérrez, A. ; Nieto, L. ; Jiménez-Barbero, J. ; Faulds, C. B.;

Kim, H. ; Ralph, J. ; Martínez, A. T.; del Río, J. C. Lignin composition and
structure in young versus adult Eucalyptus globulus plants. Plant Physiol. 2011,
155, 667–682.

(27) Granata, A.; Argyropoulos, D. S. 2-Chloro-4,4,5,5-tetramethyl-1,3,2-

dioxaphospholane, a reagent for the accurate determination of the uncondensed and
condensed phenolic moieties in lignins. J. Agric. Food Chem. 1995, 43, 1538–
1544.

(28) Gierer, J.; Wannstrom, S. Formation of ether bonds between lignins and
carbohydrates during alkaline pulping processes. Holzforschung 1986, 40, 347–
352.

(29) Rencoret, J.; Marques, G.; Gutiérrez, A.; Ibarra, D.; Li, J.; Gellerstedt, G.;
Santos, J. I.; Jiménez-Barbero, J.; Martínez, A. T.; del Río, J. C. Structural
characterization of milled wood lignins from different eucalypt species.
Holzforschung 2008, 62, 514–526.

(30) Jiang, Z. H. Advances and applications of quantitative 31P-NMR for the

structural elucidation of lignin. PhD Thesis, McGill University: Montréal, Canada,
1997.

(31) Dyer, T. Elucidating the formation and chemistry of chromophores during

kraft pulping. PhD Thesis, Institute of Paper Science and Technology, University
of Wisconsin-Stevens Point: Atlanta, GA, USA, 1999.

(32) Gellerstedt, G.; Lindfors, E. Structural changes in lignin during kraft

pulping. Part 4. Phenolic hydroxyl groups in wood and kraft pulps. Svensk
Papperstidn. 1984, 15, R115–R118.

(33) Johansson, E.; Ljunggren, S. The Kinetics of lignin reactions during

oxygen bleaching, Part 4. The reactivity of different lignin model compounds and

282
 V. RESULTADOS Y DISCUSIÓN

the influence of metal ions on the rate of degradation. J. Wood Chem. & Technol.
1994, 14, 507–525.

(34) Ljunggren, S.; Johansson, E. The kinetics of lignin reactions during oxygen
bleaching. Part 3. The reactivity of 4-n-propyl guaiacol and 4,4'-di-n-propyl-6,6'-
biguaiacol. Holzforschung 1990, 44, 291–296.

(35) Karhunen, P.; Rummakko, P.; Sipila, J.; Brunow, G.; Kilpelainen, I.
Dibenzodioxocins; a novel type of linkage in softwood lignins. Tetrahedron Lett.
1995, 36, 169–170.

283
VI. CONCLUSIONES
 VI. CONCLUSIONES

En la presente Tesis se ha realizado un estudio exhaustivo de la composición

química de diversos materiales lignocelulósicos de interés industrial, tanto de
origen maderero (diferentes especies híbridas de eucalipto) como no maderero
(hierba elefante y paja de trigo). Se ha prestado especial atención a la composición
química de los compuestos extraíbles lipofílicos y a las ligninas, ya que estas
fracciones tienen un papel importante en el uso industrial de los materiales
lignocelulósicos. Finalmente, se han estudiado los cambios estructurales de la
lignina durante los procesos de deconstrucción alcalina. Las principales
conclusiones obtenidas se citan a continuación:

In the present PhD Thesis we have performed an exhaustive study of the chemical
composition of several lignocellulosic materials of industrial interest, including
woody (eucalypt hybrid species) and non-woody materials (elephant grass and
wheat straw). Special emphasis was paid to the chemical composition and
structural characteristics of their lipophilic extractives and lignins, as they play a
major role during their industrial use. Finally, we have studied the structural
modification of an eucalypt feedstock by different alkaline deconstruction
processes. The main conclusions obtained from the results are cited below:

1. La madera de eucalipto presentó un contenido más elevado en carbohidratos, y

un contenido más bajo en compuestos extraíbles, proteínas y cenizas, en
comparación con la hierba elefante o la paja de trigo. En principio, estas
características favorecen la utilización de eucalipto para la producción de pasta
de celulosa y de bioetanol.

The eucalypt woods presented slightly higher carbohydrate contents, and lower
contents of extractives, proteins and ash, with respect to elephant grass and
wheat straw. At first sight, this feature may favor the utilization of eucalypt
wood for the production of cellulose pulp and bioethanol.

2. La madera de las diferentes especies híbridas de eucalipto analizadas en esta

Tesis, presentaron un contenido en compuestos extraíbles lipofílicos similar
(0.2-0.3%). Los esteroles libres y conjugados (en forma de ésteres y glicósidos)
fueron los compuestos más abundantes. El eucalipto híbrido E. urophylla × E.
urophylla (U1×U2) presentó un contenido más bajo en compuestos esteroidales,
que son los principales responsables de la formación de depósitos de ´´pitch´´.
Por otra parte, las maderas de los eucaliptos presentaron un contenido en lignina
muy similar (27-28%). Además, la composición de sus ligninas fue similar

287
VI. CONCLUSIONES

(relación S/G 2.6-2.7), excepto para el híbrido G1×UGL que presentó una
relación ligeramente mayor (3.1), lo que en principio podría facilitar la
deslignificación y disminuir el consumo de reactivos durante los procesos de
cocción. Los principales enlaces entre las diferentes unidades de lignina fueron
éteres β-O-4´ (75-79% del total), y en menor proporción enlaces β-β´ resinol (9-
11%), β-5´ fenilcumarano (5%) y β-1´ espirodienona (3-4%).

The woods of the different eucalypt hybrid species analyzed presented a similar
content of lipophilic extractives (0.2-0.3%). Among them, free and conjugated
(as esters and glycosides) sterols were the most abundant compounds. The
extractives of E. urophylla × E. urophylla (U1×U2) wood contained the lowest
amount of these steroid compounds, which are among the main compounds
responsible for the formation of pitch deposits during the production of cellulose
pulp. On the other hand, the different eucalypt woods showed a similar lignin
content (27-28%). Moreover, the composition of their lignins was also similar
(S/G ratio of 2.6-2.7), except for the lignin of E. grandis × [E.urophylla × E.
globulus] (G1×UGL) wood, which presented a slightly higher S/G ratio (3.1),
which might favor the delignification efficiency during the cooking process. The
main inter-unit linkages present in the eucalypt lignins are alkyl-aryl β-O-4´
ether linkages (75-79% of total side linkages content), with minor proportions of
β-β´ resinol (9-11%), β-5´ phenylcoumarane (5%) and β-1´ spirodienone
linkages (3-4%).

3. El contenido en compuestos extraíbles lipofílicos de la hierba elefante se situó

alrededor de 0.9%. El análisis detallado por GC/MS indicó que están
compuestos principalmente por ácidos grasos (45%), compuestos esteroidales
(32%) y ésteres de glicerol (11%). También se encontraron, en cantidades
menores, compuestos alifáticos lineales como β-dicetonas (6%), ceras (3%), n-
alquil ferulatos (1%), alcoholes grasos (0.6%) y n-alcanos (0.6%), tocoferoles
(0.4%) y lignanos (0.4%). Se analizaron dos fracciones bien separadas en la
caña de hierba elefante (corteza y médula). La corteza representó el 84% del
peso total y la médula el 16%. El contenido de extraíbles lipofílicos en la
médula fue casi el doble que el contenido en la corteza, y contuvo relativamente
más compuestos esteroidales y menos ácidos grasos. Los alcoholes grasos, β-
dicetonas, n-alquil ferulatos y lignanos fueron más abundantes en la corteza.

The content of lipophilic extractives of elephant grass was around 0.9%. A

detailed GC/MS analysis indicated that fatty acids (45%), steroid compounds
(32%) and glycerol esters (11%) represented the main fraction of the lipophilic

288
 VI. CONCLUSIONES

extract. Minor amounts of linear aliphatic compounds, such as β-diketones

(6%), waxes (3%), n-alkyl ferulates (1%), fatty alcohols (0.6%) and n-alkanes
(0.6%), tocopherols (0.4%) and lignans (0.4%) were also found. Elephant grass
contained two well differentiated fractions, the cortex and pith, which
represented 84% and 16% of the total weight of the elephant grass material.
The lipophilic extractives content in the pith was almost twice as high as in the
cortex, and contained more steroid compounds and less fatty acids, β-diketones,
n-alkyl ferulates and lignans.

4. El contenido en lignina de la hierba elefante fue 20%. El análisis por pirólisis

reveló la presencia de ácidos hidroxicinámicos, y se observó que el ácido p-
cumárico está unido principalmente a la lignina, mientras que el ácido ferúlico
se une principalmente a los carbohidratos. La lignina presentó una relación
H:G:S de 3:40:57. El análisis por 2D-NMR demostró que la mayoría de las
unidades de la lignina están unidas por enlaces éter β-O-4′ (82% del total), y en
menor proporción por enlaces β-5´ fenilcumarano (8%), β-β′ resinol (2%) y β-1´
espirodienona (2%). Alrededor del 39% de las unidades en la lignina de la
corteza, y 55% en la lignina de la médula, estaban aciladas en la posición γ de la
cadena lateral. El análisis por DFRC demostró que el ácido p-cumárico acila el
grupo hidroxilo en el carbono γ, y predominantemente en unidades de siringilo.

The lignin content of elephant grass was ca. 20%. The analysis by Py-GC/MS
and Py-TMAH-GC/MS revealed the presence of p-hydroxycinnamic acids, with
p-coumaric acid predominantly bound to the lignin fraction and ferulic acid
predominantly bound to the carbohydrate fraction. The elephant grass lignin
presented a H:G:S ratio of 3:40:57. The 2D-NMR analysis showed that β-O-4′
ether linkages were the main inter-unit linkages (82% of total side chain
linkages), with minor amounts of β-5´ phenylcoumaranes (8%), β-β′ resinols
(2%) and β-1´ spirodienones (2%). The lignins from the cortex and pith were
partially acylated at the γ-carbon of the side-chain (39% and 55% acylations,
respectively). DFRC analyses confirmed that p-coumarate groups acylate the γ-
carbon of these lignins, and predominantly on syringyl units.

5. El contenido en extraíbles lipofílicos de la paja de trigo alcanzó el 2.0%,

predominando compuestos alifáticos lineales como ácidos grasos (25%),
alcoholes grasos (19%), ceras (11%), ésteres de glicerol (5%), n-alcanos (4%) y
n-aldehídos (1%). Se encontraron cantidades relativamente altas de β-dicetonas
(11%), siendo la 14,16-hentriacontano-diona el compuesto predominante. Los

289
VI. CONCLUSIONES

compuestos esteroidales también se presentaron en cantidades importantes

(25%), de los cuales el 14% fueron esteroles libres.

Wheat straw presented a high content of lipophilic extractives (2.0%), with a

predominance of linear aliphatic compounds, such as fatty acids (25%), fatty
alcohols (19%), waxes (11%), glycerol esters (5%), n-alkanes (4%) and n-
aldehydes (1%). β-diketones were also present in rather high amounts (11%),
with hentriacontane-14,16-dione being the major compound. Steroid compounds
accounted for around 25% of the lipophilic extracts, of which 14% were free
sterols.

6. El contenido en lignina de la paja de trigo fue 18%. Se estudió la lignina por Py-
GC/MS, 2D-NMR y DFRC. Al igual que en la hierba elefante, el ácido p-
cumárico está principalmente unido a la lignina, mientras que el ácido ferúlico
está unido principalmente a los carbohidratos. Los análisis de la lignina de la
paja de trigo indicaron una relación H:G:S de 6:64:30. El análisis por 2D-NMR
indicó que alrededor de 75% de las diferentes unidades estaban unidas por
enlaces éter β-O-4′, seguidos por enlaces β-5´ fenilcumarano (11%) y otros
enlaces minoritarios. Se observó que la lignina estaba parcialmente (10%)
acilada en el carbono γ de la cadena lateral, predominantemente con acetatos,
que acilan preferentemente unidades de guayacilo. Los p-cumaratos apenas se
detectaron por DFRC (que rompe selectivamente los enlaces β-O-4´), indicando
que pueden estar presentes en estructuras condensadas o terminales. Finalmente,
se identificó la presencia de tricina (un flavonoide) en la estructura de la lignina,
unida por enlaces β-O-4´ a unidades G. Este descubrimiento amplía el espectro
de posibles monolignoles que pueden acoplarse durante la lignificación en la
pared celular.

The lignin content of wheat straw was 18%. The composition and the structural
characteristics of the lignin from wheat straw lignin were analyzed by Py-
GC/MS, 2D-NMR and DFRC. As already observed in elephant grass, p-
coumaric was predominantly bound to the lignin fraction while ferulic acid was
predominantly bound to the carbohydrate fraction. The analyses of wheat straw
lignin revealed a H:G:S ratio of 6:64:30. HSQC analysis showed that β-O-4′
ether linkages were the main inter-unit linkages (75% of total side-chain
linkages), followed by β-5´ phenylcoumaranes (11%) and other linkages in
minor amounts. The lignin was partially acylated (10%) at the γ-carbons of the
lignin side-chain, predominantly with acetate groups acylating G units. p-
Coumarates, which also acylate exclusively the γ-carbon, were barely detected

290
 VI. CONCLUSIONES

by DFRC (that selectively cleaves β-ether bonds), indicating that they might be
present in condensed or terminal substructures. Finally, a new compound was
identified in wheat straw lignin by HMBC analysis, that corresponded to tricin
(a flavone), bound to a G-lignin unit by β-O-4´ ether linkages. The presence of
this novel lignin compound may extend the possible range of lignin monomers
which participate in the coupling reactions taking place during lignification in
the cell wall of grasses.

7. Se estudió la modificación de la estructura de la lignina durante la

deconstrucción química de la madera del eucalipto híbrido G1 × UGL mediante
diferentes cocciones alcalinas (kraft, sosa-AQ y sosa-O2). La composición y las
características estructurales de las ligninas residuales de las pastas y de la
lignina nativa de dicha madera fueron parecidas, mientras las ligninas
precipitadas de los correspondientes licores de cocción mostraron un grado de
degradación mucho mayor. A números kappa 50 y 35 las ligninas residuales de
las pastas sosa-O2 mostraron una relación S/G y una proporción de enlaces β-O-
4´ menor que las de las pastas sosa-AQ, igual que el peso molecular y el
contenido en grupos hidroxilos en la cadena lateral. A número kappa 15 las
características de las ligninas residuales de pastas kraft, sosa-AQ y sosa-O2
fueron muy parecidos, excepto el contenido en grupos fenólicos. El proceso
sosa-O2 degradó la lignina más extensivamente que el proceso sosa-AQ, y se
podría usar como pretratamiento alcalino de madera de eucalipto con vistas de la
obtención de bioetanol.

The structural modification of the lignin during the chemical deconstruction of

the G1×UGL eucalypt wood by different alkaline cooking processes (kraft,
soda-AQ and soda-O2) was studied. The composition and the structural
characteristics of the residual lignin from the different eucalypt pulps were
similar to the native lignin from the corresponding wood, while the lignins
precipitated from the corresponding cooking liquors showed a much higher
degradation degree. At kappa numbers 50 and 35, the residual lignins from
soda-O2 pulps presented a lower S/G ratio and a lower proportion of β-O-4´
ether linkages than the residual lignins from soda-AQ pulps, as well as their
corresponding molecular weights and the content of hydroxyl groups in the side-
chains. At kappa number 15, the residual lignins from kraft, soda-AQ and soda-
O2 pulps showed rather similar structural characteristics, except for the content
of phenolic groups which was significantly lower in the residual lignin from
soda-O2 pulp due to the oxidative conditions. The soda-O2 cooking process

291
VI. CONCLUSIONES

degraded the eucalypt lignin more extensively than the soda-AQ process, and
may be used as alkaline pretreatment for eucalypt feedstocks intended for
bioethanol production.

En conclusión, el estudio mediante técnicas analíticas avanzadas de las

características químicas y estructurales de los materiales lignocelulósicos, y su
modificación durante los procesos de deconstrucción química, contribuyen a
optimizar el aprovechamiento industrial de estos materiales.

In conclusion, the study by advanced analytical techniques of the chemical and

structural characteristics of lignocellulosic materials, and its modification during
chemical deconstruction processes, contribute to a more sustainable industrial use
of these materials.

292