Está en la página 1de 294

Valorizacin de diferentes cultivos lignocelulsicos

para la fabricacin de pasta de papel:


Caracterizacin qumica, modificacin estructural de sus
constituyentes orgnicos durante los procesos de
coccin y blanqueo y aplicaciones biotecnolgicas

Gisela Marques Silva


Sevilla, 2010
Valorizacin de diferentes cultivos lignocelulsicos para la
fabricacin de pasta de papel: Caracterizacin qumica,
modificacin estructural de sus constituyentes orgnicos
durante los procesos de coccin y blanqueo y aplicaciones
biotecnolgicas

Memoria que presenta

Gisela Marques Silva


para optar al ttulo de Doctor en Ciencias
Qumicas por la Universidad de Sevilla.
Sevilla, a 13 de Abril de 2010.
Valorizacin de diferentes cultivos lignocelulsicos para la
fabricacin de pasta de papel: Caracterizacin qumica,
modificacin estructural de sus constituyentes orgnicos
durante los procesos de coccin y blanqueo y aplicaciones
biotecnolgicas
Visado en Sevilla, a 13 de Abril de 2010

LOS DIRECTORES

Dr. D. Jos C. del Ro Andrade Dra. Da. Ana Gutirrez Surez


Investigador Cientfico del CSIC Investigador Cientfico del CSIC
IRNAS-CSIC IRNAS-CSIC

EL TUTOR

Dr. D. Alfonso Guiram Prez


Catedrtico de la Universidad de Sevilla

Memoria que presenta

Gisela Marques Silva


para optar al grado de Doctor en Ciencias
Qumicas por la Universidad de Sevilla.
DOCTOR D. LUIS CLEMENTE SALAS, DIRECTOR DEL INSTITUTO DE
RECURSOS NATURALES Y AGROBIOLOGA DE SEVILLA DEL
CONSEJO SUPERIOR DE INVESTIGACIONES CIENTFICAS

CERTIFICA: Que la presente Memoria de Investigacin titulada Valorizacin


de diferentes cultivos lignocelulsicos para la fabricacin de pasta de papel:
Caracterizacin qumica, modificacin estructural de sus constituyentes
orgnicos durante los procesos de coccin y blanqueo y aplicaciones
biotecnolgicas, presentada por Gisela Marques Silva para optar al grado de
Doctor en Ciencias Qumicas, ha sido realizada en el Departamento de
Biotecnologa Vegetal, bajo la direccin de los Drs. D. Jos C. del Ro Andrade
y Da. Ana Gutirrez Surez, reuniendo todas las condiciones exigidas a los
trabajos de Tesis Doctorales.

En Sevilla, a 13 de Abril de 2010


AGRADECIMIENTOS

Este trabajo se ha llevado a cabo en el Instituto de Recursos Naturales y


Agrobiologa de Sevilla (IRNAS-CSIC). Ha sido financiado por una beca I3P de
postgrado del CSIC, por una beca FPI del Ministerio de Educacin y Ciencia, por los
proyectos nacionales de investigacin AGL2005-01748 y AGL2008-00709, y por el
proyecto europeo NMP2-CT-2006-026456.

Quiero expresar mi sincero agradecimiento a las personas que tanto directa como
indirectamente han hecho posible la realizacin de esta Tesis:

A los Dres. Jos Carlos del Ro y Ana Gutirrez, directores de esta Tesis, por todo
lo que me han aportado tanto a nivel cientfico, por sus conocimientos y enseanzas,
como a nivel personal, por la confianza en su trato personal y por sus consejos, y por
estar siempre que los he necesitado. Por su esfuerzo y dedicacin en esta Tesis.

Al Prof. ngel T. Martnez, del Centro de Investigaciones Biolgicas (CIB-CSIC,


Madrid), por ofrecerme la posibilidad de realizar en el CIB algunas estancias breves,
aportndome numerosos y valiosos conocimientos, y por haber seguido el desarrollo
de esta Tesis.

Al Prof. Dmitry Evtuguin, de la Universidad de Aveiro (Aveiro, Portugal), por


ofrecerme la posibilidad de realizar dos estancias en su grupo de investigacin, por
acogerme como un miembro ms de su familia y por aportarme valiosos
conocimientos. Por su apoyo y consejos, y su excelente trato personal. Tambin quiero
agradecer al Dr. Jos Antonio Gamelas y a la Dra. Paula Pinto por su apoyo durante
las estancias en la Universidad de Aveiro.

Al Prof. Alfonso Guiram, Catedrtico de la Universidad de Sevilla, tutor de esta


Tesis, por toda su ayuda en la parte burocrtica.

A mi compaero de laboratorio durante los primeros aos de la Tesis y amigo, el


Dr. Jorge Rencoret, por ayudarme siempre que lo he necesitado tanto en el
laboratorio como fuera, por su alegra, dndole siempre una vida muy suya al
laboratorio sin dejar de lado la profesionalidad.

A mis compaeras la Dra. Isabel Mara Rodrguez y Setefilla Molina, con las que
he coincidido en el inicio y mediados de esta Tesis, y a mis compaeros Pepijn
Prinsen, Alejandro Rico y Esteban Babot con los que he coincidido en el final de esta
Tesis. Un agradecimiento muy especial a Esteban por su apoyo y los buenos
momentos compartidos.

A Da. Trinidad Verdejo por hacer las pirlisis de mis numerosas muestras.
Al Prof. Jess Jimnez-Barbero, al Dr. Iaki Santos y a Lidia Nieto del CIB-
CSIC, por sus mltiples anlisis de NMR.

A Gerardo Artal (CELESA) por suministrarme las muestras de diversas fibras y sus
pastas, al Dr. Javier Romero (ENCE) por las pastas de eucalipto y al Dr. Manuel J.
Daz Blanco (Universidad de Huelva) por las muestras de caa comn y tagasaste.

Al Dr. Garca Hortal (UPC, Terrassa) por las imgenes proporcionadas de las
fibras elementales de las muestras de lino, camo, kenaf, yute, sisal y abac que se
muestran en la seccin de Material y Mtodos de esta Tesis.

A mis compaeros del IRNAS, Roco, Mari Trini, Agi, Alegra, Mara Fernanda,
Ftima, Jos Mara, Antonio y Jaime, que me acompaaron durante el inicio de esta
Tesis, y en particular a Ftima Sopea que tambin me acompa durante casi toda
la Tesis brindndome muy buenos momentos y consejos y por estar siempre all
incluso durante su post-doc en el extranjero.

A los compaeros del CIB, Mario, Yuta, Mara, ngeles, Miguel, Elvira, Davinia,
Vero, Helena, Aitor, Eva y Beatriz, y en particular a Celia Mndez por su apoyo en
mi primera estancia en el CIB y por los buenos momentos brindados. Quiero
agradecer tambin al Dr. Javier Ruiz-Dueas por su apoyo en las estancias
realizadas en el CIB.

A Mara Jess Ortega, madre de Aitor del CIB, aunque slo la conozco por
Internet, por sus e-mails y por proporcionarme una de las fotos que se muestra en la
seccin de Introduccin de esta Tesis.

Y por ltimo a mi hermana, a mi madre y a Augusto por el apoyo brindado


aunque estn lejos. A Dani por su paciencia, por escucharme hablar de experimentos
que le son completamente ajenos y por estar ah cuando ms lo necesito, as como a
su familia de Sevilla que desde que los conozco me han brindado todo su apoyo.

Gracias a todos y gracias tambin, tan slo por su existencia, a una nueva
personita subacutica que ahora llevo dentro
ABREVIATURAS

AQ Antraquinona
ABTS 2,2-azinobis(3-etilbenzotiazolin-6-sulfonato)
BSTFA N,O-bis-(trimetilsilil)-trifluoroacetamida
GC Desplazamiento qumico del carbono
GH Desplazamiento qumico del protn
CED Cobre (II)-etilendiamina
COSY Espectroscopia de correlacin (Correlation Spectroscopy)
DBO Demanda biolgica de oxgeno
DCM Diclorometano
DMAC N,N-dimetilacetamida
DMSO Dimetilsulfxido
DTT Ditiotreitol
2D-NMR Espectroscopa de Resonancia Magntica Nuclear bidimensional
3D-NMR Espectroscopa de Resonancia Magntica Nuclear tridimensional
ECF Secuencia de blanqueo libre de cloro elemental (elemental
chlorine free)
FAO Organizacin de las Naciones Unidas para la Agricultura y
Alimentacin (The Food and Agriculture Organization of the
United Nations)
FID Detector de ionizacin de llama (flame ionization detector)
G Unidad guayacilpropano (o guayacilo)
GC Cromatografa de gases (gas chromatography)
GC/MS Cromatografa de gases/espectrometra de masas (gas
chromatography/mass spectrometry)
H Unidad 4-hidroxifenilpropano (o 4-hidroxifenilo)
HBT 1-Hidroxibenzotriazol
HPLC Cromatografa lquida de alta resolucin
HSQC Espectroscopa 2D de correlacin heteronuclear de cuanto simple
(heteronuclear single-quantum correlation)
HexA cidos hexenurnicos
ICP-OES Espectrometra de emisin ptica con plasma acoplado
inductivamente (inductively coupled plasma-optical emission
spectrometry)
ID Dimetro interno (internal diameter)
IK ndice Kappa
IPTG Isopropil tio--D-galactopiransido
ISO Organizacin Internacional para la Estandarizacin,
Documentacin e Informacin (International Organization for
Standardization)
ITD Detector de trampa de iones (ion trap detector)
Ligninaox Productos de degradacin oxidativa de la lignina
LiP Lignina peroxidasa
Mw Masa molecular (molecular weight)
MnP Manganeso peroxidasa
MWL Lignina de madera molida (milled wood lignin)
NMR Espectroscopa de Resonancia Magntica Nuclear (nuclear
magnetic resonance)
O Etapa de deslignificacin con oxgeno (en secuencia de blanqueo)
PCA cido p-cumrico
POM polioxometalato
POM 1- Primera etapa POM en los ensayos de deslignificacin
-POM 2- Segunda etapa POM en los ensayos de deslignificacin
POMox Polioxometalato oxidado
POMred Polioxometalato reducido
PoP Doble etapa de blanqueo con perxido de hidrgeno, la primera
bajo oxgeno presurizado
ppb Partes por billn
ppm Partes por milln
Py-GC/MS Pirlisis acoplada a cromatografa de gases/espectrometra de
masas (pyrolysis-gas chromatography/mass spectrometry)
Q Etapa de quelato (en secuencia de blanqueo)
rpm Revoluciones por minuto
S Unidad siringilpropano (o siringilo)
SPE Extraccin en fase slida (solid phase extraction)
TAPPI Technical Association of the Pulp and Paper Industry
TCF Secuencia de blanqueo totalmente libre de cloro (totally chlorine
free)
TMAH Hidrxido de tetrametilamonio
TMP Pasta termomecnica (thermomechanical pulp)
TMSD Trimetilsilildiazometano
TMS Trimetilsililo
TOCSY Espectroscopia de correlacin total (Total Correlation
Spectroscopy)
U Unidad de actividad enzimtica
UV/VIS Espectroscopa de ultravioleta/visible
VP Peroxidasa verstil
 Coeficiente de extincin molar
NDICE

RESUMEN ..........................................................................................................1

1. INTRODUCCIN.........................................................................................5

1.1. CULTIVOS LIGNOCELULSICOS .....................................................5


1.1.1. Fibras procedentes de cultivos madereros ......................................5
1.1.2. Fibras procedentes de cultivos agrcolas ........................................7
1.2. ESTRUCTURA Y COMPOSICIN QUMICA DE LOS
MATERIALES LIGNOCELULSICOS................................................9
1.2.1. Celulosa.........................................................................................10
1.2.2. Hemicelulosas ...............................................................................11
1.2.3. Lignina ..........................................................................................13
1.2.4. Componentes de bajo peso molecular ..........................................20
1.3. UTILIZACIN DE CULTIVOS LIGNOCELULSICOS:
PRODUCCIN DE PASTA DE CELULOSA .....................................23
1.3.1. Procesos de pasteado ....................................................................24
Procesos mecnicos ...................................................................24
Procesos qumicos .....................................................................24
1.3.2. Procesos de blanqueo....................................................................26
1.4. PROBLEMTICA PLANTEADA POR LA PRESENCIA DE
LIGNINA Y LPIDOS EN LA PRODUCCIN DE PASTA DE
CELULOSA. ..........................................................................................28
1.5. BIOTECNOLOGA EN EL SECTOR DE LA PASTA DE
CELULOSA. ..........................................................................................29
1.5.1. Degradacin enzimtica de la lignina...........................................30
1.5.2. Degradacin enzimtica de lpidos: Control del pitch .................33

2. OBJETIVOS ................................................................................................37

3. MATERIAL Y MTODOS .......................................................................41

3.1. MATERIALES ......................................................................................41


3.1.1. Cultivos lignocelulsicos..............................................................41
Lino ...........................................................................................41
Camo .....................................................................................42
Kenaf .........................................................................................43
Yute ...........................................................................................44
Sisal ...........................................................................................45
Abac ........................................................................................46
Curau .......................................................................................47
Caa comn...............................................................................47
Tagasaste...................................................................................47
3.1.2. Pastas de papel ..............................................................................48
Pastas de fibras no madereras ...................................................48
Pastas de fibras madereras ........................................................48
3.1.3. Enzimas y mediadores ..................................................................49
Lipoxigenasas ...........................................................................49
Peroxidasas................................................................................49
Polioxometalatos.......................................................................50
3.2. MTODOS ANALTICOS ...................................................................51
3.2.1. Aislamiento y anlisis de los compuestos lipoflicos de las
fibras y pastas............................................................................51
Fraccionamiento de los compuestos extrables
lipoflicos mediante SPE...........................................................51
Mtodos de derivatizacin de los compuestos extrables
lipoflicos ..................................................................................53
Anlisis de los extractos lipoflicos mediante GC y GC/MS ........53
3.2.2. Aislamiento y anlisis de la lignina de las fibras y pastas............54
Determinacin del contenido en lignina ...................................54
Aislamiento de la lignina de las fibras......................................55
Anlisis de la lignina mediante Py-GC/MS..............................55
Anlisis de la lignina mediante DFRC .....................................56
Anlisis de la lignina mediante 2D-NMR ................................59
3.2.3. Aislamiento y anlisis de las hemicelulosas de las fibras y
pastas.........................................................................................62
Preparacin de la holocelulosa y aislamiento de los
xilanos ......................................................................................62
Anlisis de azcares neutros tras hidrlisis cida....................62
Anlisis de azcares neutros y cidos urnicos tras
metanolisis cida ......................................................................63
Determinacin del peso molecular de los xilanos
mediante SEC............................................................................63
Anlisis de la estructura de los xilanos mediante NMR ..........64
Determinacin del contenido en cidos hexenurnicos ...........64
3.2.4. Otros anlisis.................................................................................65
Determinacin de la fraccin hidrosoluble de las fibras ..........65
Determinacin del contenido en cenizas de las fibras..............65
Anlisis de metales y otros elementos en las fibras .................66
3.2.5. Tratamientos enzimticos de las pastas ........................................66
Tratamientos con lipoxigenasas................................................66
Tratamientos con POM y peroxidasa verstil...........................67
Determinacin de las propiedades de las pastas .......................69
Determinacin de la blancura ISO........................................69
Determinacin del ndice kappa ...........................................69
Determinacin de la viscosidad intrnseca ...........................70
Determinacin del contenido en cidos hexenurnicos .......72

4. REFERENCIAS ..........................................................................................75

5. RESULTADOS Y DISCUSIN.................................................................91

Publicacin I: Marques G., Rencoret J., Gutirrez A., del Ro J.C.


(2010) Evaluation of the chemical composition of different non-
woody plant fibers used for pulp and paper manufacturing. The
Open Agriculture Journal (in press)......................................................93
Publicacin II: del Ro J.C., Marques G., Rencoret J., Martnez
A.T. and Gutirrez A. (2007) Occurence of naturally acetylated
lignin units. Journal of Agricultural and Food Chemistry, 55,
5461-5468. ...........................................................................................111
Publicacin III: del Ro J.C., Rencoret J., Marques G., Gutirrez
A., Ibarra D., Santos J.I., Jimnez-Barbero J., Zhang L. and
Martnez 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.............127
Publicacin IV: Marques G., Gutirrez A. and del Ro J.C. (2007)
Chemical characterization of lignin and lipophilic fractions from
leaf fibers of curaua (Ananas erectifolius). Journal of
Agriculture and Food Chemistry, 55, 1327-1336................................151
Publicacin V: Coelho D., Marques G., Gutirrez A., Silvestre
A.R.D. and del Ro J.C. (2007) Chemical characterization of the
lipophilic fraction of Giant reed (Arundo donax) fibers used for
pulp and paper manufacturing. Industrial Crops and Products,
26, 229-236.. ........................................................................................173
Publicacin VI: Marques G., Gutirrez A. and del Ro J.C. (2008)
Chemical composition of lignin and lipids from tagasaste
(Chamaecytisus proliferus spp. palmensis). Industrial Crops
and Products, 28, 29-36. .....................................................................187
Publicacin VII: Marques G., del Ro J.C. and Gutirrez A. (2010)
Lipophilic extractives from several nonwoody lignocellulosic
crops (flax, hemp, sisal, abaca) and their fate during alkaline
pulping and TCF/ECF bleaching. Bioresource Technology, 101,
260-267.................................................................................................203
Publicacin VIII: Marques G., Gutirrez A., del Ro J.C. and
Evtuguin D.V. (2010) Acetylated heteroxylan from Agave
sisalana and its behavior in alkaline pulping and TCF/ECF
bleaching. Carbohydrate Polymers, (doi: 10-
1016/j.carbpol.2010.02.043)................................................................221
Publicacin IX: Marques G., Gamelas J.A., Ruiz-Dueas F.J., del
Ro J.C., Evtuguin D.V., Martnez A.T. and Gutirrez A. (2010)
Delignification of eucalypt kraft pulp with manganese-
substituted polyoxometalate assisted by fungal versatile
peroxidase. Bioresource Technology, 101, 5935-5940. .....................241
Publicacin X: Marques G., Molina S., Babot E.D., Lund H., del
Ro J.C. and Gutirrez A. Exploring the potential of a fungal
manganese lipoxygenase to remove lipophilic extractives from
paper pulps. Bioresource Technology (in preparation).......................255

6. CONCLUSIONES .....................................................................................273

7. ANEXOS ....................................................................................................277
RESUMEN

La presente Tesis plantea el estudio de la composicin qumica de los


principales constituyentes de diferentes cultivos lignocelulsicos utilizados
como materia prima para la fabricacin de pastas de celulosa de alta calidad,
poniendo especial nfasis en la composicin de la fraccin lipoflica
(responsable de la formacin de los denominados depsitos de pitch) y de la
lignina (cuya composicin y estructura influyen decisivamente en el proceso de
deslignificacin), as como en la composicin de las hemicelulosas. Entre los
materiales estudiados se incluyen fibras no madereras del tallo de varias
angiospermas dicotiledneas, tales como lino (Linum usitatissimum), kenaf
(Hibiscus cannabinus), camo (Cannabis sativa) y yute (Corchorus
capsularis), as como fibras procedentes de hojas de angiospermas
monocotiledneas como sisal (Agave sisalana), abac (Musa textilis) y curau
(Ananas erectifolius). Otras fibras estudiadas fueron las procedentes de la caa
comn (Arundo donax) y de podas de rboles de tagasaste (Chamaecytisus
proliferus spp. palmensis). Se estudi tambin la evolucin de los constituyentes
de los materiales lignocelulsicos durante la produccin de pasta de papel. Para
ello, se seleccionaron diversas pastas de celulosa a lo largo de los procesos de
pasteado (coccin sosa-antraquinona) y de blanqueo (procesos TCF y ECF).
Finalmente, se ensayaron dos procedimientos biotecnolgicos basados en la
utilizacin de enzimas fngicas para la eliminacin tanto de lignina como de
lpidos residuales en pastas de celulosa.

Los resultados obtenidos muestran que las diferentes materias primas


estudiadas se caracterizan, en general, por un alto contenido en polisacridos y
un bajo contenido en lpidos y lignina, lo que las hace, en principio, favorables
para la produccin de pasta de celulosa. Los compuestos lipoflicos presentes en
las fibras, analizados por GC y GC/MS, incluyen principalmente cidos grasos,
hidroxicidos, alcoholes, ceras, alcanos y esteroles libres y conjugados (en
forma de steres y glicsidos), entre otros. Los anlisis indican que el contenido
y composicin de las diferentes clases de lpidos vara considerablemente entre
las distintas fibras. Adems, las diferentes clases de lpidos muestran distinto
comportamiento durante los procesos de coccin y blanqueo. As, las ceras se
hidrolizan durante la coccin alcalina mientras que los cidos grasos se
disuelven. Por el contrario, los alcanos, alcoholes grasos, esteroles y
triterpenoles, hidrocarburos esteroidales, cetonas y glicsidos de esteroles tienen
baja solubilidad en agua y son difciles de eliminar de la pasta, por lo que
sobreviven a la coccin. Se observ que entre los compuestos que sobreviven a
la coccin, los esteroles libres se eliminan durante el blanqueo ECF pero resisten
al blanqueo TCF, mientras que los glicsidos de esteroles se eliminan tanto en el
blanqueo TCF como ECF. Finalmente, mientras que los cidos grasos

1
insaturados se eliminan durante los procesos de blanqueo ECF y TCF, los cidos
grasos saturados, as como los alcanos y alcoholes grasos sobreviven a estas
secuencias de blanqueo.

En cuanto a las ligninas, su estructura y composicin se estudi tanto por


mtodos degradativos (pirlisis-GC/MS y DFRC) como espectroscpicos (2D-
NMR). Los anlisis mostraron un predominio de unidades de tipo siringilo (S)
en el caso de las fibras liberianas de kenaf y yute, as como en todas las fibras de
hojas (sisal, abac y curau). Por el contrario, las fibras de camo, lino y caa
comn mostraron un predominio de unidades de tipo guayacilo (G). Esto fue
especialmente evidente en las ligninas de lino y camo, con una relacin S/G
de 0,1. La mayor relacin S/G de las ligninas de kenaf, yute, sisal y abac hace
que estas fibras sean ms fciles de deslignificar a causa del menor grado de
condensacin de la lignina, a pesar de tener un mayor contenido en lignina. Los
principales enlaces entre las unidades de lignina fueron de tipo aril-ter -O-4
en todas las fibras estudiadas. Tambin se observaron enlaces condensados -
5/-O-4 (fenilcumarano), - (resinol) y -1/-O- (espirodienona). La
mayor proporcin de enlaces no condensados -O-4 se encontr en las ligninas
de kenaf, sisal, abac y curau, las cuales al tener tambin mayor proporcin de
unidades S son ms fcilmente deslignificables. Por otro lado, en las ligninas de
kenaf, sisal, abac y curau se encontraron unidades aciladas (con acetatos y/o
p-cumaratos) en el carbono  de la cadena lateral y predominantemente sobre
unidades S. Se demostr que la acilacin tiene lugar a nivel de monmero y que
el sinapil acetato, y otros monmeros acilados, se comportan como autnticos
monmeros de la lignina. Se demostr tambin, que el nivel de acilacin de la
lignina estaba relacionado con un alto contenido en unidades S y enlaces -O-4,
as como con un menor contenido en enlaces -.

Tambin se estudi la composicin qumica de las hemicelulosas y las


modificaciones de las mismas durante los procesos de pasteado y blanqueo. El
estudio de las hemicelulosas tiene importancia debido a que los polisacridos se
disuelven y/o degradan parcialmente durante el pasteado, lo que afecta al
rendimiento del proceso y a la calidad de las pastas de celulosa. Los resultados
mostraron que las hemicelulosas de las fibras liberianas presentan una mayor
variabilidad en cuanto a su composicin en azcares neutros que las fibras
procedentes de hojas. As, mientras que en las fibras de lino y camo
predominan la manosa y la galactosa, en el kenaf y yute el monosacrido
predominante es la xilosa. Por otro lado, en todas las fibras de hojas estudiadas
(sisal, abac y curau) se observ un predominio de la xilosa. Un estudio en
profundidad de la estructura de las hemicelulosas de sisal mostr que estn
constituidas fundamentalmente por un glucuronoxilano acetilado cuya cadena
principal est formada por unidades de -D-xilopiranosa parcialmente
ramificada con residuos glucuronosilos. Esta hemicelulosa sufre una

2
despolimerizacin y desacetilacin significativa durante el proceso de pasteado.
Los grupos acetilo residuales que quedaban en la pasta cruda se eliminaron
completamente tras el blanqueo.

Finalmente, se estudiaron dos procedimientos biotecnolgicos basados en el


uso de enzimas fngicas para la eliminacin de la lignina residual de pastas as
como de los compuestos extrables lipoflicos responsables de la formacin de
depsitos de pitch durante el proceso de fabricacin de pasta de papel. Estos
procedimientos incluyeron la utilizacin de un sistema compuesto de un
polioxometalato y una enzima de tipo peroxidasa producida por el hongo
Pleurotus eryngii, y una lipoxigenasa producida por el hongo Gaeumannomyces
graminis. Los resultados obtenidos mostraron la eficacia del sistema
polioxometalato-peroxidasa para eliminar la lignina residual de la pasta y de la
lipoxigenasa para degradar parte de los compuestos lipoflicos responsables de
la formacin de los depsitos de pitch.

La presente Tesis incluye los siguientes apartados: i) una introduccin general


sobre los cultivos lignocelulsicos, su inters industrial, su estructura y
composicin, y los procesos utilizados para la produccin de pasta de celulosa,
as como los principales problemas que plantean algunos de sus constituyentes y
algunas soluciones biotecnolgicas a estos problemas; ii) los objetivos
perseguidos en la Tesis; iii) una descripcin detallada de los materiales
estudiados y los mtodos analticos empleados; iv) las referencias citadas en el
primer y tercer apartado; v) los resultados obtenidos y su discusin, que se
presentan en forma de publicaciones; vi) las principales conclusiones; y vii) una
lista de tablas que se muestran como Anexos.

3
1

Lino (Linum usitatissimum)


1. Introduccin

INTRODUCCIN

1.1. CULTIVOS LIGNOCELULSICOS


Los cultivos lignocelulsicos incluyen especies tanto de origen agrcola como
forestal y poseen un gran inters industrial. Entre los principales usos de los
cultivos lignocelulsicos se encuentra la produccin de pasta de celulosa. Por
otro lado, estos cultivos presentan un gran potencial como materia prima en el
contexto de las futuras biorrefineras para la produccin de biocombustibles y
otros productos de inters, como alternativa al petrleo.

La principal fuente de fibra de celulosa virgen utilizada actualmente en la


fabricacin de pasta de celulosa la constituyen los cultivos de fibras madereras,
mientras que las fibras no madereras se utilizan en menor proporcin. La amplia
disponibilidad y concentracin de madera en zonas de fcil acceso, el elevado
contenido en fibras, el coste de manejo, transporte y facilidad de
almacenamiento, as como la estabilidad de la materia prima y su
comportamiento durante el proceso de obtencin de celulosa, han apoyado el
uso de la misma en la industria de la pasta de papel. Sin embargo, existe en la
actualidad un renovado inters en el uso de plantas de origen no maderero
debido a, entre otras razones, la gran disponibilidad de residuos agrcolas. stos
constituyen una fuente abundante de fibras de bajo coste, siendo a veces la nica
fuente aprovechable de fibras en determinadas zonas geogrficas,
principalmente en pases en vas de desarrollo. La gran variedad de
caractersticas, dimensiones fibrosas y composicin qumica de estas fibras les
confieren un gran potencial como materias primas (Garca Hortal 2007).
Adems, en los pases desarrollados, se utilizan para la fabricacin de pastas de
celulosa para papeles especiales.

1.1.1. Fibras procedentes de cultivos madereros


Las fibras madereras provienen de especies vegetales que desarrollan un
tronco donde se acumulan preferentemente las mejores fibras. Las conferas
constituyen el primer cultivo forestal a escala mundial para la obtencin de pasta
de papel, aunque tambin existe un importante mercado de pastas de frondosas
(Figura 1).

Las conferas presentan fibras largas (3 a 5 mm), que son ptimas para la
fabricacin de papeles de elevada resistencia mecnica. Las conferas, en
trminos econmicos generales, son ms valiosas que las frondosas, ya que sus
troncos son ms largos y rectos, su madera es uniforme, ligera y blanda, por lo
que es ms fcil de trabajar, y presentan una mayor proporcin de elementos

5
1. Introduccin

fibrosos que son ms adecuados para la mayora de las calidades papeleras


(Garca Hortal 2007). Las principales conferas usadas para la fabricacin de
pasta de papel son la Picea y el pino.

La madera de frondosas, por otro lado, es una madera ms dura, de fibras


cortas (entre 0,75 y 2 mm) que dan lugar a pastas menos uniformes. El papel
fabricado con maderas de frondosas es ms dbil que los fabricados con maderas
de conferas pero su superficie es ms lisa, y por lo tanto, es mejor para papel de
escritura. Otra de las ventajas es que el crecimiento de las especies de frondosas
utilizadas para la fabricacin de pasta de papel es ms rpido que el de las
conferas, dando lugar a mayor cantidad de fibra en menos tiempo. Las
principales frondosas utilizadas en el sector papelero son el eucalipto, el chopo y
el abedul.

Figura 1. Ejemplos de especies madereras usadas para la produccin de pasta de celulosa,


incluyendo conferas como la Picea (izquierda) y frondosas como el abedul (derecha).

6
1. Introduccin

1.1.2. Fibras procedentes de cultivos agrcolas


Las fibras procedentes de cultivos agrcolas constituyen una excelente materia
prima alternativa a las fibras madereras para la produccin de pasta de celulosa.
Uno de los hechos apremiantes que conducen a la utilizacin de materia prima
no maderera es su conocida abundancia que sobrepasa la utilizacin actual.

En general, las fibras de plantas no madereras tienen una estructura menos


densa y ms porosa, lo que implica un menor requerimiento de energa y
productos qumicos para la separacin de las fibras durante la produccin de
pasta de papel. Adems, presentan ciclos de crecimiento ms cortos, alcanzando
la madurez ms rpidamente que las especies madereras y en muchos casos los
rendimientos de pasta obtenidos son mayores (Tabla 1). Algunas pastas de fibras
largas no madereras, tienen propiedades superiores a las mejores pastas del
mercado de conferas, pues son extremadamente resistentes. El principal
inconveniente de este tipo de materias primas es que la mayora slo estn
disponibles en ciertas pocas del ao.

Tabla 1. Rendimiento promedio de algunas materias primas (Pierce 1991).

Rendimiento Rendimiento
Especies materia seca (t/ha) Pasta (t/ha)
Trigo 2,5 1,1
Avena 1,6 0,7
Centeno 2,2 1,1
Arroz 3,0 1,2
Caa de azcar (bagazo) 9,0 4,2
Bamb 4,0 1,6
Miscanthus sinensis 12,0 5,7
Canary grass 6,0 3,0
Caa comn 9,0 4,3
Kenaf 15,0 6,5
Camo 12,0 6,7
Frondosa de zona templada (abedul) 3,4 1,7
Frondosa de crecimiento rpido (Eucalyptus) 15,0 7,4
Confera escandinava 1,5 0,7
Confera de crecimiento rpido 8,6 4,0

7
1. Introduccin

Las fibras no madereras se pueden clasificar en tres categoras: i) fibras


procedentes del tallo de diversas plantas como lino, camo, kenaf y yute, y de
hojas como abac y sisal; ii) residuos agrcolas como la paja de trigo, maz y
arroz o el bagazo de caa; y iii) hierbas silvestres como bamb o hierba elefante.
Actualmente, las fibras no madereras representan una alternativa para la
produccin de pasta de celulosa en pases con baja disponibilidad de madera y
en los que disponen de abundantes residuos agrcolas fibrosos o cultivos de
plantas fibrosas no madereras. As, el uso de estas fibras para la produccin de
pasta de celulosa ha ido aumentando, especialmente en los pases en vas de
desarrollo, como India, China y algunos pases latinoamericanos.

En los pases desarrollados, las fibras no madereras se usan principalmente


para la produccin de papeles especiales. En Espaa, existen varias empresas
que fabrican pasta de papel a partir de fibras no madereras. Entre ellas destaca la
empresa CELESA que utiliza fibras liberianas (del tallo) de lino, camo y yute,
y fibras de hojas de sisal y abac para fabricar pasta de celulosa para papeles
especiales de distintas caractersticas, tales como papeles para cigarrillos, filtros
especiales o papeles dielctricos (Figura 2). Dicha empresa ha suministrado la
mayora de las fibras y sus respectivas pastas de papel que se han estudiado en
esta Tesis.

Lino Yute
Papel para bolsas de vaco

Bolsas de t

Camo Sisal
Papeles para filtros

Papeles
decorativos
Kenaf Abac

Papeles para circuitos Papel para cigarrillos


elctricos

Figura 2. Papeles especiales (izquierda) obtenidos de las pastas de papel fabricadas por la
empresa CELESA (Tortosa, Tarragona) y sus principales materias primas (derecha).

8
1. Introduccin

1.2. ESTRUCTURA Y COMPOSICIN QUMICA DE LOS


MATERIALES LIGNOCELULSICOS
Los materiales lignocelulsicos, incluyendo los productos de origen agrcola
y forestal, representan la mayor fuente de energa y materia orgnica renovables
de la biosfera. Son materiales heterogneos cuya estructura y composicin
qumica varan dentro de amplios lmites y condicionan su utilizacin industrial
y la posible aplicacin de mtodos biotecnolgicos. Los principales
componentes de estos materiales son los polmeros constituyentes de todas las
paredes celulares de materiales vegetales: celulosa, hemicelulosas y lignina
(Figura 3), y una serie de compuestos de bajo peso molecular solubles en agua o
en solventes orgnicos, as como pequeos contenidos en protena y sales
minerales (Fengel y Wegener 1984, Sjstrm 1993).

Lignina

Hemicelulosas

Celulosa

Enlaces de Celulosa Lignina Hemicelulosas


hidrgeno

Figura 3. Representacin esquemtica de los principales constituyentes de la pared vegetal


correspondiente a una angiosperma no leosa (adaptado de Bidlack et al. 1992).

9
1. Introduccin

1.2.1. Celulosa
La celulosa es el componente principal de las clulas vegetales, que
comprende aproximadamente del 10 al 20% del peso seco de las hojas, entre un
43 y un 47% de la madera de conferas, entre un 42 y un 44% de la madera de
frondosas y el 90% del peso de las fibras de algodn (Streitwieser y Heathcock
1983, Aitken et al. 1988). Estructuralmente, es un polmero lineal constituido
por unidades de -D-glucopiranosa unidas por enlaces glicosdicos  (1o4), en
los que dos molculas de glucosa se unen con eliminacin de una molcula de
agua entre dos hidroxilos de los carbonos 1 y 4. La configuracin  slo es
posible por la rotacin de la unidad de glucosa siguiente del eje C1-C4 del anillo
de piranosa, por eso la unidad de cadena de celulosa que se repite es la celobiosa
(disacrido), con una longitud de 1,03 nm (Fengel y Wegener 1984, Sjstrm
1993, Sjstrm y Westermark 1999). Los numerosos grupos hidroxilo favorecen
la formacin de enlaces de hidrgeno intra e inter-moleculares, formando cada
unidad de glucosa dos enlaces intramoleculares y uno intermolecular (Figura 4).
Los enlaces de hidrgeno intermoleculares se establecen con otras cadenas que
estn en el mismo plano, as como con cadenas en planos superiores e inferiores,
de este modo, las cadenas de celulosa se unen dando lugar a la formacin de
microfibrillas, y la unin de stas entre s a la fibra de celulosa, cuyos agregados
forman la pared celular (Lennholm y Henriksson 2007).

Unidad celobiosa
H
H
O
O
OH 1 4 OH 1
O HO O
 HO
O  O
O O
 O

O
OH 1 4 HO O
HO OH 1 O
4
O
H H H
H
O
O
OH OH
O HO O HO
O O
O O
O
O O
OH HO OH
HO O O
H H
H
H
O
O
OH OH
O HO O HO
O O
O O
O
O HO O
OH
HO OH O
O
H H

Figura 4. Estructura de la celulosa donde se muestra la unidad de celobiosa, la conformacin


 (14) y los enlaces por puentes de hidrgeno intra e inter-moleculares.

10
1. Introduccin

En su estructura supramolecular, la celulosa se organiza en zonas cristalinas y


zonas amorfas. Son los enlaces de hidrgeno inter-moleculares los que permiten
una estructura ordenada, esto es, una alta cristalinidad. En las zonas amorfas, el
nmero de enlaces por puentes de hidrgeno establecidos es menor y bastante
ms desorganizado que en las zonas cristalinas, siendo por lo tanto la celulosa
amorfa ms fcil de disolver y ms reactiva, pues la accesibilidad a los grupos
hidroxilo es mayor (Garca Hortal 2007, Annergren 1996). Las propiedades de
los materiales lignocelulsicos estn relacionadas con el grado de
polimerizacin de la molcula de celulosa, que es de al menos 15000 (Brett y
Waldron 1996). La resistencia del papel es debida, en parte, a la resistencia
individual de las cadenas de celulosa, que diminuye si estas se degradan.

1.2.2. Hemicelulosas
Las hemicelulosas comprenden aproximadamente del 25 al 30% del peso seco
de la madera de conferas, entre un 20 y 43% de la madera de frondosas, entre
un 12 y 18% de las fibras liberianas de lino y un 12% de las fibras de hojas de
sisal (Aitken et al. 1988, Fengel y Wegener 1984, Garca Hortal 2007). Actan
como matriz de soporte para las microfibrillas de celulosa y estructuralmente
son ms complejas que la celulosa.

Las hemicelulosas son polisacridos heterogneos constituidos por una


cadena lineal de diferentes monosacridos unidos principalmente por enlaces 
(14) y en algunos casos  (13), de la que parten diversas ramificaciones
(Sjstrm 1993). Los principales monosacridos que las constituyen incluyen
pentosas (D-xilosa y L-arabinosa), hexosas (D-glucosa, D-galactosa, D-manosa,
L-ramnosa y L-fucosa), y cidos urnicos (cido D-glucurnico y cido D-
galacturnico) (Figura 5) con un grado de polimerizacin entre 200 y 300,
siendo ms fciles de disolver y de degradar que la celulosa (Sjstrm y
Westermark 1999). Contrariamente a la celulosa, la naturaleza de las
hemicelulosas vara entre las diferencies especies (Tabla 2). En el caso de las
maderas de conferas se suele apreciar una mayor cantidad de hexosanos, como
la manosa y galactosa, siendo predominantes los galactoglucomananos y los
glucomananos (Garca Hortal 2007, Sjstrm y Westermark 1999) aunque
tambin se observan en las conferas los arabinoglucuronoxilanos (Sjstrm y
Aln 1999). La xilosa es ms abundante en las frondosas donde predominan
pentosanos como los glucuronoxilanos (Fengel 1989, Sjstrm 1993, Shimizu
2001) aunque tambin se observan glucomananos (Garca Hortal 2007, Sjstrm
y Aln 1999). En las plantas no madereras, las hemicelulosas presentan una gran
variedad en su composicin dependiendo de la especie, siendo en algunas, como
en el lino, predominantes la manosa y la galactosa (Morrison et al. 1999) y en
otras, como en el kenaf, la xilosa (Han 1998, Neto et al. 1996).

11
1. Introduccin

PENTOSAS HEXOSAS CIDOS HEXURNICOS

H H OH H

H H O H O COOH H O
HO HO HO
HO OH HO OH HO OH
H H OH H OH
OH
H H H H H H

-D-Xilosa -D-Glucosa cido -D-Glucurnico

OH H OH H
COOH H
H H O OH O O
H HO H 3C O
HO H HO OH HO H
H OH H H OH
H
H OH H H H OH

-L-Arabinopiranosa -D-Manosa cido -D-4-O-Metilglucurnico

O HO H OH
H H
O COOH
H O H O
HO H
H H
HO H HO H
H O H 2C OH H H
OH OH
H OH
H OH H OH

-L-Arabinofuranosa -D-Galactosa cido -D-Galacturnico

DESOXI-HEXOSAS

H
OH
H H O
H H O
HO
H
H H
H H
OH
OH
OH OH
OH OH

-L-Ramnosa -L-Fucosa

Figura 5. Monosacridos componentes de las hemicelulosas (adaptado de Fengel y Wegener


1984).

12
1. Introduccin

Tabla 2. Tipos y estructuras simplificadas de las principales hemicelulosas en diversos


materiales lignocelulsicos, X, xilosa; A, arabinosa; G, glucosa; Gal, galactosa; M, manosa;
Ac, grupo acetilo; Gl, cido 4-O-metilglucurnico (adaptado de Garca Hortal 2007)
Tipo hemicelulosa Estructura simplificada Presencia

X- X - X - X
Glucuronoxilanos Frondosas, plantas no madereras
Ac Gl
7

Glucomananos G-M-G-M-M Conferas, frondosas

G- M - M - M
Galactoglucomananos Conferas
Gal Ac

X - X - X- X - X5
Arabinoglucuronoxilanos Conferas, plantas no madereras
Gl A
2

Las hemicelulosas, con estructura ramificada y amorfa, son muy hidroflicas y


desempean un papel fundamental en el proceso de fabricacin de papel al
promover el hinchamiento de la fibra y aumentar su plasticidad, flexibilidad y
capacidad de enlace, con la consiguiente mejora de la densidad de la hoja. Sin
embargo, durante el secado de la pasta tambin tienden a mantener dura o rgida
la fibra, lo que puede impedir la subsiguiente rehidratacin de la pasta (Garca
Hortal 2007).

1.2.3. Lignina
Despus de la celulosa, la lignina es el polmero ms abundante en el mundo
vegetal, representando entre un 25 y un 33% de la madera de conferas y entre
un 18 y un 34% de la madera de frondosas (Aitken et al. 1988). En el caso de las
plantas no madereras hay un menor porcentaje de lignina con respecto a las
especies madereras, con un 8-9% para fibras de hojas (abac y sisal), entre un 3-
13% para fibras liberianas (lino, camo, yute y kenaf), entre un 12 y un 21%
para pajas (paja de arroz, paja de trigo) y entre un 19 y un 22% para caas

13
1. Introduccin

(azcar, bambes) (Garca Hortal 2007). La lignina acta como aglomerante de


las fibras debido a su carcter hidrfobo siendo una de las molculas orgnicas
ms recalcitrantes.

Estructuralmente, la lignina es un heteropolmero aromtico con una


estructura tridimensional irregular, constituida por unidades de fenilpropano con
diferentes patrones de substitucin y unidas por diferentes tipos de enlaces, que
varan considerablemente entre las especies vegetales e incluso dependiendo de
la edad (Freudenberg y Lehmann 1963), de la parte del rbol/planta (Bland
1966), tipo de clulas (Fergus y Goring 1970, Hardell et al. 1980a, 1980b) y del
lugar de la pared celular donde se sintetice (Fergus y Goring 1970, Fukushima y
Terashima 1991, Christierini et al. 2005), por lo que la lignina no puede ser
descrita por una frmula simple.

Los precursores de la lignina son los alcoholes p-hidroxicinamlicos (Figura


6), que incluyen los alcoholes p-cumarlico (4-hidroxicinamlico, I), coniferlico
(4-hidroxi-3-metoxicinamlico, II) y sinaplico (4-hidroxi-3,5-
dimetoxicinamlico, III), que difieren entre s en el nmero de grupos metoxilo
sustituyentes. Estos precursores se sintetizan a su vez a partir de la fenilalanina a
travs de la ruta de los cidos cinmicos (Higuchi 1997, Boerjan et al. 2003,
Freudenberg y Neish 1968, Adler 1977, Ralph et al. 2004). Recientemente, se ha
descrito la existencia de otros precursores de la lignina, tales como derivados
acilados (acetatos y/o p-cumaratos) de los correspondientes alcoholes p-
hidroxicinamlicos (IV y V) (del Ro et al. 2004, 2007, 2008a, 2008b, Martnez
et al. 2008) observados en diversas plantas angiospermas, as como alcoholes
dihidroxicinamlicos (VI), o aldehdos cinamlicos (VII), observados en la
lignina de especies modificadas genticamente (Ralph et al. 1997, 1998,
Sederoff et al. 1999). Su deshidrogenacin oxidativa, catalizada por peroxidasas
o lacasas en presencia de perxido de hidrgeno u oxgeno, respectivamente,
conlleva a la formacin de radicales fenoxilo estabilizados por resonancia que
luego se acoplan entre s y con el polmero creciente de lignina mediante
diversos tipos de enlaces (Figura 7).

Aunque la variedad de uniones para formar el polmero de lignina es amplia


(Figura 8), se pueden diferenciar dos tipos: uniones de tipo ter y uniones de
tipo carbono-carbono. La formacin de enlaces ter-alquil-arlico es la ms
favorable termodinmicamente, como es el caso del enlace -O-4, en el que se
encuentran involucrados la posicin  del monolignol radical y el radical
fenoxilo del polmero de lignina creciente. En menor proporcin existen uniones
de tipo aril-aril ter, como por ejemplo la unin 4-O-5. Los enlaces de tipo
carbono-carbono, conocidos tambin como enlaces condensados, son ms
difciles de romper que los de tipo ter, e incluyen las uniones de dos cadenas
alifticas (- resinol), la unin de un carbono de un anillo bencnico con el de
una cadena aliftica (-1 y -5 fenilcumarano) y las uniones entre carbonos de

14
1. Introduccin

dos anillos bencnicos (5-5). Se ha descrito que el enlace 5-5 no se encuentra


tal cual, sino en forma de trmero, ya que incorpora una nueva unidad mediante
un enlace -O-4 y un enlace -O-4, dando lugar a una estructura de tipo
dibenzodioxocina (Karkunen et al. 1995). Igualmente, estudios recientes indican
que la mayora de las uniones -1 se encuentran en forma de espirodienonas
(Zhang y Gellersted 2001, Zhang et al. 2006).

OH OH OH O
J

 O


OMe (MeO) OMe (MeO) (OMe)

OH OH OH OH

I II III IV

OH

O OH H O

(MeO) (OMe) HO OMe (MeO) (OMe)

OH OH OH

V VI VII

Figura 6. Estructuras de precursores de la lignina: I, alcohol p-cumarlico; II, alcohol


coniferlico; III, alcohol sinaplico; IV, derivado acetilado de los alcoholes p-
hidroxicinamlicos; V, derivados p-cumaroilados de los alcoholes p-hidroxicinamlicos; VI,
alcohol 5-hidroxiconiferlico y VII, aldehdos correspondientes a los alcoholes p-
hidroxicinamlicos.

15
1. Introduccin

OH
J OH
OH OH OH OH


Peroxidasa
Lacasa
- (e- + H+)

OMe
OMe
OMe OMe
OH OMe OMe
O
O O O O
a

Lignina Lignina
(MeO) (MeO)
Lignina Lignina
(MeO)
HO HO HO (MeO)
HO O HO
OMe OMe
R
O O
OMe
Acoplamiento OMe
Oxidacin radicalar ROH
(MeO) OMe (MeO) OMe (MeO) OMe
(MeO) OMe
OH O O OH

Figura 7. Sntesis de la lignina: (a) deshidrogenacin del alcohol coniferlico y formas


resonantes del radical fenoxilo (adaptado de Adler 1977) y (b) mecanismo de la unin de los
monolignoles libres al polmero de lignina (Freudenberg y Neish 1968).

La cantidad de lignina, su distribucin a travs de las paredes celulares y la


estructura bsica de la misma, difieren segn su origen entre conferas,
frondosas y fibras no madereras. En la lignina de las conferas la estructura que
se repite predominantemente es la unidad guayacilo (G), que contiene un nico
grupo metoxilo en el anillo de fenilpropano y deriva del alcohol coniferlico
(ms de 95% de las unidades estructurales). En el caso de la lignina de las
maderas de frondosas, hay predominantemente dos unidades que se repiten, la
unidad guayacilo (G) y la unidad siringilo (S), conteniendo esta ltima dos
grupos metoxilo por ncleo de fenilpropano y deriva del alcohol sinaplico
(Parhan 1983, Sarkanen y Hergert 1971, Lin y Dence 1992). Por otro lado, la
lignina de fibras no madereras presenta unidades del tipo p-hidroxifenilo (H),
procedentes del alcohol p-cumarlico, y unidades S y G, en proporciones
variables dependiendo de la planta. Las unidades G, al contrario que las S,
tienen un nico grupo metoxilo y la posicin C-5 est libre y disponible para la
formacin de enlaces carbono-carbono, por lo que ligninas con mayor cantidad
de unidades G tienen una estructura ms condensada y por lo tanto la lignina se
degrada con mayor dificultad. En la Figura 9 se muestran modelos estructurales
del polmero de lignina de conferas y frondosas, y en la Figura 10 se muestran
modelos estructurales del polmero de lignina de algunas plantas no madereras.

16
1. Introduccin

6
MeO 1
HO MeO 5
6 2
J 5 1 OH
J 4 3
D E E
4 2
HO 3
O HO D O
OMe
OMe OMe
1 1 2 1
6 2 6 2 1 3 6 2

5 3 5 3 6 4 5 3
4 4 5 4
MeO OMe OMe O OMe
OMe

O O OMe O

-O-4 -O-4  4-O-5 

OMe
1
6 2
O
5 3 3
HO 2 4
4
OMe J
1 5
O
HO
E E 6
J HO D OMe J D
OMe
D O 2
1 3
E E
6 4
5
1 1
O MeO D J
6 2 6 2 6
5 1 O
5 3 5 3
4 4 OMe 4 2
MeO OMe MeO OMe 3
O
O O
OMe

-5  /-O-4  -1  -  /-O-  /-O- 


fenilcumarano resinol

1 1
2 6 6 2
3 5 5 3
4 4 OMe
MeO OMe
O 1''
6'' 2''
3
4
O O 5'' 3''
2
4''
5 MeO OMe
Dcc Ecc MeO 1
6 D
O O
OH J E D E
6 1
5 J
MeO 2 OH
HO
1
4 3 6 2
5 3
OMe 4
O MeO OMe
O

5-5  / -O-4  / -O-4  -1  /-O- 


dibenzodioxocina espirodienona

Figura 8. Enlaces tipo ter y carbono-carbono presentes en el polmero de lignina.

17
1. Introduccin

A B

(a) (b)
Figura 9. Modelos del polmero de lignina en maderas: (a) confera (Picea) (Brunow 2001) y
(b) frondosa (lamo) (Boerjan et al. 2003).

18
1. Introduccin

OMe OH
OH O OH
HO O
OMe
O OH
O O MeO OMe
OMe O O O
O
OH OMe O
HO HO
OMe OH OH
O
O OMe
HO
OMe
MeO OMe
O
O O O
O O
O MeO OMe
OMe O
HO
OH
O
O OMe
O
O MeO OMe
O
O
MeO OMe OMe
O HO HO
OH
OMe O
HO HO OMe OH
O O
O OH
O
O MeO OMe
O OMe
HO
MeO OMe O
O O
O O
OMe OMe
HO OH
MeO
O
OMe OH O OMe
O
OH
HO O
HO
MeO OMe OMe
O OMe
HO
O MeO O
O
OMe O OH
OH O
MeO MeO

O OMe O OMe
O
HO O HO OH

OMe OMe

O O

O OH O OH
O O
MeO MeO
O OMe O OMe
HO OH
HO O
O O OH
OMe
OMe
MeO O OMe
O
MeO O
O O OH OH
O HO
OMe MeO
OH OMe MeO HO
O O OMe
MeO O O O
O OH
O OH OH

MeO MeO

HO OMe HO OMe

(a) (b)

Figura 10. Modelos del polmero de lignina en plantas no madereras: (a) kenaf y (b) abac
(del Ro et al. 2008a).

19
1. Introduccin

La lignina aparece tambin asociada a los polisacridos en la pared celular y


es esta asociacin la que determina la rigidez y la resistencia estructural del
material. Las hemicelulosas estn asociadas a la lignina principalmente a travs
de las unidades de arabinosa, xilosa y galactosa por enlaces de tipo glicosdico,
ter benclico y ster benclico formando complejos lignina-polisacridos
(Watanabe 2003).

1.2.4. Componentes de bajo peso molecular


Adems de los carbohidratos (celulosa y hemicelulosas) y lignina, existen en
los materiales lignocelulsicos pequeas cantidades de componentes que no
influyen en la estructura morfolgica de las clulas pero que tienen gran
influencia en el procesamiento de estos materiales. Algunos de estos
componentes protegen a la madera de los insectos y son responsables de su
color, olor y gusto. Atendiendo a su solubilidad se pueden dividir en compuestos
extrables en solventes apolares, que incluyen los extrables lipoflicos, y
compuestos extrables en disolventes polares (extrables hidroflicos), as como
diversos compuestos insolubles tales como sustancias proteicas, pcticas y de
naturaleza inorgnica (Garca Hortal 2007, Hillis 1962, Fengel y Wegener 1984,
Rowe 1989, Sjstrm 1993).

Los extrables lipoflicos (Figura 11) incluyen tpicamente alcanos (a),


alcoholes grasos (b), aldehdos (c), cidos grasos (d), esteroles (e), cidos
resnicos (f), ceras (steres de cidos grasos con alcoholes de cadena larga, g) y
glicridos (steres de cidos grasos con glicerol, h). Los esteroles pueden
encontrarse libres o esterificados con cidos grasos (steres de esteroles, i) y
tambin pueden estar formando glicsidos y acilglicsidos (Gutirrez y del Ro
2001), siendo el ms abundante el 3-D-glucopiransido (j).

Los extrables lipoflicos afectan negativamente al proceso de fabricacin de


pasta de celulosa as como al producto final, formando depsitos insolubles
comnmente denominados depsitos de pitch, que se describen ms adelante en
el apartado 1.4. Debido a su alto grado de pegajosidad, los esteroles libres y
conjugados se encuentran en el origen de muchos depsitos de pitch (Back y
Allen 2000, del Ro et al. 1998, 2000). El estudio de los compuestos extrables
lipoflicos de cada una de las materias primas constituye un requisito
fundamental para identificar los compuestos que originan los depsitos de pitch
y disear estrategias adecuadas para su control.

20
1. Introduccin

OH

a b

O O

H OH

c d

COOH
HO

e fF

O
g

CO-O-CH2

CO-O-CH

CO-O-CH2

CH2OH
O O
OH O
O HO
OH j
i

Figura 11. Estructuras de compuestos representativos de las principales familias de


extrables lipoflicos: (a) pentacosano, (b) docosanol, (c) octacosanal, (d) cido palmtico, (e)
sitosterol, (f) cido abitico, (g) octacosanil hexadecanoato, (h) trilinolena, (i) sitosteril
linoleato, (j) sitosteril 3-D-glucopiransido.

21
1. Introduccin

Por otro lado, los extrables polares engloban diferentes compuestos fenlicos
libres de bajo peso molecular (Figura 12), los cuales incluyen precursores de la
lignina (cidos p-hidroxicinmicos y aldehdos p-hidroxicinamlicos), cidos
bencenocarboxlicos relacionados (cido p-hidroxibenzoico, vainllico y
sirngico), aldehdos y cetonas aromticas (p-hidroxibenzaldehdo, vainillina,
siringaldehdo y propioguayacona), e incluyen taninos hidrolizables (steres del
cido glico y sus dmeros), flavonoides (estructuras derivadas del anillo de
flavona, 2-fenilbenzopirona) y taninos no hidrolizables (varias unidades de
flavonoides condensadas). Adems de incrementar el consumo de reactivos
durante la coccin, estos compuestos pueden dificultar las reacciones de
pasteado impidiendo la difusin de los reactivos en la materia prima, y los
taninos, cuando estn presentes en cantidades importantes, forman complejos
coloreados con cationes metlicos afectando el color de las pastas de papel y su
blanqueabilidad (Garca Hortal 2007).

HO O

HO O O H O

O O O O O

OH OH OH OH

a b c d

HO O

HO OH

OH
O

e f

Figura 12. Estructuras de compuestos representativos de los compuestos extrables polares:


(a) cido sirngico, (b) cido p-hidroxibenzoico, (c) vainillina, (d) acetosiringona, (e) cido
glico y (f) 2-fenilbenzopirona.

22
1. Introduccin

1.3. UTILIZACIN DE CULTIVOS LIGNOCELULSICOS:


PRODUCCIN DE PASTA DE CELULOSA
La fabricacin de pasta de celulosa consiste bsicamente en la separacin de
las fibras de celulosa de la madera u otros materiales fibrosos a travs de
procesos mecnicos y/o qumicos (Fengel y Wegener 1984, Sjstrm 1993). Se
cree que la fabricacin de papel tuvo su origen en China hacia el ao 100 d.C. y
para su fabricacin se utilizaban trapos, camo, paja y hierba como materias
primas, que se golpeaban contra morteros de piedra para separar la fibra
original. Aunque con el tiempo gan terreno la mecanizacin, hasta el siglo XIX
siguieron utilizndose los mtodos de produccin por lotes y las fuentes de fibra
agrcolas. Las primeras mquinas continuas de papel se patentaron a finales del
siglo XIX y principios del siglo XX. Entre 1844 y 1884 se desarrollaron los
primeros mtodos para la obtencin de pasta a partir de madera, una fuente de
fibra ms abundante que los trapos o hierbas; estos mtodos implicaban la
abrasin mecnica y la aplicacin de diversos procedimientos qumicos. La
fabricacin de papel fue una labor artesana e individualizada, pero con los
descubrimientos de la ciencia y los avances tecnolgicos, as como con el
desarrollo y expansin de la cultura, la industria de fabricacin del papel se
desarroll a un ritmo acelerado. La industria de pasta celulsica muestra an una
tendencia creciente en su produccin, segn los datos de la FAO (FAO 2004) de
la Tabla 3.

Tabla 3. Estadsticas sobre la capacidad de produccin de papel en los principales pases


productores (2003-2008) (FAO 2004)

2003 2004 2005 2006 2007 2008


Capacidad total , 1000Mt

Pases desarrollados 244169 247669 251387 254304 255786 256589


Norteamrica 114550 114325 115062 115405 115405 115405
Europa 91894 95866 98824 101360 102827 103620
Oceana 4049 4143 4208 4246 4261 4271
Otros 33676 33335 33293 33293 33293 33293

Pases en desarrollo 21201 21774 22434 22964 23296 23493


frica 2681 2688 2721 2865 2865 2872
Amrica Latina 13181 13469 13841 14052 14219 14261
Asia 5339 5617 5872 6047 6212 6360

23
1. Introduccin

El proceso de produccin de pasta de celulosa comprende fundamentalmente


el proceso de pasteado y el proceso de blanqueo. El proceso de pasteado tiene
por objeto separar las fibras de celulosa del resto de los componentes de la
madera, fundamentalmente de la lignina ya que las fibras de celulosa se
encuentran cementadas por ella. Por otro lado, el blanqueo de la pasta tiene por
objeto disolver o modificar la lignina residual que no se elimina durante el
pasteado, para mejorar las propiedades de la pasta y consecuentemente del
producto final.

1.3.1. Procesos de pasteado


Dependiendo de las caractersticas de las fibras, el tratamiento aplicado para
destruir o debilitar los enlaces interfibras vara, con la finalidad de obtener una
pasta de caractersticas adecuadas y el mayor rendimiento posible. Los procesos
de obtencin de pasta de papel se clasifican bsicamente en mecnicos y
qumicos. Combinaciones de stos dan lugar a procedimientos intermedios o
semiqumicos.

Procesos mecnicos
El pasteado mecnico tiene como objeto la separacin fsica de las fibras,
realizndose el desfibrado por fragmentacin mecnica, utilizando molinos y
refinadores de discos. La fabricacin de pastas mecnicas ofrece la ventaja de
dar como resultado rendimientos elevados (hasta un 98% del material inicial),
obtenindose pastas ventajosas para algunos tipos de papel por su rigidez,
volumen y opacidad (Garca Hortal 2007). Sin embargo, como en este proceso la
lignina slo se ablanda (no se disuelve), el alto contenido en lignina va en
detrimento de la calidad del papel ya que las fibras muy lignificadas son rgidas,
poco flexibles, no estn bien unidas entre s, proporcionando papeles con bajas
caractersticas de resistencia y muy sensibles al envejecimiento ptico.

Procesos qumicos
En el pasteado o coccin qumica, la deslignificacin se lleva a cabo con la
ayuda de agentes qumicos cidos o bsicos, en digestores o reactores a altas
temperaturas y presiones. La pasta se produce con disolucin de la lignina que
se encuentra entre las fibras del material lignocelulsico y los productos de
degradacin se disuelven en la leja de la coccin. En el pasteado qumico, se
eliminan muchos de los componentes no fibrosos de la materia prima y los
rendimientos son normalmente del 35 al 65%, sin embargo, la pasta se blanquea
mejor y el producto es ms resistente y de mejor calidad que en el caso de los
procesos mecnicos (Sjstrm 1993).

24
1. Introduccin

Los procesos de pasteado qumico pueden realizarse en condiciones alcalinas,


como el pasteado a la sosa y el proceso kraft, o en condiciones cidas como el
pasteado al sulfito. Otro tipo de procesos utiliza solventes orgnicos (pasteado
organosolv) (Gilarranz et al. 1999).

El proceso a la sosa es el ms antiguo y el ms simple de los procesos


qumicos alcalinos. En este proceso, la fibra se somete a un proceso de coccin
con sosa custica y vapor a alta presin y temperatura. El hidrxido de sodio es
un producto muy til para la deslignificacin de materias primas vegetales,
principalmente de maderas, pajas de cereales y plantas fibrosas en general. En
este proceso, se puede utilizar antraquinona (AQ) como catalizador ya que
presenta dos efectos fundamentales como son la aceleracin del proceso de
deslignificacin alcalino y la estabilizacin de los carbohidratos, mejorando los
rendimientos respecto al proceso convencional en las mismas condiciones de
operacin (Abarca y Blanco 2008).

El proceso kraft para la obtencin de pasta de papel es un proceso qumico


alcalino que deriva del proceso a la sosa. En este proceso, adems de hidrxido
de sodio se utiliza sulfuro sdico, siendo estos agentes de coccin conocidos
como lejas blancas. El proceso se lleva a cabo en digestores que pueden ser
tanto discontinuos como continuos, en los que se introducen las astillas junto a
las lejas blancas llevndose a cabo la coccin a elevada temperatura (150-
170C) y presin. Generalmente el proceso tiene lugar con una concentracin de
reactivos del 16-20% (expresados como peso de Na2O, en relacin al peso de la
madera). Este tipo de pasteado permite obtener pastas con una gran resistencia,
aunque con menor rendimiento (entre un 40 y 60%), ya que se elimina mucha
cantidad de lignina (hasta el 90%) (Garca Hortal y Colom 1992, Santos et al.
1997). La ventaja de este proceso es que requiere tiempos de coccin
relativamente cortos pues el sulfuro acelera la deslignificacin reduciendo la
degradacin del material celulsico y produciendo as pastas de mejor calidad.
Para este proceso, se pueden utilizar todo tipo de maderas, aunque los mejores
resultados se obtienen con maderas de frondosas.

El proceso al sulfito es un proceso qumico cido donde se utilizan sulfitos y


bisulfitos para la deslignificacin. Es un proceso ms fuerte que el proceso
alcalino y permite una mejor separacin de la celulosa. Este proceso est
limitado en cuanto al tipo de materia prima, pues no se pueden utilizar maderas
de conferas ya que a pH bajos los fenoles y los cidos resnicos se condensan
con la lignina formando complejos insolubles y coloreados que manchan la
pasta. El licor de coccin es una disolucin de cido sulfuroso (H2SO3) y
bisulfito de calcio (Ca(HSO3)2), que se prepara disolviendo dixido de azufre en
agua y hacindola reaccionar con CaCO3. Los digestores operan a temperaturas
comprendidas entre los 125 y 180C segn la aplicacin que se quiera dar al
producto final (papel, cartn, etc.), obtenindose rendimientos entre el 40 y

25
1. Introduccin

60%. En estos procesos tambin se degradan los hidratos de carbono por rotura
de los enlaces glicosdicos, lo que provoca una disminucin del grado de
polimerizacin todava mayor que en los procesos kraft siendo la pasta
resultante menos resistente, pero por lo contrario estas pastas son ms fciles de
blanquear. El mtodo al sulfito ha sido relegado en parte por el proceso kraft
(Bryce 1990).

Alternativamente, se han desarrollado los procesos organosolv que utilizan


solventes orgnicos para la deslignificacin. Estos procesos presentan una
mayor selectividad y por lo tanto, dan lugar a rendimientos mayores. Por otro
lado, permiten la utilizacin de cualquier materia prima fibrosa (conferas,
frondosas y plantas no madereras) dando lugar a la obtencin de pastas con bajo
contenido en lignina que pueden ser blanqueadas sin el uso de compuestos
clorados. Se han empleado multitud de disolventes orgnicos (etanol, metanol,
butanol, alcohol benclico, glicerol, glicol, etilenglicol, trietilenglicol, fenol,
acetona, cido frmico, cido actico, dioxano, dimetilsulfxido,
hexametilendiamina, etc.) puros o en disolucin acuosa, con la adicin o no de
catalizadores. Los elevados precios de los reactivos, la dificultad en su
recuperacin y en muchos casos su elevada toxicidad, ha favorecido el uso de
alcoholes alifticos de bajo peso molecular (etanol y metanol) como solventes
para los procesos organosolv (Herrero et al. 2002). Estos solventes combinan su
alta velocidad de deslignificacin en condiciones de operacin favorables y su
fcil recuperacin. Sin embargo, en general, las propiedades de resistencia de las
pastas organosolv son inferiores a las pastas kraft.

1.3.2. Procesos de blanqueo


En el proceso de blanqueo se trata qumicamente la pasta de celulosa para
eliminar o modificar la lignina residual que queda despus del proceso de
coccin. Los componentes coloreados de la lignina se degradan, disuelven o se
decoloran (Sjstrm 1993). El proceso de blanqueo se lleva a cabo hasta el
punto de blancura que se pretende, por lo que el nmero de etapas depender de
la calidad de la pasta que se desee obtener (Figura 13). Los reactivos
comerciales ms utilizados para el blanqueo son el cloro gas, el hipoclorito, el
perxido de hidrgeno y el dixido de cloro; y el lcali utilizado es el hidrxido
de sodio que se usa en la operacin de extraccin alcalina.

26
1. Introduccin

Figura 13. Diferentes grados de blancura de una pasta de celulosa.

El blanqueo ha sido la etapa de la produccin de pasta de celulosa que ha


sufrido ms cambios durante los ltimos aos. La decisin de eliminar el cloro
molecular y, en algunos casos tambin el dixido de cloro, de las secuencias de
blanqueo se debe a la necesidad de reducir las emisiones de compuestos
clorados orgnicos, de haluros orgnicos absorbibles y dioxinas en los efluentes
de las plantas de blanqueo. El desarrollo de leyes ms restrictivas con respecto a
los procesos contaminantes ha llevado a una parte importante de la industria
europea de pasta y papel a introducir secuencias de blanqueo totalmente libres
de cloro (blanqueo TCF, totally chlorine free) (Brooks et al. 1994). Estas
secuencias incluyen blanqueo con perxido de hidrgeno, oxgeno y ozono. Otra
parte de la industria papelera mundial, ha eliminado el cloro elemental pero
contina utilizando dixido de cloro en el blanqueo (blanqueo ECF, elemental
chlorine free).

El desarrollo del blanqueo con oxgeno ha sido bastante lento por la


degradacin de la celulosa y dems polisacridos de la madera. Las ventajas del
perxido de hidrgeno se apoyan en su facilidad de manipulacin y aplicacin,
su versatilidad y la naturaleza relativamente inocua de los productos de reaccin.
La novedad de las secuencias de blanqueo TCF obliga a solucionar nuevos
problemas que surgen al introducir mtodos de blanqueo menos agresivos y que
se describen en el apartado siguiente.

En Espaa, diversas empresas productoras de pasta de papel, incluyendo


ENCE y CELESA (que han suministrado pastas de papel para la presente Tesis),
han realizado un considerable esfuerzo de inversin y modernizacin de sus
fbricas con objeto de adaptarlas a las modernas tecnologas ECF y TCF.

27
1. Introduccin

1.4. PROBLEMTICA PLANTEADA POR LA PRESENCIA DE


LIGNINA Y LPIDOS EN LA PRODUCCIN DE PASTA DE
CELULOSA
Una parte de los problemas originados durante la produccin de pasta de
celulosa est relacionada con los compuestos extrables lipoflicos de los
materiales lignocelulsicos. Estos compuestos causan tanto problemas
medioambientales como problemas tcnicos durante el proceso de produccin.
Entre los compuestos lipoflicos ms problemticos estn los cidos grasos
libres, cidos resnicos, ceras, alcoholes, esteroles tanto libres como
esterificados, glicridos, cetonas y otros compuestos (Hillis 1962, Fengel y
Wegener 1984, Rowe 1989, Gutirrez et al. 1999). Durante el proceso de
produccin de pasta de celulosa los compuestos lipoflicos se liberan formando
partculas coloidales que pueden unirse y formar gotas que luego se depositan en
la pasta o en la maquinaria formando los llamados depsitos de pitch (Figura
14). La formacin de estos depsitos da lugar a importantes prdidas
econmicas como consecuencia de pastas contaminadas, paradas en la
produccin, as como por el coste de los aditivos qumicos utilizados para el
control del pitch (Hillis 1989, Allen 2000). Adems, algunos compuestos
lipoflicos de los materiales lignocelulsicos tambin tienen un impacto negativo
sobre el medio ambiente, considerndose algunos de ellos como primera fuente
de toxicidad cuando se liberan en los vertidos (Ali y Sreekrishnan 2001, Rigol et
al. 2004). Esto es especialmente importante en los procesos modernos donde el
blanqueo con cloro ha sido sustituido por el blanqueo libre de cloro elemental
(ECF) o totalmente libre de cloro (TCF). El blanqueo ECF evita los problemas
asociados a la formacin de compuestos clorados producidos en el blanqueo con
cloro. Sin embargo, algunos de los extrables lipoflicos que son destruidos por
el dixido de cloro no se eliminan en el blanqueo TCF, ya que estas secuencias
que utilizan oxgeno y perxido de hidrgeno no afectan prcticamente a la
fraccin lipdica de las materias primas.

La problemtica del pitch es muy compleja porque vara con la materia prima
as como con el proceso empleado para la fabricacin de pasta y papel. En el
caso de las pastas mecnicas, los depsitos de pitch muestran una composicin
similar a los extractos lipoflicos de la materia prima. En el caso de las pastas
alcalinas, slo algunos de los compuestos extrables presentes en la materia
prima sobreviven al proceso de coccin. En condiciones alcalinas los steres de
glicerol se saponifican y los cidos grasos y resnicos se disuelven. Los steres
de esteroles, los esteroles libres y las ceras, se saponifican ms lentamente que
los steres de glicerol, no forman jabones solubles como en el caso de los cidos
libres, por lo que tienen tendencia a depositarse (Gutirrez et al. 2001).

28
1. Introduccin

Figura 14. Imagen de una gota de resina en el rbol (izquierda) y de un depsito de pitch
en una pasta kraft TCF (cedidas por Mara Jess Ortega y Javier Romero, respectivamente).

Por otro lado, la lignina est tambin relacionada con la problemtica


existente en la produccin de pasta de celulosa ya que la variabilidad en su
composicin y estructura influye decisivamente en el proceso de
deslignificacin. Por otro lado, la formacin de compuestos oxidados de la
lignina durante el proceso de pasteado (lignina residual) es responsable del color
oscuro de las pastas. La fabricacin de pastas de papel mediante tecnologas
menos contaminantes ha trado consigo nuevos problemas en el blanqueo de la
pasta, que no se daban al utilizar reactivos ms agresivos (aunque tambin ms
contaminantes) y/o en sistemas con un menor grado de cierre en los circuitos.
De momento ni el oxgeno ni la combinacin de oxgeno y perxido pueden
igualar la eficacia de la cloracin para la eliminacin de los productos derivados
de la lignina, responsable del color de las pastas.

1.5. BIOTECNOLOGA EN EL SECTOR DE LA PASTA DE CELULOSA


La produccin de pasta y papel ha sido tradicionalmente un proceso industrial
con un fuerte impacto medioambiental. El gran incremento en la demanda de
papel ha agravado el impacto negativo sobre el medio ambiente, por lo que se
han desarrollado leyes ms restrictivas con respecto a los procesos
contaminantes. Por consiguiente, las empresas papeleras han tenido que realizar
un considerable esfuerzo de inversin y modernizacin de sus fbricas con
objeto de adaptarlas a tecnologas ms limpias y adems, con un mayor grado de
cierre en los circuitos para reducir los efluentes lquidos. La biotecnologa
aplicada a este sector ofrece nuevas posibilidades de utilizar mtodos biolgicos

29
1. Introduccin

basados en el uso de hongos y enzimas para reducir o remediar el impacto


medioambiental, reduciendo el consumo de reactivos qumicos, as como el
gasto energtico durante la fabricacin de pasta de papel.

Durante los ltimos aos, el nmero de aplicaciones enzimticas en la


industria de la pasta de celulosa ha aumentado considerablemente, y varias han
alcanzado o se estn acercando a su uso comercial. stas incluyen el uso de
xilanasas para ayudar al blanqueo, la deslignificacin directa con enzimas
oxidativas, el ahorro de energa de refino con celulasas, as como la reduccin
de depsitos de pitch con lipasas (Bajpai 1999, 2006). Adems de las enzimas,
los tratamientos microbianos tambin tienen una potencial aplicacin para
aumentar la eficiencia en la fabricacin de pasta de celulosa, para la reduccin
de los problemas de pitch y la mejora en la reutilizacin de las aguas del
proceso.

En la presente Tesis, se incluye el estudio de enzimas para dos de estas


aplicaciones, como son el blanqueo de la pasta de papel y el control del pitch,
que se mencionan a continuacin.

1.5.1. Degradacin enzimtica de la lignina


El uso de enzimas en la industria papelera ha crecido rpidamente a partir de
mediados de los aos 80. Las enzimas ms utilizadas en el blanqueo de las
pastas son enzimas hidrolticas como las xilanasas, que se utilizan para limitar el
uso de cloro en los procesos de blanqueo de la pasta (Viikari et al. 1994). Las
xilanasas no actan directamente sobre la lignina, sino catalizando la hidrlisis
de los xilanos que se encuentran entre las microfibrillas de la celulosa y la
lignina. Sin embargo, enzimas de tipo oxidoreductasa (lacasas y peroxidasas)
tienen mayor potencial que las xilanasas porque actan directamente sobre la
lignina. Durante aos se concedi mayor atencin a las peroxidasas
ligninolticas que a las lacasas en la degradacin de la lignina y en el desarrollo
de aplicaciones biotecnolgicas (Paice et al. 1995) ya que los bajos potenciales
redox de las lacasas (0.3 a 0.8 V) comparados con los de las peroxidasas
ligninolticas (>1 V) slo permiten a las lacasas la degradacin directa de
compuestos fenlicos de bajo potencial redox, que constituyen nicamente un
20% del total de la lignina (Kawai et al. 1987a, 1987b). El inters por las lacasas
como biocatalizadores industriales en la produccin de pasta de papel se ha
incrementado enormemente tras el descubrimiento de compuestos mediadores
que amplan la accin de la lacasa a sustratos no fenlicos, lo que aumenta el
potencial en la degradacin de la lignina y de otros compuestos aromticos (Call
y Mcke 1997). En la Figura 15 se pueden observar las estructuras de estas tres
enzimas.

30
1. Introduccin

(a) (b) (c)

Figura 15. Enzimas de inters en la industria de la pasta y papel: (a) xilanasa, (b) lacasa y (c)
peroxidasa verstil.

Las peroxidasas catalizan la oxidacin de una gran variedad de compuestos


tanto orgnicos como inorgnicos en presencia de perxidos. Hay dos tipos de
peroxidasas ligninolticas: la lignina peroxidasa (LiP) y las manganeso
peroxidasas (MnP) (Tien y Kirk 1983, Glenn et al. 1983, Kuwahara et al. 1984).
La LiP oxida compuestos aromticos de alto potencial redox, como el alcohol
veratrlico (alcohol 3,4-dimetoxibenzlico) y dmeros modelo de lignina de tipo
no fenlico. Esta enzima es una glicoprotena con hierro protoporfirnico IX
como grupo prosttico, dependiente de H2O2 para su actividad. Inicialmente es
oxidada por perxido de hidrgeno, oxidando ncleos aromticos de la molcula
de lignina (fenlicos y no fenlicos), generando radicales catinicos. Estos
interactan espontneamente con nuclefilos (principalmente H2O) y con
oxgeno molecular, generando una combustin enzimtica donde los enlaces
C-C e C-O se rompen, despolimerizando la lignina y abriendo los anillos
aromticos. Las MnP son glicoprotenas con hierro protoporfirnico IX como
grupo prosttico y dependiente de H2O2 para su actividad, pero la oxidacin por
esta enzima es tambin dependiente de la disponibilidad de iones manganeso.
Un tercer tipo de peroxidasa ligninoltica, la VP (peroxidasa verstil), se ha
descrito por primera vez en Pleurotus eryngii (Martnez et al. 1996, Ruiz-
Dueas et al. 1999a, Camarero et al. 1999) y se caracteriza por combinar
propiedades catalticas de las otras dos peroxidasas ligninolticas (Ruiz-Dueas
et al. 1999b) pudiendo oxidar lignina y compuestos de manganeso. En la Figura
16 se puede observar una vista axial de la regin del hemo de la VP donde se
muestra los tres sitios de la oxidacin y el ciclo cataltico de sta enzima. La VP
se ha utilizado en esta Tesis en ensayos de deslignificacin con
polioxometalatos (POMs).

31
1. Introduccin

Alcohol Hlice B
veratrlico
(VA) VA
Trp164 Glu36
VP VA
[Fe3+]
Glu40 ROOH Mn3+ C-IIB
[Fe3+ Trp ]
Mn2+ ABTS+
agua Mn2+
ROH

Grupo Hemo ABTS


Asp175
ABTS ABTS+

C-IA C-IIA
[Fe4+=O P] [Fe4+=O]
Mn2+ Mn3+
VA

VA
C-IB
ABTS [Fe4+=O Trp ]
(fenoles)

(a) (b)

Figura 16. Vista axial de la regin del hemo de la VP que muestra los tres sitios de la
oxidacin: Mn2+ (que incluye tres residuos acdicos de amino-cidos), ABTS (en el lmite del
grupo hemo) y VA (a travs de Trp164) (a) y ciclo cataltico propuesto para la VP (b) (Ruiz-
Dueas y Martnez 2010, Prez-Boada et al. 2005).

Los POMs son catalizadores de oxidacin conocidos por sus sntesis


orgnicas homogneas y heterogneas y se han sugerido como alternativas a los
reactivos de blanqueo basados en cloro y al blanqueo convencional alcalino con
oxgeno (Gamelas et al. 2007). Se utilizan en procesos de deslignificacin
oxidativa con oxgeno, cuyo objetivo principal es la oxidacin selectiva de la
lignina residual, pudiendo ser regenerados y reutilizados en el proceso de
deslignificacin. Se han intentado regenerar POMs por reoxidacin con lacasas,
pero los resultados no fueron muy esperanzadores para su aplicacin industrial
debido a los elevados tiempos de reoxidacin, que adems no era completa
(Gamelas et al. 2007, Gaspar et al. 2007, Tavares et al. 2004, Gamelas et al.
2008). En esta Tesis se han utilizado por primera vez peroxidasas para la
reoxidacin de POMs. En la Figura 17 se puede observar el ciclo cataltico
propuesto para la oxidacin de la lignina por el POM as como su reoxidacin
por la VP. Los POMs se caracterizan por ser aniones que se pueden visualizar
estructuralmente como conjuntos agregados metal-oxgeno (Pope 1983). La
unidad bsica y ms comn, los octaedros, est formada por un metal rodeado
de seis oxgenos (MO6) (Weinstock et al. 1997). Adems de M y O, otros
elementos que usualmente son designados por X, pueden formar parte de la
estructura de los POMs. Los POMs con estructura de Keggin son los ms
importantes y los ms estudiados actualmente ya que son ms estables y
fcilmente disponibles (Gamelas et al. 2003). El anin de Keggin est
compuesto por un tetraedro central XO4 rodeado por 12 octaedros MO6, metal-
oxgeno, compartiendo aristas y vrtices. Los octaedros se encuentran en cuatro
grupos M3O13, que comparten los tomos de oxgeno de los vrtices formando el
POM con la frmula [XM12O40]m-, en que XM12 es la frmula abreviada.

32
1. Introduccin

H2O
H2O2
Compuesto 1
(enzima activa)

POMred
H2O Mn2+

Ligninaox
Mn3+
POMox Lignina

Mn3+
POMred POMox
Lignina
Mn2+
Ligninaox

Compuesto 2

(a) (b)

Figura 17. Representacin polidrica de un polioxometalato con estructura de Keggin (a) y


ciclo cataltico propuesto para la oxidacin de la lignina y reoxidacin del POM por la VP en
los ensayos de deslignificacin realizados durante esta Tesis (b).

1.5.2. Degradacin enzimtica de lpidos: Control del pitch


El control enzimtico del pitch en pastas mecnicas de conferas mediante el
uso de lipasas, se puso en prctica como una operacin rutinaria en la
produccin industrial de papel a principios de los aos 90 (Hata et al. 1996) y
fue el primer caso en que una enzima se aplic con xito en el proceso de
produccin papelero. Las lipasas (Figura 18a) son un grupo de hidrolasas
producidas por una gran variedad de organismos. Los tratamientos de pasta con
lipasas se iniciaron en Japn con una enzima de Candida cylindracea (Irie 1990)
y se continuaron en diversas pruebas de fbrica utilizando una lipasa mejorada y
comercializada por Novo Nordisk (actualmente Novozymes) bajo el nombre
comercial de Resinase (Matsukura et al. 1990, Fujita et al. 1991, 1992,
Gutirrez et al. 2001), que es capaz de hidrolizar aproximadamente el 95% de
los triglicridos presentes en una pasta mecnica de pino. A lo largo de los aos
siguientes, varias fbricas en Japn y China han introducido esta tecnologa del
control del pitch basada en lipasas en pasteados mecnicos. Las lipasas actan
sobre los glicridos pero no sobre otros extrables lipdicos. Teniendo en cuenta
que los triglicridos se hidrolizan con facilidad en el pasteado alcalino (kraft y
sosa), las lipasas no son de inters para el control del pitch en estos procesos, as
como cuando otros compuestos como esteroles libres y conjugados, alcoholes
grasos y alcanos, son los responsables del pitch (del Ro et al. 1998, 1999, 2000,
Gutirrez y del Ro 2005). Otras enzimas de tipo hidrolasa estudiadas son las
esterol esterasas, que tambin podran ser efectivas para el control del pitch ya
que los steres de esteroles son a menudo causa de pitch, debido a su

33
1. Introduccin

pegajosidad y resistencia a la coccin alcalina. Sin embargo, estas enzimas


liberan esteroles libres que son tan problemticos como sus steres.

Por otro lado, las lacasas constituyen un grupo de enzimas oxidativas que
ha sido objeto de gran inters en el desarrollo de tecnologas respetuosas con el
medio ambiente (Mayer y Staples 2002). Como se ha comentado anteriormente,
la accin directa de las lacasas en principio est restringida a las unidades
fenlicas, pero en presencia de mediadores redox (Bourbonnais y Paice 1990,
Call 1994) amplan la accin de la lacasa a sustratos no fenlicos, lo que
aumenta su potencial en la degradacin de lignina y de otros compuestos
aromticos. Recientemente, se ha mostrado la gran eficacia del sistema lacasa-
mediador en la eliminacin de extrables lipoflicos de pastas de conferas,
frondosas as como de fibras no madereras (Gutirrez et al. 2006a, 2006c,
2006b, 2007, 2009, Molina et al. 2008).

Finalmente, hay que mencionar que se ha sugerido recientemente el uso


de una lipoxigenasa (de soja) para el control del pitch en pasta termomecnica
de conferas (Zhang et al. 2007). Las lipoxigenasas (Figura 18b) son una clase
de dioxigenasas que contienen hierro (no en forma hemo) y catalizan la
oxigenacin de cidos grasos insaturados y sus steres. El descubrimiento
reciente de una lipoxigenasa fngica (que contiene manganeso) ha revelado la
existencia de otro tipo de lipoxigenasas con una habilidad nica para oxidar
cidos grasos (Hamberg et al. 1998; Su and Oliw 1998). Esta lipoxigenasa se ha
evaluado en la Tesis para la eliminacin de lpidos y lignina.

(a) (b)

Figura 18. Estructura de una lipasa (a) y de una lipoxigenasa (b).

34
1. Introduccin

35
2

Sisal (Agave sisalana)


2. Objetivos

OBJETIVOS

En esta Tesis se aborda el estudio exhaustivo de la composicin qumica de


los principales constituyentes de diversos cultivos lignocelulsicos utilizados
como materia prima para la fabricacin de pasta de papel, as como de la
evolucin de los mismos durante el proceso de coccin y blanqueo. Estos
estudios tienen por objeto obtener un mejor aprovechamiento industrial de estas
fibras, que ayudar a optimizar los procesos de coccin y blanqueo utilizando
tecnologas menos contaminantes, incluyendo la biotecnologa.

Los objetivos especficos de esta Tesis son los siguientes:

- Realizar una caracterizacin qumica de los diferentes cultivos


lignocelulsicos seleccionados, poniendo especial nfasis en la
composicin de lpidos, lignina y hemicelulosas.

- Estudiar la modificacin estructural de los constituyentes orgnicos de


los diferentes cultivos lignocelulsicos durante los procesos de coccin
y blanqueo.

- Estudiar y desarrollar diferentes aplicaciones biotecnolgicas que


permitan degradar tanto la lignina residual como los compuestos
extrables lipoflicos presentes en las pastas de papel mediante
tecnologas menos contaminantes.

37
2. Objetivos

39
3

Camo (Cannabis sativa)


3. Material y Mtodos

MATERIAL Y MTODOS

3.1. MATERIALES
En esta Tesis se ha estudiado la composicin qumica de fibras de diversos
cultivos agrcolas utilizadas para la fabricacin de pastas de celulosa de alta
calidad y el comportamiento de sus componentes a lo largo del proceso de
fabricacin de dichas pastas. Entre los materiales estudiados se encuentran fibras
del tallo de varias plantas anuales, tales como lino, kenaf, camo y yute; y
fibras procedentes de hojas de sisal, abac y curau. Tambin se seleccionaron
pastas crudas y pastas blanqueadas (TCF y ECF). Tanto las muestras
correspondientes a las materias primas como sus pastas fueron suministradas por
la empresa Celulosa de Levante S.A., CELESA (Tortosa, Tarragona). Por otro
lado, tambin se estudiaron otras fibras de potencial aplicacin en el sector
procedentes de tallos de caa comn, as como fibras de residuos de poda de
rboles de tagasaste, que fueron suministradas por la Universidad de Huelva.
Finalmente, para la realizacin de los diversos tratamientos biotecnolgicos se
utilizaron pastas de eucalipto y de lino, suministradas por las empresas ENCE
(Pontevedra) y CELESA, respectivamente.

3.1.1. Cultivos lignocelulsicos

Lino
El lino (Linum usitatissimum) (Figura 19) es una planta herbcea
dicotilednea de la familia de las Linceas, originaria de Asia que se cultiva por
el aceite de su semilla y por las fibras de sus tallos. Se cultiva principalmente en
regiones fras y templadas como Europa, Asia, Australia, Argentina y Brasil.

Figura 19. Planta de lino (izquierda) y morfologa de sus fibras elementales (derecha) (Garca
Hortal 2007).

41
3. Material y Mtodos

De las 150 especies del gnero Linum, slo la especie L. usitatissimum


produce fibras tiles comercialmente. Las plantas de lino textil alcanzan una
altura de 0,9-1,25 m y un dimetro de 0,25-0,5 cm. Los haces fibrosos se extraen
fcilmente por un procedimiento denominado enriado, que consiste en un
proceso hidroltico en el que enzimas de hongos y bacterias disuelven las
pectinas que mantienen adheridas las fibras, proporcionando unas fibras muy
puras y de alta calidad que tienen una longitud de 30 a 90 cm. La fibra elemental
tiene una longitud comprendida entre 10 y 55 mm y un dimetro de 12 a 30 m,
estando entre las ms altas de todas las fibras utilizadas en la industria papelera,
siendo excedida slo por la fibra de algodn (Garca Hortal 2007).

La pasta procedente de fibras de lino es ideal para la produccin de papeles


delgados a los que se exige una gran resistencia, papeles densos y permanentes,
tales como el papel para cigarrillos, papeles muselina, papel biblia y papeles
para imprimir de bajo gramaje. Debido a su pureza y resistencia, a su capacidad
para soportar un refinado intenso y por consiguiente dar una pasta
extremadamente porosa, es ideal para bolsas de t, papeles para registro, papeles
de seguridad y papel moneda.

Camo

El camo (Cannabis sativa) (Figura 20) es una planta herbcea dicotilednea


de la familia de las Cannabinceas, originaria de Asia central que se cultiva por
el aceite y la protena de su semilla as como por las fibras de sus tallos. Se
cultiva principalmente en los pases de la antigua URSS y en China, pero se
encuentra tambin en la India, pases de Europa Central, Paquistn, Turqua,
Italia y Colombia. Es una planta que se acomoda a todos los climas, muy
competitiva con las malas hierbas, por lo que no exige grandes inversiones de
herbicidas (Struik et al. 2000).

Figura 20. Planta de camo (izquierda) y morfologa de sus fibras elementales (derecha)
(cedida por Garca Hortal).

42
3. Material y Mtodos

Las plantas de camo presentan tallos que pueden alcanzar alturas de 1 a 5 m


y dimetros entre 5 y 10 mm. Los haces fibrosos son ms largos que los de lino,
de 100-300 cm, ms rgidos y gruesos. La fibra elemental tiene caractersticas
morfolgicas similares a la del lino, aunque no son tan uniformes, son menos
transparentes y con numerosos nudos. Las fibras secundarias, ms cortas, son
ms delgadas y ms lignificadas (Garca Hortal 2007).

La pasta procedente de camo, al igual que la pasta de lino, es ideal para la


produccin de papeles especiales que requieren una fibra fuerte, tales como el
papel para cigarrillos, papel biblia, filtros de caf, bolsitas de t, papel aislante y
paales.

Kenaf
El kenaf (Hibiscus cannabinus) (Figura 21) es una planta herbcea
dicotilednea de la familia de las Malvceas, probablemente originaria de
frica. Se cultiva en regiones tropicales y subtropicales de la India, Sudeste de
Asia y Amrica Central. Es una planta de crecimiento rpido, que se est
desarrollando como fuente de fibras para la fabricacin de papel.

Las plantas de kenaf crecen en delgadas caas de hasta 6 m de altura y 4 cm


de dimetro. En el tallo, al igual que en el lino y camo, se distinguen dos
regiones distintas. Las fibras de la corteza exterior (bast), que constituye
alrededor de 30-40% del tallo, son moderadamente largas, mientras que las
fibras del ncleo (core) son ms cortas. La longitud de las fibras liberianas es de
2 a 6 mm, sin embargo, cabe destacar que las fibras liberianas de las plantas
tardas son ms cortas que las de las plantaciones normales, ya que el
crecimiento vegetativo de esta planta est influenciado por la duracin del da
(Garca Hortal 2007, Pande y Roy 1998).

Figura 21. Planta de kenaf (izquierda) y morfologa de sus fibras elementales (derecha)
(Garca Hortal 2007).

43
3. Material y Mtodos

Se pueden obtener pastas de kenaf tanto de la corteza exterior como del


ncleo, siendo las fibras largas de la corteza exterior especialmente adecuadas
para la fabricacin de papeles de calidad especial. Las pastas de kenaf son
ideales para papel de prensa, impresin y escritura, y se pueden usar en la
mayora de calidades de papel.

Yute
El yute (Corchorus capsularis) (Figura 22) es una planta herbcea
dicotilednea de la familia de las Tiliceas, que crece principalmente en climas
hmedos y clidos de Paquistn, India, Bangladesh, Brasil y otros pases
tropicales. Sus haces fibrosos, que tienen una longitud de 20 a 50 cm y
contienen de 10 a 30 fibras elementales, se usan principalmente para la
produccin de sacos, tapices, cuerdas, textiles y materiales para embalaje.

Las plantas de yute presentan tallos que pueden alcanzar alturas de 2,5-3 m y
10-20 mm de dimetro. La fibra elemental tiene una longitud variable de 1,5 a 5
mm, el menos elevado de las otras fibras liberianas mencionadas, y un ancho de
10 a 25 m. Los tallos son enriados en estanques de agua, donde se liberan los
haces de fibras liberianas de la corteza y la porcin leosa de la planta (Garca
Hortal 2007, Han 1998).

Dureza y durabilidad son las principales caractersticas aportadas por las


fibras de yute en mezclas con pastas kraft. Las pastas mecnicas o poco cocidas
de yute, debido a la mayor lignificacin de sus haces fibrosos respecto a los del
lino y camo, se usan para cartones, papeles de estraza y de embalaje, y las
pastas qumicas y blanqueadas se usan para papeles finos como bolsas de t y
papel de fumar.

Figura 22. Planta de yute (izquierda) y morfologa de sus fibras elementales (derecha)
(Garca Hortal 2007).

44
3. Material y Mtodos

Sisal
El sisal (Agave sisalana) (Figura 23) es una planta robusta de climas
tropicales originaria de Amrica Central y Mxico, cultivada actualmente en
Brasil, Venezuela, Tanzania, Kenia, Mozambique, Angola, Madagascar y otras
zonas tropicales. Es una planta monocotilednea de la familia de las Agavceas
y sus fibras, denominadas fibras duras, ms rgidas y bastas que las fibras
liberianas, se han utilizado durante mucho tiempo en las industrias textil y de
cordelera.

La planta de sisal crece hasta alturas de 2 m, con un tronco corto de 15-23 cm


de dimetro. Las fibras se extraen de las hojas, que tienen entre 100 y 150, tras
ser cortadas y descortezadas. El descortezado se debe hacer tan pronto como sea
posible una vez que se cortan las fibras, para reducir el deterioro de las mismas.
El espesor, longitud y resistencia de la fibra depende de la madurez de la hoja y
de su posicin a lo largo de sta, ya que las hojas ms maduras contienen las
fibras ms largas y bastas, y las fibras ms gruesas se hallan en el extremo
terminal de la hoja.

La hoja madura del sisal mide 1-2 m de largo, 10-15 cm de ancho y 6 mm de


espesor (en el centro) y las fibras se disponen longitudinalmente en la hoja. Los
haces fibrosos tienen longitudes de 60-150 cm, y la fibra elemental tiene una
longitud de 1-8 mm y un ancho de 8-40 m, (Garca Hortal 2007, McDougall et
al. 1993).

Las fibras del sisal se aplican fundamentalmente en la fabricacin de cuerdas


y papeles especiales como bolsas de t, papeles dielctricos y papel de filtro,
debido, sobre todo, a la alta porosidad de sus pastas. Por otra parte, tambin se
utilizan como refuerzo para los papeles delgados (Moore 1996)

Figura 23. Plantas de sisal (izquierda) y morfologa de sus fibras elementales (derecha)
(Garca Hortal 2007).

45
3. Material y Mtodos

Abac
El abac (Musa textilis) (Figura 24) es una planta perenne nativa de Filipinas
y que se cultiva actualmente tambin en Indonesia y Amrica tropical. Es una
planta monocotilednea que pertenece a la familia de las Musceas y sus fibras
tambin estn clasificadas como fibras duras.

La planta se asemeja a la del platanero, pero con hojas ms pequeas y frutos


no comestibles. Los tallos alcanzan alturas de 3 a 7,5 m y dimetros de 12 a 30
cm y consisten en un corazn central envuelto por vainas foliares. Las fibras
tiles comercialmente se encuentran en las vainas externas del tallo, que se
extraen por descortezado puramente mecnico de las vainas, de las que se
separan los haces por simple rasgadura.

Los haces fibrosos son muy largos, de hasta 2 m y la longitud de la fibra


elemental vara de 2,5 a 12 mm segn el espesor y la posicin en el tallo de la
vaina madre, donde las vainas ms externas tienen fibras ms cortas, ms
gruesas y de color ms oscuro, por lo que son las de menor calidad (Garca
Hortal 2007, Moore 1996).

Los papeles producidos por las fibras de abac son altamente porosos, por lo
que son ideales para filtros y envolturas de embutidos. Se considera una
excelente materia prima para papeles de alta calidad, como papel moneda,
paales, servilletas, papel tis, accesorios para hospitales, etc. El abac de
primera calidad tambin se emplea mezclado con algodn o pasta de madera en
la fabricacin de papeles superfinos, papel para imprimir de bajo gramaje, de
registro, moneda y seguridad y, sobre todo, papel para filtro poroso de uso en
laboratorio o industrial.

Figura 24. Planta de abac (izquierda) y morfologa de sus fibras elementales (derecha)
(cedida por Garca Hortal).

46
3. Material y Mtodos

Curau
El curau (Anans erectifolius) es una planta herbcea de fruto no comestible
(Figura 25a) perteneciente a la familia de las Bromeliceas, de origen americano
y hbitat tropical, muy comn en el Amazonas. Tiene un tallo tan corto que
parece carecer de l y con hojas rgidas. Normalmente produce entre 20 y 24
hojas, proporcionando aproximadamente 2 kg de fibra. En la ltima dcada, ha
ganado reconocimiento comercial como material para composites en la industria
del automvil (Leao et al. 1998, Silva et al. 2001). La fibra del curau tambin
se ha propuesto como una materia prima alternativa para la produccin de pastas
qumicas en Brasil.

Caa comn
La caa comn (Arundo donax) es una planta monocotilednea perenne,
perteneciente a la familia de las Poceas que parece ser originaria de Asia y que
ha colonizado el rea del Mediterrneo, en pases como Portugal y Espaa. Es
considerada una de las mayores gramneas, con una estructura tubular
segmentada semejante al bamb (Figura 25b), con alturas entre 2 y 8 m (Seca et
al., 2000). Debido a la fcil adaptabilidad de esta gramnea a diferentes
condiciones ecolgicas, a la elevada productividad de biomasa y capacidad de
cultura intensiva, es una de las especies no madereras ms atractivas como
fuente alternativa de fibras para la industria de pasta de papel (Shatalov y
Pereira 2002).

Tagasaste
El tagasaste (Chamaecytisus proliferus spp. palmensis) es un arbusto robusto
de crecimiento rpido perteneciente a la familia de las Fabceas. Es nativo de las
Islas Canarias y se cultiva en Australia, Nueva Zelanda y otros pases. Debido a
su alto contenido proteico, se usa como alimento para el ganado y tambin como
cultivo fijador del nitrgeno para mejorar la fertilidad del suelo. Con el fin de
fomentar la formacin de tallos mltiples, el arbusto debe ser podado con
regularidad, lo que conduce a una alta acumulacin de residuos de poda. Estos
residuos se han evaluado recientemente como materia prima alternativa para la
produccin de pasta de celulosa (Daz et al. 2004, Lpez et al. 2004, Jimnez et
al. 2006, 2007, Garca et al. 2008). En la Figura 25c se muestra una fotografa
de un arbusto de tagasaste.

47
3. Material y Mtodos

a c

Figura 25. Diversas plantas estudiadas como materias primas potenciales para la produccin
de pasta de celulosa: (a) curau, (b) caa comn y (c) tagasaste.

3.1.2. Pastas de papel

Pastas de fibras no madereras


Se han estudiado pastas de lino, camo, sisal y abac, tanto crudas (proceso
de coccin sosa-AQ) como blanqueadas (procesos de blanqueo TCF y ECF)
suministradas por CELESA. Las pastas crudas se obtuvieron por una coccin
sosa-AQ que emplea hidrxido de sodio y antraquinona (hasta 0,05%) como
agentes de coccin, durante 2-4 h a una temperatura de 160-170 C.

Las pastas blanqueadas se obtuvieron tras secuencias de blanqueo TCF y


ECF. La secuencia de blanqueo TCF (Q-Po), incluy una etapa quelato (Q),
seguida de una etapa con perxido de hidrgeno con oxgeno presurizado (Po).
La etapa de quelato se lleva a cabo para capturar los iones metlicos y evitar que
dichos iones destruyan el perxido de hidrgeno. La secuencia de blanqueo ECF
usada (D-Po) incluy una etapa de dixido de cloro (D), seguida por una etapa
de perxido de hidrgeno con oxgeno presurizado (Po).

Pastas de fibras madereras


Para la realizacin de los diferentes tratamientos biotecnolgicos se us pasta
kraft cruda (no blanqueada) de eucalipto (E. globulus) suministrada por ENCE.

48
3. Material y Mtodos

3.1.3. Enzimas y mediadores


En esta Tesis se han estudiado dos aplicaciones biotecnolgicas: (i) enzimas
de tipo lipoxigenasa para la degradacin de lpidos y lignina en pastas kraft de
eucalipto y pastas sosa-AQ de lino, y (ii) un sistema compuesto de un
polioxometalato y una enzima de tipo peroxidasa para la degradacin de lignina
en pastas kraft de eucalipto.

Lipoxigenasas
La lipoxigenasa utilizada se obtuvo del hongo ascomiceto Gaeumannomyces
graminis y fue suministrada por la empresa Novozymes (Bagsvaerd,
Dinamarca). La unidad de actividad de esta enzima se define como la cantidad
de enzima que da lugar a un aumento en la Absorbancia a 234 nm, de 0,001 por
minuto a pH 7 y 30C, cuando se usa el cido linoleico como sustrato (volumen
de reaccin = 1,0 mL y camino ptico = 1 cm). Los tratamientos enzimticos
con lipoxigenasa incluyeron reacciones con varios lpidos modelo, como alcanos
(nonacosano), alcoholes (octacosanol), cidos grasos (cido linoleico), esteroles
libres (sitosterol) y steres de esteroles (colesteril linoleato), representativos de
los extrables lipoflicos de las pastas estudiadas. El sitosterol fue suministrado
por Calbiochem y los dems compuestos por Sigma-Aldrich.

Peroxidasas
Se utiliz la peroxidasa verstil del hongo basidiomiceto Pleurotus eryngii
(VP), producida en el Centro de Investigaciones Biolgicas (CIB, CSIC,
Madrid) como se describe a continuacin.

La cepa de E. coli se cultiv en matraces de 1 L con 500 mL de medio TB


(Terrific Broth) y 100 g/mL de ampicilina. El medio TB consiste en un medio
tamponado para el crecimiento de la cepa de E. coli transformada con el
plsmido portador del gen que codifica la VP de P. eryngii. Se incub a 37C y
200 rpm durante 3 horas. Se indujo la expresin aadiendo isopropil tio--D-
galactopiransido (IPTG) a una concentracin final de 1 mM y se continu con
la incubacin durante 4 horas ms. El pellet bacteriano, que contiene la
protena recombinante en forma de cuerpos de inclusin, se obtuvo por
centrifugacin, se resuspendi en tampn de lisis (tampn con Tris-HCl pH 8,0
50 mM, EDTA pH 8,0 10 mM y DTT 5 mM), se le aadi lisozima a una
concentracin final de 2 mg/mL y se incub durante 1 hora en hielo. Se le
aadi 0,1 mg/mL DNAasa, se sonic en 3 ciclos de 1 minuto a 20.000 Hz de
frecuencia y se centrifug 30 min a 12.500 rpm para favorecer la precipitacin
de los cuerpos de inclusin.

49
3. Material y Mtodos

Una vez lavados los cuerpos de inclusin con una solucin de Tris-HCl pH
8,0 20 mM, EDTA pH 8,0 1 mM y DTT 5 mM, se resuspendieron en el mnimo
volumen posible de la misma solucin de lavado, se homogenizaron y
centrifugaron. Se resuspendieron de nuevo los cuerpos de inclusin en el
mnimo volumen de solucin desnaturalizante (solucin de Tris-HCl pH 8,0 50
mM, EDTA pH 8,0 1 mM, DTT 5 mM y urea 8 M), se incubaron a temperatura
ambiente durante 15 min y se centrifugaron para eliminar los restos no disueltos.
El sobrenadante, con la protena completamente desplegada, se diluy 5 veces
con una solucin de Tris-HCl pH 8,0 50 mM, EDTA pH 8,0 1 mM y DTT 1
mM, para llegar a una concentracin final de urea de 1,6 M y de 1-2 mg/mL de
protena. Esta solucin se aadi a la mezcla de replegado (CaCl2 5 mM,
glutatin oxidado 0,5 mM, Tris-HCl pH 9,5 50 mM y hemina 20 M) en una
proporcin 1:10 para obtener una concentracin final de urea de 0,16 M y se
incub en oscuridad a temperatura ambiente durante 16 horas.

La mezcla de replegado con la enzima activada se concentr por filtracin


tangencial hasta 80-100 mL, a travs de una membrana de 3 KDa de tamao de
poro (Membrane Casette de Filtron) con una bomba peristltica (Masterflex
modelo 7518-02), y posteriormente por ultrafiltracin con membrana de 3-10
KDa de tamao de poro en un sistema Amicon de Millipore hasta 20-40 mL.
Se dializ la mezcla concentrada en una solucin tampn acetato (acetato 20
mM, pH 4,3 y CaCl2 1 mM) y el material insoluble precipitado se elimin por
centrifugacin. Se dializ de nuevo en solucin tampn tartrato (tartrato 10 mM,
pH 5,5 y CaCl2 1 mM) y se purific por cromatografa lquida de alta resolucin
(HPLC) en un equipo kta de Amersham Pharmacia Biotech de intercambio
aninico previamente equilibrada con el mismo tampn. La protena se eluy
con un gradiente de NaCl de 0 a 0,3 M en tampn tartrato (tartrato 10 mM, pH
5,5, CaCl2 1 mM) con un flujo de 2 mL/min. Las fracciones obtenidas se
dializaron con tartrato 10 mM a pH 5,0 y se realiz el espectro UV-VIS
(espectrofotmetro Shimazdu UV-160) para verificar la correcta incorporacin
del hemo durante el proceso de replegado. La concentracin de la enzima pura
se determin a partir de su coeficiente de extincin molar (406= 150.000 M-1cm-1)
(Ruiz-Dueas et al. 1999b). La enzima se congel en nitrgeno lquido y se
conserv a -80C hasta su utilizacin.

Polioxometalatos
El POM utilizado fue [SiW11MnIII(H2O)O39]5- denominado SiW11MnIII en
forma simplificada. La solucin acuosa del POM se prepar a partir de
-K8[SiW11O39]13H2O, KMnO4 y Mn(CH3COO)24H2O. Se disolvieron 9,6 g
de K8[SiW11O39]13H2O en 13 mL de agua, a 95C. En vasos separados se
disolvieron, a temperatura ambiente, 0,095 g de KMnO4 en 10 mL de HCl
0,6mM y 10 mL de agua y 0,6 g de Mn(CH3COO)24H2O en 10 mL de agua. Al

50
3. Material y Mtodos

vaso conteniendo K8[SiW11O39]13H2O a 95C se le adicion poco a poco y


consecutivamente las disoluciones de los otros dos vasos y se dej unos 25
minutos a 95C. Se guard la disolucin preparada hasta el da siguiente para
filtrarla y diluirla para 100 mL de modo que la concentracin final fuera de
aproximadamente 30 mM.

3.2. MTODOS ANALTICOS

3.2.1. Aislamiento y anlisis de los compuestos lipoflicos de las fibras y


pastas
El anlisis de los compuestos extrables lipoflicos de las fibras y pastas
requiri su aislamiento previo. Dichos compuestos se extrajeron con acetona en
un extractor de tipo Soxhlet durante 8 horas. A continuacin se evapor el
disolvente a sequedad en un rotavapor y la cantidad de extracto se determin por
gravimetra. Los extractos lipoflicos obtenidos se redisolvieron en CHCl3 para
su posterior anlisis por cromatografa de gases (GC) y cromatografa de
gases/espectrometra de masas (GC/MS), descritos ms adelante.

Fraccionamiento de los compuestos extrables lipoflicos mediante SPE


Para una caracterizacin ms detallada de los compuestos presentes en los
extractos lipoflicos, se procedi a su aislamiento y purificacin mediante SPE
(extaccin en fase slida) segn un mtodo previamente descrito (Gutirrez et
al. 1998, 2004), tal como se muestra en la Figura 26.

Los extractos lipdicos se fraccionaron en cartuchos (500 mg) de


aminopropilo (Waters, Millipore). Los extractos secos (5-10 mg) se
resuspendieron en un volumen mnimo (<0,5 mL) de hexano-CHCl3 (4:1) con el
que se carg el cartucho, previamente acondicionado con hexano (4 mL). La
primera fraccin se eluy con 8 mL de hexano, que contiene los compuestos
ms apolares como steres de esteroles, ceras e hidrocarburos; la segunda
fraccin se eluy con 6 mL de hexano-CHCl3 (5:1), rica en triglicridos; la
tercera con 10 mL de CHCl3, que contiene principalmente esteroles y alcoholes
libres y por ltimo la cuarta fraccin se eluy con 10 mL de una mezcla de ter
etlico-cido actico (98:2), conteniendo compuestos cidos. Cada fraccin
aislada se sec con nitrgeno, se pes y se procedi a su anlisis por GC y
GC/MS.

51
3. Material y Mtodos

Esteroles

Hidrocarburos
esteroidales
steres de esteroles
Escualeno Triglicridos
cidos grasos

10 20 30 min

Extracto lipdico total en


Hexano /CHCl3 (4:1) 0.5 mL
Hexano Hexano Hexano /CHCl3 CHCl3 ter/AcOH
4 mL 8 mL (5:1) 6mL 10 mL (98:2) 10 mL

Fase aminopropilo

Acondicionamiento
A B C D
Escualeno
steres de
Hidrocarb. esteroles
esteroidales
Ceras
Triglicridos Esteroles
10 20 30 min
C16
C18:2
cidos grasos
10 20 30 min
C18:1
C26
10 20 30 min C18
C22C24 C28C
C20 30

10 20 30 min

Figura 26. Esquema del fraccionamiento de un extracto lipdico por SPE (Gutirrez et al.
2004).

52
3. Material y Mtodos

Mtodos de derivatizacin de los compuestos extrables lipoflicos


Para el anlisis por GC y GC/MS es esencial que los compuestos existentes en
la muestra sean suficientemente voltiles, por lo que es necesario recurrir a
mtodos de derivatizacin cuando los compuestos a analizar no son voltiles,
esto es, mtodos de conversin de ciertos compuestos en otros que sean
compatibles con el mtodo analtico de GC/MS. Las tcnicas empleadas en esta
Tesis incluyeron la metilacin de grupos carboxilo y la silanizacin de grupos
hidroxilo.

La metilacin de los grupos carboxlicos de cidos grasos libres,


hidroxicidos y grupos fenlicos se realiz con (trimetilsilil)diazometano
(TMSD) suministrado por Sigma-Aldrich. Para ello, una vez seca la muestra, se
aadi 100 l de metanol y 50 l de una solucin de TMSD 2,0 M en hexano y
se mantuvo 20 min en el bao de ultrasonidos. A continuacin se sec con
nitrgeno, para resuspenderla en CHCl3 y analizarla por GC/MS.

La silanizacin de los grupos hidroxilo de alcoholes, esteroles, etc, se realiz


con N,O-bis-(trimetilsilil)-trifluoroacetamida (BSTFA) suministrado por Sigma-
Aldrich. Para ello, una vez seca la muestra, se aadi 0,2 mL de BSTFA y 0,1
mL de piridina. A continuacin se calent a 70C durante 2 h y se sec con
nitrgeno. Posteriormente, se redisolvi en CHCl3 para analizarla por GC/MS.

Anlisis de los extractos lipoflicos mediante GC y GC/MS


Para el anlisis de los compuestos extrables lipoflicos por GC y GC/MS, las
caractersticas de las columnas cromatogrficas utilizadas fueron las adecuadas
para separar e identificar los compuestos de alto peso molecular como ceras,
steres de esteroles, triglicridos, etc. Previamente se haban realizado estudios
sobre procedimientos para el anlisis de los extractos lipoflicos de maderas
(Gutirrez et al. 1998a, 2004) por GC y GC/MS en donde se usaron diversas
columnas de diferente longitud y diferentes programas de temperatura. En estos
estudios, las columnas capilares seleccionadas para el anlisis de lpidos por GC
fueron de longitud corta (5 m) ya que proporcionan una conveniente elucin y
separacin de lpidos de alto peso molecular en un corto perodo de tiempo (20
min). Columnas menores de 5 m no son convenientes ya que no proporcionan la
resolucin necesaria para anlisis cuantitativos. En el caso de los anlisis 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 vaco a las que opera, no se pueden
usar columnas tan cortas, por lo que se usaron columnas de 10-15 m. Esta
longitud de columna es apropiada para el anlisis de lpidos de alto peso

53
3. Material y Mtodos

molecular por GC/MS proporcionando resultados en un perodo de tiempo corto


(30 min).

Los anlisis cromatogrficos de los extractos lipoflicos, tanto de las muestras


derivatizadas como sin derivatizar, se llevaron a cabo en un cromatgrafo de
gases Agilent 6890N equipado con un detector de ionizacin de llama (FID) y
una columna capilar corta de slice fundida (DB-5HT, J&W; 5 m x 0,25 mm ID
y 0,1 m de espesor de pelcula). El programa de calentamiento del horno
comenz a 100C (1 min), seguido de un incremento de temperatura hasta
350C (3 min) a 15C/min. Las temperaturas del inyector y del detector se
mantuvieron a 300C y 350C, respectivamente. El gas portador que se utiliz
fue Helio y la inyeccin se realiz en modo splitless.

El anlisis mediante GC/MS se llev a cabo en un cromatgrafo de gases


Varian 3800 acoplado a un detector de trampa de iones (ITD, Varian 4000),
usando una columna capilar de slice fundida (DB-5HT, J&W; 12 m x 0,25 mm
ID, con espesor de pelcula de 0,1 m). El horno se calent de 120C (1 min) a
380C (5 min) a 10C/min. La lnea de transferencia se mantuvo a 300C. La
temperatura del inyector se program de 120C (0,1 min) a 380C con una
rampa de 200C/min y mantenindose hasta el final del anlisis. El gas portador
utilizado fue Helio. La identidad de cada componente se determin por
comparacin de sus espectros de masas con los espectros existentes en las
libreras (Wiley y NIST) y con espectros publicados anteriormente, por sus
fragmentaciones y, cuando fue posible, por comparacin con patrones
suministrados por Sigma-Aldrich (octadecano, cido palmtico, sitosterol,
colesteril oleato y sitosteril 3-D-glucopiransido). Los picos cromatogrficos se
cuantificaron a partir de sus reas en los cromatogramas. Se utiliz una recta de
calibrado, realizada con los patrones anteriormente descritos. En todos los casos
se obtuvo un coeficiente de correlacin mayor de 0,99.

3.2.2. Aislamiento y anlisis de la lignina de las fibras y pastas

Determinacin del contenido en lignina


El contenido en lignina de las muestras se determin por el mtodo Klason
segn la norma Tappi T222 om-88 (Tappi 2004), con algunas modificaciones.
En este mtodo, las muestras molidas de las distintas fibras seleccionadas, libres
de compuestos extrables, se sometieron a una hidrlisis con H2SO4 al 72%
(p/p), a 30C durante 1 h. Posteriormente, la solucin se diluy hasta alcanzar
una concentracin del 4% en H2SO4 y se autoclavaron (1h a 110C). A
continuacin, las muestras se filtraron, guardndose los primeros 100 mL para el
posterior anlisis de los azcares libres y el residuo insoluble (lignina Klason) se

54
3. Material y Mtodos

lav con agua destilada hasta pH neutro y se sec para su cuantificacin


gravimtrica.

Aislamiento de la lignina de las fibras


La lignina se extrajo de las muestras, previa eliminacin de los compuestos
lipoflicos e hidrosolubles de las fibras, segn el protocolo desarrollado por
Bjrkman (1956) que consiste en extraer la lignina de la muestra finamente
molida. El grado de molienda deseado se alcanz utilizando un molino de bolas
centrfugo modelo Retsch S100 durante 100 h. La lignina de las muestras
finamente molidas se extrajo con dioxano-agua (9:1, utilizando 250 mL por cada
10 g de muestra) durante 12 h. Posteriormente, se centrifug (25 min, 4C,
11000 rpm) y se recogi el sobrenadante (que contiene la lignina) en un matraz,
repitindose este proceso dos veces de forma consecutiva, y se sec en rotavapor
a 40C. A continuacin, el residuo seco se disolvi en una mezcla de cido
actico:agua (9:1), aadindose 20 mL por cada gramo de residuo seco. La
lignina se precipit en agua destilada (225 mL por cada gramo de lignina) en
constante agitacin y se recogi tras centrifugacin. Una vez seco el residuo, se
tritur en un mortero de gata para facilitar su disolucin en una mezcla 1,2-
dicloroetano-etanol (2:1) y de nuevo se volvi a centrifugar a baja velocidad
(5000 rpm durante 5 min) recuperando el sobrenadante, que contiene la lignina.
ste se dispers gota a gota, sin agitar, en ter dietlico (225 mL es suficiente
para 0,5-1 g de lignina), precipitando de nuevo la lignina. Se volvi a centrifugar
(5000 rpm durante 5 min) y el residuo se resuspendi en ter de petrleo en el
que se dej durante 12 horas. Finalmente, se centrifug y se sec mediante
corriente de nitrgeno y se conserv a 4C, preservndola de la luz y el aire para
evitar su oxidacin hasta el momento del anlisis. Con este mtodo se obtuvo
una lignina poco degradada y representativa de la lignina nativa de las fibras.

Anlisis de la lignina mediante Py-GC/MS


La pirlisis es un mtodo degradativo que transforma compuestos complejos
no voltiles en una mezcla de fragmentos voltiles por descomposicin trmica
en ausencia de oxgeno (Meier y Faix 1992; Fullerton y Franich 1983) que se
lleva a cabo habitualmente a temperaturas de 400-800C. En la pirlisis se
producen roturas de los enlaces por accin del calor, ya que cuando la energa
aplicada a la molcula es mayor que la energa de enlaces especficos ocurre la
disociacin de stos de una forma predecible y reproducible, pudindose obtener
informacin sobre la molcula original a travs del anlisis de los productos de
degradacin. Los fragmentos resultantes de la pirlisis se pueden separar por GC
e identificar por MS. La Py-GC/MS es un mtodo poderoso para el anlisis de
materiales lignocelulsicos, especialmente de la lignina. La lignina se piroliza

55
3. Material y Mtodos

produciendo una mezcla de compuestos fenlicos que resultan de la rotura no


slo de enlaces ter, sino tambin de ciertos enlaces C-C, reteniendo estos
fenoles las caractersticas de sustitucin del polmero de lignina y siendo posible
por lo tanto identificar los diferentes componentes de ligninas provenientes de
unidades H, G y S. La Py-GC/MS presenta diversas ventajas frente a otros
mtodos degradativos, pues es una tcnica analtica rpida que proporciona
resultados en apenas un paso, que necesita poca cantidad de muestra y una
simple preparacin de la misma. Tambin presenta ventajas frente a los mtodos
clsicos de anlisis de la lignina, pues no es necesario aislar la lignina de la
muestra, permitiendo su anlisis in situ.

En la presente Tesis, la pirlisis de las muestras se llev a cabo en un


pirolizador Frontier Laboratories Ltd., a 500C durante 10 s. El pirolizador
estaba conectado a un cromatgrafo de gases Agilent 6890 con una columna
capilar HP 5MS (30 m x 0,25 mm ID, y un espesor de pelcula de 0,25 m)
acoplado a un espectrmetro de masas Agilent 5973 N. El cromatgrafo se
program de 50C (1 min) a 100C con un incremento de 30C/min y de 100 a
300C con un incremento de 10C/min. La temperatura final se mantuvo durante
10 min. El gas portador utilizado fue Helio con un flujo controlado de 1
mL/min. Para las pirlisis en presencia de hidrxido de tetrametilamonio
(TMAH), se adicion 0,5 L de TMAH 25% a 100 g de muestra. Los
compuestos obtenidos mediante Py-GC/MS se identificaron por comparacin
con la literatura (Faix et al. 1990, Ralph y Hatfield 1991) y con los incluidos en
las libreras de espectros de masas Wiley y NIST. Se calcularon las reas
molares para los productos de la pirlisis (lignina y carbohidratos), se
normaliz al 100% y se hizo una media entre las repeticiones de las pirlisis. La
desviacin estndar fue inferior al 5% de la media.

Anlisis de la lignina mediante DFRC


El mtodo de degradacin qumica de la lignina conocido como DFRC
(derivatization followed by reductive cleavage) es un mtodo simple y poderoso
que rompe de manera selectiva y eficiente los enlaces ter -O-4 y -O-4
presentes en la lignina. Permite el anlisis cuantitativo de las unidades
estructurales de la lignina eterificadas y tambin ofrece informacin sobre los
enlaces carbono-carbono a travs del anlisis de las estructuras dimricas que se
forman.

El mtodo DFRC incluye dos pasos fundamentales: (i) la solubilizacin de la


lignina por bromacin y acetilacin con bromuro de acetilo y (ii) la
fragmentacin reductora de los enlaces aril ter en la lignina con polvo de zinc.
La identificacin de los productos resultantes de la degradacin (monmeros y
dmeros) por GC/MS proporciona informacin valiosa sobre la estructura de la

56
3. Material y Mtodos

lignina (Lu y Ralph 1997a, 1997b, 1998). Una de las ventajas de este mtodo es
que deja intacto el carbono  de la cadena lateral de la lignina, por lo que lo hace
muy adecuado para estudiar la presencia de unidades aciladas (con acetatos, p-
cumaratos o p-hidroxibenzoatos) en el carbono . Sin embargo, el mtodo de
DFRC usa reactivos acetilantes que interfieren en el anlisis de los grupos
acetatos nativos en la lignina, pero con las modificaciones apropiadas, mediante
la sustitucin de la acetilacin por la propionilacin (DFRC), es posible obtener
una informacin significativa sobre la presencia de unidades acetiladas en la
lignina nativa (Ralph y Lu 1998). En la Figura 27a se puede observar la rotura
selectiva de los enlaces ter en el mtodo DFRC y en la Figura 27b la
modificacin de este mtodo (DFRC) para el anlisis de ligninas naturalmente
acetiladas.

La degradacin DFRC se llev a cabo con 10 mg de lignina aislada que se


trat con bromuro de acetilo en cido actico (8:92) durante 2 h a 50C. Despus
de proceder a la eliminacin del disolvente en rotavapor, se disolvi en una
mezcla de dioxano/cido actico/agua (5:4:1) con polvo de zinc y se dej en
agitacin durante 30-40 min a temperatura ambiente. A continuacin se ajust el
pH hasta un valor inferior a 3 por adicin de HCl 3% y se pas la disolucin a
un embudo de extraccin para separar las fases orgnica y acuosa por extraccin
lquido-lquido con diclorometano. A la fase orgnica extrada se le aadi
sulfato de sodio anhidro y se filtr, recogindose el filtrado para eliminacin del
disolvente en rotavapor. El residuo que qued tras la evaporacin, se acetil
durante 1 h con anhdrido actico, diclorometano y piridina. Se evapor el
disolvente en rotavapor con etanol, se resuspendi en diclorometano y se
procedi a su anlisis por CG/MS. En el caso de la DFRC el protocolo aplicado
fue el mismo pero sustituyendo todos los reactivos acetilantes por reactivos con
propionilo.

Los anlisis mediante GC/MS se llevaron a cabo en un cromatgrafo de gases


Varian Star 3400 acoplado a un detector de trampa de iones (ITD, Varian Saturn
2000), usando una columna capilar de slice fundida (DB-5HT, J&W; 12 m x
0,25 mm ID, con espesor de pelcula de 0,1 m). El horno se calent de 50C
(0,2 min) a 100C a 30C/min y se elev a 300C a 5C/min. El inyector y la
lnea de transferencia se mantuvieron a 300C. El gas portador utilizado fue
Helio. La cuantificacin de los monmeros se realiz usando tetracosano como
patrn externo asumiendo factores de respuesta similares a los obtenidos por Lu
y Ralph (1997).

57
3. Material y Mtodos

HO AcO OAc (a)


HO Br
O O

AcBr Zn
Ac2O/Py
CH3O OCH3 CH3O OCH3 CH3O OCH3
O O OAc

RO RO OR

HO Br
O O

AcBr Zn
Ac2O/Py
CH3O OCH3 CH3O OCH3 CH3O OCH3
O O OAc

(b)
HO PropO OProp

HO Br
O O

PropBr Zn
Prop2O/Py
CH3O OCH3 CH3O OCH3 CH3O OCH3
O O OProp

AcO AcO OAc

HO Br
O O

PropBr Zn
Prop2O/Py
CH3O OCH3 CH3O OCH3 CH3O OCH3
O O OProp

Figura 27. Rotura selectiva de los enlaces ter: (a) Mtodo DFRC (Lu y Ralph 1997a) y (b)
Mtodo DFRC modificado (DFRC) para el anlisis de ligninas naturalmente acetiladas
(Ralph y Lu 1998).

58
3. Material y Mtodos

Anlisis de la lignina mediante 2D-NMR


La NMR, tanto de 1H como de 13C, ofrece una informacin detallada de la
estructura de la lignina, incluyendo los diferentes tipos de unidades y los enlaces
que se establecen entre ellas (Robert 1992). En los ltimos aos, se ha
desarrollado la NMR bidimensional (2D-NMR) y tridimensional (3D-NMR), en
las que se establecen correlaciones 1H-13C, entre otras, que resuelve seales que
aparecan solapadas en los espectros unidimensionales (Capanema et al. 2001,
Ralph et al. 2001, Liiti et al. 2003). La 2D-NMR se considera en la actualidad
la tcnica ms potente para el anlisis de la estructura de la lignina y a travs de
ella se han podido identificar nuevas sub-estructuras tales como las
dibenzodioxocinas (Karhunen et al. 1995) o las espirodienonas (Zhang y
Gellerstedt 2001, Zhang et al. 2006).

El HSQC (Heteronuclear Single Quantum Correlation) proporciona


correlaciones a travs de acoplamiento escalar a un enlace entre un protn y el
heteroncleo al que est directamente unido. Los espectros HSQC de la lignina
presentan tres regiones bien diferenciadas: regin aliftica, regin aliftica
oxigenada y regin aromtica (Figura 28). La regin aliftica oxigenada (Figura
29) es la ms importante para el estudio de la estructura de la lignina ya que en
esta zona se encuentran las correlaciones de la mayora de los enlaces que tienen
lugar entre las distintas unidades estructurales. La regin aromtica (Figura 30)
es la ms importante desde el punto de vista de la composicin de la lignina, ya
que en esta zona aparecen las correlaciones de las distintas unidades H, G y S.

0
0

Regin
aliftica

50
50

Regin aliftica
C (ppm)
oxigenada

100
100

Regin
aromtica

150
150
ppm (t1

10
10.0 5
5.0 H (ppm) 0
0.0
ppm (t2)

Figura 28. Espectro HSQC de la lignina aislada de sisal donde se pueden observan sus tres
regiones caractersticas.

59
3. Material y Mtodos

-OMe J

J-OAc J-OH

D
E (J-Oac)
D D
E (J-OH)

R O

HO MeO
O MeO J 6
6 5 1
J 1
5
HO E 4 2
HO E 4 2 D 3
D 3 O
O
OMe
OMe 6
1
2
1
6 2 3
5
5 3 4
4 MeO OMe
MeO OMe
O
O

OMe
OMe
O O 1
3
2 4 6 2
3
4 3
5
O 1
6
5 2 4
J D OMe MeO
5 1 MeO OMe
6
D O
O
E E D
J E E
MeO D J
6
O HO J OH
5 1 1
6 2
4 2
3 5 3
O MeO
4
OMe
OMe O

Figura 29. Regin aliftica oxigenada del espectro HSQC de la lignina aislada de sisal y sus
correspondientes estructuras.

60
3. Material y Mtodos

S2,6 S2,6
S2,6

G2 2
G5
G6 6'

O O OH
HO
D D
D

S S S
1 1
1 6 2 6 2
6 2
5 3 5 3
5 3 4 4
4 MeO OMe MeO OMe
MeO OMe
OH O
O

OMe
HO O 1
6 2
3
D 4
5 3
2 4
MeO
5 1 MeO OMe
G
1 6
6 2 D O
3
O
5
4 J E D E
OMe
HO J OH
O 6
1
2

5 3
4
MeO

OMe
O

Figura 30. Regin aromtica del espectro HSQC de la lignina aislada de sisal y sus
correspondientes estructuras.

61
3. Material y Mtodos

Los espectros de NMR de ligninas se registraron en un espectrmetro Bruker


AVANCE 500 MHz, equipado con una sonda triple con gradientes en el eje z, a
una temperatura de 298K. Se disolvieron 40 mg en 0,75 mL de dimetilsulfxido
deuterado (DMSO-d6). Los experimentos HSQC se realizaron empleando una
matriz de datos de 256 incrementos en la dimensin indirecta, y adquiriendo
1024 puntos en la dimensin de adquisicin. La constante de acoplamiento 1JCH
utilizada fue de 140 Hz. La intensidad de las seales en los espectros HSQC
dependen del valor de esta constante as como del tiempo de relajacin T2
(Zhang y Gellerstedt 2007). Por ello, la integracin de las seales se realiz por
separado en cada una de las regiones del espectro, utilizando las seales
correspondientes a correlaciones C-H qumicamente anlogas, con constantes de
acoplamiento 1JCH similares. En la regin aliftica oxigenada, las abundancias
relativas de las diferentes subestructuras de la lignina se estimaron mediante la
integracin de las correlaciones C-H. En la regin aromtica, las
correlaciones 1H-13C de las unidades S y G se usaron para estimar las relaciones
S/G.

3.2.3. Aislamiento y anlisis de las hemicelulosas de las fibras y pastas

Preparacin de la holocelulosa y aislamiento de los xilanos

Las holocelulosas se obtuvieron por deslignificacin de las fibras y pastas


crudas (5,0 g) con cido peractico 10% (pH 3,5) durante 20 minutos a 85C.
Una vez ocurrida la deslignificacin, la holocelulosa se filtr, se lav con
acetona y agua destilada caliente y se sec para su determinacin gravimtrica.

Los xilanos se aislaron de las correspondientes holocelulosas (molidas


previamente) con DMSO a 50C durante 24 h y se precipitaron con un exceso de
7:2:1 etanol:metanol:agua acidificado con cido actico a 4C durante 3-4 das.
El precipitado se recogi tras centrifugacin, se lav con metanol y se sec
sobre vaco. En el caso de las pastas blanqueadas, los xilanos se aislaron
directamente de dichas pastas sin previo aislamiento de las holocelulosas
correspondientes.

Anlisis de azcares neutros tras hidrlisis cida


Los xilanos aislados se sometieron a una hidrlisis con H2SO4 72% durante 3
horas a 20C, seguido de otra hidrlisis con H2SO4 4% durante 2,5 horas a
100C. Los monosacridos neutros se determinaron como acetatos de alditol por
cromatografa de gases (Selvendran et al. 1979).

62
3. Material y Mtodos

El anlisis por cromatografa de gases se realiz en un equipo Varian 3350


con un detector FID y columna DB-225 J&W (30 m 0,25 mm ID y un espesor
de pelcula de 0,15 m). El programa de temperatura se program de 220C (5
min) a 230C con un incremento de 2C/min. La temperatura final se mantuvo
durante 5 min. Las temperaturas del inyector y detector fueron de 230C,
respectivamente. La cuantificacin se realiz con ayuda de curvas de calibracin
realizadas con patrones.

Anlisis de azcares neutros y cidos urnicos tras metanolisis cida


La fibra o pasta seca (4,5 mg) se someti a metanolisis cida aadiendo 2 mL
de una disolucin de HCl 2 M en metanol anhidro a 100C durante 4 horas
(Sundberg et al. 1996). Una vez fro, se adicion 80 L de piridina para
neutralizar la disolucin cida y se adicion 1 mL de una disolucin 0,1 mg/mL
de sorbitol como patrn interno. Se extrajo 2 mL del sobrenadante, se sec en un
rotavapor a 40-50C, se resuspendi en 70 L de piridina y se le aadi 150 L
de hexametildisiloxano y 80 L de trimetilclorosilano para promover la
sililacin durante 12 horas a temperatura ambiente.
Los anlisis de las muestras sililadas se llevaron a cabo en un cromatgrafo
de gases Hewlet-Packard 5890 equipado con un detector de masas MSD series
II, usando Helio como gas portador (35 cm/s) y una columna capilar (DB-1
J&W; 30 m x 0,32 mm ID y 0,25 m de espesor de pelcula). El horno se
calent de 100C a 175C a 4C/min y de 175C a 290C a 12C/min. La lnea
de transferencia se mantuvo a 300C. La temperatura del detector (FID) fue de
290C. La identidad de cada componente se determin por comparacin de sus
espectros de masas con los espectros existentes en las libreras (Wiley y NIST) y
con espectros publicados anteriormente (Sundberg et al. 1996, Bertaud et al.
2002, Bleton et al. 1996).

Determinacin del peso molecular de los xilanos mediante SEC


El peso molecular de los xilanos solubles en Me2SO se valor mediante
cromatografa de exclusin molecular. Los xilanos aislados con Me2SO se
disolvieron en una disolucin al 10% de LiCl en N,N-dimetilacetamida
(DMAC) y luego se diluyeron con DMAC hasta una concentracin de
aproximadamente 0,5% (5 mg/mL). Los anlisis por SEC se realizaron en
columnas de PLgel 10m (Mixed B 300 7,5 mm) con una pre-columna PLgel
10m (Polymer Laboratories, UK) usando un sistema PL-GPC 110 (Polymer
Laboratories). Las columnas, el inyector y el detector se mantuvieron a 70C
durante el anlisis. El eluyente fue LiCl en DMAC 0,1 M con un flujo
controlado de 0,9 mL/min. Las columnas se calibraron con patrones (Polymer

63
3. Material y Mtodos

Laboratories) en un rango de 0,8-100 kDa. El volumen de muestra que se


inyect fue de 100 L.

Anlisis de la estructura de los xilanos mediante NMR


Los xilanos aislados se analizaron tanto por 1H-NMR como por 2D-NMR.
Los espectros 1H-NMR se obtuvieron en un espectrmetro Bruker AVANCE
300. Se utiliz como patrn interno ( 0,00) 3-(trimetilsilil) propionato de sodio-
d4. Los experimentos 1H-NMR se realizaron a 30C con un pulso de 90, un
tiempo de relajacin de 16 s y adquiriendo 400 puntos.

Los espectros 2D-NMR se obtuvieron en un espectrmetro Bruker AVANCE


300. Los experimentos 2D 1H-1H COSY se realizaron a 50C usando una
secuencia COSY patrn (pulso de 90 y un tiempo de relajacin de 2 s). Los
experimentos 2D 1H-1H TOCSY (
mix= 0.050 s) se realizaron empleando una
anchura espectral de 2185 Hz en ambas dimensiones a 60C. El tiempo de
relajacin fue de 2 s. Se emplearon 128 acumulaciones para cada FID,
adquiriendo 1024 puntos en t1 512 en t2. Finalmente, los experimentos HSQC
se obtuvieron a 50C empleando una anchura espectral de 12,000 Hz (F1) y de
2000 Hz (F2), una matriz de datos de 2048 1024 incrementos y 128
acumulaciones. El tiempo de relajacin fue de 2 s y la constante de
acoplamiento 1JCH utilizada fue de 148 Hz.

Determinacin del contenido en cidos hexenurnicos


Durante el proceso kraft, los cidos 4-O-metil-D-glucurnicos, presentes en la
cadena lateral de los xilanos, se convierten en los cidos insaturados
correspondientes (cidos hexenurnicos, HexA) a travs de la prdida del grupo
metoxilo. Los HexA interaccionan con los agentes qumicos de blanqueo y otros
reactivos disminuyendo la blancura de las pastas. Los HexA interfieren en el
mtodo de determinacin del ndice kappa, ya que reaccionan con el
permanganato de potasio, haciendo que se obtenga un valor ms elevado del
contenido de lignina residual que el real (Gran y Liebing 1996).

El contenido en HexA se determin a travs del mtodo de la hidrlisis cida,


por el que los cidos hexenurnicos se convierten selectivamente en derivados
del furano por hidrlisis en tampn formato de sodio a pH 3,0 (Vuorinen et al.
1999). La determinacin de la cantidad de HexA, se basa en la cuantificacin de
los derivados del furano que se forman a travs del anlisis del espectro de
UV/VIS.

Se dejaron impregnar aproximadamente 0,75 g de pasta en 0,75 mL de


tampn formato de sodio 10 mM sobre agitacin durante una noche. Esta

64
3. Material y Mtodos

mezcla se transfiri al reactor, se substituy el aire por nitrgeno, se encendi la


agitacin mecnica y se llev la temperatura hasta 110C durante 1h. Una vez
acabada la reaccin se filtr la pasta y se lav, recogiendo siempre el lquido de
filtrado, diluyndose todo hasta 500 mL. Se ley entonces el valor de la
absorbancia a 245 nm y 480 nm. Se realiz tambin un ensayo en blanco.

El contenido en HexA existentes en las pastas se determin mediante la


frmula:
( A245  A480 )
C
8,7 u m

Siendo,

C = cantidad de HexA en la pasta (meq/Kg)

A245 = Absorbancia a 245 nm

A480 = Absorbancia a 480 nm

M = peso de la pasta seca (Kg)

8,7 mM-1cm-1 = coeficiente de extincin molar a 245 nm con relacin a los


HexA.

3.2.4. Otros anlisis

Determinacin de la fraccin hidrosoluble de las fibras


El porcentaje de compuestos hidrosolubles en las fibras se determin segn la
norma Tappi T 207 om-88 (Tappi 2004). Para ello, los cartuchos de las muestras
extradas con acetona, una vez secos, se colocaron en matraces con 100 mL de
agua destilada y se tuvieron en un bao a 100C durante 3 h, al cabo de las
cuales el extracto se concentr en rotavapor y se sec a 100C para su
determinacin gravimtrica.

Determinacin del contenido en cenizas de las fibras


El contenido en cenizas se determin mediante la norma Tappi 211 om-85
(Tappi 2004). Para ello se depositaron 200 mg de cada una de muestras en
crisoles de porcelana previamente tarados y se introdujeron en la mufla a 575 C
durante 6 h. Para tararlos se limpiaron con HCl y se introdujeron en la mufla a
575C durante 1 h. Posteriormente se sacaron los crisoles de la mufla y se

65
3. Material y Mtodos

pesaron una vez que alcanzaron la temperatura ambiente. Los contenidos en


cenizas se expresaron como porcentajes de la materia prima inicial.

Anlisis de metales y otros elementos en las fibras


Las fibras seleccionadas, una vez lavadas y secas, se molieron en un molino
de cuchillas y se les realiz una digestin con 4 mL de HNO3 concentrado por
0,5 mg de muestra, dejndolas 15 min en un horno microondas (Jones y Case
1990). Posteriormente se filtraron con filtro Whatman del nmero 2, y se
recogieron en un matraz que se enras hasta 50 mL. La concentracin de
metales en la disolucin obtenida se determin por espectrometra de emisin
por plasma (ICP-OES) en un espectrmetro Termo Jarrel Ash, modelo IRIS
Advantages.

3.2.5. Tratamientos enzimticos de las pastas

Tratamientos con lipoxigenasas


Los tratamientos de pastas de eucalipto y lino con lipoxigenasa se realizaron
con 5 g de pasta seca, al 1% de consistencia (peso/peso) en tampn dihidrgeno
fosfato sdico 100 mM (pH 7). La dosis de enzima fue de 10 mg lipoxigenasa/g
de pasta de eucalipto y 20 mg de lipoxigenasa/g de pasta de lino, la temperatura
de 30C, y el tiempo de reaccin de 4 horas. Los tratamientos se realizaron en
matraces de 1L, con burbujeo de oxgeno en un bao trmico con agitacin (170
rpm). En una etapa posterior, las pastas al 5% de consistencia, se sometieron a
una etapa de blanqueo con perxido, usando H2O2 al 3% (peso/peso) y NaOH
1,5% (peso/peso), ambos referidos al peso de la pasta seca, a 90C durante 2
horas, en unas bolsitas selladas de plstico termorresistente. Los controles para
la evaluacin de la accin de la lipoxigenasa se trataron bajo las mismas
condiciones pero sin enzima.

Una vez tratadas las pastas, se extrajeron con acetona en un extractor de tipo
Soxhlet durante 6 horas. A continuacin se evapor el disolvente a sequedad en
un rotavapor y los extractos lipoflicos obtenidos se redisolvieron en CHCl3 para
su posterior anlisis mediante GC y GC/MS, en las condiciones descritas
anteriormente para el anlisis de los compuestos lipoflicos. Se realizaron
anlisis posteriores de las propiedades de las pastas (blancura ISO, ndice kappa,
viscosidad intrnseca y cidos hexenurnicos).

Tambin se llevaron a cabo reacciones enzimticas con compuestos modelo.


Para las reacciones con compuestos modelo, se us 1 mg de cada compuesto y
se aadi 0,1 mg de lipoxigenasa, y Tween 20 como dispersante (1% v/v). El

66
3. Material y Mtodos

tratamiento se realiz a pH 7, usando tampn dihidrgeno fosfato sdico 100


mM, durante 2 horas. Se burbuje oxgeno dentro de los matraces de reaccin, y
la reaccin se llev a cabo en un bao con agitacin a una velocidad de 100 rpm.
En los controles de los experimentos, los lpidos se someten a las mismas
condiciones de reaccin, sin lipoxigenasa. En una etapa posterior se realiz una
fase de perxido, en la que se aada a cada matraz de reaccin 50 L de H2O2 al
30% (p/v) y 37,5 L de NaOH 5N, se taparon los matraces y se colocaron en el
bao trmico de agitacin, a 90C y 100 rpm, durante 2 horas. Las dispersiones
de los lpidos se secaron en rotavapor, se recogieron con cloroformo-metanol
(1:1), que se sec luego con nitrgeno, y se redisolvieron en cloroformo para su
anlisis mediante GC y GC/MS.

Tratamientos con POM y peroxidasa verstil


Los ensayos de blanqueo de pasta kraft de eucalipto con POM se llevaron a
cabo en un reactor PARR modelo 4842 (0,25 L) equipado con un sistema de
control de temperatura y mecanismo de agitacin (Figura 31).

Figura 31. Sistema usado para los ensayos de deslignificacin.

67
3. Material y Mtodos

Previamente a la realizacin de los ensayos de deslignificacin, se optimiz la


cintica de oxidacin del POM reducido (SiW11MnII) con peroxidasa verstil
(VP) para conocer las cantidades a utilizar de perxido y enzima en los
diferentes ensayos de deslignificacin. La reoxidacin del POM por la VP fue
monitorizada por espectroscopa de UV/VIS. El color amarillo del POMred,
SiW11MnII, cambia gradualmente durante la oxidacin con la VP a un color rosa
caracterstico del POMox, SiW11MnIII, (Figura 32a), mostrando una banda de
transicin d-d* con un mximo de Absorbancia a 490-495 nm (Figura 32b). Se
realizaron varias pruebas en tampn acetato 0,1 M, pH 4,5 con objeto de
maximizar la oxidacin del POM con VP/H2O2, variando para esto la relacin
H2O2/POM y POM/VP. Las lecturas se realizaron a un valor de longitud de onda
fijo (490 nm) para intervalos de tiempo de 1 minuto, considerando  de
(SiW11MnIII)= 327 y de (SiW11MnII)= 19 en cm-1mol-1L. Todas las medidas
espectrofotomtricas se realizaron en un espectrofotmetro Jasco V-560 UV/Vis
a temperatura ambiente.

Una vez conocidas las condiciones ptimas para la reoxidacin del POM, se
procedi a la realizacin de los diferentes ensayos de deslignificacin. Para cada
uno de los ensayos, se dej la pasta (8 g) en agua destilada (600 mL) durante
una noche en agitacin constante. Se filtr, se pas al reactor y se adicion 67
mL de tampn acetato de sodio 0,2 M pH 4,5 (concentracin final de 0,1 M),
disolucin de POM concentrado (entre 12,5 a 13,5 mL para que la concentracin
final de POM en la mezcla fuera de aproximadamente 3 mM) y agua destilada
hasta 134 mL de volumen total (considerando todava la contribucin del agua
retenida en la pasta filtrada), para obtener una consistencia final del 6%. Se
presuriz el reactor con pO2=5 bar, se encendi la agitacin mecnica y se llev
hasta la temperatura de 110C. Una vez acabado el ensayo, se filtr la pasta y se
lav con agua destilada (POM 1-). En los casos en que los lquidos de filtrado
conteniendo el POM fueron reoxidados (-VP-), se adicion a estos la enzima y
el perxido de hidrgeno en las cantidades necesarias (obtenidas a travs de los
ensayos de optimizacin descritos anteriormente), manteniendo agitacin
constante y a temperatura ambiente (para no desnaturalizar la enzima), durante
aproximadamente 20 min (aunque a los seis minutos la reoxidacin ya se
estimaba como completa). El filtrado conteniendo el POM reoxidado se aadi
nuevamente a la pasta filtrada reiniciando una nueva etapa de deslignificacin
(-POM 2-).

Finalmente, todas las pastas deslignificadas se sometieron a una extraccin


alcalina con NaOH (-E) 2% durante 1 hora a 70 C. Despus de las extracciones
alcalinas, se lavaron las pastas con agua destilada, hasta alcanzar pH neutro en el
lquido de filtrado y se dejaron secar a temperatura ambiente durante 4 das. Se
realizaron anlisis posteriores de las propiedades de las pastas (blancura ISO,
ndice kappa, viscosidad intrnseca y cidos hexenurnicos).

68
3. Material y Mtodos

(a) (b)

Figura 32. (a) Colores caractersticos de SiW11MnIII (izquierda) y de SiW11MnII (derecha) y


(b) Espectros de UV/VIS de disoluciones acuosas de a: SiW11MnII, b-d: diferentes porcentajes
de oxidacin hasta e: SiW11MnIII (la oxidacin mxima).

Determinacin de las propiedades de las pastas

Determinacin de la blancura ISO

Las medidas de blancura en porcentaje ISO de las pastas fueron realizadas en


ENCE y UPC (Terrassa) en el caso de los tratamientos con lipoxigenasas y en el
Instituto Raz de Aveiro (Portugal) en el caso de las pastas tratadas con POM y
peroxidasa verstil.

El mtodo para la determinacin de la blancura ISO mide el factor de


reflectancia difusa en azul (grado de blancura ISO) de pastas de papel y
cartones. El alcance de esta norma est restringido a pastas de papel y cartones
blancos o casi blancos.

Determinacin del ndice kappa

El procedimiento estndar utilizado en la industria para determinar el grado


de deslignificacin en una pasta qumica es la determinacin del ndice kappa
por la norma TAPPI T 236cm-85 (Tappi 2006) y consiste en el volumen en mL
de una disolucin de KMnO4 0,1 N consumido por 1 g de pasta. La lignina de la
pasta reacciona con el permanganato y la cuantificacin del permanganato
consumido se determina por valoracin con tiosulfato de sodio.

Con ayuda de una batidora de mano, se desintegraron los gramos de pasta


necesarios para el ensayo (ver Anexo 1) en 140 mL de agua destilada y se lav
el pie de la batidora con 50 mL de agua. Con agitacin constante, se adicion

69
3. Material y Mtodos

una mezcla de 25 mL de KMnO4 0,1 N y 25 mL de H2SO4 0,2 N. Al cabo de 5


minutos se midi la temperatura y despus de 10 minutos se par la reaccin con
5 mL de KI 1,0 N, se colocaron unas gotas de solucin indicadora de almidn al
0,2% y se valor el I2 liberado con una disolucin de Na2S2O3 0,2 N.
Previamente, se realiz siempre un ensayo en blanco que no contena pasta.

El ndice kappa se calcul por la expresin:

pu f
IK >1  0,013(25  t )@
w

(b  a) N
p
0,1

Con:

IK = ndice kappa.

p = Volumen de permanganato de potasio 0,1 N consumido en el ensayo


(mL).

f = factor de correccin para un consumo de 50% de permanganato de potasio


y que depende de p (Anexo 2).

w = peso de pasta seca (g).

b = volumen consumido de tiosulfato de sodio para determinacin del blanco


(mL).

a = volumen consumido de tiosulfato de sodio para determinacin de la


muestra (mL).

N = normalidad de la disolucin de tiosulfato de sodio.

t = Temperatura del medio de reaccin (C).

Determinacin de la viscosidad intrnseca

La viscosidad de las pastas est directamente relacionada con el grado de


polimerizacin de las molculas de celulosa y por lo tanto con la resistencia de

70
3. Material y Mtodos

las fibras. Se determin la viscosidad intrnseca de las pastas a travs de la


norma SCAN-CM 15:85 (SCAN 1994). Este mtodo permite determinar la
viscosidad de las pastas celulsicas solubles en una disolucin de cobre (II)-
etilendiamina (CED) en un viscosmetro capilar.

Se pes la cantidad de pasta necesaria para el ensayo (ver Anexo 3) y se


transfiri a frascos de 61 mL, se adicion 25 mL de agua destilada y 5 hilos de
cobre. Se colocaron los frascos a agitar durante aproximadamente 30 minutos en
un agitador de brazos. Seguidamente, se adicion 25 mL de CED 1,0 M para dar
lugar a la disolucin de los polisacridos. Se complet el volumen del frasco con
disolucin de CED 0,5 M usando una bureta, se cerraron los frascos para no
dejar burbujas de aire y se colocaron de nuevo sobre agitacin durante otros 30
minutos. Finalmente, se anot el tiempo de escurrido de 1 mL de pasta disuelta
en CED, repitiendo la lectura 3 veces para cada frasco, usando un viscosmetro
capilar a una temperatura controlada de 25C.

La viscosidad relativa de las pastas fue determinada por la expresin:

rel = htn

Con:

h = constante del viscosmetro, obtenida por calibracin (0,0928 s-1).

tn = tiempo de escurrido (s).

A partir de la tabla que se encuentra en anexo (Anexo 4), se lee el valor del
producto [ ]C, que corresponde al valor de la viscosidad relativa obtenido:

rel = [ ]C

Con:

[ ] = viscosidad intrnseca (mL/g).

C = concentracin de la pasta seca en CED (g/mL).

71
3. Material y Mtodos

Determinacin del contenido en cidos hexenurnicos

La determinacin del contenido de cidos hexenurnicos (HexA) en las pastas


tratadas con POM se realiz segn el procedimiento explicado anteriormente en
el apartado 3.2.3.

72
3. Material y Mtodos

73
4

Abac (Musa textilis)


4. Referencias

REFERENCIAS

Abarca, R. y Blanco, M. L. (2008) Soda- Antraquinone Pulp of Tectona grandis in


Costa Rica. V Congreso Iberoamericano de investigacin en celulosa y papel,
CIADICYP, Octubre 2008, Guadalajara, Jalisco, Mxico.
Adler, E. (1977) Lignin chemistry - Past, present and future. Wood Science and
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.
Ali, M. y Sreekrishnan, T. R. (2001) Aquatic toxicity from pulp and paper mill
effluents: A review. Advances in Environmental Research, 5, 175-196.
Allen, L. H. (2000) Pitch control in paper mills. En: Pitch control, wood resin and
deresination (Eds.: Back, E.L. y Allen, L.H.), TAPPI Press, Atlanta, pp. 307-
328.
Annergren G. E. (1996) Pulp bleaching. Principles and practice. Captulo VII-3:
Strength properties and characteristics of bleached chemical and
(chemi)mechanical pulps. Ed. C. W. Dence y D. W. Reeve. Tappi press, pp:
717-748
Back, E. L. y Allen, L. H. (2000) Pitch control, wood resin and deresination.
TAPPI Press, Atlanta.
Bajpai, P., Anand, A., Sharma, N., Mishra, S. P., Pramod K. Bajpai P. K. y
Lachenal, D. (2006) Enzymes improve ECF Bleaching of pulp. BioResources, 1
(1), 34-44.
Bajpai, P. (1999) Application of enzymes in the pulp and paper industry.
Biotechnology Progress, 15, 147-157. ISSN 8756-7938
Bajpai, P. (2004) Biological bleaching of chemical pulps. Critical Reviews in
Biotechnology, 24, 1-58.
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.
Bjrkman, A. (1956) Studies on finely divided wood. Part I. Extraction of lignin
with neutral solvents. Svensk Papperstindning, 13, 477-485.
Bertaud, F., Sundberg, A. y Holmbom, B. (2002). Evaluation of acid methanolysis
for analysis of wood hemicelluloses and pectins. Carbohydrate Polymers, 48,
319-324.

75
4. 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.
Bleton, J., Mejanelle, P., Sansoulet, J., Goursaud, S. y Tchapla, A. (1996)
Characterization of neutral sugars and uronic acids after methanolysis and
trimethylsilylation for recognition of plant gums. Journal of Chromatography A,
720, 27-49.
Boerjan, W., Ralph, J. y Baucher, M. (2003) Lignin biosynthesis. Annual Review
of Plant Biology, 54, 519-546.
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.
Brett, C. y Waldron, K. (1996) Physiology and Biochemistry of Plant Cell Walls.
Brooks, R. T., Edwards, L. L., Nepote, J. C. y Caldwell, M. R. (1994) Bleach plant
closeup and conversion to TCF: A case study using mill data and computer
simulation. Tappi Journal, 77, 83-92.
Brunow, G. (2001) Methods to reveal the structure of lignin. Biopolymers.Vol.1 -
Lignin, Humic substances and Coal, 1, 89-116.
Bryce J.R. (1990) Produccin de pulpa al sulfito. En Pulpa y Papel: Qumica y
Tecnologa Qumica. Vol. 1. Casey, J.P. Ed Limusa, Mxico, 29-64.
Call, H. P (1994) Verfahren zur Vernderung, Abbau oder Bleichen von Lignin,
ligninhaltigen Materialien oder hnlichen Stoffen. Patent (International) WO
94/29510.
Call, H. P. y Mcke, I. (1997) History, overview and applications of mediated
lignolytic systems, especially laccase-mediator-systems (Lignozym(R)-process).
Journal of Biotechnology, 53, 163-202.
Camarero, S., Sarkar, S., Ruiz-Dueas, F. J., Martnez, M. J. y Martnez, A. T.
(1999) Description of a versatile peroxidase envolved in natural degradation of
lignin that has both Mn-peroxidase and lignin-peroxidase substrate binding sites.
The Journal of Biological Chemistry, 274, 10324-10330
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.
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.

76
4. Referencias

del Ro, J. C., Gutirrez, A. y Gonzlez-Vila, F. J. (1999) Analysis of impurities


occurring in a totally-chlorine free-bleached Kraft pulp. Journal of
Chromatography A, 830, 227-232.
del Ro, J. C, Gutirrez, A., Gonzlez-Vila, F. J., Martn, F. y Romero, J. (1998)
Characterization of organic deposits produced in the Kraft pulping of Eucalyptus
globulus wood. Journal of Chromatography A, 823,457-465
del Ro, J. C., Gutirrez, A. y Martnez, 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 Ro, J. C., Marques, G., Rencoret, J., Martnez, A. T. y Gutirrez, A. (2007)
Occurrence of naturally acetylated lignin units. Journal of Agriculture and Food
Chemistry, 55, 5461-5468.
del Ro, J. C., Rencoret, J., Marques, G., Gutirrez, A., Ibarra, D., Santos, J. I.,
Jimnez-Barbero, J., Zhang, L. y Martnez, A.T. (2008a) Highly acylated
(acetylated and/or p-coumaroylated) native lignins from diverse herbaceous
plants. Journal of Agriculture and Food Chemistry, 56, 9525-9534.
del Ro, J. C., Rencoret, J., Marques, G., Gutirrez, A., Ibarra, D., Santos, J. I.,
Jimnez-Barbero, J. y Martnez, A.T. (2008b) Occurrence and structural
characteristics of highly acylated (acetylated and/or p-coumaroylated) native
lignins from diverse herbaceous plants. 10th European Workshop on
lignocelluloses and Pulp. Stockholm Sweden, 26-29 August.
del Ro J. C., Romero J. y Gutirrez A. (2000) Analysis of pitch deposits produced
in Kraft pulp mills using a totally chlorine free bleaching sequence. Journal of
Chromatography A, 874, 235-245
Daz, M. J., Alfaro, A., Garca, M. M., Eugenio, F. M. E., Ariza, J. y Lpez, F.
(2004) Ethanol pulping from tagasaste (Chamaecytisus proliferus L.F. ssp.
Palmensis). A new promising source for cellulose pulp. Industrial &
Engineering Chemistry Research, 43, 1875-1881.
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.
Faix, O., Meier, D. y 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.
FAO (2004) Base de datos de la FAO "FAOSTAT".

77
4. Referencias

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.
Fuji, K., Ichikawa, K., Node, M. y Fujita, E. (1979) Hard acid and soft nucleophile
system. New efficient method for removal of benzyl protecting group. Journal
of Organic Chemistry, 44, 1661-1664.
Fujita, Y., Awaji, H., Matsukura, M. y Hata, K. (1991) Enzymic pitch control in
papermaking process. Kami Pa Gikyoshi, 45, 905-921
Fujita, Y., Awaji, H., Taneda, H., Matsukura, M., Hata, K., Shimoto, H., Sharyo,
M., Sakaguchi, H. y Gibson, K. (1992) Recent advances in enzymic pitch
control. Tappi Journal, 75 (4), 117-122
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.
Fullerton, T. J. y Franich, R. A. (1983) Lignin analysis by pyrolysis-GC-MS.
Holzforschung, 37, 267-269.
Gamelas J. A. F., Cavaleiro A., Santos I., Balula M. 2003. Os polioxometalatos.
Do anio de Keggin s nanocpsulas. Boletim da Sociedade Portuguesa de
Qumica. 90, 45-51.
Gamelas, J. A. F., Evtuguin, D. V. y Gaspar, A. R. (2008) Transition metal
complexes in the delignification catalysis. In: Varga, B., Kis, L. (Eds.),
Transition metal chemistry: New research. Nova Science Publishers, Inc., New
York, pp. 15-57.
Gamelas J. A. F., Pontes A. S. N., Evtuguin D. V., Xavier A. M. R. B. y Esculcas
A. P. (2007) New polyoxometalete-laccase integrated sistem for kraft pulp
delignification. Biochemical Engineering Journal, 33, 141-147.

78
4. Referencias

Gaspar, A. R., Gamelas, J. A. F., Evtuguin, D. V. y Neto, C. P. (2007) Alternatives


for lignocellulosic pulp delignification using polyoxometalates and oxygen: a
review. Green Chemistry, 9, 717-730.
Garca Hortal. (2007) Fibras Papeleras. Ediciones UPC (Universitat Politcnica de
Catalunya), Terrassa (Spain).
Garca Hortal, J. A. y Colom, J. F. (1992) El proceso al sulfato. Vol. I. Universitat
Politcnica de Catalunya, Terrassa (Spain).
Garca, M. M., Lpez, F., Alfaro, A., Ariza, J., Tapias, R. (2008) The use of
tagasaste (Chamaecytisus proliferus) from different origins for biomass and
paper production. Bioresource technology, 99(9), 3451-7.
Gellerstedt, G. y Li, J. (1996) An HPLC method for the quantitative determination
of hexenuronic acid groups in chemical pulps. Carbohydrate Research, 294, 41-
51.
Gellerstedt, G. Pranda J. y Lindfors E-L. (1994) Structural and molecular
properties of residual birch kraft lignins. Journal of Wood Chemistry and
Technology, 14(4), 467-482.
Gilarranz, M. A., Oliet, M., Rodrguez, F. and Tijero, J. (1999) Methanol-based
pulping of Eucalyptus globulus. Canadian Journal of Chemical Engineering, 77,
515-521.
Glenn, J. K., Morgan, M. A., Mayfield, M. B., Kuwahara, M. y Gold, M. H. (1983)
An extracellular H2O2-requiring enzyme preparation involved in lignin
biodegradation by the write rot basidiomycete Phanerochaete chrysosporium.
Biochemical and Biophysical Research Communications, 114, 1077-1083.
Grabber, J. H., Quideau, S. y Ralph, J. (1996) p-Coumaroylated syringyl units in
maize lignin: Implications for -ether cleavage by thioacidolysis.
Phytochemistry, 43, 1189-1194.
Gutirrez, A. y del Ro, J. C. (2001) Gas chromatography-mass spectrometry
demonstration of steryl glycosides in eucalypt wood, kraft pulp and process
liquids. Rapid Communications in Mass Spectrometry, 15, 2515-2520.
Gutirrez, A. y del Ro, J. C. (2005) Pitch episodes produced during the
manufacturing of high-quality paper pulps from hemp fibers. Bioresourse
Technology, 96, 1445-1450.
Gutirrez, A., del Ro, J. C., Gonzlez-Vila, F. J. y Martn, 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.

79
4. Referencias

Gutirrez, A., del Ro, J. C., Gonzlez-Vila, F. J. y Martn, F. (1999) Chemical


composition of lipophilic extractives from Eucalyptus globulus Labill. wood.
Holzforschung, 53, 481-486.
Gutirrez, A., del Ro, J. C., Ibarra, D., Rencoret, J., Romero, J., Speranza, M.,
Camarero, S., Martnez, M. J. y Martnez, A. T. (2006a) Enzymatic removal of
free and conjugated sterols forming pitch deposits in environmentally sound
bleaching of eucalypt paper pulp. Environmental Science and Technology, 40,
3416-3422.
Gutirrez, A., del Ro, J. C. y Martnez, A. T. (2004) Chemical analysis and
biological removal of wood lipids forming pitch deposits in paper pulp
manufacturing. In Spencer, J. F. T. (ed), Protocols in Environmental
Microbiology. Humana Press, Totowa, USA.
Gutirrez, A., del Ro, J. C. y Martnez, A. T. (2009) Microbial and enzymatic
control of pitch in the pulp and paper industry. Applied Microbiology and
Biotechnology, 82, 1005-1018.
Gutirrez, A., del Ro, J. C., Martnez, M. J. y Martnez, A. T. (2001) The
biotechnological control of pitch in paper pulp manufacturing. Trends in
Biotechnology, 19, 340-348.
Gutirrez, A., del Ro, J. C., Rencoret, J., Ibarra, D. y Martnez, A. T. (2006b)
Main lipophilic extractives in different paper pulp types can be removed using
the laccase-mediator system. Applied Microbiology and Biotechnology, 72, 845-
851.
Gutirrez, A., del Ro, J. C., Rencoret, J., Ibarra, D., Speranza, A. M., Camarero,
S., Martnez, M. J. y Martnez, A. T. (2006c) Sistema enzima-mediador para el
control de los depsitos de pitch en la fabricacin de pasta y papel. Patent
(International) ES2282020B1.
Gutirrez, A., Rencoret, J., Ibarra, D., Molina, S., Camarero, S., Romero, J., del
Ro, J. C. y Martnez, A. T. (2007) Removal of lipophilic extractives from paper
pulp by laccase and lignin-derived phenols as natural mediators. Environmental
Science & Technology, 41, 4124-4129.
Hamberg, M., Su, C. y Oliw, E. (1998) Manganese lipoxygenase - Discovery of a
bis-allylic hydroperoxide as product and intermediate in a lipoxygenase reaction.
The Journal of Biological Chemistry, 273, 13080-13088.
Han, J. S. (1998) Properties of Nonwood Fibers. Proceedings of the Korean
Society of Wood Science and Technology Annual Meeting.

80
4. Referencias

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.
Hata, K., Matsukura, M., Taneda, H. y Fujita, Y. (1996) Mill-scale application of
enzymatic pitch control during paper production. In: Viikari, L. and Jeffries, T.
W. (eds) Enzymes for pulp and paper processing. ACS, Washington, pp 280-
296.
Higuchi, T. (1997) Biochemistry and molecular biology of wood. Springer Verlag,
London.
Hillis, W. E. (1962) Wood extractives. Academic Press, London.
Hillis, W. E. y Sumimoto, M. (1989) Effect of extractives on pulping. In Rowe, J.
W. (ed), Natural products of woody plants. II. Springer-Verlag, Berlin, pp. 880-
920.
Hoareau, W., Trindade, W. G., Siegmund B., Castellan, A. y Frollini E. (2004)
Sugar cane bagasse and curaua lignins oxidized by chlorine dioxide and reacted
with furfuryl alcohol: characterization and stability. Polymer Degradation and
Stability, 86, 567-576.
Irie, Y. (1990) Enzymatic pitch control in papermaking system. Abs 1990
Papermakers Conference, 23-25 April, Atlanta.
Jimnez, L., Prez, A., Rodrguez, A. y de La Torre, M.J. (2006) New raw
materials and pulping processes for production of pulp and paper. Afinidad 63
(525), 362-369.
Jimnez, L., Prez, A., de la Torre, M.J., Moral A. y Serrano, L. (2007)
Characterization of vine shoots, cotton stalks, Leucaena leucocephala and
Chamaecytisus proliferus, and of their ethyleneglycol pulps. Bioresource
Technology, 98, 3487-3490.
Jones, J. B. y Case, V. W. (1990) Sampling, handling, and analyzing plant tissue
samples. Soil Testing and Plant Analysis, 389-427.
Karhunen, P., Rummakko, P., Sipila, J., Brunow, G. y Kilpelinen, I. (1995)
Dibenzodioxocins - A novel type of linkage in softwood lignins. Tetrahedron
Letters, 36, 169-170.

81
4. Referencias

Kawai, S., Umezawa, T. y Higuchi, T. (1987a) p-Benzoquinone monoketals, novel


degradation products of -O-4 lignin model compounds by Coriolus versicolor
and lignin peroxidase of Phanerochaete chrysosporium. FEBS Letters, 210, 61-
65.
Kawai, S., Umezawa, T., Shimada, M., Higuchi, T., Koide, K., Nishida, T.,
Morohoshi, N. y Haraguchi, T. (1987b) C-C cleavage of phenolic -1 lignin
substructure model compound by laccase of Coriolus versicolor. Mokuzai
Gakkaishi, 33, 792-797.
Kuwahara, M., Glenn, J. K., Morgan, M. A. Y Gold, M. H. (1984) Separation and
characterization of two cellular H2O2-dependent oxidases from ligninolytic
cultures of Phanerochaete chrysosporium. FEBS Letters, 1969, 247-250.
Lapierre, C., Monties, B. y Rolando, C. (1985) Thioacidolysis of lignin:
Comparison with acidolysis. Journal of Wood Chemistry and Technology, 5,
277-292.
Lapierre, C., Monties, B. y Rolando, C. (1986) Preparative thioacidolysis of spruce
lignin: Isolation and identification of main monomeric products. Holzforschung,
40, 47-50.
Lapierre, C., Pollet, B. y Monties, B. (1991) Thioacidolysis of spruce lignin: GC-
MS analysis of the main dimers recovered after Raney nickel desulphuration.
Holzforschung, 45, 61-68.
Leao, A. L., Rowell, R. y Tavares, N. (1998) Applications of natural fibres in
automotive industry in Brazil-Thermoforming process. In: Science and
Technology of Polymers Advanced Materials. Emerging Technologies and
Business Opportunities; Prasad, P. N., Mark, J. E., Kandil, S. H., Kafafi, Z. H.,
Eds., Plenum Press: New York, 755-761.
Lennholm, H. y Henriksson, G. (2007) Chapter 4. Cellulose. In Fibre and Polymer
Technology, K. I. I. (ed), Ljungberg Textbook Pulp and Paper Chemistry and
Technology Book 1. Wood Chemistry and Technology.
Liiti, T. M., Maunu, S. L., Hortling, B., Toikka, M. y Kilpelinen, I. (2003)
Analysis of technical lignins by two- and three-dimensional NMR spectroscopy.
Journal of Agricultural and Food Chemistry, 51, 2136-2143.
Lin, S. Y. y Dence, C. W. (1992) Methods in lignin chemistry. Springer-Verlag,
Berlin.
Lpez, F., Alfaro, A., Garca, M.M., Diaz, M.J., Calero, A.M. y Ariza, J. (2004)
Pulp and paper from tagasaste (Chamaecytisus proliferus ssp palmensis).
Chemical Engineering Research and Design, 82, 1029-1036.

82
4. Referencias

Lu, F. 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. y Ralph, J. (1997b) The DFRC method for lignin analysis. Part 1. A new
method for Earyl ether cleavage: lignin model studies. Journal of Agricultural
and Food Chemistry, 45, 4655-4660.
Lu, F. y Ralph, J. (1998) The DFRC method for lignin analysis. 2. Monomers from
isolated lignin. Journal of Agricultural and Food Chemistry, 46, 547-552.
Martnez A. T., Rencoret J., Marques G., Gutirrez A., Ibarra D., Jimnez-Barbero
J., and del Ro J. C. (2008) Monolignol acylation and lignin structure in some
nonwoody plants: A 2D-NMR study. Phytochemistry, 69, 2831-2843.
Martnez, M. J., Ruiz-Dueas, F. J., Guilln, F. y Martnez, A. T. (1996)
Purification and catalytic properties of two manganese-peroxidase isoenzymes
from Pleurotus eryngii. European Journal of Biochemistry, 237, 424-432.
Matsukura, M., Fujita, Y. y Sakaguchi, H. (1990) On the use of Resinase for
pitch control. Novo Publication. A, 6122, 1-7.
Mayer, A. M. y Staples, R. C. (2002) Laccase: new functions for an old enzyme.
Phytochemistry, 60, 551-565.
McDougall, G. J., Morrison, I. M., Stewart, D., Weyers, J. D. B. y Hillman, J. R.
(1993) Plant fibres: Botany, chemistry and processing for industrial use. Journal
of the Science of Food and Agriculture, 62, 1-20.
Meier, D. y Faix, O. (1992) Pyrolysis-gas chromatography-mass spectrometry. In
Lin, S. Y. and Dence, C. W. (eds), Methods in lignin chemistry. Springer-
Verlag, Berlin, pp. 177-199.
Molina, S., Rencoret, J., del Ro, J. C., Lomascolo, A., Record, E., Martnez, A. T.
y Gutirrez, A. (2008) Oxidative degradation of model lipids representative for
main paper pulp lipophilic extractives by the laccase-mediator system. Applied
Microbiology and Biotechnology, 80, 211-222.
Morrison, W. H. III., Akin, D. E., Himmelsbach, D. S. y Gamble, G. R. 1999.
Chemical, microscopic, and instrumental analysis of graded flax fibre and yarn.
Journal of the Science of Food and Agriculture, 79, 3-10.
Neto, C. P., Seca, A. M. L., Fradinho, D., Coimbra, M. A., Domingues, F. M. J.,
Evtuguin, D., Silvestre, A. J. D. y Cavaleiro, J. A. S. (1996) Chemical
composition and structural features of the macromolecular components of
Hibiscus cannabinus grown in Portugal. Industrial Crops and Products, 5, 189-
196.

83
4. Referencias

Node, M., Hori, H. y Fujita, E. (1976) Demethylation of aliphatic methyl ethers


with a thiol and boron trifluoride. Journal of the Chemical Society, Perkin
Transactions, 1, 2237-2240.
Norma Finlandesa TYOHJE C/6 15/07/98.
Paice, M. G., Bourbonnais, R., Reid, I. D., Archibald, F. S. y Jurasek, L. (1995)
Oxidative bleaching enzymes: A review. Journal of Pulp and Paper Science, 21,
J280-J284.
Pande, H. y Roy, D. N (1998) Influence of fibre morphology and chemical
composition on the papermaking potential of kenaf fibres. Pulp & Paper
Canada, 99(11), 31-34.
Parhan A., Gray R. L. (1990) The practical identification of wood pulp fibers.
Tappi Press. USA.
Prez-Boada, M., Ruiz-Dueas, F. J., Pogni, R., Basosi, R., Choinowski, T.,
Martnez, M. J., Piontek, K. y Martnez, A. T. (2005) Versatile peroxidise
oxidation of high redox potential aromatic compounds: site-directed
mutagenesis, spectroscopic and crystallographic investigations of three long-
range electrn transfer pathways. Journal of Molecular Biology, 354, 385-402.
Pope M. T. Springer Verlag. (1983) Heteropoly and Isopoly Oxometalates.
Inorganic Chemistry Concepts, V 8. Berlin.
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., Hatfield, R. D., Piquemal, J., Yahiaoui, N., Pean, M., Lapierre, C. y
Boudet, A. M. (1998) NMR characterization of altered lignins extracted from
tobacco plants down-regulated for lignification enzymes cinnamyl-alcohol
dehydrogenase and cinnamoyl-CoA reductase. Proceedings of the National
Academy of Sciences, 95, 12803-12808.
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, J. y Lu, F. (1998) The DFRC method for lignin analysis. 6. A simple
modification for identifying natural acetates in lignin. Journal of Agricultural
and Food Chemistry, 46, 4616-4619.

84
4. Referencias

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.
Rigol, A., Latorre, A., Lacorte, S. y Barcel, D. (2004) Bioluminescence inhibition
assays for toxicity screening of wood extractives and biocides in paper mill
process waters. Environmental Toxicology and Chemistry, 23, 339-347.
Robert, D. (1992) Carbon-13 nuclear magnetic resonance. In Lin, S. Y. and Dence,
C. W. (eds), Methods in lignin chemistry. Springer-Verlag, Berlin, pp. 250-273.
Rolando, C., Monties, B. y Lapierre, C. (1992) Thioacidolysis. In Lin, S. Y. and
Dence, C. W. (eds), Methods in lignin chemistry. Springer-Verlag, Berlin, pp.
334-349.
Rowe, J. W. (1989) Natural products of woody plants. I and II. Chemicals
extraneous to the lignocellulosic cell wall. Springer-Verlag, Berlin.
Ruiz-Dueas, F. J. y Martnez, A. T. (2010) Structural and functional features of
peroxidases with a potential as industrial biocatalysts. In: Torres, E. y Ayala, M.
(eds.). Biocatalysis Based on Heme Peroxidases (ISBN: 978-3-642-12626-0).
Springer.
Ruiz-Dueas, F. J., Martnez, M. J. y Martnez, A. T. (1999a) Heterologous
expression of Pleurotus eryngii peroxidase confirms its ability to oxidize Mn2+
and different aromatic substrates. Applied and Environmental Microbiology, 65,
4705-4707.
Ruiz-Dueas, F. J., Martnez, M. J. y Martnez, A. T. (1999b) Molecular
characterization of a novel peroxidase isolated from the ligninolytic fungus
Pleurotus eryngii. Molecular Microbiology, 31, 223-236.
Santos, A., Rodrguez, F., Gilarranz, M. A., Moreno, D. y Garca-Ochoa, F. (1997)
Kinetic modeling of kraft delignification of Eucalyptus globulus. Industrial &
Engineering Chemistry Research, 36, 4114-4125.
Saito, K. y Fukushima, K. (2005) Distribution of lignin interunit bonds in the
differentiating xylem of compression and normal woods of Pinus thunbergii.
Journal of Wood Science, 51, 246-251.
Sarkanen, K. V. y Hergert, H. L. (1971) Classification and distribution. In
Sarkanen, K. V. and Ludwig, C. H. (eds), Lignins - Occurrence, Formation,
Structure and Reactions. Wiley-Interscience, New York, pp. 43-94.
Scandinavian Pulp Paper and Board Committee, 1994. SCAN Test Methods.
Sweden.

85
4. Referencias

Seca, A. M., Cavaleiro, J. A. S., Domingues, F. M. J., Silvestre, A. J. D, Evtuguin,


D. y Neto, C. P. (2000) Structural characterization of the lignin from the nodes
and Internodes of Arundo donax Reed. Journal of Agricultural and Food
Chemistry, 48, 817-824.
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.
Selvendran, R. R., March, J. F. y Ring, S. G., (1979). Determination of aldoses and
uronic acids content of vegetable fiber. Analytical Biochemistry, 96, 282-292.
Shatalov, A. A. y Pereira, H. (2002) Influence of stem morphology on pulp and
paper properties of Arundo donax L. reed. Industrial Crops and Products, 15,
77-83.
Shimizu, K. (2001) Wood and Cellulosic Chemistry, 2nd edition, Eds. D.N.-S.
Hon, N. Shiraishi, Marcel Dekker Inc., New York, USA, 177-214.
Sigoillot, C., Camarero, S., Vidal, T., Record, E., Asther, M., Prez-Boada, M.,
Martnez, M.J., Sigoillot, J.C., Asther, M., Colom, J. y Martnez, A.T. (2005)
Comparison of different fungal enzymes for bleaching high-quality paper pulps.
Journal of Biotechnology, 115, 33334.
Silva, G. S., Assis, M. B. y Barbosa, W. L. R. (2001) Investigaao fitoqumica e
microbiologica da espcie Ananas erectifolius (curau). ReV. Virtual Iniciaao
Acad. UFPA, 1.
Sjstrm, E. y Westermark, U. (1999) Chemical Composition of Wood and Pulps:
Basic constituents and Their Distribution. In: Analytical Methods in Wood
Chemistry, Pulping and Papermaking. E. Sjstrm and R. Aln (Eds.), Springer-
Verlag, Berlin, Germany, Chap. 1, pp. 1-19.
Sjstrm, E. (1981) Wood Chemistry: Fundamentals and Applications. Academic
Press, Orlando, pp. 6882.
Sjstrm, E. (1993) Wood chemistry. Fundamentals and Applications. Academic
Press, San Diego.
Streitwieser, A. Jr. y Heathcock, C. H. (1983). Qumica Orgnica. 3 edicin.
Importecnica, S.A. 726-730.
Struik, P. C., Amaducci, S., Bullard, M. J., Stutterheim, N. C., Venturi G. y
Cromack, H. T. H. (2000) Agronomy of fibre hemp (Cannabis sativa L.) in
Europe. Industrial Crops and Products, 11, 107-118.
Su, C. y Oliw, E. (1998) Manganese lipoxygenase - Purification and
characterization. The Journal of Biological Chemistry, 273, 13072-13079.

86
4. Referencias

Sundberg, A., Sundberg, K., Lillandt, C. y Holmbom, B. (1996). Determination of


hemicelluloses and pectins in wood and pulp fibers by acid methanolysis and
gas chromatography. Nordic Pulp and Paper Research Journal, 4, 216-226.
Tappi (2004) 2004-2005 TAPPI Test Methods. TAPPI Press, Norcoss, GA 30092,
USA.
Tappi (2006) 2006-2007 TAPPI Test Methods. TAPPI Press, Norcoss, GA 30092,
USA.
Tavares, A. P. M., Gamelas, J. A. F., Gaspar, A. R., Evtuguin, D. V. y Xavier, A.
M. R. B. (2004) A novel approach for the oxidative catalysis employing
polyoxometalate-laccase system: application to the oxygen bleaching of kraft
pulp. Catalysis Communications, 5, 485-489.
Tien, M. y Kirk, T. K. (1983) Lignin-degrading enzyme from the hymenomycete
Phanerochaete chrysosporium Burds. Science, 221, 661-663.
Viikari, L., Kantelinen, A., Sundquist, J. y Linko, M. (1994) Xylanases in
bleaching - From an idea to the industry. FEMS Microbiology Reviews, 13, 335-
350.
Vuorinen, T., Fagerstrm, P., Buchert, J., Tenkanen, M., y Teleman, A. (1999).
Selective hydrolysis of hexenuronic acid groups and its application in ECF and
TCF bleaching in kraft pulps. Journal of Pulp and Paper Science, 25(5), 155-
162.
Watanabe, T. (2003) Analysis of native bonds between lignin and carbohydrate by
specific chemical reactions. In Springer-Verlag, B. H. (ed), Timell, T.E. (Ed.).
Springer series in wood science, Association between lignin and carbohydrates
in wood and other plant tissues. pp. 91-130.
Weinstock, I. A., Atalla, R.H., Reiner, R. S., Moen, M.A. Hammel, K. E.,
Houtman, C. J., Hill, C. L. y Harrup, M. K. (1997) A new environmentally
bening technology for transforming wood pulp into paper. Engineering
polyoxometalates as catalysts for multiple processes. Journal of Molecular
Catalysis A: Chemical, 116, 59-84.
Zhang, L. y Gellerstedt, G. (2001) NMR observation of a new lignin structure, a
spiro-dienone. Chemical Communications, 2744-2745.
Zhang, L. M. 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.

87
4. Referencias

Zhang, L. M., Gellerstedt, G., Ralph, J. y Lu, F. C. (2006) NMR studies on the
occurrence of spirodienone structures in lignins. Journal of Wood Chemistry and
Technology, 26, 65-79.
Zhang, X., Nguyen, D., Paice, M. G., Tsang, A. y Renaud, S. (2007) Degradation
of wood extractives in thermo-mechanical pulp by soybean lipoxygenase.
Enzyme and Microbial Technology, 40, 866-873.

88
4. Referencias

89
5

Kenaf (Hibiscus cannabinus)


5. Resultados y discusin

RESULTADOS Y DISCUSIN

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


muestran a continuacin en forma de publicaciones. Durante la realizacin de
esta Tesis se han publicado los resultados principales de los estudios realizados
sobre: i) la composicin qumica de la lignina, lpidos y hemicelulosas de
diversos cultivos lignocelulsicos (Publicaciones I-VI); ii) el comportamiento de
los constituyentes orgnicos de diversos cultivos lignocelulsicos durante la
fabricacin de pastas de papel de alta calidad, que incluyen procesos de coccin
sosa-AQ y blanqueo TCF y ECF (Publicaciones VII-VIII); y iii) el desarrollo de
dos procedimientos biotecnolgicos para la degradacin de lignina y lpidos
residuales en pastas de papel (Publicaciones IX-X).

91
5. Resultados y discusin

Publicacin I:

Marques G., Rencoret J., Gutirrez A., del Ro J.C. (2010) Evaluation of the
chemical composition of different non-woody plant fibers used for pulp and paper
manufacturing. The Open Agriculture Journal (in press).

92
5. Resultados y discusin

Evaluation of the chemical composition of different non-woody plant fibers


used for pulp and paper manufacturing

Gisela Marques, Jorge Rencoret, Ana Gutirrez, Jos C. del Ro

Instituto de Recursos Naturales y Agrobiologa de Sevilla, CSIC, P.O. Box 1052, 41080-
Seville, Spain

Abstract
The chemical composition of several non-woody plant fibers (bast fibers from
flax, hemp, kenaf, jute; leaf fibers from sisal, abaca and curaua; and giant reed),
which are used as raw materials for pulp and papermaking, has been evaluated.
Particular attention was paid to the composition of the lipophilic compounds and
the structure of the lignin polymer since they are important components of the
fiber that strongly influence the pulping and bleaching performances.
Keywords: non-woody fibers; flax, hemp, kenaf, jute, sisal, abaca, giant reed,
paper pulp; lipophilic extractives; lignin

1. Introduction
An alternative to woody raw materials for pulp and paper production in
developing countries is the use of non-woody fibers from field crops and
agricultural residues. In developed countries, non-woody fibers are mainly used
for the production of specialty papers, i.e., tea bags, filter papers, bank notes,
etc. On the other hand, there is a growing need within Europe to consider
alternative agricultural strategies that move an agricultural industry purely
focused on food production to one that also supplies the needs of other industrial
sectors, such as paper and textiles. Non-wood fibers, therefore, could become
important raw materials in this transformation [1-3]. The main sources of non-
woody raw materials are agricultural residues from monocotyledons, including
cereal straw and bagasse, or plants grown specifically for the fiber, such as
bamboo, reeds, and some other grass plants such as flax, hemp, kenaf, jute, sisal,
or abaca. Non-woody plants offer several advantages including short growth
cycles, moderate irrigation requirements and low lignin content, which in
principle would result in reduced energy and chemicals consumption during
pulping [4].
Plant fibers are constituted by three structural polymers (the polysaccharides
cellulose, and hemicelluloses and the aromatic polymer lignin) as well as by
some minor non-structural components (i.e. proteins, extractives, minerals).
Pulping and bleaching performances are highly dependent on the relative
content, structure and reactivity of the plant components. In particular, the lignin
content and its composition in terms of p-hydroxyphenyl (H), guaiacyl (G) and
syringyl (S) moieties and the different inter-unit linkages are important factors

93
5. Resultados y discusin

in pulp production affecting the delignification rate. It has been shown that
higher S/G ratios in woods implied higher delignification rates, less alkali
consumption and therefore higher pulp yield [5]. On the other hand, among the
non-structural components, lipophilic extractives present special relevance due
to their high impact in paper pulp manufacturing [6]. Lipophilic extractives
include different classes of compounds (i.e. alkanes, fatty alcohols, fatty acids,
free and conjugated sterols, terpenoids, triglycerides and waxes), which have
different behavior during pulping and bleaching [6-8]. These lipophilic
compounds, even when present in low amounts in the raw material, may play an
important role during the industrial wood processing for pulp and paper
production since they are at the origin of the so-called pitch deposits. 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 [6].
In order to maximize the exploitation of non-woody plant fibers for paper
pulp production, a more complete understanding of its chemistry is required.
Most studies have been devoted to the chemical characterization of woody
materials, while studies on non-woody fibers have been comparatively scarce. In
this context, the main objective of this work is to revise and evaluate the
chemical composition of different non-woody plant fibers used for pulp and
papermaking, that will help improving the industrial processes in which they are
used as raw materials.

2. Analytical methodologies
2.1. Samples
The samples selected for this study were bast fibers from flax (Linum
usitatissimum), hemp (Cannabis sativa), kenaf (Hibiscus cannabinus) and jute
(Corchorus capsularis); leaf fibers from sisal (Agave sisalana), abaca (Musa
textilis) and curaua (Ananas erectifolius); as well as giant reed (Arundo donax).

2.2. Chemical analyses


For hemicellulose and Klason lignin content estimation, milled samples were
extracted with acetone in a Soxhlet apparatus for 8h and subsequently extracted
with hot water (3h at 100 C). The acetone extracts were evaporated to dryness
and resuspended in chloroform for chromatographic analysis of the lipophilic
fraction. Klason lignin was estimated as the residue after sulfuric acid hydrolysis
of the pre-extracted material. The acid-soluble lignin was determined, after
filtering off the insoluble lignin, by spectrophotometric determination at 205 nm
wavelength. Neutral sugars from polysaccharide hydrolysis were analyzed as
alditol acetates by GC according to Tappi rules T222 om-88 and T249 om85,
respectively [9]. Ash content was estimated as the residue after 6h at 575 C.

94
5. Resultados y discusin

2.3. Analysis of lipids


The broad range of molecular masses of lipophilic extractives and their
structural diversity represent two important difficulties for their chemical
analysis. High-temperature, short-length (5m) capillary columns with thin films
was used for the rapid identification and quantification of lipophilic wood
extractives with no prior derivatization nor fractionation [10], resulting in an
optimal analysis of high-molecular-weight lipids such as waxes, sterol esters,
and triglycerides. This method enables elution and separation of compounds
with a wide range of molecular weights (from fatty and resin acids to sterol
esters and triglycerides) in the same chromatographic analysis. For GC-MS,
medium-length high-temperature capillary columns (12 m) were used [10].
When a more accurate characterization of some compounds was required, the
extracts were fractionated by solid-phase extraction (SPE) procedures [10-11].

2.4. Analysis of lignin


Pyrolysis coupled to gas chromatography/mass spectrometry (Py-GC/MS) was
used for the in situ analysis of the chemical composition of the lignin, in terms
of their H:G:S distribution. Lignin is thermally degraded to produce a mixture of
relatively simple phenols, which result from cleavage of ether and certain C-C
inter-unit linkages. The released methoxylated phenols retain the substitution
patterns of the different lignin monomers, and it is thus possible to identify
components from the p-hydroxyphenylpropanoid (H), guaiacylpropanoid (G)
and syringylpropanoid (S) lignin units [5, 12, 13]. For a more detailed structural
study, the milled wood lignins were isolated according to a known procedure
[14] and analyzed by bidimensional nuclear magnetic resonance (2D-NMR).
2D-NMR can provide information of the structure of the whole macromolecule
and is a powerful tool for lignin structural elucidation since signals overlapping
in the 1H and 13C NMR spectra are resolved revealing both the aromatic units
and the different interunit linkages present in lignin [15-18].

3. Characterization of the selected non-woody plant fibers


3.1. Morphological characteristics of the fibers
The morphological characteristics of a fiber, such as fiber length and width, are
important parameters in estimating pulp qualities. Fiber length is the most
important physical property for pulping as it generally influences the tearing
strength of paper. Greater the fiber length, higher will be the tearing resistance
of paper. On the other hand, longer fibers tend to give a more open and less
uniform sheet structure. Table 1 shows the morphological characteristics of the
non-woody fibers used for this study [19]. An important feature of non-wood
fibers is the wide variability among the lengths of the fibers of different species.
Some of these fibers have short lengths (i.e. giant reed, with only 1180 m fiber

95
5. Resultados y discusin

length), similar to the short fibers of hardwoods, while others, and particularly
flax and hemp bast fibers, present remarkably high lengths (up to 28000 m
fiber length).

Table 1. Morphological characteristics (length and width) of the selected fibers [19].
Fiber Source Length (m) Width (m) L/W ratio
Bast fibers
flax 28000 21 1350:1
hemp 20000 22 1000:1
kenaf 2740 20 135:1
jute 2000 20 100:1

Leaf fibers
sisal 3030 17 180:1
abaca 6000 20 300:1
curaua n.a. n.a. n.a.

Reeds
giant reed 1180 15 78:1

Woods for comparison


softwoods 3000 30 100:1
hardwoods 1250 25 50:1
n.a. not available

Among the studied fibers, flax and hemp pulps have traditionally been used as
the primary furnish for cigarette paper (burning tube), where strength, opacity
and control of air permeability are required. Banknote paper often incorporates
flax or hemp to enhance general strength characteristics. Jute pulp is used for
high porosity papers. Its fiber length plus low diameter makes it very suitable for
finishing paper purposes. Sisal and abaca pulps have an unusually high tearing
resistance and high porosity and are well suited for the production of papers
where high strength and high porosity are required.

3.2. Raw chemical composition of the fibers


The chemical composition of the main constituents of the selected non-woody
fibers is shown in Table 2. In general, they are characterized by a high
polysaccharide content and low contents of lignin, lipids and ash [20-22]. Giant
reed presents the lowest holocellulose content and the higher content of lignin,
which makes it less interesting for pulp and papermaking. The low lignin
content of the rest of the fibers, with a lignin content as low as 2.9 % in flax, is
in principle advantageous for pulping. Moreover, the acetone extractives content
is also low, and usually less than 2%, except for curaua fibers (5.3% of total

96
5. Resultados y discusin

fiber weight). However, most of the acetone extracts in curaua corresponds to


polar compounds, while only 1.3% corresponds to lipophilic compounds, which
were estimated by redissolving the acetone extracts in chloroform. Thus, in
general, the lipophilic content of the selected non-woody fibers ranges from 0.5
to 1.3%. Finally, the ash content for all the selected fibers was low in
comparison to other raw materials used for pulp and papermaking, as the cereal
straws, with an ash content generally higher than 15% [1]. Therefore, according
to their chemical composition, most of these fibers seem suitable raw materials
for pulp and papermaking.

Table 2. Composition of the main constituents of the selected fibers (% of dry matter) [20-
22].
Ash Acetone Water-solubles Klason Acid-soluble Holocellulose
extractives lignin lignin
Bast fibers
flax 1.5 0.7 1.3 2.9 1.6 92.0
hemp 2.0 0.5 1.2 4.6 1.5 90.3
kenaf 1.8 1.0 1.1 11.4 3.0 81.9
jute 2.4 0.5 0.4 13.3 2.8 81.6

Leaf fibers
sisal 1.0 0.7 2.3 5.9 3.0 85.0
abaca 0.9 0.5 1.7 7.7 1.4 85.6
curaua 1.3 5.3 5.1 4.9 1.6 92.5

Reeds
giant reed 4.2 1.6 8.5 24.7 n.d. 49.8

3.3. Carbohydrate composition of the fibers


The results of the analyses of neutral sugars of the non-woody fibers selected for
this study are reflected in Table 3. The hemicelluloses fraction of the bast fibers
presents a higher variability than those of the leaf fibers. Thus, hemicelluloses
from flax and hemp are mainly constituted by mannose followed by galactose,
while the hemicelluloses from kenaf and jute are predominantly constituted by
xylose. On the other hand, all the leaf fibers (sisal, abaca and curaua) show a
predominance of xylose. Finally, giant reed presents a strikingly high content of
xylose, that amounts up to 39.2% of the total neutral sugars.

97
5. Resultados y discusin

Table 3. Composition of neutral monosaccharides (as percentage of total neutral


carbohydrates) [20-22].
Rhamnose Arabinose Xylose Mannose Galactose Glucose

Bast fibers
flax 0.4 0.9 1.1 8.8 3.5 85.3
hemp 0.4 0.6 1.0 9.9 1.6 86.4
kenaf 0.5 2.1 10.5 4.9 0.5 81.5
jute 0.5 1.5 7.9 4.2 0.5 85.4

Leaf fibers
sisal 0.3 1.9 12.0 3.6 0.6 81.7
abaca 0.3 1.6 7.5 3.5 0.3 86.9
curaua 0.0 2.7 8.0 3.5 0.2 85.6

Reeds
giant reed 0.0 3.4 39.2 0.3 0.8 56.3

3.4. Lipid composition of the fibers


As shown in Table 2, all the studied fibers present low extractives contents.
However, due to the wide structural heterogeneity of the compounds that may
occur and their different behavior during pulping, the knowledge of the chemical
nature of these components, especially the lipophilic compounds, is important in
order to predict and control the eventual pitch problems that may occur during
pulping and bleaching and to establish appropriate methods and strategies for
their control.
The composition of the lipids present in the different fibers was studied by
GC and GC-MS and is shown in Table 4. The main lipid classes found in the
non-woody fibers are shown in Figure 1 and consists mainly of alkanes (A),
fatty alcohols (B), aldehydes (C), fatty acids (D), sterols (G), sterol esters (H),
sterol glycosides (I), steroid hydrocarbons (J), steroid ketones (K) and waxes
(L). Other compounds found are alkyl ferulates (M), glycerides (N), -hydroxy
monoesters (O) and -hydroxy acylesters of glycerol (P). The detailed
composition of the lipophilic compounds present in these fibers has been
addressed [7, 8, 22-27]. The content and composition of the different lipid
classes vary considerable among the fibers. In the case of flax bast fibers the
predominant lipophilic compounds are fatty acids and aldehydes, accounting for
34% and 23% of total extract, respectively, followed by ester waxes (18%) and
fatty alcohols (13%). Fatty acids are also the predominant compounds (27% of
total extracts) in hemp bast fibers, followed by alkanes (15%), free sterols (12%)
and steroid hydrocarbons (12%). The predominant lipophilic compounds in

98
5. Resultados y discusin

kenaf and jute are fatty acids (28 and 35% respectively), followed by ester
waxes (26 and 27% respectively). Among the selected leaf fibers, free sterols
and fatty acids predominate in both sisal (20 and 24%, respectively) and abaca
(45 and 19%, respectively), while in curaua fibers, fatty acids and ester waxes
predominate (38 and 34%, respectively), followed by free sterols (10%). Finally,
in giant reed, the predominant lipophilic compounds are fatty acids (40%),
followed by free sterols (19%) and ester waxes (15%).
Generally, these fibers are pulped by an alkaline process, usually
soda/anthraquinone pulping. Therefore, we discuss the behavior and fate of the
different fiber components during alkaline cooking. In this context, the lipids
present in these fibers can be classified, in general terms, into two principal
groups, namely fatty acids (including - and -hydroxyfatty acids) and neutral
components, including wax esters, long-chain n-fatty alcohols, alkanes, and
steroids and triterpenoids. The different lipids classes have different behavior
during cooking and bleaching [7, 8]. The wax esters, which are abundant
lipophilic compounds in some of these fibers (i.e. flax and curaua), are
hydrolyzed during alkaline cooking and the fatty acids dissolved. At sufficiently
high pH (as in alkaline pulping), the acids dissociate and form fatty acid soaps
and can thus dissolve in water to quite a high extent. By contrast, alkanes, fatty
alcohols, sterols and triterpenols, steroid hydrocarbons and ketones, and steryl
glycosides do not form soluble soaps under the alkaline pulping conditions and
therefore survive cooking. These compounds have a very low solubility in water
and are difficult to remove, and therefore can be at the origin of pitch deposition.
The low amounts of these neutral compounds in most of the fibers, and
particularly the low abundances of free and conjugated sterols, which have a
high propensity to form pitch deposits [28-30] would point to a low pitch
deposition tendency of the lipophilics from these fibers. On the other hand, fatty
acid soaps are effective solubilizing agents facilitating the removal from pulp of
these sparingly soluble neutral substances. Therefore, the ratio of saponifiables-
to-unsaponifiables has been suggested to be a better index for predicting pitch
problems than the total amount of lipids (Back and Allen, 2000). In fact, the
higher abundances of unsaponifiable compounds (neutrals) with respect to the
saponifiable ones is the main cause for pitch problems during pulping of some
woods, such as aspen or eucalypt [28-31]. Fatty alcohols, alkanes and sterols are
among the compounds responsible for pitch deposits formed during pulping of
nonwoody plants [8, 32]. In most of the fibers, as in flax, hemp, kenaf or jute
fibers, the content of free fatty acids (including - and -hydroxyfatty acids) is
high, and therefore the fatty acid soaps formed during alkaline pulping may
possess sufficient micellar-forming properties to carry the less polar compounds
into solution. However, in other fibers, such as sisal and particularly abaca, the
fatty acids amounts up to only 20% of total lipophilic compounds, and therefore
they would be more prone to produce pitch deposition.

99
5. Resultados y discusin

 OH
A B
O O

H OH
C D
O
HO
OH
E
O

OH
F OH

O
G H
HO

CH2OH
O
O
OH O
HO
I J K
OH

O
L

O O
O

O-CH2

HO-O-CH
OCH3 M N
OH HO-O-CH2
O
HO
O
O
O
HO
O-CH2

P HO-O-CH

HO-O-CH2

Figure 1. Structures of compounds representing the main classes of lipophilic extractives


found in the different fibers selected. : A, pentacosane; B, docosanol; C, octacosanal; D,
palmitic acid; E, 26-hydroxyhexacosanoic acid; F, 2-hydroxytetracosanoic acid; G, sitosterol;
H, sitosteryl linoleate; I, sitosteryl 3-D-glucopyranoside; J, stigmasta-3,5-diene; K,
stigmasta-3,5-dien-7-one; L, octacosyl hexadecanoate; M, trans-docosanylferulate; N, 1-
monodocosanoylglycerol; O, docosanyl, 16-hydroxyhexadecanoate; P, 1-mono(24-
hydroxytetracosanoyl)glycerol.

100
5. Resultados y discusin

Table 4. Composition of lipophilic extractives (mg/100g) in the different fibers [8, 22-26].
Bast fibers Leaf fibers Reeds
flax hemp kenaf jute sisal abaca curaua giant reed
n-alkanes (A) 27 43 27 5 15 - - 8
fatty alcohols (B) 220 2 13 13 8 <1 55 19
n-aldehydes (C) 371 25 1 - 1 - - 8
fatty acids (D) 552 78 33 13 11 9 81 114
-hydroxyfatty acids (E) - - - 3 6 1 142 -
-hydroxyfatty acids (F) 11 9 - 10 7 1 23 -
free sterols/triterpenols (G) 92 36 5 4 20 25 62 53
sterol/triterpenol esters (H) 6 7 1 - <1 1 9 7
sterol glycosides (I) 5 13 <1 1 2 2 27 15
steroid hydrocarbons (J) 14 30 2 2 14 3 12 13
steroid/triterpenoid ketones (K) 33 27 4 3 3 4 6 4
n-alkyl ferulates (M) - - - - 6 3 - -
ester waxes* (L, N, O, P) 284 17 30 20 8 7 222 42

* including mono- and diglycerides and -hydroxyfatty acid esters.

3.5. Lignin composition of the fibers


The lignin content of the non-woody plant fibers selected for this study
(estimated as Klason lignin) ranges from 2.9 for flax bast fiber to 24.7 for giant
reed (Table 2). The low lignin content of most of the fibers seems to be
advantageous for their use in paper pulp manufacturing, as they would require
fewer chemicals and less drastic conditions during pulping and bleaching.
However, not only the content but the lignin composition also strongly affects
delignification rates, chemical consumption and pulp yields [5, 33]. In general,
the efficiency of pulping is directly proportional to the amount of syringyl (S)
units in lignin [5]. The G units have a free C-5 position available for carbon-
carbon inter-unit bonds, which make them fairly resistant to lignin
depolymerization in pulping, while the S lignin is relatively unbranched and has
a lower condensation degree and therefore is easier to delignify. The higher
reactivity of the S lignin with respect to the G lignin in alkaline systems is
known [34, 35] and, therefore, the lignin S/G ratio directly affects the
delignification behavior. Higher S/G ratios would imply higher delignification
rates, less alkali consumption and, therefore, higher pulp yield [5, 33].
The lignin composition of the selected fibers was characterized in situ by Py-
GC/MS [20-24, 26, 36]. The relative composition of the H, G and S-lignin units

101
5. Resultados y discusin

for the non-woody fibers studied here is listed in Table 5. A predominance of S-


over G-lignin was found in the bast fibers of kenaf [24] and jute [20, 36] and in
all the leaf fibers of sisal, abaca and curaua [20, 22, 23]. By contrast, the bast
fibers from flax and hemp, as well as the giant reed showed a predominance of
G-lignin [20, 21, 26]. This is especially evident in the lignin of flax, with an
extremely low S/G ratio of 0.1. The low S/G ratio of the lignins from flax and
hemp, despite having very low lignin contents (less than 5% Klason lignin),
makes them fairly resistant to alkaline delignification. Also, the high lignin
content of giant reed (24.7 % Klason lignin) together with its low S/G ratio of
0.7 makes it especially hard to delignify. By contrast, the rest of the fibers
(kenaf, jute, sisal, abaca, curaua) present both high S/G ratios and low lignin
contents, which will make them easily delignifiable under alkaline pulping,
requiring lower energy and less drastic conditions.

Table 5. Composition of H, G and S moieties for all the raw materials studied by Py-GC/MS
[20-25, 35].
Bast fibers Leaf fibers Reeds
flax hemp kenaf jute sisal abaca curaua giant reed
%H 56,4 12.8 1.3 2.1 1.3 20.2 29.8 25.6
%G 40.8 53.0 15.4 32.2 18.7 13.5 29.1 44.2
%S 2.8 34.2 83.3 65.7 80.0 66.2 41.1 30.2
S/G ratio 0.1 0.6 5.4 2.0 4.3 4.9 1.4 0.7

For a more complete structural characterization of the lignins from these non-
woody fibers, the milled-wood lignins (MWL) were isolated and analyzed by
2D-NMR spectroscopy [36-38]. Signals from S and G lignin units were
observed in all spectra, whereas signals for p-hydroxyphenyl (H) lignin units
could only be detected in the HSQC spectra of flax and hemp lignins, in
agreement with the high amounts of these units observed by Py-GC/MS and
shown in Table 5. In general, the relative proportions of the different lignin
units (S/G ratios in Table 6) are in close agreement with the Py-GC/MS data
shown above. In addition, prominent signals corresponding to p-coumarate
structures were observed in the lignins of abaca and curaua [37]. In these
lignins, p-coumaric acid has been reported to be esterified to the lignin polymer
[23, 37, 39]. The side-chain region of the HSQC spectra gave additional
information about the different inter-unit linkages (i.e. -O-4 aryl ether, -
resinol, -5 phenylcoumaran, -1/-O- spirodienones, etc) present in the
structure of these lignins. The main substructures found in these lignins are
depicted in Figure 2. The relative abundances of the main inter-unit linkages
present in the MWL of the non-woody fibers selected for this study are shown in
Table 6. -O-4 aryl ether substructures (I) were predominant in all of the
lignins. Interestingly, the lignins from kenaf, sisal, abaca and curaua are

102
5. Resultados y discusin

especially enriched in -O-4 structures (more than 84% of all side-chains) [37].
Phenylcoumaran (-5 linkages) substructures (II) were observed in most of the
fibers, being especially abundant in flax and hemp, but were completely absent
in abaca. The presence of these low amounts of phenylcoumaran substructures
was expected due to the low levels of guaiacyl lignin units in these samples.
Resinol (- linkages) substructures (III) were also observed in important
amounts in flax, hemp, and jute, and in low amounts in kenaf and sisal, but were
completely absent in abaca and curaua lignins. Finally, spirodienone structures
(IV) were also present, although in lower amounts in most of the fibers, being
absent in flax and hemp. The high abundance of non-condensed linkages in the
lignins of kenaf, sisal, abaca and curaua makes them particularly easily to
delignify, in contrast to the rest of the lignins, with a high content of condensed
linkages, particularly in flax and hemp lignins.

Table 6. Structural characteristics (relative abundance of the main interunit linkages as


percentages of side-chains involved, percentage of -acylation and S/G ratio) observed from
the HSQC spectra of the MWL of selected fibers (curaua, hemp, kenaf, jute, sisal and abaca)
[35-37].
Bast fibers Leaf fibers
flax hemp kenaf jute sisal abaca curaua
Linkage relative abundance (% of side-chains involved)

-O-4 alkyl-aryl units (I, I', I'') 71 69 84 72 89 94 94


Phenylcoumarans (II) 16 9 2 4 2 0 2
Resinols (III) 13 22 8 16 4 0 0
Spirodienones (IV) 0 0 6 4 5 6 4

Percentage of -acylation 0 0 58 4 68 80 69

S/G ratio 0.1 0.8 5.6 2.0 3.9 8.7 4.9

Interestingly, the spectra of some of these lignins (kenaf, sisal, abaca, curaua)
revealed the presence of intense signals corresponding to acylated -carbon
(Figure 2, structures I' and I'') [37]. An estimation of the percentage of -
acylation of the lignin side-chain was calculated by integration of the signals
corresponding to the hydroxylated and acylated -carbon (Table 6) and ranged
from 4% in jute lignin to 80% in abaca lignin. The high level of acylation of the
-carbon has been correlated with the high abundances of -O-4 linkages and
the low abundances of the - resinol structures [37, 38]. The nature of the acyl
group esterifying the -carbon was studied by the so-called Derivatization
Followed by Reductive Cleavage (DFRC) degradation method [40, 41], which
selectively and efficiently cleaves -ether and -ether linkages but leaves -
esters intact. This method allowed confirming that p-coumarate groups are

103
5. Resultados y discusin

attached at the -carbon of abaca and curaua lignins, and predominantly on


syringyl units [37, 41]. In addition, acetate units were also found esterifying the
-carbon in the lignins of all the studied fibers, although at different extents. In
all cases, acetate and p-coumarate groups were found to be preferentially
attached to syringyl units [37, 41-43]. It must be noted that, although these ester
moieties will, in principle, consume additional alkaline reagents during cooking,
it has been shown above that the highly acylated lignins are extremely enriched
in easily hydrolysable non-condensed -O-4 linkages, which will be more
amenable to delignification.
HO
5
4 6
3 1 O
O 2 E
D J

HO MeO O MeO O MeO


6 6 6'
J 5 1 J 5 1 J 5' 1'

E 4 2 E 4 2 E 4' 2'
HO D 3 HO D 3 HO D 3'
O O O

1
OMe 1
OMe 1
OMe
6 2 6 2 6 2
5 3 5 3 5 3
4 4 4
MeO OMe MeO OMe MeO OMe

O O O

I I' I''

1 OMe
6 2
OMe
5 3 O
4 3 O 1''
OMe 2 4
3
6'' 2''
HO 4 3''
E 1 5 2 5''
J O O 6 4''
D J OMe 5 MeO OMe
D MeO 6
1
D
E E
O O
J E D
1 E
6 2 MeO D J
J
5 3 6
1 O HO 1
OH
4 5
MeO 6 2
OMe 4 2
3 5 3
O O 4
MeO OMe
OMe O

II III IV

Figure 2. Main substructures present in the lignins studied here: I, -O-4 linked
substructures; I, -O-4 linked substructures with acetylated -carbon; I, -O-4 linked
substructures; with p-coumaroylated -carbon; II, phenylcoumaran structures formed by -5
and -O-4 linkages; III, resinol structures formed by - , -O-, and -O-R linkages; IV,
spirodienone structures formed by -1, and -O-  linkages.

4. Conclusions
The chemical composition of different non-woody plant fibers used as raw
materials for pulp and papermaking has been summarized, with especial
emphasis in the chemistry of lipids and lignin and their fate during alkaline

104
5. Resultados y discusin

pulping. This study offers valuable information that will lead to a better
industrial utilization of these non-woody plant species of high socioeconomic
interest.

Acknowledgements
This study has been supported by the Spanish Projects AGL2005-01748 and
AGL2008-00709 and the EU BIORENEW project (NMP2-CT-2006-26456).
We thank CELESA (Tortosa, Spain) and University of Huelva for providing the
samples. G.M. thanks the Spanish Ministry of Education for a FPI fellowship.
J.R. thanks the Spanish CSIC for a I3P fellowship.

References
[1] Moore G. Nonwood Fibre Applications in Papermaking, Pira International,
Leatherhead, Surrey, UK, 1996.

[2] Paavilainen L. European prospects for using nonwood fibres. Pulp Pap Int
1998; 61-86.

[3] Saijonkari-Pahkala K. Non-wood plants as raw materials for pulp and


paper. PhD Thesis, University of Helsinki, Finland, 2001; pp. 101.

[4] Hurter RW, Riccio FA. Why CEOS dont want to hear about nonwoods-or
should they? In: TAPPI Proceeding, NA Nonwood Fiber Symposium,
Atlanta, GA, USA, 1998; 1-11.

[5] del Ro JC, Gutirrez A, Hernando M, Landn P, Romero J, Martnez AT.


Determining the influence of eucalypt lignin composition in paper pulp
yield using Py-GC/MS. J Anal Appl Pyrol 2005; 74: 110-115.

[6] Back EL, Allen LH. Pitch Control, Wood Resin and Deresination, Tappi
press, Atlanta, GA., 2000; pp. 392.

[7] Gutirrez A, del Ro JC. Lipids from flax fibers and their fate after alkaline
pulping. J Agric Food Chem 2003; 51: 4965-4971.

[8] Marques G, del Ro JC, Gutirrez 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.

[9] Technical Association of the Pulp and Paper Industry. Test methods, 1992-
1993. TAPPI, Atlanta, Ga. 1993.

105
5. Resultados y discusin

[10] Gutirrez A, del Ro JC, Gonzlez-Vila FJ, Martn F. Analysis of lipophilic


extractives from wood and pitch deposits by solid-phase extraction and gas
chromatography. J Chromatogr A 1998; 823: 449-455.

[11] Christie WW. In Christie WW, Ed. Solid-phase extraction columns in the
analysis of lipids. Advances in Lipid MethodologyOne, The Oily Press,
Dundee, Scotland 1992; pp. 118.

[12] Faix O, Meier D, Fortmann I. Thermal degradation products of wood. A


collection of electron of electron-impact (EI) mass spectra of monomeric
lignin derived products. Holz Roh- Werkst 1990; 48 (9): 351-354.

[13] Ralph J, Hatfield RD. Pyrolysis-GC/MS characterization of forage


materials. J Agric Food Chem 1991; 39 (8): 1426-1437.

[14] Bjrkman A. Studies on finely divided wood. Part I. Extraction of lignin


with neutral solvents. Sven Papperstidn 1956; 59: 477-485.

[15] Capanema EA, Balakshin MY, Chen CL, Gratzl JS, Gracz H. Structural
analysis of residual and technical lignins by 1H-13C correlation 2D NMR-
spectroscopy. Holzforschung 2001; 55: 302-308.

[16] Capanema EA, Balakshin MY, Kadla JF. Quantitative characterization of a


hardwood milled wood lignin by nuclear magnetic resonance spectroscopy.
J Agric Food Chem 2005; 53 (25): 96399649.

[17] Liiti TM, Maunu SL, Hortling B, Toikka M, Kilpelinen I. Analysis of


technical lignins by two- and three-dimensional NMR spectroscopy. J
Agric Food Chem 2003; 51 (21): 21362143.

[18] Ralph J, Marita JM, Ralph SA, Hatfield RD, Lu F, Ede RM, Peng J,
Quideau S, Helm RF, Grabber JH, Kim H, Jimenez-Monteon G, Zhang Y,
Jung HJG, Landucci LL, MacKay JJ, Sederoff RR, Chapple C, Boudet
AM. Solution-state NMR of lignin. In: Argyropoulos DS, Ed. Advances in
Lignocellulosics Characterization, Tappi Press, Atlanta, GA 1999; 55-108.

[19] Garca Hortal JA. Fibras Papeleras, Edicions UPC, Barcelona, Spain, 2007;
pp. 243.

[20] Rodrguez MI. Caracterizacin qumica de plantas hebceas utilizadas para


la fabricacin de pasta de papel de alta calidad. PhD Thesis, University of
Seville, Spain 2006.

106
5. Resultados y discusin

[21] Coelho DS Estudo sistemtico da composiao quimica das fibras de


Arundo donax e a sua evoluao durante a producto de pasta de papel
atravs do processo organosolv. MSc Thesis, University of Aveiro,
Portugal, 2006.

[22] Marques G, Gutirrez A, del Ro JC. Chemical characterization of lignin


and lipophilic fractions from leaf fibers of curaua (Ananas erectifolius). J
Agric Food Chem 2007; 55: 1327-1336.

[23] del Ro JC, Gutirrez A. Chemical composition of abaca (Musa textilis)


leaf fibers used for manufacturing of high quality paper pulps. J Agric Food
Chem 2006; 54 (13): 46004610.

[24] Gutirrez A., Rodrguez IM, del Ro JC. 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.

[25] del Ro JC, Marques G, Rodrguez IM, Gutirrez A. Chemical composition


of lipophilic extractives from jute (Corchorus capsularis) fibers used for
manufacturing of high-quality paper pulps. Ind Crops Prod 2009; 30: 241-
249.

[26] Gutirrez A, Rodrguez IM, del Ro JC. 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.

[27] Coelho DS, Marques G, Gutirrez A, Silvestre ARD, del Ro JC. Chemical
characterization of the lipophilic fraction of Giant reed (Arundo donax)
fibres used for pulp and paper manufacturing. Ind Crops Prod 2007; 26:
229-236.

[28] del Ro JC, Gutirrez A, Gonzlez-Vila FC, Martn F, Romero J.


Characterization of organic deposits produced in kraft pulping of
Eucalyptus globulus wood. J Chromatogr A 1998; 823: 457465.

[29] del Ro JC, Romero J, Gutirrez A. Analysis of pitch deposits produced in


kraft pulp mills using a totally chlorine free bleaching sequence. J
Chromatogr A 2000; 874: 235245.

[30] Gutirrez A, del Ro JC. Gas chromatography/mass spectrometry


demonstration of steryl glycosides in eucalypt wood, kraft pulp and process
liquids. Rapid Commun Mass Spectrom 2001; 15: 25152520.

[31] Chen T, Wang Z, Zhou Y, Breui, C, Aschim OK, Yee E, Nadeau L. Using
solid-phase extraction to assess why aspen causes more pitch problems
than softwoods in kraft pulping. Tappi J 1995; 78: 143-149.

107
5. Resultados y discusin

[32] Gutirrez A, del Ro JC. Chemical characterization of pitch deposits


produced in the manufacturing of high-quality paper pulps from hemp
fibers. Biores Technol 2005; 96: 1445-1450.

[33] Gonzlez-Vila FJ, Almendros G, del Ro JC, Martn F, Gutirrez A,


Romero J. Ease of delignification assessment of wood from different
Eucalyptus species by pyrolysis (TMAH)-GC/MS and CP/MAS 13C NMR
spectrometry. J Anal Appl Pyrol 1999; 49: 295-305.

[34] Chang HM, Sarkanen KV. Species variation in lignin. Effect of species on
the rate of Kraft delignification. Tappi press 1973; 56: 132134.

[35] Tsutsumi Y, Kondo R, Sakai K, Imamura H. The difference of reactivity


between syringyl lignin and guaiacyl lignin in alkaline systems.
Holzforschung 1995; 49 (5): 423-428.

[36] del Ro JC, Rencoret J, Marques G, Li J, Gellerstedt G, Jimnez-Barbero J,


Martnez AT, Gutirrez A. Structural characterization of the lignin from
jute (Corchorus capsularis) fibers. J Agric Food Chem 2009; 57: 10271-
10281.

[37] del Ro JC, Rencoret J, Marques G, Gutirrez A., Ibarra D, Santos JI,
Jimnez-Barbero J, Zhang L, Martnez AT. Highly acylated (acetylated
and/or p-coumaroylated) native lignins from diverse herbaceous plants. J
Agric Food Chem 2008; 56: 9525-9534.

[38] Martnez AT, Rencoret J, Marques G, Gutirrez A, Ibarra D, Jimnez-


Barbero J, del Ro JC. Monolignol acylation and lignin structure in some
nonwoody plants: A 2D NMR study. Phytochemistry 2008; 69: 2831-2843.

[39] Sun RC, Fang JM, Goodwin A, Lawther JM, Bolton J. Fractionation and
characterization of ball-milled and enzyme lignins from abaca fibre. J Sci
Food Agric 1999; 79: 1091-1098.

[40] 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 (7): 2590-2592.

[41] del Ro JC, Marques G, Rencoret J, Martnez AT, Gutirrez A. Occurrence


of naturally acetylated lignin units. J Agric Food Chem 2007; 55:5461-
5468.

[42] 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.

108
5. Resultados y discusin

[43] Ralph J. An unusual lignin from kenaf. J Nat Prod 1996; 59 (4): 341-342.

109
5. Resultados y discusin

Publicacin II:

del Ro J.C., Marques G., Rencoret J. Martnez A.T. and Gutirrez A. (2007)
Occurence of naturally acetylated lignin units. Journal of Agricultural and Food
Chemistry, 55, 5461-5468.

110
5. Resultados y discusin

Occurrence of naturally acetylated lignin units

Jos C. del Ro, Gisela Marques, Jorge Rencoret, ngel T. Martnez, Ana Gutirrez


Instituto de Recursos Naturales y Agrobiologa de Sevilla, CSIC, P.O. Box 1052, 41080-
Seville, Spain

Centro de Investigaciones Biolgicas, CSIC, Ramiro de Maeztu 9, E-28040 Madrid, Spain

Abstract
In this work, we have studied the occurrence of native acetylated lignin in a
large set of vascular plants, including both angiosperms and gymnosperms, by a
modification of the so-called Derivatization Followed by Reductive Cleavage
(DFRC) method. Acetylated lignin units were found in all angiosperms selected
for this study, including mono- and eudicotyledons, but were absent in the
gymnosperms analyzed. In some plants (e.g. abaca, sisal, kenaf or hornbeam),
lignin acetylation occurred at a very high extent, exceeding 45% of the
uncondensed syringyl lignin units. Acetylation was observed exclusively at the
J-carbon of the lignin-side chain and predominantly on syringyl units, although
a predominance of acetylated guaiacyl over syringyl units was observed in some
plants. In all cases, acetylation appears to occur at the monomer stage and
sinapyl and coniferyl acetates seem to behave as real lignin monomers
participating in lignification.
Keywords: lignin, angiosperms, gymnosperms, eudicotyledons,
monocotyledons, coniferyl acetate, sinapyl acetate, abaca, sisal, kenaf,
hornbeam, Derivatization Followed by Reductive Cleavage (DFRC).

1. Introduction
Lignin is a principal structural component of cell walls in higher terrestrial
plants. In addition to structural support and pathogen defense, lignin functions in
water transport as a hydrophobic constituent of vascular phloem and xylem cells.
Lignins are complex polymers formed by the dehydrogenative polymerization of
three main monolignols, p-coumaryl, coniferyl and sinapyl alcohols.
Gymnosperm lignins are mainly formed from coniferyl alcohol, together with
small proportions of p-coumaryl alcohol. Angiosperm lignins are mainly formed
from coniferyl and sinapyl alcohols with small amounts of p-coumaryl alcohol.
A considerable variation in the contribution of these three alcohol precursors is
observed in lignins from annual plants (1-4). After their synthesis, the lignin
monomers are transported to the cell wall where they are polymerized in a
combinatorial fashion by free-radical coupling mechanisms in a reaction
mediated by peroxidases, generating a variety of structures within the lignin
polymer (5-7).

111
5. Resultados y discusin

Some lignins are known to be naturally acetylated at the J-carbon of the side
chain. Acetates have been reported to occur in the lignin of a limited number of
hardwoods and non-woody plants (1, 8, 9). In particular, kenaf bast lignin has
been found to be extensively J-acetylated and predominantly on syringyl units,
although the role of such a highly acetylated lignin is not yet known. Recent
studies have provided strong evidence that sinapyl acetate is implicated as a
monomer in lignification in kenaf bast fibers and that the naturally acetylated
polymeric lignin derives not from acetylation of the lignin polymer but from
polymerization of pre-acetylated monolignols (9, 10). Other acids are also
known to occur naturally acylating lignin; thus, grass lignins are partially p-
coumaroylated and some hardwood lignins such as poplar, aspen or willow are
p-hydroxybenzoylated (7, 11-18).
Naturally acetylated lignins may also occur in other plants but their
occurrence has probably being missed due to the limitations of the analytical
procedures used for their isolation and/or structural characterization. Natural
acetates present on lignin might have been hydrolyzed and removed when using
traditional isolation methods (such as alkaline extraction often applied to non-
wood lignins) and degradative procedures for chemical characterization (such as
nitrobenzene oxidation, CuO oxidation or thioacidolysis). Indeed, for
spectroscopic analysis, e.g. using Nuclear Magnetic Resonance (NMR), lignin is
frequently acetylated for improved solubility and spectroscopic properties and,
therefore, native acetates cannot be seen. Recently, we reported the occurrence
of acetylated lignins in some herbaceous plants, including kenaf, jute, sisal and
abaca, characterized by a high syringyl (S) to guaiacyl (G) ratio, by the use of
Pyrolysis coupled to Gas Chromatography-Mass Spectrometry (Py-GC/MS)
(19-21), although the method used hindered the determination of the extent of
acetylation. In this work we have studied the occurrence and abundance of
native acetylated lignin units in the milled wood lignins (MWL) isolated from a
wide set of vascular plants, including gymnosperms and angiosperms (mono-
and eudicotyledons). For this purpose, we have used a modification of the so-
called Derivatization Followed by Reductive Cleavage (DFRC) degradation
method (22-24). DFRC is a simple and powerful method which selectively and
efficiently cleaves D-ether and E-ether linkages and allows quantitative analysis
of structural units in etherified lignin, and also provides some information on
carbon-carbon linked lignin by analysis of the dimeric structures released.
DFRC includes two key steps, (i) solubilization of lignin by bromination and
acetylation with acetyl bromide and (ii) reductive cleavage of the aryl ether
bonds in lignin with zinc dust. Identification of the resulting monomeric and
dimeric degradation products by GC/MS gives valuable information on the
lignin structure. However, and most importantly, DFRC leaves J-esters intact
allowing the analysis of native J-acylated lignin. Thus, the method has allowed
to confirm that p-coumarate groups are attached at the J-carbon of grass lignins,
predominantly on syringyl units (17). However, the DFRC method uses

112
5. Resultados y discusin

acetylating reagents that interfere in the analysis of native acetates in lignin, but
with appropriate modification by substituting acetylation by propionylation (25),
it is possible to obtain significant information and clues about the occurrence
and extent of native lignin acetylation. In this paper, we use this method to
investigate the presence of naturally acetylated lignin units in a set of vascular
plants, including angiosperms and gymnosperms.

2. Material and methods


2.1. Samples
The plant samples selected for this study are listed in Table 1. They consist of
both woody and nonwoody angiosperms (mono- and eudicotyledons) and
gymnosperms. Among the woody angiosperms, wood of beech (Fagus
sylvatica), European hornbeam (Carpinus betulus), aspen (Populus tremula),
and eucalyptus (Eucalyptus globulus) were selected. The nonwoody angiosperm
samples consisted of bast fibers obtained from the stalk phloem layer of bamboo
(Bambusa sp.), hemp (Cannabis sativa), kenaf (Hibiscus cannabinus) and jute
(Corchorus capsularis); leaf fibers of sisal (Agave sisalana) and abaca (Musa
textilis); and coir, a coarse fiber obtained from the outer shell of coconut from
the palm tree (Cocos nucifera). Among gymnosperm woods, Scots pine (Pinus
sylvestris) and Norway spruce (Picea abies) were selected for this study. Wood
and nonwoody plants were finely ground to sawdust using a knife mill
(Analysenmhle A10, Janke and Kunkel GmbH, Staufen, Germany) before
analysis. MWL was extracted from finely ball-milled (150 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 (26). The final
yields ranged from 5-15% of the original lignin content. Extension of milling
time, which would increase yield, was avoided in order to prevent chemical
modifications on the lignin structure.

2.2. DFRC
A modification of the standard DFRC method by using propionyl instead of
acetyl reagents (DFRC) was used (25). Lignins (10 mg) were stirred for two
hours at 50 C with propionyl bromide in propionic acid (8:92, v/v). The
solvents and excess of bromide were removed by rotary evaporation. The
products were then dissolved in dioxane/propionic acid/water (5:4:1, v/v/v), and
50 mg powered 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 aqueous phase was adjusted to pH < 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 NaSO4 and the filtrate was
evaporated in a rotary evaporator. The residue was subsequently propionylated

113
5. Resultados y discusin

for 1 h in 1.1 mL of dichloromethane containing 0.2 mL of propionic anhydride


and 0.2 mL pyridine. The propionylated lignin degradation compounds were
collected after rotary evaporation of the solvents, and subsequently analyzed by
GC/MS.

2.3. GC/MS analysis


The GC/MS analyses were performed with a Star 3400 GC (Varian) equipped
with a Saturn 2000 ion trap detector (Varian) using a 12m x 0.25 mm i.d., 0.1
Pm, DB-5HT length capillary column (J&W Scientific, Folsom, CA, USA). The
oven was heated from 50 (held 0.2 min) to 100 C at 30 C/min, then raised to
300 C at 5C/min, and held for 5 min at the final temperature. The injector and
transfer line were kept at 300 C. Helium was used as the carrier gas at a rate of
2 mL/min.
Quantification of the released individual monomers was performed using
tetracosane as external standard and by assuming similar response factors as
those of the acetylated monomers reported in Lu and Ralph (22), although
without authentication on our instrument. Molar yields were calculated on the
basis of molecular weights of the respective propionylated (or acetylated)
compounds.

3. Results and discussion


The MWL isolated from the different vascular plants selected for this study,
including angiosperms (mono- and eudicotyledons) and gymnosperms, were
analyzed by the modified DFRC method developed by Lu and Ralph (25) in
order to investigate the occurrence of naturally acetylated lignin moieties.
However, we have to emphasize here that the monomeric degradation products
released by DFRC originate from etherified lignin units since this method only
cleaves D- and E-aryl ether bonds. The abundance of guaiacyl lignin units,
which are predominantly in condensed form (i.e. forming 5-5 linkages), may
also be underestimated.
The chromatograms of the DFRC degradation products of selected MWL
samples are shown in Figure 1. All the analyzed lignins released the cis and
trans isomers of guaiacyl (c-G and t-G) and syringyl (c-S and t-S) lignin
monomers (as their propionylated derivatives) in different proportions, arising
from normal (J-OH) units in lignin. In addition, the presence of originally J-
acetylated guaiacyl (c-Gac and t-Gac) and syringyl (c-Sac and t-Sac) lignin units
could also be clearly observed in the chromatograms of most of the analyzed
lignins. The structures and mass fragments of these compounds arising from J-
OH and from J-acetylated lignin units are depicted in Figure 2. Acetylation
occurred exclusively at the J-carbon of the lignin side-chain, as already reported
for kenaf lignin (1, 25). In all samples, coniferyl and sinapyl acetates presented a

114
Table 1. Abundance (Molar Yields) of the DFRC Degradation Monomers of the MWL Isolated from the Different Plants Selected for
This Study, S/G Ratios and Relative Abundances of Acetylated Lignin Moieties.

Monomers (Pmol/g lignin)

Order Family Species Name G Gac S Sac S/G %Saca %Gacb


Angiosperms
Monocotyledons
Asparagales Agavaceae Agave sisalana Sisal 122 124 108 378 2.0 77.7 50.4
Arecales Arecaceae Cocos nucifera Coir 819 5 174 14 0.2 7.4 0.6
Poales Poaceae Bambusa sp. Bamboo 256 13 280 4 1.1c 1.2c 4.8
Zingiberales Musaceae Musa textilis Abaca 50 3 21 131 3.0c 80.3c 5.6

Eudicotyledons
Fagales Fagaceae Fagus sylvatica Beech 126 2 165 20 1.4 10.8 1.6
Fagales Betulaceae Carpinus betulus Hornbeam 146 4 230 185 2.8 44.6 2.7
Rosales Cannabaceae Cannabis sativa Hemp 286 2 177 2 0.6 1.1 0.7
Malvales Malvaceae Hibiscus cannabinus Kenaf 390 38 543 780 3.1 59.0 8.9
Malvales Malvaceae Corchorus capsularis Jute 299 1 336 23 1.2 6.4 0.3
Malpighiales Salicaceae Populus tremula Aspen 651 5 662 8 1.0 1.2 0.8
Myrtales Myrtaceae Eucalyptus globulus Eucalypt 154 8 275 3 2.3 1.1 4.9

Gymnosperms
Coniferales Pinaceae Picea abies Spruce 520 0 0 0 0.0 - 0.0

Coniferales Pinaceae Pinus sylvestris Pine 402 0 0 0 0.0 - 0.0

a
%Sac is the percentage of acetylated S units respect to the total S units. b %Gac is the percentage of acetylated G units respect to the total G units. cSome amounts of
J-p-coumaroylated S units were found (27 an 11 Pmol/g lignin for bamboo and abaca, respectively) and were included in the estimation of total S units for calculation
of S/G and %Sac.
5. Resultados y discusin

115
5. Resultados y discusin

predominance of the trans- over the cis- form, as also occurred with the
corresponding non-acetylated alcohols.
The results from the DFRC analysis of the MWL selected for this study,
namely the molar yields of the released monomers, the S/G ratios and the
percentages of naturally acetylated guaiacyl (%Gac) and syringyl (%Sac) lignin
moieties, are presented in Table 1. The yields of the released monomers were in
the same range as previously observed in the DFRC degradation of other
isolated lignins (17, 24). As shown in Table 1, naturally acetylated lignin units
were found to occur in all angiosperms analyzed in the present study, including
both mono- and eudicotyledons. However, no traces of acetylated lignin units
could be found in the MWL of the two gymnosperms (pine and spruce) studied
here. The data also indicated that in most lignin samples acetylation occurred
predominantly on syringyl units, whereas only traces of acetylated guaiacyl
units were detected, although in bamboo and eucalyptus lignins, with a low
extent of acetylation, this occurred preferentially on guaiacyl units. We can
exclude acetylation as an artifact produced during the lignin isolation protocol
since MWL from pine and spruce (where no traces of acetylated units could be
detected) were also isolated using the same procedure as the rest of the samples.
Indeed, acetates were found predominantly on S lignin units and exclusively at
the J-carbon, which suggests that they are naturally present. And finally, direct
DFRC of some whole fibers (such as sisal and kenaf), without previous lignin
isolation, gave also similar results.
The occurrence of naturally acetylated lignin units seems to be widespread
among angiosperms and restricted only to this group of vascular plants, being
particularly abundant in syringyl-rich lignins. Especially important is the high
extent of lignin acetylation observed in the MWL from the herbaceous plants
abaca, sisal and kenaf, and in the hardwoods hornbeam and, in a minor extent,
beech, all of them characterized by high S/G ratios. However, we also noted the
acetylation of S units in coir lignin, which is characterized by a very low S/G
ratio (0.2). The high extent of acetylation of kenaf lignin has been previously
reported by NMR and DFRC (1, 25). The occurrence of naturally acetylated
lignin units was also reported in kenaf, jute, sisal and abaca by Py-GC/MS of the
whole fibers, although the extent of acetylation could not be determined due to
the limitations of the technique (19-21). Interestingly, the high extent of
acetylation of sisal MWL included both S units (78%) and G units (50%),
whereas in the case of abaca, kenaf or hornbeam lignins, acetylation occurred
almost predominantly on S units (80, 59 and 45%, respectively) and only a
minor degree of acetylation was observed on G units (6, 9 and 3%, respectively).

116
5. Resultados y discusin

t-Sac
100

Sisal

relative intensity

t-G
t-G ac t-S

c-G
c-Sac c-S
c-G ac

0
t-Sac
100
t-S
Kenaf
t-G
relative intensity

c-G c-Sac c-S

t-G ac
c-G ac
0
t-S
100
t-Sac
Hornbeam
t-G
relative intensity

c-Gt-G ac c-S
c-Sac

0
t-Sac
100

Abaca
relative intensity

t-G

t-S
c-Sac
c-G
t-G ac c-S

0
10 15 20
Retention time (min)

Figure 1. Chromatograms of the DFRC degradation products of selected MWL from sisal,
kenaf, hornbeam and abaca. c-G, t-G, c-S and t-S are the cis- and trans-guaiacyl and syringyl
monomers, respectively. c-Gac, t-Gac, c-Sac and t-Sac are the originally acetylated cis- and
trans-guaiacyl and syringyl monomers, respectively.

On the other hand, it has been recently and elegantly proved that acetylated
lignin in kenaf derives from polymerization of pre-acetylated monolignols and
not from acetylation of the lignin polymer (9, 10). Acetylation of lignin at
monomer stage only partially affects the course of the coupling reactions, and

117
5. Resultados y discusin

these acetylated monolignols can still undergo the EO4-coupling reactions


that incorporate them into the lignin polymer. Obviously, the resulting
acetylated lignin polymer is produced and has the mechanical properties
required by the plant. However, the presence of J-acetylated monolignols alters
to some extent the structure of the lignins because the J-OH group of a
monolignol participates in some post-coupling reactions, such as EE coupling,
internally trapping the quinone methide. With the -OH group acetylated, such
internal reactions are no longer possible and the quinone methide must be
rearomatized by trapping an external nucleophile, usually water, and forming as
a result new EE products in the lignin different to the expected resinol
structures formed from non-acetylated monolignols (9). Figure 3 shows the
different tetrahydrofuran structures arising from the EE homo- and cross-
coupling of the two sinapyl (acetylated and non-acetylated) monolignols. It is
clear that structures I and II can only be formed if sinapyl alcohol is pre-
acetylated and then undergoes EE coupling. These EE homo- (I) and cross-
coupled (II) structures arising from pre-acetylated sinapyl alcohol were found in
kenaf lignin by DFRC and 2D-NMR (9, 10), which unequivocally demonstrates
that sinapyl alcohol is pre-acetylated prior to lignification and that sinapyl
acetate behaves as a real monolignol in kenaf lignin.

OProp OProp

OCH 3 CH 3O OCH3
OProp OProp

G S
M+= 292; [M+-56]= 236 M+= 322; [M+-56]= 266

O O
=
=

O-C-CH3 O-C-CH 3

OCH 3 CH3O OCH 3


OProp OProp

Gac Sac
M+= 278; [M+-56]= 222 M+= 308; [M+-56]= 252

Figure 2. Structures and mass fragments of the J-OH (G and S) and naturally J-acetylated
(Gac and Sac) lignin monomers released after DFRC of MWL.

118
5. Resultados y discusin

O O

=
O-C-CH3 O-C-CH 3 O O
J J

=
J J
CH3-C-O O-C-CH 3
E E
D D E E

EE D D
CH3O OCH3
O
+ homo-coupling
HO OH
CH 3O OCH3 CH3O OCH3
OH OH OCH 3 I OCH3

OH
O CH3O OCH3
=

OH O-C-CH 3
J J

E E J
D D E
D
EE HO
O
J
+ cross-coupling O E D

=
CH 3-C-O
CH3O OCH3 CH3O OCH3
OCH3
OH OH

CH 3O OH
II

OCH3

OH
OH OH
J J O
J D OCH3
E E
D D E E

EE D J
CH3O O
+ homo-coupling
HO III
CH3O OCH3 CH3O OCH3
OH OH OCH3

Figure 3. Structures of the tetrahydrofuran dimers arising from the EE coupling of sinapyl
alcohol and sinapyl acetate. I: EEcoupling product of two sinapyl acetates; II: EEcoupling
product of a sinapyl alcohol and a sinapyl acetate; III: EEcoupling product (syringaresinol)
of two sinapyl alcohols.

We also investigated the presence of the tetrahydrofuran structures arising


from EE coupling of sinapyl acetates in the MWL selected for this study, by
using the DFRC method. The DFRC degradation products of the
tetrahydrofuran structures depicted in Figure 3 are presented in Figure 4, and
their mass spectra have already been published (9). Compounds I, IIa and IIb,
arising from originally acetylated sinapyl alcohol, together with compound III
arising from the resinol structure, were found in most of the samples analyzed.
Figure 5 shows the chromatograms (sum of the single ion chromatograms of the
respective base peaks) of the DFRC degradation products of the tetrahydrofuran

119
5. Resultados y discusin

structures arising from EEcoupling of sinapyl alcohol (and its acetylated


counterpart) in selected samples. It is interesting to note the relatively high
amounts of compounds I, IIa and IIb, arising from native acetylated lignin
units in kenaf and especially in sisal lignin and the complete absence of EE
coupling structures in abaca lignin. However, we also observed that the
respective tetrahydrofuran structures arising from the EEcoupling of the
acetylated guaiacyl counterparts could not be found, even in the case of sisal
lignin (with 50% acetylated ether-linked G units). The presence of compound
I(arising from structure I in lignin) and compounds IIa and IIb (arising from
structure II) in the DFRC degradation products of sisal, kenaf, jute, hornbeam,
and other lignins clearly indicates that in these samples sinapyl alcohol is pre-
acetylated and behave as a real monolignols participating in post cross-coupling
reactions. Therefore, it is possible that in all angiosperms, in which we have
shown the occurrence of acetylated lignin units, sinapyl and possibly coniferyl
acetates participate in lignification as true lignin precursors. This implies that the
naturally acetylated polymeric lignins in all these plants derive not from
acetylation of the lignin polymer but from polymerization of pre-acetylated
monolignols, as already suggested (9). Thus, the traditional concepts of both
lignin biosynthesis and structure must be reconsidered. This indicates, in
agreement with other authors (7, 27) that the lignification process is very
flexible, and that the definition of lignin must not be restricted to a polymer of
the three traditional hydroxycinnamyl alcohols.
The relative abundance of the compounds released in Figure 5 gives some
additional information. In sisal, the relative molar abundance of the acetylated
versus the non-acetylated sinapyl alcohols forming EE linkages is 44:56, with a
predominance of the non-acetylated sinapyl alcohol, whereas their relative molar
abundances in ether-linked structures is 78:22, with a strong predominance of
sinapyl acetate. This indicates that sinapyl acetate has a lower affinity to form
EE linkages than the normal sinapyl alcohol and therefore those lignins having
a high extent of acetylation would produce lower amounts of EE linkages. This
is in agreement with the high proportions of E-O-4 substructures present in sisal
lignin, as indicated by 2D-NMR (data not shown). Interestingly, abaca MWL,
with a very high extent of acetylation (80% of S units), lacks EE linkages,
including those arising from normal non-acetylated lignin units (Figure 5), and
produces almost exclusively EO-4 substructures. Similar results were observed
after normal DFRC and 2D-NMR (data not shown). Therefore, it seems that the
high extent of J-acetylation would favor the formation of a predominantly
E2lignin structure. The difficulty in finding EE substructures arising from
naturally acetylated guaiacyl lignin in sisal, despite the high abundance of this
type of units, would be perhaps due to the low amounts of EE linkages
produced.

120
5. Resultados y discusin

CH 3O O

=
O-C-CH 3
O-C-CH 3

=
PrO
O
DFRC
I CH 3O

CH 3O OCH3

OPr

I; M+=616; [M-56]+= 560

CH 3O CH 3O O

=
OPr O-C-CH3
O-C-CH 3 OPr

=
PrO PrO
O
DFRC
II CH 3O + CH3O

CH 3O OCH3 CH 3O OCH 3

OPr OPr

IIa; M+= 630; [M-56]+= 574 IIb; M+= 630: [M-56]+= 574

CH 3O
OPr
OPr
DFRC PrO
III
CH3O

CH 3O OCH 3

OPr

III; M+= 644; [M-56]+= 588

Figure 4. Aryltetralin products resulting from the DFRC degradation of the di- (I), mono-
(IIa and IIb) and none-acetylated (III) EE coupling structures. The molecular mass and
base peaks are indicated under the structures.

Although it is now evident that native acetylated lignin units are widespread,
and probably ubiquitous, in angiosperms, the role of such lignin acetylation in
the plant is not yet known. Some studies indicated that J-acylation with p-
coumarates may function as radical transfer carriers to help sinapyl alcohol
incorporate into lignin when the wall peroxidases have a low reactivity with
sinapyl alcohol directly (28). However, this is not the case for acetylated lignin
monomers and the function of such acetylated lignin remains unknown.

121
5. Resultados y discusin

IIa,IIb
100

Sisal
III

relative intensity
I

0
III
100
Kenaf
relative intensity

IIa,IIb

0
III
100
Hornbeam
relative intensity

IIa,IIb

0
100
Abaca
relative intensity

0
35 36 37 38 39 40
Retention time (min)

Figure 5. Detail of reconstructed (sum of the ions at m/z 560, 574 and 588) chromatograms of
the DFRC degradation products of selected MWL from sisal, kenaf, hornbeam and abaca,
showing the presence of aryltetralin EE products containing two (I), one (II and IIb) and
none (III) native acetates.

The high extent of lignin acetylation of sisal and other Agaves (preliminary
results on another species, Agave americana, by DFRC of the whole fiber,
without previous lignin isolation, also indicate more than 75% of acetylated S
units) can provide some answers about its role in this plant. Since the resultant

122
5. Resultados y discusin

acetylated lignin polymer is more hydrophobic than normal lignin, the role of
lignin acetylation could be associated with drought tolerance, as already
advanced by Ralph (29). Sisal, as other Agaves, is a drought tolerant desert
perennial plant successfully copying with high temperatures and desiccation.
Agaves are characterized by tissue succulence and Crassulacean Acid
Metabolism to minimize water loss. The presence of a highly acetylated lignin
in Agaves will increase hydrophobicity of the vascular tissues thus helping to
reduce water loss in the plant. However, abaca, despite being highly acetylated,
has not succulent leaves and is not drought tolerant, and therefore the highly
acetylated lignin might have a different role for this plant. In this case, the role
of the high extent of lignin acetylation could be more related with the lower
affinity of acetylated monolignols to form EE linkages producing a more non-
condensed lignin enriched in E-O-4 linkages. Whatever the reason for the
occurrence of acetylated lignin in plants, it seems that the mechanism for lignin
acetylation confers the plant a high flexibility to produce different types of
lignins with different degree of acetylation to adapt to different environmental
conditions.
In conclusion, the presence of naturally acetylated lignin units, which in some
plants make up to 80% of the uncondensed S lignin, has been largely
underestimated. This has mainly been due to the analytical methodologies used
for their isolation and structural characterization, which are not appropriate for
the analysis of native acetylated lignin. Therefore, all subsequent lignin
structural studies should take into account the occurrence of these moieties and
accordingly adapt the methodological protocols used.

Acknowledgements
This study has been supported by the Spanish MEC (project AGL2005-01748)
and EU contract NMP2-CT-2006-26456. JR thanks the Spanish CSIC for an I3P
fellowship; GM thanks the Spanish Ministry of Education for a FPI fellowship.
Literature cited
[1] Ralph, J. An unusual lignin from kenaf. J. Nat. Prod. 1996, 59, 341-342.
[2] Sun, R.C.; Fang, J.M.; Goodwin, A.; Lawther, J.M., Bolton, A.J.
Fractionation and characterization of ball-milled and enzyme lignin from
abaca fibre. J. Sci. Food Agric. 1999, 79, 1091-1098.
[3] Sun, R.C.; Sun, X.F.; Wang, S.Q.; Zhu, W.; Wang, X.Y. Ester and ether
linkages between hydroxycinnamic acids and lignin from wheat, rice, rye,
and barley straws, maize stems, and fast-growing poplar wood. Ind. Crops
& Prod. 2002, 15, 179-188.

123
5. Resultados y discusin

[4] del Ro, J.C.; Gutirrez, A.; Rodrguez, I.M., Ibarra, D., Martnez, A.T.
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.
[5] Ros Barcel, A. Lignification in plant cell walls. Int. Rev. Cytol. 1997, 176,
87-132.
[6] Boerjan, W.; Ralph, J.; Baucher, M. Lignin biosynthesis. Annu. Rev. Plant
Biol. 2003, 54, 519546
[7] 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. et al.
Lignins: natural polymers from oxidative coupling of 4-
hydroxyphenylpropanoids. Phytochem. Rev. 2004, 3, 29-60.
[8] Sarkanen, K.V.; Chang, H. -M.; Allan, G.C. Species variation in lignins.
III. Hardwood lignins. Tappi 1967, 50, 587-590.
[9] Lu, F.; Ralph, J. Preliminary evidence for sinapyl acetate as a lignin
monomer in kenaf. Chem. Commun. 2002, 9091.
[10] Lu, F.; Ralph, J. Novel EEstructures in lignins incorporating acylated
monolignols. In Proceedings of 13th International Symposium on Wood,
Fiber, and Pulping Chemistry, Auckland, New Zealand, May 16-19, 2005;
Volume 3, pp. 233-237.
[11] Smith, D. C. C. p-Hydroxybenzoate groups in the lignin of aspen (Populus
tremula). J. Chem. Soc. 1955, 23472351.
[12] Nakano, J.; Ishizu, A.; Migata, N. Studies on lignin. XXXII. Ester groups
of lignin. Tappi 1961, 44, 3032.
[13] Monties, B.; Lapierre, C. Donns rcentes sur lhtrognite de la lignine
(Recent data on the heterogeneity of lignin). Physiol. Veg. 1981, 19, 327-
348.
[14] Landucci, L. L.; Deka, G. C.; Roy, D. N. A. 13C NMR study of milled
wood lignins from hybrid Salix clones. Holzforschung 1992, 46, 505-511.
[15] 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.
[16] Sun, R. C.; Fang, J. M.; Tomkinson, J. Fractional isolation and structural
characterization of lignins from oil palm trunk and empty fruit bunch
fibres. J. Wood Chem. Technol. 1999, 19, 335356.
[17] Lu, F. and Ralph, J. Detection and determination of p-coumaraloylated
units in lignin. J. Agric. Food Chem. 1999, 47, 1985-1992.

124
5. Resultados y discusin

[18] Meyermans, H.; Morreel, K.; Lapierre, C.; Pollet, B.; De Bruyn, A.;
Busson, R.; Herdewijn, P:; Devreese, B.; Van Beeumen, J.; Marita, J. M.;
et al. Modification 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. J. Biol.
Chem. 2000, 275, 3689936909.
[19] del Ro, J. C.; Gutirrez, A.; Martnez A. T. Identifying acetylated lignin
units in non-wood fibers using pyrolysis-gas chromatography/mass
spectrometry. Rapid Commun. Mass Spectrom. 2004, 18, 1181-1185.
[20] Gutirrez, A.; Rodrguez, I. M.; del Ro 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.
[21] del Ro, J. C.; Gutirrez, A. Chemical composition of abaca (Musa textilis)
leaf fibers used for manufacturing of high quality paper pulps. J. Agric.
Food Chem. 2006, 54, 4600-4610.
[22] 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.
[23] Lu, F.; Ralph, J. The DFRC method for lignin analysis. Part 1. A new
method for Earyl ether cleavage: lignin model studies. J. Agric. Food
Chem. 1997, 45, 4655-4660.
[24] Lu, F.; Ralph, J. The DFRC method for lignin analysis. 2. Monomers from
isolated lignin. J. Agric. Food Chem. 1998, 46, 547-552.
[25] 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.
[26] Bjrkman, A. Studies on finely divided wood. Part I. Extraction of lignin
with neutral solvents. Sven. Papperstidn. 1956, 59, 477-485.
[27] Sederoff, R. D.; MacKay, J.; Ralph, J.; Hatfield, R. Unexpected variation in
lignin. Curr. Opin. Plant Biol. 1999; 2: 145.
[28] Takahama, U.; Oniki, T.; Shimokawa, H. A possible mechanism for the
oxidation of sinapyl alcohol by peroxidase-dependent reactions in the
apoplast: enhancement of the oxidation by hydroxycinnamic acids and
components of the apoplast. Plant Cell Physiol. 1996, 37, 499-504.
[29] Ralph, J. Elucidation of new pathways in normal and perturbed
lignification. Appita 2005, 3-13.

125
5. Resultados y discusin

Publicacin III:

del Ro J.C., Rencoret J., Marques G., Gutirrez A., Ibarra D., Santos J.I.,
Jimnez-Barbero J., Zhang L. and Martnez 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.

126
5. Resultados y discusin

Highly acylated (acetylated and/or p-coumaroylated) native lignins from


diverse herbaceous plants
Jos C. del Ro, Jorge Rencoret, Gisela Marques, Ana Gutirrez, David Ibarra, J. Ignacio
Santos, Jess Jimnez-Barbero, Liming Zhang, ngel T. Martnez


Instituto de Recursos Naturales y Agrobiologa de Sevilla, CSIC, P.O. Box 1052, 41080-
Seville, Spain

Centro de Investigaciones Biolgicas, CSIC, Ramiro de Maeztu 9, E-28040 Madrid, Spain

Royal Institute of Technology (KTH), Fiber and Polymer Technology, SE-100 44
Stockholm, Sweden

Abstract
The structure of lignins isolated from the herbaceous plants sisal (Agave
sisalana), kenaf (Hibiscus cannabinus), abaca (Musa textilis) and curaua
(Ananas erectifolius) has been studied upon spectroscopic (2D-NMR) and
chemical degradative (Derivatization Followed by Reductive Cleavage)
methods. The analyses demonstrate that the structure of the lignins from these
plants is highly remarkable, being extensively acylated at the -carbon of the
lignin side-chain (up to 80% acylation) with acetate and/or p-coumarate groups,
and preferentially over syringyl units. While the lignins from sisal and kenaf are
-acylated exclusively with acetate groups, the lignins from abaca and curaua
are esterified with acetate and p-coumarate groups. The structures of all these
highly-acylated lignins are characterized by a very high syringyl/guaiacyl ratio,
a large predominance of -O-4 linkages (up to 94% of all linkages) and a
strikingly low proportion of traditional -linkages, which indeed are
completely absent in the lignins from abaca and curaua. The occurrence of -
homo-coupling and cross-coupling products of sinapyl acetate in the lignins
from sisal and kenaf indicates that sinapyl alcohol is acetylated at monomer
stage and that, therefore, sinapyl acetate should be considered as a real
monolignol involved in the lignification reactions.
Keywords: lignin, herbaceous plants, sinapyl acetate, sinapyl p-coumarate, 2D-
NMR, HSQC, DFRC, sisal, kenaf, abaca, curaua

1. Introduction
Lignins are complex natural biomacromolecules characteristics of vascular
plants, where they provide mechanical support. In addition, lignin waterproofs
the cell wall, enabling transport of water and solutes through the vascular
system, and plays a role in protecting plants against pathogens (1). The lignin
polymer results from the random oxidative coupling of p-hydroxycinnamyl
monolignols mediated by laccases and/or peroxidases (2, 3). The three primary

127
5. Resultados y discusin

monolignols are p-coumaryl, coniferyl and sinapyl alcohols, which produce,


respectively, p-hydroxyphenyl (H), guaiacyl (G), and syringyl (S)
phenylpropanoid units when incorporated into the lignin polymer.
However, it is now widely accepted that other monomers also participate in
coupling reactions giving rise to the lignin macromolecule. This is the case of J-
acylated (with acetate, p-hydroxybenzoate and/or p-coumarate groups) lignins
which have been discovered in many plants. Different grass lignins are partially
p-coumaroylated and some hardwood lignins such as poplar, aspen or willow
lignins are p-hydroxybenzoylated (3-11). Acetylated lignin units have also been
reported to occur in many plants (12-16). Characteristic products from sinapyl
and coniferyl acetate coupling have been detected upon degradative techniques
(Py-GC/MS and derivatization followed by reductive cleavage, DFRC) in the
lignin of several plants characterized by having a high S/G ratio such as sisal,
kenaf, abaca and jute (15, 16). Previous studies have shown that lignin from
these plants is acetylated exclusively at the J-position of the side-chain and that
this acetylation occurred predominantly on S-units (12-16). Moreover, these
studies have provided strong evidence that sinapyl acetate is implicated as a
monomer in lignification in several plants and that the naturally acetylated lignin
derives not from acetylation of the lignin polymer but from polymerization of
pre-acetylated monolignols (14, 16, 17). The same seems also to occur with
sinapyl p-coumarate and sinapyl p-hydroxybenzoate (17-19).
We have recently shown, using a previously developed modification of the
DFRC method to allow detection of natural acetate groups (13) (the so-called
DFCR methodology) that lignin J-acetylation is widespread, and probably
ubiquitous, among angiosperms, although at different extents, but is absent from
conifers (16). Moreover, the lignins of many plants (e.g. the non-woody sisal,
kenaf, abaca or the hardwood hornbeam) are particularly extensively acetylated
(up to 80% of the S-lignin moieties) (16). However, the DFRC degradation
method only cleaves D- and E-aryl ether bonds allowing only the analysis of the
monomeric degradation products released from non-condensed etherified lignin
units. Therefore, the DFRC method, although extremely useful, does not give
information of the whole macromolecule, and the extent of lignin acylation may
be actually different from that estimated.
Spectroscopic techniques, and particularly 1D- and 2D-NMR, can provide
information of the structure of the whole macromolecule and are powerful tools
for lignin structural elucidation (7, 20-24). Therefore, they can be very useful to
estimate the actual extent of lignin acylation. The main advantage of
spectroscopic techniques over degradation methods is the analysis of the whole
lignin structure and direct determination of the different lignin moieties and
inter-unit linkages. In this paper, we study the structure of some naturally
extensively acylated lignins occurring in several herbaceous plants (namely
sisal, kenaf, abaca and curaua) by a combination of chemical degradative
(DFRC) and spectroscopic (2D-NMR) techniques.

128
5. Resultados y discusin

2. Material and methods


2.1. Samples
The plant samples selected for this study consist of bast fibers obtained from the
stalk phloem layer of kenaf (Hibiscus cannabinus) and leaf fibers of sisal (Agave
sisalana), abaca (Musa textilis) and curaua (Ananas erectifolius). The fibers
were finely ground to sawdust using a knife mill (Analysenmhle A10, Janke
and Kunkel GmbH, Staufen, Germany) before analysis. Milled-wood lignin
(MWL) was extracted from finely ball-milled (150 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 (25). The final
yields ranged from 5-15% of the original lignin content. Extension of milling
time, which would increase yield, was avoided to prevent chemical
modifications on the lignin structure.

2.2. DFRC (derivatization followed by reductive cleavage)


The DFRC degradation was performed according to the developed protocol (26-
28). Lignins (10 mg) were stirred for two hours at 50C with acetyl bromide in
acetic acid (8:92). The solvents and excess of bromide were removed by rotary
evaporation. The products were then dissolved in dioxane/acetic acid/water
(5:4:1, v/v/v), and 50 mg powered 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% 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 NaSO4 and the filtrate was evaporated in 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 pyridine. The acetylated
lignin degradation products were collected after rotary evaporation of the
solvents, and subsequently analyzed by GC/MS. To asses the presence of
naturally acetylated lignin units, a modification of the standard DFRC method
by using propionylating instead of acetylating reagents (DFRC) was used in the
present study (13, 16).
The GC/MS analyses were performed with a Varian model Star 3800 GC
equipped with an ion trap detector (Varian model Saturn 4000) using a medium-
length (15 m) capillary column (DB-5HT, 5 m 0.25 mm I.D., 0.1 Pm film
thickness) from J&W Scientific. The oven was heated from 120 (1 min) to 330
C at 6 C/min, and held for 4 min at the final temperature. The injector was
programmed from 60C to 350C at a rate of 200C/min and held until the end of
the analysis. The transfer line was kept at 300 C. Helium was used as the carrier
gas at a rate of 2 mL/min. Quantification of the released individual monomers
was performed using tetracosane as external standard, and by assuming similar

129
5. Resultados y discusin

response factors as those of the acetylated monomers previously reported (26),


although without authentication on our instrument. Molar yields were calculated
on the basis of molecular weights of the respective acetylated and/or
propionylated compounds.

2.3. NMR spectroscopy


NMR spectra were recorded at 25 C on a Bruker AVANCE 500 MHz equipped
with a z-gradient triple resonance probe. Around 40 mg of lignin were dissolved
in 0.75 mL of deuterated dimethylsulfoxide (DMSO-d6) and 2D-NMR spectra
were recorded in HSQC (heteronuclear single quantum correlation) experiments.
The spectral widths were 5000 Hz and 13200 Hz for the 1H- and 13C-
dimensions, respectively. The number of collected complex points was 2048 for
1
H-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 transform
the data matrixes were zero filled up to 1024 points in the 13C-dimension. The
central solvent peak was used as an internal reference ( C 40.1; H 2.50 ppm).
HSQC cross-signals were assigned by combining the results of the different
experiments, and comparing with the literature (20-23, 29-32). A
semiquantitative analysis of the intensities of the HSQC cross-signal intensities
was performed (29, 33). Since the cross-signal intensity depends on the
particular 1JCH value, as well on the T2 relaxation time, a direct analysis of the
intensities is impossible. A more accurate quantification of the 2D HSQC NMR
data analysis was achieved by using the recently published method in which the
errors caused by T2 relaxation, off-resonance effect and coupling constant
deviation could mostly be eliminated (34). Thus, the integration on the cross-
signals was performed separately for the different regions of the HSQC spectra,
which contain signals that correspond to chemically analogous carbon-proton
pairs. For these signals, the 1JCH coupling value is relatively similar, their
chemical shifts are also similar to each other hence the error of off-resonance
effect is small, and therefore can be used semiquantitatively to estimate the
relative abundance of the different species. In the aliphatic oxygenated region,
inter-unit linkages were estimated from C-H correlations to avoid possible
interference from homonuclear 1H-1H couplings, and the relative abundance of
side-chains involved in inter-unit linkages were calculated. In the aromatic
region, C2,6-H2,6 correlations from S units and C2-H2 plus C6-H6 correlations
from G units were used to estimate the S/G ratio of lignin, and p-coumaric acid
content was estimated from its C2,6-H2,6 correlation signal. Lignin acylation was
estimated from the intensities of C-H correlations in acylated and non-acylated
side-chains.

130
5. Resultados y discusin

3. Results and discussion


MWL is a lignin preparation considered as the most representative of the whole
native lignin in the plant (25), in spite of its low yield and the possibility of
some modifications during milling (35). Therefore, in this work, the MWL
isolated from the selected herbaceous plants (sisal, kenaf, abaca and curaua)
were analyzed by 2D-NMR and DFRC to get insight into their structural
characteristics. However, we must still keep in mind that the results obtained
here reflect only the structure of isolated MWL, which only represents a part of
the whole lignin in the plant.

3.1. HSQC-NMR spectra of highly acetylated lignins


The HSQC NMR spectra of the different MWL showed three regions
corresponding to aliphatic, side-chain and aromatic 13C-1H correlations. The
aliphatic (non-oxygenated) region showed signals with no structural information
(except for the presence of acetate signals at GC/GH 20.7/1.74 ppm) and therefore
is not discussed in detail. The side-chain regions (GC/GH 50-90/2.5-5.5 ppm) and
the aromatic regions (GC/GH 95-150/5.5-8.5 ppm) of the MWL selected for this
study are shown in Figures 1 and 2, respectively, and the main substructures
found in these lignins are depicted in Figure 3. The main lignin cross-signals
assigned in the HSQC spectra are listed in Table 1.
In the side-chain region, cross-signals of methoxyls (GC/GH 56.2/3.73 ppm)
and side-chains in EO-4 substructures were the most prominent in all lignins.
Interestingly, the spectra clearly show the presence of intense signals
corresponding to acylated J-carbon in the range from GC/GH 63.5/3.83 and 4.30
ppm in all lignin samples, together with the presence of signals from normal
hydroxylated J-carbon (at GC/GH 60.2/3.30 and 3.70 ppm). The HSQC spectra
indicate that these lignins are extensively acylated and that acylation occurs
exclusively at the J-position of the lignin side-chain. HSQC can not identify the
nature of the ester, although it was shown in previous studies that acetates (and
p-coumarates) occurred in these lignins (12, 13, 15, 16, 36). Traces (less than
0.5% of acylated J-carbon) of a signal at GC/GH 5.87/74.66 ppm corresponding to
acylated D-carbon were found in the HSQC of sisal and kenaf MWL although it
was absent in the rest of the lignins studied here. This signal could be due to a
migration of the acetyl group from the J-carbon to the D-carbon in the lignin
side-chain, as already advanced by Ralph (12).
An estimation of the percentage of J-acylation of the lignin side-chain was
calculated from the HSQC spectra by integration of the signals corresponding to
the hydroxylated and acylated J-carbon (Table 2), and ranged from 58% in
kenaf bast lignin up to 80% in abaca lignin. Although kenaf lignin has already
been known for long to be highly acetylated (12), up to our knowledge, this is
the first time that this type of highly acylated lignin has been described in other

131
5. Resultados y discusin

plants. The high extent of lignin acylation observed in different herbaceous


plants (including both mono- and eudicotyledoneous) indicates that this type of
lignin might be more frequent than previously thought. Naturally acetylated
lignins may also occur in many other plants but their occurrence has probably
being biased due to the limitations of the analytical procedures used for their
isolation and/or structural characterization. Natural acetates present on lignin
might have been hydrolyzed and removed when using traditional isolation
methods (such as alkaline extraction often applied to non-wood lignins) and
degradative procedures for chemical characterization (such as nitrobenzene
oxidation, CuO oxidation or thioacidolysis). Indeed, for spectroscopic analysis,
e.g. using NMR, lignin is frequently in vitro acetylated for improved solubility
and spectroscopic properties, which prevented the detection of natural lignin
acetylation.

GC GC
(a) MeO CE (b) MeO CE
55 55
DE DE
AJ AJ
A' J 60 A' J 60

65 65
AD/AD CJ 70 AD/AD CJ 70

75 75

DD carbohydrates 80 DD 80
A' E A' E carbohydrates
D D CD
AE D D CD AE
85 85

GH 5.0 4.5 4.0 3.5 3.0 GH 5.0 4.5 4.0 3.5 3.0

GC GC
(c) MeO (d) MeO
DE 55 55

AJ AJ
A' J/A'' J 60 A' J/A'' J 60

65 65
AD/AD/AD 70 AD/AD/AD 70

75 75
carbohydrates
DD A' E/A'' E 80 DD carbohydrates 80
A' E/A'' E
D D AE D D AE
85 85

GH 5.0 4.5 4.0 3.5 3.0 GH 5.0 4.5 4.0 3.5 3.0

Figure 1. Expanded side-chain region, GC/GH 50-90/2.5-5.5 ppm, of the HSQC spectra of the
lignins from (a) sisal, (b) kenaf, (c) abaca and (d) curaua. Carbohydrate signals are presented
in grey color. See Table 1 for signal assignment and Figure 3 for the main lignin structures
(A-D) identified.

132
5. Resultados y discusin

GC GC
(a) S2,6 100
(b) S2,6 100
S2,6 (CD=O) S2,6 (CD=O)
S/G 3.9 S/G 5.6
D 2` 110 D 2` 110
G2 G2
G5 D 6` G5
D 6`
G6 120 G6 120

130 130

140 140

GH 8.0 7.5 7.0 6.5 6.0 GH 8.0 7.5 7.0 6.5 6.0

GC GC
(c) S 2,6 100 (d) S2,6 100
S2,6 (CD=O) S 2,6 (CD=O)
S/G 8.7 S/G 4.9
D 2` 110 D 2` 110
G2 A'' E G2 A'' E
G5 G5
G6 D 6` D 6` 120
120 G6
A'' 3,5 A'' 3,5
A'' 2,6 A'' 2,6
130 130

A'' D 140 140


A'' D

GH 8.0 7.5 7.0 6.5 6.0 GH 8.0 7.5 7.0 6.5 6.0

Figure 2. Expanded aromatic region, GC/GH 95-150/5.5-8.5 ppm, of the HSQC spectra of the
lignins from (a) sisal, (b) kenaf, (c) abaca and (d) curaua. See Table 1 for signal assignment
and Figure 3 for the main lignin structures (A-D) identified. G and S are the guaiacyl and
syringyl aromatic units, respectively.

The side-chain region of the HSQC spectra gives also additional information
about the inter-unit linkages present in the structure of these lignins. All the
spectra showed prominent signals corresponding to EO-4 aryl ether linkages.
The CD-HD correlations in EO-4 substructures were observed at GC/GH
72.3/4.86 ppm (structures A, A' and A''), while the CE-HE correlations were
observed at GC/GH 86.5/4.10 ppm in normal J-OH EO-4 aryl ether
substructures (A) but shifted to GC/GH 83.6/4.32 ppm in J-acylated EO-4 aryl
ether substructures (A', A''). EO-4 aryl ether substructures were highly
predominant in all the lignins analyzed here although other substructures were
also observed. Small signals corresponding to spirodienone (E1, DOD
linkages) substructures (D) can be observed in the spectra of sisal, kenaf, abaca
and curaua lignins. Signals of spirodienone CD-HD, CD-HD and CE-HE
correlations were observed at GC/GH 85.4/4.64, 85.4/4.80 and 56.1/3.09 ppm,
respectively. Spirodienone substructures were previously reported in the lignin

133
5. Resultados y discusin
13
Table 1. Assignment of Main Lignin C-1H Cross-Signals in the MWL HSQC Spectra
Shown in Figures 1 and 2.
C/H (ppm) Assignment
53.7/3.12 C-H in EE' (resinol) substructures (C)
56.1/3.09 C-H in E'(spirodienone) substructures (D)
60.0/3.38-3.71 CH in -O-4' substructures (A)
63.8/3.83-4.30 CH in J-acylated -O-4' substructures (A' and A'')
71.7/3.81 and 4.17 C-H in EE' (resinol) substructures (C)
72.3/4.86 C-H in -O-4' substructures (A, A' and A'')
82.1/5.12 CD-HD in E'(spirodienone) substructures (D)
83.6/4.32 C-H in J-acylated -O-4' substructures (A' and A'')
85.4/4.64 C-H in EE' (resinol) substructures (C)
85.4/4.80 CD-HD in E'(spirodienone) substructures (D)
86.5/4.10 C-H in J-OH -O-4' substructures (A)
87.7/5.45 CD-HD in phenylcoumaran substructures (B)
103.8/6.68 C2-H2 and C6-H6 in syringyl units
106.7/7.36 and 7.21 C2-H2 and C6-H6 in oxidized (C=O) syringyl units
111.5/6.99 C2-H2 in guaiacyl units
111.6/6.23 C2-H2 in E'(spirodienone) substructures (D)
114.3/6.24 CE-HE in p-coumaroylated substructures (A'')
115.2/6.71 and 6.94 C5-H5 in guaiacyl units
116.2/6.77 C3-H3 and C5-H5 in p-coumaroylated substructures (A'')
118.3/6.19 C6-H6 in E(spirodienone) substructures (D)
119.5/6.83 C6-H6 in guaiacyl units
130.5/7.4 C2-H2 and C6-H6 in p-coumaroylated substructures (A'')
145.1/7.39 CD-HD in p-coumaroylated substructures (A'')

from kenaf bast fibers by Zhang et al. (37). Phenylcoumaran (E5 linkages)
substructures (B) were also found, although in very small proportions. Very
weak signals corresponding to CD-HD correlations of phenylcoumaran
substructures at GC/GH 87.7/5.45 ppm were observed in the spectra of sisal, kenaf
and curaua lignins, but were absent in the spectrum of abaca. The presence of
these low amounts of phenylcoumaran substructures was expected due to the
very low levels of guaiacyl lignin units in all these samples. Finally, resinol
(EE linkages substructures (C) were clearly observed in the spectrum of
kenaf. Signals for the CD-HD, CE-HE and the double CJ-HJ correlations of resinol
substructures were observed at GC/GH 85.4/4.64, 53.7/3.12 and 71.7/3.81 and
4.17 ppm, respectively. Resinol substructures could also be observed, although
in very small traces, in the spectrum of sisal, but were completely absent in the
spectra of abaca and curaua lignins. The relative abundances of the main inter-
unit linkages present in the MWL selected for this study were calculated from
the HSQC spectra and are shown in Table 2. All these highly acetylated lignins
share a common characteristic, the strikingly high proportion of EO-4 ether
linkages (up to 94% of all linkages) and a very low proportion of condensed
linkages (i.e. E, E and EE). Some of these condensed linkages (E
and EE) are even absent in some lignins (abaca and curaua).

134
5. Resultados y discusin

Table 2. Structural characteristics (percentage of J-acylation, relative abundance of the main


inter-unit linkages, and S/G ratio) observed from the HSQC spectra of the selected MWL.
sisal kenaf abaca curaua

Percentage of J-acylation 68 58 80 69

Linkage relative abundance (% of side-chains involved)


E-O-4' alkyl-aryl ether 89 84 94 94
E-1' (spirodienone) 5 6 6 4
E-5' (phenylcoumaran) 2 2 0 2
E-E' (syringaresinol) 4 8 0 0

S/G ratio 3.9 5.6 8.7 4.9

The main cross-signals in the aromatic region of the HSQC spectra (Figure 2)
correspond to the aromatic rings of the different lignin units. Signals from
syringyl- (S) and guaiacyl- (G) lignin units can be observed in all spectra. The
syringyl units show a prominent signal for the C2,6-H2,6 correlation at GC/GH
103.8/6.68 ppm, while guaiacyl units showed different correlations for C2-H2
(GC/GH 111.5/6.99 ppm), C5-H5 (GC/GH 115.2/6.71 and 6.94) and C6-H6 (GC/GH
119.5/6.83 ppm). Signals corresponding to C2,6-H2,6 correlations in CD-oxidized
S-lignin units were observed at GC/GH 106.7/7.36 and 7.21 ppm. No signals for p-
hydroxyphenyl (H) lignin units could be detected in the HSQC spectra of these
lignins. An estimation of the relative proportions of the S and G-lignin units in
the HSQC spectra revealed that all the lignins selected for this study present a
very high S/G ratio, ranging from 3.9 in sisal to 8.7 in abaca (Table 2). Other
signals present in this region of the HSQC spectra are from spirodienone
substructures (D) with C2-H2 and C6-H6 correlations at GC/GH 111.6/6.23 and
118.3/6.19, respectively. Prominent signals corresponding to p-coumarate
structures were observed in the lignins of abaca and curaua. Cross-signals
corresponding to the correlations C2,6-H2,6 at GC/GH 130.5/7.40 ppm and C3,5-
H3,5 at GC/GH 116.2/6.77 ppm of the aromatic ring and signals for the
correlations of the unsaturated CD-HD at GC/GH 145.1/7.39 and CE-HE at
114.3/6.24 ppm of the p-coumarate unit in structure A'' of Figure 3, were
observed in this region of the HSQC spectra of abaca and curaua. In abaca
lignin, p-coumaric acid has already been reported to be esterified to the lignin
polymer (8, 16, 35, 38).

3.2. Degradation Followed by Reductive Cleavage (DFRC and DFRC)


The HSQC data shown above indicate that these lignins are extensively acylated
at the J-position of the side-chain, but cannot provide additional information on

135
5. Resultados y discusin

the nature of the acyl group (besides the occurrence of acetate and p-coumarate
moieties). A sensitive and selective method is therefore needed to reveal the

HO
5
4 6

3 1
O 2 O
E
D J

HO MeO O MeO O MeO


6 6 6'
J 1 J 1 J 1'
5 5 5'
E 4 2 E 4 2 E 4' 2'
HO HO HO
D 3 D 3 D 3'
O O O

1
OMe 1
OMe OMe
1
6 2 6 2 6 2
5 3 5 3 5 3
4 4 4
MeO OMe MeO OMe MeO OMe

O O O

A A' A''

1 OMe
6 2
OMe
5 3 O
4 3 O 1''
OMe 2 4 6'' 2''
3
HO 4
1 5 5'' 3''
E O 2
J O 6 4''
D J OMe 5 MeO OMe
D MeO 1
6
D
E E
O O
1 J E D
E
6 2 D J
MeO J
5 3 6 HO OH
5 1 O 1
4
6 2
MeO OMe 4 2
3 5 3
O O 4
MeO OMe
OMe O

B C D
Figure 3. Main structures present in the highly acylated lignins studied here: (A) E2-4 aryl
ether linkages; (A') E2-4 aryl ether linkages with acetylated J-carbon; (A'') E2-4 aryl
ether linkages with p-coumaroylated J-carbon; (B) phenylcoumaran structures formed by
E5 and D2-4 linkages; (C) resinol structures formed by EE, DOJand JOD
linkages; and (D) spirodienone structures formed by E1,DO-D linkages.

nature of the acyl group that is esterifying the J-carbon of the lignin side-chain
and to know to which lignin moiety it is attached. The DFRC degradation
method, which cleaves D- and E-ether linkages in the lignin polymer leaving J-
esters intact (26-28), seems to be the most appropriate method for the analysis of
native J-acylated lignin.
DFRC analysis of the lignin samples selected for this study allowed
confirming that p-coumarate groups are attached to the J-carbon of abaca and
curaua lignins, and predominantly on syringyl units (Figure 4). Saturated p-
coumarate (dihydro-p-coumarate) esterified to sinapyl alcohol (as its acetate
derivative, Sdpc) was expected to be the major DFRC degradative compound,

136
5. Resultados y discusin

according to Lu and Ralph (9), and this was the only degradation product that
was quantified in our previous paper (16). However, a closer look to other major
degradation compounds produced upon DFRC of abaca and curaua lignins
indicated the release of important amounts of the unsaturated counterpart, that
is, intact sinapyl p-coumarate (as its acetate derivative, Spc), that was biased, and
therefore not quantified, in our previous paper. Therefore, in this work, we have
now taken into account both compounds to quantify the total abundance of
sinapyl p-coumarate units present in these lignins (Table 3). Trace amounts of
the respective coniferyl p-coumarate could also de detected in abaca and curaua
lignins. Moreover, some amounts of free p-coumaric acid (as its acetate
derivative) could be observed among the DFRC degradation products of curaua
lignin as a broad peak (Figure 4), that could probably correspond to p-coumaric
acid moieties linked to lignin through EO-4 aryl-ether bonds.
The original DFRC degradation method, however, does not allow the analysis
of native acetylated lignin because the degradation products are acetylated
during the degradation procedure, but with appropriate modification of the
protocol by substituting acetylating reagents with propionylating reagents,
DFRC, it is also possible to obtain information about the occurrence of native
lignin acetylation (13). Figure 5 shows the chromatograms of the DFRC
products released from the lignin samples selected in this study. All the analyzed
lignins released the cis and trans isomers of guaiacyl (c-G and t-G) and syringyl
(c-S and t-S) lignin monomers (as their propionylated derivatives) arising from
normal J-OH units in lignin. In addition, the presence of originally J-acetylated
guaiacyl (c-Gac and t-Gac) and syringyl (c-Sac and t-Sac) lignin units could also be
clearly observed in the chromatograms of all of the selected lignins indicating
that acetylation occurred exclusively at the J-carbon of the lignin side-chain, as
already observed in the HSQC spectra.

Table 3. Abundance (Molar Yields) of the DFRC and DFRC degradation monomers of the
MWL isolated from the different plants selected for this study, and relative abundances of the
different acylated (acetylated and p-coumaroylated) lignin moieties.

Monomers (Pmol/g lignin)


G Gac Gpc S Sac Spc %Saca %Spcb %Gacc %Gpcd S/G
sisal 122 124 0 108 378 0 78 0 50 0 2.0
kenaf 390 38 0 543 780 0 59 0 9 0 3.1
abaca 50 3 1 21 131 124 48 45 6 2 5.1
curaua 250 252 3 515 595 195 46 15 50 1 2.6
a
%Sac is the percentage of acetylated S units respect to the total S units. b %Spc is the percentage of p-
coumaroylated S units respect to the total S units. c %Gac is the percentage of acetylated G units respect to the
total G units. d %Gpc is the percentage of p-coumaroylated G units respect to the total G units

137
5. Resultados y discusin

100 tS
(a)
relative intensity

Spc

tG
cS
Spc
S
Sdpc pc
0

100 tS
(b)
relative intensity

Spc

tG

pC
cS Spc
Sdpc Spc

0
5 10 15 20 25
Retention time (min)
Figure 4. Chromatograms of the DFRC degradation products of the MWL from (a) abaca and
(b) curaua, showing the presence of sinapyl alcohol esterified to p-coumarate moieties. Sdpc
and Spc are the sinapyl alcohol esterified with dihydro-p-coumarate and p-coumarate,
respectively (as their acetyl derivative). c-G, t-G, c-S and t-S are the normal cis- and trans-
guaiacyl and syringyl monomers, respectively, (as their acetyl derivatives).

The results from the DFRC and DFRC analysis 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) and p-coumaroylated
guaiacyl (%Gpc) and syringyl (%Spc) lignin moieties, and the S/G ratios, are
presented in Table 3. The data indicate that a high extent of J-acetylation occurs

138
5. Resultados y discusin

in all lignins studied here, and that p-coumaric acid is also found partially
esterifying the lignin of abaca and curaua, in agreement with the NMR data. In
all cases, acetate and p-coumarate groups are preferentially attached to syringyl
units, as previously noted for other lignins (7-9, 12-16, 39). Interestingly, the
high extent of acetylation observed in sisal and curaua also included the G-lignin
units (around 50% of acetylation in both cases). By contrast, in kenaf and abaca
lignins, the J-carbon of G-lignin units is mostly not esterified. On the other
hand, the high extent of acylation of the lignin monomers observed by the
DFCR (and DFRC) method, which can only analyze non-condensed lignin
moieties, is in accordance with the high extent of lignin acylation observed by
HSQC technique, which allows the analysis of the entire MWL structure,
including both condensed and non-condensed linkages. This fact indicates that
both condensed and non-condensed moieties have similar extent of acylation.
However, we must now convey again that the results presented here reflect only
the structure of the isolated MWL, which only represent a small part of the
entire native lignin. However, similar lignin S/G ratio and acylation degree have
been found by in situ analysis in HSQC spectra of the whole cell wall material
(without lignin isolation) at the gel state (40), indicating that MWL can still be
considered as the most representative preparation for the plant native lignin, in
spite of its low yield.
Previous papers describing the structure of some of these lignins have failed
to detect their high levels of acetylation. A recent paper describing the structure
of sisal lignin (41) did not detect the high levels of acetylation, despite of using
spectroscopic techniques. Probably, this was due to the method used for
isolation (acidolysis) that might have hydrolyzed the acetyl groups, or to a
misassignment of the spectral bands. Previous structural studies on abaca lignin
(8, 16, 36, 38), using different degradation methods, also suggested the
occurrence of p-coumaroylated units attached to the J-carbon of the lignin side-
chain. The presence of acetylated J-carbons was also observed in abaca fibers
directly by Py-GC/MS (15) although other authors failed to detect their presence
(8).
On the other hand, the question to whether acylated lignin derives from
polymerization of acylated monolignols or from acylation of the lignin polymer
has recently been addressed and sinapyl acetate has been demonstrated to
behave as a monomer in lignification participating in coupling reactions (14, 16,
17, 42). Part of the evidence comes from the  coupling reactions. If the -
carbon of a monolignol is pre-acylated, the formation of the normal  resinol
structures can not occur because the absence of free -hydroxyls needed to re-
aromatize the quinone methide moiety. Instead, new tetrahydrofuran structures
are formed from the EE homo- and cross-coupling of two sinapyl (acylated
and non-acylated) monolignols, as advanced by Lu and Ralph (14) (Figure 6). It
is clear that tetrahydrofuran structures I and II can only be formed if sinapyl
alcohol is pre-acetylated (at monomer stage) and then undergoes EE coupling.

139
5. Resultados y discusin

t-Sac
100
relative intensity (a)

t-S

t-G
t-G ac
c-Sac
c-G acc-G c-S
0
t-Sac
100
t-S (b)
relative intensity

t-G

t-G ac c-Sac c-S


c-G acc-G
0
t-Sac
100
(c)
relative intensity

t-G
t-S

c-Sac
t-G ac c-S
c-G
0
100 t-Sac
t-S (d)
relative intensity

t-G
t-G ac

c-G c-Sac
c-G ac c-S

0
5.0 7.5 10.0 12.5 15.0
Retention time (min)
Figure 5. Chromatograms of the DFRC degradation products of MWL from (a) sisal, (b)
kenaf, (c) abaca and (d) curaua. c-G, t-G, c-S and t-S are the normal cis- and trans-guaiacyl
and syringyl monomers, respectively (as their propionylated derivatives). c-Gac, t-Gac, c-Sac
and t-Sac are the originally acetylated cis- and trans-guaiacyl and syringyl monomers,
respectively (as their propionylated derivatives).

140
5. Resultados y discusin

Therefore, the presence of these tetrahydrofuran substructures in the lignin


polymer would be indicative of the occurrence of pre-acylated monolignols. In
this work, we have investigated the presence of the tetrahydrofuran structures
arising from EE coupling of sinapyl acetate in the MWL selected for this study
by DFRC. The expected DFRC degradation products of the tetrahydrofuran
EE structures, as suggested by Lu and Ralph (14), are also indicated in Figure
6. Figure 7 shows the reconstructed chromatograms (sum of the single ion
chromatograms of the respective base peaks) of the DFRC degradation products
of the expected tetrahydrofuran dimers arising from the EEcoupling of the
sinapyl monolignols. Interestingly, compounds derived from the DFRC of
homo-coupling (I) and cross-coupling (IIa and IIb) of sinapyl acetate were
clearly observed in the lignins of sisal and kenaf, clearly indicating that in these
lignins sinapyl alcohol is preacetylated and behaves as a real monolignol
participating in post-coupling reactions. However, in the case of abaca and
curaua lignins, no traces of any type of EE linkage (including syringaresinol
and the new tetrahydrofuran structures) could be detected after DFRC, in
agreement with the absence of these linkages observed in the HSQC spectra.
The presence of cross-coupling structures of sinapyl alcohol and sinapyl
acetate indicates that both monolignols are produced simultaneously by the
plant. Moreover, the relative abundance of the compounds released in Figure 7
gives some additional information. In sisal, the relative molar abundance of the
acetylated versus the non-acetylated sinapyl alcohols forming EE linkages
(taking into account that dimer I consists of two sinapyl acetates, dimers II
consist of one sinapyl acetate and one sinapyl alcohol, and dimer III consists of
two sinapyl alcohols) is 44:56, with a slight predominance of the non-acetylated
sinapyl alcohol, whereas their relative molar abundances in ether-linked
structures is 78:22, with a strong predominance of sinapyl acetate. A similar
trend is also observed in kenaf lignin, where the relative molar abundance of the
acetylated versus the non-acetylated sinapyl alcohols forming EE linkages
is23:77, with a predominance of the non-acetylated sinapyl alcohol, whereas
their relative molar abundances in ether-linked structures is 59:41, with a strong
predominance of sinapyl acetate. This indicates that sinapyl acetate has a lower
tendency to form EE linkages than the normal sinapyl alcohol and therefore
those lignins having a high extent of acetylation would produce lower amounts
of EE linkages, as already advanced (16). This means that, probably, the high
level of lignin acetylation is related in some way with the low presence of EE
linkages. This is in agreement with the high proportions of E-O-4 aryl ether
substructures and the low proportion of EE substructures present in these
lignins, as observed in the HSQC spectra shown above. Therefore, it seems that
the high extent of J-acetylation would favor the formation of a predominantly -
O-4 lignin structure, which is indeed devoid of - linkages.

141


142
O O

=
=
CH 3O O

=
O-C-CH 3 O-C-CH 3 O O

=
=
J J O-C-CH 3
J J
CH 3-C-O O-C-CH 3
E E O-C-CH 3
E E

=
D D PrO
EE D D DFRC O
CH 3O OCH 3
O CH 3O
+ homo-coupling
HO OH
CH 3O OCH 3 CH 3O OCH 3 CH 3O OCH 3
OH OH OCH 3 I OCH 3
OPr

I; M +=616; [M-56]+= 560


5. Resultados y discusin

OH
O CH 3O OCH 3 O

=
CH 3O
=

CH 3O
OH O-C-CH 3 OPr O-C-CH 3
J J
O-C-CH 3 OPr

=
E E J
D D E
PrO O PrO
D
EE HO DFRC
+ O CH 3O + CH 3O
J
E D
cross-coupling O

=
CH 3-C-O
CH 3O OCH 3 CH 3O OCH 3 CH 3O OCH 3 CH 3O OCH 3
OCH 3
OH OH
OPr OPr
CH 3O OH
IIa; M += 630; [M-56]+= 574 IIb; M += 630: [M-56]+= 574
II
OCH 3
OH
OH OH CH 3O
J J O
J D OCH 3 OPr
E E
D D E E OPr
EE D J DFRC PrO
+ CH 3O O
homo-coupling CH 3O
HO III
CH 3O OCH 3 CH 3O OCH 3
OH OH OCH 3 CH 3O OCH 3
OPr

III; M += 644; [M-56]+= 588


Figure 4

Figure 6. Structures of the tetrahydrofuran dimers arising from theEE coupling of sinapyl alcohol and sinapyl acetate. I:EE coupling product of
two sinapyl acetates; II: EE coupling product of a sinapyl alcohol and a sinapyl acetate; III: EEcoupling product (syringaresinol) of two sinapyl
alcohols. The aryltetralin products expected from the DFRC degradation of these tetrahydrofuran moieties are also shown, with indication of their
molecular weight and base peak in their mass spectra. Adapted from Lu and Ralph (14).
O
5. Resultados y discusin

IIa,IIb
100
I III
relative intensity
(a)

0
III
100

(b)
relative intensity

IIa,IIb

0
100

(c)
relative intensity

0
100

(d)
relative intensity

0
20.0 22.5 25.0 27.5 30.0 32.5
Retention time (min)
Figure 7. Detail of reconstructed chromatogram (sum of the characteristic ions at m/z 560,
574 and 580) of the DFRC degradation products of the MWL from (a) sisal, (b) kenaf, (c)
abaca and (d) curaua, showing the presence of aryltetralin EE products containing two (I),
one (IIa and IIb) and no (III) native acetates.

143
5. Resultados y discusin

It has been reported that in vitro peroxidase-H2O2 oxidation of equimolar


amounts of sinapyl alcohol and J-acylated sinapyl alcohol produced equal
amounts of the expected EE coupled and cross-coupled products (I, II and III)
shown in Figure 6, in a ratio 1:2:1, suggesting that the coupling reactions were
insensitive to the acylation of the J-carbon (17, 18). However, this is not the case
of what we have seen that occur in the plants, where sinapyl acetate seems to
have lower tendency to form EE linkages and, therefore, a high abundance of
acetylated lignin monomers will ultimately produce a lignin with very low levels
of EE structures during lignification. Therefore, a discrepancy exists between
what is observed in vitro and in the plant. Moreover, as indicated by Lu and
Ralph (17), from the two possible stereoisomers of the EE homo-dimerization
product of sinapyl acetate (structure I in Figure 6), the isomer produced during
in vitro coupling reactions is not the same that is present in plants. Whether
there is a protein (or any other mechanism) in the plant controlling the EE
coupling reaction needs additional investigations.

3.3. Structural features of highly acylated lignins


The lignins selected for this study share some common structural features. First,
they are characterized for being extensively acylated (with either acetate or p-
coumarate groups), exclusively at the Jcarbon of the lignin side-chain, and
preferentially over syringyl units. Moreover, all these lignins present a high
predominance of syringyl over guaiacyl lignin units, a very high predominance
of EO-4 linkages and a very low proportion of EE and E5 linkages and
other condensed bonds that make these lignins very linear and unbranched. In
particular, sisal and kenaf lignins present a high extent of J-acylation,
exclusively with acetate groups, and preferentially on S-lignin moieties in the
case of kenaf lignin and over both S- and G-lignin moieties in the case of sisal
lignin. In both cases, EO-4 aryl ether linkages predominate although some
EE (resinol) and E1 (spirodienones) linkages are observed in low
proportions. On the other hand, the structure of abaca lignin is assembled mostly
with syringyl units with a high extent of acylation of the J-carbon with both
acetate and p-coumarate groups. EO-4 linkages are also predominant in this
lignin. No EE linkages are present but some E1 (spirodienones) linkages can
be observed in abaca lignin. Finally, the lignin of curaua has a predominance of
S-lignin units, a predominance of EO-4 linkages and a high extent of acylation
at the J-carbon with acetate and p-coumarate groups, acetate groups being also
esterifying to a high extent the J-carbon of G-lignin units. In general, all these
structural features make these lignins very different from the structural models
already proposed for softwood (43, 44) and hardwood (45). lignins. All
subsequent lignin structural studies, including those in plant genetics or plant
breeding projects, should take into account the possible occurrence of lignin

144
5. Resultados y discusin

acylation, which in many plants takes place at very high levels, as seen above,
and which have often been overlooked in the past; otherwise the conclusions
drawn may not be representative of the real native lignin structure.

4. Conclusions
The structure of the MWL isolated from the herbaceous plants sisal, kenaf,
abaca and curaua has been elucidated by 2D-NMR and DFRC techniques. The
analyses indicated that the lignins from these plants are extensively acylated at
the J-carbon of the lignin side-chain (with either acetate and/or p-coumarate
groups) and preferentially on syringyl moieties. The structure of these highly
acetylated lignins can be essentially regarded as syringyl units linked mostly
through EO-4 ether bonds, where the J-carbons of the side-chains are
extensively acylated. The lignin polymer is therefore extremely linear and
unbranched. The study of highly acylated lignins will significantly contribute to
redefine the structure of lignin and to complete the lignin biosynthetic pathway.

Acknowledgements
This study has been supported by the Spanish MEC (projects AGL2005-01748
and BIO2007-28719-E) and the EU contract NMP2-CT-2006-26456
(BIORENEW). JR thanks the Spanish CSIC for an I3P fellowship; GM thanks
the Spanish Ministry of Education for a FPI fellowship.

References
(1) Sarkanen, K. V.; Ludwig, C. H. Definition and Nomenclature. In Lignins:
Occurrence, Formation, Structure, and Reactions, Sarkanen, K.V. and
Ludwig, C.H., Ed. Wiley- Intersci.: New York, 1971: pp 1-16.

(2) Boerjan, W.; Ralph, J.; Baucher, M. Lignin biosynthesis. Annu. Rev. Plant
Biol. 2003, 54, 519546.

(3) 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, 29-60.

(4) Smith, D. C. C. p-Hydroxybenzoate groups in the lignin of aspen (Populus


tremula). J. Chem. Soc. 1955, 2347-2351.

(5) Nakano, J.; Ishizu, A.; Migata, N. Studies on lignin. XXXII. Ester groups
of lignin. Tappi 1961, 44, 30-32.

(6) Landucci, L. L.; Deka, G. C.; Roy, D. N. A. 13C NMR study of milled
wood lignins from hybrid Salix clones. Holzforschung 1992, 46, 505-511.

145
5. Resultados y discusin

(7) 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.

(8) Sun, R. C.; Fang, J. M.; Goodwin, A.; Lawther, J. M.; Bolton, J.
Fractionation and characterization of ball-milled and enzyme lignins from
abaca fibre. J. Sci. Food Agric. 1999, 79, 1091-1098.

(9) Lu, F.; Ralph, J. Detection and determination of p-coumaraloylated units in


lignin. J. Agric. Food Chem. 1999, 47, 1985-1992.

(10) Meyermans, H.; Morreel, K.; Lapierre, C.; Pollet, B.; De Bruyn, A.;
Busson, R.; Herdewijn, P.; Devreese, B.; Van Beeumen, J.; Marita, J. M.;
Ralph, J.; Chen, C.; Burggraeve, B.; Van Montagu, M.; Messens, E.;
Boerjan, W. Modification 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. J. Biol.
Chem. 2000, 275, 36899-36909.

(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 D-O-4 substructures. J. Agric. Food Chem. 1997, 45, 1212-1219.

(12) Ralph, J. An unusual lignin from kenaf. J. Nat. Prod. 1996, 59, 341-342.

(13) 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.

(14) Lu, F.; Ralph, J. Preliminary evidence for sinapyl acetate as a lignin
monomer in kenaf. Chem. Commun. 2002, 90-91.

(15) del Ro, J. C.; Gutirrez, A.; Martnez A. T. Identifying acetylated lignin
units in non-wood fibers using pyrolysis-gas chromatography/mass
spectrometry. Rapid Commun. Mass Spectrom. 2004, 18, 1181-1185.

(16) del Ro, J. C.; Marques, G.; Rencoret, J.; Martnez A. T.; Gutirrez, A.
Occurrence of naturally acetylated lignin units. J. Agric. Food Chem. 2007,
55, 5461-5468.

(17) Lu, F.; Ralph, J. Novel EEstructures in lignins incorporating acylated


monolignols. Appita 2005, 233-237.

(18) Lu, F.; Ralph, J.; Morreel, K.; Messens, E.; Boerjan, W. Preparation and
relevance of a cross-coupling product between sinapyl alcohol and sinapyl
p-hydroxybenzoate. Org. Biomol. Chem. 2004, 2888-2890.

146
5. Resultados y discusin

(19) Morreel, K.; Ralph, J.; Kim, H.; Lu, F.; Goeminne, G.; Ralph, S. A.;
Messens, E.; Boerjan, W. Profiling of oligolignols reveals monolignols
coupling conditions in lignifying poplar xylem. Plant Physiol. 2004, 136,
3537-3549.

(20) 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; pp 55-108.

(21) 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, 2004
(http://ars.usda.gov/Services/docs.htm?docid=10491) (accessed: July
2006): 2004.

(22) 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.

(23) 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.

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

(25) Bjrkman, A. Studies on finely divided wood. Part I. Extraction of lignin


with neutral solvents. Sven. Papperstidn. 1956, 59, 477-485.

(26) 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.

(27) Lu, F.; Ralph, J. The DFRC method for lignin analysis. Part 1. A new
method for Earyl ether cleavage: lignin model studies. J. Agric. Food
Chem. 1997, 45, 4655-4660.

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

147
5. Resultados y discusin

(29) Liiti, T. M.; Maunu, S. L.; Hortling, B.; Toikka, M.; Kilpelinen, I.
Analysis of technical lignins by two- and three-dimensional NMR
spectroscopy. J. Agric. Food Chem. 2003, 51, 2136-2143.

(30) mmlahti, E.; Brunow, G.; Bardet, M.; Robert, D.; Kilpelinen, I.
Identification of side-chain structures in a poplar lignin using three-
dimensional HMQC-HOHAHA NMR spectroscopy. J. Agric. Food Chem.
1998, 46, 5113-5117.

(31) Ibarra, D.; Chvez, M. I.; Rencoret, J.; del Ro, J. C.; Gutirrez, A.;
Romero, J.; Camarero, S.; Martnez, M. J.; Jimnez-Barbero, J.; Martnez,
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,
3477-3490.

(32) Ibarra, D.; Chvez, M. I.; Rencoret, J.; del Ro, J. C.; Gutirrez, A.;
Romero, J.; Camarero, S.; Martnez, M. J.; Jimnez-Barbero, J.; Martnez
A. T. Structural modification of eucalypt pulp lignin in a totally chlorine
free bleaching sequence including a laccase-mediator stage. Holzforschung
2007, 61, 634-646.

(33) Heikkinen, S.; Toikka, M. M.; Karhunen, P. T.; Kilpelinen, I. A.


Quantitative 2D HSQC (Q-HSQC) via suppression of J-dependence of
polarization transfer in NMR spectroscopy: Application to wood lignin. J.
Am. Chem. Soc. 2003, 125, 4362-4367.

(34) Zhang, L.; Gellerstedt, G. Quantitative 2D HSQC NMR determination of


polymer structures by selecting the suitable internal standard references.
Magn. Reson. Chem., 2007, 45(1), 37-45.

(35) Holtman, K. H.; Chang, H. M.; Jameel, H.; Kaddla, J. F. Quantitative 13C
NMR characterization of milled wood lignins isolated by different milling
techniques. J. Wood Chem. Technol. 2006, 26, 21-34.

(36) del Ro, J. C.; Gutirrez, A.; Rodrguez, I. M.; Ibarra, D.; Martnez, 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.

(37) Zhang, L; Gellerstedt, G.; Ralph, J.; Lu, F. NMR studies on the occurrence
of spirodienone structures in lignins. J. Wood Chem. Technol. 2006, 26, 65-
79.

148
5. Resultados y discusin

(38) del Ro, J. C.; Gutirrez, A. Chemical composition of abaca (Musa textilis)
leaf fibers used for manufacturing of high quality paper pulps. J. Agric.
Food Chem. 2006, 54, 4600-4610.

(39) Grabber, J. H.; Quideau, S.; Ralph, J., p-Coumaroylated syringyl units in
maize lignin; implications for -ether cleavage by thioacidolysis.
Phytochemistry 1996, 43, 1189-1194.

(40) Rencoret, J.; Marques, G.; Gutirrez, A.; Nieto, L.; Santos, J. I.; Jimnez-
Barbero, J.; Martnez, A. T.; del Ro, J. C. In situ analysis of lignin by
2D-NMR of wood (Eucalyptus globulus and Picea abies) and non-woody
(Agave sisalana) plant materials at the gel state. Proc. EWLP-2008,
Stockholm, 26-29 August.

(41) Megiatto, J. D.; Hoareau, W.; Gardrat, C.; Frollini, E.; Castellain, A. Sisal
fibers: surface chemical modification using reagent obtained from a
renewable source; characterization of hemicellulose and lignin as model
study. J. Agric. Food Chem. 2007, 55, 8576-8584.

(42) 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.

(43) Adler, E. Lignin chemistry past, present and future. Wood Sci. Technol.
1977, 11, 169-218.

(44) Brunow, G. Methods to Reveal the Structure of Lignin. In: Hofrichter M &
Steinbchel A, (ed), Lignin, Humic Substances and Coal, Vol 1, 2001, pp.
89116, Wiley-VHC, Weinheim.

(45) Nimz, H. Beech lignin- Proposal of a constitutional scheme. Agnew. Chem.


1974, 13(5), 313-321.

149
5. Resultados y discusin

Publicacin IV:

Marques G., Gutirrez A. and del Ro J.C (2007) Chemical characterization of


lignin and lipophilic fractions from leaf fibers of curaua (Ananas erectifolius).
Journal of Agriculture and Food Chemistry, 55, 1327-1336.

150
5. Resultados y discusin

Chemical characterization of lignin and lipophilic fractions from leaf fibers


of curaua (Ananas erectifolius)

Gisela Marques, Ana Gutirrez and Jos C. del Ro

Instituto de Recursos Naturales y Agrobiologa de Sevilla, CSIC, P.O. Box 1052, 41080-
Seville, Spain

Abstract
The chemical composition of leaf fibers of curaua (Ananas erectifolius), an
herbaceous plant native of Amazonia, was studied. Special attention was paid to
the content and composition of lignin and lipophilic compounds. The analysis of
lignin in the curaua fibers was performed in situ by pyrolysis-gas
chromatography/mass spectrometry (Py-GC/MS) and showed a lignin
composition with a p-hydroxyphenyl:guaiacyl:syringyl units (H:G:S) molar
proportion of 30:29:41 (S/G molar ratio of 1.4). The presence of p-
hydroxycinnamic acids (p-coumaric and ferulic acids) in curaua fibers was
revealed upon pyrolysis in the presence of tetramethylammonium hydroxide. On
the other hand, the main lipophilic compounds, analysed by GC/MS, were series
of long-chain n-fatty acids, n-fatty alcohols, D- and Z-hydroxyacids,
monoglycerides, sterols and waxes. Other compounds, such as Z-hydroxy
monoesters and Z-hydroxy acylesters of glycerol were also found in this fiber in
high amounts.
Keywords: Curaua; Ananas erectifolius; lipids; lignin; pyrolysis; hydroxy
monoesters; glyceryl esters; paper pulp.

1. Introduction
An alternative to wood raw materials for pulp and paper production in
developing countries is the use of non-woody fibers from herbaceous field
crops. In developed countries non-woody fibers are mainly used for the
production of specialty papers, i.e. tea bags, filter papers, bank notes, etc. The
main sources of non-woody raw materials are agricultural residues from
monocotyledons, including cereal straw and bagasse. Bamboo, reeds and some
other grass plants such as flax, hemp, kenaf, jute, sisal or abaca are also grown
or collected for the pulp industry but increased attention has been paid in recent
years to find new non-wood raw materials for pulp production.
Curaua (Ananas erectifolius), an herbaceous plant native of the Amazonian
region and member of the bromeliad family, has been recognized since pre-
Columbian days for its valuable fibers. In the last decade, it has gained
commercial recognition as material for composites for automotive industry [1-
4]. The curaua fiber has also been promoted for paper pulp in Brazil [5] and it is

151
5. Resultados y discusin

being now investigated as an alternative lignocellulosic material for the


production of chemical pulps.
Studies on the chemical composition of curaua fibers are important to
evaluate this fiber as a potential raw material for pulp and papermaking,
however only limited studies have been performed so far on this interesting fiber
[1-4]. In this work, we have performed a chemical characterization of curaua
fibers, paying especial attention to the content and composition of the lipophilic
compounds and the structural characterization of lignin, since these two organic
fractions are of high importance during pulping and papermaking. It is known
that the efficiency of pulping is directly proportional to the amount of syringyl
(S) units in lignin [6, 7]. This is because the S-lignin is mainly linked by more
labile ether bonds, is relatively unbranched and has lower condensation degree
that G-lignin [8, 9]. Indeed, the S-lignin has higher reactivity in alkaline systems
than G-lignin [10]. On the other hand, the lipophilic compounds present in raw
materials cause significant environmental and technical problems in the
manufacturing of paper pulp. During pulping, lipids are released from the fibers
forming colloidal pitch, which can deposit in either pulp or machinery causing
production troubles [11-13]. Moreover, such extractives might also contribute to
the toxicity of paper pulp effluents and products [14, 15].
In the present study, the lignin in curaua fibers was characterized in situ by
using analytical pyrolysis coupled to gas chromatography/mass spectrometry
(Py-GC/MS), which is a powerful analytical tool for the rapid analysis of
complex polymer mixtures, including lignocellulosic materials [16, 17].
Pyrolysis in the presence of a methylating reagent, tetramethylammonium
hydroxide (TMAH), was used for the analysis of p-hydroxicinnamic acids (p-
coumaric and ferulic acids). On the other hand, the lipid composition in curaua
fibers was analyzed by gas chromatography (GC) and gas chromatography/mass
spectrometry (GC/MS), using short- and medium-length high-temperature
capillary columns, respectively [18], which enable the elution and analysis of
intact high molecular weight lipids such as waxes, sterol esters, and
triglycerides.

2. Material and methods


2.1. Samples
Curaua (Ananas erectifolius) fibers were supplied by CELESA pulp mill
(Tortosa, Spain). The dried samples were milled using a knife mill (Janke and
Kunkel, Analysenmhle). For the isolation of lipids, the milled samples were
extracted with acetone in a Soxhlet apparatus for 8 h. The acetone extracts were
evaporated to dryness and weighted. Then, the extracts were resuspended in
chloroform for chromatographic analysis of the lipophilic fraction. Two
replicates were used for each sample, and all samples were subjected to GC and
GC/MS analyses. For carbohydrate analysis and estimation of the Klason lignin

152
5. Resultados y discusin

content, the acetone extracted samples were subsequently extracted with hot
water (3 h at 100 C) to remove the water-soluble material. Holocellulose was
isolated from the pre-extracted fibers by delignification for 4 hours using the
acid chlorite method (19). The D-cellulose content was determined by removing
the hemicelluloses from the holocellulose by alkali extraction (19). Klason
lignin was estimated as the residue after sulfuric acid hydrolysis of the pre-
extracted material according to Tappi rule T222 om-88 [20]. The acid-soluble
lignin was determined, after filtering off the insoluble lignin, by
spectrophotometric determination at 205 nm wavelength. Neutral sugars from
polysaccharide hydrolysis were analyzed as alditol acetates by GC according to
Tappi rule T249 om85 [20]. Ash content was estimated as the residue after 6 h at
575 C. The general composition (as percent of whole fiber) was as follows:
holocellulose, 92.5%; D-cellulose, 66.4%; ash, 1.3%; acetone extractives, 5.3%;
water-soluble extract, 5.1%; Klason lignin, 4.9%; acid-soluble lignin, 1.6%. The
composition of neutral monosaccharides (as percent of total neutral
carbohydrates) included arabinose, 2.7%; xylose, 8.0%; mannose, 3.5%;
galactose, 0.2%; and glucose, 85.6%. No uronic acid determination was
performed in this study. The composition of metals and other elements was
analyzed by inductively coupled plasma spectrophotometry (ICP-OES) after
oxidation with concentrated HNO3 under pressure in a microwave digestor, with
the following results: K, 2770 ppm; Ca, 2025 ppm; Mg, 945 ppm; Mn, 120 ppm;
Na, 95 ppm; Al, 86 ppm; Fe, 82 ppm; Sr, 10 ppm; Zn, 4 ppm.

2.2. Solid Phase Extraction (SPE) fractionation


For a better characterization of the different homologous series, the lipid extracts
were fractionated by a SPE procedure using aminopropyl-phase cartridges (500
mg) from Waters Division of Millipore (Mildford, MA, USA), as already
described [18]. Briefly, the dried chloroform extracts were taken up in a minimal
volume (< 0.5 mL) of hexane:chloroform (4:1) and loaded into the cartridge
column previously conditioned with hexane (4 mL). The cartridge was loaded
and eluted by gravity. The column was first eluted with 8 mL of hexane and
subsequently with 6 ml of hexane:chloroform (5:1), then with 10 mL of
chloroform and finally with 10 mL of diethyl ether:acetic acid (98:2). Each
isolated fraction was dried under nitrogen and analyzed by GC and GC/MS.

2.3. GC and GC/MS analyses


The GC analyses of the extracts were performed in an Agilent 6890N Network
GC system using a 5 m 0.25 mm i.d., 0.1 Pm DB-5HT fused silica capillary
column from J&W Scientific (Folsom, CA, USA). The temperature program
was started at 100 C with a 1-min hold and then raised to a final temperature of
350 C at 15 C/min, and held for 3 min. The injector and flame-ionization
detector temperatures were set at 300 and 350 C, respectively. Helium was used

153
5. Resultados y discusin

as the carrier gas at a rate of 5 mL/min, and the injection was performed in
splitless mode. Peaks were quantified by area, and a mixture of standards
(octadecane, palmitic acid, sitosterol and cholesteryl oleate) was used to
elaborate calibration curves. The data from the two replicates were averaged. In
all cases the standard derivations from replicates were below 10% of the mean
values.
The GC/MS analyses were performed with a Varian model Star 3400 GC
equipped with a model Saturn 2000 ion trap detector using a medium-length (12
m) capillary column of the same characteristics described above. The oven was
heated from 120 C (1 min) to 380 C at 10 C/min and held for 5 min. The
transfer line was kept at 300 C. The injector was temperature programmed from
120 C (0.1 min) to 380 C at a rate of 200 C/min and held until the end of the
analysis. Helium was used as the carrier gas at a rate of 2 mL/min. Methylation
with trimethylsilyldiazomethane and silylation with
bis(trimethylsilyl)trifluoroacetamide (BSTFA) was used when required.
Compounds were identified by comparing their mass spectra with mass spectra
in Wiley and NIST libraries, by mass fragmentography, and when possible, by
comparison with authentic standards.

2.4. Py-GC/MS
The pyrolysis of curaua fibers (approximately 100 Pg) was performed in
duplicate with a model 2020 micro-furnace pyrolyzer (Frontier Laboratories
Ltd., Yoriyama, Japan) directly connected to an Agilent 6890 GC/MS system
equipped with a 30 m 0.25 mm i.d., 0.25 Pm HP 5MS fused silica capillary
column. The detector consisted of an Agilent 5973 mass selective detector (EI at
70 eV). The pyrolysis was performed at 500 C. The final temperature was
achieved at a rate of 20 C/min. The GC/MS conditions were as follows: oven
temperature was held at 50 C for 1 min and then increased up to 100 C at
30 C/min, from 100 to 300 C at 10 C/min and isothermal at 300 C for
10 min. The carrier gas used was helium with a controlled flow of 1 ml/min. For
the pyrolysis in the presence of TMAH, approximately 100 Pg of sample was
mixed with 0.5 PL of 25% TMAH. The pyrolysis was carried out as described
above. The compounds were identified by comparing the mass spectra obtained
with those of the Wiley and NIST computer libraries and that reported in the
literature [16, 17]. Relative peak molar areas were calculated for carbohydrate
and lignin pyrolysis products. The summed molar areas of the relevant peaks
were normalized to 100%, and the data for two repetitive pyrolysis experiments
were averaged. The relative standard deviation for the pyrolysis data was less
than 5%.

154
5. Resultados y discusin

3. Results and discussion


The curaua fiber was characterized by a high holocellulose and D-cellulose
contents (92.5 and 66.4, respectively), and a low lignin content (6.5% of the
total fiber weight). This lignin content is similar to other non-wood fibers such
as flax or hemp and lower than other non-wood fibers such as kenaf or abaca
[21-26]. The extractives content (5.3% of total fiber weight) is very high, and
much higher than other nonwood fibers, which are usually less than 1% [21-26].
However, most of the acetone extract corresponds to polar compounds, while
only 1.3% corresponded to lipophilic compounds, which were estimated by
redissolving the acetone extracts in chloroform. On the other hand, the
hemicellulose fraction was mainly constituted by xylose. Finally, the ash content
(1.3% of total fiber weight) was low in comparison to cereal straw [27], and the
composition of the different metals revealed a predominance of Ca and K, and a
very low content of other metals.

3.1. Lignin composition


The lignin composition of curaua fibers was analyzed in situ by Py-GC/MS.
The Py-GC/MS chromatogram of curaua fibers is shown in Figure 1 and the
identities and relative molar abundances of the released compounds are listed in
Table 1. The Py-GC/MS of curaua fibers released predominantly compounds
arising from carbohydrates, with only minor amounts of lignin-derived phenols.
Carbohydrate pyrolysis products represented 88% on average and phenols from
lignin represented only 12% of the total identified compounds, which is in
agreement with the low lignin content estimated as Klason lignin. Among the
lignin derived compounds, the pyrogram of curaua fibers showed compounds
derived from p-hydroxyphenyl (H), guaiacyl (G) and syringyl (S) lignin units,
with a slight predominance of the S units. The main lignin-derived compounds
identified were 4-methylphenol (13), guaiacol (14), 4-vinylphenol (19), 4-
vinylguaiacol (23), syringol (24), 4-ethylsyringol (33), 4-vinylsyringol (34) and
trans-4-propenylsyringol (40). The relative molar distribution of the different
lignin units (H:G:S) estimated by Py-GC/MS were 30:29:41 with a S/G molar
ratio of 1.4. The predominance of S-lignin observed in the curaua fiber is
advantageous for delignification during pulping because the S-lignin is
relatively unbranched and has lower condensation degree than H- and G-lignins.
Moreover, S-lignin is more reactive in alkaline media [10].
It must be noted that the relatively high abundances of 4-vinylphenol (19)
observed in the pyrogram of curaua fibers could be due to the presence of p-
coumaric acid, which upon pyrolysis will decarboxylate to produce 4-
vinylphenol [28]. p-Hydroxycinnamic acids (p-coumaric and ferulic acids) occur
widely in the cell walls of herbaceous plants forming cross-linkages between
lignin and polysaccharides [29-34]. The presence of p-hydroxycinnamic acids
constitutes a complication for lignin analyses by analytical pyrolysis since they

155
5. Resultados y discusin

yield pyrolysis products similar to those of corresponding lignin units. However,


this problem can be solved by the use of pyrolysis in the presence of TMAH
(Py/TMAH), which prevents decarboxylation and releases intact p-
hydroxycinnamic acids (as their methyl derivatives), in addition to different
lignin degradation products [28, 35-38].
Py/TMAH of curaua fibers released significant amounts of the methyl
derivative of p-coumaric acid (25% of the lignin and cinnamic acids released
products) as well as minor amounts of the methyl derivative of ferulic acid (5%
of the lignin and cinnamic acids released products). p-Hydroxycinnamic acids
are present in curaua fiber in relatively high amounts (cinnamic acids/lignin
ratio of 0.4, estimated after Py/TMAH) and agrees with the relatively high
content of 4-vinylphenol released by Py-GC/MS. Studies on maize [39], wheat
[40] and other grasses including bamboo [41] revealed that p-coumaric acid is
esterified at the J-position of lignin side-chains, and predominantly to S units
[41, 42]. Therefore, probably the major part of the p-coumaric acid in curaua
fibers also attaches at the J-position of the lignin side-chain by ester bonds. The
relatively high content of p-hydroxycinnamic acids in curaua fibers would also
be advantageous for pulping since ester bonds are easily cleaved during cooking.

11,12
6

19

10
5
3 18
2

15 23
34

24 32,33
13 16
14 20
7
40
17 21 37
22 27 29
25
38
26 28 30 31 35 36 39 42 43
41 44

2 4 6 8 10 12 14 16 18 20 22
Retention time (minutes)

Figure 1. Py-GC/MS chromatogram of curaua fibers. The identities and relative molar
abundances of the compounds are listed in Table 1.

156
5. Resultados y discusin

Table 1. Identification and Relative Molar Abundances (%) of the Compounds Released after
Py-GC/MS of Curaua Fibers.
No Compound Mass Fragments MW Origin %
1 acetic acid 45/60 60 C 35.8
2 2-hydroxypropanal 43/74 74 C 3.1
3 (3H)-furan-2-one 55/84 84 C 3.1
4 1,3-hydroxydihydro-2-(3H)-furanone 58/102 102 C 6.0
5 (2H)-furan-3-one 55/84 84 C 1.4
6 2-furaldehyde 67/95/96 96 C 6.4
7 cyclopent-1-ene-3,4-dione 54/68/96 96 C 0.9
8 (5H)-furan-2-one 55/84 84 C 4.2
9 2,3-dihydro-5-methylfuran-2-one 55/69/98 98 C 8.5
10 4-hydroxy-5,6-dihydro-(2H)-piran-2-one 58/85/114 114 C 2.1
11 3-hydroxy-2-methyl-2-cyclopenten-1-one 55/85/112 112 C 1.1
12 2-hydroxy-3-methyl-2-cyclopenten-1-one 55/85/112 112 C 4.8
13 4-methylphenol 77/107/108 108 LH 0.7
14 guaiacol 81/109/124 124 LG 0.5
15 2 furoic acid, methyl ester 67/95/126 126 C 1.3
16 4-methylguaiacol 95/123/138 138 LG 0.1
17 3,4-dihydroxybenzaldehyde 81/109/137/138 138 L 0.5
18 catechol 64/81/92/110 110 L/C 1.0
19 4-vinylphenol 65/91/120 120 LH/pCA 2.2
20 5-hydroxymethyl-2-furaldehyde 69/97/109/126 126 C 1.9
21 3-methoxycatechol 60/97/125/140 140 L 0.3
22 4-ethylguaiacol 122/137/152 152 LG 0.2
23 4-vinylguaiacol 107/135/150 150 LG 1.1
24 syringol 111/139/154 154 LS 0.7
25 eugenol 131/149/164 164 LG 0.2
26 4-propylguaiacol 122/136/166 166 LG <0.1
27 vanillin 109/151/152 152 LG 0.3
28 cis-isoeugenol 131/149/164 164 LG <0.1
29 4-methylsyringol 125/153/168 168 LS 0.2
30 trans-isoeugenol 131/149/164 164 LG 0.1
31 acetoguaiacone 123/151/166 166 LG 0.1
32 levoglucosane 60/98 162 C 7.7
33 4-ethylsyringol 167/182 182 LS 0.8
34 4-vinylsyringol 137/165/180 180 LS 0.9
35 4-allylsyringol 167/179/194 194 LS 0.1
36 4-propylsyringol 123/167/196 196 LS 0.1
37 cis-4-propenylsyringol 167/179/194 194 LS 0.2
38 syringaldehyde 167/181/182 182 LS 0.2
39 4-propinylsyringol 106/131/177/192 192 LS 0.1
40 trans-4-propenylsyringol 167/179/194 194 LS 0.5
41 trans-coniferaldehyde 107/135/147/178 178 LG 0.1
42 acetosyringone 153/181/196 196 LS 0.1
43 syringylacetone 123/167/210 210 LS 0.1
44 trans-sinapaldehyde 137/165/180/208 208 LS 0.1
%H 29.8
%G 29.1
%S 41.1
S/G 1.4
%L 11.7
%C 88.3
L/C 0.13
C, carbohydrates; L, lignin; LH, p-hydroxyphenyl lignin units, H; LG, guaiacyl lignin units, G; LS, syringyl
lignin units, S; pCA, p-coumaric acid.

157
5. Resultados y discusin

3.2. Lipid composition


The total lipid extract of the curaua fibers accounted for 1.3% of the total fiber
weight. The extracts were analyzed by GC and GC/MS according to the method
developed by Gutirrez et al. [18]. The chromatogram of the curaua extracts (as
methyl ester and TMSi ether derivatives) is shown in Figure 2 and the detailed
list with the identities and abundances of the main compounds present are
summarized in Table 2. The main compounds identified were series of n-fatty
acids, n-fatty alcohols, D- and Z-hydroxyacids, monoglycerides, sterols and
waxes. Other series of high molecular weight compounds such as Z-hydroxy
monoesters and Z-hydroxy acylesters of glycerol, as well as sterol esters and
sterol glycosides, were also present in important amounts. The structures of the
main lipophilic compounds identified in the curaua extract are shown in Figure
3. The Z-hydroxy fatty acids both in free or esterified form (forming esters with
both fatty alcohols and glycerol), was the main series of compounds present in
the extracts.

OHM24

3,4

OH20 OH24

OHM26
Al22

Al24
M24

OH22 OH22
1
OH24
2
M26

W38
FA18:2 FA24 SG
FA16:1 + FA22
FA18:1 OHM28
FA18 FA20 M28
FA16 W40 CG

W42

CE SE

5 10 15 20 25
Retention time (minutes)

Figure 2. GC/MS chromatogram of the methyl ester and TMSi ether derivative of the lipid
extract from curaua fibers. FA(n), n-fatty acid series; Al(n), alcohol series; W(n), wax series;
DOH(n) and ZOH(n), D and Z-hydroxy fatty acids series; M(n), monoglyceride series;
ZOHM(n), Z-hydroxy acylesters of glycerol series; SG, sitosteryl 3E-D-glucopyranoside; CG,
campesteryl 3E-D-glucopyranoside;1, campesterol; 2, ergostanol; 3, sitosterol; 4,
stigmastanol; CE, campesterol ester; SE, sitosterol ester; n denotes the total carbon atom
number.

158
5. Resultados y discusin

Table 2. Composition and Abundance (mg/Kg) of Lipophilic Compounds in Curaua Fibers

compound mass fragments (m/z) MW abundance

fatty acids 813.2


n-tetradecanoic acid 73/117/145/285* 285* 4.3
n-pentadecanoic acid 73/117/145/299* 299* 3.3
9-hexadecenoic acid 55/69/236/254 254 7.6
n-hexadecanoic acid 60/73/129/256 256 162.5
9,12-octadecadienoic acid 67/81/280 280 23.0
9-octadecenoic acid 55/69/264 282 91.6
n-octadecanoic acid 60/73/129/284 284 262.0
n-eicosanoic acid 60/73/129/312 312 24.2
n-docosanoic acid 60/73/129/340 340 138.2
n-tetracosanoic acid 60/73/129/368 368 96.5

Z-hydroxy fatty acids 1423.3


16-hydroxyhexadecanoic acid 311/343/359# 374# 20.2
18-hydroxyoctadecanoic acid 339/371/387# 402# 28.4
20-hydroxyeicosanoic acid 367/399/415# 430# 30.8
22-hydroxydocosanoic acid 395/427/443# 458# 235.0
24-hydroxytetracosanoic acid 423/455/471# 486# 367.0
26-hydroxyhexacosanoic acid 451/483/499# 514# 532.5
28-hydroxyoctacosanoic acid 479/511/527# 542# 119.2
30-hydroxytriacontanoic acid 507/539/555# 570# 90.2

D-hydroxy fatty acids 226.5


2-hydroxyeicosanoic acid 73/117/355* 472* 62.7
2-hydroxydocosanoic acid 73/117/149/383* 500* 19.9
2-hydroxytetracosanoic acid 73/117/411* 528* 79.2
2-hydroxyhexacosanoic acid 73/117/439* 556* 64.7

fatty alcohols 552.4


n-eicosanol 75/103/355* 370* 8.9
n-docosanol 75/103/383* 398* 247.2
n-tetracosanol 75/103/411* 426* 242.2
n-hexacosanol 75/103/439* 454* 44.3
n-octacosanol 75/103/467* 482* 9.8

sterols 618.7
campesterol 55/145/213/382/400 400 56.9
ergostanol 215/402 402 145.7
sitosterol 145/213/396/414 414 226.4
stigmastanol 215/416 416 189.7

tocopherols 31.4
D-tocopherol 165/205/430 430 31.4

steroid hydrocarbons 119.4


ergostatriene 135/143/380 380 23.8

159
5. Resultados y discusin

ergostadiene 81/147/367/382 382 14.4


stigmastadiene 81/147/381/396 396 71.8
stigmasta-3,5,22-triene 135/143/394 394 2.6
stigmasta-3,5-diene 81/147/381/396 396 6.8

steroid ketones 57.9


stigmasta-3,5-dien-7-one 174/269/410 410 7.7
stigmast-4-en-3-one 124/229/412 412 24.4
stigmastadienone isomer 57/136/174/269/410 410 16.4
stigmastane-3,6-dione 245/287/428 428 9.4

sterol esters 89.4


campesterol ester 49.3
sitosterol ester 40.1

steryl glycosides 264.9


campesteryl 3-D- 204/217/361/383* 850* 141.3
glucopyranoside
sitosteryl 3-D-glucopyranoside 204/217/361/397* 864* 123.6

waxes 173.2
C36 201/229/257/285/536 536 3.5
C37 243/257/550 550 3.3
C38 257/564 564 56.8
C39 243/257/271/285/299/578 578 5.9
C40 257/285/313/592 592 46.5
C40:1 264/283/590 590 2.9
C41 257/271/285/299/313/327/341/355/606 606 2.3
C42 257/285/313/341/620 620 24.9
C42:1 264/283/618 618 1.2
C43 257/271/285/299/313/327/355/369/634 634 1.5
C44 257/285/313/341/648 648 19.1
C46 257/285/313/341/369/397/676 676 5.3

Z-hydroxy monoesters 369.5


C36 73/129/237/311/609* 624* 29.1
C37 73/129/237/311/623* 638* 3.0
C38 73/129/237/311/339/637* 652* 229.8
C40 73/129/237/311/339/367/395/665* 680* 95.6
C42 73/129/237/311/339/367/395/693* 708* 11.7
C44 73/129/367/395/721* 736* 0.3

monoglycerides 714.6
1-monotetradecanoylglycerol 73/103/129/147/343/431* 446* 5.1
1-monohexadecanoylglycerol 73/103/129/147/371/459* 474* 5.3
1-monooctadecanoylglycerol 73/103/129/147/399/487* 502* 5.3
1-monoeicosanoylglycerol 73/103/129/147/427/515* 530* 31.6
1-monodocosanoylglycerol 73/103/129/147/455/543* 558* 288.2
1-monotetracosanoylglycerol 73/103/129/147/483/571* 586* 179.3
1-monohexacosanoylglycerol 73/103/129/147/511/599* 614* 168.2

160
5. Resultados y discusin

1-monooctacosanoylglycerol 73/103/129/147/539/627* 642* 23.4


1-monotriacontanoylglycerol 73/103/129/147/567/655* 670* 8.2

Z-hydroxy acylesters of 960.8


glycerol
1-mono(22- 73/103/129/147/203/486/543/631* 646* 116.8
hydroxydocosanoyl)glycerol
1-mono(24- 73/103/129/147/203/514/571/659* 674* 482.8
hydroxytetracosanoyl)glycerol
1-mono(26- 73/103/129/147/203/542/599/687* 702* 301.7
hydroxyhexacosanoyl)glycerol
1-mono(28- 73/103/129/147/203/570/627/715* 730* 59.5
hydroxyoctacosanoyl)glycerol
tr
traces,* as TMSi ether derivates, # as methyl ester and TMSi ether derivates.

Waxes (esters of fatty acids to fatty alcohols) were also important components
of the curaua fiber extracts and were found in the range from C36 to C46. Among
the waxes, the GC/MS analysis revealed that each chromatographic peak
consisted of a complex mixture of different long-chain fatty acids esterified to
different long-chain fatty alcohols. The identification and quantification of the
individual long-chain esters in each chromatographic peak was resolved based
on 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 2H atoms from the alcohol chain to the acid chain giving a protonated
acid ion [24, 43-45]. Therefore, the base peak gives the number of carbon atoms
in the acid moiety and the molecular ion the total number of carbon atoms in the
ester. It is possible then to determine the individual contribution of the esters to
every chromatographic peak by mass spectrometric determination of the
molecular ion and the base peak. Quantification of individual esters was
accomplished by integrating areas in the chromatographic profiles of ions
characteristic for the acidic moiety. The detailed structural composition and
abundance of the high molecular weight waxes identified in the curaua fiber is
shown in Table 3. The esterified fatty acids ranged from C12 to C25 and the
esterified fatty alcohols ranged from C16 to C30. Waxes with unsaturated fatty
acids (C40:1 and C42:1) were also found in lower amounts, the unsaturated fatty
acid being in all cases oleic acid.

161
5. Resultados y discusin

Table 3. Composition of the Different Waxes (mg/kg) Identified in Curaua Fibers.


Wax Fatty acid:fatty alcohol Abundance

wax C36 3.5


C12:C24 0.2
C14:C22 1.1
C16:C20 2.0
C18:C18 0.2

wax C37 3.3


C16:C21 0.9
C15:C22 2.4

wax C38 56.8


C16:C22 56.8

wax C39 5.9


C15:C24 0.9
C16:C23 2.6
C17:C22 2.2
C18:C21 0.2
C19:C20 <0.1

wax C40 46.5


C16:C24 25.3
C18:C22 15.5
C20:C20 5.7

wax C40:1 2.9


C18:1:C22 2.9

wax C41 2.3


C15:C26 0.1
C16:C25 0.6
C17:C24 0.6
C18:C23 0.4
C19:C22 0.4
C20:C21 0.2
C21:C20 <0.1
C22:C19 <0.1
C23:C18 <0.1
C24:C16 <0.1

wax C42 32.2


C16:C26 3.4
C18:C24 4.0
C20:C22 22.1
C22:C20 2.7

wax C42:1 1.2

162
5. Resultados y discusin

C18:1:C24 1.2

wax C43 1.5


C15:C28 <0.1
C16:C27 <0.1
C17:C26 0.1
C18:C25 0.1
C19:C24 0.1
C20:C23 0.2
C21:C22 0.7
C22:C21 0.2
C23:C20 0.1
C24:C19 <0.1
C25:C18 <0.1

wax C44 19.1


C16:C28 0.9
C18:C26 0.5
C20:C24 1.7
C22:C22 16.0

wax C46 5.3


C16:C30 0.2
C18:C28 0.2
C20:C26 0.2
C22:C24 1.8
C24:C22 2.9
tr: traces.

Other waxes, consisting of a complex mixture of different long-chain Z-


hydroxy fatty acids esterified to different long-chain fatty alcohols, were also
found in high amounts. These waxes are similar to those described among the
waxes normally secreted by bees [46-48]. The mass spectrum of the TMSi ether
derivative of a selected Z-hydroxy monoester (C38) is shown in Figure 4. The
mass spectrum of this compound is characterized by a base peak at m/z 637
corresponding to the [M-15]+ fragment ion and a fragment formed by the loss of
the fatty alcohol at m/z 311. The elimination of trimethylsilanol (TMSOH) from
the molecular ions also can be observed at m/z 563 [49, 50]. As occurred with
the esters of fatty acids with fatty alcohols, each chromatographic peak is
composed of a complex mixture of compounds. Quantification of individual
compounds was performed by integrating the chromatographic profiles of the
characteristic ions. The detailed structural composition of the Z-hydroxy
monoesters is shown in Table 4. The esterified Z-hydroxy fatty acids ranges
from C16 to C22 and the esterified fatty alcohols ranged from C20 to C26.

163
5. Resultados y discusin

Table 4. Composition and abundance (mg/kg) of the different Z-hydroxy monoesters


identified in curaua fibers.
Z-Hydroxy monoester Z-hydroxy fatty acid: fatty alcohol abundance

Z-hydroxy monoester C36 29.1


Z-OHC16:C20 29.1

Z-hydroxy monoester C37 3.0


Z-OHC16:C21 3.0

Z-hydroxy monoester C38 229.8


Z-OHC16:C22 223.9
Z-OHC18:C20 5.9

Z-hydroxy monoester C40 95.6


Z-OHC16:C24 59.5
Z-OHC18:C22 35.2
Z-OHC20:C20 0.9

Z-hydroxy monoester C42 11.7


Z-OHC16:C26 4.8
Z-OHC18:C24 0.7
Z-OHC20:C22 5.3
Z-OHC22:C20 0.9

Z-hydroxy monoester C44 0.3


Z-OHC20:C24 0.1
Z-OHC12:C22 0.2

Z-Hydroxy fatty acids esterified to glycerol were also found in high amounts
in the curaua fiber. The mass spectra of the TMSi derivatives of Z-hydroxy
acylesters of glycerol are characterized by the presence of an abundant fragment
arising from the loss of a methyl group at [M-15]+. The cleavage between the C-
2 and C-3 carbons in the glyceryl moiety gives rise to the fragments at m/z 103
and [M-103]+. Other diagnostic ions are derived from the glyceryl moiety- i.e. at
m/z 205 as a result of the cleavage between the C-2 and C-1 (the esterified
carbon), and at m/z 219 due to the loss of the acyloxy moiety. The same loss of
the acyloxy group from M+ and M-15+, but with the H rearrangement, gives rise
to the ions at m/z 218 and 203, respectively. Other significant ions in the low-
mass region occur at m/z 73 (the TMSi group), m/z 129 (the glycerol carbon
backbone with a TMSi group [H2C=CHCH=O+Si(CH3)3]) and m/z 147
(produced by the rearrangement of two TMSi groups) [51]. The Z-hydroxy fatty
acids esterified to the glycerol ranges from C22 to C28. The structure and mass
spectrum of the TMSi derivative of 1-mono-(22-hydroxydocosanoyl)glycerol is
shown in Figure 5.

164
5. Resultados y discusin

OH
A

OH
B
O
HO
OH
C
O

OH
D OH

O
HO
O
E

O-CH2

HO-O-CH
F
HO-O-CH2

O
HO
O-CH2

HO-O-CH
G
HO-O-CH2

HO HO HO HO

H I J K

CH2OH CH2OH
O O
OH O OH O
HO HO
OH L OH M

Figure 3. Structures of the main lipids present in the curaua fibers. A: stearic acid, B: n-
docosanol, C: 26-hydroxyhexacosanoic acid, D: 2-hydroxytetracosanoic acid, E: docosanyl,
16-hydroxyhexadecanoate, F: 1-monodocosanoylglycerol, G: 1-mono(24-
hydroxytetracosanoyl)glycerol, H: campesterol, I: ergostanol, J: sitosterol, K: stigmastanol, L:
campesteryl 3E-D-glucopyranoside, M: sitosteryl 3E-D-glucopyranoside.

165
5. Resultados y discusin

 [M-15]+
[M-15-C22H45OH] +
637
100% 311

[C15H 31COO-H 2O]+


[M-C 22H 45OH-H] +
237
55 327
73
129 [M-TMSOH]+
343 563

100 200 300 400 500 600 m/z

Figure 4. Mass spectrum of trimethylsilylated hydroxy monoester C38.

n-Fatty alcohols ranging from C20 to C28 were present in the curaua extracts
with the presence of only the even carbon atom homologs, docosanol (C22) and
tetracosanol (C24) being the most abundant. Monoglycerides, accounting for
690.7 mg/Kg of the fibers, were present in important amounts, from C14 to C30,
C22 (1-monodocosanoylglycerol) being the most prominent. Di- and
triglycerides were only identified in trace amounts.
Sterols were also present among the lipids of curaua fibers in high amounts.
Sitosterol was the most abundant among the free sterols with the presence of
minor amounts of stigmastanol, ergostanol and campesterol. Lower amounts of
sitosterol and campesterol could also be found in ester form. Sterol glycosides,
such as sitosteryl and campesteryl 3E-D-glucopyranosides were also identified
in high amounts, the former being the most predominant. The identification of
steryl glycosides was accomplished, after BSTFA derivatization of the lipid
extract, by comparison with the mass spectra and relative retention times of
authentic standards [52]. Finally, other compounds identified among the curaua
fiber extractives were D-tocopherol, several steroid hydrocarbons and steroid
ketones, as reflected in Table 2.
In conclusion, curaua fiber is characterized by a high content of holocellulose
and D-cellulose and low lignin content which would make this fiber suitable for
papermaking. Moreover, the lignin composition indicates a slight predominance
of S-lignin units (S/G molar ratio of 1.4). On the other hand, the high extractive
content can be considered as a detrimental aspect, however most of the acetone
extracts are due to polar compounds and only 1.3% corresponds to lipophilic
compounds. Indeed, most of the lipophilic compounds are easily saponifiable,
and therefore can be hydrolyzed and dissolved during alkaline cooking.

166
5. Resultados y discusin

 73
OTMS

100%
OTMS

O O

TMS O

129

103 147
203
[M-15]+
321 631
[M-145]+ [M-88]+
411 486
381 543
265

100 200 300 400 500 600 m/z


Figure 5. Mass spectrum and structure of the TMSi ether derivative of 1-mono(22-
hydroxydocosanoyl)glycerol.

Acknowledgements
This study has been supported by the Spanish MEC (project AGL2005-01748).
We thank CELESA (Tortosa, Spain) for providing the curaua fibers.

Literature cited
[1] Fujihashi, G. A.; Barbosa, W. L. R. Ananas erectifolius (curau):
padronizao dos extractos, fraes e do material vegetal. Revista Cientfica da
UFPA. Vol 3, 2002 .
[2] Silva, G. S.; Assis, M. B.; Barbosa, W. L. R. Investigao fitoqumica e
microbiologica da espcie Ananas erectifolius (curau). Revista Virtual de
Iniciao Acadmica da UFPA. Vol 1, 2001.
[3] Hoareau, W.; Trindade, W. G.; Siegmund B.; Castellan, A.; Frollini E. Sugar
cane bagasse and curaua lignins oxidized by chlorine dioxide and reacted with
furfuryl alcohol: characterization and stability. Polym. Degrad. Stabil. 2004, 86,
567-576.
[4] Kelley, S. S.; Rowell, R. M.; Davis, M.; Jurich, Ch. K.; Ibach R. Rapid
analysis of the chemical composition of agricultural fibers using near infrared
spectroscopy and pyrolysis molecular beam mass spectrometry. Biomass
Bioenergy. 2004, 27, 77-88.
[5] Leao, A.L., Rowell, R., and Tavares, N., 1998. Applications of natural fibres
in automotive industry in Brazil - Thermoforming process. In: (P.N. Prasad, J.E.

167
5. Resultados y discusin

Mark, S.H. Kandil, and Z.H. Kafafi, eds) Science and Technology of Polymers
Advanced Materials. Emerging technologies and business opportunities. Plenum
press, New York, pp 755-761.
[6] Gonzlez-Vila, F. J.; Almendros, G.; del Ro, J. C; Martn, F.; Gutirrez, A.;
Romero, J. Ease of delignification assessment of wood from different
Eucalyptus species by pyrolysis (TMAH)-GC/MS and CP/MAS 13C NMR
spectrometry. J. Anal. Appl. Pyrol. 1999, 49, 295-305.
[7] del Ro, J. C.; Gutirrez, A.; Hernando, M.; Landn, P.; Romero, J.;
Martnez, A. T. Determining the influence of eucalypt lignin composition in
paper pulp yield using Py-GC/MS. J. Anal. Appl. Pyrol. 2005, 74, 110-115.
[8] Nimz, H. Beech lignin- Proposal of a constitutional scheme. Angew. Chem.
Int. Edit. 1974, 13, 313-321.
[9] Adler, E. Lignin chemistry - Past, present and future. Wood Sci. Technol.
1977, 11, 169-218.
[10] Tsutsumi, Y.; Kondo, R.; Sakai, K.; Imamura, H. The difference of
reactivity between syringyl lignin and guaiacyl lignin in alkaline systems.
Holzforschung 1995, 49, 423-428.
[11] Hillis, W. E.; Sumimoto, M. Effect of extractives on pulping. In Natural
Products of Woody Plants II; Rowe, J. W., Ed.; Springer-Verlag, Berlin,
Germany, 1989; pp 880-920.
[12] Back, E. L.; Allen, L. H. Pitch Control, Wood Resin and Deresination;
Tappi Press: Atlanta, GA, 2000.
[13] Gutirrez, A.; del Ro, J. C.; Martnez, M. J.; Martnez, A. T. The
biotechnological control of pitch in paper pulp manufacturing. Trends
Biotechnol. 2001, 19, 340-348.
[14] Ali, M.; Sreekrishnan, T. R. Aquatic toxicity from pulp and paper mill
effluents: A review. Adv. Environ. Res. 2001, 5, 175-196.
[15] Rigol, A.; La Torre, A.; Lacorte, S.; Barcel, D. Bioluminiscence
inhibition assays for toxicity screening of wood extractives and biocides in
paper mill process waters. Environ. Toxicol. Chem. 2003, 23, 339-347.
[16] Faix, O.; Meier, D.; Fortmann, I. Thermal degradation products of wood. A
collection of electron of electron-impact (EI) mass spectra of monomeric lignin
derived products. Holz Roh- Werkst. 1990, 48, 351-354.
[17] Ralph, J.; Hatfield, R. D. Pyrolysis-GC/MS characterization of forage
materials. J. Agric. Food Chem. 1991, 39, 1426-1437.

168
5. Resultados y discusin

[18] Gutirrez, A.; del Ro, J. C.; Gonzlez-Vila, F. J.; Martn, F. Analysis of
lipophilic extractives from wood and pitch deposits by solid-phase extraction
and gas chromatography. J. Chromatogr. A. 1998, 823, 449-455.
[19] Browning, B.L. Methods of Wood Chemistry, vol. II; Wiley-Interscience
Publishers, New York, USA, 1967.
[20] Technical Association of the Pulp and Paper Industry. Test methods, 1992-
1993. TAPPI, Atlanta, GA. 1993.
[21] Gutirrez, A.; Rodrguez, I. M.; del Ro, 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.
[22] van Dam, J. E. G.; van Vilsteren, G. E. T.; Zomers, F. H. A.; Shannon, W.
B.; Hamilton, I. T. Increased application of domestically produced plant fibers
in textiles, pulp and paper production and composite materials. Directorate-
General XII, Science, Research and Development, European Commision:
Brussels, 1994.
[23] del Ro, J. C.; Gutirrez, A. Chemical composition of abaca (Musa textilis)
leaf fibers used for manufacturing of high quality paper pulps. J. Agric. Food
Chem. 2006, 54, 4600-4610.
[24] Gutirrez. A.; del Ro, J. C. Lipids from flax fibers and their fate in alkaline
pulping. J. Agric. Food Chem. 2003, 51, 4965-4971.
[25] Gutirrez. A.; del Ro, J. C. Lipids from flax fibers and their fate in alkaline
pulping. J. Agric. Food Chem. 2003, 51, 6911-6914.
[26] Gutirrez, A.; Rodrguez, I. M.; del Ro, 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.
[27] Moore, G. Nonwood Fibre Applications in Papermaking; Pira International,
Leatherhead, Surrey, UK, 1996
[28] del Ro, J. C.; Martn, F.; Gonzlez-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.
[29] Lam, T. B. T.; Kadoya, K.; Liyama, K. Bonding of hydroxycinamic acids
to lignin: ferulic and p-coumaric acids are predominantly linked at the benzyl
position of lignin, not the E-position, in grass cell walls. Phytochemistry 2001,
57, 987-992.

169
5. Resultados y discusin

[30] 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,
5122-5129.
[31] 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.
[32] Scalbert, A.; Monties, B.; Lallemand, J. Y.; Guittet, R.; Rolando, C. Ether
linkage between phenolic acids and lignin fractions from wheat straw.
Phytochemistry 1985, 24, 1359-1362.
[33] 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.
[34] Lam, T. B. T.; liyama, K.; Stone, B. A. Cinnamic acids bridges between
cell wall polymers in wheat and Phalaris internodes. Phytochemistry 1992, 31,
1179-1183.
[35] del Ro, 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.
[36] Martn, F.; del Ro, J. C.; Gonzlez-Vila, F. J.; Verdejo, T. Thermally
assisted hydrolysis and alkylation of lignins in the presence of tetra-
alkylammonium hydroxides. J. Anal. Appl. Pyrol. 1995, 35, 1-13.
[37] Challinor, J. M. Characterization of wood by pyrolysis derivatization-gas
chromatography/ mass spectrometry. J. Anal. Appl. Pyrol. 1995, 35, 93-107.
[38] Clifford, D. J.; Carson, D. M.; Mackinney. D. E.; Bortiatynski, J. M.;
Hatcher, P. G. A new rapid technique for the characterization of lignin in
vascular plants: thermochemolysis with tetramethylammonium hydroxide
(TMAH). Org. Geochem. 1995, 23, 169-175.
[39] 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. Chem. Soc. 1994, 34, 1-12.
[40] Crestini, C.; Argyropoulos, D.S. Structural analysis of wheat straw lignin
by quantitative 31P and 2D NMR spectroscopy. The occurrence of ester bonds
and D-O-4 substructures. J. Agric. Food Chem. 1997, 45, 1212-1219.

170
5. Resultados y discusin

[41] Lu, F.C.; Ralph, J. Detection and determination of p-coumaroylated units in


lignins. J. Agric. Food Chem. 1999, 47, 1988-1992.
[42] Grabber, J.H.; Quideau, S.; Ralph, J. p-Coumaroylated syringyl units in
maize lignin: Implications for E-ether cleavage by thioacidolysis.
Phytochemistry 1996, 43, 1189-1194.
[43] Sharkey, Jr. A. G.; Shultz, J. L.; Friedel, R. A. Mass spectra of esters.
Formation of rearrangement ions. Anal. Chem. 1959, 31, 87-94.
[44] 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, 952, 193-204.
[45] Reiter, B.; Lechner, M.; Lorbeer, E.; Aichholz, R. Isolation ad
characterization of wax esters in fennel and caraway seed oils by SPE-GC. J.
High Res. Chromatogr. 1999, 22, 514-520.
[46] Tulloch, A. P. The composition of beeswax and other waxes secreted by
insects. Lipids. 1969, 5, 247-258.
[47] Tulloch, A. P. Beeswax: structure of the esters and their component
hydroxy acids and diols. Chem. Phys. Lipids. 1971, 6, 235-265.
[48] Aichholz, R.; Lorbeer, E. Investigation of combwax of honeybees with
high-temperature gas chromatography and high-temperature gas
chromatography-chemical ionization mass spectrometry: I. High-temperature
gas chromatography. J. Chromatogr. A. 1999, 855, 601-615.
[49] Aichholz, R.; Lorbeer, E. Investigation of combwax of honeybees with
high-temperature gas chromatography and high-temperature gas
chromatography-chemical ionization mass spectrometry: II. High-temperature
gas chromatography-chemical ionization mass spectrometry. J. Chromatogr. A.
2000, 883, 75-88.
[50] Kimpe, K.; Jacobs, P. A.; Waelkens, M. Mass spectrometric methods prove
the use of beeswax and ruminant fat in late Roman cooking pots. J. Chromatogr.
A. 2002, 968, 151-160.
[51] Graa, J.; Schreiber, L.; Rodrigues, J.; Pereira, H. Glycerol and glyceryl
esters of Z-hydroxyacids in cutins. Phytochemistry. 2002, 61, 205-215.
[52] Gutirrez, A.; del Ro, 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.

171
5. Resultados y discusin

Publicacin V:

Coelho D., Marques G., Gutirrez A., Silvestre A.R.D. and del Ro J.C. (2007)
Chemical characterization of the lipophilic fraction of Giant reed (Arundo donax)
fibers used for pulp and paper manufacturing. Industrial Crops and Products, 26,
229-236.

172
5. Resultados y discusin

Chemical characterization of the lipophilic fraction of Giant reed (Arundo


donax) fibres used for pulp and paper manufacturing
Dora Coelho1,2, Gisela Marques1, Ana Gutirrez1, Armando J.D. Silvestre2, Jos C. del Ro1
1
Instituto de Recursos Naturales y Agrobiologa, Consejo Superior de Investigaciones
Cientficas, P.O. Box 1052, 41080-Seville, Spain
2
Departamento de Qumica, Universidade de Aveiro, 3810-193 Aveiro, Portugal

Abstract
The chemical composition of lipophilic extractives from Arundo donax stems
(including nodes and internodes), used for pulp and papermaking, was studied.
The lipid fraction was extracted with acetone and redissolved in chloroform, and
then fractionated by solid-phase extraction (SPE) on aminopropyl-phase
cartridges into four different fractions of increasing polarity. The total lipid
extract and the resulting fractions were analysed by gas chromatography and gas
chromatography-mass spectrometry, using short- and medium-length high-
temperature capillary columns, respectively. The main compounds identified in
the fibres included series of long-chain n-fatty acids, n-alkanes, n-aldehydes, n-
alcohols, monoglycerides, free and esterified sterols and triterpenols, steryl
glucosides, steroid hydrocarbons and steroid and triterpenoid ketones. Minor
amounts of other compounds such as diglycerides, waxes and tocopherols were
also identified among the lipids of A. donax.
Keywords: Arundo donax, lipophilic extractives, pitch, fatty acids, sterols, steryl
glucosides, GC, GC/MS.

1. Introduction
In the last decades, fast growing plants have received particular attention as
alternative sources of cellulose fibres (van Dam et al. 1994; Moore, 1996).
These non-wood plants are the common fibre source for paper pulp production
in developing countries where wood fibres are not available. In the developed
world, although wood is still by far the main raw material for pulp and paper
manufacture, a market exists for high-value-added papers from these fibres.
Arundo donax L. (giant reed) is a widely distributed naturally growing perennial
rhizomatous grass with a segmented tubular structure like bamboo (Seca et al.,
2000), which has been considered as one of the promising non-wood plants for
pulp and paper industry (Shatalov and Pereira, 2002). The easy adaptability to
different ecological conditions, the annual harvesting period and the high
biomass productivity (32-37 t per year-1ha-1 of dry biomass) reached by intensive
cultivation (Vecchiet et al., 1996), combined with appropriate chemical
composition (Shatalov et al., 2001), make A. donax very attractive as an
alternative source of fibres (Shatalov and Pereira, 2005).

173
5. Resultados y discusin

To improve the utilisation of A. donax fibres, it is necessary to broaden the


knowledge of structural features of its components. Previous chemical research
on A. donax includes chemical composition, general features of macromolecular
components (Pascoal Neto et al., 1997) and structures of isolated hemicelluloses
(Driss et al., 1973, Joseleau and Barnoud, 1974, 1975, 1976). A few studies on
the lignin composition (Joseleau and Barnoud, 1976, Joseleau et al., 1976, Faix
et al., 1989) showed that it is composed of guaiacyl- and syringyl-propane units
with minor amounts of p-hydroxyphenylpropane units (Faix et al., 1989) and
associated with phenolic acids (Tai et al., 1987). However, until now no studies
about the composition of A. donax lipophilic fraction have been performed.
The amount and composition of lipophilic extractives is an important
parameter in wood processing for pulp and paper production and it is dependent
on factors such as the plant species, age, and growth location. The different lipid
classes have different chemical behaviour during pulping and bleaching
(Gutirrez and del Ro, 2003; Freire et al., 2005). The lipophilic extractives are
also responsible for the formation of sticky deposits on the machinery, giving
rise to dark spots in bleached pulp and paper, the so-called pitch, both with
negative economic impact on pulp and paper industry (del Ro et al., 1998,
2000; Gutirrez et al., 2004; Gutirrez and del Ro, 2005; Silvestre et al., 1999).
The accumulation of lipophilic compounds leads also to higher chemical
consumption during pulping and bleaching and therefore increasing production
costs. On the other hand, extractives or their derivatives, might contribute to the
toxicity of paper pulp effluents and products (McCubbin and Folke 1995; Rigol
et al., 2003). The detailed identification of such lipophilic components is
therefore an important step in the study of the behaviour and fate of extractives
during pulp and paper production and consequently in the search for new
solutions to control pitch deposition as well as to decrease effluent toxicity.
In the present paper, the chemical composition of the lipophilic extractives
from A. donax fibres was studied. Gas chromatography (GC) and GC-mass
spectrometry (GC/MS) using, respectively, short- and medium-length high-
temperature capillary columns with thin films, that enable elution and separation
of high-molecular-mass lipids such as waxes, steryl esters and triglycerides, are
employed. For a more detailed characterization of the different homologous
series and other minor compounds, the extract was fractionated by a simple
solid-phase extraction (SPE) method using aminopropyl phase cartridges, as
described previously (Gutirrez et al., 1998, 2004).

2. Experimental
2.1. Samples
Samples of A. donax L. reed stems (including nodes and internodes) were
supplied by University of Huelva, Spain. The samples were air-dried and milled
using a knife mill (Janke and Kunkel, Analysenmhle). For the isolation of

174
5. Resultados y discusin

lipids, the milled samples were Soxhlet extracted with acetone for 8h. The
lipophilic extractives were obtained by redissolving the dried acetone extract in
chloroform and evaporated to dryness under nitrogen.

2.2. Solid Phase Extraction (SPE) fractionation


The chloroform extracts (5-20 mg) were fractionated by a SPE procedure in
aminopropyl phase cartridges (500 mg) from Waters (Dvision of Millipore,
Milford, MA, USA), as already described (Gutirrez et al., 1998, 2004). Briefly,
the dried extract was taken up in a minimal volume (<0.5 mL) of
hexane:chloroform (4:1) and loaded into the cartridge column previously
conditioned with hexane (4 mL). The cartridge was loaded and eluted by
gravity. The column was first eluted with 8 mL of hexane and subsequently with
6 mL of hexane:chloroform (5:1), then with 10 mL of chloroform and finally
with 10 mL of diethyl ether:acetic acid (98:2). Each isolated fraction was dried
under nitrogen.

2.3. GC and GC/MS analyses


For identification and quantification, the total extracts and the SPE fractions
were analysed by GC and GC/MS. For GC analysis, a Hewlett-Packard HP 5890
gas chromatograph equipped with split-splitless injector and a flame ionization
detector (FID) system was used (Hewlett-Packard, Hoofddorp, Netherlands).
The injector and detector temperatures were set at 300 and 350C, respectively.
Duplicate samples (1 L) were injected in the splitless mode. Helium was used
as the carrier gas. The capillary column used was a 5m 0.25 mm i.d., 0.1 m
film thickness, high-temperature, polyimide-coated fused silica tubing DB-5HT
from J&W Scientific (Folsom, CA), especially processed for use at 400C. The
oven was temperature programmed from 100C (1 min) to 350C (3 min) at
15C/min. Peaks were quantified by area and a mixture of standards
(tetracosane, hexadecanoic acid, sitosterol, cholesteryl oleate and
triheptadecanoin) was used for quantitation. The data from the two replicates
was averaged.
The 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 m0.25 mm i.d., 0.1 m film thickness; J&W). Helium was used as carrier
gas at a rate of 2 ml/min. The oven was heated from 120C (1 min) to 380C (5
min) at 10C/min. The temperature of the injector during the injection was
120C, and 0.1 min after injection was programmed to 380C at a rate of
200C/min and held for 10 min. The temperature of the transfer line was set at
300C. Bis(trimethylsilyl)trifluoroacetamide (BSTFA) silylation was used when
required. Compounds were identified by comparing their mass spectra with

175
5. Resultados y discusin

mass spectra in Wiley and NIST libraries, by mass fragmentography, and, when
possible, by comparison with authentic standards.

3. Results and discussion


The total acetone extract from A. donax fibres accounted for 1.56% of total fibre
weight. The lipophilic chloroform soluble compounds represented 0.62%,
while the remaining 0.94% corresponded to polar compounds non-soluble in
chloroform. The lipid extracts were analyzed by GC and GC/MS according to
the method previously described (Gutirrez et al. 1998, 2004). The GC/MS
chromatogram of the A. donax fibres extract, as trimethylsilyl (TMS)
derivatives, is shown in Figure 1. For a better characterization of the
compounds present in the lipid extracts, these were subsequently fractionated by
SPE in aminopropyl-phase cartridges into four major fractions of increasing
polarity. The chromatograms of the different SPE fractions are shown in Figure
2. The first fraction (A), eluted with hexane, was enriched in steryl esters, waxes
and hydrocarbons. The second fraction (B), eluted with hexane:chloroform
(5:1), contained steroid ketones. The third fraction (C), eluted with chloroform,
contained sterols, fatty alcohols and mono- and diglycerides. A final fraction (D)

Sitosterol

FA28

FA16 Steryl glycosides

SG

Campesterol
FA30
FA18:1
Stigmasterol
FA18:2

CG
FA26
StG
FA24
FA17 FA18 MG26
MG16 MG18 Steryl/triterpenyl esters
FA20

5 10 15 20 30
Retention time (min)
Figure 1. GC/MS chromatogram of the derivatized (TMS) chloroform extract of Arundo
donax fibres. FA: fatty acids; MG: monoglycerides; CG: campesteryl 3E-D-glucopyranoside;
StG: stigmasteryl 3E-D-glucopyranoside; SG: sitosteryl 3E-D-glucopyranoside.

176
5. Resultados y discusin

Ak 29 A
Steryl esters
Ak 27

Ak 25 Ak 31

5 10 15 20 25
Cycloartenone
B

-amyrenone

-amyrenone

5 10 15 20 25
-Sitosterol + Stigmastanol + -amyrin
C

Campesterol
Stigmasterol -amyrin

7-oxositosterol

5 10 15 20 25

FA28 D

FA26
FA16
FA18:1
+
FA18:2 FA24
FA18

5 10 15 20 25

Figure 2. GC/MS chromatograms of the different SPE fractions isolated from the A. donax
fibres extracts. Fraction A, eluted with 8 mL of hexane; fraction B, eluted with 6mL of
hexane:chloroform (5:1); fraction C, eluted with 10 mL of chloroform; and fraction D, eluted
with 10 mL diethyl ether:acetic acid (98:2). FA: fatty acids; AK: n-alkanes.

177
5. Resultados y discusin

Table 1. Chemical composition of lipophilic extractives in Arundo donax reed (mg/Kg of


fibre). Each value is the average of two extractions with variation coefficients within 0.1-
4.5% .
Compound Mass Fragments MW Amount
n-Alkanes 77.9
n-docosane 57/71/85/310 310 0.5
n-tricosane 57/71/85/324 324 0.2
n-tetracosane 57/71/85/338 338 0.6
n-pentacosane 57/71/85/352 352 6.3
n-hexacosane 57/71/85/366 366 3.9
n-heptacosane 57/71/85/380 380 15.8
n-octacosane 57/71/85/394 394 6.7
n-nonacosane 57/71/85/408 408 37.0
n-triacontane 57/71/85/422 422 0.8
n-hentriacontane 57/71/85/436 436 5.4
n-dotriacontane 57/71/85/450 450 0.3
n-tritriacontane 57/71/85/464 464 0.4

Steroid hydrocarbons 127.4


ergostatriene 135/143/380 380 14.5
ergostadiene 81/147/367/382 382 9.3
estigmastadiene 81/147/381/396 396 8.4
estigmasta-3,5,22-triene 135/143/394 394 49.2
estigmasta-3,5-diene 81/147/381/396 396 46.0

Fatty acids 1137.7


n-tetradecanoic acid 73/117/132/145/285/300 * 300* 3.5
n-pentadecanoic acid 73/117/132/145/299/314 * 314* 1.8
n-hexadecanoic acid 60/73/129/256 256 276.3
n-heptadecanoic acid 73/117/132/145/327/342 * 342 * 10.0
9,12-octadecadienoic acid 67/81/280 280 30.0
9-octadecanoic acid 55/69/264 282 55.7
n-octadecanoic acid 60/73/129/284 284 73.6
n-nonadecanoic acid 73/117/132/145/355/370 370* 3.1
n-eicosanoic acid 60/73/129/312 312 50.0
n-heneicosanoic acid 55/69/129/326 326 3.3
n-docosanoic acid 60/73/129/340 340 35.7
n-tricosanoic acid 60/73/129/354 354 25.3
n-tetracosanoic acid 60/73/129/368 368 55.7
n-pentacosanoic acid 60/73/129/382 382 33.5
n-hexacosanoic acid 73/117/132/145/453/468 * 468 * 144.1
n-heptacosanoic acid 73/117/132/145/467/482 482* 14.3
n-octacosanoic acid 73/117/132/145/482/497 * 497 * 134.9
n-nonacosanoic acid 73/117/132/145/495/510 * 510 * 53.9
n-triacontanoic acid 73/117/132/145/509/525 * 525 * 109.9
n-hentriacontanoic acid 73/117/132/145/523/538 538* 6.2
n-dotriacontanoic acid 73/132/145/117/537/552 * 552 * 16.9

Fatty alcohols 194.3


n-hexacosanol 75/103/439* 454* 33.4

178
5. Resultados y discusin

n-octacosanol 75/103/467* 482* 54.9


n-triacontanol 75/103/495* 510* 57.7
n-dotriacontanol 75/103/523* 538* 48.3

Aldehydes 81.6
n-hexacosanal 82/96/362 380 10.4
n-octacosanal 82/96/390 408 22.9
n-triacontanal 82/96/418 436 48.3

Sterols/Triterpenols 528.1
campesterol 55/145/213/382/400 400 90.6
stigmasterol 55/81/255/394/412 412 46.4
sitosterol 145/213/396/414 414 281.0
stigmastanol 215/416 416 71.9
7-oxo-sitosterol 135/161/187/396/428 428 6.5
-amyrin 189/203/218/409/426 426 8.2
-amyrin 189/203/218/409/426 426 23.5

Tocopherol 17.7
-tocopherol 151/416 416 6.8
-tocopherol 165/430 430 10.9

Triterpenoid and steroid ketones 43.9


-amyrenone 189/203/218/409/424 424 10.2
-amyrenone 189/203/218/409/424 424 5.9
cycloartenone 189/205/313/409/424 424 14.2
stigmasta-3,5-dien-7-one 174/269/410 410 3.2
stigmast-4-en-3-one 124/229/412 412 4.6
stigmast-4-en-3,6-dione 137/398/408/411/426 426 3.6
stigmastane-3,6-dione 245/287/428 428 2.5

Steryl /triterpenyl esters 68.1


sitosteryl ester 147/381/397 - 16.1
-amyrinyl ester 189/203/218 - 14.0
-amyrinyl ester 189/203/218 - 38.0

Steryl glucosides 151.6


campesteryl 3--D-glucopyranoside 204/217/361/383 * 850 * 30.6
stigmasteryl 3--D-glucopyranoside 204/217/361/395 * 864 * 8.0
sitosteryl 3--D-glucopyranoside 204/217/361/397 * 862 * 113.0

Monoglyceride 367.5
2,3-dihydroxypropyl tetradecanoate 73/103/129/147/343/431 * 446 * 5.5
2,3-dihydroxypropyl hexadecanoate 73/103/129/147/371/459 * 474* 94.2
2,3-dihydroxypropyl octadecanoate 73/103/129/147/399/487 * 502 * 86.6
2,3-dihydroxypropyl eicosanoate 73/103/129/147/427/515 * 530 * 35.1
2,3-dihydroxypropyl docosanoate 73/103/129/147/455/543 * 558 * 43.0
2,3-dihydroxypropyl tetracosanoate 73/103/129/147/483/571 * 586 * 46.9
2,3-dihydroxypropyl hexacosanoate 73/103/129/147/511/599 * 614 * 56.2

179
5. Resultados y discusin

Diglycerides 47.6
dipalmitin, 1,2- (P2) 57/129/313/386/625 * 640 * 7.8
dipalmitin, 1,3- (P2) 57/129/314/371/385/625 * 640 * 12.1
palmitoylstearin (PS) 57/129/314/372/399/579 * 668 * 16.8
distearin, 1,2- and 1,3- (S2) 57/129/342/399/607 * 696 * 10.9
* as TMSi ether derivates; bold mass fragments indicate base peaks.

enriched in free fatty acids was eluted with diethyl ether-acetic acid (98:2). The
identities and abundances of the main compounds identified are listed in Table
1. The most predominant lipid classes identified among the A. donax lipid
extracts were series of n-fatty acids (41% of total lipids identified), sterols
(19%), monoglycerides (13%), fatty alcohols (7%) and steryl glucosides (6%).
Minor amounts of alkanes, aldehydes, tocopherols, steroid hydrocarbons, steroid
and triterpenoid ketones and steryl/triterpenyl esters, were also present in these
fibres. The structures of main and representative compounds are shown in
Figure 3.
The series of free fatty acids were identified in A. donax fibres ranging from
tetradecanoic (C14) to dotriacontanoic (C32) acids, with strong even-over-odd
carbon atom predominance. Hexadecanoic acid (palmitic acid, I) was the most
abundant fatty acid, however a bimodal distribution, with a second maximum
for octacosanoic acid (C28) was observed. The unsaturated 9-octadecenoic
(oleic acid, II) and 9,12-octadecadienoic (linoleic acid, III) acids were also
present in important amounts. The series of n-alkanes was also identified in the
A. donax fibre ranging from docosane (C22) to tritriacontane (C33), with a
strong odd-over-even carbon atom number predominance, and nonacosane (IV)
being the most predominant homolog. n-Fatty alcohols ranging from
hexacosanol (C26) to dotriacontanol (C32) were present in the A. donax extracts
with the presence of only the even carbon atom number homologues,
triacontanol (V) being the most abundant. Significant amounts of a series of n-
aldehydes ranging from hexacosanal (C26) to triacontanal (C30) were identified
in the A. donax fibres with triacontanal (VI) predominating. Monoglycerides
were also present in high amounts in A. donax fibres. The series of
monoglycerides was identified in the range from C14 to C26, with maximum for
monopalmitin, C16, (VII).
Steroids and triterpenoids, including free sterols, steryl esters, steryl glucosides,
steroid ketones and hydrocarbons are among the most predominant compounds
in the lipophilic extract of A. donax fibre. Free sterols were the major compound
class among steroids and triterpenoids, sitosterol (VIII) being the main sterol
present. Other sterols, such as campesterol (IX), stigmasterol (X), stigmastanol
(XI) and the oxidized 7-oxositosterol, were also present. Steryl esters were also
present in A. donax extract, although in low amounts. The complete
identification of the individual steryl esters by GC-MS was not possible since
they only show fragments arising from the sterol moiety by electro-impact MS

180
5. Resultados y discusin

Figure 3. Structures of the main lipophilic compounds present in A. donax fibres. (I) palmitic
acid, (II) oleic acid, (III) linoleic acid, (IV) nonacosane, (V) triacontanol, (VI) triacontanal,
(VII) monopalmitin, (VIII) sitosterol, (IX) campesterol, (X) stigmasterol, (XI) stigmastanol,
(XII) sitosteryl 3E-D-glucopyranoside, (XIII) D-amyrin, (XIV) E-amyrin, (XV) stigmasta-3,5-
diene, (XVI) stigmasta-3,5,7-triene, (XVII) E-amyrenone, (XVIII) D-amyrenone, (XIX)
cycloartenone, (XX) stigmasta-3,5-dien-7-one, (XXI) stigmast-4-en-3-one, (XXII) stigmasta-
3,6-dione.

181
5. Resultados y discusin

and rarely give detectable molecular ions (Lusby et al. 1984, Evershed et al.
1989). By monitoring the ions corresponding to the different sterol moieties in
the SPE fraction enriched in steryl esters, it was possible to identify series of
sitosterol as well as D- and E-amyrin esters. Steryl glucosides, such as
campesteryl, stigmasteryl and sitosteryl -D-glucopyranosides (XII), were
identified in significant amounts, the latter being the most predominant.
The identification of steryl glucosides was accomplished (after BSTFA
derivatization of the lipid extract) by comparison with the mass spectra and
relative retention times of authentic standards (Gutirrez and del Ro, 2001).
Among triterpenols, D-amyrin (XIII) and E-amyrin (XIV) occurred in free and
esterified form, with the latest being detected in low amounts. Finally, several
steroid hydrocarbons, such as stigmasta-3,5-diene (XV) and stigmasta-3,5,7-
triene (XVI) and triterpenoid and steroid ketones, such as E-amyrenone (XVII),
D-amyrenone (XVIII), cycloartenone (XIX), stigmasta-3,5-dien-7-one (XX),
stigmast-4-en-3-one (XXI) and stigmasta-3,6-dione (XXII), were also identified.
The different lipid classes present in A. donax fibres will have different
behavior during pulping and bleaching and therefore the problematic of pitch
will be different depending the type of pulping (i.e. mechanical, chemical) and
bleaching (ECF, TCF) processes. The knowledge of the chemical composition
of the lipophilic components of A. donax fibres shown here will assist to predict
pitch problems during pulp and papermaking of this fibre and to establish
appropriate methods for their control.

Acknowledgements
This study has been funded by the Spanish project AGL2005-01748. GM thanks
the Spanish Ministry of Education and Science for a FPI fellowship. We thank
M.J. Diaz (University of Huelva) for the Arundo donax fibres.

References
del Ro, J.C., Gutirrez A., Gonz.lez-Vila F.C., Martn F. and Romero J., 1998.
Characterization of organic deposits produced in kraft pulping of Eucalyptus
globulus wood. J. Chromatogr. A. 823, 457-465.

del Ro, J.C.; Romero, J.; Gutirrez, A., 2000. Analysis of pitch deposits
produced in Kraft pulp mills using a totally chlorine free bleaching sequence.
J. Chromatogr. A 874, 235-245.

Driss, M., Rozmarin, G. and Chene, M., 1973. Some physicochemical properties
of two xylans of reed (Phragmites communis and Arundo donax) in solution.
Cell. Chem. Technol. 7, 703-713.

182
5. Resultados y discusin

Evershed, R.P., M.C. Prescott, N. Spooner and L.J. Goad. 1989. Negative ion
ammonia chemical ionization and electron impact ionization mass
spectrometric analysis of steryl fatty acyl esters. Steroids 53, 285309.

Faix, O., Meier, D., Beinhoff, O., 1989. Analysis of lignocelluloses and lignins
from Arundo donax and Miscanthus sinensis Anderss and hydroliquefaction
of Miscanthus. Biomass 18, 109.

Freire, C.S.R., Silvestre, A.J.D. and Pascoal Neto, C., 2005. Lipophilic
extractives in Eucalyptus globulus Kraft pulps. Behaviour during ECF
bleaching. J. Wood Chem. Technol. 25, 67-80.

Gutirrez, A., del Ro, J.C., Gonzlez-Vila F.J. and Martn F., 1998. Analysis of
lipophilic extractives from wood and pitch deposits by solid-phase extraction
and gas chromatography. J. Chromatogr. A. 823, 449-455.

Gutirrez, A.; del Ro, 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.

Gutirrez, A.; del Ro, J. C., 2003. Lipids from flax fibers and their fate in
alkaline pulping. J. Agric. Food Chem., 51, 4965-4971.

Gutirrez, A., del Ro, J.C. and Martnez, A.T., 2004. Chemical Analysis and
Biological Removal of Wood Lipids forming Pitch deposits in paper pulp
manufacturing. In: F.J.T. Spencer and A.L. Ragout de Spencer (Eds.),
Protocols in Environmental Microbiology. In: Methods in Molecular Biology.
Chapter 19, Humana Press, 2004, pp. 189-202.

Gutirrez, A. and del Ro, J.C., 2005. Chemical characterization of pitch


deposits produced in the manufacturing of high-quality paper pulps from
hemp fibers. Biores. Technol. 96, 1445-1450.

Joseleau, J.P. and Barnoud, F., 1974. Hemicelluloses of young internodes of


Arundo donax. Phytochemistry 13, 1155-1158.

Joseleau, J.P. and Barnoud, F., 1975. Hemicelluloses of Arundo donax at


different stages of maturity. Phytochemistry 14, 71-75.

Joseleau, J.P. and Barnoud, F., 1976. Cell wall carbohydrates and structural
studies of xylan in relation to growth in the Arundo donax. Appl. Polym.
Symp. 28, 983-992.

Joseleau, J.P., Miksche, G.E. and Yasuda, S., 1976. Structural variation of
Arundo donax lignin in relation to growth. Holzforschung 31, 19-20.

183
5. Resultados y discusin

Lusby, W. R., M. J. Thompson and J. Kochansky. 1984. Analysis of sterol esters


by capillary gas chromatography electron impact and chemical ionization-
mass spectrometry. Lipids 19(11), 888901.

McCubbin, N. and J. Folke. 1995. Significance of AOX vs. unchlorinated


organics. Pulp Paper Can. 96, 43-48.

Moore, G., 1996. Nonwood Fibre Applications in Papermaking; Pira


International, Leatherhead, Surrey, UK.

Pascoal Neto, C., Seca, A., Nunes, A.M., Coimbra, M.A., Domingues, F.,
Evtuguin, D., Silvestre, A.J.D, Cavaleiro, J.A.S., 1997. Variations in chemical
composition and structure of macromolecular components in different
morphological regions and maturity stages of Arundo donax. Ind. Crops Prod.
6, 51-58.

Rigol, A.; La Torre, A.; Lacorte, S.; Barcel, D., 2003. Bioluminiscence
inhibition assays for toxicity screening of wood extractives and biocides in
paper mill process waters. Environ. Toxicol. Chem., 23, 339-347.

Seca, A.M., Cavaleiro, J.A.S., Domingues, F.M.J., Silvestre, A.J.D, Evtuguin,


D., Neto, C.P., 2000. Structural characterization of the lignin from the nodes
and internodes of Arundo donax Reed. J.Agric.Food Chem. 48, 817-824.

Shatalov, A.A., Quilh, T., Pereira, H., 2001. Arundo donax L. reed new
perspectives for pulping and bleaching. 2. Raw material characterization.
TAPPI J. 84 (1), 1-12.

Shatalov, A.A., Pereira, H., 2002. Influence of stem morphology on pulp and
paper properties of Arundo donax L. reed. Ind. Crops Prod. 15, 77-83.

Shatalov, A.A., Pereira, H., 2005. Kinetics of organosolv delignification of fibre


crop Arundo donax L. Ind. Crops Prod. 21, 203-210.

Silvestre, A. J.D., C. L.C. Pereira, C. Pascoal Neto, A.C. Duarte, J. A. S.


Cavaleiro and F. P. Furtado. 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(5), 375-
382.

Tai, D., Cho, W. And Ji, W., 1987. Studies of Arundo donax lignins.
Proceedings of the 4th ISWPC, Vol. II, April 1987, Paris, pp. 13-17.

van Dam, J. E. G.; van Vilsteren, G. E. T.; Zomers, F. H. A.; Shannon, W. B.;
Hamilton, I. T., 1994. Industrial fibre crops - study on: increased application
of domestically produced plant fibres in textiles, pulp and paper production

184
5. Resultados y discusin

and composite materials. Directorate-General XII, Science, Research and


Development, European Commission.

Vecchiet, M., Jodice, R., Schenone, G., 1996. Agronomic research on giant reed
(Arundo donax L.). Management system and cultivation of two different
provenances. In: Chartier, Ph., Ferrero, G.L., Henius, U.M., Hultberg, S.,
Sachau, J., Wiinblab, M. (Eds.), Biomass for Energy and the Environment. In:
Proceedings of the Ninth European Biomass Conference, Copenhagen,
Pergamon, UK, pp. 644-648.

185
5. Resultados y discusin

Publicacin VI:

Marques G., Gutirrez A. and del Ro J.C. (2008) Chemical composition of


lignin and lipids from tagasaste (Chamaecytisus proliferus spp. palmensis).
Industrial Crops and Products, 28, 29-36.

186
5. Resultados y discusin

Chemical composition of lignin and lipids from tagasaste (Chamaecytisus


proliferus spp. palmensis)

Gisela Marques, Ana Gutirrez and Jos C. del Ro


Instituto de Recursos Naturales y Agrobiologa de Sevilla, CSIC, P.O. Box 1052, 41080-
Seville, Spain

Abstract
The chemical characterization of trimming residues of tagasaste (Chamaecytisus
proliferus spp. palmensis), a hardy leguminous shrub that has been recently
explored for pulp and paper production, was performed with especial emphasis
in the composition of lignin and lipophilic extractives. Tagasaste was
characterized by a high content of holocellulose (81%) and D-cellulose (41%),
while having a relatively low lignin content of 18.9%. The analysis of lignin was
performed in situ by pyrolysis-gas chromatography/mass spectrometry (Py-
GC/MS) and showed a composition with a guaiacyl:syringyl (G:S) molar
proportion of 38:62 (S/G molar ratio of 1.6) and the absence of p-hydroxyphenyl
(H) units. The high S/G ratio, together with its low lignin content makes
tagasaste an adequate raw material for pulping. On the other hand, the relatively
high acetone extractive content (1.4%) was mostly due to polar compounds and
only 0.2% corresponded to lipophilic compounds. The lipophilic compounds,
analyzed by GC and GC/MS, were mainly composed of fatty acids, including D-
hydroxyfatty acids, and steroid compounds, such as free and conjugated (esters
and glycosides) sterols.
Keywords: Tagasaste; lipids; lignin; pyrolysis; paper pulp.

1. Introduction
Tagasaste (Chamaecytisus proliferus spp. palmensis), also known as tree
lucerne, is a hardy leguminous and fast-growing shrub of the Fabaceae
(Genisteae) family. It is indigenous of the Canary Islands (Spain) but is now
being cultivated in Australia, New Zealand and other countries (Francisco-
Ortega et al., 1991). The shrub is being mainly exploited for high-protein fodder
to maintain livestock (Borens and Poppi, 1990; Ventura et al., 2002) and also as
N-fixing crops to improve soil fertility (Kindu et al., 2006). In order to
encourage the formation of bushes with multiple stems the shrub must be grazed
with regularity, which leads to a high accumulation of trimming residues. These
residues are nowadays considered as agricultural waste since they cannot be
converted to valuable products.
As attempts to reduce the adverse environmental impact and to use this
renewable biomass, they have been recently explored as an alternative raw
material for pulp production (Daz et al. 2004; Lpez et al., 2004; Jimnez et al.,

187
5. Resultados y discusin

2006; Jimnez et al., 2007; Garca et al., 2007). Tagasaste has been found to be
an excellent raw material for paper pulp production, similar to eucalypt wood,
with high hollocellulose and D-cellulose contents, and low lignin and extractives
content, giving high yields (Daz et al. 2004; Lpez et al., 2004; Jimnez et al.,
2006). However, despite all this previous work, a detailed chemical composition
of tagasaste trimming residues has not been addressed so far, which is of high
importance for optimizing the use of this raw material for paper pulp production.
The content and chemical structure of woody components, in particular the
lignin content and its composition in terms of its p-hydroxyphenyl (H), guaiacyl
(G) and syringyl (S) moieties are important parameters in pulp production in
view of delignification rates, chemical consumption and pulp yields. The higher
reactivity of the S lignin with respect to the G lignin in alkaline systems is
known (Chang and Sarkanen 1973; Tsutsumi et al. 1995) and therefore, the
lignin S/G ratio in hardwoods affects the pulping efficiency. 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 (Gonzlez-Vila et
al. 1999; del Ro et al. 2005).
On the other hand, the composition of extractives, especially the lipophilic
compounds, is also important for pulp and paper production. The different
classes of lipids have different behavior during cooking and bleaching. The
lipids can be classified into two principal groups, namely fatty acids and neutral
components, the latter including waxes, long chain n-fatty alcohols, alkanes and
steroids and triterpenoids. And the behavior of the fatty acids in an aqueous
environment is quite different from that of the neutrals. In alkaline pulping, the
acids dissociate and can dissolve in water to quite a high extent, forming fatty
acid soaps. The neutrals, however, have a very low solubility in water and
survive the cooking process and remain in the pulp being at the origin of the so-
called pitch deposits, which are responsible of reduced product quality and
higher operating costs due to production stops for cleaning the equipment (Hillis
and Sumimoto 1989; Back and Allen 2000). The increasing trend in
recirculating water in pulp mills to accomplish environmental demands is
aggravating these problems.
Therefore, the main objective of this work is to perform a thorough chemical
characterization of tagasaste trimming residues, with especial emphasis in the
chemical composition of lignin and lipophilic extractives. In this work, the
lignin composition of tagasaste was characterized in situ using analytical
pyrolysis coupled to gas chromatography/mass spectrometry (Py-GC/MS), a
powerful analytical tool for the rapid analysis of complex polymer mixtures
including lignocellulosic materials (Ralph and Hatfield 1991; Faix et al. 1990;
del Ro et al. 2001; del Ro et al. 2005) that can give information on the lignin
composition in terms of the H, G and S moieties. On the other hand, the
chemical characterization of the lipophilic extractives was performed by GC and
GC/MS by using high-temperature short- and medium-length capillary columns,

188
5. Resultados y discusin

respectively. This method enables the elution and analysis of intact high
molecular weight lipids (Gutirrez et al. 1998). The knowledge of the chemical
composition of the main components of tagasaste trimming residues will be
useful for a better utilization of this agricultural waste.

2. Experimental
2.1. Samples
Tagasaste trimming residues were supplied by University of Huelva, Spain. The
dried samples were milled using a knife mill. For the isolation of lipids, the
milled samples were 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, and all samples were subjected to GC and GC/MS analyses. For
Klason lignin content estimation, the samples extracted with acetone were
subsequently extracted with hot water (3 h at 100 C) to remove the water-
soluble material. Holocellulose was isolated from the pre-extracted fibers by
delignification for 4 hours using the acid chlorite method (Browning, 1967). The
D-cellulose content was determined by removing the hemicelluloses from the
holocellulose by alkali extraction (Browning, 1967). Klason lignin was
estimated as the residue after sulfuric acid hydrolysis of the pre-extracted
material according to Tappi rule T222 om-88 (Tappi, 1993). The acid-soluble
lignin was determined, after filtering off the insoluble lignin, by
spectrophotometric determination at 205 nm wavelength. Ash content was
estimated as the residue after 6 h at 575 C.

2.2. Solid Phase Extraction (SPE) fractionation


For a better characterization of the different homologous series, the lipid extracts
were fractionated by a SPE procedure using aminopropyl-phase cartridges (500
mg) from Waters (Division of Millipore), as already described (Gutirrez et al.
1998). Briefly, the dried chloroform extracts were taken up in a minimal volume
(< 0.5 ml) of hexane:chloroform (4:1) and loaded into the cartridge column
previously conditioned with hexane (4 ml). The cartridge was loaded and eluted
by gravity. The column was first eluted with 8 ml of hexane and subsequently
with 6 ml of hexane:chloroform (5:1), then with 10 ml of chloroform and finally
with 10 ml of diethyl ether:acetic acid (98:2). Each isolated fraction was dried
under nitrogen and analyzed by GC and GC/MS.

2.3. GC and GC/MS Analyses


The GC analyses of the extracts were performed in an Agilent 6890N GC
system using a short-fused silica capillary column (DB-5HT, 5 m 0.25 mm
I.D., 0.1 m film thickness). The temperature program was started at 100 C

189
5. Resultados y discusin

with a 1-min hold and then raised to a final temperature of 350 C at 15 C/min,
and held for 3 min. The injector and flame-ionization detector temperatures were
set at 300 and 350 C, respectively. Helium was used as the carrier gas at a rate
of 5 mL/min, and the injection was performed in splitless mode. Peaks were
quantified by area, and a mixture of standards (octadecane, palmitic acid,
sitosterol and cholesteryl oleate) was used to elaborate calibration curves.
The GC/MS analyses were performed with a Varian model Star 3400 GC
equipped with an ion trap detector (Varian Saturn 2000) using a medium-length
(12 m) capillary column of the same characteristics described above. The oven
was heated from 120 (1 min) to 380 C at 10 C/min and held for 5 min. The
transfer line was kept at 300 C. The injector was temperature programmed from
120 (0.1 min) to 380 C at a rate of 200 C/min and held until the end of the
analysis. Helium was used as the carrier gas at a rate of 2 mL/min.
Trimethylsilyldiazomethane methylation and BSTFA (bis(trimethylsilyl)-
trifluoroacetamide) silylation, in the presence of pyridine, were used to produce
the appropriate derivatives, when required. Compounds were identified by
comparing their mass spectra with mass spectra in Wiley and NIST libraries, by
mass fragmentography, and when possible, by comparison with authentic
standards.

2.4. Py-GC/MS.
The pyrolysis of tagasaste (1 mg) was performed in a micro-furnace pyrolyzer
(model 2020, Frontier Laboratories Ltd) directly connected to a GC/MS system
Agilent 6890 equipped with a fused silica capillary column HP 5MS
(30 m 0.25 mm 0.25 m I.D.). The detector consisted of an Agilent 5973
mass selective detector. The pyrolysis was performed at 500 C. The final
temperature was achieved at a rate of 20 C/min. The GC/MS conditions were as
follows: oven temperature was held at 50 C for 1 min and then increased up to
100 C at 30 C/min, from 100 to 300 C at 10 C/min and isothermal at 300 C
for 10 min using a heating rate of 20 C/min in the scan modus. The carrier gas
used was helium with a controlled flow of 1 ml/min. The compounds were
identified by comparing the mass spectra obtained with those of the Wiley and
NIST computer libraries and that reported in the literature (Faix et al. 1990;
Ralph and Hatfield 1991). Relative peak molar areas were calculated for
carbohydrate and lignin pyrolysis products. The summed molar areas of the
relevant peaks were normalized to 100%, and the data for two repetitive
pyrolysis experiments were averaged.

3. Results and Discussion


Tagasaste was characterized by a high content of holocellulose (81%) and D-
cellulose (41%), and a lignin content of 18.9% (Klason lignin, 16.6%; acid-
soluble lignin: 2.3%). Similar values have been reported by other authors (Daz

190
5. Resultados y discusin

et al., 2004). This low lignin content is similar to that found in eucalypt wood
(Rencoret et al. 2007), a widely raw material for pulp and papermaking, and
together with the high holocellulose and D-cellulose contents makes tagasaste an
interesting raw material for pulp and paper production, as already advanced by
other authors (Daz et al. 2004; Lpez et al., 2004; Jimnez et al., 2006; Jimnez
et al. 2007; Garca et al., 2007). On the other hand, the acetone extractives
accounted for 1.4% but the chloroform-soluble lipids accounted for only 0.2%,
which is a value lower than that found in most lignocellulosic materials (Back
and Allen, 2000). However, while the content of the main organic fractions is an
important parameter in wood processing for pulp and papermaking, the chemical
composition of these organic fractions, in particular the lignin composition in
terms of the relative proportions of the S- and G-units, and the lipid
composition, in terms of the presence of saponifiable or unsaponifiable
components, strongly influences the pulping and bleaching behavior of a raw
material.

3.1. Lignin composition


The lignin composition of tagasaste was characterized in situ by Py-GC/MS.
The pyrogram of tagasate sample is shown in Figure 1 and the identities and
relative molar abundances of the released compounds are listed in Table 1. Py-
GC/MS of tagasaste wood trimming residues released compounds arising from
carbohydrates (67%) and lignin-derived phenols (34%). The lignin-derived
phenols arise from guaiacyl (G) and syringyl (S) lignin units whereas no traces
of p-hydroxyphenyl (H) units were found in tagasaste. The main lignin-derived
compounds identified were 4-vinylsyringol (38), syringol (24), 4-methylsyringol
(29), 4-ethylsyringol (35) and 4-vinylguaiacol (23). Syringaldehyde (41),
syringylacetone (47) and trans-sinapaldehyde (49) were also identified. Relative
peak molar areas were calculated for carbohydrate, and lignin G- and S-type
degradation products. The Py-GC/MS data showed a lignin composition in
tagasaste with a predominance of the S units (S/G molar ratio of 1.6), typical of
hardwoods. The predominance of S-lignin units in tagasaste makes this material
advantageous for delignification due to the higher reactivity of the S-lignin in
alkaline systems (Chang and Sarkanen, 1973; Tsutsumi et al. 1995; Gonzlez-
Vila et al., 1999) and its lower condensation degree. The G units have a free C-5
position available for carbon-carbon inter-unit bonds, which make them fairly
resistant to lignin depolymerization in pulping, while the S lignin is relatively
unbranched and has a lower condensation degree and therefore is easier to
delignify. The high S/G ratio, together with its low lignin content, explains the
excellent performances of tagasaste during alkaline pulping.

191
5. Resultados y discusin

Table 1.- Identification and relative molar abundance of the compounds released by Py-
GC/MS of tagasaste.
No. Compound Origin Mass fragments MW %
1 acetic acid C 45/60 60 39.4
2 (3H) furan-3-one C 55/84 84 0.7
3 1,3-hydroxydihydro-2-(3H)-furanone C 58/102 102 3.1
4 (3H) furan-3-one C 55/84 84 1.0
5 2-furaldehyde C 67/95/96 96 3.8
6 2-(hydroxymethyl)-furan C 43/70/81/98 98 0.5
7 cyclopent-1-ene-3,4-dione C 54/68/96 96 0.6
8 (5H) furan-2-one C 55/84 84 2.9
9 2,3-dihydro-5-methylfuran-2-one C 55/69/98 98 3.1
10 5-methyl-2-furfuraldehyde C 53/109/110 110 0.4
11 4-hydroxy-5.6-dihydro-(2H)-pyran-2-one C 58/85/114 114 2.6
12 3-hydroxy-2-methyl-2-cyclopenten-1-one C 55/85/112 112 0.6
13 2-hydroxy-3-methyl-2-cyclopenten-1-one C 55/85/112 112 0.3
14 dihydroxypyran-1-one C 56/84/114 114 0.8
15 guaiacol LG 81/109/124 124 1.9
16 dimethyldihydropyranone C 55/70/83/111/126 126 0.3
17 4-methylguaiacol LG 95/123/138 138 1.2
18 catechol C 64/81/92/110 110 1.1
19 5-hydroxymethyl-2-tetrahydrofuraldehyde-3-one C 57/69/85/98/99 138 0.4
20 5-hydroxymethyl-2-furaldehyde C 69/97/109/126 126 0.3
21 methoxycatechol L 60/97/125/140 140 0.9
22 4-ethylguaiacol LG 122/137/152 152 0.4
23 4-vinylguaiacol LG 107/135/150 150 2.1
24 syringol LS 111/139/154 154 3.6
25 eugenol LG 131/149/164 164 0.4
26 4-propylguaiacol LG 122/136/166 166 0.1
27 vanillin LG 109/151/152 152 0.7
28 cis-isoeugenol LG 131/149/164 164 0.4
29 4-methylsyringol LS 125/153/168 168 1.8
30 trans-isoeugenol LG 131/149/164 164 1.2
31 homovanillin LG 122/137/166 166 0.2
32 4-propinylguaiacol LG 77/91/119/147/162 162 1.0
33 4-propinylguaiacol isomer LG 77/91/119/147/162 162 1.3
34 acetovanillone LG 123/151/166 166 0.2
35 4-ethylsyringol LS 167/182 182 1.5
36 guaiacylacetone LG 122/137/180 180 0.7
37 levoglucosane C 60/98 162 4.8
38 4-vinylsyringol LS 137/165/180 180 4.2
39 4-allylsyringol LS 167/179/194 194 0.7
40 cis-4propenylsyringol LS 167/179/194 194 0.5
41 syringaldehyde LS 167/181/182 182 1.2

192
5. Resultados y discusin

42 4-propynylsyringol LS 106/131/177/192 192 1.0


43 4-propynylsyringol isomer LS 106/131/177/192 192 0.8
44 trans-4-propenylsyringol LS 167/179/194 194 2.5
45 acetosyringone LS 153/181/196 196 0.6
46 trans-coniferaldehyde LG 107/135/147/178 178 0.8
47 syringylacetone LS 123/167/210 210 0.5
48 propiosyringone LS 151/181/210 210 0.3
49 trans-sinapaldehyde LS 137/165/180/208 208 1.3

%G 38
%S 62
S/G 1.6
C: cellulose; LG: lignin guaiacyl; LS: lignin syringyl

3.2. Lipid composition


The lipid composition in tagasaste was analyzed by gas chromatography (GC)
and gas chromatography/mass spectrometry (GC/MS), using short- and
medium-length high-temperature capillary columns, respectively (Gutirrez et
al., 1998), which enable the elution and analysis of intact high molecular weight
lipids such as waxes, sterol esters, and triglycerides. The GC/MS chromatogram
of the lipids from tagasaste is shown in Figure 2. The identities and abundances

38

24

5
44

3
23
11 15
9
29
8
30
2 13 4142 49
17 35
32 33
36 39 47
6 25 27 40 45 46
10 21
14 16 18 22 28 48

2 4 6 8 10 12 14 16 18 20 22
Retention time (min)
Figure 1. Py-GC/MS of tagasaste wood. The identities of the peaks are shown in Table 2.

193
5. Resultados y discusin

of the main lipid classes identified are summarized in Table 2. The structures of
the main lipophilic compounds identified in the tagasaste extracts are shown in
Figure 3. The main lipids identified in tagasaste were series of fatty acids,
including D-hydroxy acids, and steroid compounds, including steroid
hydrocarbons, steroid ketones, sterols, sterol esters and sterol glycosides. Other
compounds, such as series of alkanes and monoglycerides were also found in
minor amounts.
The series of free fatty acids (141.4 mg/kg) were present in the range from n-
tetradecanoic (C14) to n-hexacosanoic (C26) acids, with a strong even-over-odd
carbon atom predominance. Hexadecanoic (palmitic) acid (I, C16:0) and
octadecanoic (stearic) acid (C18:0) were the most abundant fatty acids followed
by n-tetracosanoic (C24) acid. Unsaturated fatty acids were also present, 9-
octadecenoic (oleic) acid (II, C18:1) being especially abundant. Fatty acids also
included a series of D-hydroxyfatty acids (51.8 mg/kg) that was present in the
range from 2-hydroxyoctadecanoic acid (C18) to 2-hydroxyhexacosanoic acid
(C26) with maximum at C24 (III), and the presence of exclusively the even
carbon atom number homologs.

34+5

steroid 2 steroid
hydrocarbons ketones
FA16
FA18:1
FA18
1
78
6
FA20 FA22 FA24
sterol esters

5 10 15 20
Retention time (min)
Figure 2. GC/MS of the lipophilic extracts from tagasaste wood. Key labels are: Fn: fatty
acids; 1: D-tocopherol; 2: campesterol; 3: stigmasterol; 4: sitosterol; 5: stigmastanol; 6: 7-
oxocampesterol; 7: 7-oxostigmasterol; 8: 7-oxositosterol.

194
5. Resultados y discusin

Table 2.- Composition and abundance (mg/Kg) of lipophilic extractives in tagasaste trimming
residues.
compound mass fragments MW abundance
n-alkanes 8.8
n-eicosane 57/71/85/282 282 0.1
n-heneicosane 57/71/85/296 296 0.2
n-docosane 57/71/85/310 310 0.1
n-tricosane 57/71/85/324 324 0.3
n-tetracosane 57/71/85/338 338 0.2
n-pentacosane 57/71/85/352 352 0.5
n-hexacosane 57/71/85/366 366 0.3
n-heptacosane 57/71/85/380 380 0.5
n-octacosane 57/71/85/394 394 0.3
n-nonacosane 57/71/85/408 408 1.2
n-triacontane 57/71/85/422 422 0.2
n-hentriacontane 57/71/85/436 436 4.8
n-tritriacontane 57/71/85/464 464 0.1
fatty acids 141.4
n-tetradecanoic acid 60/73/228 228 1.1
n-pentadecanoic acid 60/73/242 242 1.0
9-hexadecenoic acid 55/69/236/254 254 2.2
n-hexadecanoic acid 60/73/129/256 256 55.4
n-heptadecanoic acid 60/73/129/270 270 2.9
9-octadecenoic acid 55/69/264 282 29.4
n-octadecanoic acid 60/73/129/284 284 10.7
n-nonadecanoic acid 60/73/129/298 298 1.4
n-eicosanoic acid 60/73/129/312 312 7.4
n-heneicosanoic acid 60/73/129/327 327 1.1
n-docosanoic acid 60/73/129/340 340 5.4
n-tricosanoic acid 60/73/129/354 354 6.0
n-tetracosanoic acid 60/73/129/368 368 10.3
n-pentacosanoic acid 60/73/129/382 382 2.6
n-hexacosanoic acid 60/73/129/396 396 4.5
-hydroxy fatty acids# 51.8
2-hydroxyoctadecanoic acid 73/327/371 386 0.6
2-hydroxyeicosanoic acid 73/355/399 414 0.9
2-hydroxydocosanoic acid 73/383/427 442 22.1
2-hydroxytetracosanoic acid 73/411/455 470 26.3
2-hydroxyhexacosanoic acid 73/439/483 498 1.9
steroid hydrocarbons 30.1
ergostatriene 135/143/380 380 5.6
ergostadiene 81/147/367/382 382 3.4
stigmasta-3,5,22-triene 135/143/394 394 11.7

195
5. Resultados y discusin

stigmasta-3,5-diene 81/147/381/396 396 9.4


sterols 113.1
campesterol 55/145/213/382/400 400 14.0
stigmasterol 55/83/255/394/412 412 35.7
sitosterol 145/213/396/414 414 30.7
stigmastanol 215/416 416 2.9
7-oxocampesterol 382/397/414 414 10.1
7-oxostigmasterol 286/426 426 9.0
7-oxositosterol 396/413/428 428 10.7
-tocopherol 165/205/430 430 2.1
steroid ketones 30.5
stigmasta-7,22-dien-3-one 55/269/298/367/410 410 7.1
stigmasta-3,5-dien-7-one 174/269/410 410 14.1
stigmast-4-en-3-one 124/229/412 412 4.5
stigmastadienone isomer 57/136/174/269/410 410 2.0
stigmastane-3,6-dione 245/287/428 428 2.8
sterol glycosides* 13.2
campesterol 3-D-glucopyranoside 204/217/361/383 850 1.9
stigmasterol 3-D-glucopyranoside 204/217/361/413 880 4.3
sitosterol 3-D-glucopyranoside 204/217/361/397 864 7.0
sterol esters 28.3
campesterol esters n.d 2.3
stigmasterol esters n.d 9.2
sitosterol esters n.d 16.8
monoglycerides* 8.1
1-monopalmitin 73/103/129/147/371/459 474 4.2
1-monostearin 73/103/129/147/399/487 502 3.9
* as TMS ethers; # as methyl esters and TMS ethers

Steroid compounds were the second most important lipid class found among
the tagasaste extractives and included steroid hydrocarbons, steroid ketones and
sterols (in free and conjugated form). Sterols, in free and conjugated (esters and
glycosides) form were the most important steroids identified, free sterols being
present in important amounts (113.1 mg/kg). Stigmasterol (IV) was the most
abundant among the free sterols with the presence of important amounts of
campesterol (V) and sitosterol (VI) and minor amounts of stigmastanol (VII) as
well as the presence of the oxidized 7-ketocampesterol, 7-ketostigmasterol and
7-ketostigmastanol. Lower amounts of sterols could also be found in ester form
(28.3 mg/kg), with a predominance of sitosterol esters. Sterol glycosides, such
as campesteryl, stigmasteryl and sitosteryl 3E-D-glucopyranosides (VIII) were

196
5. Resultados y discusin

Figure 3. Structures of the main lipophilic compounds present in tagasaste. I: hexadecanoic


acid; II: 9-hexadecenoic acid; III: 2-hydroxyhexacosanoic acid; IV: stigmasterol; V:
campesterol; VI: sitosterol; VII: stigmastanol; VIII: sitosteryl 3E-D-glucopyranoside; IX:
stigmasta-3,5-dien-7-one; X: stigmasta-3,5,22-triene; XI: n-hentriacontane; XII: 1-
monopalmitin.

also identified in lower amounts (13.2 mg/kg), the latter being the most
predominant. The identification of steryl glycosides was accomplished, after
BSTFA derivatization of the lipid extract, by comparison with the mass spectra
and relative retention times of authentic standards (Gutirez and del Ro, 2001).
It is important to point out the low content of free and conjugated (esters and
glycosides) sterols, in comparison to that found in eucalypt wood (Rencoret et
al., 2007), since these are the main compounds responsible for pitch deposition
during kraft cooking of hardwoods, such as eucalypt wood (del Ro et al. 1998,
2000; Silvestre et al. 1999; Gutirrez and del Ro 2001). On the other hand,

197
5. Resultados y discusin

steroid ketones were also found in important amounts (30.5 mg/kg), being
mainly constituted by stigmasta-7,22-dien-3-one, stigmasta-3,5-dien-7-one (IX),
stigmast-4-en-3-one and stigmasta-3,6-dione. Different steroid hydrocarbons
(di- and triunsaturated) were also identified, although in low amounts (30.1
mg/kg), stigmasta-3,5,22-triene (X) being the most predominant.
Finally, a series of n-alkanes ranging from n-eicosane (C20) to n-tritriacontane
(C33) with n-hentriacontane (X, C31) being the most prominent, was found in low
amounts (8.8 mg/kg). And minor amounts of monoglycerides (8.1 mg/kg), such
as 1-monopalmitin (XI) and 1-monostearin, were also present among tagasaste
extractives.

4. Conclusions
Tagasaste is characterized by a high content of holocellulose and D-cellulose
and a relatively low lignin content. Moreover, the chemical composition of
tagasaste lignin indicates a predominance of S-lignin units over the G-lignin
ones (S/G molar ratio of 1.6) and the absence of H-lignin units. The total
acetone extractives accounted for 1.4% but the content on lipophilic compounds
is very low (only 0.2%). The main lipids identified in tagasaste were series of
fatty acids, including D-hydroxyfatty acids, and steroid compounds, including
steroid hydrocarbons, steroid ketones, sterols, sterol esters and sterol glycosides.

Acknowledgements
This study has been supported by the Spanish project AGL2005-01748. G.
Marques acknowledges a FPI Fellowship from the Spanish Ministry of
Education and Science. We thank Manuel J. Daz (University of Huelva, Spain)
for providing the tagasaste samples.

References
Back, E.L., Allen, L.H., 2000. Pitch Control, Wood Resin and Deresination.
Tappi Press, Atlanta.

Borens, F., Poppi, D.P., 1990. The nutritive and feeding value for ruminants of
tagasaste (Chamaecytisus palmensis), a leguminous tree. Anim. Feed Sci.
Technol. 28, 275292.

Browning, B.L., 1967. Methods of Wood Chemistry, vol. II; Wiley-Interscience


Publishers, New York.

Chang, H-M., Sarkanen, K.V., 1973. Species variation in lignin. Effect of


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

198
5. Resultados y discusin

del Ro, J.C., Gutirrez, A., Gonzlez-Vila, F.J., Martn, F., Romero, J., 1998.
Characterization of organic deposits produced in the Kraft pulping of
Eucalyptus globulus wood. J. Chromatogr. A 823, 457-465.

del Ro, J.C., Romero, J., Gutirrez, 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 Ro J.C., Gutirrez A., Romero J., Martnez M.J. and Martnez A.T., 2001.
Identification of residual lignin markers in eucalypt kraft pulp by Py-GC-MS.
J. Anal. Appl. Pyrolysis 58/59, 425-439.

del Ro, J.C., Gutirrez, A., Hernando, M., Landn, P., Romero, J., Martnez,
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.

Daz, M.J., Alfaro, A., Garca, M.M., Eugenio, F.M.E., Ariza, J., Lpez, F.,
2004. Ethanol pulping from tagasaste (Chamaecytisus proliferus L.F. ssp.
Palmensis). A new promising source for cellulose pulp. Ind. Eng. Chem. Res.
43, 1875-1881.

Faix, O., Meier, D., Fortmann, I., 1990. Thermal degradation products of wood.
A collection of electron of electron-impact (EI) mass spectra of monomeric
lignin derived products. Holz Roh- Werkst. 48, 351-354.

Francisco-Ortega, J., Jackson, M.T., Santos-Guerra, A., Fernndez-Galvn, M.,


1991. Historical aspects of the origin and distribution of tagasaste
(Chamaecytisus proliferus (L. fil.) Link ssp. palmensis (Christ) Kunkel), a
fodder tree from the Canary Islands. J. Adelaide Botanical Garden 14, 6776.

Garca, M.M., Lpez, F., Alfaro, A., Ariza, J., Tapias, R., 2007. The use of
tagasaste (Chamaecytisus proliferus) from different origins for biomass and
paper production. Biores. Technol. (in press).

Gonzlez-Vila, F.J., Almendros, G., del Ro, J.C., Martn, F., Gutirrez, A.,
Romero, J. 1999. Ease of delignification assessment of wood from different
Eucalyptus species by pyrolysis (TMAH)-GC/MS and CP/MAS 13C NMR
spectrometry. J. Anal. Appl. Pyrol. 49, 295-305.

Gutirrez, A., del Ro, J.C., Gonzlez-Vila, F.J., Martn, F., 1998. Analysis of
lipophilic extractives from Wood and pitch deposits by solid-phase extraction
and gas chromatography. J. Chromatogr. A 823, 449-455.

199
5. Resultados y discusin

Gutirrez, A., del Ro, 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.

Hillis, W.E., Sumimoto, M., 1989. Effect of extractives on pulping. In: Rowe,
J.W. (Ed.), Natural Products of Woody Plants II; Springer-Verlag, Berlin. pp.
880-920.

Jimnez, L., Prez, A., Rodrguez, A., de La Torre, M.J., 2006. New raw
materials and pulping processes for production of pulp and paper. Afinidad 63
(525), 362-369.

Jimnez, L., Prez, A., de la Torre, M.J., Moral A., Serrano, L., 2007.
Characterization of vine shoots, cotton stalks, Leucaena leucocephala and
Chamaecytisus proliferus, and of their ethyleneglycol pulps. Biores. Technol.
98, 3487-3490.

Kindu, M., Glatzel, G., Tadesse, Y., Yosef, A., 2006. Tree species screened on
nitosols of central Ethiopia: Biomass production, nutrient contents and effect
on soil nitrogen. J. Trop. Forest Sci. 18, 173-180.

Lpez, F., Alfaro, A., Garca, M.M., Diaz, M.J., Calero, A.M., Ariza, J., 2004.
Pulp and paper from tagasaste (Chamaecytisus proliferus LF ssp palmensis).
Chem. Eng. Res. & Des. 82, 1029-1036.

Ralph, J., Hatfield, R.D., 1991. Pyrolysis-GC/MS characterization of forage


materials. J. Agric. Food Chem. 39, 1426-1437.

Rencoret, J., Gutirrez, A., del Ro, J.C., 2007. Lipid and lignin composition of
woods from different eucalypt species. Holzforschung 61, 165-174.

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.

Technical Association of the Pulp and Paper Industry, 1993. Test methods,
1992-1993. TAPPI, Atlanta, GA..

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.

200
5. Resultados y discusin

Ventura, M.R., Castanon, J.I.R., Rey, L., Flores, M.P., 2002. Chemical
composition and digestibility of tagasaste (Chamaecytisus proliferus)
subspecies for goats. Small Ruminant Res. 46, 207-210.

201
5. Resultados y discusin

Publicacin VII:

Marques G., del Ro J.C. and Gutirrez A. (2010) Lipophilic extractives from
several nonwoody lignocellulosic crops (flax, hemp, sisal, abaca) and their fate
during alkaline pulping and TCF/ECF bleaching. Bioresource Technology, 101,
260-267.

202
5. Resultados y discusin

Lipophilic extractives from several nonwoody lignocellulosic crops (flax,


hemp, sisal, abaca) and their fate during alkaline pulping and TCF/ECF
bleaching
Gisela Marques, Jos C. del Ro and Ana Gutirrez

Instituto de Recursos Naturales y Agrobiologa, CSIC, PO Box 1052, E-41080, Seville, Spain

Abstract
The fate of lipophilic extractives from several nonwoody species (flax, hemp,
sisal and abaca) used for the manufacturing of cellulose pulps, was studied
during soda/anthraquinone (AQ) pulping and totally chorine free (TCF) and
elemental chlorine free (ECF) bleaching. With this purpose, the lipophilic
extracts from the raw materials and their unbleached and bleached industrial
pulps, were analyzed by gas chromatography-mass spectrometry. Aldehydes,
hydroxyfatty acids and esterified compounds such as ester waxes, sterol esters
and alkylferulates strongly decreased after soda/AQ pulping while alkanes,
alcohols, free sterols and sterol glycosides survived the cooking process. Among
the lipophilic extractives that remained in the unbleached pulps, some amounts
of free sterols were still present in the TCF pulps whereas they were practically
absent in the ECF pulps. Sterol glycosides were also removed after both TCF
and ECF bleaching. By contrast, saturated fatty acids, fatty alcohols and alkanes
were still present in both bleached pulps.
Keywords: pitch; lipophilic extractives; paper pulp; nonwoody fibers

1. Introduction
Lipophilic extractives, i.e. the non-polar extractable fraction from wood and
other lignocellulosic crops, include different classes of compounds, such as
alkanes, fatty alcohols, fatty acids, free and conjugated sterols, terpenoids,
triglycerides and waxes. These lipophilic compounds, even when present in low
amounts in the raw material, may play an important role in industrial wood
processing for bleached pulp and paper production since they are at the origin of
the so-called pitch deposits along the pulp and paper manufacturing processes.
Pitch deposition is a serious problem in the pulp and paper industry since it is
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 (Back and Allen, 2000).
The nature and severity of pitch deposition depend not only on the raw
materials used (and hence on the nature of the lipophilic compounds), but also
on the industrial processes of pulping and bleaching applied at the mill. In the
manufacture of alkaline pulps, a large part of the lipids originally present in the

203
5. Resultados y discusin

raw material is removed during the cooking and bleaching stages. However,
some chemical species survive these processes and are found as pulp extractives
(Bergelin and Holmbom, 2003; Freire et al., 2005; 2006; Gutirrez et al.,
2001a), suspended in process waters (Gutirrez et al., 2001b) or forming the so-
called pitch deposits in circuits, equipments and final product (Bergelin et al.,
2005; del Ro et al., 1998; 2000; Gutirrez and del Ro, 2005; Silvestre et al.,
1999). Growing pressure for closing water loops in pulp and paper mills leads to
increasing build up of lipophilic compounds in the processes and therefore, to an
increasing number of severe and costly pitch related problems.
The different classes of lipids have different behavior during pulping and
bleaching. Several studies have provided information on the behavior of wood
lipids, such as fatty and resin acids, triglycerides and sterols and triterpenols
during pulping and bleaching (Back and Allen, 2000; Bergelin and Holmbom,
2003; 2008; Freire et al., 2005; 2006; Gutirrez et al., 2001a; Shin and Kim,
2006). However, the chemistry of lipids from nonwoody pulps in pulping and
bleaching has not been examined much so far (Gutirrez et al., 2004; Gutirrez
and del Ro, 2003a; 2003b). Likewise, to the best of our knowledge, there is
very little information available dealing with pitch problems on nonwoody pulps
(Gutirrez and del Ro, 2005). In this context, the aim of this work was to
identify the specific lipophilic constituents of different nonwoody fibers, which
are used for the manufacturing of cellulose pulps for specialty papers, and to
study their behavior during soda/anthraquinone (AQ) pulping and totally chorine
free (TCF) and elemental chlorine free (ECF) bleaching. For this, a series of
pulps (crude pulps taken after soda/AQ pulping and bleached pulps taken after
TCF and ECF bleaching) from different nonwoody raw materials (flax, hemp,
sisal, abaca) were selected for this study. The composition of the lipophilic
compounds in the fibers and their respective pulps was analyzed 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 intact
high molecular weight lipids such as waxes or sterol glycosides (Gutirrez et al.,
1998; Gutirrez and del Ro, 2001). The knowledge of the behavior of lipophilic
extractives during pulping and bleaching is an important step towards
understanding and predicting the pitch problems and designing effective
solutions for its control.

2. Materials and methods


2.1. Samples
Two bast fibers, flax (Linum usitatissimum) and hemp (Cannabis sativa), and
two leaf fibers, sisal (Agave sisalana) and abaca (Musa textilis), were selected
for this study. The raw materials and their respective crude (unbleached) pulps
(pulps taken after soda/AQ pulping), as well as pulp samples after TCF and ECF

204
5. Resultados y discusin

bleaching, were supplied by CELESA pulp mill (Tortosa, Spain). General


conditions of soda/AQ pulping included the use of sodium hydroxide and
anthraquinone (up to 0.05%) as cooking chemicals, and 2-3 h of cooking time at
a temperature of about 165C. These general conditions can be slightly modified
depending on the raw material used. The TCF bleaching sequence used (QPoP)
included a quelating stage (Q) and two hydrogen peroxide stages (P), the first
one under pressurized oxygen (Po). The ECF bleaching sequence used (DPo)
included a chlorine dioxide stage (D) followed by a hydrogen peroxide stage
under pressurized oxygen (Po).

2.2. Lipid extraction


Raw materials and pulps were air-dried until constant weight and the samples
were Soxhlet-extracted with acetone for 8 h. All the extracts were evaporated to
dryness and redissolved in chloroform for chromatographic analysis of the
lipophilic fraction as described below.

2.3. GC and GC-MS analyses


The GC analyses of the lipids from the raw materials and pulps were performed
in an Agilent 6890N Network GC system using a short fused-silica DB-5HT
capillary column (5 m x 0.25 mm internal diameter, 0.1 m film thickness) from
J&W Scientific, enabling simultaneous elution of the different lipid classes
(Gutirrez et al., 1998). The temperature program was started at 100C with 1
min hold, and then raised to 350C at 15C/min, and held for 3 min. The injector
and flame-ionization detector (FID) temperatures were set at 300C and 350C,
respectively. Helium (5 ml/min) was used as carrier gas, and the injection was
performed in splitless mode.
The GC-MS analyses were performed with a Varian 3800 chromatograph
coupled to an ion-trap detector (Varian 4000) using a medium-length (12 m)
capillary column of the same characteristics described above for GC/FID. The
oven was heated from 120C (1 min) to 380C at 10C/min, and held for 5 min.
The transfer line was kept at 300C, the injector was programmed from 120C
(0.1 min) to 380C at 200C/min and held until the end of the analysis, and
helium was used as carrier gas at a rate of 2 ml/min. The compounds were
identified by mass fragmentography, and by comparing their mass spectra with
those of the Wiley and NIST libraries and standards.
Both underivatized and derivatized samples were analyzed by GC and GC-
MS. Two derivatized samples, silylated samples and methylated and silylated
samples, were analyzed. The silylation was peformed using
bis(trimethylsilyl)trifluoroacetamide (BSTFA) in the presence of pyridine at
80C for 90 min (Gutirrez and del Ro, 2001). Trimethylsilyl-diazomethane
methylation, in the presence of methanol (at room temperature for 10 min)
followed by BSTFA silylation, in the presence of pyridine, were also performed.

205
5. Resultados y discusin

Fatty alcohols, fatty acids and sterol glycosides were determined as silylated
derivatives in the derivatized samples. Hydroxyfatty acids were determined in
the samples after methylation followed by silylation. The rest of compounds
were determined in the underivatized samples. Peaks were quantified by area
and a mixture of standards (octadecane, palmitic acid, sitosterol, cholesteryl
oleate, and sitosteryl 3-D-glucopyranoside) with a concentration range between
0.1 and 1mg/ml was used to elaborate calibration curves. The data from the two
replicates were averaged.

3. Results and discussion


The lipid content of the nonwoody fibers used in this study and their
corresponding crude and TCF and ECF bleached pulps is shown in Table 1.
Flax fibers have the highest content on lipophilic extractives followed by hemp,
sisal and abaca fibers. Regardless the higher lipid content of flax fibers, the
content of lipophilic extractives of its crude pulp was similar or even lower than
in the other fibers. On the other hand, the lipid content of the pulps decreased
further after TCF or ECF bleaching, ranging from 0.03 to 0.18%. The
differences observed in the lipid content among the different crude and bleached
pulps of the nonwoody fibers is related to the behavior of lipophilic extractives
present in the raw materials in alkaline pulping and to their reactivity towards
the bleaching agents used as described below.

3.1. Composition of lipophilic extractives in the nonwoody fibers selected


for this study
With the aim of studying the fate of the different lipophilic extractives during
soda/AQ pulping, the lipid extracts from the different nonwoody fibers selected
for this study were analyzed by GC and GC-MS and then compared with those
in the crude pulps. The composition of the lipophilic extractives of the selected
raw materials are listed in Table 2 and the chemical structures of the main
compounds identified are represented in Fig. 1.

Table 1. Lipophilic extractives content (%) in the selected nonwoody raw materials and
pulps

Flax Hemp Sisal Abaca

Fiber 1.70 0.88 0.68 0.51


Crude pulp 0.21 0.26 0.30 0.20
TCF pulp 0.05 0.18 0.09 0.04
ECF pulp 0.13 0.18 0.14 0.03

206
5. Resultados y discusin

OH
I II

O O

H OH
III IV

HO O
V VI

CH2OH
O
OH O HO
HO
VII VIII
OH

O
IX

Fig.1. Chemical structures of compounds representing the main classes of lipophilic


extractives found in lignocellulosic fibers: I, pentacosane; II, octacosanol; III, octacosanal;
IV, palmitic acid; V, sitosterol; VI, sitosteryl linoleate; VII sitosteryl 3E-D-glucopyranoside;
VIII, E-amyrin; and IX, octacosyl hexadecanoate.

The main lipid classes present in the nonwoody fibers selected for this study
were series of alkanes (I), fatty alcohols (II), aldehydes (III), fatty acids (IV),
steroids, including free (V) and conjugated sterols (VI-VII), triterpenoids (VIII)
and ester waxes (IX). The detailed composition of the lipophilic compounds
present in these fibers have been previously addressed (del Ro et al., 2004; del
Ro and Gutirrez, 2006; Gutirrez et al., 2006; 2008; Gutirrez and del Ro,
2003a; 2003b).
In the case of flax bast fibers the predominant lipophilic compounds present
were series of fatty acids and aldehydes, accounting for 34% and 23% of total
extract, respectively, followed by ester waxes (18%), and fatty alcohols (13%).
Fatty acids were also the predominant compounds (27% of total extracts) in
hemp bast fibers, followed by alkanes (15%), free sterols (12%) and steroid
hydrocarbons (12%). Among the selected leaf fibers, free sterols predominated
in both sisal (20%) and abaca (45%), followed by alkanes (14%) and fatty acids
(10%) in sisal, and by fatty acids (16%) in abaca fibers.

207
5. Resultados y discusin

These lipophilic compounds can be classified into two principal groups,


namely organic acids and neutral components. Organic acids include fatty acids,
and the neutrals include alkanes, aldehydes, fatty alcohols, and free sterols as
well as esterified and conjugated compounds such as ester waxes, sterol esters,
alkylferulates and sterol glycosides. Organic acids and neutral compounds, the
latter including saponifiable and unsaponifiable compounds present different
behavior during alkaline pulping, as shown below.

3.2. Fate of lipophilic extractives during soda/AQ pulping


The fate of the different lipophilic extractives during soda/AQ pulping was
studied by analyzing the respective crude pulps. This is exemplified in Figs. 2
and 3 for the hemp bast fibers and the sisal leaf fibers, respectively. The
composition of main lipids present in the crude pulps from the studied fibers is
listed in Table 2. It can be observed that the lipophilic compounds from the raw
materials are modified to a different extent during alkaline (soda/AQ) pulping
depending of their chemical nature. Fatty acids, which are among the main
lipophilic extractives predominating in all these fibers, were also present in
significant amounts in the crude pulps. The content of fatty acids decreased
significantly after soda/AQ pulping of flax fibers whereas an increase in fatty
acids content was produced after pulping of the other fibers, being especially
evident in the case of sisal and abaca. In contrast, -hydroxyfatty acids present
in all the fibers and Z-hydroxyfatty acids present in sisal and abaca were
completely absent in crude pulps.
On the other hand, esterified compounds such as ester waxes, sterols esters
and ferulic acid esters, were completely hydrolyzed during soda/AQ pulping.
This is especially evident in the case of flax fibers where ester waxes
predominate. In contrast, other conjugated compounds, namely sterol
glycosides, which are present in all nonwoody fibers studied here, resisted the
alkaline cooking conditions and were present intact in the crude pulps, as
already reported in the pulping of woody (Gutirrez and del Ro, 2001;
Nilvebrant and Bystrm, 1995) and nonwoody plants (Gutirrez and del Ro,
2003a; 2003b; Gutirrez et al., 2004). The importance of the presence of sterol
glycosides after alkaline pulping is due to their high hydrophilic-lipophilic
balance, high melting point and very low solubility in water, alkali and the usual
organic solvents (Hillis and Sumimoto, 1989). Due to these properties, sterol
glycosides constitute a part of protecting layers that prevent the cooking and
bleaching chemicals to reach the resin and thereby keep them and other
extractives in the pulp. Other neutral compounds, such as aldehydes, which were
present in significant amounts in flax and hemp fibers, decreased largely during
pulping and therefore were practically absent in the crude pulps. The content of
steroid hydrocarbons and steroid ketones also decreased after soda/AQ cooking.

208
5. Resultados y discusin

Ak 29 5+6+7
+
Al26 11
Hemp bast fibers

2
1 9
Fa18:1 +
+ 10
Fa16 Fa18:2 Ak 27 8
+ 34 W44
Fa18 Al24 12 W42 W46
SE W48 W50

5+6+7

Ak 31
+
Al28 Crude pulp
Ak 29
+
Al26
Fa18:1 4 Al30
+ Ak 27 3
Fa18:2
AQ + 11
Al24
Fa16 Fa18 1

5+6+7

Ak 29 Ak 31
+ +
Al26 Al28 TCF pulp
Def oamer

Fa16 Al30
Fa18
Ak 27 11
AQ
+ 1 34
Al24

Ak 31
+
Al28
Ak 29
+
Al26
ECF pulp
Al30
11
7
Ak 27
Al22 +
AQ Al24
Fa16 Fa18 1

5 10 15 20 25
Retention time (minutes)

Fig. 2. GC-MS chromatograms of underivatized lipophilic extractives from hemp fibers (raw
material), and their crude, TCF and ECF pulps. Peak identification, 1: stigmasta-3,5,22-triene;
2: stigmasta-3,5-diene; 3: campesterol; 4: stigmasterol; 5: sitosterol; 6: stigmastanol; 7: -
amyrin; 8: stigmasta-3,5-dien-7-one; 9: -amyrin acetate; 10: stigmast-4-en-3-one; 11:
friedelan-3-one; 12: stigmastane-3,6-dione; AQ: anthraquinone; Fa(n): fatty acids; Ak(n):
alkanes; Al(n): fatty alcohols; SE: sterol esters; W(n): ester waxes; n denotes the total carbon
atom number.

209
5. Resultados y discusin

5+6+7

Sisal leaf fibers

Al28
1 4
Ak 29
+ 2 8
Al26
Fa16 3
Ak 25 F24 9
Ak 21 Ak 23 Ak 27
Fe26 Fe28
Fe24

5+6+7

Al28
Crude pulp
Ak 29
+
Al26 4
AQ Ak 25 Ak 27
Al30
Fa16 + + 3 9
Al22 Al24 1 8
Fa18

Al28
5+6

Ak 29
+ TCF pulp
Al26

Def oamer Ak 25 4
+ Ak 27
Al22 + Al30
Al24 9
AQ Fa16 Fa18 1 3

Al28
Ak 29
+
Al26
5+6 ECF pulp

Ak 25 Ak 27 Al30
Fa16 + +
Al24 4
AQ Al22
1 9
Fa18

5 10 15 20 25
Retention time (minutes)

Fig. 3. GC-MS chromatograms of underivatized lipophilic extractives from sisal fibers (raw
material), and their crude pulp, TCF pulp and ECF pulps. Peak identification, 1: stigmasta-
3,5,22-triene; 2: stigmasta-3,5-diene; 3: campesterol; 4: stigmasterol; 5: sitosterol; 6:
stigmastanol; 7: -amyrin; 8: stigmasta-3,5-dien-7-one; 9: hecogenin; AQ: anthraquinone;
Fa(n): fatty acids; Ak(n): alkanes; Al(n): fatty alcohols; Fe(n): n-alkylferulates; n denotes the
total carbon atom number.

210
5. Resultados y discusin

Table 2. Composition of lipophilic extractives (mg/100 g) in nonwoody raw


materials (M) and crude (C), TCF (T) and ECF (E) pulps
Flax Hemp Sisal Abaca
Compound M C T E M C T E M C T E M C T E
Fatty acids 552 96 29 39 78 60 12 80 9 50 10 42 9 57 19 28
n-hexadecanoic acid 121 27 4 18 34 15 4 29 4 24 8 27 4 20 9 10
9,12-octadecadienoic 1 tr - - 3 1 1 - - - - - 1 3 - -
acid
9-octadecenoic acid 235 39 2 1 15 1 tr 1 - - - - 2 6 - -
n-octadecanoic acid 52 22 8 12 6 10 3 20 <1 6 7 16 1 21 11 16
n-nonadecanoic acid tr 3 tr tr tr 5 1 tr - - - - tr - tr tr
n-eicosanoic acid 20 2 4 4 4 6 <1 6 tr 1 <1 2 <1 2 tr 1
n-heneicosanoic acid 4 tr tr tr tr 1 1 1 - - - - tr - - tr
n-docosanoic acid 36 2 4 2 5 7 1 8 tr tr 1 2 <1 tr tr tr
n-tricosanoic acid 16 tr tr tr 2 3 1 4 1 tr <1 1 tr tr tr tr
n-tetracosanoic acid 40 1 4 2 4 7 <1 6 1 1 1 3 1 3 <1 1
n-pentacosanoic acid 8 tr tr tr 2 1 <1 1 <1 2 tr - tr - tr tr
n-hexacosanoic acid 19 - 3 tr 3 3 tr 1 1 10 1 2 tr 2 - 1
n-octacosanoic acid - - - - - - tr 3 2 17 tr tr - - - -
-Hydroxyfatty acids 11 - - - 9 tr - - 7 - - - 1 - - -
2-hydroxydocosanoic 3 - - - - - - - 1 - - - - - - -
acid
2-hydroxytricosanoic 2 - - - - - - - 1 - - - - - - -
acid
2-hydroxytetracosanoic 6 - - - 5 tr - - 4 - - - 1 - - -
acid
2-hydroxypentacosanoic tr - - - 4 - - - 1 - - - tr - - -
acid
-Hydroxyfatty acids - - - - - - - - 6 - - - 1 - - -
22-hydroxydocosanoic - - - - - - - - 1 - - - 1 - - -
acid
26-hydroxyhexacosanoic - - - - - - - - 2 - - - tr - - -
acid
28-hydroxyoctacosanoic - - - - - - - - 3 - - - - - - -
acid
Fatty alcohols 220 68 1 52 2 44 13 59 8 74 18 51 tr 6 1 <1
n-docosanol - - - - - 4 1 4 - 5 2 3 tr 2 <1 <1
n-tetracosanol 4 2 tr - <1 3 1 4 - 6 1 5 tr 1 <1 tr
n-hexacosanol 49 15 <1 11 <1 8 2 12 2 20 5 14 tr 1 <1 tr
n-octacosanol 116 38 1 29 1 21 6 28 5 35 9 25 tr 2 <1 tr
n-triacontanol 51 13 <1 12 1 8 3 11 1 8 1 4 tr <1 - -
Alkanes 27 22 1 3 43 44 14 23 15 18 4 22 - - - -
n-heneicosane - - - - - - - - 2 - - - - - - -
n-docosane - - - - - - - - 1 - - - - - - -
n-tricosane - - - - - - - - 2 - <1 1 - - - -
n-tetracosane - - - - - - - - 1 tr - tr - - - -
n-pentacosane 2 2 <1 tr <1 1 - - 5 5 2 5 - - - -
n-hexacosane - - - - - - - - 1 tr tr tr - - - -
n-heptacosane 5 3 <1 1 3 5 1 2 3 3 1 4 - - - -
n-nonacosane 20 17 <1 2 38 31 12 19 <1 10 1 12 - - - -
n-hentriacontane - - - - 2 7 1 2 - tr tr tr - - - -

211
5. Resultados y discusin

Aldehydes 371 3 <1 8 25 <1 tr 6 1 tr tr 2 - - - -


n-hexacosanal 58 tr tr 1 3 - - 1 tr tr tr - - - - -
n-octacosanal 174 3 <1 5 9 <1 tr 3 1 tr 1 2 - - - -
n-triacontanal 139 tr tr 2 13 tr - 2 tr tr 1 4 - - - -
Steroid hydrocarbons 14 3 1 2 30 5 1 1 14 11 1 3 3 4 1 <1
stigmasta-3,5,22-triene 5 3 1 2 12 4 1 1 10 11 1 3 2 3 1 <1
stigmasta-3,5-diene 9 <1 tr tr 18 1 <1 tr 4 tr - - 1 1 <1 tr
Steroid ketones 33 6 <1 tr 27 6 3 9 3 5 1 7 4 6 5 1
cycloartenone - - - - - - - - - - - - 2 1 <1 -
-amirenone - - - - tr 1 tr - - - - - - - - -
stigmasta-3,5-dien-7-one 5 3 <1 tr 3 1 1 <1 2 3 1 5 1 3 3 1
stigmast-4-en-3-one 9 1 tr - 5 - - - 1 1 tr tr 1 1 1 <1
stigmastadienone isomer 3 1 - - 2 <1 <1 2 <1 tr <1 2 <1 1 1 tr
friedelan-3-one - - - - 15 4 2 7 - - - - - - - -
ergostane-3,6-dione 6 tr - - - - - - - - - - - - - -
stigmastane-3,6-dione 10 1 - - 2 - - - <1 1 tr - - - - -
Free 92 33 3 - 36 78 15 6 20 10 14 14 25 20 5 <1
sterols/triterpenols 7
campesterol 27 8 1 - 3 9 2 - <1 9 tr - 2 3 1 tr
stigmasterol 23 5 <1 - 3 9 2 - 3 15 2 2 2 3 1 <1
sitosterol 36 17 2 - 18 37 7 - 11 56 8 8 10 11 3 <1
stigmastanol 6 3 <1 - 2 6 1 - 1 6 1 1 tr 1 <1 <1
-amyrin - - - - 9 15 3 5 1 6 tr tr - - - -
-amyrin - - - - 1 2 <1 1 tr 1 tr tr - - - -
7-oxositosterol - - - - - - - - 2 2 1 tr 11 2 <1 tr
hecogenin - - - - - - - - 2 12 2 2 - - - -
Sterol glycosides* 5 8 1 tr 13 16 <1 tr 2 25 2 1 2 8 <1 tr
campesterol 3-D-Glcp tr 2 <1 tr 2 3 tr - tr 6 <1 tr <1 3 <1 tr
stigmasterol 3-D- Glcp tr 1 tr tr 1 2 tr - <1 3 <1 tr tr 1 tr tr
sitosterol 3-D-Glcp 5 5 1 tr 10 11 <1 tr 2 16 2 1 2 4 <1 tr
Sterol/triterpenol 6 - - - 7 - - - <1 - - - 1 - - -
esters
-amyrin acetate - - - - 6 - - - - - - - - - - -
sitosterol esters 6 - - - 1 - - - <1 - - - 1 - - -
Ester waxes* 284 - - - 17 - - - <1 - - - - - - -
C40 8 - - - tr - - - - - - - - - - -
C42 19 - - - 4 - - - tr - - - - - - -
C44 68 - - - 5 - - - <1 - - - - - - -
C46 130 - - - 4 - - - tr - - - - - - -
C48 56 - - - 2 - - - <1 - - - - - - -
C50 4 - - - 2 - - - <1 - - - - - - -
Ferulates - - - - - - - - 5 - - - - - - -
t-tetracosanil ferulate - - - - - - - - 1 - - - - - - -
t-hexacosanil ferulate - - - - - - - - 2 - - - - - - -
t-octacosanil ferulate - - - - - - - - 2 - - - - - - -

* Abreviations: Glcp (glucopyranoside); C(n), n denotes the total carbon atom number; tr: traces

212
5. Resultados y discusin

Finally, alkanes, fatty alcohols and free sterols and triterpenols survived the
alkaline conditions and therefore were the main lipophilic constituents in crude
pulps.
The behavior of the fatty acids in an aqueous environment is quite different
from that of the neutrals. At alkaline pH, the acids dissociate and can dissolve in
water to quite a high extent forming fatty acid soaps that can form micelles.
Fatty acid soaps are effective solubilizing agents facilitating the removal from
pulp of sparingly soluble neutral substances (Back, 2000). The ratio of
saponifiables-to-unsaponifiables has been suggested to be a better index for
predicting pitch problems than the total amount of lipids (Back and Allen,
2000). In fact, the higher abundances of unsaponifiable compounds with respect
to the saponifiable ones is the main cause for pitch problems during pulping of
some woods, such as aspen or eucalypt (Chen et al., 1995; del Ro et al., 1998;
2000). This fact can explain why the lipid content of crude flax pulps is similar
or even lower than the other pulps regardless flax fibers had the highest
extractive content. As mentioned above, fatty acids are the predominat lipids in
flax fibers and can help to solubilize other water-insoluble components such as
fatty alcohols and sterols. In contrast, the lipid content of crude sisal pulp is
higher than the other three pulps regardless the relatively low content of
lipophilic extractives in the raw material. Sisal fibers have low fatty acid content
whereas the content in neutrals such as alkanes, fatty alcohols and steroids is
relatively high. Particularly, the high abundances of free sterols, which have a
high propensity to form pitch deposits (del Ro et al., 1998; 2000), in the crude
pulps from hemp and sisal, would point to a high pitch deposition tendency of
the lipophilics from these fibers.

3.3. Fate of lipophilic extractives during TCF and ECF bleaching


The lipophilic extractives remaining in the crude pulp are carried over to the
bleach plant, where they react with the bleaching agents used (Bjrklund-
Jansson et al., 1995; Holmbom, 2000). Pulp bleaching technology radically
changed in the 1990s when the previously used chlorine was replaced and
several new bleaching chemicals were introduced (Sjstrm, 1993) and ECF and
TCF sequences were adopted. As a consequence, new pitch problems arose due
to the different reactivity of pulp lipids with the new bleaching agents. With the
aim of studying the behavior of the different classes of lipophilic extractives
under different bleaching conditions, both TCF and ECF bleached pulps were
analyzed and compared with their respective crude pulps.
In general terms, the qualitative composition of the lipophilic extractives from
the TCF pulps was very similar to that of their respective crude pulps (Figs. 2
and 3) although the lipid content was significantly lower (Table 2). Fatty
alcohols, alkanes, free sterols and triterpenols, and fatty acids were the main
lipophilic compounds present in the TCF bleached pulps, although in lower
amount than in their respective crude pulps. The low reactivity of most pulp

213
5. Resultados y discusin

lipids in TCF bleaching sequences (using hydrogen peroxide as a bleaching


agent) has been reported (Freire et al., 2003; Gutirrez et al., 2001a). The
decrease in the content of these lipophilic compounds in the TCF pulps,
regardless the low reactivity of hydrogen peroxide towards them, might be due
to the extensive washing in alkaline conditions carried out in the industrial TCF
bleaching sequences used. Moreover, it is interesting to note that sterol
glycosides, which resisted soda/AQ pulping conditions, were removed to a high
extent under the industrial TCF bleaching conditions used in this work. The high
efficiency of hydrogen peroxide in degrading sterol glycosides was previously
reported using model compounds (Nilvebrant and Bystrm, 1995). However, a
lower extent in the removal of sterol glycosides were observed during TCF
bleaching of eucalypt pulps (Gutirrez et al., 2006). In addition to the lipophilic
extractives arising from the raw materials, several non-resolved peaks
corresponding to the defoamers used in the process could also be observed in
TCF pulps (Figs. 2 and 3). The excessive use of defoamers can also produce
problems of pitch deposition (Allen, 2000). The presence of defoamer in pitch
deposits produced during manufacturing of paper pulp from hemp fibers has
been reported (Gutirrez and del Ro, 2005).
On the other hand, the composition of the lipophilic extractives from ECF
pulps was somewhat different compared to that from their respective crude
pulps (Figs. 2 and 3). The main difference observed was the large removal of
free sterols in all the ECF pulps, although some minor amounts of free sterols
still remained in sisal ECF pulp. The complete degradation of unsaturated
sterols, such as sitosterol, during ECF bleaching has been previously reported in
pitch deposits, pulps and in reactions with pure compounds (Freire et al., 2003;
Gutirrez et al., 2001a; Bjrklund-Jansson et al., 1995). On the other hand,
sterol glycosides were largely removed during the ECF bleaching and were
practically absent in these pulps. A less reactivity of these compounds with
chlorine dioxide compared to hydrogen peroxide has been reported (Nilvebrant
and Bystrm, 1995). Therefore, the complete removal of sterol glycosides in the
ECF pulps studied here may have been due to the use of hydrogen peroxide in
the ECF sequence.
The removal of lipophilic extractives in bleaching is mainly achieved by two
mechanisms: by dissolving or dispersing the pulp lipids followed by removal
with washing liquors, and by oxidation of lipids with bleaching chemicals
resulting in more hydrophilic compounds, which then may be dissolved in
bleaching liquors and subsequently removed in washing (Holmbom, 2000). The
decrease in the content of the lipophilic compounds in the TCF pulps studied
here might be due to the first mechanism as mentioned above, taking into
account the low reactivity of these lipids towards hydrogen peroxide. In
contrast, the removal of unsaturated compounds during ECF bleaching was due
to the second mechanism (Holmbom, 2000). The lower reactivity of pulp lipids
in TCF bleaching sequences using hydrogen peroxide as a bleaching agent

214
5. Resultados y discusin

compared to ECF bleaching using chlorine dioxide (Back and Allen, 2000;
Gutirrez et al., 2009) may cause pitch problems to be, in principle, more severe
in the former bleaching sequences. This is specially evident in the case of
unsaturated steroids and triterpenoids as well as unsaturated fatty acids, which
are strongly modified by chlorine dioxide but remain practically unaltered by
oxygen and hydrogen peroxide (Holmbom, 2000; Bergelin and Holmbom, 2003;
Freire et al., 2005; 2006; Gutirrez et al., 2001a). However, in the nonwoody
pulps studied here, the major lipophilic compounds present are saturated fatty
acids, fatty alcohols, alkanes, which do not show reactivity towards chlorine
dioxide, and therefore there are not great differences in the composition of the
lipophilic extractives between ECF and TCF bleached pulps, with the exception
of abaca pulp which lacks fatty alcohols and alkanes and where unsaturated
sterols are the predominant lipophilic compounds. Therefore, both ECF and TCF
bleached pulps will undergo similar pitch problems. Fatty acids, fatty alcohols
and alkanes have been reported to be the compounds responsible for pitch
deposits formed during pulping of nonwoody plants (Gutirrez and del Ro,
2005).

4. Conclusions
A thorough chemical characterization of the lipophilic extractives from different
nonwoody fibers (flax, hemp, sisal and abaca) at different stages of pulp
production (soda/AQ pulping and TCF/ECF bleaching) has been carried out.
This study provides useful information into the extent of their removal along the
cooking and bleaching processes. The soda/AQ pulping stage led to the removal
of aldehydes, hydroxyfatty acids and the complete hydrolysis of esterified
compounds such as ester waxes, sterol esters and alkyl ferulates. Among the
bleaching processes, ECF bleaching showed high reactivity towards unsaturated
sterols and both ECF and TCF bleaching were very effective in removing sterol
glycosides from nonwoody pulps. In contrast, saturated fatty acids, fatty
alcohols and alkanes, which are the main lipophilic compounds in most of the
studied fibers, survived pulping and bleaching conditions and were the
predominating compounds in both TCF and ECF bleached pulps.

Acknowledgements
This study has been supported by the Spanish projects BIO2007-28719-E and
AGL2008-00709 and the EU project BIORENEW (NMP2-CT-2006-026456).
We thank CELESA pulp mill (Tortosa, Spain) for providing the samples. G.M.
thanks the Spanish MEC for a FPI fellowship.

215
5. Resultados y discusin

References
Allen, L.H., 2000. Pitch control in pulp mills, in: Back, E. L., Allen, L. H.
(Eds.), Pitch control, wood resin and deresination. TAPPI Press, Atlanta, pp.
265-288.

Back, E.L., 2000. Deresination in pulping and washing, in: Back, E. L., Allen,
L. H. (Eds.), Pitch Control, Wood Resin and Deresination. TAPPI Press,
Atlanta, pp. 205-230.

Back, E.L., Allen, L.H., 2000. Pitch control, wood resin and deresination.
TAPPI Press, Atlanta.

Bergelin, E., Holmbom, B., 2003. Deresination of birch kraft pulp in bleaching.
J. Pulp Pap. Sci. 29, 29-34.

Bergelin, E., Holmbom, B., 2008. Reactions and distribution of birch extractives
in kraft pulp oxygen delignification. J. Wood Chem. Technol. 28, 261-269.

Bergelin, E., Mller, R., Holmbom, B., 2005. Analysis of pitch and deposit
samples in kraft pulp production. Pap. Puu-Pap. Tim. 87, 399-403.

Bjrklund-Jansson, M., Wormald, P., Dahlman, O., 1995. Reactions of wood


extractives during ECF and TCF bleaching of kraft pulp. Pulp Pap. Can. 96,
T134-T137.

Chen, T., Wang, Z., Zhou, Y., Breuil, C., Aschim, O.K., Yee, E., Nadeau, L.,
1995. Using solid-phase extraction to assess why aspen causes more pitch
problems than softwoods in kraft pulping. Tappi J. 78, 143-149.

del Ro, J.C., Gutirrez, A., 2006. Chemical composition of abaca (Musa
textilis) leaf fibers used for manufacturing of high quality paper pulps. J. Agr.
Food Chem. 54, 4600-4610.

del Ro, J.C., Gutirrez, A., Gonzlez-Vila, F.J., Martn, F., Romero, J., 1998.
Characterization of organic deposits produced in the kraft pulping of
Eucalyptus globulus wood. J. Chromatogr. A 823, 457-465.

del Ro, J.C., Rodrguez, I.M., Gutirrez, A., 2004. Identification of intact long-
chain p-hydroxycinnamate esters in leaf fibers of abaca (Musa textilis) using
gas chromatography/mass spectrometry. Rapid Commun. Mass Sp. 18, 2691-
2696.

216
5. Resultados y discusin

del Ro, J.C., Romero, J., Gutirrez, A., 2000. Analysis of pitch deposits
produced in kraft pulp mills using a totally chlorine free bleaching sequence.
J. Chromatogr. A 874, 235-245.

Freire, C.S.R., Silvestre, A.J.D., Neto, C.P., 2003. Oxidized derivatives of


lipophilic extractives formed during hardwood kraft pulp bleaching.
Holzforschung 57, 503-512.

Freire, C.S.R., Silvestre, A.J.D., Neto, C.P., 2005. Lipophilic extractives in


Eucalyptus globulus kraft pulps. Behavior during ECF bleaching. J. Wood
Chem. Technol. 25, 67-80.

Freire, C.S.R., Silvestre, A.J.D., Neto, C.P., Evtuguin, D.V., 2006. Effect of
oxygen, ozone and hydrogen peroxide bleaching stages on the contents and
composition of extractives of Eucalyptus globulus kraft pulps. Bioresource
Technol. 97, 420-428.

Gutirrez, A., del Ro, J.C., 2001. Gas chromatography-mass spectrometry


demonstration of steryl glycosides in eucalypt wood, kraft pulp and process
liquids. Rapid Commun. Mass Sp. 15, 2515-2520.

Gutirrez, A., del Ro, J.C., 2003a. Lipids from flax fibers and their fate in
alkaline pulping. J. Agr. Food Chem. 51, 4965-4971.

Gutirrez, A., del Ro, J.C., 2003b. Lipids from flax fibers and their fate in
alkaline pulping (Vol 51, pg 4965, 2003). J. Agr. Food Chem. 51, 6911-6914.

Gutirrez, A., del Ro, J.C., 2005. Chemical characterization of pitch deposits
produced in the manufacturing of high-quality paper pulps from hemp fibers.
Bioresource Technol. 96, 1445-1450.

Gutirrez, A., del Ro, J.C., Gonzlez-Vila, F.J., Martn, F., 1998. Analysis of
lipophilic extractives from wood and pitch deposits by solid-phase extraction
and gas chromatography. J. Chromatogr. A 823, 449-455.

Gutirrez, A., del Ro, J.C., Ibarra, D., Rencoret, J., Romero, J., Speranza, M.,
Camarero, S., Martnez, M.J., Martnez, A.T., 2006. Enzymatic removal of
free and conjugated sterols forming pitch deposits in environmentally sound
bleaching of eucalypt paper pulp. Environ. Sci. Technol. 40, 3416-3422.

Gutirrez, A., del Ro, J.C., Martnez, A.T., 2009. Microbial and enzymatic
control of pitch in the pulp and paper industry. Appl. Microbiol. Biot. 82,
1005-1018.

217
5. Resultados y discusin

Gutirrez, A., Rodrguez, I.M., del Ro, J.C., 2004. Chemical characterization of
lignin and lipid fractions in kenaf bast fibers used for manufacturing high-
quality papers. J. Agr. Food Chem. 52, 4764-4773.

Gutirrez, A., Rodrguez, I.M., del Ro, J.C., 2006. Chemical characterization of
lignin and lipid fractions in industrial hemp bast fibers used for manufacturing
high-quality paper pulps. J. Agr. Food Chem. 54, 2138-2144.

Gutirrez, A., Rodrguez, I.M., del Ro, J.C., 2008. Chemical composition of
lipophilic extractives from sisal (Agave sisalana) fibers. Ind. Crop. Prod. 28,
81-87.

Gutirrez, A., Romero, J., del Ro, J.C., 2001a. Lipophilic extractives from
Eucalyptus globulus pulp during kraft cooking followed by TCF and ECF
bleaching. Holzforschung 55, 260-264.

Gutirrez, A., Romero, J., del Ro, J.C., 2001b. Lipophilic extractives in process
waters during manufacturing of totally chlorine free kraft pulp from eucalypt
wood. Chemosphere 44, 1237-1242.

Hillis, W.E., Sumimoto, M., 1989. Effect of extractives on pulping, in: Rowe, J.
W. (ed.), Natural products of woody plants. II. Springer-Verlag, Berlin, pp.
880-920.

Holmbom, B., 2000. Resin reactions and deresination in bleaching, in: Back, E.
L., Allen, L. H. (Eds.), Pitch control, wood resin and deresination. Tappi
Press, Atlanta, pp. 231-244.

Nilvebrant, N.-O., Bystrm, S., 1995. Demonstration of glucosidic linked sterols


in birch. Proc. 8th Intern. Symp. Wood and Pulping Chem., Helsinki, 6-9 June
2, 135-140.

Shin, S.J., Kim, C.H., 2006. Residual extractives in unbleached aspen and pine
kraft pulps and their fate on oxygen delignification. Nord. Pulp Pap. Res. J.
21, 260-263.

Silvestre, A.J.D., Pereira, C.C.L., Neto, C.P., Evtuguin, D.V., Duarte, A.C.,
Cavaleiro, J.A.S., Furtado, F.P., 1999. 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.
52, 375-382.

218
5. Resultados y discusin

Sjstrm, E., 1993. Wood chemistry. Fundamentals and applications. Academic


Press, San Diego.

219
5. Resultados y discusin

Publicacin VIII:

Marques G., Gutirrez A., del Ro J.C. and Evtuguin D.V. (2010) Acetylated
heteroxylan from Agave sisalana and its behavior in alkaline pulping and
TCF/ECF bleaching. Carbohydrate Polymers, 81, 517-523.

220
5. Resultados y discusin

Acetylated heteroxylan from Agave sisalana and its behavior in alkaline


pulping and TCF/ECF bleaching
Gisela Marquesa, Ana Gutirreza, Jos C. del Roa and Dmitry V. Evtuguinb
a
Instituto de Recursos Naturales y Agrobiologa de Sevilla, CSIC, P.O. Box 1052, 41080-
Seville, Spain
b
CICECO and University of Aveiro, Department of Chemistry,
3810-193 Aveiro, Portugal

Abstract
The heteroxylan from sisal (Agave sisalana), an O-acetyl-(4-O-
methylglucurono)xylan with a molecular weight (Mw) of 18 kDa, was isolated
by extraction of peracetic holocellulose with Me2SO and thoroughly
characterized by wet chemistry, and NMR spectroscopy. The heteroxylan
backbone is composed of (14)-linked -D-xylopyranosyl units (Xylp)
partially branched with terminal (12)-linked 4-O-methyl--D-glucuronosyl
(MeGlcpA, 9%mol.) and a small proportion of -D-glucuronosyl (GlcpA,
<1%mol.) residues. Roughly 61%mol. of Xylp residues are acetylated (DS
=0.70). During soda/AQ pulping of sisal fibers, MeGlcpA and GlcpA are mostly
removed or converted to 4-deoxy--L-threo-hex-4-enopyranosyluronic acid
(HexA), although about 15% of the initially present MeGlcpA was maintained
intact upon cooking. The major part of acetyl groups (95%) was hydrolyzed
during pulping. It was proposed that during bleaching, a low molecular weight
xylan fraction associated to residual lignin was removed from pulp and small
proportion of MeGlcpA was additionally converted to HexA. The profiles of
uronosyl residues in xylans from TCF and ECF bleached sisal pulps were rather
different.
Keywords: Agave sisalana; soda pulping; bleaching; black liquor; xylan;
structural analysis; NMR spectroscopy

1. Introduction
Sisal fibers are hard fibrous material isolated from the leaves of the sisal plant
(Agave sisalana), a monocotyledonous plant from the Agavaceae family (Li et
al. 2000; Gutirrez et al. 2008). Originally from Central America and Mexico,
sisal plant is widely cultivated in South America (e.g. Brazil), Australia and
Africa (e.g. Kenya) (Gutirrez et al, 2008; Mwaikambo, 2006). The sisal fibers
find out numerous applications in the manufacture of ropes for boats, goods for
the agricultural industry and for the reinforcement of polymeric matrices (Li et
al., 2000; Gutirrez et al., 2008; Mwaikambo, 2006; Megiatto et al., 2007). Sisal
cellulosic pulp possesses such characteristics as high tear resistance, alpha
cellulose content, porosity, bulk, moister absorbency and high folding
endurance, that offer unique opportunities for the papermaking (Gutirrez et al.,

221
5. Resultados y discusin

2008; Mwaikambo, 2006; Megiatto et al., 2007; Hurter, 2001; Idrraga et al.,
2001). Easily bleachable sisal chemical pulp is industrially produced by soda
pulping in the presence of athroquinone (AQ) as catalyst.
The basic knowledge of the chemical composition of sisal fibers, as well as
the behavior of its components during pulping and bleaching, is essential for the
better understanding and improving the pulping and bleaching operations and
for the assessment of pulp and paper properties. Previous papers have reported
the composition of the lipophilic compounds (Gutirrez et al., 2008) and the
structure of the lignin (del Ro et al., 2007, 2008) of sisal fibers, but only limited
work has been performed on the structural characterization of the carbohydrate
fraction of this fiber (Das Gupta & Mukherjee, 1967; Stewart et al., 1997;
Megiatto et al., 2007). In the present work, we report the structural
characterization of a heteroxylan in sisal fibers, as well as in their soda/AQ
pulps (unbleached and TCF/ECF bleached). The study of hemicelluloses is of
fundamental and practical interest, since their partial degradation and dissolution
during pulping is responsible for significant consumption of pulping chemicals,
the decrease of pulp yield and the papermaking properties of bleached pulps
(Evtuguin et al., 2003; Pinto et al., 2005; Lisboa et al., 2004).
Hemicelluloses provide a supporting function to the cell wall being bounded
to cellulose fibrils. Hemicelluloses are mainly branched polymers of low
molecular weight (DP  80-200) and are composed by diverse sugar residues
(D-xylose, L-arabinose, D-glucose, D-galactose, D-mannose, D-glucuronic acid,
4-O-methyl-D-glucuronic acid, D-galacturonic acid, and to a lesser extent, L-
rhamnose, L-fucose, and various O-methylated neutral sugars) (Shimizu, 1991;
Sun et al., 2000; Ebringerov et al., 2005). In particular, glucuronoxylan (GX) is
the major hemicellulose in such important industrial crops produced by agro-
industry as kenaf, bamboo, flax, sisal and jute and is structurally similar to
hardwood xylans (Rowell et al., 1998; Gorshkova et al., 1996; Neto et al., 1996;
Vignon & Gey, 1997; Stewart et al., 1997). Among the above mentioned plants,
GX of sisal is one of the less investigated. Previous structural/compositional
studies of GX from sisal were carried out mainly with alkali-extracted GX
(Stewart et al., 1997; Das Gupta & Mukherjee, 1967). Hence, some essential
structural information, such as substitution patterns with acetyl groups, was not
assessed. It was suggested, however, that sisal GX is an O-acetyl-4-O-
methylglucuronoxylan with moderate molecular weight, around 15-20 kDa
(Stewart et al., 1997; Das Gupta & Mukherjee, 1967). According to data from
methylation analysis of sisal xylan, its backbone is constituted by E-(1-4)-linked
D-xylopyranose residues, carrying a low degree of substitution (8-10 % mol.)
with terminal 4-O-methyl-D-glucopyranosyluronic acid residue linked through
the O-2 position (Das Gupta & Mukherjee, 1967).
In this report, the heteroxylan from sisal fibers was isolated by extraction of
peracetic holocellulose with dimethyl sulfoxide and thoroughly characterized by
wet chemistry and NMR spectroscopy. This isolation procedure allowed to

222
5. Resultados y discusin

obtain a xylan sample with intact O-acetyl moieties. Simultaneously, the fate of
this heteroxylan structure during soda/anthraquinone pulping and TCF and ECF
bleaching was also studied.

2. Experimental
2.1. Samples
Sisal (Agave sisalana) leaf fibers from Africa (Kenya), their soda/AQ chemical
pulps (unbleached and ECF/TCF bleached) and cooking black liquor were
supplied by CELESA pulp mill (Tortosa, Spain). The TCF bleaching sequence
E(O)P-EP included two hydrogen peroxide stages at 90C, the first under
oxygen pressure (E(O)P stage) and the second without oxygen (EP stage). The
ECF bleaching sequence D-EP included a chlorine dioxide (D) at 60C and a
hydrogen peroxide (EP) stages at 90C.
The samples were air-dried, milled using a knife mill (Janke and Kunkel,
Analysenmhle), and extracted with acetone in a Soxhlet apparatus for 8 h. For
estimation of the Klason lignin content, the acetone extracted samples were
subsequently extracted with hot water (3 h at 100 C). Klason lignin was
estimated as the residue after 72% sulfuric acid hydrolysis of the pre-extracted
material according to Tappi standard procedures (Tappi Test Methods, 1993).
Ash content was estimated as the residue after 6 h calcination at 575 C.

2.2. Preparation of holocellulose


The holocelluloses of sisal fibers and their unbleached pulp were obtained from
extractives-free sawdust (5.0 g) by delignification with 10% peracetic acid at pH
3.5 for 20 min at 85 C. After delignification, the holocellulose was filtered off
on a porous glass filter, washed with acetone and further with warm water and
air-dried. The holocellulose yield was 69.7% and 94.7% for the sisal fibers and
unbleached pulp, respectively.

2.3. Isolation of xylans from pulps


The isolation of acidic heteroxylans was carried out by two consecutive
extractions with Me2SO (1.5 g of holocelluloses with 50 mL of Me2SO in each
assay) at 50 C for 24 h under stirring and further precipitation of the resulted
extracts in an excess of 7:2:1 EtOH-MeOH-water acidified by HCOOH. The
complete precipitation of the heteroxylan was accomplished in 3 days at 4 C.
The heteroxylan was isolated by centrifugation, washed four times with
anhydrous MeOH and quickly dried under vacuum at room temperature. For the
ECF and TCF bleached pulps the xylans were extracted directly from the pulp
without previous preparation of holocelluloses.

223
5. Resultados y discusin

2.4. Isolation of xylan from the black liquor


The xylan of the black liquor was isolated following a procedure previously
described, with minor modifications (Engstrm et al, 1995). 200 ml of 1,4-
dioxane were slowly added with agitation to 100 ml of diluted with distilled
water 1/2 black liquor, followed by the addition of glacial acetic acid until pH 2-
3. The solution and resulting precipitate was kept at 5 C during 2 days. The
black liquor precipitated polysaccharides (BLPS) were separated by
centrifugation and the solution decanted off. The precipitate was sequentially
washed up with 150 ml of a 1,4-dioxane-water (2:1) solution, 150 ml of 1,4-
dioxane, 150 ml of methanol and 150 ml of acetone and, finally, dried under
vacuum with phosphorus pentoxide.

2.5. Carbohydrate analysis


The heteroxylan was subjected to hydrolysis with 72% H2SO4 at 20 C for 3 h,
followed by 2.5 h hydrolysis with diluted 1 M H2SO4 at 100 C (Saeman
hydrolysis) and the released neutral monosaccharides were determined as alditol
acetate derivatives by gas chromatography (Selvendran et al., 1979). The
quantitative analysis was carried out on a Varian 3350 gas chromatograph
equipped with a FID detector and with a DB-225 J&W column (30 m0.25 mm
i.d. 0.15 m film thickness). The temperature program was started at 220 C
with a 5 min hold, and then raised to a final temperature of 230 C at 2 C/min,
and held for 5 min. The injector and detector temperatures were set at 230 C.
The quantification was made using calibration curves with standards.

2.6. Acid methanolysis for analysis of sugars and uronic acids


About 4.5 mg of freeze-dried sisal fibers and their pulps (unbleached, ECF and
TCF bleached pulps) were subjected to acid methanolysis by the addition of 2
mL, 2 M solution of HCl in anhydrous methanol at 100 C for 4 hours
(Sundberg et al., 1996). After cooling to room temperature, about 80 L of
pyridine was added to neutralize the acidic solution. Additionally, 1 mL of
internal standard solution containing 0.1 mg/mL of sorbitol was added. To avoid
fibers silylation, 2 mL of the supernatant reaction solution was separated from
fiber suspension and evaporated in a rotary evaporator with a water bath kept at
40-50 C. The samples were dissolved by addition of 70 L pyridine. For
silylation, 150 L hexamethyldisiloxane and 80 L trimethylchlorosilane were
added and the samples were shaken well. After 12 hours at room temperature,
the samples were ready for analysis. GC-MS analysis were performed using a
Hewlet-Packard Gas Chromatograph 5890 equipped with a mass selective
detector MSD series II, using helium as carrier gas (35 cm/s), equipped with a
DB-1 J&W capillary column (30 m0.32 mm i.d. 0.25 m film thickness). The
column temperature program was 100 4 C/min 175 C followed by 175 12
C/min 290 C. The detector (FID) temperature was 290 C. The different

224
5. Resultados y discusin

peaks were identified by comparing their mass spectra with mass spectra in
Wiley and NIST libraries and that reported in the literature (Sundberg et al.,
1996; Bertaud et al., 2002; Bleton et al., 1996).

2.7. Size-exclusion chromatography (SEC)


The xylan samples were dissolved in a small amount of 10% LiCL solution in
N,N-dimethylacetamide (DMAC) at 70-80 C and further diluted with DMAC to
a xylan concentration of about 0.5% (5mg/mL). The SEC analysis has been
carried out on two PLgel 10 m MIXED B 300 7.5 mm columns protected by
a PLgel 10 m pre-column (Polymer Laboratories, UK) using a PL-GPC 110
system (Polymer Laboratories). The columns, injector system and the detector
(RI) were maintained at 70 C during the analysis. The eluent (0.1 M LiCl
solution in DMAC) was pumped at a flow rate of 0.9 mL/min. The analytical
columns were calibrated with pollulan standards (Polymer Laboratories) in the
range 0.8-100 kDa. The injected volume was 100 L.

2.8. NMR spectroscopy


One-dimensional 1H NMR spectra of the xylan samples were recorded in D2O
(30 C) on a Bruker Avance 300 spectrometer operating at 300.13 MHz. Sodium
3-(trimethylsilyl)-propionate-d4 was used as internal standard ( 0.00). The
relaxation delay was 16 s, r.f. 90-pulse width of 10.2 s and about 400 pulses
were collected. All 2D NMR spectra were recorded on a Bruker Avance 300
spectrometer operating at 300.1 MHz for proton and at 75.2 MHz for carbon.
2D 1H1H COSY spectroscopy was performed at 50 C using a standard COSY
sequence (90 pulse, relaxation delay 2 s). Two-dimensional 1H-1H TOCSY
(Total Correlation Spectroscopy) spectra (
mix= 0.050 s) were acquired at a
spectral width of 2185 Hz in both dimensions at 60 C. The relaxation delay was
2.0 s. For each FID, 128 transients were acquired; the data size was 1024 in t1
512 in t2. The phase sensitive 1H-detected HSQC (Heteronuclear Single
Quantum Coherence) spectrum was acquired at 50C over a F1 spectral weight
of 12,000 Hz and a F2 width of 2000 Hz with a 2048 1024 matrix and 128
transients per increment. The delay between scans was 2 s and the delay for
polarization transfer was optimized for 1JCH= 148 Hz.

2.9. Hexenuronic acid content


The amount of hexenuronic acids (HexA) was determined by acidic hydrolysis
in sodium formate buffer at pH 3.0 followed by UV detection of furan
derivatives at 245 nm (Vuorinen et al. 1999).

225
5. Resultados y discusin

3. Results and discussion


3.1. Chemical composition of sisal fibers
The chemical composition of sisal (Agave sisalana) fibers is presented in Table
1. The sugar analysis confirmed the data previously reported by Stewart et al.
(1997) indicating that xylan is the principal hemicellulose of sisal fibers and the
second most abundant polysaccharide after cellulose. At the same time, taking
into account the small amount of lignin (around 6%) and extractives (slightly
more than 3%) in sisal fibers, the relatively low yield of peracetic holocellulose
(70%) indicates a significant content of easily removable hemicelluloses, other
than xylan. These may be pectins and, in particular, glucans that are known to be
present in the leaves of the genus Agave in noticeable amounts (Nobel, 2003).
The misbalance in cellulose content and the amount of glucose detected upon
sugars analysis (Table 1) allows suggesting that non-cellulosic glucans may
contribute to at least 15% (w/w) of sisal fibers. This fact was further confirmed
by analysis of the hemicelluloses dissolved in the black liquor from soda pulping
of sisal fibers.

3.2. Isolation and structural characterization of xylan from sisal fibers


The heteroxylan (yield of about 60% w/w) from sisal fibers was isolated from
peracetic holocellulose by two consecutive extractions with Me2SO followed by
precipitation of the extracted polyose in 7:2:1 ethanol/methanol/water. Such
procedure guaranteed the isolation of intact and representative xylan sample,
which can be structurally characterized including the quantification and
distribution of O-acetyl moieties (Evtuguin et al. 2003).

Table 1. Chemical composition of sisal fibers, unbleached soda pulp and ECF and TCF
bleached pulps (% w/w).
Component Sisal fibers Unbleached pulp TCF pulp ECF pulp
Ash 1.0 1.0 0.4 0.4
Extractives (acetone) 0.8 0.3 0.1 0.1
Extractives (water) 2.3 0.7 0.6 0.4
Klason lignin 5.9 0.7 - -
Holocellulose 70.0 95.0 - -
Cellulose* 54.5 - - -

Neutral sugars
Rha 0.7 0.7 tr tr
Ara 1.3 tr tr tr
Xyl 20.0 19.0 19.4 20.6
Man 0.8 - - -
Gal 1.0 tr tr tr
Fuc <0.5 - - -
Glc 75.7 80.4 80.6 79.4
* Krschner-Hoffer method of determination; tr: traces;

226
5. Resultados y discusin

The composition of the isolated xylan was assessed by analyses of neutral


sugars and easily hydrolyzed sugars after methanolysis (Tables 2 and 3). The
high purity of the isolated xylan was confirmed using neutral monosaccharides
analysis, that showed the predominance of xylose (Xyl) and only small amounts
of glucose (Glc), galactose (Gal), arabinose (Ara) and rhamnose (Rha) (Table
2). The presence of Gal, Ara and Rham may indicate the eventual small
contamination of xylan with pectin compounds. This was confirmed by
methanolysis studies (Table 3, Figure 1), which revealed a much higher amount
of galacturonic acid than could be expected if arisen only from the terminal
structural fragment [3)-D-L-Rhap-(12)-D-D-GalpA-(14)-D-Xylp]
suggested to be present in xylans (Shimizu, 1991). The presence of D-
glucopyranosyluronic acid (GlcpA), besides the expected 4-O-methyl-D-
glucopyranosyluronic acid (MeGlcpA), may indicate that, at least part of
glucuronosyl moieties attached to xylan backbone, is not methylated. The ratio
between internal xylopyranosyl units (Xylp) in the backbone and terminal
attached glucuronosyl residues (MeGlcpA and GlcpA) was estimated to be
around 9:1. The molecular weight (Mw) of sisal xylan was about 18 kDa, as
revealed by SEC analysis (Figure 2).

Table 2. Neutral monosaccharide composition (% w/w) of xylans isolated from sisal fibers,
pulps and black liquor.
Rha Ara Xyl Man Fuc Gal Glc
Sisal fibers 0.9 0.7 93.7 tr - 1.2 3.5
Unbleached pulp tr tr 99.5 - - tr 0.5
TCF bleached pulp tr tr 99.0 - - tr 1.0
ECF bleached pulp tr tr 98.7 - - tr 1.3
Black liquor precipitated polysaccharides
(BLPS) 0.7 3.6 10.9 2.6 1.1 27.1 54.0
tr: traces

According to previously reported methylation analysis of alkali extracted


sisal xylan, its backbone is constituted by E-(14)-linked D-xylopyranose
residues branched at O-2 with terminal 4-O-methyl-D-glucopyranosyluronic
acid residues (Das Gupta & Mukherjee, 1967). These structural features were
confirmed by 1D and 2D NMR techniques. Single (COSY) and multiple
(TOCSY) bonds 1H-1H correlation analyses and 1H-13C (HSQC) correlations
(Figure 3) allowed assignment of proton and carbon signals in sisal heteroxylan,
as summarized in Table 4. Chemical shifts of protons and carbons were
practically the same as those reported for acetylated heteroxylans from other
plant sources (Teleman et al. 2002; Evtuguin et al. 2003). The anomeric region
in the TOCSY spectrum (Figure 4) revealed the characteristic proton
correlations that are normally found in heteroxylans containing attached to
backbone non-methylated GlcpA residues (Vignon & Gey, 1997; Gonalves et
al. 2008). Hence, NMR results corroborates the data obtained by methanolysis

227
5. Resultados y discusin

Table 3. Neutral monosaccharide and uronic acids composition (% w/w) determined by


methanolysis of xylans isolated from sisal fibers and their pulps.
Rha Ara Xyl Man Gal Glc GalA GlcA MeGlcpA
Sisal fibers 0.8 0.6 83.0 - 0.8 1.3 2.8 0.3 10.5
Unbleached pulp - - 98.3 - - 0.2 - - 1.5
TCF bleached pulp - - 98.0 - - 0.8 - - 1.2
ECF bleached pulp - - 97.5 - - 1.2 - - 1.3

(Table 3), evidencing that a small proportion of glucuronic residues in sisal


heteroxylan are not methylated. Therefore, it can be suggested that the backbone
of sisal heteroxylan is composed of partially acetylated (14)-linked -D-Xylp
units O-2 ramified with terminal (12)-linked MeGlcpA and GlcpA.

 Xyl
SISAL FIBERS
GalA
Gal

4-O-MeGLCA
GlcA

Xyl
Ara

Xyl
GalA

GalA
Ara

IS Dimers
GalA
Ara

Rha

Glc
Gal

Glc
Gal

Xyl

SISAL UNBLEACHED PULP

SISAL TCF PULP

SISAL ECF PULP

8 10 12 14 16 18 20 22 24 26

Retention Time (min)

Figure 1. Gas chromatogram of methylated and silylated sugars obtained by acid


methanolysis of xylans isolated from sisal fibers and their pulps. Xyl: xylose, Gal: galactose,
Glc: Glucose, Rha: rhamnose, GlcA: glucuronic acid, 4-O-MeGlcA: 4-O-methylglucuronic
acid, GalA: galacturonic acid.

228
5. Resultados y discusin

Table 4. Proton/carbon chemical shifts ( ) of heteroxylan from sisal (30C, D2O).


Structural unit Assignments
H1/C1 H2/C2 H3/C3 H4/C4 H5/C5
ax eq
Xyl (isol) 4.47/102.9 3.28/73.7 3.55/74.5 3.80/77.4 3.40/63.9 4.10/63.9
Xyl-3Ac 4.57/102.4 3.49/71.9 4.98/76.3 3.93/76.5 3.47/63.8 n.d/63.8
Xyl-2Ac 4.68/101.2 4.68/74.6 3.80/72.5 3.87/77.2 3.45/63.8 n.d/63.8
Xyl-2,3Ac 4.80/100.5 4.82/74.6 5.16/74.0 4.06/76.6 3.54/63.9 n.d/63.9
Xyl-3Ac-2GlcA 4.73/102.1 3.70/75.8 5.06/75.8 3.98/77.2 3.50/n.d n.d/n.d
MeGlcA 5.27/98.9 3.57/72.3 3.88/73.7 3.27/83.2 n.d/n.d _a
GlcA 5.25/n.d 3.60/n.d 3.82/n.d 3.52/n.d n.d/n.d _a
n.d
non determined, anot relevant. The following designations were used: Xyl (isol.), non-acetylated Xylp in the
backbone isolated from other acetylated Xylp units; Xyl-3Ac, 3-O-acetylated Xylp; Xyl-2Ac, 2-O-acetylated
Xylp; Xyl-2,3Ac, 2,3-di-O-acetylated Xylp; Xyl-3Ac-2GlcA, MeGlcA 2-O-linked and 3-O-acetylated Xylp;
MeGlcA, 2-O-linked MeGlcpA; GlcA, 2-O-linked GlcpA.

The acetylation patterns in the heteroxylan backbone were assessed by 1H


NMR spectroscopy based on signal assignment employing 2D NMR techniques
(Table 4). The total 1H NMR spectrum of sisal xylan and its expanded anomeric
region with specified groups of protons in particular substructures, are presented
in Figure 5. According to previously published methodology (Evtuguin et al.
2003), internal non-acetylated, 3-O- and 2-O-acetylated xylose residues,
MeGlcpA residues were assessed based on their anomeric proton resonances,
whereas the amounts of 2,3-di-O-acetylated and 3-O-acetylated/ MeGlcpA O-2
substituted internal xylose residues were estimated based on H-3 resonances in
corresponding structures (Figure 5). This allowed the integration of protons
from particular structural fragments and their quantification (Table 5). Around
61%mol. of the Xylp residues were acetylated; among them, 39 %mol.
corresponded to 3-O-acetylated, 13% mol. corresponded to 2-O-acetylated and


1200

960
Response

720
14 kDa (D)

480 12 kDa (C)

240 (B)
10 kDa
18 kDa (A)
0

11.2 12.3 13.5 14.6 15.7

Retention time (min)

Figure 2. The GPC elution curves of xylans isolated from sisal fibers and their pulps. (A)
sisal fibers, (B) unbleached pulp, (C) TCF bleached pulp, (D) ECF bleached pulp.

229
5. Resultados y discusin

9 %mol. corresponded to 2,3-di-O-acetylated residues. Accordingly, sisal


heteroxylan possessed a substitution degree with acetyl groups of 0.70. Worth
notably, Xylp units in backbone of sisal heteroxylan are predominantly 3-O-
acetylated. The proportion of 3-O-acetylated Xylp units in backbone was much
higher (almost twice) in sisal xylan than in xylans from woody sources such as
birch and beech (Teleman et al. 2002), eucalypt (Evtuguin et al. 2003),
paulownia (Gonalves et al. 2008) and aspen (Teleman et al. 2000). Almost all
Xylp linked at O-2 with MeGlcpA (9 mol%) were 3-O-acetylated (Table 5).

3.3. Changes in xylan structure during alkaline pulping


During the soda/AQ pulping, around 40% of the xylan was dissolved in the
black liquor. This conclusion was made based on the xylose content in
delignified unbleached sisal fibers (Table 1) and the pulp yields (ca. 60%). The
chemical changes in the xylan during pulping were examined comparing the
composition and structural features of the xylan from sisal fibers and their soda
pulp, both isolated by Me2SO extraction of the corresponding peracetic
holocelluloses (Tables 2 and 3, Figure 6).
The molecular weight of the xylan remaining in the pulp decreased to 10 kDa,
reflecting significant alkali-induced depolymerization (Figure 2). The xylan
suffered also a significant deacetylation (about 95%) and the major part of the
uronic moieties (at least 75%) were converted to 4-deoxy--L-threo-hex-4-
enopyranosyluronic acid (hexenuronic acid or HexA), as revealed from 1H NMR


X5Ac3MeGlcA -OCH3 in MeGlcA

X5 60
MeGlcA2
MeGlcA3 X3Ac2 X5 X2Ac3
70
X2Ac2
X3Ac3MeGlcA
X4Ac3MeGlcA X3 X2
X3Ac3 X4 80
X4Ac2

X2Ac3MeGlcA
MeGlcA4 90
X1Ac2
MeGlcA1
X1Ac3 100
X1Ac3MeGlcA
X1

5.5 5.0 4.5 4.0 3.5 ppm

Figure 3. HSQC spectrum (D2O, 50 C) of heteroxylan from sisal fibers.

230
5. Resultados y discusin

Table 5. Relative content in acetyl groups in structural units of sisal heteroxylan.


Relative abundance
Structural fragment and short designation (per 100 Xylp units)
4)--D-Xylp-(1 (Xyl) 39
4)[3-O-Ac]--D-Xylp-(1 (Xyl-3Ac) 30
4)[2-O-Ac]--D-Xylp-(1 (Xyl-2Ac) 13
4)[3-O-Ac][2-O-Ac]--D-Xylp-(1 (Xyl-2,3Ac) 9
4)[4-O-Me--D-GlcpA-(12)][3-O-Ac]--D-Xylp-
-(1 (Xyl-3Ac-2MeGlcA) 9
4-O-Me--D-GlcpA-(1 (MeGlcA) 9

spectra (Figure 6). The presence of HexA, formed under alkaline pulping
conditions via E-elimination of methoxyl group, was confirmed applying the
total correlation spectroscopy (TOCSY), according to previously published
proton chemical shifts (Teleman et al. 1995). The HexA residues may be easily
detected in the anomeric region of 1H NMR spectra, which showed the
appearance of new signals at 5.36 and at 5.82 ppm that were assigned to H-1 and
H-4 in corresponding structures (Figure 6). The HexA content in sisal soda pulp

 III
I
IV V IV
II

H1/H4 H3/H5
H1/H2 H1/H4
H3/H2 3.5
H1/H2
H1/H3 H3/H4

4.0

H3/H1 4.5
H3/H1 H1
H3/H1

H3 H1/H3 5.0
H1 H3
H3

5.5
5.5 5.4 5.3 5.2 5.1 5.0 4.9 4.8 ppm

I- MeGlcpA-(1 IV, IV- 4)[3-O-Ac][2-O-Ac]--D-Xylp-(1

II- GlcpA-(1 V- 4)[4-O-Me--D-GlcpA-(12][3-O-Ac]--D-Xylp-(1

III- 4)[3-O-Ac]--D-Xylp-(1
Figure 4. Anomeric region of the TOCSY spectrum (D2O, 60 C) of heteroxylan from sisal
fibers

231
5. Resultados y discusin

was 60.6 mmol/kg of pulp as determined after pulp treatment under acidic
conditions followed by detection of furoic acid derivatives by UV-spectroscopy
at 245 nm (Vuorinen et al. 1999).
The balance of uronic moieties in the initial xylan and in the xylan remaining
in the pulp was estimated based on the ratio of anomeric protons in uronic
groups at 5.27 ppm and in internal xylopyranose units at 4.47 ppm (pulp xylan)
or at 4.40-4.65 ppm (fiber xylan). This analysis indicated a removal of about
30% of all uronic units (MeGlcpA and HexA) from xylan during pulping.
The polysaccharides dissolved in the black liquor (BLPS) during pulping
were isolated according to a previously published procedure (Engstrm et al,
1995) and chemically characterized (Table 2). The aim of this study was to
compare the structure of the xylan remaining in the pulp with that dissolved in
the pulping liquor. Surprisingly, the heteroxylan was a minor polysaccharide
dissolved in the liquor and its purification by fractional precipitation failed. At
the same time, the analysis of neutral sugars of BLPS revealed glucans as the
major precipitated polysaccharides (glucose represents around 54 % of BLPS
weight) followed by galactans. The preliminary study on BLPS using multiple
bonds 1H-1H correlation NMR spectroscopy (TOCSY) gave additional insights


H2O CH3-CO-
*

5.5 5.0 4.5 4.0 3.5 3.0 2.5 ppm

H1/H2
Xyl-2Ac H1
H1
H3 Xyl-3Ac
Xyl-3Ac Xyl

H1 H3 Xyl-
MeGlcA 3Ac-2GlcA
H3
Xyl-2,3Ac

5.5 5.4 5.3 5.2 5.1 5.0 4.9 4.8 4.7 4.6 4.5 4.4
ppm

Figure 5. 1H NMR spectrum (D2O, 30 C) of heteroxylans from sisal fibers (top image) and
the expanded region of anomeric protons (bottom image). The designations for the structural
fragments are the same as in Table 5. * solvent impurities.

232
5. Resultados y discusin

SISAL UNBLEACHED PULP

H1 Xyl

H1
MeGlcA

H1
HexA
H4 HexA

SISAL TCF PULP

SISAL ECF PULP

5.9 5.7 5.5 5.3 5.1 4.9 4.7 4.5 4.3


ppm

Figure 6. 1H NMR spectra (D2O, 30 C ) of heteroxylans isolated from sisal unbleached pulp
and TCF and ECF bleached pulps.

into the type of glucans dissolved during pulping from sisal fibers (Figure 7).
These were suggested to be mixture of E-glucans, in particular, E-(13)-
glucans with a low degree of ramification at C6 (callose type), by comparison of
the proton-proton correlations with previously published data on proton
resonances in E-(13)-glucan (Torosantucci et al. 2005). However, a more
detailed study is required to elucidate the exact structure of the E-glucans in sisal
fibers.

3.4. Changes in xylan structure during TCF and ECF bleaching


The structural changes in the heteroxylan from sisal soda pulp during industrial
TCF and ECF bleaching were also investigated. The chemical composition and
structural features were assessed in xylan samples isolated directly from
bleached pulps by Me2SO extraction. The TCF pulp was obtained by E(O)P-EP
bleaching and the ECF pulp was obtained by D-EP sequence. TCF pulp was
bleached essentially by hydrogen peroxide under alkaline conditions and
included two hydrogen peroxide stages, the first with oxygen and the second
without oxygen, at 90 C,, whereas ECF pulp bleaching included a treatment
with chlorine dioxide (D) at 60C followed by hydrogen peroxide stage under
alkaline conditions (EP) at 90 C.

233
5. Resultados y discusin

 OH
OH H
H
H O
H O
HO O
HO O
O OH
OH H
H
H H
H6ax H5
H1 H4
H3 H2

3.50

4.00

4.50

ppm (t1)

4.50 4.00 3.50


ppm (t2)

Figure 7. TOCSY spectrum (D2O, 60 C) of BLPS fraction isolated from black liquor. Top
image represents the fragment of E-(1-3)-D-glucan backbone.

The chemical analysis of the sugars of TCF and ECF bleached pulps did not
show significant changes when compared to the unbleached pulp (Table 1). This
indicates that no specific removal of xylan from pulp took place during
bleaching. The chemical composition of the xylans isolated from TCF and ECF
bleached pulps was also similar to the xylan from unbleached pulp, although a
very small decrease in MeGlcpA content in pulp xylans after bleaching was
detected (Tables 2 and 3). This fact may be explained by a partial degradation
of MeGlcpA to HexA under alkaline conditions, which are inaccessible for the
analysis by methanolysis. This explanation was further supported by 1H NMR
analysis, that showed a relative increase of the HexA content and a decrease of
MeGlcpA moieties in the xylan from TCF bleached pulp (Figure 6). In contrast
to the xylan from TCF bleached pulp, the xylan from ECF bleached pulp did not
contain HexA residues, which were degraded upon bleaching with chlorine
dioxide (Figure 6). Taking into account that uronic moieties strongly affect the
papermaking properties (Lindstrm, 1992) and brightness stability (Buchert at
al. 1997) of cellulosic pulps this knowledge may be important to explain the
different response of pulps bleached employing TCF and ECF sequences.

234
5. Resultados y discusin

The xylans from bleached pulps (either TCF or ECF) did not show any acetyl
groups, as revealed by 1H NMR analysis. Hence, alkaline bleaching stages
favored the removal of residual acetyl groups from xylan of unbleached pulp.
Xylans from bleached pulps possessed slightly higher molecular weight (12 kDa
in TCF pulp and 14 kDa in ECF pulp), when compared to this in unbleached
pulp (Figure 2). This fact may be explained by a predominant removal of low
molecular weight xylan fractions structurally associated to residual lignin during
bleaching.

4. Conclusions
The structure of the heteroxylan isolated from sisal fibers has been characterized
and its behavior during soda/AQ pulping and TCF/ECF bleaching has been
studied. The data indicates that the heteroxylan backbone is composed by
(14)-linked -D-xylopyranosyl units (Xylp) partially ramified with terminal
(12)-linked 4-O-methyl--D-glucuronosyl (MeGlcpA, 9 %mol.) and a small
proportion of -D-glucuronosyl (GlcpA) residues. Around 61mol% of the Xylp
residues are acetylated, the major proportion of acetyl groups being attached at
the O-3 position of the Xylp residues (39 %mol.), followed by acetylation at the
O-2 position (13 %mol.) and diacetylation at both O-2 and O-3 positions
(9%mol.). The molecular weight (Mw) of the heteroxylan was of 18 kDa.
Around 40% of xylan was removed during soda pulping. However, the major
polysaccharides found in the black liquor were E-glucans rather than xylans.
Sisal xylan suffered a significant depolymerisation (Mw decreased to almost
half) and deacetylation (95%) during pulping. Terminal MeGlcpA residues were
partially removed (to about 30%) or converted to HexA in a large extent. HexA
revealed to be relatively stable during TCF bleaching with hydrogen peroxide
and were predominant among uronic moieties of xylan. Since all HexA were
degraded during ECF bleaching with chlorine dioxide, the final pulp contained a
xylan with rather small amount of uronosyls (MeGlcpA residues). A small
proportion of MeGlcpA residues (15% from initial amounts), remaining intact
during soda pulping, were additionally converted to HexA residues during alkali
bleaching stages. After bleaching, the residual acetyl groups were completely
removed from the pulp xylan. It was suggested that a low molecular weight
fraction of xylan, probably associated to residual lignin, was removed from upon
bleaching.

Acknowledgements
This study has been supported by the Spanish Projects AGL2005-01748 and
AGL2008-00709 and the EU BIORENEW project (NMP2-CT-2006-26456).
We thank CELESA (Tortosa, Spain) for providing the samples. G.M. thanks the
Spanish Ministry of Education for a FPI fellowship

235
5. Resultados y discusin

References
Bertaud, F., Sundberg, A., & Holmbom, B. (2002). Evaluation of acid
methanolysis for analysis of wood hemicelluloses and pectins. Carbohydrate
Polymers, 48, 319-324.

Bleton, J., Mejanelle, P., Sansoulet, J., Goursaud, S., & Tchapla, A. (1996).
Characterization of neutral sugars and uronic acids after methanolysis and
trimethylsilylation for recognition of plant gums. Journal of Chromatography
A, 720, 27-49.

Buchert, J., Bergnor, E., Lindblad, G., Viikari, L., & Ek, M. (1997).
Significance of xylan and glucomannan in the brightness reversion of kraft
pulps. Tappi Journal, 80(6), 165-170.

Das Gupta P. C., & Mukherjee P. P. (1967) The hemicellulose of sisal fibre
(Agave sisalana). Journal of the Chemical Society C: Organic, 1179-1182.

del Ro, J. C., Marques, G., Rencoret, J., Martnez, A. T., & Gutirrez, A.
(2007). Occurrence of naturally acetylated lignin units. Journal of Agriculture
and Food Chemistry, 55, 5461-5468.

del Ro, J. C., Rencoret, J., Marques, G., Gutirrez, A., Ibarra, D., Santos, J. I.,
Jimnez-Barbero, J., & Martnez, A. T., (2008) Highly acylated (acetylated
and/or p-coumaroylated) native lignins from diverse herbaceous plants.
Journal of Agriculture and Food Chemistry, 56, 9525-9534.

Ebringerov, A., Hromdkov, Z., & Heinze, T. (2005). Polysaccharides I.


Advances in. Polymer Science, 186, 1-67.

Engstrm, N., Vikkula, A., Teleman A., & Vuorinen, T. (1995). Structure of
hemicellulose in pine kraft cooking liquors. Proceedings of the eighth
International Symposium on Wood and Pulping Chemistry, (pp. 195-200), 6-9
June, Helsinki, Finland.

Evtuguin, D. V., Toms, J. L., Silva, A. M. S., & Pascoal Neto, C. (2003).
Characterization of an acetylated heteroxylan from Eucalyptus globulus labill.
Carbohydrate Research, 338, 597-604.

Gonalves, V., Evtuguin D. V., & Domingues, M. R. (2008). Structural


characterization of acetylated heteroxylan from the natural hybrid Paulownia
elongate/Paulownia fortunei. Carbohydrate Research, 343, 256-266.

Gorshkova, T. A., Wyat, S. E., Salnikov, V. V., Gibeaut, D. M., Ibragimov, M.


R., Lozovaya, V. V., & Carpita, N.C. (1996). Cell-wall polysaccharides of
developing flax plants. Plant Physiology, 110, 721-729.

236
5. Resultados y discusin

Gutirrez, A., Rodrguez, I. M., del Ro, J. C. (2008). Chemical composition of


lipophilic extractives from sisal (Agave sisalana) fibers. Industrial Crops &
Products, 28, 81-87.

Hurter, R. W. (2001). Sisal fibre: market opportunities in the pulp & paper
industry. FAO/CFC Technical Paper, 14, 61-74.

Idrraga, G., Ramos, J., Young, R. A., Denes, F., & Zuiga, V. (2001).
Biomechanical pulping of Agave sisalana. Holzforschung, 55, 42-46.

Li, Y., Mai, Y.-W., & Ye, L. (2000). Sisal fibre and its composites: a review of
recent developments. Composites Science and Technology, 60, 2037-2055.

Lindstrm, T. (1992). Chemical factors affecting the behaviour of fibres during


papermaking. Nordic Pulp and Paper Research Journal, 4(7), 181-192.

Lisboa, S. A., Evtuguin, D. V., & Pascoal Neto, C. (2004). Isolation and
structural characterization of polysaccharides dissolved in Eucalyptus
globulus kraft black liquors. Carbohydrate Polymers, 60, 77-85.

Megiatto, J. D., Houreau, Jr. W., Gardrat, C., Frollini, E., Castellan, A. (2007).
Sisal fibers: surface chemical modification using reagent obtained from
renewable source, characterization of hemicellulose and lignin as model
study. Journal of Agriculture and Food Chemistry, 55, 8576-8584.

Mwaikambo, L. Y. (2006). Review of the history, properties and application of


plant fibers. African Journal of Science and Technology, 7, 120-133.

Neto, P. C., Seca, A., Fradinho, D., Coimbra, M. A., Domingues, F., Evtuguin,
D., Silvestre, A., & Cavaleiro, J. A. S. (1996). Chemical composition and
structural features of the macromolecular components of Hibiscus cannabinus
grown in Portugal. Industrial Crops and Products, 5, 189-196.

Nobel, P. S. (2003). Environmental Biology of Agaves and Cacti. Cambridge:


Cambridge Press.

Pinto, P. C., Evtuguin, D. V., & Pascoal Neto, C. (2005). Structure of hardwood
glucuronoxylans: modifications and impact on pulp retention during wood
kraft pulping. Carbohydrate Polymers, 60, 489-497.

Rowell R. M., Han J. S., & Rowell, J. S. (2000). Characterization and factors
affecting fiber properties. In: Natural Polymers and Agrofibers Composites
(Frollini E, Leo AL, Mattoso LHC eds), IQSC/USP, UNESP and Embrapa
Instrumentao Agropecuria. So Carlos - Brazil, pp. 115-134.

237
5. Resultados y discusin

Selvendran, R. R., March, J. F. & Ring, S. G., (1979). Determination of aldoses


and uronic acids content of vegetable fiber. Analytical Biochemistry, 96, 282-
292.

Shimizu, K. (1991). Chemistry of hemicelluloses. In: Hon, D. H. -S., Shiraishi,


N. (Eds.), Wood and Cellulosic Chemistry. (pp. 3-57), New York: Marcel
Dekker.

Stewart, D., Azzini, A. Hall, A. T., & Morrison, I. M. (1997). Sisal fibres and
their constituent non-cellulosic polymers. Industrial Crops & Products, 6, 17-
26.

Sun, R. C., Tomkinson, J., Ma, P. L., & Liang, S. F. (2000). Comparative study
of hemicelluloses from rice straw by alkali and hydrogen peroxide treatments.
Carbohydrate Polymers, 42, 111-122.

Sundberg, A., Sundberg, K., Lillandt, C., & Holmbom, B. (1996). Determination
of hemicelluloses and pectins in wood and pulp fibers by acid methanolysis
and gas chromatography. Nordic Pulp and Paper Research Journal, 4, 216-
226.

Tappi Test Methods (1993), Atlanta: TAPPI Press.

Teleman, A., Hajunp, V., Tenkanen, M., Buchert, J., Hausalo, T., Drakenberg,
T., & Vuorinen, T. (1995). Characterization of 4-deoxy-E-L-threo-hex-4-
enopyranosyluronic acid attached to xylan in pine kraft pulp and pulping
liquor by 1H and 13C NMR spectroscopy. Carbohydrate Research, 272, 55-71.

Teleman, A., Lundqvist, J., Tjerneld, F., Stalbrand, H., & Dahlman, O. (2000).
Characterization of acetylated 4-O-methylglucuronoxylan isolated from aspen
employing 1H and 13C NMR spectroscopy. Carbohydrate Research 329, 807-
815.

Teleman, A.; Tenkanen, M.; Jacobs, A.; & Dahlman, O. (2002).


Characterization of O-acetyl-(4-O-methylglucurono)xylan isolated from birch
and beech. Carbohydrate Research, 337, 373-377.

Torosantucci, A., Bromuro, C., Chiani, P., De Bernardis, F., Berti, F., Galli, C.,
Norelli, F., Bellucci, C., Polonelli, L., Costantino, P., Rappuoli, R., &
Cassone, A. (2005) A novel glyco-conjugate vaccine against fungal
pathogens. The Journal of Experimental Medicine, 202, 597- 606.

Vignon, M. R., & Gey, C. (1997). Isolation, 1H and 13C NMR studies of (4-O-
methyl-D-glucurono)-D-xylans from luffa fruit fibers, jute bast fibers and
mucilage of quince tree seeds. Carbohydrate Research, 307, 107-111.

238
5. Resultados y discusin

Vuorinen, T., Fagerstrm, P., Buchert, J., Tenkanen, M., & Teleman, A. (1999).
Selective hydrolysis of hexenuronic acid groups and its application in ECF
and TCF bleaching in kraft pulps. Journal of Pulp and Paper Science, 25(5),
155-162.

239
5. Resultados y discusin

Publicacin IX:

Marques G., Gamelas J. A., Ruiz-Dueas F.J., del Ro J.C., Evtuguin D.V.,
Martnez A.T. and Gutirrez A. (2010) Delignification of eucalypt kraft pulp
with manganese-substituted polyoxometalate assisted by fungal versatile
peroxidase. Bioresource Technology, 101, 5935-5940.

240
5. Resultados y discusin

Delignification of eucalypt kraft pulp with manganese-substituted


polyoxometalate assisted by fungal versatile peroxidase

Gisela Marques a, Jos A.F. Gamelas b,, Francisco J. Ruiz-Dueas c, Jos C. del Rio a, Dmitry
V. Evtuguin b, Angel T. Martnez c, Ana Gutirrez a
a
Instituto de Recursos Naturales y Agrobiologa de Sevilla, CSIC, PO Box 1052, E-41080
Seville, Spain
b
University of Aveiro, CICECO, 3810-193, Aveiro, Portugal
c
Centro de Investigaciones Biolgicas, CSIC, Ramiro de Maeztu 9, E-28040 Madrid

Abstract
Oxidation of the manganese-substituted polyoxometalate [SiW11MnII(H2O)O39]6-
(SiW11MnII) to [SiW11MnIII(H2O)O39]5-, one of the most selective
polyoxometalates for the kraft pulp delignification, by versatile peroxidase (VP)
was studied. First, SiW11MnII was demonstrated to be quickly oxidized by VP at
room temperature in the presence of H2O2 (Km= 6.40.7 mM and kcat= 472 s-1).
Second, the filtrate from eucalypt pulp delignification containing reduced
polyoxometalate was treated with VP/H2O2, and 95-100% reoxidation was
attained. In this way, it was possible to reuse the liquor from a first SiW11MnIII
stage for further delignification, in a sequence constituted by two
polyoxometalate stages, and a short intermediate step consisting of the addition
of VP/H2O2 to the filtrate for SiW11MnII reoxidation. When the first ClO2 stage
of a conventional bleaching sequence was substituted by the two-stage
delignification with polyoxometalate (assisted by VP) a 50% saving in ClO2 was
obtained for similar mechanical strength of the final pulp.
Keywords: Polyoxometalate, versatile peroxidase, oxidative delignification, pulp
bleaching, eucalypt kraft pulp

1. Introduction
The residual lignin remaining after wood pulping is the target of the bleaching
process to produce high quality pulp for the papermaking. This plant polymer,
which is responsible for the undesirable dark color and photoyellowing of pulp,
must be attacked with minimal polysaccharides damage to preserve the physical
properties of the bleached pulp.
In the middle of the 1990's, polyoxometalates (POM) were proposed as an
environmentally-friendly alternative to the chlorine-based bleaching reagents,
as well as to conventional alkaline oxygen pre-bleaching (Evtuguin and Neto,
1997; Weinstock et al., 1997). POM have been evaluated for the
bleaching/delignification of pulps both as reagents under anaerobic conditions
(in this case a second stage is required for POM reoxidation and reuse) or as
catalysts under aerobic conditions (Gamelas et al., 2008; Gaspar et al., 2007;

241
5. Resultados y discusin

Weinstock et al., 1997). Applied as catalysts, POM oxidizes the residual lignin
in pulp, and the reduced form of POM is reoxidized by molecular oxygen at the
same stage. Therefore, it is possible to reuse the POM solutions in a closed loop.
The thermodynamic conditions required for lignin oxidation and reoxidation of
the POM are related to the corresponding redox potentials as follows: E (Lignin)
< E (POM) < E (O2) = 1.22 0.059 pH.
Several POM types, mostly with the Keggin-type structure (Fig. 1), have
been considered for kraft pulp delignification, such as [SiW11VO40]5-,
[SiW10V2O40]6-, SiW10.1Mo1.0V0.9O40, and [SiW11Mn(H2O)O39]5- (SiW11Mn)
(Gamelas et al., 2005a; 2008; Gaspar et al., 2003; 2007; 2009; Weinstock et al.,
1997; 2001). However, some of them possessing high M(n+1)/n redox potentials
(E = +0.70.8 V), although lower than oxygen redox potential, are hardly
reoxidized even at extreme conditions of oxygen pressure and temperature
(Gamelas et al., 2005a; Gaspar et al., 2003; Weinstock et al., 1997), thus,
limiting their reuse for pulp delignification.
In particular, SiW11Mn has been found to be highly selective in pulp
delignification (Gamelas et al., 2005a; Gaspar et al., 2003). The SiW11Mn/O2
catalytic system has been compared to the conventional alkaline oxygen process
already used by the pulp industry. In addition to lignin removal, an important
advantage of the SiW11Mn-based process, when applied to eucalypt kraft pulps,
is the higher reduction of kappa number than in the alkaline oxygen process, due
to the degradation of hexenuronic acids at the low pH used in these reactions
(Gamelas et al., 2005a). However, SiW11MnII is very slowly reoxidized under
these conditions, limiting its practical application.

Fig. 1. Structural representation of the Mn-substituted polyoxometalate, D-


[SiW11MnIII(H2O)O39]5-. The dark octahedron represents the MnIIIO5(H2O) group with Mn at
the centre of the octahedron.

Enzymatic catalysis is a promising approach to regenerate some of the POM


species that are difficult to be reoxidized by O2 and other chemical oxidizers. In
this context, fungal laccase (from Trametes versicolor) has been assayed for the
reoxidation of [SiW11VIVO40]6- and [SiW11MnII(H2O)O39]6- (SiW11MnII)

242
5. Resultados y discusin

(Gamelas et al., 2005b; Tavares et al., 2004). Although laccase easily oxidized
VIV to VV in the former POM, the corresponding oxidation of MnII to MnIII in
the manganese-based POM was slow (less that 50% after 4 h at 45 C, and under
0.3 bar oxygen pressure) (Gamelas et al., 2005b). This urged the search for
alternative faster methods of oxidation of MnII-substituted POM.
In contrast to laccase, versatile peroxidase (VP) produced by fungi of the
genera Pleurotus and Bjerkandera is a high redox-potential enzyme able to
oxidize a variety of substrates, including free MnII, due to the presence of
different catalytic sites in its molecular architecture (Ruiz-Dueas et al., 2009).
VP is activated by H2O2 in a two-electron reaction yielding highly reactive
intermediate states. Activated VP can oxidize two substrate molecules in two
successive one-electron reactions. It has been demonstrated that MnIII, resulting
from MnII oxidation by VP or related manganese peroxidase (Ruiz-Dueas et
al., 2007), is stabilized in solution by the chelation of dicarboxylic acids of small
size produced by ligninolytic fungi. The manganic cation can, then, act as an
oxidizer of lignin contributing to wood delignification in nature (Wariishi et al.,
1992).
In the present work, reoxidation of the MnII-containing POM, SiW11MnII, by
the VP/H2O2 system, was studied for the first time. Based on the easy oxidation
of MnII (as a free ion or in POM complexes) by the enzyme a novel approach for
the delignification catalysis was developed. Reduced POM in the liquor from a
first eucalypt pulp delignification stage was reoxidized by VP, and the resultant
liquor mixed with the partially delignified pulp for a further delignification stage
in a simple POM-VP-POM trial. In addition, delignification of eucalypt pulp by
POM in a VP-assisted process was tested as a pre-bleaching stage to substitute
the first Cl2O stage in a conventional elemental chlorine free (ECF) bleaching
sequence.

2. Materials and methods


2.1. Pulp samples and POM synthesis
The delignification assays were carried out with Eucalyptus globulus
unbleached kraft pulp supplied by ENCE pulp mill (Spain). The pulp had a
kappa number of 13.7, and an intrinsic viscosity of 1180 cm3/g.
For the kinetic studies of SiW11MnII oxidation by VP/H2O2, the potassium
salt of the MnII-containing POM, K6[SiW11MnII(H2O)O39].10 H2O, was prepared
(Tourn et al., 1970). For the delignification experiments, a solution containing
2.8 r 0.1 mmol/L of [SiW11MnIII(H2O)O39]5- (SiW11MnIII) was prepared as
previously reported (Galli et al., 2007).

2.2. VP expression, in vitro activation and purification


Recombinant VP was obtained from E. coli W3110 transformed with the
pFLAG-VPL2 expression vector as previously described (Prez-Boada et al.,

243
5. Resultados y discusin

2002). The enzyme was activated in vitro after solubilization of inclusion bodies
in 8 M urea. The folding conditions included 0.15 M urea, 5 mM Ca2+, 20 M
hemin, a 4:1 oxidized-glutathione/reduced-glutathione ratio and 0.1 mg/mL
protein at pH 9.5. The active enzyme was purified in a single chromatographic
step (Resource-Q column, Pharmacia-Biotech) using a 00.3 M NaCl gradient
(2 mL/min) in 10 mM sodium tartrate (pH 5.5) containing 1 mM CaCl2. The
concentration of the enzyme was determined by spectrophotometry (H407 150
000 M-1 cm-1) (Ruiz-Dueas et al., 1999).

2.3. SiW11MnII oxidation by VP


Oxidation of the MnII-substituted POM was followed at 20 C in a quartz cuvette
(1 cm optical path) under stirring: 3.0 mL of 0.1 M acetate solution (pH 4.5)
containing 2.7 mM SiW11MnII (K6[SiW11MnII(H2O)O39].10H2O), 0.56-1.26 PM
VP and 0.57-2.24 mM H2O2 were mixed inside the cell. The increase of
absorbance at 490 nm was followed for 1 min, until a constant value was
reached. The oxidation degree was estimated using the molar extinction
coefficients of the oxidized and reduced POM (SiW11MnIII H490 325 M-1 cm-1;
and SiW11MnII H490 22 M-1 cm-1) (Tourn et al., 1970). For the assays with the
delignification liquor, the H2O2 amount varied between 0 and 2.06 mM, with the
amount of enzyme kept at 1.20 PM. The total absorbance was corrected for the
liquor contribution.
Steady-state kinetic constants were calculated during VP oxidation of
increasing SiW11MnII concentrations in 0.1 M sodium tartrate, pH 5, containing
0.1 mM H2O2. The enzymatic activity at 20 C was measured as the initial
velocity, taking linear increments. Mean values and standard errors for the
apparent affinity constant (Michaelis constant, Km) and maximal enzyme
turnover (catalytic constant, kcat) were obtained by non-linear least-squares
fitting of the experimental measurements to the Michaelis-Menten model.
Fitting of these constants to the normalized equation v = (kcat/Km)[S]/(1+[S]/Km)
yielded the efficiency value (kcat/Km) with its corresponding standard error.

2.4. Pulp delignification experiments


Pulp delignification was carried out in a PARR reactor, model 4843 (0.25 L)
equipped with an automatic temperature control system, pressure control and
mechanical stirring (220 rpm). 7.5g of pulp (dry weight), 67 mL of 0.2 M
sodium acetate (pH 4.5), 13 mL of 28 mM POM (SiW11MnIII) solution, and
water to make a final volume of 132 mL were put inside the reactor. The final
concentration of POM was 2.7 mM. At the end of the reactions, the reactor was
quickly cooled with water and degasified.
In the two-stage experiments, including intermediate POM reoxidation with
VP (POM-VP-POMreox), the pulp from the first stage was filtered and pressed,
the required amounts of enzyme and H2O2 were added to the delignification

244
5. Resultados y discusin

liquor, and the solution was stirred at 20-25 C for 10 min. The liquor containing
the reoxidized POM (verified by visible absorption spectroscopy) was mixed
again with the filtered pulp and a second delignification stage was applied under
the same experimental conditions of the first stage. A two-stage experiment not
including the reoxidation step of POM by VP/H2O2 was also performed by
adding fresh POM (SiW11MnIII), acetate buffer and water to the washed pulp
obtained after the first stage.
After each delignification sequence the pulps were filtered and washed with
water until neutral filtrate. Alkaline extraction of pulps was carried out at 70 C
during 1 h and NaOH charge of 2% (on the dry pulp weight).

2.5. Modified ECF bleaching sequence


Bleaching with Cl2O was performed on untreated kraft pulp and with pulp
delignified with POM, at 10% pulp consistency, in plastic bags in a Grant model
Y28 thermostatic bath. Two bleaching sequences, D-Ep-D-Ep-D and POM-VP-
POMreoxE-D-Ep-D, were studied (D refers to Cl2O stage; Ep to peroxide-
reinforced alkaline extraction, POM-VP-POMreox corresponds to VP-assisted
two stage [2 h + 2 h] POM treatment; and E to alkaline extraction). The
bleaching conditions in the D-Ep-D-Ep-D sequence were as follows: first D
stage at 50 C for 1 h; second D stage at 70 C for 2 h; third D stage at 70 C for
2.5 h; first Ep stage at 70 C for 1 h, using 2.0% NaOH and 0.2% H2O2; second
Ep stage at 70 C for 1 h, using 1.5% NaOH and 0.1% H2O2.
The pulp delignified with POM (2 h)-VP-POMreox (2 h) and extracted with
NaOH was subjected to D-Ep-D bleaching (POM-VP-POMreoxE-D-Ep-D
sequence). The conditions of the last stages in this sequence were as follows:
first D stage at 50 C for 1 h; second D stage at 70 C for 2.5 h; Ep stage at 70 C
for 1 h, using 1.5% NaOH and 0.2% H2O2. The loads of ClO2 for each stage in
both sequences are discussed in the text.

2.6. Pulp characterization


The treated pulps were characterized using TAPPI T236 cm99 standard for the
kappa number (Tappi, 2006), and the SCAN-CM 15:88 standard for the intrinsic
viscosity (Scandinavian Pulp Paper and Board Committee, 1994). Hexenuronic
acid content was determined by acid hydrolysis in sodium formate (pH 3.0)
followed by spectrophotometric (245 nm) quantitation of the furan derivatives
formed (Vuorinen et al., 1999). The acid hydrolysis treatment of pulp was
carried out in the same PARR reactor mentioned above (see 2.4). The strength
properties, brightness, and opacity of the bleached pulps were determined
according to ISO (International Organisation for Standardization Documentation
and Information, 2003) and TAPPI (Tappi, 2006) standards.

245
5. Resultados y discusin

3. Results and discussion


3.1. Kinetics of SiW11MnII oxidation by VP
The ability of VP, a peroxidase acting on free MnII and other substrates, to
oxidize this metal ion in Mn-substituted POM is demonstrated here for the first
time. The steady-state kinetic constants for SiW11MnII oxidation, obtained by
non-linear fitting of initial velocities vs substrate concentration (Fig. 2), revealed
high VP turnover on SiW11MnII, with a kcat of 47 2 s-1, and a moderate affinity
for this compound, with a Km of 6.4 0.7 mM. This VP activity was lower than
on free MnII, with a reported kcat value near 300 s-1, but the main difference
between both substrates concerned Km that was around 0.19 mM for free MnII,
revealing over 30-fold higher affinity of VP on the free metal ion (Ruiz-Dueas
et al., 2007). As a result, the global catalytic efficiency of VP oxidizing
SiW11MnII (7.36 0.6 mM-1 s-1) was around 200-fold lower than that for
oxidation of free MnII (1600 100 mM-1 s-1).

40

30
k obs (s )
-1

20
K m (6.4 0.7) mM
k cat (47.3 2) s-1
10 Efficiency (7.36 0.6) mM-1 s -1

0
0 5 10 15 20 25 30
[SiW 11MnII] (mM)

Fig. 2. Michaelis-Menten kinetics of SiW11MnII oxidation by VP. The Km, kcat and efficiency
values (means and standard errors) are shown inside the plot.

The high affinity of VP for free MnII is due to the existence of a specific
catalytic site in this enzyme constituted by three acidic residues forming a small
channel on the internal heme propionate, where the free metal cation is oxidized
(Ruiz-Dueas et al., 2007). The lower efficiency observed for SiW11MnII
oxidation by VP was in the order of those reported both for veratryl alcohol
oxidation taking place at a tryptophan residue exposed to the solvent (Prez-
Boada et al., 2005), and for oxidation of phenols at the edge of the main heme
access channel (Ruiz-Dueas et al., 2008). This suggested that SiW11MnII could
be oxidized in one of these two easily accessible catalytic sites, and not in the
narrow channel described for free MnII that most probably provides a limited
access to the bulky SiW11MnII. A detailed kinetic study of different VP variants
mutated at the three catalytic sites mentioned above would be necessary to
definitively identify the SiW11MnII oxidation site in VP.

246
5. Resultados y discusin

3.2. Optimization of SiW11MnII oxidation by VP (in the presence of H2O2)


A set of assays was carried out aiming to optimize the oxidation of SiW11MnII to
SiW11MnIII by VP (in the presence of H2O2), either by using an aqueous solution
of SiW11MnII buffered at pH 4.5, and the liquor from previous eucalypt kraft
pulp delignification with POM (Table 1). The assays were performed at 20 C,
with 2.7 mM POM concentration, and varying the H2O2/POM and POM/VP
molar ratios. During the oxidation of POM by VP (and H2O2) no indication
about the formation of other species besides SiW11MnIII was obtained.

Table 1. Oxidation of SiW11MnII by VP/H2O2 in two distinct reaction media using


different H2O2, POM and VP molar ratiosa
H2O2/POM H2O2/VP POM/VP Oxidation (%) Time (min)
A) Oxidation of SiW11MnII in aqueous solution buffered at pH 4.5
0.19 446 2382 40 2
0.40 883 2226 78 2
0.50 1117 2226 95 5
0.61 1351 2226 100 5
0.81 1793 2226 57 22
0.50 2495 5009 55 29
0.50b 2495 5009 94 10
B) Oxidation of SiW11MnII in the filtrate from one-stage delignification with POM
0 0 2227 18 1
0.21 472 2227 57 2
0.40 896 2227 93 6
0.50 1116 2230 97 2
0.59 1321 2227 74 27
0.79 1769 2227 44 18
a
The assays were performed at 20 C, with 2.6-2.8 mM POM concentration
b
H2O2 was added in five portions each of them including 20% of the total volume required

In the assays performed with the SiW11MnII solution (Table 1A), the extent
of POM oxidation (for a fixed amount of enzyme) increased with the H2O2/POM
ratio until a 0.5-0.6 molar ratio, and then decreased at higher ratios. Using this
H2O2/POM ratio (0.5-0.6), 95-100% POM oxidation was obtained in less than 5
min, with a POM/VP ratio ~2200. These values were in agreement with the
stoichiometry of the global enzymatic reaction, which predicts that 0.5 moles of
H2O2 will be needed to oxidize 1 mole of SiW11MnII. For the H2O2/POM ratio of
0.8, only 57% oxidation of SiW11MnII was obtained, indicating enzyme
inactivation by the excess of H2O2 (Valderrama et al., 2002). If the amount of
enzyme was reduced to about 50%, keeping the H2O2/POM ratio of 0.5, the
oxidation extent also decreased (to only 55%) due to the increased H2O2/VP
ratio. However, when the later assay was carried out by adding the H2O2 in
several steps, without exceeding a 500-fold molar excess of H2O2 in each
addition, the extent of oxidation (94%) was similar to that attained using a
higher amount of enzyme. These data confirmed VP inactivation by H2O2 (even

247
5. Resultados y discusin

in the presence of enough amount of SiW11MnII to consume all the H2O2) and
showed that the enzyme dose can be reduced by stepwise addition of H2O2 (to
prevent VP inactivation). No POM oxidation was observed in the absence of
H2O2 or without enzyme.
In the assays with the liquor from POM delignification of eucalypt kraft pulp
(Table 1B), which were performed using a POM/VP ratio ~2200, and varying
the H2O2/POM ratio between 0 and 1, the highest POM oxidation degrees (over
90%) were obtained at the H2O2/POM ratios of 0.4-0.5. In the absence of H2O2,
some POM oxidation occurred suggesting that some substances present in the
delignification liquor may act as the enzyme oxidizing agents. In fact, for all the
assays performed with a H2O2/POM ratio up to 0.5, the oxidation of POM was
higher when the reactions were conducted in the delignification liquor than
those in the SiW11MnII aqueous solutions. Again, the use of H2O2/POM ratios t
0.6 did not improve POM reoxidation in the delignification liquor, and lower
rates were obtained. It was concluded that a H2O2/POM ratio around 0.5 and a
POM/VP ratio of 2000-3000 should be used to obtain near complete oxidation
of the manganese-substituted POM.

3.3. Two-stage POM delignification of pulp assisted by VP


As a continuation of the above studies, which showed easy oxidation of
SiW11MnII by VP/H2O2, a novel approach for the paper pulp delignification was
developed. A first delignification stage using chemically-prepared SiW11MnIII
and O2 was followed by pulp filtration, and a short intermediate step consisting
of the addition of VP and H2O2 to the filtrate. Fig. 3 shows the visible absorption
spectra of the filtrate before and after the enzymatic treatment (resulting in the
complete reoxidation of the previously reduced POM). Then, the filtrate with
reoxidized POM was mixed again with the pulp, and a second delignification
stage (under the same conditions of the first one) was applied. The results were
compared with those obtained when the second delignification stage was
performed by adding chemically-prepared SiW11MnIII, as well as when only
one-stage POM delignification was performed (Table 2).
After one-stage POM delignification at 110 C, decreases of kappa number (a
rough measure of the lignin content in pulp) of 40% and 50%, with viscosity
losses of only 3% and 6%, were obtained after 1-h and 2-h reaction, respectively
(Table 2). Besides residual lignin, hexenuronic acids contribute significantly to
the kappa number in E. globulus kraft pulps and to the consumption of bleaching
reagents (Furtado et al., 2001). In fact, a significant removal of hexenuronic
acids (up to 70% after 2 h) was detected after the POM treatment. It is
noteworthy that the POM/O2 system was highly selective for delignification
when compared with the oxygen-delignification control, which showed a
viscosity loss of 28% (near 10-fold higher than that obtained with POM
delignification) (Gamelas et al., 2005a).

248
5. Resultados y discusin

1,5

Absorbance
b

0,5

a
0
300 400 500 600 700 800
Wavelength (nm)

Fig. 3. UV-vis spectra of the filtrate after 2-h treatment of eucalypt kraft pulp with POM/O2
(a), and after short incubation of this delignification liquor with VP and H2O2 at 20-25C (b),
revealing the typical SiW11MnII and SiW11MnIII spectra, respectively.

After two-stage (2-h each) POM delignification including the intermediate


reoxidation step with VP and H2O2, kappa number was reduced of 62% and the
viscosity had a drop of 11% (Table 2). Interestingly, this treatment also
degraded almost 90% of the hexenuronic acids present in the pulp. The
delignification degree corrected for the hexenuronic acids content was of 51%.
Similar results in terms of pulp kappa number, viscosity and hexenuronic acid
degradation were obtained in parallel assays in which freshly-prepared
SiW11MnIII was added after the first POM stage, revealing that the presence of
the enzyme did not exert a negative effect on the performance and selectivity of
the SiW11MnIII/O2 system.

Table 2. Delignification of eucalypt kraft pulp with SiW11MnIII/O2 assisted by VP/H2O2


(effect of different treatments on pulp kappa number, intrinsic viscosity and hexenuronic
acid content (HexA))a
Kappa Viscosity Kappa Viscosity HexA
number (cm3/g) decrease (%)d loss (%) (mmol/kg)
Initial kraft pulp 13.6 1215 - - 61.2
O2 (without POM, 2 h) 7.3 875 46 (33) 28 15.7
POM (1 h) 8.2 1180 40 (33) 3 28.5
POM (2 h) 6.8 1140 50 (40) 6 18.4
POM (1 h)-VP-(1 h)b 6.5 1130 52 (42) 7 16.8
b
POM (2 h)-VP-(2 h) 5.2 1080 62 (51) 11 9.3
POM (2 h)-POM (2 h)c 5.2 1085 62 (50) 11 8.2
a
Pulp consistency of 5.3%; 2.7 mM POM; pH 4.5; pO2 of 0.5 MPa; 110 C; and 220 rpm
b
The pulp after the first stage was filtered, and the POM in the filtrate reoxidized by VP/H2O2.
c
The pulp after the first stage was washed, and fresh POM (SiW11MnIII) was added
d
Kappa number reduction corrected for HexA (kappacor= kappa - 0.073 u [HexA]) in parentheses

249
5. Resultados y discusin

3.4. Modified ECF bleaching including a VP-assisted POM stage


Pulp treatment with the above VP-assisted two-stage (2-h each) POM
delignification followed by an alkaline extraction (POM-VP-POMreox-E) was
investigated to substitute the first Cl2O stage in a conventional D-Ep-D-Ep-D
ECF bleaching sequence for eucalypt kraft pulp. Results from the conventional
D-Ep-D-Ep-D bleaching sequence (see Materials and methods) and the
sequence including VP-assisted two-stage POM delignification, (POM-VP-
POMreoxE-D-Ep-D), were compared in terms of Cl2O savings for the same final
brightness (~89% ISO). Pulp bleached by the sequence including VP and POM
showed a Cl2O consumption 50% lower than the conventional ECF sequence
(Table 3). The Cl2O oxidation equivalents (OXE) per kappa number unit in the
modified sequence were higher than in the conventional sequence. Moreover,
the main strength properties of the unbeaten pulps after the two bleaching
sequences were very similar (Table 4). The results obtained suggest that VP-
assisted continuous reutilization of SiW11MnIII in a two-reactor system (Gamelas
et al., 2007) may be implemented in future industrial ECF sequences, with no
apparent deterioration of the pulp strength properties, while significantly
reducing the Cl2O consumption, and consequently lowering the environmental
impact of the bleaching process.

Table 3. Cl2O consumption and oxidation equivalents (OXE) for eucalypt pulp
bleaching in a conventional ECF sequence and after substituting the first D-stage by
VP-assisted two-stage POM delignification (89% ISO final brightness)

D-Ep-D-Ep-D POM-VP-POMreoxE-D-Ep-D
a
ClO2 consumption 25 + 9 + 6 15 + 5
OXE b 90 134
a
As active chlorine in each D stage (kg/ton)
b
As moles of active chlorine per ton of dry pulp and kappa unit

4. Conclusions
In this work, we demonstrate that the reduced Mn-substituted POM, SiW11MnII,
can be oxidized by VP (in the presence of H2O2) following Michaelis-Menten
kinetics. This POM, whose oxidized form is highly selective for delignification,
was fully oxidized by VP/H2O2 at 20-25 C in less than 10 min. In this way, a
two-stage POM delignification process, including a fast intermediate step
consisting of the addition of VP (and H2O2) to the delignification filtrate for
POM reoxidation, was performed, resulting in 62% reduction of the pulp kappa
number and a viscosity loss of only 10%. The substitution of the first Cl2O stage
by a POM-VP-POMreox treatment in a conventional ECF bleaching sequence
allowed 50% Cl2O saving without decreasing the pulp strength properties.

250
5. Resultados y discusin

Hence, the continuous reutilization of SiW11MnIII in a two-reactor system may


be implemented in the future.

Table 4. Physical properties of unbeaten bleached pulps (89% ISO, and 65 g/m2) from a
conventional ECF sequence and after substituting the first Cl2O stage by VP-assisted two-stage
POM delignification
D-Ep-D-Ep-D POM-VP-POMreox-E-D-Ep-D
Beating degree (SR) 19 20
Bulk density (g/cm3) 0.56 0.57
Burst index (kPa.m2/g) 1.39 1.45
Tensile strength (N.m/g) 30.4 28.9
Tear index (mN.m2/g) 5.2 6.0
Elongation (%) 2.2 2.1
Stiffness (kN/m) 409 396
Opacity (%) 75.8 77.2
Internal bonds (Scott test, J/m2) 106 114
Air resistance (Gurley-100 mL, s) 0.8 0.8

Acknowledgements
This study has been supported by the EU project BIORENEW (NMP2-CT-
2006-026456) and the Spanish projects AGL2008-00709 and BIO2008-01533.
We thank ENCE (Pontevedra, Spain) for pulp samples. G. M. and F. J. R.-D.
thank the Spanish MICINN for a FPI fellowship and a Ramon y Cajal contract,
respectively.

References
Evtuguin, D.V., Neto, C.P., 1997. New polyoxometalate promoted method of
oxygen delignification. Holzforschung 51, 338-342.
Furtado, F.P., Evtuguin, D.V., Gomes, T.M., 2001. Effect of the acid stage in
ECF bleaching on Eucalyptus globulus kraft pulp bleachability and strength.
Pulp Paper Can. 102, 89-92.
Galli, C., Gentili, P., Pontes, A.S.N., Gamelas, J.A.F., Evtuguin, D.V., 2007.
Oxidation of phenols employing polyoxometalates as biomimetic models of
the activity of phenoloxidase enzymes. New J. Chem. 31, 1461-1467.
Gamelas, J.A.F., Evtuguin, D.V., Gaspar, A.R., 2008. Transition metal
complexes in the delignification catalysis. In: Varga, B., Kis, L. (Eds.),
Transition metal chemistry: New research. Nova Science Publishers, Inc.,
New York, pp. 15-57.

251
5. Resultados y discusin

Gamelas, J.A.F., Gaspar, A.R., Evtuguin, D.V., Neto, C.P., 2005a. Transition
metal substituted polyoxotungstates for the oxygen delignification of kraft
pulp. Appl. Cat. A:General 295, 134-141.
Gamelas, J.A.F., Pontes, A.S.N., Evtuguin, D.V., Xavier, A.M.R.B., Esculcas,
A.P., 2007. New polyoxometalate-laccase integrated system for kraft pulp
delignification. Biochem. Eng. J. 33, 141-147.
Gamelas, J.A.F., Tavares, A.P.M., Evtuguin, D.V., Xavier, A.M.B., 2005b.
Oxygen bleaching of kraft pulp with polyoxometalates and laccase applying a
novel multi-stage process. J. Mol. Catal. B Enzym. 33, 57-64.
Gaspar, A., Evtuguin, D.V., Neto, C.P., 2003. Oxygen bleaching of kraft pulp
catalysed by Mn(III)-substituted polyoxometalates. Appl. Cat. A:General
239, 157-168.
Gaspar, A.R., Gamelas, J.A.F., Evtuguin, D.V., Neto, C.P., 2007. Alternatives
for lignocellulosic pulp delignification using polyoxometalates and oxygen: a
review. Green Chem. 9, 717-730.
Gaspar, A.R., Gamelas, J.A.F., Evtuguin, D.V., Neto, C.P., 2009.
Polyoxometalate-catalyzed oxygen delignification process: kinetic studies,
delignification sequences and reuse of HPA-5-MnII aqueous solution. Chem.
Eng. Commun. 196, 801-811.
International Organisation for Standardization Documentation and Information,
2003. ISO Standards Collection on CD-ROM. Paper, board and pulps. ISO,
Geneva.
Prez-Boada, M., Doyle, W.A., Ruiz-Dueas, F.J., Martnez, M.J., Martnez,
A.T., Smith, A.T., 2002. Expression of Pleurotus eryngii versatile peroxidase
in Escherichia coli and optimisation of in vitro folding. Enzyme Microb.
Technol. 30, 518-524.
Prez-Boada, M., Ruiz-Dueas, F.J., Pogni, R., Basosi, R., Choinowski, T.,
Martnez, M.J., Piontek, K., Martnez, A.T., 2005. Versatile peroxidase
oxidation of high redox potential aromatic compounds: Site-directed
mutagenesis, spectroscopic and crystallographic investigations of three long-
range electron transfer pathways. J. Mol. Biol. 354, 385-402.
Ruiz-Dueas, F.J., Martnez, M.J., Martnez, A.T., 1999. Molecular
characterization of a novel peroxidase isolated from the ligninolytic fungus
Pleurotus eryngii. Mol. Microbiol. 31, 223-236.
Ruiz-Dueas, F.J., Morales, M., Garca, E., Miki, Y., Martnez, M.J., Martnez,
A.T., 2009. Substrate oxidation sites in versatile peroxidase and other
basidiomycete peroxidases. J. Exp. Bot. 60, 441-452.

252
5. Resultados y discusin

Ruiz-Dueas, F.J., Morales, M., Prez-Boada, M., Choinowski, T., Martnez,


M.J., Piontek, K., Martnez, A.T., 2007. Manganese oxidation site in
Pleurotus eryngii versatile peroxidase: A site-directed mutagenesis, kinetic
and crystallographic study. Biochemistry 46, 66-77.
Ruiz-Dueas, F.J., Morales, M., Rencoret, J., Gutirrez, A., del Ro, J.C.,
Martnez, M.J., Martnez, A.T., 2008. Improved peroxidases. Patent (Spain)
P200801292.
Scandinavian Pulp Paper and Board Committee, 1994. SCAN Test Methods.
Sweden.
Tappi, 2006. 2006-2007 TAPPI Test Methods. TAPPI Press, Norcoss, GA
30092, USA.
Tavares, A.P.M., Gamelas, J.A.F., Gaspar, A.R., Evtuguin, D.V., Xavier,
A.M.R.B., 2004. A novel approach for the oxidative catalysis employing
polyoxometalate-laccase system: application to the oxygen bleaching of kraft
pulp. Catal. Commun. 5, 485-489.
Tourn, C.M., Tourn, G.F., Malik, S.A., Weakley, T.J.R., 1970.
Triheteropolyanions containing copper(II), manganese(II), or manganese(III).
J. Inorg. Nucl. Chem. 32, 3875-3890.
Valderrama, B., Ayala, M., Vzquez-Duhalt, R., 2002. Suicide inactivation of
peroxidases and the challenge of engineering more robust enzymes. Chem.
Biol. 9, 555-565.
Vuorinen, T., Fagerstrom, P., Buchert, J., Tenkanen, M., Teleman, A., 1999.
Selective hydrolysis of hexenuronic acid groups and its application in ECF
and TCF bleaching of kraft pulps. J. Pulp Paper Sci. 25, 155-162.
Wariishi, H., Valli, K., Gold, M.H., 1992. Manganese(II) oxidation by
manganese peroxidase from the basidiomycete Phanerochaete
chrysosporium. Kinetic mechanism and role of chelators. J. Biol. Chem. 267,
23688-23695.
Weinstock, I.A., Atalla, R.H., Reiner, R.S., Moen, M.A., Hammel, K.E.,
Houtman, C.J., Hill, C.L., Harrup, M.K., 1997. A new environmentally
benign technology for transforming wood pulp into paper. Engineering
polyoxometalates as catalysts for multiple processes. J. Mol. Catal. A Chem.
116, 59-84.
Weinstock, I.A., Barbuzzi, E.M.G., Wemple, M.W., Cowan, J.J., Reiner, R.S.,
Sonnen, D.M., Heintz, R.A., Bond, J.S., Hill, C.L., 2001. Equilibrating
metal-oxide cluster ensembles for oxidation reactions using oxygen in water.
Nature 414, 191-195.

253
5. Resultados y discusin

Publicacin X:

Marques, G., Molina, S., Babot, E.D., Lund, H., del Ro, J.C. y Gutirrez, A.
Exploring the potential of a fungal manganese-containing for pitch control and
pulp delignification. Bioresource Technology (in preparation).

254
5. Resultados y discusin

Exploring the potential of a fungal manganese-containing for pitch control


and pulp delignification

Gisela Marquesa, Setefilla Molinaa, Esteban D. Babota, Henrik Lundb, Jos C. del Roa, Ana
Gutirreza
a
Instituto de Recursos Naturales y Agrobiologa de Sevilla, CSIC, PO Box 1052,
E-41080, Seville, Spain
b
Novozymes A/S, Krogshoejvej 36, DK-2880 Bagsvaerd, Denmark

Abstract
The potential of the lipoxygenase from Gaeumannomyces graminis to remove
lipophilic extractives from eucalypt and flax pulps is explored. Pulp treatments
with the lipoxygenase were performed in the presence and absence of linoleic
acid, and with and without a subsequent hydrogen peroxide stage. The highest
removal of lipophilic extractives from eucalypt pulp, including conjugated
sterols (about 40% removal), and free sterols (up to 10% removal), was attained
in the lipoxygenase treatment with linoleic acid followed by the peroxide stage.
Different degradation patterns were observed among the lipophilic compounds
in flax pulp with the lipoxygenase treatment, although a high removal (from
55% to 70%) of all extractives classes, including alkanes, fatty alcohols, and
free and glycosylated sterols, was attained in most cases. Reactions of the
lipoxygenase with model lipid mixtures were carried out to better understand the
degradation patterns observed in pulps. Pulp delignification by the lipoxygenase
treatments was also evaluated.
Keywords: Lipoxygenase, Gumannomyces graminis, Fungal enzymes, Paper
pulps, Pitch deposits, Lignin removal.

1. Introduction
The non-polar extractable fraction from wood and nonwood fibers, commonly
referred to as lipophilic extractives, includes fatty and resin acids, fatty alcohols,
alkanes, sterols, sterol esters and triglycerides. These lipophilic compounds are
the precursors of the so-called pitch deposits within the pulp and paper
manufacturing processes (Back and Allen 2000). Pitch deposition results in low
quality pulp, and can cause the shutdown of pulp mill operations. Specific issues
related to pulps with high extractives content include runnability problems, spots
and holes in the paper, and sheet breaks.
In addition to physicochemical methods, biological methods including both
enzymes and microorganisms (Gutirrez et al. 2001a; 2009), have been
investigated to control pitch problems in the pulp and paper industry. Lipases,
which hydrolyze triglycerides, are applied with success in softwood (mainly
pine) mechanical pulping at mill scale (Fujita et al. 1992) and find widespread

255
5. Resultados y discusin

use nowadays. However, pitch problems in most of the chemical and mechanical
processes using other raw materials have not yet been solved. Other compounds,
such as free and esterified sterols, resin acids, fatty alcohols and alkanes, are
responsible for pitch problems in these processes (del Ro et al. 1999; 2000;
Gutirrez and del Ro 2005b). In addition to lipases, the use of sterol esterases
has also been suggested (Barfoed 2000; Calero-Rueda et al. 2004; Kontkanen et
al. 2004) because sterol esters are often at the origin of pitch deposits owing to
their high tackiness and resistance to kraft cooking. However, free sterols
released by the action of these esterases are as problematic as sterol esters.
On the other hand, the modification of some lipophilic extractives by the use
of laccases has been reported (Buchert et al. 2002; Molina et al. 2008; Zhang et
al. 2000; 2005). In contrast to lipases and sterol esterases, laccases are oxidative
enzymes whose action is directed toward some aromatic compounds (such as
phenols and anilines), although their reactivity with some unsaturated lipids was
demonstrated. The interest on laccases as industrial biocatalysts has increased
after discovering the effect of some synthetic compounds (Bourbonnais and
Paice 1990; Call 1994) expanding the action of laccases to non-phenolic
aromatics and, therefore, increasing their potential in the degradation of lignin
and other aromatic compounds (Riva 2006; Rodrguez-Couto and Toca 2006;
Widsten and Kandelbauer 2008). Moreover, the use of laccases in the presence
of redox mediators has recently been described for the removal of the lipophilic
extractives responsible for pitch deposition from wood and nonwood paper
pulps (Gutirrez et al. 2006a; 2006b; 2007; Molina et al. 2008).
In addition to laccases, other oxidative enzymes, such as soybean
lipoxygenases have been tested for the degradation of extractives in softwood
thermo-mechanical pulp (Zhang et al. 2007). Earlier work had also suggested
the use of lipoxygenases to reduce model wood pitch mixtures in pulp and
paper (Borch et al. 2003). Lipoxygenases are non-heme iron-containing
dioxygenases which catalyze the oxidation of polyunsaturated fatty acids to
hydroperoxides. Despite extensive studies on the properties, genetic information
and physiological functions of this group of enzymes, there is a lack of
utilization of these natural abundant enzymes in industrial processing. The
specific activity of lipoxygenases to degrade linoleic acid leads to a potential
application in papermaking processes to degrade wood extractives that have
adverse effects on pulp and paper properties. In the present work, we study the
capability of the lipoxygenase from the fungus Gaeumannomyces graminis, to
remove lipophilic extractives from hardwood (eucalypt) and nonwoody (flax)
pulps. This enzyme is a manganese-containing lipoxygenase with several
exceptional features different from other lipoxygenases (Hamberg et al. 1998;
Su and Oliw 1998). It has a broad pH range (being active between pH 5 and 11),
and good heat stability (being active at temperatures up to 60C) (Su and Oliw
1998) which confer great potential of use under mill process conditions. Since
the oxidation of polyunsaturated fatty acids by lipoxygenase leads to the

256
5. Resultados y discusin

formation of reactive fatty acid hydroperoxides and lipid radicals (Prigge et al.
1997) that can co-oxidize lignin in addition to lipids (Kapich et al. 2010),
delignification of this pulp mediated by lipoxygenase treatment in the presence
of linoleic acid was also investigated.

2. Materials and methods


2.1. Pulp samples
Unbleached kraft pulp from eucalypt (Eucalyptus globulus) wood, with 44%
ISO brightness and 13.5 kappa number was obtained from ENCE (Pontevedra,
Spain). Unbleached soda/anthraquinone (AQ) pulp from flax (Linum
usitatissimum) with 34% ISO brightness and 11 kappa number was provided by
CELESA (Tortosa, Spain).

2.2. Model lipids


Linoleic acid, cholesteryl linoleate, nonacosane and octacosanol (from Sigma-
Aldrich) and sitosterol (from Calbiochem) were used.

2.3. Lipoxygenase
The lipoxygenase (GLOX) used was provided by Novozymes (Bagsvaerd,
Denmark) and obtained from the fungus G. graminis. GLOX activity was
130000 units per mg, where one unit will cause an absorbance increase at 234
nm of 0.001 units per min, at pH 7.0 and 30C, when linoleic acid is used as
substrate in a reaction volume of 1.0 ml (light path 1 cm).

2.4. Enzymatic treatments of paper pulps


Pulp treatments with GLOX (10 mg/g eucalypt pulp and 20 mg/g flax pulp)
were carried out using 5 g pulp (dry weight) at 1% consistency (w:w) in 100
mM monosodium phosphate (pH 7), 30C, with oxygen bubbling, in a
thermostatic shaker at 170 rev/min for 4 h. Pulp treatments were performed in
the presence and absence of linoleic acid (0.1 mg/g pulp). In a subsequent step,
pulps at 5% consistency (w:w) were submitted to a bleaching stage using 3%
(w:w) H2O2 and 1.5% (w:w) NaOH, both referred to pulp dry weight, at 90C
for 2 h. Controls without GLOX were performed. Treated pulp samples were
Soxhlet extracted with acetone (8 h) and the extracts were evaporated to dryness
and redissolved in chloroform for gas chromatography/mass spectrometry (GC-
MS) and GC analyses. When required, bis(trimethylsilyl)trifluoroacetamide
(BSTFA, from Supelco) in the presence of pyridine was used to prepare
trimethylsilyl derivatives.

257
5. Resultados y discusin

2.5. Enzymatic reactions with model lipids


Enzymatic reactions with mixtures of model compounds listed in section 2.2 (1
mg) were performed using GLOX (0.1 mg/mg lipid), and Tween 20 as
dispersant (1% v/v) in 100 mM monosodium phosphate (pH 7), 30C, with
oxygen bubbling, in a thermostatic shaker at 100 rev/min for 2h. In a subsequent
step, the model mixtures after the enzymatic reaction were submitted to a
hydrogen peroxide stage, adding 50 l H2O2 at 30% (w:w) and 37.5 l 5 N
NaOH to each reaction flask, at 90C and 100 rev/min for 2 h. Controls without
GLOX were performed.
After the enzymatic treatments, the lipid dispersions were immediately
evaporated, and the reaction products recovered with chloroform: methanol
(1:1), dried and redissolved in chloroform for GC and GC-MS analyses. When
required, BSTFA in the presence of pyridine was used to prepare trimethylsilyl
derivatives.

2.6. GC and GC-MS analyses of lipids


The GC analyses of lipids from both the enzymatic treatments of pulps and
enzymatic reactions with model compounds were performed in an Agilent
6890N Network GC system using a short fused-silica DB-5HT capillary column
(5 m x 0.25 mm internal diameter, 0.1 Pm film thickness) from J&W Scientific,
enabling simultaneous elution of the different lipid classes (Gutirrez et al.
1998). The temperature program was started at 100C with 1 min hold, and then
raised to 350C at 15C/min, and held for 3 min. The injector and flame-
ionization detector (FID) temperatures were set at 300C and 350C,
respectively. Helium (5 ml/min) was used as carrier gas, and the injection was
performed in splitless mode. Peaks were quantified by area, and data from
replicates were averaged. In all cases the standard deviations were below 7% of
the mean values.
The GC-MS analyses were performed with a Varian 3800 chromatograph
coupled to an ion-trap detector (Varian 4000) using a medium-length (12 m)
capillary column of the same characteristics described above for GC/FID. The
oven was heated from 120C (1 min) to 380C at 10C/min, and held for 5 min.
In all GC/MS analyses, the transfer line was kept at 300C, the injector was
programmed from 120C (0.1 min) to 380C at 200C/min and held until the
end of the analysis, and helium was used as carrier gas at a rate of 2 ml/min.
Compounds were identified by mass fragmentography, and by comparing their
mass spectra with those of the Wiley and NIST libraries and standards.

2.7. Pulp evaluation


Pulp brightness, kappa number and intrinsic viscosity were analyzed following
ISO 3688:1999, ISO 302:1981 and ISO 5351/1:1981 standard methods,

258
5. Resultados y discusin

respectively (International Organisation for Standardization Documentation and


Information (ISO) 2003).

3. Results and discussion


3.1. Composition of lipophilic extractives in eucalypt and flax pulps
The lipophilic extractives from eucalypt and flax pulps were analyzed by GC
and GC-MS (Fig. 1). The compounds from unbleached eucalypt kraft pulp
(Fig.1a) include sterols (predominantly sitosterol) in free (peak 8) and
conjugated form, both as glycosides (peak 10) and esters (peak 11), as well as
fatty acids, mainly palmitic acid (peak 1). The detailed composition of lipophilic
extractives from eucalypt pulp has been published elsewhere (Gutirrez et al.
2001b; Gutirrez and del Ro 2001). Among these compounds, free and
esterified sterols have been reported to be the main responsible for pitch
deposition during the manufacturing of eucalypt pulp (del Ro et al. 1998; 1999;
2000). On the other hand, fatty alcohols including hexacosanol (peak 6),
octacosanol (peak 7), and triacontanol (peak 9), and free sterols with sitosterol
predominating (peak 8) and fatty acids, predominantly palmitic acid (peak 1),
linoleic acid (peak 2), oleic acid (peak 3) and stearic acid (peak 4) are the main
lipophilic extractives identified in flax pulp (Fig.1b). Minor amounts of alkanes
such as nonacosane (peak 5) and sterol glycosides (peak 10) were also present.
The detailed characterization of lipophilic extractives from flax pulp has already
been reported (Gutirrez and del Ro 2003a; 2003b; Marques et al. 2010). Fatty
alcohols and alkanes have been shown as the main responsible for pitch
problems during manufacturing of nonwoody soda pulps (Gutirrez and del Ro
2005a).
From the analysis of pulp extractives it would not be expected to see a
significant change in their total content after treatment with GLOX given that
the amount of extractives with the 1-cis, 4-cis-pentadiene (i.e. linoleic acid) is
quite scarce. Still it was chosen to pursue testing of GLOX on both pulps and
model substrates since other studies (Gutirrez et al. 1999) had shown the
presence of linoleic acid esterified with sitosterol as the main sterol ester among
eucalypt extractives and, at the same time, other studies had surprisingly shown
an ability of lipoxygenase to significantly impact other components in
unbleached eucalypt kraft pulp (Borch et al. 2003).

3.2. Removal of pulp lipophilic extractives by lipoxygenase treatment


The eucalypt and flax pulps were treated with the lipoxygenase from G.
graminis (GLOX) to evaluate the potential of this enzyme to remove lipophilic
extractives from these pulps. Additional assays adding linoleic acid to the
enzymatic reactions were also performed to study the mediating effect of
peroxidation products on the lipid removal. After the enzymatic treatments, the
pulps were subsequently submitted to a hydrogen peroxide bleaching stage,

259
5. Resultados y discusin

which is a common stage in the bleaching of these alkaline pulps. The


composition of the lipophilic extractives in pulps after the treatment with GLOX
was analyzed by GC and GC-MS and compared with that of control pulp treated
under the same conditions but without enzyme addition.

100% 8

(a)
Relative response

1 10
2+3 11
4

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

100% 7

(b)
Relative response

1 2+3 8
6
4
5

10

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

Figure 1. GC chromatograms of silylated lipid extract from eucalypt kraft (a), and flax soda
(b) pulps. Peak identification: 1, palmitic acid; 2, oleic acid; 3, linoleic acid; 4, stearic acid; 5,
nonacosane; 6, hexacosanol; 7, octacosanol; 8, sitosterol; 9, triacontanol; 10, sterol
glycosides; and 11, sterol esters

3.2.1. Eucalypt pulp treatments


Table 1 shows the percentage of degradation of the main eucalypt pulp
extractives after the enzymatic treatment with GLOX. GLOX treatment
produced a removal of 22% and 20% of the sterol esters and sterol glycosides,
respectively, but the amount of free sterols remained unchanged. A similar lack
of degradation of free sterols was also observed after TMP pulp treatment with
soybean lipoxygenase (Zhang et al. 2007).

260
5. Resultados y discusin

Table 1. Removal (percentage of reduction) of the main lipophilic extractives from


eucalypt pulp after treatment with lipoxygenase (GLOX) in the absence and presence
of linoleic acid (LA), without and with a subsequent H2O2 stage (P).
Without H2O2 With H2O2
GLOX GLOX/LA GLOX-P GLOX/LA-P
Free sterols 0 0 0 7
Sterol glycosides 22 24 20 39
Sterol esters 20 29 27 40

Accordingly it appears that the effect of the lipoxygenase treatment can


extend well beyond the primary substrates of the enzyme via the fatty acid
hydroperoxides. When the GLOX treatments were performed in the presence of
linoleic acid, a higher decrease of sterol esters and glycosides (up to 24% and
29%, respectively) was observed but the content of free sterols was not
modified. On the other hand, in the GLOX treatments followed by a hydrogen
peroxide stage, the addition of linoleic acid to the enzymatic reaction caused a
higher removal of sterol esters (from 20% to 39%) and sterol glycosides (from
27% to 40%). A slight decrease of free sterols (from 0 to 7%) was also
observed.
To get further insight into these enzymatic reactions, GLOX reactions with
model compounds (Fig. 2) representative for the main lipophilic extractives in
eucalypt pulp, including free and esterified sterols (sitosterol and cholesteryl
linoleate, respectively), and free fatty acids (linoleic acid) were carried out. The
reactivity of the different lipids was studied by GC and GC-MS. The content of
cholesteryl linoleate decreased about 20% while the content of free sitosterol
remained unchanged by the GLOX treatment. In addition, linoleic acid was
completely degraded, as expected. When the GLOX reactions were performed
with a mixture of the three model compounds, reductions of 83% and 25% of
cholesteryl linoleate and sitosterol, respectively, were observed in the reactions
of lipoxygenase, suggesting that peroxidation reactions could mediate the co-
oxidation of these model lipids. The co-oxidation of sitosterol observed in the
reaction of GLOX with the model mixture was not observed in the eucalypt pulp
treatment. One potential reason for the limited effect of the treatment on the free
sterols may be the predominant localization of these compounds inside the pulp
elements (Speranza et al. 2002).
In the reaction of lipoxygenase with mixtures of the three model lipids,
oxidative derivatives of cholesteryl linoleate and free sitosterol were observed,
and these were more evident after a hydrogen peroxide stage (Fig. 3). The
chemical structures of the oxidation products identified are shown in Fig. 4. The
reaction products of cholesteryl linoleate were, as mentioned above, cholesta-
3,5-dien-7-one (peak 2) and the cholesteryl ester core aldehyde (peak 6).
Oxidized derivatives of free sitosterol, namely 7E-hydroxysitosterol (peak 4)

261
5. Resultados y discusin

(and traces of 7-hydroxysitosterol) and 7-ketositosterol (peak 5) can also be


observed in Fig. 3. These oxidized derivatives were also identified in the
enzymatic reactions of these model lipids with laccase in the presence of 1-
hydroxybenzotriazole (HBT) as redox mediator (Molina et al. 2008).

OH HO
a b

OH

d e
Figure 2. Chemical structures of the model compounds representative for main paper pulp
lipophilic extractives used in the enzymatic reactions: (a) linoleic acid; (b) sitosterol; (c)
cholesteryl linoleate; (d) nonacosane and (e) octacosanol.

3.2.2. Flax pulp treatments


The percentage of removal of lipophilic extractives from flax pulp by the GLOX
treatment is shown in Table 2. All lipophilic extractives from flax pulp
decreased significantly in the several GLOX treatments. Surprisingly, the
content of free sterols in flax pulp decreased up to 55% after the GLOX
treatments, contrasting with the lack of removal in eucalypt pulp where no effect
on free sterols was observed. A high removal of alkanes (up to 48%), fatty
alcohols (up 60%) and sterol glycosides (up to 65%) was also observed. The
sterols present in flax is likely distributed throughout the surface layers of the
fibers alongside the other lipophilic components and may be partially co-
solubilized with them, following the reaction with the lipoxygenase on the cis-
cis-pentadiene structures present in flax pulp.
The addition of linoleic acid to these enzymatic reactions caused different
effects upon the different lipophilic compounds. A higher removal of alkanes
and sterol glycosides was observed by the addition of linoleic acid whereas the
contrary happened with fatty alcohols and free sterols. In the GLOX treatments
of flax pulp followed by a hydrogen peroxide stage, the addition of linoleic
caused an increase in the removal of all the lipophilic extractives.

262
5. Resultados y discusin

100% 1 3
(a)

Relative response

4 6 8 10 12 14 16
Retention time (minutes)

100%
(b)
3
Relative response

5
4 7
2 6

4 6 8 10 12 14 16
Retention time (minutes)

Figure 3. Behaviour of a mixture of different model compounds (linoleic acid, sitosterol,


cholesteryl linoleate) representative for eucalypt pulp lipophilic extractives after enzymatic
treatment with lipoxygenase. Shown are the GC chromatograms of silylated model
compounds before treatment (a), and products after reaction with lipoxygenase (b). Peak
identification: 1, linoleic acid; 2, cholesta-3,5-dien-7-one; 3, sitosterol; 4, 7E-
hydroxysitosterol; 5, 7-ketositosterol; 6, cholesteryl 9-oxononanoate; and 7, cholesteryl
linoleate.

Reactions of GLOX with mixtures of four model compounds (Fig. 2)


representative for the main lipophilic extractives in flax pulp, including alkanes
(nonacosane), fatty alcohols (octacosanol), free sterols (sitosterol), and free fatty
acids (linoleic acid) were also carried out. The GC and GC-MS analyses showed
reductions of 100, 51, 65 and 55% of linoleic acid, nonacosane, octacosanol and
sitosterol, respectively (Fig. 5) after the GLOX treatment in agreement with the
results observed in pulps, suggesting that peroxidation reactions could mediate
the co-oxidation of these lipids. Similar findings were reported in the removal of
alkanes and fatty alcohols from flax pulps with laccase in the presence of HBT
as mediator (Molina et al. 2008). The fact that lipid radicals generated from
peroxidation of unsaturated lipids such as linoleic acid participate in the

263
5. Resultados y discusin

oxidation of the less reactive lipophilic compounds can be used as a way to


remove lipophilic extractives from pulps where linoleic acid is present using
GLOX.

Sitosterol oxidation products:

HO O
HO OH
f g

Cholesteryl linoleate oxidation products:

O O
O H O
h i

Figure 4. Chemical structures of the main oxidized derivatives identified after the
lipoxygenase reactions with free sterols (sitosterol) and sterol esters (cholesteryl linoleate): (f)
7-ketositosterol; (g) 7E-hydroxysitosterol; (h) cholesta-3,5-dien-7-one and (i) cholesteryl 9-
oxononanoate.

Table 2. Removal (percentage of reduction) of main lipophilic extractives in flax pulp


after treatment with the G. graminis lipoxygenase (GLOX) in the absence and presence of
linoleic acid (LA) without and with a subsequent H2O2 stage (P)
Without H2O2 With H2O2
GLOX GLOX/LA GLOX-P GLOX/LA-P
Alkane (C27) 26 39 21 46
Alkane (C29) 48 54 35 55
Fatty alcohol (C26) 48 44 45 50
Fatty alcohol (C28) 55 44 42 49
Fatty alcohol (C30) 61 43 52 52
Free sterols 55 30 16 34
Sterol glycosides 65 51 45 71

264
5. Resultados y discusin

3.4. Pulp properties after lipoxygenase treatment


In addition to the enzymatic removal of lipophilic extractives from pulps, the
effect of the lipoxygenase treatment on some selected properties of the pulps
were assessed, including kappa number (a rough estimation of the lignin content
in pulp), brightness and intrinsic viscosity (an estimation of cellulose integrity).
Table 3 shows the results for eucalypt pulp. Pulps with lower kappa number (1.2
points decrease) and increased brightness (2.6 and 3.4 points increase, in the
absence and presence of linoleic acid, respectively) were obtained after
treatment with GLOX and subsequent peroxide stage. No improvement of pulp
brightness (and only a decrease of 0.7-0.8 points in kappa number) was observed
when the enzymatic treatment was not followed by a hydrogen peroxide stage,
revealing the need of a peroxide stage after the lipoxygenase treatment to
improve pulp properties. Pulp viscosity was maintained after the enzymatic
treatment although decreased after the peroxide stage.

Table 3. Properties of eucalypt pulp treated with the G. graminis lipoxygenase (GLOX)
in the absence and presence of linoleic acid (LA) before and after a H2O2 stage (P), and
control without enzyme
Control GLOX GLOX/LA
initial P initial P initial P
Kappa number 13.5 10.7 12.8 9.5 12.9 9.5
Brightness (% ISO) 44.0 55.9 43.8 58.5 43.9 59.3
Intrinsic viscosity (mL/g) 1140 925 1148 800 1143 767

In the case of flax pulp (Table 4), an increase of 0.8 and 1.2 points of ISO
brightness was achieved after the enzymatic treatment followed by the hydrogen
peroxide stage, in the absence and presence of linoleic acid, respectively (no
increase of brightness was observed before the hydrogen peroxide stage). No
significant decrease of kappa number (lower than 1 point) was observed after the
enzymatic treatment of flax pulp (before the peroxide stage). After peroxide, the
kappa number did not decrease and the viscosity increased.

Table 4. Properties of flax pulp treated with the G. graminis lipoxygenase (GLOX) in
the absence and presence of linoleic acid (LA) before and after a H2O2 stage (P), and
control without enzyme
Control GLOX GLOX/LA
initial P initial P initial P
Kappa number 9.3 5.1 9.1 5.2 8.6 5.5
Brightness (% ISO) 35.1 61.2 35.1 62 35.3 62.4
Intrinsic viscosity (mL/g) 787 535 762 648 779 602

265
5. Resultados y discusin

The better delignification values observed in the GLOX treatment of eucalypt


pulp compared with those of flax pulp treatment can be related with the different
composition and structure of lignin in these two types of pulp. The lignin from
eucalypt pulp is characterized by a high abundance of syringyl units (Ibarra et al.
2005; 2007) and, therefore, it is easier to delignify than lignin from flax pulp
that is mainly constituted by guaiacyl units (Camarero et al. 2004).

2
100%
1 4
(a)

3
Relative response

4 6 8 10 12 14
Retention time (minutes)

100%
(b)
Relative response

2
4

4 6 8 10 12 14
Retention time (minutes)

Figure 5. Behaviour of a mixture of different model compounds (linoleic acid, nonacosane,


octacosanol, sitosterol) representative for flax pulp lipophilic extractives after enzymatic
treatment with lipoxygenase. Shown are the GC chromatograms of silylated model
compounds before treatment (a), and after reaction with lipoxygenase (b). Peak identification:
1, linoleic acid: 2, nonacosane; 3, octacosanol; and 4, sitosterol.

4. Conclusions
The potential of the lipoxygenase from G. graminis to remove lipophilic
extractives from a hardwood (eucalypt) and a nonwood (flax) pulp has been
studied. A removal up to 40% of esterified and glycosylated sterols was

266
5. Resultados y discusin

achieved by the GLOX treatment of eucalypt pulps, while only a decrease of


10% of free sterols was observed. Higher decreases (up to 70%) of lipophilic
extractives from flax pulp were produced by the lipoxygenase treatment,
including free sterols that decreased 55%. In addition, some pulp properties
were determined on the enzymatically treated pulps observing a significant
increase of brightness and decrease of kappa number for eucalypt pulp following
subsequent peroxide treatment, while only limited brightness enhancement was
observed for flax pulp. Given the significant alkaline activity of lipoxygenases it
is suggested that these may have potential use in brown stock treatment as a
means for reducing down-stream deposition problems as well as reducing the
amount of chemicals in the subsequent bleaching stages. Further studies are
needed to gain insight into the chemistry of the reactions of the lipoxygenase
with different lipids, and in the potential applicability of this treatment in the
pulp and paper industry.

Acknowledgements
This study has been supported by the Spanish projects BIO2007-28719-E and
AGL2008-00709 and the EU project BIORENEW (NMP2-CT-2006-026456).
Novozymes (Bagsvaerd, Denmark) is acknowledged for GLOX supply and
ENCE (Pontevedra, Spain) and CELESA (Tortosa, Spain) for eucalypt and flax
paper pulp samples, respectively. J. Romero (ENCE) and T. Vidal (UPC,
Barcelona, Spain) are acknowledged for pulp properties determination.
References
Back EL, Allen LH (2000) Pitch control, wood resin and deresination. TAPPI
Press, Atlanta
Barfoed M (2000) Methods of hydrolyzing cholesterol esters by using a
Pseudomonas fragi cholesterol esterase. Patent (International) WO9423052:
Borch K, Franks N, Lund H, Xu H, Luo J (2003) Oxidizing enzymes in the
manufacturing of paper materials. Patent (USA)US 2003-0124710 A1
Bourbonnais R, Paice MG (1990) Oxidation of non-phenolic substrates. An
expanded role for laccase in lignin biodegradation. FEBS Lett 267:99-102
Buchert J, Mustranta A, Tamminen T, Spetz P, Holmbom B (2002)
Modification of spruce lignans with Trametes hirsuta laccase. Holzforschung
56:579-584
Calero-Rueda O, Gutirrez A, del Ro JC, Prieto A, Plou FJ, Ballesteros A,
Martnez AT, Martnez MJ (2004) Hydrolysis of sterol esters by an esterase
from Ophiostoma piceae: Application for pitch control in pulping of
Eucalyptus globulus wood. Intern J Biotechnol 6:367-375

267
5. Resultados y discusin

Call H-P (1994) Verfahren zur Vernderung, Abbau oder Bleichen von Lignin,
ligninhaltigen Materialien oder hnlichen Stoffen. Patent (International) WO
94/29510:
Camarero S, Garca O, Vidal T, Colom J, del Ro JC, Gutirrez A, Gras JM,
Monje R, Martnez MJ, Martnez AT (2004) Efficient bleaching of non-wood
high-quality paper pulp using laccase-mediator system. Enzyme Microb
Technol 35:113-120
del Ro JC, Gutirrez A, Gonzlez-Vila FJ (1999) Analysis of impurities
occurring in a totally-chlorine free-bleached Kraft pulp. J Chromatogr 830:227-
232
del Ro JC, Gutirrez A, Gonzlez-Vila FJ, Martn F, Romero J (1998)
Characterization of organic deposits produced in the Kraft pulping of
Eucalyptus globulus wood. J Chromatogr 823:457-465
del Ro JC, Romero J, Gutirrez A (2000) Analysis of pitch deposits produced in
Kraft pulp mills using a totally chlorine free bleaching sequence. J Chromatogr
A 874:235-245
Fujita Y, Awaji H, Taneda H, Matsukura M, Hata K, Shimoto H, Sharyo M,
Sakaguchi H, Gibson K (1992) Recent advances in enzymic pitch control.
Tappi J 75 (4):117-122
Gutirrez A, del Ro JC (2001) Gas chromatography-mass spectrometry
demonstration of steryl glycosides in eucalypt wood, kraft pulp and process
liquids. Rapid Commun Mass Spectrom 15:2515-2520
Gutirrez A, del Ro JC (2003a) Lipids from flax fibers and their fate in alkaline
pulping. J Agric Food Chem 51:4965-4971
Gutirrez A, del Ro JC (2003b) Lipids from flax fibers and their fate in alkaline
pulping (Vol 51, pg 4965, 2003). J Agric Food Chem 51:6911-6914
Gutirrez A, del Ro JC (2005a) Chemical characterization of pitch deposits
produced in the manufacturing of high-quality paper pulps from hemp fibers.
Bioresource Technol 96:1445-1450
Gutirrez A, del Ro JC (2005b) Pitch episodes produced during the
manufacturing of high-quality paper pulps from hemp fibers. Bioresour
Technol (submitted):
Gutirrez A, del Ro JC, Gonzlez-Vila FJ, Martn F (1998) Analysis of
lipophilic extractives from wood and pitch deposits by solid-phase extraction
and gas chromatography. J Chromatogr 823:449-455
Gutirrez A, del Ro JC, Gonzlez-Vila FJ, Martn F (1999) Chemical
composition of lipophilic extractives from Eucalyptus globulus Labill. wood.
Holzforschung 53:481-486

268
5. Resultados y discusin

Gutirrez A, del Ro JC, Ibarra D, Rencoret J, Romero J, Speranza M, Camarero


S, Martnez MJ, Martnez AT (2006a) Enzymatic removal of free and
conjugated sterols forming pitch deposits in environmentally sound bleaching
of eucalypt paper pulp. Environ Sci Technol 40:3416-3422
Gutirrez A, del Ro JC, Martnez AT (2009) Microbial and enzymatic control
of pitch in the pulp and paper industry. Appl Microbiol Biotechnol 82:1005-
1018
Gutirrez A, del Ro JC, Martnez MJ, Martnez AT (2001a) The
biotechnological control of pitch in paper pulp manufacturing. Trends
Biotechnol 19:340-348
Gutirrez A, del Ro JC, Rencoret J, Ibarra D, Martnez AT (2006b) Main
lipophilic extractives in different paper pulp types can be removed using the
laccase-mediator system. Appl Microbiol Biotechnol 72:845-851
Gutirrez A, Rencoret J, Ibarra D, Molina S, Camarero S, Romero J, del Ro JC,
Martnez AT (2007) Removal of lipophilic extractives from paper pulp by
laccase and lignin-derived phenols as natural mediators. Environ Sci Technol
41:4124-4129
Gutirrez A, Romero J, del Ro JC (2001b) Lipophilic extractives from
Eucalyptus globulus pulp during kraft cooking followed by TCF and ECF
bleaching. Holzforschung 55:260-264
Hamberg M, Su C, Oliw E (1998) Manganese lipoxygenase - Discovery of a
bis-allylic hydroperoxide as product and intermediate in a lipoxygenase
reaction. J Biol Chem 273:13080-13088
Ibarra D, Chvez MI, Rencoret J, del Ro JC, Gutirrez A, Romero J, Camarero
S, Martnez MJ, Jimnez-Barbero J, Martnez AT (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
Ibarra D, del Ro JC, Gutirrez A, Rodrguez IM, Romero J, Martnez MJ,
Martnez AT (2005) Chemical characterization of residual lignins from
eucalypt paper pulps. J Anal Appl Pyrolysis 74:116-122
International Organisation for Standardization Documentation and Information
(ISO) (2003) ISO Standards Collection on CD-ROM. Paper, board and pulps,
second edn. ISO, Geneva
Kapich AN, Korneichik TV, Hatakka A, Hammel KE (2010) Oxidizability of
unsaturated fatty acids and of a non-phenolic lignin structure in the manganese
peroxidase-dependent lipid peroxidation system. Enzyme Microb Technol
46:136-140

269
5. Resultados y discusin

Kontkanen H, Tenkanen M, Fagerstrom R, Reinikainen T (2004)


Characterisation of steryl esterase activities in commercial lipase preparations.
J Biotechnol 108:51-59
Marques G, del Rio JC, Gutirrez A (2010) Lipophilic extractives from several
nonwoody lignocellulosic crops (flax, hemp, sisal, abaca) and their fate during
alkaline pulping and TCF/ECF bleaching. Bioresource Technol 101:260-267
Molina S, Rencoret J, del Ro JC, Lomascolo A, Record E, Martnez AT,
Gutirrez A (2008) Oxidative degradation of model lipids representative for
main paper pulp lipophilic extractives by the laccase-mediator system. Appl
Microbiol Biotechnol 80:211-222
Prigge ST, Boyington JC, Faig M, Doctor KS, Gaffney BJ, Amzel LM (1997)
Structure and mechanism of lipoxygenases. Biochemie 79:629-636
Riva S (2006) Laccases: blue enzymes for green chemistry. Trends Biotechnol
24:219-226
Rodrguez-Couto S, Toca JL (2006) Industrial and biotechnological applications
of laccases: A review. Biotechnol Adv 24:500-513
Speranza M, Martnez MJ, Gutirrez A, del Ro JC, Martnez AT (2002) Wood
and pulp localization of sterols involved in pitch deposition using filipin
fluorescent staining. J Pulp Paper Sci 28:292-297
Su C, Oliw EH (1998) Manganese lipoxygenase - Purification and
characterization. J Biol Chem 273:13072-13079
Widsten P, Kandelbauer A (2008) Laccase applications in the forest products
industry: A review. Enzyme Microb Technol 42:293-307
Zhang X, Nguyen D, Paice MG, Tsang A, Renaud S (2007) Degradation of
wood extractives in thermo-mechanical pulp by soybean lipoxygenase.
Enzyme Microb Technol 40:866-873
Zhang X, Renaud S, Paice M (2005) The potential of laccase to remove
extractives present in pulp and white water from TMP newsprint mills. J Pulp
Paper Sci 31:175-180
Zhang X, Stebbing DW, Saddler JN, Beatson RP, Kruus K (2000) Enzyme
treatments of the dissolved and colloidal substances present in mill white water
and the effects on the resulting paper properties. J Wood Chem Technol
20:321-335

270
5. Resultados y discusin

271
6

Yute (Corchorus capsularis)


6. Conclusiones

CONCLUSIONES

En la presente Tesis se ha estudiado la composicin qumica de los


principales constituyentes (lignina, lpidos y hemicelulosas) de diferentes
cultivos lignocelulsicos utilizados como materia prima para la fabricacin de
pastas de celulosa de alta calidad, as como su comportamiento durante los
procesos de pasteado y blanqueo. Adems, se han estudiado dos procedimientos
biotecnolgicos que incluyen el uso de enzimas fngicas para la eliminacin de
lignina y lpidos residuales en pastas. Las principales conclusiones obtenidas se
citan a continuacin:

1. En general, las materias primas estudiadas se caracterizan por presentar


un alto contenido en polisacridos y bajo en lpidos y lignina, lo que es,
en principio, favorable para el proceso de produccin de pasta de
celulosa.

2. El contenido y composicin de las diferentes clases de lpidos vara


entre las distintas fibras estudiadas. Las fibras provenientes de tallos se
caracterizan por un alto contenido en cidos grasos y, en particular, las
fibras de lino, kenaf y yute poseen tambin un alto contenido en ceras.
Por otro lado, las fibras provenientes de hojas adems de tener un alto
contenido en cidos grasos, poseen un alto contenido en esteroles (sisal
y abac) y ceras (curau).

3. Las diferentes clases de lpidos muestran distinto comportamiento


durante los procesos de coccin y blanqueo. Durante la coccin sosa-
AQ, los compuestos esterificados se hidrolizan, los cidos grasos se
disuelven y los aldehdos se eliminan, mientras que los compuestos
neutros (esteroles, alcanos, alcoholes, cetonas) sobreviven a la coccin.
Los esteroles libres que sobreviven a la coccin alcalina se degradan en
el blanqueo ECF mientras que permanecen prcticamente inalterados en
el blanqueo TCF, pudiendo originar problemas de deposicin de pitch en
estos procesos. Los glicsidos de esteroles, por otra parte, se eliminan
tanto en el blanqueo TCF como ECF. Los cidos grasos saturados,
alcoholes y alcanos sobreviven tanto a la coccin como al blanqueo ECF
y TCF, y por tanto pueden originar tambin problemas de pitch.

4. La composicin de las ligninas de las fibras liberianas de kenaf y yute,


as como las de todas las fibras de hojas (sisal, abac y curau) son
fundamentalmente de tipo S. Por el contrario, la lignina de las fibras de
camo, lino y caa comn tienen un predominio de unidades de tipo G.
A pesar de que las fibras de lino y camo presentan un bajo contenido

273
6. Conclusiones

en lignina (< 5%), son ms difciles de deslignificar puesto la lignina de


tipo G es ms condensada.

5. Los principales enlaces entre las unidades de lignina en todas las fibras
estudiadas son de tipo aril-ter -O-4. Tambin estn presentes unidades
de tipo resinol -, fenilcumarano -5/-O-4 y espirodienona -1/-O-
. La mayor proporcin de enlaces -O-4 se encuentra en las ligninas
del kenaf, sisal, abac y curau, las cuales al tener tambin mayor
proporcin de unidades S, son ms fcilmente deslignificables.

6. Las ligninas de kenaf, sisal, abac y curau estn extensamente aciladas


en el carbono  de la cadena lateral (con grupos acetatos y/o p-
cumaratos) y principalmente sobre unidades S. Se demostr que la
acilacin de la lignina tiene lugar a nivel de monmero y que el sinapil
acetato (y otros monmeros acilados) se comportan como autnticos
monmeros de la lignina. Adems, estas ligninas estn muy enriquecidas
en unidades S y en enlaces -O-4, y con poca presencia de enlaces
condensados, lo que las hace extremadamente lineales.

7. Las hemicelulosas de las fibras liberianas presentan una mayor


variabilidad en cuanto a su composicin en azcares neutros que las
fibras procedentes de hojas. As, mientras que en las fibras de lino y
camo predominan la manosa y la galactosa, en las de kenaf y yute el
monosacrido predominante es la xilosa. Por otro lado, en todas las
fibras de hojas estudiadas (sisal, abac y curau) se observ un
predominio de xilosa.

8. Las hemicelulosas de sisal se caracterizan por estar constituidas


fundamentalmente por un glucuronoxilano acetilado, cuya cadena
principal est formada por unidades de -D-xilopiranosa parcialmente
ramificada con residuos glucuronosilos (MeGlcpA y GlcpA). El 61%
mol. de las unidades de xilopiranosa estn acetiladas, principalmente en
la posicin O-3 y O-2 (39% mol. y 13% mol. respectivamente) y di-
acetiladas (9% mol.) tambin en esas posiciones.

9. El heteroxilano de sisal sufre una despolimerizacin y una


desacetilacin significativas durante el proceso de pasteado, en el que
los residuos MeGlcpA y GlcpA se eliminan parcialmente o se
convierten en HexA y los grupos acetilo se hidrolizan mayoritariamente.
Durante el proceso de blanqueo, una pequea fraccin de los xilanos
asociados a la lignina residual se eliminan y los grupos acetilo residuales
se eliminan completamente. El comportamiento de los residuos de HexA
durante los procesos de blanqueo TCF y ECF son diferentes. Durante el
blanqueo TCF, una pequea proporcin de los residuos MeGlcpA

274
6. Conclusiones

existentes en la pasta se convierten en HexA mientras que durante el


blanqueo ECF todos los HexA son eliminados por la accin del dixido
de cloro.

10. El procedimiento biotecnolgico basado en la utilizacin de un POM y


la peroxidasa verstil (VP) del hongo Pleurotus eryngii muestra una
gran eficacia para eliminar la lignina residual en pastas de celulosa. El
POM en su forma oxidada es altamente selectivo para la deslignificacin
y se reoxida completamente por la VP en tiempos muy cortos. De esta
manera, es posible la sustitucin de etapas de blanqueo que usan dixido
de cloro por tratamientos POM-VP-POMreox pudiendo utilizarse sistemas
de este tipo en procesos industriales.

11. El procedimiento biotecnolgico basado en el uso de la lipoxigenasa del


hongo Gaeumannomyces graminis muestra una gran eficacia para
eliminar los lpidos residuales responsables de los problemas de pitch
durante la produccin de pastas de celulosa, especialmente en el caso de
pastas de lino. Tambin se observa una mejora de algunas propiedades
de las pastas (especialmente en el caso de pastas de eucalipto) como es
el aumento de la blancura y la disminucin del nmero kappa.

En conclusin, el estudio de la composicin qumica de los cultivos


lignocelulsicos utilizados como materia prima para la fabricacin de pastas
de celulosa as como la evolucin de sus principales componentes durante los
procesos de coccin y blanqueo, contribuye a optimizar su aprovechamiento
industrial mediante tecnologas menos contaminantes. Este conocimiento
contribuir a un aprovechamiento industrial ms sostenible de estos
materiales lignocelulsicos as como al desarrollo de nuevas especies de
inters socioeconmico.

275
7

Fibras liberianas de kenaf (Hibiscus cannabinus)


7. Anexos

ANEXOS

Anexo 1. Peso de pasta ideal (c/  7,5% humedad) para la determinacin del
ndice kappa. Adaptado de un documento realizado por Armindo Gaspar de la
Universidad de Aveiro.

IK peso ideal/g peso seco/g


70,0 0,190 0,176
65,0 0,204 0,189
60,0 0,221 0,205
55,0 0,242 0,224
50,0 0,266 0,246
45,0 0,295 0,273
40,0 0,332 0,308
35,0 0,380 0,351
30,0 0,443 0,410
25,0 0,531 0,492
20,0 0,664 0,615
19,0 0,699 0,647
18,0 0,738 0,683
17,0 0,781 0,724
16,0 0,830 0,769
15,0 0,886 0,820
14,0 0,949 0,879
13,0 1,022 0,946
12,0 1,107 1,025
11,0 1,208 1,118
10,0 1,328 1,230
9,0 1,476 1,367
8,0 1,661 1,538
7,0 1,898 1,757
6,0 2,214 2,050
5,0 2,657 2,460
4,0 3,321 3,705
3,0 4,428 4,100
2,0 6,642 6,150
1,0 13,284 12,300

277
7. Anexos

Anexo 2. Factores f de correccin del consumo de permanganato usado en la


determinacin del ndice kappa ( Tappi 2006).

f+ 0 1 2 3 4 5 6 7 8 9
30 0,958 0,960 0,962 0,964 0,966 0,968 0,970 0,973 0,975 0,997
40 0,979 0,981 0,983 0,985 0,987 0,989 0,991 0,994 0,996 0,998
50 1,000 1,002 1,004 1,006 1,099 1,011 1,013 1,015 1,017 1,019
60 1,002 1,024 1,026 1,028 1,030 1,033 1,035 1,037 1,039 1,042
70 1,004

278
7. Anexos

Anexo 3. Peso de pasta ideal (c/  7,5% humedad) para la determinacin de la


viscosidad intrnseca. Adaptado de un documento realizado por Armindo Gaspar
de la Universidad de Aveiro.

Volumen del frasco/ml


Viscosidad 56 58 61
1400 0,129 0,134 0,141
1350 0,134 0,139 0,146
1300 0,139 0,144 0,152
1250 0,145 0,150 0,158
1200 0,151 0,156 0,165
1150 0,158 0,163 0,172
1100 0,165 0,171 0,179
1050 0,173 0,179 0,188
1000 0,181 0,188 0,197
950 0,191 0,198 0,208
900 0,201 0,209 0,219
850 0,213 0,221 0,232
800 0,227 0,235 0,247
750 0,242 0,250 0,263
700 0,259 0,268 0,282

279
7. Anexos

Anexo 4. Valores del producto []C para diferentes valores de rel


(Scandinavian Pulp Paper and Board Committee 1994).

0,00 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08 0,09
1,0 0,000 0,010 0,020 0,030 0,040 0,049 0,059 0,069 0,078 0,088
1,1 0,097 0,107 0,116 0,125 0,134 0,144 0,153 0,162 1,171 0,180
1,2 0,189 0,198 0,207 0,216 0,224 0,233 0,242 0,250 0,259 0,268
1,3 0,276 0,285 0,293 0,302 0,310 0,318 0,326 0,335 0,343 0,351
1,4 0,359 0,367 0,375 0,383 0,391 0,399 0,407 0,415 0,423 0,431
1,5 0,438 0,446 0,454 0,462 0,469 0,477 0,484 0,492 0,499 0,507
1,6 0,514 0,522 0,529 0,537 0,544 0,551 0,558 0,566 0,573 0,580
1,7 0,587 0,594 0,601 0,608 0,615 0,622 0,629 0,636 0,643 0,650
1,8 0,657 0,664 0,671 0,678 0,684 0,691 0,698 0,705 0,711 0,718
1,9 0,725 0,731 0,738 0,744 0,751 0,757 0,764 0,770 0,777 0,783

2,0 0,790 0,796 0,802 0,809 0,815 0,821 0,827 0,834 0,840 0,846
2,1 0,852 0,858 0,865 0,871 0,877 0,883 0,889 0,895 0,901 0,907
2,2 0,913 0,919 0,925 0,931 0,937 0,943 0,949 0,954 0,960 0,966
2,3 0,972 0,978 0,983 0,989 0,995 1,001 1,006 1,012 1,018 1,023
2,4 1,029 1,035 1,040 1,046 1,051 1,057 1,062 1,068 1,073 1,079
2,5 1,084 1,090 1,095 1,101 1,106 1,111 1,117 1,122 1,127 1,133
2,6 1,138 1,143 1,149 1,154 1,159 1,164 1,170 1,175 1,180 1,185
2,7 1,190 1,196 1,201 1,206 1,211 1,216 1,221 1,226 1,231 1,236
2,8 1,241 1,246 1,251 1,256 1,261 1,266 1,271 1,276 1,281 1,286
2,9 1,291 1,296 1,301 1,306 1,310 1,316 1,320 1,325 1,330 1,335

3,0 1,339 1,344 1,349 1,354 1,358 1,363 1,368 1,373 1,377 1,382
3,1 1,387 1,391 1,396 1,401 1,405 1,410 1,414 1,419 1,424 1,428
3,2 1,433 1,437 1,442 1,446 1,451 1,455 1,460 1,464 1,469 1,473
3,3 1,478 1,482 1,487 1,491 1,496 1,500 1,504 1,509 1,513 1,517
3,4 1,522 1,526 1,531 1,535 1,539 1,544 1,548 1,552 1,556 1,561
3,5 1,565 1,569 1,573 1,578 1,582 1,586 1,590 1,595 1,599 1,603
3,6 1,607 1,611 1,615 1,620 1,624 1,628 1,632 1,636 1,640 1,644
3,7 1,648 1,653 1,657 1,661 1,665 1,669 1,673 1,677 1,681 1,685
3,8 1,689 1,693 1,697 1,701 1,705 1,709 1,713 1,717 1,721 1,725
3,9 1,729 1,732 1,736 1,740 1,744 1,748 1,752 1,756 1,760 1,764

4,0 1,767 1,771 1,775 1,779 1,783 1,787 1,790 1,794 1,798 1,802
4,1 1,806 1,809 1,813 1,817 1,821 1,824 1,828 1,832 1,836 1,839
4,2 1,843 1,847 1,851 1,854 1,858 1,862 1,865 1,869 1,873 1,876
4,3 1,880 1,884 1,887 1,891 1,894 1,898 1,902 1,905 1,909 1,912
4,4 1,916 1,920 1,923 1,927 1,930 1,934 1,937 1,941 1,944 1,948

280
7. Anexos

0,00 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08 0,09
4,5 1,952 1,955 1,959 1,962 1,966 1,969 1,973 1,976 1,979 1,983
4,6 1,986 1,990 1,993 1,997 2,000 2,004 2,007 2,010 2,014 2,017
4,7 2,021 2,024 2,028 2,031 2,034 2,038 2,041 2,044 2,048 2,051
4,8 2,054 2,058 2,061 2,064 2,068 2,071 2,074 2,078 2,081 2,084
4,9 2,088 2,091 2,094 2,098 2,101 2,104 2,107 2,111 2,114 2,117

5,0 2,120 2,124 2,127 2,130 2,133 2,137 2,140 2,143 2,146 2,149
5,1 2,153 2,156 2,159 2,162 2,165 2,168 2,172 2,175 2,178 2,181
5,2 2,184 2,187 2,191 2,194 2,197 2,200 2,203 2,206 2,209 2,212
5,3 2,215 2,219 2,222 2,225 2,228 2,231 2,234 2,237 2,240 2,243
5,4 2,246 2,249 2,252 2,255 2,258 2,261 2,264 2,267 2,270 2,273
5,5 2,276 2,280 2,283 2,286 2,288 2,291 2,294 2,297 2,300 2,303
5,6 2,306 2,309 2,312 2,315 2,318 2,321 2,324 2,327 2,330 2,333
5,7 2,336 2,339 2,342 2,345 2,347 2,350 2,353 2,356 2,359 2,362
5,8 2,365 2,368 2,371 2,374 2,376 2,379 2,382 2,385 2,388 2,391
5,9 2,394 2,396 2,399 2,402 2,405 2,408 2,411 2,413 2,416 2,419

6,0 2,422 2,425 2,427 2,430 2,433 2,436 2,439 2,441 2,444 2,447
6,1 2,450 2,452 2,455 2,458 2,461 2,463 2,466 2,469 2,472 2,475
6,2 2,477 2,480 2,783 2,485 2,488 2,491 2,494 2,496 2,499 2,502
6,3 2,504 2,507 2,510 2,512 2,515 2,518 2,521 2,523 2,526 2,529
6,4 2,531 2,534 2,537 2,539 2,542 2,545 2,547 2,550 2,552 2,555
6,5 2,558 2,560 2,563 2,566 2,568 2,571 2,573 2,576 2,579 2,581
6,6 2,584 2,587 2,589 2,592 2,594 2,597 2,599 2,602 2,605 2,607
6,7 2,610 2,612 2,615 2,617 2,620 2,623 2,625 2,628 2,630 2,633
6,8 2,635 2,638 2,640 2,643 2,645 2,648 2,651 2,653 2,656 2,659
6,9 2,661 2,663 2,666 2,668 2,671 2,673 2,676 2,678 2,681 2,683

7,0 2,686 2,688 2,690 2,693 2,695 2,698 2,700 2,703 2,705 2,708
7,1 2,710 2,713 2,715 2,718 2,720 2,722 2,725 2,727 2,730 2,732
7,2 2,735 2,737 2,739 2,742 2,744 2,747 2,749 2,752 2,754 2,756
7,3 2,758 2,761 2,764 2,766 2,768 2,771 2,773 2,775 2,778 2,780
7,4 2,783 2,785 2,787 2,790 2,792 2,794 2,797 2,799 2,801 2,804
7,5 2,806 2,809 2,811 2,813 2,816 2,818 2,820 2,823 2,825 2,827
7,6 2,829 2,832 2,834 2,836 2,839 2,841 2,843 2,846 2,848 2,850
7,7 2,853 2,855 2,857 2,859 2,862 2,864 2,866 2,869 2,871 2,873
7,8 2,875 2,878 2,880 2,882 2,885 2,887 2,889 2,891 2,894 2,896
7,9 2,898 2,900 2,903 2,905 2,907 2,909 2,911 2,914 2,916 2,918

281
7. Anexos

0,00 0,01 0,02 0,03 0,04 0,05 0,06 0,07 0,08 0,09
8,0 2,920 2,923 2,295 2,927 2,929 2,932 2,934 2,936 2,938 2,940
8,1 2,943 2,945 2,947 2,949 2,951 2,954 2,956 2,958 2,960 2,962
8,2 2,964 2,967 2,969 2,971 2,973 2,975 2,978 2,980 2,982 2,984
8,3 2,986 2,988 2,991 2,993 2,995 2,997 2,999 3,001 3,003 3,006
8,4 3,008 3,010 3,012 3,014 3,016 3,018 3,020 3,023 3,025 3,027
8,5 3,029 3,031 3,033 3,035 3,037 3,040 3,042 3,044 3,046 3,048
8,6 3,050 3,052 3,054 3,056 3,058 3,061 3,063 3,065 3,067 3,069
8,7 3,071 3,073 3,075 3,077 3,079 3,081 3,083 3,085 3,087 3,090
8,8 3,092 3,094 3,096 3,098 3,100 3,102 3,104 3,106 3,108 3,110
8,9 3,112 3,114 3,116 3,118 3,120 3,122 3,124 3,126 3,128 3,130

9,0 3,132 3,134 3,136 3,138 3,340 3,142 3,144 3,147 3,149 3,151
9,1 3,153 3,155 3,157 3,159 3,161 3,163 3,165 3,166 3,168 3,170
9,2 3,172 3,174 3,176 3,178 3,180 3,182 3,184 3,186 3,188 3,190
9,3 3,192 3,194 3,196 3,198 3,200 3,202 3,204 3,206 3,208 3,210
9,4 3,212 3,214 3,216 3,218 3,220 3,222 3,223 3,225 3,227 3,229
9,5 3,231 3,233 3,265 3,237 3,239 3,241 3,242 3,245 3,247 3,249
9,6 3,250 3,252 3,254 3,256 3,258 3,260 3,262 3,264 3,266 3,268
9,7 3,270 3,271 3,273 3,175 3,277 3,273 3,281 3,283 3,285 3,887
9,8 3,288 3,290 3,292 3,294 3,296 3,298 3,300 3,302 3,303 3,305
9,9 3,307 3,309 3,311 3,313 3,315 3,316 3,318 3,320 3,322 3,324

10 3,326 3,344 3,363 3,381 3,399 3,416 3,434 4,452 3,469 3,487
11 3,504 3,521 3,538 3,554 3,571 3,588 3,604 3,620 3,636 3,653
12 3,669 3,684 3,700 3,716 3,731 3,747 3,762 3,777 3,792 3,807
13 3,822 3,837 3,852 3,866 3,881 3,895 3,910 3,924 3,938 3,952
14 3,966 3,980 3,994 4,008 4,021 4,035 4,048 4,062 4,075 4,088
15 4,101 4,115 4,128 4,141 4,153 4,166 4,179 4,192 4,204 4,217
16 4,229 4,242 4,254 4,266 4,279 4,291 4,303 4,315 4,327 4,339
17 4,351 4,362 4,374 4,386 4,397 4,409 4,420 4,432 4,443 4,455
18 4,466 4,477 4,488 4,499 4,510 4,521 4,532 4,543 4,554 4,565
19 4,576 4,586 4,597 4,608 4,618 4,629 4,639 4,650 4,660 4,670

282