Proceso Kraft
Proceso Kraft
Proceso Kraft
Madrid, 2012
MEMORIA
Madrid, 2011
UNIVERSIDAD COMPLUTENSE DE MADRID
DEPARTAMENTO DE INGENIERÍA QUÍMICA
INFORMAN:
En primer lugar, gracias a mis directores de tesis Dr. Juan Carlos Villar y Dr.
Juan Antonio Martín. Gracias Juan Carlos por haber hecho posible que llevase a cabo
este trabajo en el INIA y por haberme no sólo permitido, sino animado, a realizar
estancias fuera del INIA, para poder completar así mi formación. En estos cuatro años
he aprendido muchas cosas y directa o indirectamente han sido gracias a ti. Gracias por
demostrarme tu confianza en mí para llevar a cabo los diferentes trabajos, experimentos,
etc. En resumen, gracias por creer en mi como futura investigadora.
Gracias a los doctores Jiebing Li y Göran Gellerstedt del KTH (Suecia), Orlando
Rojas y Ewellyn Capanema de NCSU (EEUU), Manuel Jesús Díaz y Juan Carlos
García de la Universidad de Huelva, e Ilari Filpponen y Janne Laine de Aalto University
(Finlandia). Gracias a todos ellos por la buena acogida en sus grupos de investigación,
por su dedicación y paciencia y por todo lo que he aprendido con ellos. Quisiera
expresar mi especial agradecimiento al Dr. Orlando Rojas, ya que él me ha acogido no
una sino dos veces, tanto en NCSU como en Aalto University, guiándome siempre en
mi trabajo con buenas ideas y atendiendo todas mis dudas con increíble rapidez.
Sinceramente, no sé como puedes organizarte tan bien y abarcar todo el trabajo que
llevas a cabo. Eres un ejemplo a seguir.
Gracias también a todas las personas que han sido mi familia, mis amigos y mis
compañeros de trabajo en estas estancias. Gracias a David, Vivi, Cristiane y Helena en
KTH; a Sole, Ingrid, Carlos, Ronalds y toda la banda en NCSU; a Quique, Minerva e
Isabel en Huelva; y a Tiina, Laura, Xiaomeng, Ingrid y Sole (che! vosotras dos otra vez,
y espero que coincidamos muchas más veces!) en Aalto. Sin todos vosotros no hubiese
podido integrarme tan bien tanto dentro como fuera del laboratorio. Habéis hecho que
siempre tenga ganas de irme de estancia, y que la distancia de mis seres queridos sea
mucho más llevadera.
Y por supuesto, gracias a mis compañeros del INIA. Gracias Esteban por tus
ideas, tus reactores, tu ayuda, tu coche (jeje) y tu amistad. Gracias Chema por tener
siempre un ratito para ayudarme, no solo a mí, sino a todos en el laboratorio, por estar
siempre ahí para lo que necesitemos. Gracias Sara por tu sonrisa y por tus
conversaciones, aunque hayamos trabajado poco juntas, creo que nos compenetraríamos
muy bien. Y Luisa!! Aunque llegaste ya con mi tesis a la mitad espero que podamos
seguir compartiendo laboratorio mucho tiempo, y sino, por lo menos sigamos
compartiendo emails, cañas o lo que haga falta! Y gracias a todos los demás, a “los de
arriba” Nuria, Reme, Ester y Carlos, a Serfaty, Ángela, Jesús, Pedro, etc. Todos, de una
forma u otra, habéis puesto vuestro granito de arena.
Y como no, mil gracias a María. Gracias por tu ayuda y apoyo incondicionales,
por tu alegría, tu sinceridad, tu compañerismo, y sobre todo gracias por tu amistad.
Sabes que esta tesis no hubiese sido posible sin ti. Siempre has sacado tiempo de donde
hiciese falta para ayudarme y aconsejarme, aunque tuvieses mil cosas más en la cabeza.
Contigo he aprendido muchísimo y me has contagiado las ganas de investigar, de
trabajar, de seguir en este mundillo. Sin duda puedo decir que estos 4 años no hubiesen
sido lo mismo sin ti. No cambies nunca, y sobre todo, no dejes que te quiten esas ganas
de aprender y de avanzar.
Y por último, gracias a ti, Mario. Gracias por ser como eres, por todo lo que
haces por mí, por hacerme sentir feliz a cada momento. Gracias por las largas
conversaciones a altas horas de la noche, las risas, los abrazos, la complicidad, etc.
Gracias no sólo por no cortarme las alas, sino por volar conmigo. Tú sabes, casi mejor
que yo, lo que necesito siempre. Te necesito a ti. ALB.
ÍNDICE GENERAL
1 - RESUMEN / ABSTRACT............................................................................................. 1
2 - INTRODUCCIÓN ......................................................................................................... 9
3 - OBJETIVOS................................................................................................................ 53
i
4.1.2 - Estudio termogravimétrico ....................................................................... 68
4.4 - Combinación de tratamientos de explosión por vapor y LMS antes del pasteado
como alternativa al biopulpeo con hongos ....................................................... 110
4.4.1 - Biopulpeo basado en explosión por vapor y tratamiento enzimático ..... 111
ii
ANEXOS: Publicaciones .............................................................................................. 169
iii
LISTA DE ABREVIATURAS
v
IFPB Biorefinería Integrada de Productos Forestales (Integrated Forest
Product Biorefinery)
Ind Pasta industrial
INIA Instituto Nacional de Investigación y Tecnología Agraria y
Agrolimentaria
ISO Organización Internacional de Normalización
L Tratamiento enzimático con LMS
Lc Tratamiento enzimático control sin adición de LMS
LCC Complejo lignina - carbohidratos
LiP Lignina peroxidasa
LM-NoO2 Tratamiento enzimático con LMS sin adición de oxígeno
LMO2 Tratamiento enzimático con LMS y oxígeno
LMS Sistema laccasa - mediador
MnP Manganeso peroxidasa
MWL Milled Wood Lignin
nd No disponible
Nº Rev Número de revoluciones
NoE Muestras sin tratamiento de extracción
No-LMO2 Tratamiento enzimático control sin adición de LMS ni oxígeno
NREL Laboratorio Nacional de Energías Renovables del departamento de
Energía de EEUU
OHph Grupos hidroxilo fenólicos
OHpr Grupos hidroxilo primarios
OHsec Grupos hidroxilo secundarios
OMe Grupos metoxilo
P Blanqueo con peróxido de hidrógeno
Relación L/D Relación longitud / diámetro
RMN Resonancia Magnética Nuclear
Rto Rendimiento
S Sulfidez
S Unidad siringilo de la lignina
S0 Factor de severidad del tratamiento de explosión por vapor
SE Explosión por vapor (steam explosion)
vi
SEi Tratamiento “i” de explosión por vapor de Eucalyptus globulus
SEi’ Tratamiento “i” de explosión por vapor de Hesperalo funifera
SEM Microscopio de barrido electrónico
ºSR º Schopper - Riegler
T Temperatura
t Tiempo
TCF Secuencia de blanqueo totalmente libre de cloro
TGA Análisis termogravimétrico
UA Unidades de actividad
W Extracción con agua
X Tratamiento con xilanasa
XL Tratamiento en serie con xilanasa y lacasa
vii
1
RESUMEN / ABSTRACT
1 – Resumen / Abstract
1. RESUMEN / ABSTRACT
3
Proceso Kraft en Biorrefinería
4
1 – Resumen / Abstract
5
Proceso Kraft en Biorrefinería
ABSTRACT
Based on this concept, a chemical pulp mill could be transformed into a forest
biorefinery, where ethanol, bio-polymers and other chemical products could be
produced in addition to pulp. In a conventional Kraft pulp mill, most of the
hemicelluloses and lignin present in the wood initially are extracted by the cooking
liquor and burned in the recovery boiler to produce steam and electricity and to
regenerate the pulping chemicals. These compounds represent about 50% of the wood
weight, and while lignin has a high heating value, hemicelluloses only provide half the
energy that lignin does. Therefore, the hemicelluloses could be used to produce ethanol
or other high value-added products, increasing the total revenue stream for the pulp
mill. To achieve this goal, the hemicelluloses should be extracted from wood prior to
pulping, allowing for their recovery without degradation and their utilization in the
synthesis of other products.
This is the context that frames the present doctoral thesis, whose main objective
is to study and to evaluate the potential integration of the Kraft pulping process into a
lignocellulosic biorefinery. To this end, a steam explosion process (SE) was used as
method for hemicelluloses extraction prior to chemical pulping of Eucalyptus globulus.
Compared to other methods, steam explosion does not use chemical reagents, avoiding
the recovery and/or neutralization costs and reducing the environmental impact.
Furthermore, as it has been proved in this doctoral thesis, this method yields a solid
fraction with a more open structure (greater macropore volume and increased water
retention capacity) that might favor an efficient diffusion of cooking liquor into the
fibers during pulping. On the other hand, depolymerization reactions led to a decrease in
the β-O-4 structure of the lignin remaining in the solid fraction after the SE pre-
treatment. Nevertheless, repolymerization reactions also took place, resulting in the
6
1 – Resumen / Abstract
presence of condensed lignin in the solid fraction. Finally, a 47% of the wood xylan was
extracted, while the cellulose content remained almost constant.
7
Proceso Kraft en Biorrefinería
exploded pulps had lower kappa number, higher brightness and better colorimetric
properties than the biobleached control pulp (without SE pre-treatment). In addition,
there was a reduction in hydrogen peroxide consumption of 20%. Therefore, pulps
obtained by this biorefinery process (SE+Kraft+biobleaching) could be used in
applications for which optical properties are important but not so mechanical strength,
achieving also a reduction in cooking time and chemical consumption and obtaining as
a byproduct a liquid current rich in hemicelluloses.
8
2
INTRODUCCIÓN
2 - Introducción
2. INTRODUCCIÓN
nd: no disponible
1
Hayn et al. (1993)
2
Wiselogel et al. (1996)
3
Henriksson et al. (2007)
4
Perez et al. (2008)
5
Garcia-Aparicio et al. (2006)
11
Proceso Kraft en Biorrefinería
S3
S2
S1
LM
Figura 2.1. Capas de la pared celular de las fibras vegetales. (Côte, 1977)
12
2 - Introducción
Figura 2.2. Representación esquemática de las relaciones entre los principales constituyentes de
la pared secundaria (celulosa, lignina y hemicelulosa) de una angiosperma herbácea. Las
uniones entre estos polímeros se completan mediante puentes intermoleculares formados por los
ácidos p-hidroxicinámicos (Bidlack et al., 1992).
2.1.1 - Celulosa
13
Proceso Kraft en Biorrefinería
2.1.2 - Hemicelulosa
La hemicelulosa actúa junto con la lignina como matriz soporte para las
microfibrillas de celulosa en la pared celular, por tanto se encuentra unida tanto a
celulosa como a lignina por distintos tipos de enlace. Son un grupo de polisacáridos
bien diferenciados de la celulosa, fácilmente extraíbles con álcali y mucho menos
resistentes a la hidrólisis que la celulosa (Tenkanen et al., 1995). A diferencia de la
celulosa, son heteropolímeros que pueden estar ramificados con mono-, di- u
14
2 - Introducción
HOOC
OH 3 C O
HO
OH COH 3 C
O OH
O O HO O O O O O
HO O O O O HO O
O O
OH OH
OH O
O
HOH 2 C
OH OH
HOH 2 C
OH
15
Proceso Kraft en Biorrefinería
2.1.3 - Lignina
CH 3O OH3C CH 3O
OH OH OH
Las unidades fenilpropano están interconectadas entre sí por una serie de enlaces
de tipo éter (C-O-C) y carbono-carbono (C-C), en los que participan tanto los anillos
aromáticos como las cadenas laterales, dando lugar a una serie de subestructuras tales
como: guayacilglicerol-β-aril éter, fenilcumarano, diarilpropano, resinol, bifenilo y
difenil-éter, así como otras menos abundantes (figura 2.6). El enlace intermonomérico
más frecuente, β-O-4’, está presente en la subestructura guayacilglicerol-β-aril éter y es
16
2 - Introducción
El tipo de unidades presentes en la lignina, así como los tipos de unión entre las
mismas puede variar entre los distintos grupos de plantas (coníferas, frondosas y
herbáceas). Incluso en una misma planta la composición de la lignina puede variar en
los distintos tejidos y capas de la pared celular. La lignina de coníferas se forma
mayoritariamente a partir del alcohol coniferílico (unidad G), mientras que la lignina de
frondosas se sintetiza a partir de los alcoholes coniferílico y sinapílico (unidades G y S,
respectivamente). La lignina de las plantas herbáceas está compuesta de unidades H
procedentes del alcohol p-cumarílico, además de las unidades G y S en distintas
proporciones. La existencia de dos, una o ninguna posición libre en las posiciones orto
del grupo fenólico de las unidades H, G y S respectivamente es la causa de la distinta
reactividad de cada tipo de lignina. Así por ejemplo, los dos grupos metoxilo de las
unidades S evitan la condensación de lignina en los procesos de cocción.
17
Proceso Kraft en Biorrefinería
(Howard et al., 2003). Así mismo, su estructura proporciona rigidez y dureza a los
tejidos vegetales, a la vez que minimiza la permeabilidad de dichos tejidos al agua
(Evans y Betts, 1991). Sin embargo, es esta misma complejidad la que hace que el
polímero sea difícilmente biodegradable.
18
2 - Introducción
19
Proceso Kraft en Biorrefinería
Dentro de las especies no madereras, algunas de las más utilizadas son sisal,
lino, cáñamo, abacá, bambú, etc. Estudios recientes han demostrado el potencial de
Hesperaloe funifera para la producción de papel (McLaughlin, 2000; Sanchez et al.,
2010; Wong y McLaughlin, 2000). Las principales características de esta planta son sus
largas y delgadas fibras (longitud 3,5-4,5 mm y anchura 7,0-10 μm) y su bajo contenido
en lignina (10% lignina Klason). Estas características, junto con sus bajos
20
2 - Introducción
requerimientos hídricos (debido a su metabolismo ácido crasuláceo) hacen que sea una
planta con gran potencial papelero en áreas con escasos recursos acuíferos. Tras 5 años
de cultivo, se puede recoger la primera cosecha, y posteriormente se recogen cosechas
cada 3 años. Así, en plantaciones de alta densidad (27.000 plantas por hectárea) se
pueden obtener aproximadamente 20 toneladas de biomasa seca por hectárea y año, tras
la primera cosecha (Wong y McLaughlin, 2000). Por todas estas características, en la
presente tesis doctoral se ha evaluado el uso de H. funifera en el esquema de
biorrefinería forestal planteado, comparando las pastas obtenidas con aquellas obtenidas
a partir de E. globulus.
2.2.2 - Pasteado
21
Proceso Kraft en Biorrefinería
Previo al pasteado, se deben llevar a cabo una serie de etapas comunes para los
distintos tipos de pasteado. Así, los troncos de madera que llegan a la fábrica se
descortezan y se astillan. Las astillas obtenidas se tamizan para seleccionar aquellas de
10-30 mm de longitud y 2-8 mm de espesor. Tanto las cortezas como las astillas de
tamaño más pequeño se llevan a una caldera de biomasa, donde son quemadas para
generar vapor y a partir de él, energía eléctrica. Las astillas seleccionadas se introducen
en el esquema de producción de pasta Kraft que se detalla en la figura 2.7.
Agua de lavado
Pasta
Lavadores de cruda Pasta
Astillas Digestor Kraft Blanqueo
pasta blanqueada
Lejía Verde
Planta de (Na2CO3, Na2S) Residuos,
caustificación impurezas
Figura 2.7. Esquema del proceso de pasteado Kraft, incluyendo el circuito de recuperación de
lejía negra.
22
2 - Introducción
23
Proceso Kraft en Biorrefinería
Tras la cocción, la pasta se tamiza para separar los trozos de madera que hayan
quedado sin digerir, se lava para separar la mezcla de cocción utilizada (lejía negra) y se
envía bien al proceso de blanqueo o a la máquina de producción de papel. La madera sin
digerir se devuelve al digestor o a la caldera de biomasa para quemarla y producir
energía.
Una de las ventajas importantes del método Kraft es la recuperación de sus lejías
negras, que no sólo permite la recuperación de los productos químicos, sino que,
además, aprovecha el poder calorífico de las mismas para generar el vapor necesario
para el proceso, incluso obteniendo un excedente de energía. La lejía negra recogida del
digestor contiene sustancias orgánicas disueltas cuya composición química exacta
depende de las características de la especie de la madera y de las condiciones de la
cocción. Normalmente la lejía negra presenta un contenido en sólidos disueltos del 14-
18%, que debe concentrarse considerablemente previamente a su combustión. Para ello
la lejía negra se concentra mediante una evaporación multiefecto hasta un contenido en
sólidos de 65-75%. A medida que el contenido en sólidos se incrementa se debe tener
en cuenta que la viscosidad de la lejía negra también se incrementa, y puede ser
demasiado alta para ser bombeada. A presión atmosférica el límite se encuentra
alrededor de 72-74% de sólidos. La lejía negra puede ser concentrada hasta un
contenido en sólidos del 80% en superconcentradores, que operan a mayor presión y
temperatura.
24
2 - Introducción
Las cenizas y otras impurezas son eliminadas del proceso por decantación de la
lejía verde. Los lodos de carbonato cálcico procedentes del proceso de caustificación se
separan por filtración de las lejías blancas, se lavan y se calcinan en un horno de cal
para regenerar la cal viva. El óxido de calcio se hace reaccionar con agua y se obtiene
hidróxido de calcio, que se emplea en el proceso de caustificación. La calcinación de los
lodos se efectúa en un horno rotatorio, con aporte de combustible (coque, carbón, fuel o
gas).
2.2.3 - Blanqueo
Para producir pastas químicas de alta calidad con elevados grados de blancura es
necesario utilizar métodos de blanqueo que continúen la deslignificación de la pasta
iniciada en la cocción y reduzcan el color. El blanqueo total de una pasta química se
realiza en sucesivas etapas, utilizando diferentes productos químicos (cloro, hidróxido
sódico, hipoclorito sódico, dióxido de cloro, peróxido de hidrógeno, oxígeno, ozono,
hidrosulfitos de sodio o ditiosulfato sódico) y diferentes condiciones de consistencia,
concentración del reactivo químico, pH, temperatura, tiempo de retención, etc. Las
condiciones y/o reactivos varían en cada etapa pero siempre teniendo en cuenta el
compromiso entre el aumento del grado de blancura y la pérdida de propiedades
mecánicas debido a la degradación de las fibras de celulosa. Se ha observado que se
puede conseguir una mayor blancura con una menor degradación de las fibras si se
aplican varias etapas en serie con cantidades más pequeñas de reactivos y con un lavado
entre etapas. En las primeras etapas se utilizan reactivos que contribuyen a la
25
Proceso Kraft en Biorrefinería
deslignificación de las pastas, mientras en las últimas etapas se utilizan agentes que
eliminen el color residual. El blanqueo se consigue eliminando los compuestos que son
capaces de adsorber luz visible o modificándolos para reducir esta capacidad.
26
2 - Introducción
27
Proceso Kraft en Biorrefinería
solventar las necesidades energéticas del país, y la generación de electricidad reduce los
costes y reduce emisiones de gases de efecto invernadero.
28
2 - Introducción
Alimentación de azúcares
Plataforma de procesado de azúcares
Bioquímica
Residuos
Producción Combustibles,
BIOMASA combinada de productos químicos
calor y electricidad y materiales
Gas limpio
29
Proceso Kraft en Biorrefinería
Resinas furánicas
Furfural Productos químicos
Nylos-6, 6-6
Xilosa (pentosa)
Xilitol
Hemicelulosas
Espesantes
Derivados de las Aditivos
hemicelulosas Estabilizadores
Emulsionantes
Combustible sólido
BIOMASA Lignina Formulación de polímeros
Productos químicos
Hidroximetilfurfural
Productos de fermentación
Glucosa (hexosa) Combustible: bioetanol
Ácidos orgánicos: ác. láctico
Disolventes: acetona
Celulosa
Papel y cartón
Emulsificante
Derivados de la
Estabilizante
celulosa
Gelificante
Dispersante
30
2 - Introducción
31
Proceso Kraft en Biorrefinería
32
2 - Introducción
33
Proceso Kraft en Biorrefinería
Pasta
Extracción de Sólido Lavadores de cruda Pasta
Madera Digestor Kraft Blanqueo
hemicelulosas pasta blanqueada
Líquido Químicos
(hemicelulosas) Lejía blanca Lejía negra
derivados de
(NaOH, Na2S) Lignina
Polímeros
basados en Energía Vapor
azúcares
Gasificador Energía y
Bioetanol de biomasa Vapor
Modificación Biocompuestos
de la madera de Madera
Figura 2.10. Esquema de una Biorrefinería Integrada de Productos Forestales (Van Heiningen,
2006).
34
2 - Introducción
proceso de fermentación. Además, esta fracción líquida también puede contener lignina
y productos de degradación que podrían inhibir los procesos biológicos consecutivos
(hidrólisis enzimática y fermentación). Todos estos factores deben tenerse en cuenta a la
hora de seleccionar y optimizar el método de extracción de hemicelulosas. Así,
tratamientos menos efectivos en la hidrólisis de hemicelulosas pueden ser más
adecuados para el proceso global. En este sentido, los tratamientos con vapor o agua a
temperaturas altas (explosión por vapor y autohidrólisis) pueden ser más adecuados
(Carvalheiro et al., 2008). Por otro lado, comparado con la autohidrólisis, la explosión
por vapor proporciona un residuo sólido con una estructura más abierta, que facilitará la
difusión de reactivos durante el pasteado (Ahvazi et al., 2007).
Agua de lavado
Pasta
Explosión por Sólido Lavadores de cruda Pasta
Astillas Digestor Kraft Blanqueo
vapor pasta blanqueada
Lejía Verde
Planta de (Na2CO3, Na2S) Residuos,
Bioetanol
caustificación impurezas
Figura 2.11. Integración del proceso Kraft en una biorrefinería lignocelulósica basada en la
extracción de hemicelulosas mediante un proceso de explosión por vapor.
35
Proceso Kraft en Biorrefinería
T −100
S0 = log e 14 .75 ∙ t
[1]
36
2 - Introducción
Figura 2.12. Imagen SEM de fibras de roble tras un tratamiento de explosión de vapor a 230ºC
durante 5 minutos. En el recuadro se muestra la imagen de la superficie de la fibra a escala 1μm
(Renneckar et al., 2005).
Cuando no se añaden otros reactivos, los cambios químicos originados son los
mismos que los debidos a un tratamiento con vapor. Así, el vapor a alta temperatura,
debido al alto grado de disociación del agua, provoca la hidrólisis de grupos acetilos de
la hemicelulosa, liberándose ácido acético. Éste ácido disminuye ligeramente el pH del
agua y, a la temperatura del proceso, cataliza la hidrólisis de la hemicelulosa. De esta
forma se produce una autohidrólisis del material lignocelulósico, que origina una
pérdida de hemicelulosa al solubilizarse en la fracción líquida. En la figura 2.13 se
muestra la reacción de hidrólisis de la hemicelulosa y la posterior degradación de la
xilosa a furfural que tiene lugar cuando las condiciones del tratamiento son severas
(Josefsson et al., 2001; Li et al., 2005).
O
O O OH
O H+ H+ O
O OH OH
O Hidrólisis Degradación
OH HO
OH OH
OH
n
Hemicelulosa (xilano) Xilosa Furfural
37
Proceso Kraft en Biorrefinería
OH
OH OH
HO O
O
O H+ O OH H+ O
O OH
OH
O Hidrólisis Degradación
OH HO
OH
OH
OH
n Hidroximethilfurfural
Celulosa Glucosa
(HMF)
38
2 - Introducción
OH
R
O
OH R OH R + HO
R
ón
aci O
O eriz CH3 H3C
+
lim O
spo
O HC O
De O
O O
H3C H + H3C
CH3 - ROH CH3
OH
O O Re
p ol OH
O O im O R
eri H3C
zac
i ón
O
R O
H3C
CH3
O
O
39
Proceso Kraft en Biorrefinería
Figura 2.16. Imágenes SEM de la superficie de la pared celular de bagazo de maíz antes (A) y
después (B y C) de un tratamiento ácido con 0,8% H2SO4 a 150ºc durante 20 minutos. La
imagen C es una ampliación de la región recuadrada en la imagen B (Donohoe et al., 2008).
40
2 - Introducción
Por otro lado, este pre-tratamiento se muestra menos eficaz con las maderas de coníferas
debido a su mayor contenido en lignina y menor número de grupos acetilo, lo que
provoca que el proceso de autohidrólisis no sea tan efectivo. Con este tipo de maderas
es necesario añadir un catalizador ácido, lo que implica mayor generación de productos
tóxicos.
Por otro lado, el proceso de explosión por vapor puede utilizarse para extraer
hemicelulosas y utilizar esta corriente como fuente de hexosas y pentosas que pueden
convertirse en combustibles como etanol, u otros productos de valor añadido (Li et al.,
2010) , como se ha indicado previamente. Además, el residuo sólido obtenido tras el
tratamiento, podría ser pasteado más fácilmente, debido a que su estructura más abierta
(comparado con la madera original) permitirá una mejor difusión de los reactivos de
cocción en las fibras (Ahvazi et al., 2007). Por tanto este proceso podría utilizarse como
pre-tratamiento al pasteado químico para aumentar los beneficios económicos de una
fábrica Kraft convencional, transformando esta fábrica en una biorrefinería integrada de
productos forestales (Ragauskas et al., 2006; Van Heiningen, 2006).
En las últimas décadas, la biotecnología está siendo una de las áreas más
estudiadas en la industria de la pasta y el papel. A través del uso de microorganismos y
enzimas, se busca el desarrollo de tecnologías limpias que mejoren y optimicen los
tradicionales procesos de pasteado y blanqueo y que reduzcan tanto los consumo de
energía y de reactivos químicos como las corrientes contaminantes generadas. Además,
la búsqueda de nuevos combustibles menos contaminantes obtenidos a partir de fuentes
renovables, ha potenciado el uso de la biotecnología en la producción de etanol a partir
de material lignocelulósico (bioetanol), cambiando el concepto tradicional de la
industria de la pasta y el papel hacia un concepto más amplio de biorrefinerías
forestales, como se ha señalado anteriormente.
41
Proceso Kraft en Biorrefinería
Agua de
Enzimas Biopulpeo Enzimas Bioblanqueo
lavado
Levaduras
Fermentación
Bioetanol
42
2 - Introducción
Celulasas
Las celulasas son enzimas que hidrolizan el enlace 1,4-β-D-glucosidico de las
cadenas de celulosa. Se dividen en tres grupos principales: endoglucanasas (EG),
exoglucanasas o celobiohidrolasas (CBH) y β-glucosidasas (BG).
β-glucosidasas
Exoglucanasas Endoglucanasa
(CBH-II)
Exoglucanasas
(CBH-I)
Hemicelulasas: Xilanasas
Debido a la variedad de enlaces entre los diferentes azúcares de las
hemicelulosas, existe un gran número de grupos de hemicelulasas con actividad
específica para cada tipo de enlace. Así por ejemplo, las xilanasas o endoxilanasas
atacan la cadena principal de las hemicelulosas y las β-xilosidasas hidrolizan
xilooligosacáridos a xilosa. Otras enzimas rompen las ramificaciones, como las α-D-
glucuronosidasas y las α-L-arabinosidasas que hidrolizan los enlaces entre los xilanos y
los ácidos acéticos y fenólicos (Teeri y Henriksson, 2007). La hidrólisis de la
43
Proceso Kraft en Biorrefinería
Ligninasas
Las enzimas que juegan un papel clave en la degradación de la lignina son
lignina peroxidasa (LiP), manganeso peroxidasa (MnP) y lacasa. Las lacasas usan
oxígeno molecular como sustrato y lo reducen hasta agua. Por el contrario, la LiP y
MnP emplean peróxido de hidrógeno como aceptor final de electrones, por tanto es
esencial la presencia en el medio de peróxido de hidrógeno, que a su vez es generado
por la actividad de enzimas glioxal oxidasa, glucosa oxidasa y aril alcohol oxidasa.
Las lacasas son enzimas de tipo fenoloxidasa, que contienen cuatro átomos de
cobre en su estructura, uno de los cuales está implicado en la oxidación de los sustratos,
mientras el resto participan en el transporte de electrones hasta el oxígeno. En la figura
2.19 se muestra el mecanismo catalítico propuesto por Malmström (1982) para la lacasa.
e- e-
Cu2+Cu2+(Cu2+)2 Cu+Cu2+(Cu2+)2 Cu+Cu+(Cu2+)2
H2 O
2H+
Cu+Cu2+[(Cu2+)2O-] Cu2+Cu2+(Cu+)2
O2
e- H2O 2H+
e-
Cu2+Cu2+[(Cu2+)2O-] Cu+Cu2+[(Cu2+)2O22-] Cu2+Cu2+[(Cu2+)2O22-]
44
2 - Introducción
El bajo potencial redox de las lacasas (entre 400 y 800 mV) sólo las permite
oxidar compuestos fenólicos. Sin embargo, en presencia de mediadores redox son
capaces de oxidar unidades no fenólicas. Además, los mediadores son compuestos de
bajo peso molecular, que son capaces de difundir mejor en la fibra y reaccionar con las
fracciones de lignina a las que la enzima no tiene acceso debido a su alto peso
molecular. Esta cooperación de la enzima y los mediadores es lo que se denomina
sistema lacasa mediador, en el cual la enzima oxida un mediador químico que, a su vez,
oxida la lignina, como se indica en la figura 2.20.
lignina
H2O [lacasa]OX mediador
oxidada
Figura 2.20. Mecanismo de actuación del sistema lacasa mediador sobre la lignina.
2.5.1 - Biopulpeo
45
Proceso Kraft en Biorrefinería
46
2 - Introducción
químicas en procesos de bioblanqueo (Eugenio et al., 2010a). En los casos en los que se
quieren aplicar las enzimas directamente a la madera, es recomendable someter las
astillas a vacío para facilitar la adsorción de las enzimas (Jacobs-Young et al., 1998) o a
compresión para provocar fracturas en la madera que faciliten la penetración (Maijala et
al., 2008; Pere et al., 2005). Existen otros tratamientos que provocan efectos similares, y
por tanto podrían utilizarse como pre-tratamiento en el biopulpeo, aunque hasta el
momento han sido menos estudiados para este fin. Entre ellos se encuentra el
tratamiento de explosión por vapor. Como se ha indicado en el apartado 2.5, la rápida
descompresión final provoca una separación de las fibras en paquetes y una rotura de las
paredes de las fibras, que facilita la difusión de los reactivos o enzimas aplicados en
etapas posteriores. Así, este pre-tratamiento ha demostrado mejorar tanto procesos de
pasteado (Ahvazi et al., 2007; San Martin et al., 1995) como de hidrólisis enzimática
para la obtención de azúcares fermentables (Jurado et al., 2009; Ruiz et al., 2008; Wu et
al., 1999). Sin embargo, existen pocos trabajos sobre su efecto en un biopulpeo
enzimático.
2.5.2 - Bioblanqueo
47
Proceso Kraft en Biorrefinería
Pese a que las xilanasas no degradan la lignina, éstas fueron las primeras
enzimas utilizadas en el blanqueo de pasta Kraft (Viikari et al., 1986) y son hoy día las
únicas que se aplican a nivel industrial. La aplicación de las xilanasas, previamente al
blanqueo, ha demostrado que es posible alcanzar el nivel deseado de blancura con un
ahorro de reactivos de blanqueo, reduciendo así la contaminación por cloroligninas. En
años posteriores, otros trabajos confirmaron la capacidad de las xilanasas para mejorar
los efectos de las etapas de blanqueo (Paice et al., 1988; Ragauskas et al., 1994; Senior
y Hamilton, 1992; Viikari et al., 1994). Para explicar cómo actúan las xilanasas en el
blanqueo de pasta Kraft se han descritos algunos mecanismos de actuación. Uno de
ellos supone que las xilanasas eliminan los xilanos precipitados, depositados sobre las
fibras y que dificultan el acceso de los reactivos de blanqueo. Una vez eliminada esta
barrera, la acción de los reactivos sería mucho más eficaz (Kantelinen et al., 1993; Shah
et al., 2000). Otra explicación supone que las hemicelulosas están unidas a la lignina por
enlaces covalentes y que la acción de la enzima libera estos compuestos permitiendo su
extracción del medio (Suurnäkki et al., 1996; Wong, 1997a). Por otro lado, existen
evidencias de que parte de los grupos cromóforos proceden de la degradación de
hemicelulosas durante el pasteado Kraft, como los ácidos hexenurónicos. Estas
sustancias que quedan unidas a las hemicelulosas, aumentan el número kappa,
consumen reactivos de blanqueo, retienen iones metálicos, favorecen el amarilleamiento
de la pasta y contribuyen a la formación de ácido oxálico (Roncero et al., 2003; Wong,
48
2 - Introducción
1997a). Gracias a la acción de las xilanasas, estos compuestos son liberados una vez que
se rompe su unión con las hemicelulosas (Valls et al., 2010; Wong, 1997b).
Desde su descripción, los sistemas lacasa mediador han sido investigados con
éxito para la deslignificación y blanqueo de diferentes pastas de papel, observando una
reducción del consumo de reactivos de blanqueo y una disminución de la carga
contaminante de los efluentes, mientras que las propiedades mecánicas de las pastas se
mantienen o mejoran. (Bourbonnais y Paice, 1996; Camarero et al., 2004; Eugenio et
al., 2010a; Ibarra et al., 2006; Moldes et al., 2008). Sin embargo, el proceso aún no se
ha implantado a escala industrial probablemente debido a las dificultades derivadas del
uso del mediador y a la falta de optimización del proceso. El elevado coste y los
posibles problemas de toxicidad de algunos mediadores han dirigido las investigaciones
a la búsqueda de mediadores alternativos, incluyendo algunos metabolitos fúngicos así
como productos resultantes de la degradación de la lignina (Cañas y Camarero, 2010;
Fillat et al., 2010; Moldes et al., 2008) como se ha indicado anteriormente.
49
Proceso Kraft en Biorrefinería
El pre-tratamiento es una etapa crucial que pretende romper las barreras físicas y
químicas que inhiben la susceptibilidad del material al bioprocesado, tal como la
hidrólisis enzimática y la fermentación. Este pre-tratamiento debe promover la
formación de azúcares, evitar la degradación o pérdida de polisacáridos y la formación
de productos inhibidores, y además, ser eficaz económicamente (Ruiz et al., 2008).
Entre los pre-tratamiento estudiados, la explosión por vapor es uno de los más
utilizados, ya que degrada parcialmente y solubiliza la lignina y hemicelulosas debido a
las condiciones de alta presión y temperatura (Josefsson et al., 2002). Sin embargo,
durante este pre-tratamiento pueden generarse compuestos inhibidores debido a
reacciones de degradación, como se ha señalado anteriormente. Por ello, en muchos
casos son recomendables etapas de detoxificación de los sustratos, que incluyen
métodos físicos, químicos y biológicos (Palmqvist y Hahn-Hägerdal, 2000a; b). La
aplicación de lacasas es uno de los nuevos métodos biológicos usados, ya que son
capaces de eliminar compuestos fenólicos de la biomasa (Chandel et al., 2007; Jurado et
al., 2009).
50
2 - Introducción
51
Proceso Kraft en Biorrefinería
relacionados con la producción de etanol en las bacterias entéricas como E. coli, o por la
introducción de los genes de la ruta de la fermentación de las pentosas en
microorganismos productores de etanol como la levadura S. cerevisiae o la bacteria
etanologénica Z. mobilis (Tomás-Pejó, 2010).
52
3
OBJETIVOS
3 - Objetivos
3. OBJETIVOS
Para alcanzar estos objetivos se han definido los siguientes objetivos concretos:
55
Proceso Kraft en Biorrefinería
56
4
DISCUSIÓN DE LOS RESULTADOS
4 – Discusión de los resultados
II. Martín-Sampedro, R., Capanema, E.A., Hoeger, I.C., Villar, J.C., Rojas,
O.J. Lignin changes after steam explosion and laccase-mediator
treatment of eucalyptus wood chips. Journal of Agricultural and Food
Chemistry. 2011. 59(16): 8761-8769.
III. Martín-Sampedro, R., Eugenio, M.E., García, J.C., Lopez, F., Villar,
J.C., Diaz, M.J. Steam explosion and enzymatic treatments as pre-
treatments to improve enzymatic hydrolysis of Eucalyptus globulus.
Manuscrito.
IV. Martín-Sampedro, R., Eugenio, M.E., Revilla, E., Martín, J.A., Villar,
J.C. Integration of Kraft pulping on a forest biorefiney by the addition of
a steam explosion pretreatment. BioResources. 2011. 6(1): 513-528.
59
Proceso Kraft en Biorrefinería
Agua de
Enzimas Biopulpeo Enzimas Bioblanqueo
lavado
Fermentación
Bioetanol
Figura 4.1. Esquema del proceso estudiado en la presente tesis doctoral: biorrefinería basada en
la extracción de hemicelulosas previa al pasteado Kraft, introduciendo etapas enzimáticas que
mejoren la sostenibilidad del proceso.
60
4 – Discusión de los resultados
61
Proceso Kraft en Biorrefinería
Agua de
I, II Enzimas Biopulpeo
lavado
Enzimas Bioblanqueo
Fermentación
Bioetanol
Figura 4.2. Esquema del proceso de biorrefinería forestal estudiado, indicando las partes del
proceso en las que se centran las publicaciones I y II.
62
4 – Discusión de los resultados
Tabla 4.1. Condiciones de operación de los pre-tratamientos de explosión por vapor realizados.
63
Proceso Kraft en Biorrefinería
a) b)
100 100
% remanente en el sólido
% remanente en el sólido
80 80
60 60
40 40
20 20
0 0
EG SE1 SE3 SE5 SE7 SE0 SE2 SE4 SE6 SE8
Figura 4.3. Porcentaje de cada componente de la madera que se detectó en la fracción sólida
tras la explosión por vapor, respecto a la materia prima original EG: a) pre-tratamientos sin pre-
extracción previa; b) pre-tratamientos con pre-extracción previa.
64
4 – Discusión de los resultados
65
Proceso Kraft en Biorrefinería
66
4 – Discusión de los resultados
la descompresión de la explosión por vapor (Emmel et al., 2003). Otra posible causa de
la disminución de los azúcares recuperados podría ser una recolección ineficiente de los
compuestos extraídos, ya que parte podrían estar disueltos en el líquido que impregna
las astillas después de la explosión y ser desechados con las aguas del lavado de las
astillas.
a) b)
50 50
45 45
% extraído en líquido
% extraído en líquido
40 40
35 35
30 30
25 25
20 20
15 15
10 10
5 5
0 0
EG SE1 SE3 SE5 SE7 SE0 SE2 SE4 SE6 SE8
Figura 4.4. Porcentaje de cada componente de la madera que se detectó en la fracción líquida
tras la explosión por vapor, respecto a la materia prima original EG: a) pre-tratamientos sin pre-
extracción previa; b) pre-tratamientos con pre-extracción previa.
67
Proceso Kraft en Biorrefinería
a) b)
100 EG 2,0
EG
1,8
1ª Derivada peso (%/ºC)
SE8 SE8
80 1,6
1,4
Peso (%)
60 1,2
1,0
40 0,8
0,6
20 0,4
0,2
0 0,0
0 100 200 300 400 500 600 0 200 400 600
68
4 – Discusión de los resultados
500 400
488 486 489
490 484 485 390
482
478 478
480 380
Temperatura (ºC)
Temperatura (ºC)
470
470 370
460 360
450 350
440 332 333 332 333 333 333 333 340
332 331
430 330
420 320
Lignina Polisacáridos
69
Proceso Kraft en Biorrefinería
70
4 – Discusión de los resultados
OCH3
HO γ
HO 6' O
6' 5' 1' 1'
5' 1' 6' 2' 3'
2' 4'
R β 4'
3' 2' O
4'
3' 2' 5'
4'
3'
O 1'
6' 5'
α O O OCH3 OCH3
OCH3 HO O
1
OCH3 1
6 6 2
2
5 4 3 5 4 3 6 1
5
6
1 O
OCH3 OCH3 5
4 3
2
4
3
2
O O O
O OCH3 OCH3
O OH
COOH
H3 CO O
OH
HO
6
1
2 OAr
5 4 3
OCH3 OCH3
OCH3
O O
O
F H I
R=
O
HO Carb
5'
6'
1' O
4' O
HO O HO O
3' 2'
R
O OH Carb
OCH3 O
1
6 2 HO
5 4 3
OH OCH3 H3 CO OCH3
OCH3 O
O Carb O O
O O OH
O G S
OH Carb
J
71
Proceso Kraft en Biorrefinería
20
40
60
Carbohidratos δC
80
100
120
7 6 5 4 3 2 1
δH
Figura 4.8. Espectro HSQC de la lignina MWL aislada de la muestra original de Eucalyptus
globulus.
72
4 – Discusión de los resultados
Comparando los espectros HSQC de las tres ligninas aisladas, se encontró una
mayor presencia de polisacáridos en las ligninas tratadas (SE8 y SE8+LMS) a δC/ δH
63-75/3,2-3,8. Aparte de esta diferencia, los espectros fueron muy similares. Otros
autores que estudiaron lignina residual después de un tratamiento LMS seguido de una
extracción alcalina (Balakshin et al., 2001; Ibarra et al., 2007), encontraron también
espectros HMQC muy similares a los de la lignina original (lignina Kraft residual).
Estos autores atribuyeron estos resultados a la eliminación de la lignina alterada de la
pasta durante la extracción alcalina llevada a cabo tras el tratamiento LMS. Así, la
lignina que permanece en la pasta no muestra cambios sustanciales, porque sería
principalmente la lignina no oxidada. En nuestro caso, la lignina alterada podría
eliminarse durante la extracción alcalina que se lleva a cabo en el procedimiento de
aislamiento de la lignina con el objetivo de eliminar taninos. Así, la lignina soluble en
álcali eliminada en el caso de muestra original fue un 9,6% de la lignina inicial,
mientras que para la muestra tras explosión de vapor fue un 26,9%. En el caso de la
muestra tras explosión de vapor y tratamiento LMS, la lignina soluble en álcali fue un
25,0% de la lignina inicial. Esta disminución respecto a la muestra SE8 puede deberse a
la eliminación de lignina durante el tratamiento LMS, que se observó en el estudio de la
composición química.
73
Proceso Kraft en Biorrefinería
a)
20
40
60
Carbohidratos δC
80
100
120
6 δH 4 2
b)
50
60
δC
70
Carbohidratos
80
Figura 4.9. Espectro HSQC de la lignina aislada de la muestra tras el tratamiento de explosión
por vapor SE8 (a). Incluye una ampliación de la región alifática oxigenada (b).
74
4 – Discusión de los resultados
a)
20
40
60
Carbohidratos δC
80
100
120
6 δH 4 2
b) 50
60
δC
70
80
Carbohidratos
Figura 4.10. Espectro HSQC de la lignina aislada de la muestra tras el tratamiento de explosión
por vapor seguido del tratamiento con LMS (SE8+LMS) (a). Incluye una ampliación de la
región alifática oxigenada (b).
75
Proceso Kraft en Biorrefinería
76
4 – Discusión de los resultados
autores observaron un aumento del peso molecular de la lignina, que presentó una
estructura más heterogénea, debido al predominio de las reacciones de repolimerización
a medida que aumentaba la severidad del tratamiento. En nuestro caso, no se observó un
aumento del grado de condensación (DC) después del tratamiento de explosión por
vapor, probablemente debido a que, en las condiciones ensayadas (menos severas), las
reacciones de repolimerización pueden no ser predominantes.
Por último, se observó una reducción significativa de la relación S/G después del
tratamiento de explosión por vapor y también después del tratamiento con LMS. Chua y
Wayman (1979) explicaron que esta reducción en la relación S/G durante condiciones
suaves de explosión por vapor podía deberse a una eliminación preferente de unidades
siringilo durante tiempos cortos de autohidrólisis. De igual forma, Ibarra et al. (2007)
observaron una reducción de la relación S/G después de tratamientos LMS.
77
Proceso Kraft en Biorrefinería
a) b) c)
Figura 4.11. Astillas sin pre-tratatar (a) y astillas después del pre-tratamiento SE8 (b); una
ampliación de la zona remarcada se muestra en (c).
78
4 – Discusión de los resultados
300
Sin pre-extracción
150
SE 8
100 SE 6 SE 7
SE 4 SE 5
50 SE 1 SE 2 SE 3
EG SE0
0
Control 1C: 5 1C: 10 2C: 5+3 2C: 10+3
Figura 4.12. Comparación de la capacidad de retención de agua de las astillas originales y las
astillas después de los tratamientos de explosión por vapor.
Los resultados del volumen relativo de poros mostraron una banda heterogénea
para un diámetro de poro entre 10 y 100 µm que puede corresponder al lumen y a los
vasos de las células de la madera. Este volumen no se modificó de manera importante
tras los tratamientos de explosión por vapor. Por ello, el estudio se centró en el volumen
de poros con un diámetro comprendido entre 10 y 10000 nm (Figura 4.13).
Las astillas de E. globulus sin ningún tratamiento (EG) mostraron una banda
principal de volumen de poro para diámetros comprendidos entre 10 y 200 nm, y una
banda más pequeña entre 3000 y 4000 nm. Después de un tratamiento de explosión por
vapor de 1 ciclo de 5 minutos (SE1), la banda principal se desplazó hacia diámetros de
poro mayores. Al someter las astillas a un tratamiento de 10 minutos (SE3), el volumen
de poro aumentó respecto a EG y SE1, apareciendo dos picos para diámetro de poro
entorno a 400 y 1500 nm, seguidos de un tercer pico para poros de 10000 nm. Al añadir
un segundo ciclo (figura 4.13b), se generaron poros de diámetro entre 4000-6000 nm,
siendo mayor el volumen de poros para tratamientos más largos (SE7 vs SE5).
79
Proceso Kraft en Biorrefinería
a) b)
0,05 0,05
EG EG
0,03 0,03
0,02 0,02
0,01 0,01
0,00 0,00
10 100 1000 10000 10 100 1000 10000
c) d)
0,05 0,05
SE0 SE0
SE2 SE6
0,03 0,03
0,02 0,02
0,01 0,01
0,00 0,00
10 100 1000 10000 10 100 1000 10000
Figura 4.13. Volumen relativo de poros en las astillas tras el tratamiento de explosión por
vapor: a) sin pre-extracción y con 1 ciclo, b) sin pre-extracción y con 2 ciclos, c) con pre-
extracción y con 1 ciclo d) con pre-extracción y con 2 ciclos.
80
4 – Discusión de los resultados
poros grandes, pero sin prácticamente aumentar el volumen total de poros. Es decir,
debido al tratamiento de explosión por vapor varios poros pequeños darían lugar a un
poro de mayor tamaño. Así, en la figura 4.14 observamos cómo mientras el volumen de
mesoporos (diámetro de poro entre 7 y 50 nm) disminuyó, el volumen de macroporos
(entre 50 y 10000 nm) aumentó en todas las muestras tratadas, especialmente en las pre-
extraídas con agua, excepto después del tratamiento menos severo (SE1 y SE2). A partir
de estos resultados podría concluirse que el tratamiento de explosión por vapor tiene un
efecto positivo en la formación de poros de mayor tamaño aumentando de esta forma la
accesibilidad de las astillas que permitirá una mejor y más rápida penetración tanto de
reactivos químicos como de enzimas.
a) Sin pre-extracción
b)
Sin pre-extracción
0,25 0,7
Mesoporos: Vol. esp., ml/g
Figura 4.14. Volumen específico de a) mesoporos y b) macroporos en las astillas después de los
tratamientos de explosión por vapor, comparado con las astillas antes del tratamiento (control).
81
Proceso Kraft en Biorrefinería
a) b)
7,0 7,0
EG SE0
3,0 3,0
2,0 2,0
1,0 1,0
0,0 0,0
10 100 1000 10000 10 100 1000 10000
Figura 4.15. Superficie relativa de las astillas tras el tratamiento de explosión por vapor: a) sin
pre-extracción, b) con pre-extracción.
82
4 – Discusión de los resultados
Agua de
III Enzimas Detoxificación
lavado
Enzimas Bioblanqueo
Fermentación
Bioetanol
Figura 4.16. Esquema del proceso de biorrefinería forestal estudiado, indicando las partes del
proceso en las que se centran la publicación III.
83
Proceso Kraft en Biorrefinería
84
4 – Discusión de los resultados
Tabla 4.3. Concentración de azúcares y ácido acético en las fracciones líquidas recogidas en el
tratamiento de explosión por vapor SE8, y porcentaje de azúcares recuperado en dichas
fracciones respecto al azúcar en la materia prima origina (EG). [A] hace referencia a los
monómeros y [AO] a los oligosacáridos.
120
Glucosa
Rto. de la hidrólisis (%)
100 Xilosa
80
60
40
20
0
SE8.1-EH SE8.2-EH Total-EH SE8.1-AH SE8.2-AH Total-AH
Figura 4.17. Rendimiento de la hidrólisis enzimática (EH) y la hidrólisis ácida (AH) de las
fracciones líquidas obtenidas en el tratamiento de explosión por vapor SE8.
85
Proceso Kraft en Biorrefinería
86
4 – Discusión de los resultados
Como se verá en los apartados 4.3 y 4.4, la fracción solida obtenida en el pre-
tratamiento de explosión por vapor, presenta buenas características para ser utilizada en
la producción de papel. Sin embargo, muchos autores han utilizado el proceso de
explosión por vapor para mejorar la hidrólisis enzimática de la fracción sólida en el
proceso de obtención de bioetanol (Glasser y Wright, 1998; Ohgren et al., 2006; Oliva
et al., 2003; Ramos et al., 1992). Por ello, en este apartado se ha estudiado la hidrólisis
enzimática de las fracciones sólidas obtenidas durante los distintos tratamientos de
explosión por vapor estudiados en el apartado 4.1, planteando la producción de etanol a
partir de la fracción celulósica como alternativa al pasteado.
40
Glucosa
Rto. hidrólisis enzimática (%)
35
Azúcares totales
30
25
20
15
10
0
EG SE 0 SE 1 SE 2 SE 3 SE 4 SE 5 SE 6 SE 7 SE 8
Figura 4.18. Rendimiento de la hidrólisis enzimática de la materia prima original (EG) y de las
fracciones sólidas obtenidas en los distintos tratamientos de explosión por vapor.
87
Proceso Kraft en Biorrefinería
88
4 – Discusión de los resultados
de 23,7% y 30,2% respectivamente. Sin embargo, el rendimiento real fue mayor que
estos valores, ya que se deben considerar también los azúcares recogidos en la fracción
líquida, como se verá en el apartado 4.2.3.
89
Proceso Kraft en Biorrefinería
Glucosa
40 Azúcares totales
90
4 – Discusión de los resultados
tener en cuenta (Ruiz et al., 2008). Por tanto, se ha determinado el rendimiento global
del proceso teniendo en cuenta todas estas consideraciones.
91
Proceso Kraft en Biorrefinería
Agua de
lavado
Enzimas Bioblanqueo V
IV
Astillas Explosión por Lavadores de Pasta
Digestor Kraft Blanqueo
E. globulus vapor pasta blanqueada
Fermentación
Bioetanol
Figura 4.20. Esquema del proceso de biorrefinería forestal estudiado, indicando las partes del
proceso en las que se centran las publicaciones IV y V.
92
4 – Discusión de los resultados
El efecto principal del pre-tratamiento fue una gran reducción del número kappa
de las pastas obtenidas en todas las condiciones ensayadas (figura 4.21). La mayor
reducción, compando pastas con y sin pre-tratamiento, se observó bajo las condiciones
de pasteado más suaves (factor H de 180), obteniendo una pasta con un número kappa
un 70% menor que la pasta control obtenida en las mismas condiciones de pasteado
(número kappa de 16 frente a 49). Otros autores han encontrado también una reducción
en el número kappa de las pastas obtenidas con distintos pre-tratamientos de extracción
de hemicelulosas, aunque la mayoría de ellos observaron reducciones menores que las
conseguidas en este trabajo. Así, Helmerius et al. (2010) y Schild et al. (2010)
obtuvieron reducción del 40% y del 57% utilizando autohidrólisis como pre-tratamiento
de abedul y eucalipto respectivamente. La mayor reducción del número kappa
conseguida en esta tesis se debe probablemente a que la explosión por vapor, además de
extraer hemicelulosas, proporciona un residuo sólido con una estructura más abierta que
la autohidrólisis, debido al efecto de la rápida descompresión, como se estudió en el
apartado 4.1. Sin embargo, San Martin et al. (1995) obtuvieron una reducción del 49%
en el pasteado Kraft de Pinus radiata pre-tratado mediante explosión por vapor, aunque
en este caso el hecho de que utilizase una madera de conífera en lugar de una madera de
frondosa como el eucalipto, con mayor contenido en xilanos y grupos acetilo, pudo ser
el factor clave.
93
Proceso Kraft en Biorrefinería
a) b) c)
60 40 1400
35
Viscosidad (ml/g)
30
1000
Número Kappa
40 25
800
30 20
15 600
20
10 400
10 5 200
0 0 0
150 250 350 450 150 250 350 450 150 250 350 450
Factor H
Control Factor H Control Control Factor H
SE SE SE
Figura 4.21. Número kappa (a), contenido en ácidos hexenurónicos (b) y viscosidad (c) de una
pasta Kraft obtenida a partir de una muestra control (sin pre-tratamiento) y de una muestra pre-
tratada mediante explosión de vapor (SE).
94
4 – Discusión de los resultados
peso molecular (Mendes et al., 2009). En la figura 4.22b se muestran las fibras de la
pasta pre-tratada, no observándose rotura ni degradación de las mismas.
a) b)
Figura 4.22. Imágenes tomadas con microscopio óptico de la pasta Kraft obtenida a partir de
una muestra control (sin pre-tratamiento) (a) y de una muestra pre-tratada mediante explosión
de vapor (SE) (b) tras un pasteado Kraft con factor H igual a 190.
95
Proceso Kraft en Biorrefinería
a) b)
70% 70%
Rendimiento global
60% 60%
Rendimiento útil
50% 50%
40% 40%
30% 30%
20% 20%
150 250 350 450 150 250 350 450
Control Factor H control Factor H
SE SE
Figura 4.23. Rendimiento global (a) y útil (b) del pasteado Kraft de una muestra control (sin
pre-tratamiento) y una muestra pre-tratada mediante explosión por vapor (SE).
a) b)
100% 100%
80% 80%
Consumo NaOH
Consumo Na2S
60% 60%
40% 40%
20% 20%
0% 0%
150 250 350 450 150 250 350 450
Control Factor H Control Factor H
SE SE
Figura 4.24. Consumo de reactivos durante el pasteado Kraft de una muestra control (sin pre-
tratamiento) y una muestra pre-tratada mediante explosión por vapor (SE).
96
4 – Discusión de los resultados
A partir de los resultados anteriores, se puede concluir que bajo las mismas
condiciones de pasteado, las astillas pre-tratadas con explosión por vapor se pastean de
forma más rápida y eficaz que las astillas sin tratar, o alternativamente, pueden
pastearse a una temperatura menor o con menor carga de álcali manteniendo las misma
producción de pasta (Ahvazi et al., 2007; Al-Dajani et al., 2009; San Martin et al.,
1995). Bajo condiciones similares de pasteado, el tiempo de cocción puede reducirse de
60 a 25 minutos, obteniendo una pasta de calidad similar (número kappa ~16 y
viscosidad ~ 1000 ml/g) con un buen rendimiento (~50%). Incluso teniendo en cuenta la
pérdida de rendimiento debida al pre-tratamiento, que reduciría el rendimiento total del
proceso a ~40%, la producción de pasta podría mantenerse debido a la reducción del
tiempo de pasteado. Además, gran parte de las hemicelulosas se extraerían sin apenas
degradación antes del pasteado, pudiéndose utilizar para la producción de etanol u otros
compuestos de valor añadido. Por otro lado, se obtendrían pastas con un contenido muy
bajo en ácidos hexenurónicos, que podrían reducir el consumo de reactivos durante el
proceso de blanqueo. Sin embargo, la reducción del contenido en hemicelulosas de las
pastas, puede ser perjudicial en la refinabilidad de las pastas y en las propiedades
mecánicas de las mismas.
97
Proceso Kraft en Biorrefinería
decir, las pastas obtenidas a partir de astillas sometidas a explosión por vapor
consumieron más energía que las pastas sin pre-tratar (control e industrial) para alcanzar
un determinado ºSR, probablemente debido a la eliminación de hemicelulosas durante el
pre-tratamiento. Resultados similares fueron obtenidos por otros autores (Helmerius et
al., 2010; Schild et al., 2010).
60
50
40
ºSR
30
20
10
0
0 1000 2000 3000
Nº Rev. PFI
SE Control+O2 Ind.
Las propiedades mecánicas y ópticas de las tres pastas refinadas y sin refinar se
muestran en la figura 4.26. Las tres pastas sin refinar presentaron un ºSR similar (17, 15
y 16 para SE, Control, Ind.) y propiedades mecánicas semejantes, excepto el índice de
tracción. Este índice está relacionado con la capacidad de unión entre las fibras. Por
ello, el bajo valor obtenido para la pasta pre-tratada puede deberse al bajo contenido en
hemicelulosas de dicha pasta, ya que este polímero es más amorfo y tiene mayor área
superficial por gramo que la celulosa, favoreciendo la unión entre fibras (Karlsson,
2006). El índice de desgarro de la pasta pre-tratada y la pasta industrial sin refinar fue
similar, lo que indica un grado similar de degradación de las fibras, de acuerdo con lo
publicado por San Martin et al. (1995).
98
4 – Discusión de los resultados
a) b)
Densidad aparente (g/cm3)
0,8 100
0,6 60
0,5 40
0,4 20
0,3 0
10 30 50 70 10 30 50 70
ºSR ºSR
SE Control Ind. SE Control Ind.
c) d)
12 6
10 5
8 4
6 3
4 2
2 1
0 0
10 30 50 70 10 30 50 70
ºSR ºSR
SE Control Ind. SE Control Ind.
e) f)
45 50
Índice de amarilleamiento
45
40
Blancura (% ISO)
40
35
35
30
30
25 25
20 20
10 30 50 70 10 30 50 70
ºSR ºSR
SE Control Ind. SE Control Ind.
Figura 4.26. Propiedades mecánicas y ópticas de la pasta Kraft industrial (Ind.), la pasta control
tras deslignificación con oxígeno (Control) y la pasta con pre-tratamiento de explosión por
vapor (SE): a) densidad aparente, b) índice de tracción, c) índice de desgarro, d) índice de
estallido, e) blancura y f) índice de amarilleamiento.
A medida que aumentó el grado de refino, la densidad aparente de las tres pastas
aumentó de manera análoga (figura 4.26a), indicando una cantidad de finos y un grado
de fibrilación similar (Helmerius et al., 2010). Las propiedades mecánicas de las pastas
también aumentaron con el refino, como se esperaba. Sin embargo la evolución fue
diferente para las distintas pastas. Así, las pastas pre-tratadas (SE) mostraron menores
índices de tracción y de estallido que las pastas control e industrial, debido al menor
99
Proceso Kraft en Biorrefinería
Por tanto, aunque las pastas pre-tratadas mostraron propiedades mecánicas algo
menores que las pastas control e industrial, las propiedades ópticas fueron mayores, y
no se observó degradación significativa de las fibras. Además, la evolución de las
propiedades mecánicas y ópticas tras un tratamiento de envejecimiento acelerado fue
mejor para las pastas pre-tratadas (Publicación IV).
100
4 – Discusión de los resultados
control para la pasta SE en el que se añadió LMS pero no oxígeno (LM-NoO2). Más
detalles sobre las condiciones del blanqueo se pueden encontrar en la publicación V.
30
26,8 26,4 Sin blanquear L LE LEP
26,4 25,6 26,2
25 22,5 23,0
Número Kappa
19,1
20
15,5 15,5 15,5
14,0 14,0 14,6
15 12,9
13,8 13,7
12,2 12,4
12,1 11,1 11,0
10,7
8,9 9,0
10
7,4 7,0 7,4
0
Ind: No-LMO2 Ind: LMO2 NoSE: No-LMO2 NoSE: LMO2 SE: No-LMO2 SE: LMO2 SE: LM-NoO2
Figura 4.27. Evolución del número Kappa durante la secuencia de blanqueo para los distintos
ensayos.
101
Proceso Kraft en Biorrefinería
consiguió una reducción del 47,0% (Ind:LMO2) frente a 36,6% para su control sin LMS
(Ind:No-LMO2). Otros autores han obtenido mayores reducciones del número kappa
durante el bioblanqueo de pastas Kraft, pero utilizando más etapas en la secuencia TCF,
con etapas enzimáticas o de peróxido de hidrógeno más largas y/o con mayor
concentración de reactivos (Moldes et al., 2010).
102
4 – Discusión de los resultados
alta deslignificación sin reducir de manera importante la viscosidad de las pastas. Más
detalles sobre la viscosidad se muestran en la publicación V.
80
Sin blanquear L LE LEP
67 68
70 66
63
58 59
60
Blancura (% ISO)
53
50
43 43 41 43 42 42 42 42
40 39 38 39 39 39
40 34 35 34 33 36
34
33
30
20
10
0
Ind: No-LMO2 Ind: LMO2 NoSE: No-LMO2 NoSE: LMO2 SE: No-LMO2 SE: LMO2 SE: LM-NoO2
Figura 4.28. Evolución de la blancura durante la secuencia de blanqueo para los distintos
ensayos.
Si se comparan las blancuras finales obtenidas con pastas pre-tratadas (SE) y con
pastas sin pre-tratamiento (NoSE), resulta evidente que el pre-tratamiento de explosión
por vapor favorece el aumento de blancura durante el proceso de blanqueo, obteniendo
aumentos del 25,2% para SE:LMO2 frente a 16,3% para NoSE:LMO2. La pasta
industrial bioblanqueada (Ind:LMO2) mostró también una blancura final inferior a la
pasta SE, aunque superior a la pasta NoSE. Por tanto, al igual que se observó en el
103
Proceso Kraft en Biorrefinería
43 43 44
45 41
41
39 39 39 40 40
40
35 34 34 34
35 31
30 30 30 29
30 28 28 28
24
25
20 21 20
20
15
10
5
0
Ind: No-LMO2 Ind: LMO2 NoSE: No-LMO2 NoSE: LMO2 SE: No-LMO2 SE: LMO2 SE: LM-NoO2
Figura 4.29. Evolución del amarilleamiento durante la secuencia de blanqueo para los distintos
ensayos.
104
4 – Discusión de los resultados
a) b)
95 SE 25 SE
SE: LMO2 SE: LMO2
90 20 NoSE
NoSE
NoSE: LMO2 NoSE: LMO2
85 15
Ind Ind
L* b* Ind: LMO2
80 Ind: LMO2
10
75 5
70
0
10 13 16 19 22
-5 0 5 10 15 20 25
C* a*
Figura 4.30. Propiedades colorimétricas de las pastas sin blanquear (símbolos oscuros) y
bioblanqueadas (símbolos claros): a) coordenadas CIE L*C*, b) coordenadas CIE L*a*b*.
105
Proceso Kraft en Biorrefinería
explosión por vapor ayuda a producir pastas con mejores propiedades colorimétricas. A
la misma conclusión se llega a partir de las coordenadas CIE L*a*b* mostradas en la
figura 4.30b. La secuencia de bioblanqueo redujo tanto la coordenada a* como la b* y
los valores más bajos se encontraron en la pasta SE:LMO2.
100
457
90
80
SE
70
Reflectancia (%)
60 NoSE
50 Ind
40 SE: LMO2
30 NoSE: LMO2
20
Ind: LMO2
10
0
380 420 460 500 540 580 620 660 700
Longitud de onda (nm)
106
4 – Discusión de los resultados
blanqueo (Ragnar, 2001; Vuorinen et al., 1999), es lógico esperar un menor consumo de
peróxido de hidrógeno durante el blanqueo de dichas pastas. Otros autores también
observaron una reducción en el consumo de reactivos de blanqueo cuando se llevó a
cabo un pre-tratamiento con autohidrólisis (Mendes et al., 2009).
Tabla 4.4. Consumo de peróxido de hidrógeno en los diferentes ensayos llevados a cabo.
Consumo de peróxido
Ensayo
de hidrógeno (%)
Ind: No-LMO2 79,5%
Ind: LMO2 59,1%
NoSE: No-LMO2 77,6%
NoSE: LMO2 66,8%
SE: No-LMO2 55,7%
SE: LMO2 54,0%
SE: LM-NoO2 51,2%
Envejecimiento acelerado
Para estudiar el efecto del pre-tratamiento de explosión por vapor y del
tratamiento enzimático en la estabilidad de las propiedades ópticas de las pastas
blanqueadas, se sometieron las pastas a un envejecimiento acelerado. El envejecimiento
acelerado se llevó a cabo a 80ºC y 65% de humedad relativa durante 6 días, de acuerdo
con la norma ISO 5630-3.
107
Proceso Kraft en Biorrefinería
108
4 – Discusión de los resultados
109
Proceso Kraft en Biorrefinería
La figura 4.32 muestra las partes del proceso de biorrefinería forestal en las que
se centra este apartado.
Agua de
VI Enzimas Biopulpeo
lavado
Enzimas Bioblanqueo
Fermentación
Bioetanol
Figura 4.32. Esquema del proceso de biorrefinería forestal estudiado, indicando las partes del
proceso en las que se centran la publicación VI.
Por tanto, según se indica en la figura 4.32, las etapas del sistema de biopulpeo
propuesto serán: pre-tratamiento mediante explosión por vapor, seguida de un
110
4 – Discusión de los resultados
111
Proceso Kraft en Biorrefinería
a) b)
56% 54%
Rendimiento global Kraft
54%
42% 42%
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
c) d)
100% 100%
80% 80%
Consumo NaOH
Consumo Na2S
60% 60%
40% 40%
20% 20%
0% 0%
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
e) f)
19 1400
18 1200
17
Viscosidad (ml/g)
1000
Número Kappa
16
15 800
14 600
13
400
12
11 200
10 0
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
Figura 4.33. Efecto en el pasteado del tratamiento con LMS tras un pre-tratamiento de
explosión por vapor: rendimientos global (a) y útil (b) del pasteado, consumo de Na2S (c) y
NaOH (d) y número kappa (e) y viscosidad (f) de las pastas obtenidas. El control consistió en un
tratamiento de explosión por vapor seguido de un pasteado Kraft, sin tratamiento enzimático.
112
4 – Discusión de los resultados
113
Proceso Kraft en Biorrefinería
114
4 – Discusión de los resultados
a) b)
17 0,45
15 0,40
14
0,35
13
12 0,30
11
10 0,25
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
c) d)
16 2,5
Índice de desgarro (mNm 2/g)
Índice de tracción (Nm/g)
14
2,0
12
10 1,5
8
6 1,0
4
0,5
2
0 0,0
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
e) f)
44 101
100
42
99
Blancura (% ISO)
40 98
Opacidad
97
38
96
36 95
94
34
93
32 92
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
Figura 4.34. Efecto del tratamiento con LMS tras un pre-tratamiento de explosión por vapor en
las propiedades mecánicas y ópticas del papel: a) grado de refino, b) densidad aparente, c)
índice de tracción, d) índice de desgarro, e) blancura y f) opacidad. El control consistió en un
tratamiento de explosión por vapor seguido de un pasteado Kraft, sin tratamiento enzimático.
115
Proceso Kraft en Biorrefinería
116
4 – Discusión de los resultados
Tabla 4.6. Influencia de las diferentes extracciones en la calidad de las pastas y en el proceso
Kraft.
117
Proceso Kraft en Biorrefinería
NoE – P AC – P W– P AC+W – P
NoE – T AC – T W– T AC+W – T
Figura 4.35. Colonización de los hongos (P: Pycnoporus sanguineus; T: Trametes sp. I-62) tras
28 días de tratamiento en astillas sin extraer (NoE), extraídas con acetona (AC), extraídas con
agua (W) y extraídas con acetona y agua (AC+W).
Tabla 4.7. Influencia del tratamiento fúngico en la calidad de las pastas y en el proceso Kraft
comparado con controles con la misma extracción pero sin inoculación fúngica.
118
4 – Discusión de los resultados
Al comparar las pastas biotratadas con las pastas control sometidas a la misma
extracción, las mayores reducciones en el número kappa debidas al efecto del hongo se
encontraron cuando las astillas no habían sido pre-extraídas y después de una extracción
con acetona. Estas muestran fueron las que mayor colonización del hongo y mayor
actividad lacasa mostraron, como se ha visto anteriormente. Sin embargo, considerando
el proceso global, el número kappa más bajo se obtuvo con las astillas pre-extraídas con
agua caliente y tratadas con Trametes sp. I-62.
El biopulpeo más eficaz de entre los ensayados en esta tesis, consistió en un pre-
tratamiento de explosión por vapor seguido de un tratamiento LMS a 40ºC durante 3
horas, con una dosis de lacasa de 3 UA/g, como se concluyó en el apartado 4.4.1. Este
tratamiento con LMS, aplicado después de un pre-tratamiento de explosión por vapor,
proporcionó pastas con menor número kappa y menor consumo de reactivos que los
tratamientos con xilanasa llevados a cabo en los ensayos preliminares y que los
tratamientos con xilanasa llevados a cabo por Zhao et al. (2002) antes de un pasteado
alcalino de paja de trigo.
119
Proceso Kraft en Biorrefinería
120
4 – Discusión de los resultados
En la figura 4.36 se indican las partes del proceso de biorrefinería forestal en las
que se centra este apartado, en el cual el E. globulus ha sido sustituido por el H. funifera
como materia prima.
Fermentación
Bioetanol
Figura 4.36. Esquema del proceso de biorrefinería forestal, indicando las partes del proceso en
las que se centran la publicación VII que utiliza H. funifera como materia prima.
121
Proceso Kraft en Biorrefinería
122
4 – Discusión de los resultados
a cabo una pre-extracción en agua fría similar a la llevada a cabo en los tratamientos con
E. globulus.
Tabla 4.8. Condiciones y rendimiento del pre-tratamiento de explosión por vapor y capacidad
de retención de agua de las fracciones sólidas obtenidas.
HP SE1’ SE2’
Pre-extracción No Sí Sí
er
1 ciclo (min) - 3 10
2º ciclo (min) - - 3
Severidad S0 - 2,92 3,56
Rendimiento (%) - 81,9 62,5
a
% agua retenida 267 359 434
a
Capacidad de retención de agua expresada como gramos de agua por 100 gramos de muestra
seca.
123
Proceso Kraft en Biorrefinería
SE2’ (SE8). Otros autores ya observaron en los materiales no-madereros una estructura
más porosa y accesible que los madereros (Alaejos et al., 2004; Sánchez et al., 2011).
a) b)
55% 50% 46,2
45,4
Rendimiento global Kraft
44,8 46,0
45,0 39,9
44,2
45%
40,2 40% 35,7 36,3
40% 37,0 36,6
35%
35%
30%
30%
25% 25%
20% 20%
HP SE1' SE2' HP SE1' SE2'
Figura 4.37. Rendimiento global (a) y útil (b) del pasteado para H. funifera tras un pre-
tratamiento de explosión por vapor (SE1’ y SE2’) y un tratamiento con LMS. Se incluyen los
controles sin explosión por vapor (HP), sin adición de LMS (control-tampón) y sin etapa
enzimática (control).
124
4 – Discusión de los resultados
de celulosa antes del pasteado (de 44,7% en HP a 49,5% en SE2’) aumentando tanto el
rendimiento global como el útil en SE1’ y SE2’. Sin embargo, si se tiene en cuenta el
rendimiento del pre-tratamiento, el rendimiento total del proceso para HP, SE1’ y SE2’
fue 37%, 36% y 29% respectivamente. En cuanto al tratamiento enzimático, no supuso
diferencias significativas en el rendimiento del proceso.
a) b)
16 1400
14,1 1124
1134
1200 1114
14 1064 1020
Viscosidad (ml/g)
1010
11,6 1000
Número Kappa
12 11,1
798 832 817
10,3 10,4
10,1 800
9,6
10
600
7,9 8,2
8
400
6 200
4 0
HP SE1' SE2' HP SE1' SE2'
Figura 4.38. Número kappa (a) y viscosidad (b) de las pastas Kraft de H. funifera tras un pre-
tratamiento de explosión por vapor (SE1’ y SE2’) y un tratamiento con LMS. Se incluyen los
controles sin explosión por vapor (HP), sin adición de LMS (control-tampón) y sin etapa
enzimática (control).
Por otro lado, se observó una reducción en el número kappa cuando se llevó a
cabo un tratamiento enzimático. Esta reducción fue una consecuencia del tratamiento
125
Proceso Kraft en Biorrefinería
126
4 – Discusión de los resultados
aproximadamente 23 para los controles y controles-tampón hasta 17,0, 17,4 y 18,0 para
HP-LMS, SE1’-LMS y SE2’-LMS respectivamente). Un efecto similar se observó en el
biopulpeo con LMS de E. globulus (apartado 4.4.1) cuando se utilizaron dosis altas de
lacasa. Sin embargo, el álcali activo residual mostró que el consumo total de reactivos
no varió de manera significativa al añadir LMS en el tratamiento enzimático.
Tabla 4.9. Álcali activo (AA) y sulfidez (S) residual en las lejías negras obtenidas tras el
pasteado Kraft de las distintas muestras.
Muestra AA residual S residual
HP - Control 4,0 23,1
HP - Control-tampón 4,9 24,4
HP - LMS 4,7 17,0
SE1’ - Control 4,6 25,0
SE1’ - Control-tampón 3,9 20,3
SE1’ - LMS 3,8 17,4
SE2’ - Control 5,4 19,8
SE2’ - Control-tampón 4,5 22,1
SE2’ - LMS 5,1 18,0
127
Proceso Kraft en Biorrefinería
a) b)
42 6
Índice estallido (KPam2/g)
Índice desgarro (mNm 2/g)
37 5
32
27 4
22 3
17 2
12
7 1
2 0
10 30 50 70 10 30 50 70
ºSR ºSR
SE Control Indust. HP HP-SE
Figura 4.39. Índices de desgarro (a) y de estallido (b) para pastas de Hesperaloe funifera sin
refinar sometidas al pre-tratamiento SE2’ (HP-SE) y no pre-tratadas (HP). Se muestran también
los índices obtenidos a distintos grados de refino con pastas de E. globulus sin pre-tratar
(control e industrial) y sometida a un pre-tratamiento similar a SE2’ (SE).
128
4 – Discusión de los resultados
Las pastas control HP, SE1’ y SE2’ se sometieron a una secuencia TCF de
blanqueo, incluyendo una primera etapa enzimática para estudiar el efecto del pre-
tratamiento de explosión por vapor en el bioblanqueo de H. funifera. Como control, se
utilizó la misma secuencia LEP pero sin adición de LMS en la etapa enzimática (-Lc).
129
Proceso Kraft en Biorrefinería
a)
16
14,1 14,1 Sin blanquear L LE LEP
14
12,5
12,1 11,6
12 11,6 11,2
11,2
10,1 10,1
Número Kappa
0
HP-Lc HP-L SE1'-Lc SE1'-L SE2'-Lc SE2'-L
b)
1400
Sin blanquear L LE LEP
1134 1134 1121
1200 1105 1076 1042 1051
1082 1010
1051 1040 1033 988
989 1006 989
Viscosidad (ml/g)
1000 846
829 837 834
823
798 798 780
800
600
400
200
0
HP-Lc HP-L SE1'-Lc SE1'-L SE2'-Lc SE2'-L
Figura 4.40. Evolución del número kappa (a) y la viscosidad (b) durante la secuencia de
bioblanqueo (LEP) para H. funifera sin tratar (HP) y con pre-tratamiento de explosión por vapor
(SE1’ y SE2’). Se incluye un control sin adición de LMS (-Lc).
130
4 – Discusión de los resultados
a)
80 Sin blanquear L LE LEP
67 68
70
60 61 62
58
60
Blancura (% ISO)
53
51
50 45 45 46 46
44 45 44
43 43 43
40 42
40 36 36 36 36
30
20
10
0
HP-Lc HP-L SE1'-Lc SE1'-L SE2'-Lc SE2'-L
b) c)
13 95
90
12
HP HP
HP-LEP 85 HP-LEP
b* 11 SE1' L* SE1'
SE1'-LEP 80 SE1'-LEP
SE2' SE2'
10
SE2'-LEP 75 SE2'-LEP
9 70
-1 0 1 2 3 4 10 11 12 13
a* C*
Figura 4.41. Propiedades ópticas de las pastas bioblanqueadas de H. funifera sin tratar (HP) y
con pre-tratamiento de explosión por vapor (SE1’ y SE2’): a) evolución de la blancura en la
secuencia LEP; b) coordinadas CIE L*a*b* y c) coordinadas CIE L*C* para pastas sin
blanquear (símbolos oscuros) y bioblanqueadas (símbolos claros).
Las pastas pre-tratadas mediante explosión por vapor presentaron una blancura
mayor que las pastas sin pretratar, que aumentó al incrementar la severidad del pre-
tratamiento y al incluir un tratamiento con LMS en la secuencia de blanqueo. Además
de la blancura, el color del papel es una propiedad muy importante en diferentes
aplicaciones y una buena base blanca es necesaria cuando los productos se van a teñir,
imprimir o colorear. Por esta razón, se midieron también las coordenadas CIE L*a*b* y
CIE L*C* en las pastas antes y después del bioblanqueo (figura 4.41b y 4.41c). En
todas las pastas se observó una reducción de la coordenada a* (que indica la posición
entre el rojo y el verde) después del bioblanqueo, llegando a valores de a* muy
próximos al valor neutro ideal (0). Sin embargo, en las muestras sin pre-tratar (HP) la
coordenada b* (posición entre el amarillo y el azul) apenas cambia durante el blanqueo,
e incluso aumenta (más amarillo) para las pastas pre-tratadas SE1’. Por el contrario,
cuando se llevó a cabo un pre-tratamiento más intenso (SE2’), se observó una
131
Proceso Kraft en Biorrefinería
Tabla 4.10. Consumo de peróxido de hidrógeno medido a partir de los efluentes de la etapa P.
Muestra Consumo H2O2 (%)
HP-Lc 90,3
HP-L 92,7
SE1-Lc 87,5
SE1-L 88,0
SE2-Lc 66,1
SE2-L 68,1
Otra ventaja del pre-tratamiento de explosión por vapor fue la reducción del
consumo de peróxido de hidrógeno durante el blanqueo. Como se puede observar en la
tabla 4.10, un pre-tratamiento más intenso dio lugar a un consumo menor de peróxido
de hidrógeno, observando una reducción del 24,6% con el pre-tratamiento SE2’. Un
comportamiento similar se observó en el caso del E. globulus, en el que el pre-
tratamiento de explosión por vapor redujo la cantidad de ácidos hexenurónicos en la
pasta sin blanquear, reduciendo por tanto el consumo de reactivos de blanqueo. Sin
embargo, cuando se llevó a cabo un tratamiento LMS, se observó un pequeño aumento
del consumo de peróxido de hidrógeno en todas las pastas de H. funifera,
contrariamente a lo observado en el caso del E. globulus y por otros autores utilizando
distintas materias primas (Eugenio et al., 2010a; Valls y Roncero, 2009). Sin embargo,
se debe tener en cuenta, que las pastas obtenidas al final de la secuencia que incluye
LMS presentaron un número kappa menor y una blancura mayor, y por tanto, si se
comparan pastas con similar deslignificación y nivel de blancura, el consumo de
132
4 – Discusión de los resultados
reactivos en las pastas con tratamiento LMS sería probablemente menor que en las
pastas control.
133
5
CONCLUSIONES / CONCLUSIONS
5 – Conclusiones / Conclusions
5. CONCLUSIONES / CONCLUSIONS
137
Proceso Kraft en Biorrefinería
138
5 – Conclusiones / Conclusions
40%. Aunque las pastas obtenidas necesitaron mayor energía durante el refino,
las propiedades mecánicas fueron sólo ligeramente inferiores.
139
Proceso Kraft en Biorrefinería
CONCLUSIONS
The main conclusions of the work conducted in this doctoral thesis are the
following:
3) The chemical and structural changes caused by the steam explosion treatment,
makes it a suitable pre-treatment for Kraft pulping. That way, this process could
be transformed into a lignocellulosic biorefinery with a liquid fraction rich in
hemicelluloses to be used for the production of bioethanol or other high value-
added products.
4) The main components of the liquid fraction obtained in the steam explosion pre-
treatment were xylooligosacharides, which can be used as a sugar source for
bioethanol production. Enzymatic hydrolysis of this fraction provided better
140
5 – Conclusiones / Conclusions
yields than acid hydrolysis, allowing also for milder operational conditions that
would not favor the generation of inhibiting compounds. The enzymatic yield
achieved was 83%. To improve the hydrolysis and the subsequent fermentation
yields, it would be of interest to study the detoxification of this liquid fraction by
pre-treatments based on the laccase-mediator system.
5) The enzymatic hydrolysis of the solid fraction was also evaluated, considering
the bioethanol production as an alternative to the pulp production. The
hydrolysis yield was observed to increase with the severity of the pre-treatment
and, also, when applying LMS before the enzymatic hydrolysis. However, it
should be noted that a more severe pre-treatment would elicit degradation of the
polysaccharides, reducing the recovery of xylooligosaccharides in the liquid
fraction and increasing the generation of inhibiting products. Therefore, a mild
pre-treatment, and detoxification with LMS treatments, would increase the
overall sugar yield. However, the amount of sugar obtained from the solid
fraction using the studied pre-treatments represents only 12.3% of the raw
material. This finding and the high price of the Kraft pulp, makes Kraft pulping
stand out as the best option to exploit the solid fraction, using only the liquid
fraction for the production of bioethanol.
6) The open structure of the solid residue obtained after the steam explosion pre-
treatment improved the diffusion of the reagents during a subsequent Kraft
pulping, increasing the delignification by up to 70%. Alternatively, the cooking
time was reduced from 60 to 25 minutes obtaining pulps with similar
delignification grades, reducing the pulping energy consumption and increasing
the productivity of a hypothetical mill. Nevertheless, the pulp yield of the
process decreased from 50% to 40%. Although the pulp obtained required more
energy during refining, the mechanical properties were only slightly worse.
141
Proceso Kraft en Biorrefinería
142
6
TRABAJOS FUTUROS
6 –Trabajos futuros
6. TRABAJOS FUTUROS
145
7
BIBLIOGRAFIA
TRABAJOS FUTUROS
5 – Conclusiones / Conclusions
7. BIBLIOGRAFÍA
Agenda-2020. 2010. Forest products industry technology roadmap. American Forest &
Paper Association.
Ahvazi, B., Radiotis, T., Bouchard, J. y Goel, K. 2007. Chemical pulping of steam-
exploded mixed hardwood chips. Journal of Wood Chemistry and Technology,
27(2), 49-63.
Aitken, Y., Cadel, F. y Voillot, C. 1988. Constituants fibreux des pâtes, papiers et
cartons. Centre Technique de l’Industrie des Papiers, Cartons et Cellulose y
l’Ecole Français de Papeterie et des Industries Graphiques, Grenoble.
Alaejos, J., Lopez, F., Eugenio, M. E. y Fernandez, M. 2004. Non-wood raw materials
for pulp and paper making. A rewiew. Afinidad, 61(513), 400-410.
Alvira, P., Tomás Pejó, E., Negro, M. y Ballesteros, M. 2011. Strategies of xylanase
supplementation for an efficient saccharification and cofermentation process from
pretreated wheat straw. Biotechnology Progress, DOI: 10.1002/btpr.623.
149
Proceso Kraft en Biorrefinería
Babcock, L. W. 1932. Method of producing fermentable sugars and alcohol from wood,
US Patent 1,855,464.
Bajpai, P., Bajpai, P., Akhtar, M. y Jauhari, M. 2001. Biokraft pulping of eucalyptus
with selected lignin-degrading fungi. Journal of pulp and paper science, 27(7),
235-239.
Balakshin, M., Capanema, E., Chen, C. L., Gratzl, J., Kirkman, A. y Gracz, H. 2001.
Biobleaching of pulp with dioxygen in the laccase-mediator system--reaction
mechanisms for degradation of residual lignin. Journal of Molecular Catalysis B:
Enzymatic, 13(1-3), 1-16.
Ballesteros, I., Negro, M. J., Oliva, J. M., Cabañas, A., Manzanares, P. y Ballesteros, M.
2006. Ethanol production from steam-explosion pretreated wheat straw. Applied
biochemistry and biotechnology, 130(1), 496-508.
Britt, K. W. 1970. Handbook of pulp and paper technology. 2nd Edition ed. Van
Nostrand Reinhold, New York.
150
5 – Conclusiones / Conclusions
Cadena, E. M., Vidal, T. y Torres, A. L. 2010. Influence of the hexenuronic acid content
on refining and ageing in eucalyptus TCF pulp. Bioresource technology, 101(10),
3554-3560.
Camarero, S., Garcia, O., Vidal, T., Colom, J., del Rio, J. C., Gutiérrez, A., Gras, J. M.,
Monje, R., Martinez, M. J. y Martinez, A. T. 2004. Efficient bleaching of non-
wood high-quality paper pulp using laccase-mediator system. Enzyme and
microbial technology, 35(2-3), 113-120.
151
Proceso Kraft en Biorrefinería
Chirat, C., Lachenal, D. y Dufresne, A. 2010. Biorefinery in a Kraft pulp mill: from
bioethanol to cellulose nanocrystals. Cellulose Chemistry and Technology, 44(1-
3), 59-64.
Dias, A. A., Freitas, G. S., Marques, G. S. M., Sampaio, A., Fraga, I. S., Rodrigues, M.
A. M., Evtuguin, D. V. y Bezerra, R. M. F. 2010. Enzymatic saccharification of
biologically pre-treated wheat straw with white-rot fungi. Bioresource
Technology, 101(15), 6045-6050.
152
5 – Conclusiones / Conclusions
Eriksson, J., Malmsten, M., Tiberg, F., Callisen, T. H., Damhus, T. y Johansen, K. S.
2005. Enzymatic degradation of model cellulose films. Journal of colloid and
interface science, 284(1), 99-106.
Eugenio, M., Santos, S., Carbajo, J., Martín, J., Martín-Sampedro, R., González, A. y
Villar, J. 2010a. Kraft pulp biobleaching using an extracellular enzymatic fluid
produced by Pycnoporus sanguineus. Bioresource technology, 101(6), 1866-1870.
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(9), 4252-4261.
153
Proceso Kraft en Biorrefinería
Ferraz, A., Guerra, A., Mendonça, R., Masarin, F., Vicentim, M. P., Aguiar, A. y Pavan,
P. C. 2008. Technological advances and mechanistic basis for fungal biopulping.
Enzyme and Microbial Technology, 43(2), 178-185.
Fillat, A., Colom, J. F. y Vidal, T. 2010. A new approach to the biobleaching of flax
pulp with laccase using natural mediators. Bioresource technology, 101(11),
4104-4110.
Grethlein, H. E. 1985. The effect of pore size distribution on the rate of enzymatic
hydrolysis of cellulosic substrates. Nature Biotechnology, 3(2), 155-160.
154
5 – Conclusiones / Conclusions
Gutiérrez, A., Rencoret, J., Ibarra, D., Molina, S., Camarero, S., Romero, J., José, C. y
Martínez, Á. T. 2007. Removal of lipophilic extractives from paper pulp by
laccase and lignin-derived phenols as natural mediators. Environmental science &
technology, 41(11), 4124-4129.
Hayn, M., Steiner, W., Klinger, R., Steinmüller, H., Sinner, M. y Esterbauer, H. 1993.
Basic research and pilot studies on the enzymatic conversion of lignocellulosic.
in: Bioconversion of forest and agricultural plant residues, (Ed.) Saddler, J. N.,
CAB International. Wallingford, UK.
Heitner, C., Argyropoulos, D. S., Miles, K. D., Karnis, A. y Kerr, R. D. 1993. Alkaline
sulphite ultra-yield pulping of Aspen chips – A comparison of steam-explosion
and conventional chemimechanical pulping. Journal of Pulp & Paper Science,
19(2), 58-70.
Heitz, M., Capek-Menard, E., Koeberle, P., Gagne, J. y Chornet, R. 1991. Fractionation
of Populus tremuloides at the pilot plant scale: optimization of steam pretreatment
conditions using the STAKE II technology. Bioresource technology, 35(1), 23-32.
Helmerius, J., von Walter, J. V., Rova, U., Berglund, K. A. y Hodge, D. B. 2010.
Impact of hemicellulose pre-extraction for bioconversion on birch Kraft pulp
properties. Bioresource Technology, 101(15), 5996-6005.
Henriksson, G., Brännvall, E. y Lennholm, H. 2007. The trees. in: Wood chemistry and
wood biotechnology, (Ed.) Fiber and polimer technology, K. Stockholm.
155
Proceso Kraft en Biorrefinería
Howard, R., Abotsi, E., Jansen van Rensburg, E. y Howard, S. 2003. Review-
Lignocellulose biotechnology: issues of bioconversion and enzyme production.
African Journal of Biotechnology, 2, 12.
Ibarra, D., Chávez, M. I., Rencoret, J., Suárez, G., Sánchez, R., Camarero, S.,
Hernández, M., Jesús, M., Jiménez-Barbero, J. y Ferrer, M. 2007. Structural
modification of eucalypt pulp lignin in a totally chlorine free bleaching sequence
including a laccase-mediator stage.
156
5 – Conclusiones / Conclusions
Kantelinen, A., Hortling, B., Sundquist, J., Linko, M. y Viikari, L. 1993. Proposed
mechanism of the enzymatic bleaching of kraft pulp with xylanases.
Holzforschung, 47(4), 318-324.
Karlsson, H. 2006. Fiber Guide. Fiber analysis and process applications in the pulp
and paper industry. Elanders Tofters AB, Stockholm, Sweden.
Leschinsky, M., Sixta, H. y Patt, R. 2009. Detailed mass balances of the autohydrolysis
of Eucalyptus globulus at 170 degrees C. Bioresources, 4(2), 687-703.
157
Proceso Kraft en Biorrefinería
Li, H., Saeed, A., Jahan, M. S., Ni, Y. y van Heiningen, A. 2010. Hemicellulose
removal from hardwood chips in the pre-hydrolysis step of the kraft-based
dissolving pulp production process. Journal of Wood Chemistry and Technology,
30(1), 48-60.
Maijala, P., Kleen, M., Westin, C., Poppius-Levlin, K., Herranen, K., Lehto, J.,
Reponen, P., Mäentausta, O., Mettälä, A. y Hatakka, A. 2008. Biomechanical
pulping of softwood with enzymes and white-rot fungus Physisporinus rivulosus.
Enzyme and Microbial Technology, 43(2), 169-177.
Mamers, H., Yuritta, J. y Menz, D. 1981. Explosion pulping of bagasse and wheat
straw. Tappi Journal, 64(7).
Mao, H. B., Genco, J. M., van Heiningen, A. y Pendse, H. 2010. Kraft mill biorefinery
to produce acetic acid and ethanol: Technical economic analysis. Bioresources,
5(2), 525-544.
158
5 – Conclusiones / Conclusions
Marinova, M., Mateos-Espejel, E., Jemaa, N. y Paris, J. 2009. Addressing the increased
energy demand of a Kraft mill biorefinery: The hemicellulose extraction case.
Chemical Engineering Research and Design, 87(9), 1269-1275.
Martinez, J., Granado, J., Montane, D., Salvado, J. y Farriol, X. 1995. Fractionation of
residual lignocellulosics by dilute-acid prehydrolysis and alkaline extraction:
Application to almond shells. Bioresource Technology, 52(1), 59-67.
Mason, W. H. 1926. Process and apparatus for disintegration of wood and the like, US
Patent 1,578,609.
Mendes, C. V. T., Carvalho, M., Baptista, C., Rocha, J. M. S., Soares, B. I. G. y Sousa,
G. D. A. 2009. Valorisation of hardwood hemicelluloses in the kraft pulping
process by using an integrated biorefinery concept. Food and Bioproducts
Processing, 87(C3), 197-207.
Mendonça, R. T., Jara, J. F., González, V., Elissetche, J. P. y Freer, J. 2008. Evaluation
of the white-rot fungi Ganoderma australe and Ceriporiopsis subvermispora in
biotechnological applications. Journal of Industrial Microbiology and
Biotechnology, 35(11), 1323-1330.
159
Proceso Kraft en Biorrefinería
Moldes, D., Cadena, E. y Vidal, T. 2010. Biobleaching of eucalypt kraft pulp with a two
laccase-mediator stages sequence. Bioresource technology, 101(18), 6924-6929.
Moldes, D., Díaz, M., Tzanov, T. y Vidal, T. 2008. Comparative study of the efficiency
of synthetic and natural mediators in laccase-assisted bleaching of eucalyptus
kraft pulp. Bioresource technology, 99(17), 7959-7965.
Montane, D., Salvado, J., Farriol, X., Jollez, P. y Chornet, E. 1994. Phenomenological
kinetics of wood delignification: application of a time-dependent rate constant and
a generalized severity parameter to pulping and correlation of pulp properties.
Wood Science and Technology, 28(6), 387-402.
Ohgren, K., Rudolf, A., Galbe, M. y Zacchi, G. 2006. Fuel ethanol production from
steam-pretreated corn stover using SSF at higher dry matter content. Biomass and
Bioenergy, 30(10), 863-869.
Oliva, J. M., Sáez, F., Ballesteros, I., González, A., Negro, M. J., Manzanares, P. y
Ballesteros, M. 2003. Effect of lignocellulosic degradation compounds from
steam explosion pretreatment on ethanol fermentation by thermotolerant yeast
Kluyveromyces marxianus. Applied biochemistry and biotechnology, 105(1), 141-
153.
Oriaran, T. P., Labosky Jr, P. y Blankenhorn, P. R. 1990. Kraft pulp and papermaking
properties of Phanerochaete chrysosporium-degraded aspen. Tappi Journal,
73(7), 147-152.
160
5 – Conclusiones / Conclusions
Paice, M., Bernier Jr, R. y Jurasek, L. 1988. Viscosity enhancing bleaching of hardwood
kraft pulp with xylanase from a cloned gene. Biotechnology and bioengineering,
32(2), 235-239.
Pere, J., Ellmén, J. y Viikari, L. 2005. Process for preparing mechanical pulp, WO
Patent WO/2005/056,915.
Pérez, J., Ballesteros, I., Ballesteros, M., Sáez, F., Negro, M. y Manzanares, P. 2008.
Optimizing liquid hot water pretreatment conditions to enhance sugar recovery
from wheat straw for fuel-ethanol production. Fuel, 87(17-18), 3640-3647.
Petit-Conil, M., Semar, S., Niku-Paavola, M. L., Sigoillot, J., Asther, M., Anke, H. y
Viikari, L. 2002. Potential of laccases in softwood-hardwood high-yield pulping
and bleaching. Progress in Biotechnology, 21, 193-201.
Quesada, J., Rubio, M. y Gómez, D. 1998. Direct action of molecular ozone against
lignin and its standard compounds. Afinidad, 55(478), 413-421.
Ragauskas, A. J., Nagy, M., Kim, D. H., Eckert, C. A., Hallett, J. P. y Liotta, C. L.
2006. From wood to fuels: integrating biofuels and pulp production. Industrial
Biotechnology, 2(1), 55-65.
161
Proceso Kraft en Biorrefinería
Ragnar, M. 2001. On the importance of the structural composition of pulp for the
selectivity of ozone and chlorine dioxide bleaching. Nordic Pulp & Paper
Research Journal, 16(1), 72-79.
Ramos, J., Rojas, T., Navarro, F., Dávalos, F., Sanjuán, R., Rutiaga, J. y Young, R. A.
2004. Enzymatic and fungal treatments on sugarcane bagasse for the production
of mechanical pulps. Journal of agricultural and food chemistry, 52(16), 5057-
5062.
Ramos, L., Breuil, C., Kushner, D. y Saddler, J. 1992. Steam pretreatment conditions
for effective enzymatic hydrolysis and recovery yields of Eucalyptus viminalis
wood chips. Holzforschung, 46(2), 149-154.
Renneckar, S., Johnson, R., Zink-Sharp, A., Sun, N. y Glasser, W. 2005. Fiber
modification by steam-explosion: 13C NMR and dynamic mechanical analysis
studies of co-refined wood and polypropylene. Composite Interfaces, 12(6), 559-
580.
Robert, D., Bardet, M., Lapierre, D. y Gellerstedt, G. 1988. Structural changes in aspen
lignin during steam explosion treatment. Cellulose Chemistry and Technology, 22,
221-230.
162
5 – Conclusiones / Conclusions
Ruiz, E., Cara, C., Manzanares, P., Ballesteros, M. y Castro, E. 2008. Evaluation of
steam explosion pre-treatment for enzymatic hydrolysis of sunflower stalks.
Enzyme and Microbial Technology, 42(2), 160-166.
San Martin, R., Perez, C. y Briones, R. 1995. Simultaneous production of ethanol and
kraft pulp from pine (Pinus radiata) using steam explosion. Bioresource
technology, 53(3), 217-223.
Sánchez, R., Rodríguez, A., García, J., Rosal, A. y Jiménez, L. 2011. Exploitation of
hemicellulose, cellulose and lignin from Hesperaloe funifera. Bioresource
technology, 102, 1308-1315.
Sanchez, R., Rodriguez, A., Navarro, E., Conesa, J. y Jimenez, L. 2010. Use of
Hesperaloe funifera for the production of paper and extraction of lignin for
synthesis and fuel gases. Biomass and Bioenergy, 34(10), 1471-1480.
Selig, M. J., Viamajala, S., Decker, S. R., Tucker, M. P., Himmel, M. E. y Vinzant, T.
B. 2007. Deposition of lignin droplets produced during dilute acid pretreatment of
maize stems retards enzymatic hydrolysis of cellulose. Biotechnology Progress,
23(6), 1333-1339.
163
Proceso Kraft en Biorrefinería
Shimizu, K., Sudo, K., Ono, H., Ishihara, M., Fujii, T. y Hishiyama, S. 1998. Integrated
process for total utilization of wood components by steam-explosion pretreatment.
Biomass and Bioenergy, 14(3), 195-203.
Sigoillot, C., Camarero, S., Vidal, T., Record, E., Asther, M., Pérez-Boada, M.,
Martinez, M. J., Sigoillot, J. C. y Colom, J. F. 2005. Comparison of different
fungal enzymes for bleaching high-quality paper pulps. Journal of biotechnology,
115(4), 333-343.
164
5 – Conclusiones / Conclusions
Valls, C. y Roncero, B. 2009. Using both xylanase and laccase enzymes for pulp
bleaching. Bioresource technology, 100(6), 2032-2039.
Valls, C., Vidal, T. y Roncero, M. B. 2010. The role of xylanases and laccases on
hexenuronic acid and lignin removal. Process Biochemistry, 45(3), 425-430.
Van Heiningen, A. 2006. Converting a kraft pulp mill into an integrated forest
biorefinery. Pulp and Paper Canada, 107(6), 38-43.
Viikari, L., Ranua, M., Kantelinen, A., Sundquist, J. y Linko, M. 1986. Bleaching with
enzymes. Proceeding of Biotechnology in the Pulp and Paper Industry. pp. 67-69.
Vila, C., Romero, J., Francisco, J. L., Garrote, G. y Parajo, J. C. 2011. Extracting value
from Eucalyptus wood before kraft pulping: Effects of hemicelluloses
solubilization on pulp properties. Bioresource Technology, 102(8), 5251-5254.
Vivekanand, V., Dwivedi, P., Sharma, A., Sabharwal, N. y Singh, R. 2008. Enhanced
delignification of mixed wood pulp by Aspergillus fumigatus laccase mediator
system. World Journal of Microbiology and Biotechnology, 24(12), 2799-2804.
Vuorinen, T., Fagerström, P., Buchert, J., Tenkanen, M. y Teleman, A. 1999. Selective
hydrolysis of hexenuronic acid groups and its application in ECF and TCF
bleaching of kraft pulps. Journal of pulp and paper science, 25(5), 155-162.
165
Proceso Kraft en Biorrefinería
Wolfaardt, F., Taljaard, J., Jacobs, A., Male, J. y Rabie, C. 2004. Assessment of wood-
inhabiting basidiomycetes for biokraft pulping of softwood chips. Bioresource
technology, 95(1), 25-30.
Wong, K. 1997a. Mechanisms of xylanase aided bleaching of kraft pulp. Part 1: Process
parameters. Appita journal, 50(5), 415-422.
Wong, K. 1997b. Mechanisms of xylanase aided bleaching of kraft pulp. Part 2: Target
substrates. Appita journal, 50(6), 509-518.
Wright, L., Boundy, B., Perlack, B., Davis, S. y Saulsbury, B. 2006. Biomass Energy
Data Book. U.S. Department of Energy (DOE), Oak Ridge, TN.
Wu, M. M., Chang, K., Gregg, D. J., Boussaid, A., Beatson, R. P. y Saddler, J. N. 1999.
Optimization of steam explosion to enhance hemicellulose recovery and
enzymatic hydrolysis of cellulose in softwoods. Applied Biochemistry and
biotechnology, 77(1), 47-54.
Yang, B. y Wyman, C. E. 2004. Effect of xylan and lignin removal by batch and
flowthrough pretreatment on the enzymatic digestibility of corn stover cellulose.
Biotechnology and bioengineering, 86(1), 88-98.
Zhang, X., Renaud, S. y Paice, M. 2005. The potential of laccase to remove extractives
present in pulp and white water from TMP newsprint mills. Journal of pulp and
paper science, 31(4), 175-180.
166
5 – Conclusiones / Conclusions
Zhao, J., Li, X., Qu, Y. y Gao, P. 2002. Xylanase pretreatment leads to enhanced soda
pulping of wheat straw. Enzyme and Microbial Technology, 30(6), 734-740.
167
ANEXOS:
PUBLICACIONES
PUBLICACIÓN I:
Steam explosion treatment of Eucalyptus globulus wood: Influence of operational
conditions on chemical and structural modifications.
Martín-Sampedro, R., Martín, J.A., Eugenio, M.E., Revilla, E., Villar, J.C.
171
PEER-REVIEWED ARTICLE bioresources.com
INTRODUCTION
Extensive research efforts to improve the processing and digestion of biomass for
several applications have been pursued in the last few years. Steam explosion (SE) is one
of the most widely used treatments for fractionation of lignocellulosic components. It is
defined as a process by which the lignocellulosic material is exposed to high pressure
steam before it undergoes very rapid decompression. This sudden decompression leads to
an “explosion” of the steam inside the lignocellulosic matrix, which promotes breakdown
and defibrillation of its structure, hydrolysis of the hemicelluloses, and depolymerisation/
repolymerization of the lignin (Ruiz et al. 2008). Thus, the accessibility of the material to
chemical or enzymatic degradation is greatly increased (Moniruzzaman 1996; Martin-
Sampedro et al. 2011). The advantages of SE include a significantly lower environmental
impact, low capital investment, and fewer hazardous chemicals applied in the process
compared to other methods for hemicelluloses extraction (Li et al. 2001). On the other
hand, the presence of degradation products from sugar and lignin is unavoidable and must
be taken into account in order to minimize the potential inhibitory effect on subsequent
steps such as the enzymatic hydrolysis (Tengborg et al. 2001).
EXPERIMENTAL
Raw Material
E. globulus chips were kindly supplied by La Montañanesa pulp mill (Spain). The
material was air dried until reaching constant moisture, and then homogenized in a single
stock (by conditioning inside polyethylene bags) to avoid differences in composition and
water content. The chips were stored in polyethylene bags at 25º C.
C1 No - -
C2 Yes - -
NH-5 No 1 5
H-5 Yes 1 5
NH-10 No 1 10
H-10 Yes 1 10
NH-5+3 No 2 5
H-5+3 Yes 2 5
NH-10+3 No 2 10
H-10+3 Yes 2 10
Chemical Analysis
In order to carry out the chemical analysis, all the samples were dried at room
temperature and then milled in a Wiley mill. The samples were sieved using standard
sieves to obtain 20 g of wood meal sized between 0.30 and 0.40 mm. Acetone extractives
(UNE-EN ISO 14453), hot water extractives (UNE 57-013-82), lignin content (TAPPI T
222 om-88), and holocellulose content (Wise et al. 1946) were then measured on the
wood meal. All determinations were duplicated.
The concentrations of furfural and hydroxymethyl furfural (HMF) in the liquid
fractions obtained during the experiments were also determined by High Performance
Liquid Chromatography (HPLC) using the Agilent Technology 1200 series RID (Las
Rozas, Spain) with an Aminex HPX-87H column (Bio-Rad, Alcobendas, Spain). The
equipment was operated at 50 ºC with a mobile phase containing 5 mM sulfuric acid
pumped at a rate of 0.6 mL min-1.
Chemical Analysis
The SE had a clear effect on the chemical composition of the treated chips.
Figures 2a and 2b show the acetone and hot water extractives. As can be observed in Fig.
2a, there was a clear increase in the percentage of acetone extractives in all the exploded
samples. This increase could be partly associated with the degradation of lignin caused by
the treatments (Chandra et al. 2007). Longer times of treatment resulted in higher extrac-
tive percentages (increase of 1.3-3.1% and 2.4-6.2% for 5-10 min of treatment, for
hydrated and non-hydrated samples, respectively), but when a second SE cycle was
added, similar or even lower percentages of acetone extractives were found (increase of
1.0-4.5% and 3.2-4.2% for 5-10 min of treatment, for hydrated and non-hydrated samples
respectively). This is likely to be caused by the removal effect of the second cycle. The
pre-hydration of chips did not always show a meaningful effect but seemed to be more
relevant during the first cycle with long times of treatment.
As Fig. 2b reveals, SE also increased the percentage of hot water extractives in all
the treated samples, with differences of 3.6% to 7.1%. The increase seemed to be
independent of the treatment duration, number of cycles, and previous hydration, which
suggests that a threshold severity value, causing the maximum formation of water soluble
compounds, was reached even with the mildest treatment.
There was a significant decrease in the holocellulose content after every treatment
(Fig. 2c). The relative value was reduced from the 74-77% (controls) down to 61-65%,
regardless of the SE conditions. A similar reduction of holocellulose content took place in
all the treated samples, which could indicate that the severity of the treatments was
enough to reach a constant degree of holocellulose degradation, causing an increase in
hot water extractives, as already described. However, if steam explosion yield is taken
into account, it is possible to determine the percentage of the original holocellulose that
remains in the solid after treatment, and therefore, estimate the real holocellulose
degradation. Table 2 shows the treatment yields and the percentages of holocellulose and
Klason lignin that remain in the solid fraction after each steam explosion treatment,
compared to the original amount in the raw material (C1). As can be observed, the
percentage of holocellulose underwent a progressive reduction (due to hemicelluloses
solubilization) when the severity of the treatment increased (longer treatment and/or
addition of a second SE cycle). The pre-hydration of the chips, also contributed to the
solubilization of hemicelluloses.
10% 15%
a Without previous hydration b Without previous hydration
6% 9%
4% 6%
2% 3%
0% 0%
0 5 10 5+3 10+3 0 5 10 5+3 10+3
Time of treatment, min. Time of treatment, min.
90% 30%
c Without previous hydration d Without previous hydration
80%
With previous hydration 25% With previous hydration
70%
Holocellulose ,%
Klason lignin, %
60% 20%
50%
15%
40%
30% 10%
20%
5%
10%
0% 0%
0 5 10 5+3 10+3 0 5 10 5+3 10+3
Time of treatment, min. Time of treatment, min.
Fig. 2. Relative percentages of acetone extractives (a), hot water extractives (b), holocellulose
(c), and Klason lignin (d).
300
g water adsorbed / 100 g wood without previous hydration
250 with previous hydration
200
150
100
50
0
0 5 10 5+3 10+3
Time of treatment, min.
Structural Properties
A structural characterization, assessing the changes in the porous network and the
surface configuration of the material, was developed as part of our research. This part of
our study was intended to reveal structural modifications in the treated material that
might be related with changes in the pore size distribution and accessible surface. As the
0,050 0,050
a C1 b C1
NH-5 NH-5+3
NH-10+3
0,030 0,030
0,020 0,020
0,010 0,010
0,000 0,000
10,00 100,00 1000,00 10000,00 10,00 100,00 1000,00 10000,00
Pore size diameter, nm Pore size diameter, nm
0,050 0,050
c C2 d C2
H-5 H-5+3
H-10+3
Relative vol, mLg-1
0,030 0,030
0,020 0,020
0,010 0,010
0,000 0,000
10,00 100,00 1000,00 10000,00 10,00 100,00 1000,00 10000,00
Pore size diameter, nm Pore size diameter, nm
Fig. 4.Relative pore volume versus pore size diameter. Samples without previous hydration (NH)
(a and b), samples previously hydrated (H) (c and d). C1 and C2 indicate control without and with
precious hydration respectively. The numbers after NH or H indicate the duration of the cycles in
minutes.
Figure 4 shows changes in relative pore volume (mL g-1) versus pore diameter of
samples without previous hydration, and with one (Fig. 4a) and two cycles (Fig. 4b) of
SE. It should be noted that the control is a unique sample in both figures, and that it
exhibits a band with a maximum ca. 33 nm. This broad band included the greater pore
volume of the control sample and ranged from 10 to 200 nm of pore diameter. The
control also showed a lower peak, between 3 and 4 µm, which corresponds to bigger but
less abundant pores. In the two samples treated with one cycle of SE (5 and 10 minutes),
there was a significant displacement of the band to greater pore size diameters. In this
pore size diameter range, the 10 minute treatment produced the biggest pore volume. The
addition of a second cycle generated pores with diameters between 3-4 and 6 µm. Again,
0,70
Specific volume, mLg-1
0,70
Specific volume, mLg-1
These results revealed a redistribution of the pore network in the treated material,
with a decrease of volume of small pores and an increase of volume of bigger pores.
Since there was not an increase of the total pore volume in the pore size range between
10 and 10,000 nm (data not shown), these results can be explained by assuming the
formation of pores of longer diameters derived from small pores. This can be observed in
Fig. 5, which shows the mesopore (pore diameter: 7-50 nm) and macropore (pore
diameter: 50-10,000 nm) specific volume (mL g-1) for every sample analyzed. The results
reveal a clear decrease of the mesopore relative volume in all treated samples, both
previously hydrated and non-previously hydrated (Fig. 5a). The relative volume of
macropores (Fig. 5b) increased in all treated samples, except for those exposed to the
mildest SE. In these cases (1 only cycle of 5 minutes), the results were similar to those
obtained with the control samples. These results confirm the possibility that there is a
reorganization of the pore network rather than creation of new pore volume, in this pore
size range. Thus, we may conclude that there is a global positive effect of SE on the
7,0 7,0
a C1 b C1
6,0 NH-5 Relative surface, m 2g-1 6,0 NH-5+3
Relative surface, m 2g-1
NH-10 NH-10+3
5,0 5,0
4,0 4,0
3,0 3,0
2,0 2,0
1,0 1,0
0,0 0,0
10 100 1000 10000 10 100 1000 10000
Pore size diameter, nm Pore size diameter, nm
7,0 7,0
c C2 d C2
Relative surface, m 2g-1
H-10
H-10+3
5,0 5,0
4,0 4,0
3,0 3,0
2,0 2,0
1,0 1,0
0,0 0,0
10 100 1000 10000 10 100 1000 10000
Pore size diameter, nm Pore size diameter, nm
Fig. 6.Relative surface versus pore size diameter. Samples without previous hydration (NH) (a
and b), samples previously hydrated (H) (c and d). C1 and C2 indicate control without and with
precious hydration respectively. The numbers after NH or H indicate the duration of the cycles in
minutes.
The relation between pore size and accessible surface was also evaluated in order
to get clear knowledge of the cost, in terms of surface, posed by the formation of bigger
pores derived from smaller pores. In Fig. 6 the relative surface area (m2 g-1) has been
plotted versus pore size diameter (nm) for every sample tested. When the pore diameter
ranged between 10 and 100 nm, there was a clear decrease in the relative surface in all
the treated samples, which corresponds to the disappearance of small pores in these
CONCLUSIONS
ACKNOWLEDGMENTS
The authors wish to thank the Spanish Ministry of Science and Innovation for
funding this study via Projects CTQ 2006-00385 and CTQ 2009-11268, the “Ramón y
Cajal” Programme, and the FPI fellowship (BES2007 – 14528).
REFERENCES CITED
Article submitted: September 4, 2011; Peer review completed: October 5, 2011; Revised
version received and accepted: October 14, 2011; Published: October 18, 2011.
Martín-Sampedro, R., Capanema, E.A., Hoeger, I.C., Villar, J.C., Rojas, O.J.
187
ARTICLE
pubs.acs.org/JAFC
ABSTRACT: Eucalyptus globulus chips were steam exploded followed by treatment with a laccase-mediator system (LMS) under
different experimental conditions. Removal of hemicelluloses and, to a lesser extent, lignin was observed. Thermogravimetic
analyses of whole meal obtained from chips before and after steam explosion indicated an increase in lignin degradation temperature
due to lignin condensation. In contrast, application of LMS treatment caused a reduction in lignin and polysaccharide degradation
temperatures. Lignins were isolated from wood samples before and after each treatment and analyzed by 2D NMR and 13C NMR.
An increase in carboxyl and phenolic hydroxyl groups and a significant decrease in β-O-4 structures were found in steam-exploded
samples. The most relevant changes observed after laccase treatment were increased secondary OH and degree of condensation.
KEYWORDS: Eucalyptus globulus, steam explosion, laccase-mediator system, thermogravimetric analyses (TGA), milled wood
lignin (MWL), 2D NMR, HSQC, quantitative 13C NMR, lignin thermal properties, lignin chemistry
r 2011 American Chemical Society 8761 dx.doi.org/10.1021/jf201605f | J. Agric. Food Chem. 2011, 59, 8761–8769
Journal of Agricultural and Food Chemistry ARTICLE
before and after steam explosion. Also, changes in lignin after Table 1. Summary of Conditions Used in Steam Explosion of
laccase treatment were investigated to elucidate the impact of Eucalyptus globulus Chips
steam explosion in downstream enzymatic treatments. Finally,
thermogravimetric analyses (TGA) were carried out on whole residence time during
meal biomass before and after steam explosion and enzyme steam explosion (min)
treatment to determine the effects of process conditions on the
water soaking prior to
thermal properties of the lignocellulosic material.
steam explosion severity
’ MATERIALS AND METHODS sample (16 h, 25 °C) cycle 1 cycle 2 factor (S0)
All of the chemicals used were of reagent grade and were obtained control no
from Sigma-Aldrich. SE1 no 5 3.14
Wood chips of E. globulus were kindly supplied by La Monta~ nanesa SE10 yes 5 3.14
pulp mill (Torraspapel group, Spain). The laccase used in enzymatic SE2 no 10 3.44
treatments (Novozym 51003) was supplied as a donation by Novo- SE20 yes 10 3.44
zymes (Bagsvaerd, Denmark).
SE3 no 5 3 3.35
Steam Explosion Treatment. The E. globulus chips were dried
and homogenized at room temperature for 2 weeks, and the moisture SE30 yes 5 3 3.35
content was determined (ISO 638). The size of the chips was around SE4 no 10 3 3.56
2 3 0.5 cm. Before the steam explosion treatment, the chips were SE40 yes 10 3 3.56
separated into eight 500 g batches. Four of them were immersed in SE40 +laccase a yes 10 3 3.56
distilled water during 16 h at 25 °C; the others were kept at room SE40 +control-L b yes 10 3 3.56
temperature. These two types of samples were used to evaluate the effect
of impregnation on the subsequent steam explosion treatment. The a
SE40 +laccase, same as SE40 but included a treatment of the chips
steam explosion treatment was performed in a 26 L stainless steel vessel after the second cycle of steam explosion with laccase-mediator system.
by charging steam to operation temperature and pressure of 183 °C and b
SE40 +control-L, a control sample after same treatment used in SE40
10 kg-f/cm2, respectively. After these conditions had been reached +laccase but with no laccase-mediator system added.
(typically within 3 min after steam injection), variable residence times
were applied. At the end of the steam treatment, the pressure was the isolation of MWL is shown schematically in Figure 1. Briefly, chips
reduced to 6 kg-f/cm2, and then the chips were discharged into an were ground to pass a 20 mesh screen in a Wiley mill and extracted with
expansion (blowing) tank at atmospheric pressure. After discharging, the an ethanol/toluene mixture 1:2 (v/v). The extractive-free wood meal
chips were washed with water. The steam explosion factors evaluated in was submitted to an alkaline extraction mainly to remove tannins. The
this work included the use of preimpregnation of wood chips in water, obtained extracted wood was then thoroughly washed until pH-neutral,
the number of steam explosion cycles (one or two), and the time dried under vacuum, and subjected to ball-milling in a planetary ball mill
employed during the first cycle (5 or 10 min). Table 1 summarizes the (Fritsch Planetary Mill P7). The resulting wood residue was extracted
conditions used and the nomenclature used thereafter to indicate the with dioxane/water solution (96:4) during 72 h. The dioxane extracts
respective steam-exploded (SE) samples. were combined and concentrated under vacuum. The extracted MWL
The severity factor, S0, for each treatment was calculated according to was dried under vacuum and stored in a desiccator for further analyses.
eq 1 defined by Overend and Chornet17 as a function of the temperature The liquor from the alkali extraction was acidified at pH 2 by adding
(T, °C) and the duration of the treatment (t, min). H2SO4 and the precipitated lignin recovered by filtration, water-washed
S0 ¼ logðeT100=14:75 tÞ ð1Þ until pH-neutral, air-dried, and stored for further analyses.
The milling time and intensity (as given by the number of balls per
Enzymatic Treatment. The steam-exploded sample denoted SE40 gram of sample) were adjusted to obtain yields between 30 and 50%.
was treated with a laccase-mediator system. Enzyme treatment was This was done to ensure MWL samples that could be taken as close as
performed in 500 mL flasks with air bubbling placed in a temperature- possible to the native lignin.11 Also, processing the material to a similar
controlled shaker operating at 100 rev/min. The LMS consisted of the isolation yield facilitates a better comparison between different samples.
laccase enzyme (30 U of laccase per gram of dry SE wood) and HBT Acetylation of the MWL, for NMR analysis, was carried out according to
(5 mM), which was used as mediator. The temperature, time, and pH of a procedure reported earlier by Capanema et al.19
treatment were 70 °C, 30 min, and 6 (100 mM phosphate buffer), Carbohydrate Composition and Klason Lignin. The carbo-
respectively. The final concentration of the substrate (steam-exploded hydrate composition was determined by high-performance liquid chro-
chips) was 7% (w/v). Here, 1 U of laccase is defined as the amount of matography (HPLC) analyses after acid hydrolysis of the wood
laccase required to convert 1 μmol/min of ABTS to its cationic radical samples.20 The equipment used consisted of an Agilent Technology
(0.1 M phosphate buffer, pH 6, at 24 °C). A control experiment with no 1200 series RID with an Aminex HPX-87H column operated at 50 °C
LMS was carried out (sample SE40 +control-L). with a mobile phase consisting of 5 mM sulfuric acid pumped at a rate of
Thermogravimetric Analysis. TGA of wood meal samples 0.6 mL/min. Klason lignin content in the wood samples and solid
(ground to pass a 20 mesh screen) were conducted in a TA Instruments fractions obtained during the isolation procedure (see Figure 1) were
model Q-500. Fifteen milligrams of the respective sample were used, and determined according to TAPPI T222 om-88 standard.
the thermograms were obtained by ramping the temperature from 30 to NMR Analyses. NMR spectra were recorded on a Bruker Avance
600 °C at a heating rate of 10 °C/min. An air atmosphere was 300 MHz spectrometer at 25 °C using DMSO-d6 as the solvent under
maintained at 60 mL/min flow rate. conditions optimized and reported earlier.19,21 Chemical shifts were
Isolation of Milled Wood Lignin. A modified MWL isolation referenced to TMS (0.0 ppm). Heteronuclear single-quantum coher-
procedure18 was used to isolate lignin from E. globulus wood. The same ence (HSQC) analysis was performed with a 5% solution of lignin by
procedure was employed in the isolation of lignin from wood after steam applying a 90° pulse width, a 512 ms acquisition time, a 1.2 s pulse delay,
explosion and enzymatic treatments. The detailed procedure followed in and 1JCH of 147 Hz. For the quantitative 13C NMR, the concentration
Figure 1. Schematic block diagram of milled wood lignin isolation indicating main analyses performed such as Klason lignin, TGA, 2D and 13C NMR
(ovals), and samples with respective nomenclature (squares).
of lignin was ∼20%; a 90° pulse width, 1.4 s acquisition time, and 1.7 s weight and/or crystallinity, the lower the temperature for poly-
relaxation delay were used. Chromium(III) acetylacetonate (0.01 M) mer degradation.24 Therefore, the observed increase in lignin
was added to the lignin solutions to provide complete relaxation of all degradation temperature suggests possible condensation reac-
nuclei. A total of 20000 scans were collected. A quantitative 1H NMR tions during steam explosion. Furthermore, a shoulder around
spectrum of acetylated MWL was recorded at a lignin concentration 280 °C was observed in the control sample (untreated wood).
of ∼20% in DMSO, with a 90° pulse width and a 1.3 s acquisition time. This shoulder can be attributed to the degradation of xylans,
which are the main hemicelluloses in the hardwood considered in
’ RESULTS AND DISCUSSION this study25 but are extensively reduced upon steam explosion.13
This observation was corroborated by carbohydrate analyses of
Thermal Properties. Themogravimetric analyses were used related samples that showed a progressive reduction of xylans in
to characterize the different wood chip samples. The analysis was the exploded samples (see Table 2). A reduction of the amount
based on the fact that each of the fiber cell wall polymers has a of lignin was also observed upon steam explosion treatment.
distinctive degradation temperature and rate of energy release The degradation temperatures attributed to polysaccharides
upon thermal breaking and combustion. During TGA the and lignin after the different steam explosion treatments are
lignocellulosic samples were kept under air atmosphere, and shown in Figure 2B. It can be observed that after steam explosion
two main temperature ranges of degradation were observed, the degradation temperature corresponding to polysaccharides
250350 and 400500 °C. These are attributed to degradation remained approximately constant, at about 333 °C, whereas the
of polysaccharides and lignin, respectively.22 Relative to cellulose temperature for lignin degradation increased with the severity of
and hemicelluloses, which are aliphatic structures, the higher the treatment. For instance, compared with steam explosion in a
degradation temperature for lignin is ascribed to its aromatic single cycle, two cycles raised the lignin degradation temperature
structure. For hemicelluloses and cellulose two consecutive (see data for SE1SE3, SE10 SE30 , SE2SE4, or SE20 SE40
processes have been determined during biomass combustion. pairs). Except for sample pair SE3SE4, the same observation
The first one corresponds to a thermal decomposition (volatili- applies when the residence time during steam treatment was
zation) that produces volatiles and char. The second reaction increased (see data for SE1SE2, SE10 SE20 , and SE30 SE40
involves char oxidation in the presence of oxygen. Volatilization pairs). The higher TGA degradation temperature observed with
and oxidation of lignin-derived char are considered to be con- the number of cycles or with the residence time is hypothesized to be
current processes.23 the result of lignin condensation. It is also possible that the strong
TGA spectra for samples before (control) and after the most conditions used during steam explosion may promote removal of
intense steam explosion treatment (SE40 ) were analyzed, and the carbohydrates linked to lignin (lignincarbohydrate complexes,
first derivate of the thermograms is shown in Figure 2A. Shifts in LCC), as has been reported by others.13 This hypothesis is supported
degradation temperature were taken as indirect indication of by the decrease in intensity of the NMR signal at 81.25/5.1
collective changes that occurred in wood components during ppm assigned to benzyl ether linkages between lignin and xylan.26
steam explosion. After steam explosion, an increase by 20 °C in Laccase treatment of chips after steam explosion under the
the second degradation temperature (at about 490 °C), corre- most severe conditions (sample SE40 +laccase) changed further
sponding to lignin, was observed, whereas the degradation the respective degradation temperatures of the polysaccharides
temperature for polysaccharides (at about 330 °C) remained and lignin, as observed in Figure 3. In fact, a reduction in the
approximately constant. Generally, the lower the molecular degradation temperature by 19 and 31 °C for lignin and
8763 dx.doi.org/10.1021/jf201605f |J. Agric. Food Chem. 2011, 59, 8761–8769
Journal of Agricultural and Food Chemistry ARTICLE
Figure 2. (A) Thermal properties of steam-exploded samples: first derivate of thermograms of Eucalyptus globulus wood before (control sample, black
profile) and after steam explosion under conditions denoted S40 in Table 1 (gray profile). (B) Lignin and polysaccharide degradation temperatures for
the control and steam-exploded samples.
Figure 5. HSQC spectrum of isolated lignin from Eucalyptus globulus after steam explosion treatment (SE40 ) (a). The detailed oxygenated aliphatic
region is included in panel b.
dihydroferulic acid type units (ArCH2CH2COOH), were amounts of β-O-40 and β-β0 structures decreased after steam
identified on the basis of the reported chemical shifts. Weak explosion treatment (Table 4).
signals of dihydrocinnamyl alcohol type units (ArCH2 The cross-peak at δC/ δH 83.5/5.28 was assigned to β-CH in
CH2CH2OH) (I) were detected in MWL from control β-O-40 moieties of S-type with R-carbonyl groups according to
E. globulus wood, but not in lignin isolated from E. globulus after Capanema et al.21 The signals of the γ-CH in coniferyl alcohol
steam explosion and laccase treatments. type units were observed at 61.9/4.12 ppm. These signals were
In the oxygenated aliphatic region, the signals of the major observed in the three isolated lignins.
structures of native lignin such as (β-O-40 ) (A), phenylcoumaran Weak signals of spirodienone structures (F) were observed in
(β-50 ) (D), and pino/syringyresinol (β-β0 ) (E) were observed in the HSQC spectrum of the MWL from control E. globulus wood.
the three studied lignins (Figures 4, 5, and 6), although the The signals at 79.5/4.12 and 81.2/5.1 ppm were assigned to
8766 dx.doi.org/10.1021/jf201605f |J. Agric. Food Chem. 2011, 59, 8761–8769
Journal of Agricultural and Food Chemistry ARTICLE
Figure 6. HSQC spectrum of isolated lignin from Eucalyptus globulus after steam explosion and laccase-mediator treatment (SE40 +laccase sample)
(a). The detailed oxygenated aliphatic region is included in panel b.
β0 -CH in the second side chain in the spirodienone and R-CH In the aromatic region, a predominance of syringyl (S) over
in spirodienone plus lignin xylan moieties, respectively.21,26 guaiacyl (G) units was observed in the three lignins studied,
These signals were very weak or even disappeared in lignin which is in concordance with the high proportion of S units in
isolated from wood after steam explosion and steam explosion hardwood.
plus laccase treatment. Comparison of 2D NMR spectra before and after steam
The oxygenated aliphatic region also contained cross-signals explosion followed or not by laccase-mediator treatment
for xylans and other polysaccharides (δC/ δH 6375/3.23.8) made it evident that lignin preparation isolated from SE40 and
less abundant in MWL from control E. globulus wood than in the SE40 +laccase had higher sugar content than lignin isolated from
other lignins analyzed. Lignin carbohydrate linkages were re- original wood at the same yield, as observed in the oxygenated
vealed by cross-signals for some β-O-40 substructures that are aliphatic region (Figures 4, 5b, and 6b). Additionally, 2D NMR
R-C or γ-C-etherified with carbohydrates (J).31 spectra qualitatively indicated a very similar observation. Similar
8767 dx.doi.org/10.1021/jf201605f |J. Agric. Food Chem. 2011, 59, 8761–8769
Journal of Agricultural and Food Chemistry ARTICLE
Table 4. 13C NMR Data of the Most Important Lignin condensed lignin structures. However, it should be noted that
Functionalities (per 100 Ar Units) milled wood samples and not isolated lignin was used in TGA
measurements. Therefore, indirect and collective effects of lignin
isolated lignin COOR OHpra OHseca OHph β-O-4 OMe S/G DC
and other components are factored in the thermal analyses
original (control) 5 69 57 23 64 177 3.2 13 (structural changes). These results indicate a decrease in lignin
SE40 16 66 54 47 36 170 3.1 14 molecular weight during steam explosion followed by laccase-
SE40 +laccase 17 63 70 53 36 161 3.0 19 mediator treatment. This is also supported by the observed
a
Includes sugar OH groups. higher extractives yield and ASL yield (Table 3) compared to the
control sample; also, the lignin isolated from steam-exploded
samples had a more condensed structure.
results were found by Balakshin et al.30 and Ibarra et al.,31 who The amount of methoxyl groups (OMe) was reduced from
studied the residual lignin after LMS-E treatment and found the 1.77/Ar to 1.70/Ar after steam explosion treatment and from
HSQC NMR spectra very similar to those of the original residual 1.70/Ar to 1.61/Ar after a subsequent laccase treatment. A
kraft lignin. They ascribed their observations to a removal of the similar reduction in methoxyl groups has been observed by Chua
altered lignin from the pulp during the alkaline extraction, and Wayman.29
subsequent to the LMS treatment; the residual lignin that A slight reduction in S/G ratio was observed after steam
remained in the pulp did not undergo substantial changes. In explosion treatment and also after laccase-mediator treatment.
our case, the oxidized lignin could be removed during the alkali Chua and Wayman29 explained this reduction in S/G ratio
extraction performed in the lignin isolation procedure used to during mild steam explosion condition as being due to an initial
remove tannins. preferential removal of syringyl to guaiacyl units at short auto-
A quantitative study based on 13C NMR was carried out. The hydrolysis times. The decrase in S/G ratio after laccase-mediator
quantification of the most important lignin functionalities is treatment was also observed by Ibarra et al.31
indicated in Table 4. The calculations were performed according
to the method suggested by Capanema et al.21 ’ AUTHOR INFORMATION
The amount of aliphatic and conjugated COOH groups
estimated from the spectrum of the nonacetylated MWL from Corresponding Author
control E. globulus was found similar to that previously *E-mail: [email protected]. Phone: (919) 513-7464. Fax: (919)
reported.32 However, when wood was subjected to steam explo- 515-6302.
sion, the amount of carboxyl groups increased significantly. Funding Sources
The amount of primary hydroxyl groups observed in SE40 and Financial support from the Spanish Ministry of Science and Innova-
SE40 +laccase lignin was slightly lower than in the lignin isolated tion (FPI fellowship BES2007-14528) is gratefully acknowledged.
from original milled E. globulus wood. However, the amount of
secondary OH increased significantly after the laccase treatment.
In addition, the quantity of phenolic OH increased significantly ’ ACKNOWLEDGMENT
after the steam explosion treatment, which agrees with results We thank Professor M. J. Diaz of Universidad de Huelva
found by other authors.27,33 (Spain) for assistance with HPLC carbohydrate analyses. We
The total amount of β-O-4 structures was significantly re- acknowledge the contributions of Prof. F. Somolinos of Universidad
duced after steam explosion treatment, which was in agreement Complutense de Madrid (Spain).
with other papers.6,7,13 It is well-known that steam explosion
carried out only in the presence of steam is an autohydrolytic ’ ABBREVIATIONS USED
process catalyzed by the organic acids formed from the acetylated
wood components. The predominant reactions in lignin are MWL, milled wood lignin; NMR, nuclear magnetic resonance;
fragmentation by acidolysis of β-O-4 linkages and polymeriza- SEC, size exclusion chromatography; TGA, thermogravimetric
tion by acid-catalyzed condensation between the aromatic C6 or analyses; SE, steam exploded; LMS, laccase-mediator system; HBT,
C5 and a carbonium ion, normally located at R-C of the side 1-hydroxybenzotriazole; ABTS, 2,20 -azinobis(3-ethylbenzthiazoline-
chain.13 As a result of the substantial cleavage of β-O-4 linkages, 6-sulfonate); HPLC, high-performance liquid chromatography;
the molecular weight of lignin would be expected to decrease DMSO, dimethyl sulfoxide; HSQC, heteronuclear single-quantum
sharply as a result of the steam explosion treatment. However, Li coherence; LMS-E, laccase-mediator system treatment followed by
et al.13 observed a gradual increase in molecular weight with alkaline extraction.
increasing severity of the treatment. Thus, they postulated that
the cleavage of β-O-4 linkages in lignin during the steam ’ REFERENCES
explosion process was accompanied by a comprehensive repoly- (1) Zimbardi, F.; Ricci, E.; Braccio, G. Techno-economic study on
merization, resulting in an increase in molecular size and a more steam explosion application in biomass processing. Appl. Biochem.
heterogeneous lignin structure. In the present study, the steam Biotechnol. 2002, 98, 89–99.
explosion conditions were milder than those reported before. (2) Puls, J.; Poutanen, K.; K€orner, H. U.; Viikari, L. Biotechnical
The repolymerization reaction may not be dominant, and no utilization of wood carbohydrates after steaming pretreatment. Appl.
Microbiol. Biotechnol. 1985, 22, 416–423.
increase in the degree of condensation (DC) after steam explo-
(3) Ramos, L. P.; Breuil, C.; Kushner, D. J.; Saddler, J. N. Steam
sion treatment was observed. pretreatment conditions for effective enzymatic hydrolysis and recovery
After the laccase-mediator treatment, no variation in the total yields of Eucalyptus viminalis wood chips. Holzforschung 1992, 46 (2),
amount of β-O-4 structures was observed, but an increase in 149–154.
the DC was found. In contrast, TGA results indicated a decrease (4) Heitz, M.; Capek-Menard, E.; Koeberle, P. D.; Gagne, J.;
in lignin degradation temperature, which correlates with less Chornet, E.; Overend, R. P.; Taylor, J. D.; Yu, E. Fractionation of
Populus tremuloides at the pilot plant scale: optimization of steam (26) Balakshin, M.; Capanema, E.; Gracz, H.; Chang, H-m.; Jameel,
pretreatment conditions using the Stake II technology. Bioresour. H. Quantification of lignincarbohydrate linkages with high resolution
Technol. 1991, 35, 23–33. NMR spectoscropy. Planta 2011, 233, 1097–1110.
(5) Montane, D.; Salvado, J.; Farriol, X. Fractionation of wheat straw via (27) Robert, D.; Gellerstedt, G.; Bardet, M. Carbon-13 NMR
steam explosion pretreatment. Characteristics of the lignin obtained by analysis of lignins obtained after sulfonation of steam exploded aspen
alkali delignification of the steamed straw. Holzforschung 1997, 51, 135–141. wood. Nordic Pulp Pap. Res. J. 1986, 1, 18–25.
(6) Robert, D.; Bardet, M.; Lapierre, D.; Gellerstedt, G. Structural (28) Lora, J. H.; Wayman, M. Delignification of hardwoods by
changes in aspen lignin during steam explosion treatment. Cell. Chem. autohydrolysis and extraction. Tappi 1978, 61 (6), 47–50.
Technol. 1988, 22, 221–230. (29) Chua, M. G. S.; Wayman, M. Characterization of autohydrolysis
(7) Jakobsons, J.; Hortling, B.; Erins, P.; Sundquist, J. Characteriza- aspen (P. tremuloides) lignin. Part 1. Composition and molecular weight
tion of alkali soluble fraction of steam exploded birch wood. Holz- distribution of extracted autohydrolysis lignin. Can. J. Chem. 1978,
forschung 1995, 49, 51–59. 57, 1141–1149.
(8) Josefsson, T.; Lennholm, H.; Gellerstedt, G. Changes in cellulose (30) Balakshin, M. Yu.; Capanema, E. A.; Chen, C.-L.; Gratzl, J. S.;
supramolecular structure and molecular weight distribution during Kirkman, A. G.; Gracz, H. S. Biobleaching of pulp with dioxygen in the
steam explosion of aspen wood. Cellulose 2001, 8, 289–296. laccase-mediator system reaction mechanisms for degradation of
(9) Josefsson, T.; Lennholm, H.; Gellerstedt, G. Steam explosion of residual lignin. J. Mol. Catal. B: Enzym. 2001, 13, 1–16.
aspen wood. Characterization of reaction products. Holzforschung 2002, (31) Ibarra, D.; Chavez, M. I.; Rencoret, J.; del Rio, J. C.; Gutierrez,
53 (3), 289–297. A.; Romero, J.; Camarero, S.; Martínez, M. J.; Jimenez-Barbero, J.;
(10) Lawoko, M.; Henriksson, G.; Gellerstedt, G. Characterisation Martinez, A. T. Structural modification of eucalypt pulp lignin in a totally
of lignincarbohydrate complexes (LCCs) of spruce wood (Picea abies L.) chlorine-free bleaching sequence including a laccase-mediator stage.
isolated with two methods. Holzforschung 2006, 60, 156–161. Holzforschung 2007, 61, 634–646.
(11) Balakshin, M. Yu.; Capanema, E. A.; Chang, H.-m. Recent (32) Evtuguin, D. V.; Pascoal Neto, C.; Silva, A. M. S.; Dominges,
advances in isolation and analysis of lignins and lignin-carbohydrate P. M.; Amado, F. M. L.; Robert, D.; Faix, O. Comprehensive study on the
complexes. In Characterization of Lignocellulosics; Hu, T., Ed.; Blackwell: chemical structure of dioxane lignin from plantation Eucalyptus globulus
Oxford, U.K., 2008; pp 148166. lignin. J. Agric. Food Chem. 2001, 49, 4252–4261.
(12) Bj€orkman, A. Studies on finely divided wood. Part I. Extraction (33) Shimizu, K.; Sudo, K.; Ono, H.; Ishihara, M.; Fujii, T.;
of lignin with neutral solvents. Svensk Papperstidn. 1956, 59, 477–485. Hishiyama, S. Integrated process for total utilization of wood compo-
(13) Li, J.; Henriksson, G.; Gellerstedt, G. Lignin depolymerization/ nents by steam-explosion pretreatment. Biomass Bioenerg. 1998, 14 (3),
repolymerization and its critical role for delignification of aspen wood by 195–203.
steam explosion. Bioresour. Technol. 2007, 98, 3061–3068.
(14) Chandel, A. K.; Kapoor, R. K.; Singh, A.; Kuhad, R. C.
Detoxification of sugarcane bagasse hydrolysate improves ethanol
production by Candida shehatae NCIM 3501. Bioresour. Technol. 2007,
98, 1947–1950.
(15) Jurado, M.; Prieto, A.; Martínez-Alcala, A .; Martinez, A. T.;
Martínez, M. J. Laccase detoxification of steam-exploded wheat straw for
second generation bioethanol. Bioresour. Technol. 2009, 100 (24),
6378–6384.
(16) Martín-Sampedro, R.; Eugenio, M.; Carbajo, J.; Villar, J.
Combination of steam explosion and laccase-mediator treatments prior
to Eucalyptus globulus Kraft pulping. Bioresour. Technol. 2011, 102 (14),
7183–7189.
(17) Overend, R. P.; Chornet, E. Fractionation of lignocellulosics by
steamaqueous pretreatments. Philos. Trans. R. Soc. London 1987,
A321, 523–536.
(18) Capanema, E. A.; Balakshin, M. Yu.; Kadla, J. F.; Chang, H.-m.
On isolation of milled wood lignin from eucalyptus wood. O Papel 2007,
No. (N5), 74–79.
(19) Capanema, E. A.; Balakshin, M. Yu.; Kadla, J. F. A comprehen-
sive approach for quantitative lignin characterization by NMR spectros-
copy. J. Agric. Food Chem. 2004, 52, 1850–1860.
(20) Garrote, G.; Dominguez, H.; Parajo, J. C. Mild autohydrolysis:
an environmentally friendly technology for xylooligosaccharide produc-
tion from wood. J. Chem. Technol. Biotechnol. 1999, 74, 1101–1109.
(21) Capanema, E. A.; Balakshin, M. Yu.; Kadla, J. F. Quantitative
characterization of a hardwood milled wood lignin by NMR spectros-
copy. J. Agric. Food Chem. 2005, 53, 9639–9649.
(22) Rovira, P.; Vallejo, V. R. Examination of thermal and acid
hydrolysis procedures in characterization of organic matter. Commun.
Soil Sci. Plant Anal. 2000, 31 (12), 81–100.
(23) Barneto, A. G.; Ariza, K.; Martin, J. E.; Jimenez, L. Use of
autocatalytic kinetics to obtain composition of lignocellulosic materials.
Bioresour. Technol. 2009, 100, 3963–3973.
(24) Turi, E. A. Thermal Characterization of Polymeric Materials;
Academic Press: New York, 1981.
(25) Reh, U.; Kraepelin, G.; Lamprecht, I. Use of differential
scanning calorimetry for structural analysis of fungally degraded wood.
Appl. Environ. Microbiol. 1986, 52 (5), 1101–1106.
Martín-Sampedro, R., Eugenio, M.E., García, J.C., Lopez, F., Villar, J.C., Diaz, M.J.R.
Manuscrito
199
Steam explosion and enzymatic as pre-treatments to improve enzymatic
hydrolysis of Eucalyptus globulus
a
Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria INIA.
Carretera de la Coruña, km 7.5, 28040, Madrid (Spain)
b
Departamento de Ingeniería Química. Universidad de Huelva. Campus el Carmen,
21071, Huelva (Spain)
Abstract
Keywords
201
1. Introduction
The main advantages of the steam explosion process are a significantly lower
environmental impact, lower capital investment and less hazardous chemicals released
in the process [10] in comparison with other pre-treatments. However, the degradation
of sugars and lignin during steam explosion is unavoidable, releasing degradation
products which can negatively impact further steps of the process [3,11,12]. To remove
these inhibiting products, several treatments have been assayed, including physical,
chemical and biological methods [13,14]. The application of the enzyme laccase is one
of the new biological methods used to clear inhibitors, as these enzymes are capable of
removing specifically phenolic compounds present in the biomass [15-18]. Moreover,
as the solid fraction resulting from the steam explosion process (exploded chips) shows
a more open structure compared with non-exploded chips [19], laccases have greater
accessibility to the steam exploded chips and the efficacy of the process increases [3,
16, 20-23].
202
Additionally, xylanases are known to enhance the effect of lacasses in some
applications. Thus, Valls and Roncero [24] have observed that, when a xylanase
treatment precedes the laccase treatment in the bleaching of crude eucalyptus kraft
pulps, the amount of residual lignin and hexenuronic acids is smaller than when pulps
are bleached using a laccase step alone.
Whereas a few studies have reported on the use of steam explosion together with
a laccase treatment to improve ethanol production from lignocellulosic materials, none
of them has evaluated the use of a xylanase to boost the laccase effect. Thus, the aim of
our study was not only to evaluate the different fractions obtained after applying steam
explosion and a laccase treatment to the production of ethanol, but also to assess the
contribution to the whole process of adding a xylanase treatment prior to the laccase
step. To that end, an acid or enzymatic hydrolysis was assayed over all fractions
obtained in the different steps. Sugars content before and after the hydrolysis, i.e. the
hydrolysis yield, was determined.
All chemicals used were reagent-grade and were acquired from Merck
(Barcelona, Spain), Panreac (Barcelona, Spain) or Sigma–Aldrich (Madrid, Spain),
except for 2,2´-Azino-bis-3-ethylbenzthiazoline-6-sulphonate (ABTS), which was
purchased from Roche (Madrid, Spain).
203
treatment, the pressure was reduced to 6 kg cm-2 and the chips were discharged into the
blowing tank at atmospheric pressure. The steam explosion variables were the number
of cycles and the duration of the first cycle (the second cycle had a fixed duration of 3
minutes). Addition of a pre-extraction step (immersion of the chips in water for 16
hours at 25ºC) prior to steam explosion was also evaluated in order to study the effect of
water impregnation. The operational conditions applied were based on previous work
[4,19] and are summarized in Table 1.
The severity factor of each treatment was calculated according to the following
equation defined by Overend and Chornet [25].
T 100
S0 log e 14.75 · t (1)
where T is the temperature (ºC) and t the duration of the treatment (min).
2.3. Enzymatic hydrolysis of solid fractions obtained after the several steam explosion
treatments
The solid fractions resulting from all steam explosion pre-treatments (Table 1)
were ground in a Wiley mill to pass through a 1 mm screen. Wood meals obtained were
then subjected to enzymatic hydrolysis by action of a cellulolytic complex (Celluclast
1.5L), supplemented with β-glucosidase (Novozym 188) at a 1:3 ratio [11]. Both
enzymatic mixtures were kindly provided by Novozymes (Bagsvaerd, Denmark). The
celluclast 1.5 L dose was 20 EGU g-1 of substrate (at 40ºC and pH 6.0). The enzymatic
hydrolysis treatments were carried out at pH 4.8 (50 mmol L-1 citrate – phosphate
buffer) and 50ºC (according to the specifications listed in the product sheet) in a rotary
shaker (150 revolution per minute) for 72 h with 5 % concentration of the pre-treated
material. After enzymatic hydrolysis, samples were heated in boiling water for 5 min
and, after cooling, centrifuged to separate the hydrolysate and the solid residues. The
yield of the treatment and the sugars content (as per the method described in section
2.6.) of all liquid and solid fractions were determined. The steam explosion pre-
treatment which proved to be most efficient was selected for further assays (see Figure
1).
204
2.4. Laccase and xylanase pre-treatments of the solid fraction obtained from the
selected steam explosion pre-treatment
The solid fraction obtained from the selected steam explosion pre-treatment was
subjected to the xylanase and/or laccase treatment before the enzymatic hydrolysis.
These enzymatic treatments were carried out in sequence or separately (Figure 1). The
enzymes used were Pulpzyme HC (xylanase) and Novozym 51003 (laccase) produced
by submerged fermentation of genetically modified Bacillus sp. and Aspergillus sp.,
respectively. Both enzymes were donated by Novozymes (Bagsvaerd, Denmark).
After the enzymatic treatments, the solid fractions were isolated and subjected to
an enzymatic hydrolysis according to the method described in section 2.3.
2.5. Hydrolysis of the liquid fraction obtained from the selected steam explosion
treatment
The liquid fractions obtained from the selected steam explosion pre-treatment
were subjected to both acid and enzymatic hydrolysis. Acid hydrolysis was carried out
with 2 % H2SO4 at 115ºC for 60 min in an autoclave, based on Garrote et al. [26], to
avoid degradation of the monomers. Enzymatic hydrolysis was performed at pH 6.3 and
50ºC in a rotary shaker (150 revolutions per minute) for 72 h. According to Um and van
Walsum [27], a mixture of commercial enzyme preparations increases the hydrolysis
rate of xylan-rich extracts; therefore a combination of cellulolytic complex (Celluclast
1.5L) and xylanase (Pulpzyme HP) was used at a 3:1 ratio. The Pulpzyme dose was 150
AXU ml-1 of sample. After the enzymatic hydrolysis, samples were heated in boiling
water for 5 min and then cooled at room temperature.
205
Finally, the sugar contents of the hydrolysates obtained from both hydrolysis
processes were determined as per the method described in section 2.6.
2.6. Analysis
The composition of the raw material (EG) and all solid fractions was determined
by standard analytical methods. The extractives were determined as the soluble material
after a Soxhlet extraction with ethanol for 6 hours. The free-extractive samples were
subjected to quantitative acid hydrolysis to determine the carbohydrate composition.
This acid hydrolysis was performed in two steps according to Garrote et al. [28]. The
hydrolysed liquid obtained was then analysed for sugars and acetyl group contents by
High Performance Liquid Chromatography (HPLC) using the Agilent Technology 1200
series RID with an Aminex HPX-87H column operated at 50ºC with a mobile phase
containing 5 mmol L-1 sulfuric acid pumped at a rate of 0.6 ml min-1. The solid residue
remaining after the acid hydrolysis is considered acid insoluble lignin (Klason lignin).
Additionally, the acid soluble lignin was quantified using UV spectrophotometry at 205
nm.
All treatments were run twice and samples for chemical analysis were
triplicated. Tukey's multiple comparison test (α = 0.05) was used to determine which
means, among a set of means, differ from the rest (Matlab 5.1).
206
3. RESULTS AND DISCUSSIÓN
The effects of steam explosion on the obtained solid fractions are indicated by
the data in Table 2. The pre-extraction of EG did not affect significantly the chemical
composition of the chips (EG and SE0 samples), apart from a reduction in the soluble
lignin content. However, significant differences between the EG and the rest of the
steam exploded samples (SE1-SE8) were found for all evaluated parameters except for
acetyl groups (with the exception of SE4).
Steam explosion removes hemicelluloses and also some lignin depending on the
extent of the treatment. This has been previously described by Excoffier et al. [5] who
have found that, whereas the hemicellulose is removed by hydrolysis, lignin softens
under the heat and slowly depolymerizes. In general, the pre-extraction did not affect
the solid content under lower severity factors (SE1-SE2 or SE5-SE6). However, under
the more severe treatments, significant differences were observed (SE3-SE4, SE7-SE8)
mainly in yield and extractable values. The greatest differences among treatments were
observed for SE3 and SE4. These treatments have a medium-high severity factor and a
single cycle, though the duration of this cycle is long. These data suggest that cycle
duration is a major factor that could influence the extraction of the EG components.
207
solubilization after steam explosion: xylans and arabinans remaining in the solid
fraction decreased down to 53.4 % and 6.6 %, respectively, in SE8. Similar findings
have been reported by Ruiz et al. [3] who have described a complete solubilization of
arabinan and a solubilization of a fraction of 50 % of the original xylan after subjecting
sunflower stalks to steam explosion at 190ºC. Acetyl groups, bound to the xylan in the
native wood, were also removed during medium and severe steam explosion treatments
(SE4-SE8). Other authors have observed also a similar removal during steam explosion
and autohydrolysis of hardwood [6]. Acetyl groups can be removed from the wood by
means of two different reaction pathways: direct hydrolysis of acetyl groups from the
solid xylan backbone and dissolution of acetylated xylo-oligosaccharides [30]. Lignin is
partially dissolved also whereas glucan stays unchanged. Other authors have reported a
slight solubilization of glucan, but mostly after more severe treatments than those
reported here [3,7,8].
3.2. Effects of steam explosion on the enzymatic hydrolysis of the solid fractions.
After the steam explosion treatments, the solid fractions were subjected to
enzymatic hydrolysis. The chemical characterization of the solid and liquid fractions
obtained after these treatments is shown in Tables 3 and 4, respectively. A decrease in
the extractives and sugar contents of the solid fraction were evident, being higher for
exploded samples that had been exposed to more intensive steam explosion. Enzymatic
hydrolysis is known to accelerate the extraction of soluble and relatively easily
biodegradable organic products [31]. This solubilization and hydrolysis caused an
increase in the relative content of Klason lignin in all samples after enzymatic
hydrolysis.
208
removal of hemicelluloses which increase the accessibility to cellulose [20-22].
Additionally, it is well recognized in the literature that the lignin content has an impact
on the enzymatic hydrolysis [32,33]. Nevertheless, the differences observed in the
hydrolysis yield could be due to the severity of the steam explosion pre-treatment but no
to variations in the lignin content, since differences in lignin content were barely
significant.
The highest hydrolysis yields were obtained in samples subjected to the longest
steam explosion treatments after water pre-extraction (24.1 % and 24.7 % glucose yields
and 35.8 % and 35.6 % total sugar yields for SE4 and SE8 respectively). Other authors
have reported slightly higher hydrolysis yields after steam explosions whose severity
was comparable to those performed in this study. Thus, glucose yields of 31 %, 32 %
and 35 % have been reported by Varga et al. [23] with corn stover (steam explosion at
190ºC, 5 min, severity factor of 3.35), Ruiz et al. [3] with sunflower stalk (180-190ºC, 5
min, S0 = 3.35-3.64) and Jurado et al. [16] with wheat straw (190ºC, 10 min, S0 = 3.65),
respectively. However, these authors have used non-wood materials, which have a more
porous and readily accessible tissue structure [34] that can facilitate the enzymatic
treatment. Reports of higher hydrolysis yields for eucalyptus consistently use higher
severity factors and/or the addition of a catalytic acid [11,35]. However, if the severity
or acidity of the pre-treatment were too high, greater degradation of carbohydrates
during steam explosion would result in less glucose recovery and more inhibitors
present in both the solid and liquid fractions [3,11,12]. In our case, the overall yield of
glucose and total sugars with respect to the original raw material (EG) were 23.7 % and
30.2 % respectively, with the most severe pre-treatment (SE8). Furthermore, the real
total sugar yield was higher than this value because sugar recovery in the liquid fraction
after steam explosion should be taken into account as well (see section 3.4).
209
3.3. Effects of the laccase and xylanase pre-treatments on the enzymatic hydrolysis of
the exploded solid fractions.
SE8 was subjected to xylanase and/or laccase treatments in order to study their
effect on the enzymatic hydrolysis performed subsequently. The chemical composition
of the resulting samples is shown in Table 5. Comparing the chemical composition of
the samples after the laccase treatment (SE8-L) with that of SE8, a decrease in ethanol
extractives was observed. This finding could be related with the removal of lipophilic
extractives by the laccase, which has been reported by several authors using laccase to
combat pitch problems in pulp and paper production [38-40]. Also, a decrease in xylan
and acetyl groups was found after the enzymatic treatments. Consequently, an increase
in the relative percentage of glucan and Klason lignin was observed. The removal of
these products was likely the result of partial degradation during steam explosion and
the extractive effect of the subsequent enzymatic treatment at high temperature (70ºC).
Although the LMS treatment has been combined with an alkaline extraction stage to
efficiently delignify pulps and lignocellulosic materials [41,42], no reduction in Klason
lignin was observed here. This result might indicate that LMS did oxidize the lignin, but
the resulting products were not soluble at the pH of treatment (pH 6), and a subsequent
alkaline treatment would be needed to remove the oxidized products [40]. After the
xylanase treatment (SE8-X) reduction of the xylan content was greater and reduction of
acetyl groups was smaller. However, after sequential xylanase and laccase treatments no
significant differences in chemical composition were found with respect to the laccase
treatment, apart from a higher content of acetyl groups.
After the enzymatic hydrolysis of the treated samples, a reduction in both glucan
and xylan content was observed (Table 5), secondary to the degradation of cellulose and
hemicelluloses. Thus, glucose and xylose monomers were the main components of the
hydrolysates, as Table 6 shows.
Figure 4 shows the glucose and total sugar yield of the enzymatic hydrolysis. As
it can be observed, the LMS treatment increased the glucose yield compared to SE8
(27.1 % vs 24.7 %), although the total sugar yield was slightly smaller. An increase in
the hydrolysis yield has been made evident also by other authors after laccase and white
rot fungus treatments [43,44]. This can be a result of the modification of lignin by the
laccase, probably decreasing the hydrophobicity of lignin and eventually changing the
210
polarity characteristics of the surface, which in turn may have decreased the irreversible,
unproductive binding of cellulases to lignin [17]. However, although Jurado et al. [16]
have reported no effect of LMS on the hydrolysis rate, these authors have found an
increase in ethanol production in a subsequent fermentation, owing to the removal of
inhibiting products (phenolic compounds) during the LMS treatment of steam exploded
wheat straw.
The xylanase treatment also increased the glucose yield (28.3 %), although this
treatment did not enhance the effect of the LMS treatment in the enzymatic hydrolysis
(SE8-XL). However, lack of improvement of the hydrolysis does not mean lack of
beneficial effects on a subsequent fermentation process, as mentioned above [16].
Therefore, a fermentation process should be carried out in order to clarify the effect of
the laccase and the xylanase-laccase treatments on bioethanol production.
On the other hand, the beneficial effect of the xylanase treatment on the glucose
yield could be related to greater accessibility gained by the removal of xylan, as it could
be observed also after the steam explosion treatments. Nevertheless, the overall glucose
yield of the process, taking into account both the steam explosion and the xylanase
treatment yields, was found higher than that of the SE8 sample (27.8 % vs 23.7 %),
while the overall total sugar yield was slightly lower (29.6 % vs 30.2 %).
3.4. Hydrolysis of the liquid fraction obtained after the steam explosion treatment
selected as most effective
Two liquid fractions were isolated from the SE 8 treatment, one in each cycle.
The sugar content of these fractions is listed in Table 7. This table also presents the
percentage of sugar recovery, determined from the sugar concentrations in the liquid
and the potential sugar concentration of the original raw material (EG). It should be
noted that, in the first SE cycle, 8.2 % of xylose and 4.0 % of glucose were recovered in
the liquid fraction as oligosaccharides. In the second cycle, a higher amount of
xylooligosaccharides was recovered, even when this cycle was shorter than the first one
(3 min vs 10 min). This result can indicate that, during the first steam explosion
treatment, some carbohydrates were degraded partially but not completely solubilized in
the liquid fraction, requiring a longer treatment or a second cycle to be completely
solubilized. Furthermore, the decompression in the first cycle resulted in physical
211
rupture of the fiber walls and chip structure, which is likely to have enhanced the
subsequent extraction of hemicelluloses in the second cycle. The total xylose recovery
in the two liquid fractions was smaller than that estimated based on the extraction of
hemicelluloses in the solid fraction (25.3 % vs 46.6 %). The reason could be an
inefficient recovery of the extracted hemicelluloses since part of them could be
dissolved in the liquid that impregnates the chips after the explosion and is discarded
after washing the exploded chips. Another reason could be the degradation of the
dissolved xylooligosaccharides to furfural, which has been reported by other authors
during steam explosion with high severity factor [7,11]. However, in the liquid fractions
obtained in our study, neither furfural nor hydroxymethyl furfural (HMF) was detected
in the HPLC analysis, indicating that hydrolysis reactions largely predominated over
degradation reactions. Nevertheless, it is possible that these volatile products were not
quantitatively recovered in the liquid fraction because they were partially lost to the
atmosphere after explosive decompression [35].
The xylose hydrolysis yields for both the acid and the enzymatic process were
the lowest in the liquid fraction obtained from the second cycle (SE8.2). Contrary to
this, the glucose hydrolysis yields were the lowest in SE8.1. A plausible explanation for
this can be the differences in the concentration of substrates to be hydrolyzed
(oligosaccharides). Thus, for lower initial concentrations, the hydrolysis yields were
higher. However, the total amounts converted into glucose, in SE8.1, and xylose, in
SE8.2, were higher than in SE8.2 and SE8.1 respectively. Therefore, the observed
reductions in the hydrolysis yield could be a consequence of the increased initial
amount of oligosaccharides (which is in the denominator in the yield equation).
The total hydrolysis yields from the acid hydrolysis were lower than those
obtained from the enzymatic hydrolysis: 16.5 % vs 65.3 % for glucose and 61.4 % vs
82.7 % for xylose. However, the acid hydrolysis was not optimized, and an increase in
212
reaction time, temperature and/or acid concentration could increment its yield.
Nevertheless, raising these operational conditions would also enhance degradation of
monosaccharides, increasing the amount of inhibitors such as furfural and HMF [2,26].
On the other hand, the enzymatic hydrolysis yield could be increased also by modifying
the enzyme-substrate ratio and the enzyme activity. Thus, the enzymatic hydrolysis has
the advantage of requiring less severe operational conditions (temperature, pH),
generating less inhibiting products, which would contribute to a greater bioethanol
production rate [27].
In order to optimize the overall process, several goals should be set. Recovery of
both cellulose, in the solid fraction, and hemicelluloses, in the liquid fraction, is
important in the steam explosion pre-treatment. The amount of inhibitors generated
during this pre-treatment, and the hydrolysis yields of both the solid and the liquid
fractions are also critical factors to be considered [3]. Accordingly, the overall yields of
the process were determined including all these considerations.
Taking into account that 25.3 % of the total xylan content in the original wood
(EG) was recovered as xylooligosaccharides in the liquid fractions, the enzymatic
hydrolysis rate with respect to the xylan present in EG was 21.0 %. Thus, the overall
xylose yield obtained after the enzymatic hydrolysis of both the solid and the liquid
213
fractions (section 3.2 and 3.4, respectively) in SE8 was 55.5 % of the hemicelluloses
contained in EG (8.6 % of the whole raw material). However, this yield could be
increased if recovery of extracted hemicelluloses was more efficient, as around 20 % of
the hemicelluloses were lost. On the other hand, the overall glucose yield, including the
enzymatic hydrolysis of the solid and the liquid fractions, was 26.8 % of the glucose
present in EG (12.3 % of the whole raw material); therefore, 210 g of total sugars could
be obtained from 1 kg of EG, and this yield could be increased by introducing some
modifications in the process, such as laccase detoxification, and improvements in the
recovery of the liquid fraction in SE, as it has been mentioned before. Nevertheless, a
similar overall yield (22 %) was obtained using sunflower stalks and a more severe
steam explosion pre-treatment [3], which could produce more inhibiting products and
also increase the operational cost of the pre-treatment.
4. Conclusions
Acknowledgements
The authors wish to thank the Spanish Ministry of Science and Innovation for
funding this study via the Projects CTQ 2006-00385 and CTQ 2009-11268 and the FPI
fellowship (BES2007-14528).
214
References
[2] Um BH, van Walsum GP. Acid hydrolysis of hemicellulose in green liquor pre-
pulping extract of mixed northern hardwoods. Appl Biochem Biotech
2009;153(1):127-38.
[4] Martín-Sampedro R, Eugenio ME, Revilla E, Martín JA, Villar JC. Integration of
kraft pulpin on a biorefinery by the addition of a steam explosion pretreatment.
BioResources 2011;6:513-28.
[8] Tunc MS, Lawoko M, van Heiningen A. Understanding the limitations of removal
of hemicelluloses during autohydrolysis of a mixture of southern hardwood.
BioResources 2010;5(1):356-71.
215
[10] Li J, Henriksson G, Gellerstedt G. Lignin depolymerization/repolymerization and
its critical role for delignification of aspen wood by steam explosion. Bioresour
Technol 2007;98(16):3061-8.
[12] Wu MM, Chang K, Gregg DJ, Boussaid A, Beatson RP, Saddler JN. Optimization
of steam explosion to enhance hemicellulose recovery and enzymatic hydrolysis
of cellulose in softwoods. Appl Biochem Biotechnol 1999;77(1):47-54.
[15] Chandel AK, Kapoor RK, Singh A, Kuhad RC. Detoxification of sugarcane
bagasse hydrolysate improves ethanol production by Candida shehatae NCIM
3501. Bioresour Technol 2007;98(10):1947-50.
216
[20] Ballesteros I, Negro MJ, Oliva JM, Cabañas A, Manzanares P, Ballesteros M.
Ethanol production from steam-explosion pretreated wheat straw. Appl Biochem
Biotech 2006;130(1):496-508.
[22] Yang B, Wyman CE. Effect of xylan and lignin removal by batch and
flowthrough pretreatment on the enzymatic digestibility of corn stover cellulose.
Biotechnol Bioeng 2004;86(1):88-98.
[24] Valls C, Roncero B. Using both xylanase and laccase enzymes for pulp bleaching.
Bioresour Technol 2009;100(6):2032-9.
217
[30] Maloney MT, Chapman TW, Baker AJ. Dilute acid hydrolysis of paper birch:
Kinetics studies of xylan and acetyl group hydrolysis. Biotechnol Bioeng
1985;27(3):355-61.
[31] Lee Y, Chung Y, Jung J. Effects of chemical and enzymatic treatments on the
hydrolysis of swine wastewater. Water Sci Technol 2008;58(7):1529-34.
[32] Mooney CA, Mansfield SD, Touhy MG, Saddler JN. The effect of initial pore
volume and lignin content on the enzymatic hydrolysis of softwoods. Bioresour
Technol 1998;64(2):113-9.
[33] Vinzant TB, Ehrman CI, Adney WS, Thomas SR, Himmel ME. Simultaneous
saccharification and fermentation of pretreated hardwoods. Appl Biochem
Biotechnol 1997;62(1):99-104.
[34] Alaejos J, Lopez F, Eugenio ME, Fernandez M. Non-wood raw materials for pulp
and paper making. A rewiew. Afinidad 2004;61(513):400-10.
[36] Duff SJB, Murray WD. Bioconversion of forest products industry waste
cellulosics to fuel ethanol: A review. Bioresour Technol 1996;55(1):1-33.
218
[41] Eugenio M, Santos S, Carbajo J, Martín J, Martín-Sampedro R, González A, et al.
Kraft pulp biobleaching using an extracellular enzymatic fluid produced by
Pycnoporus sanguineus. Bioresour Technol 2010;101(6):1866-70.
[43] Dias AA, Freitas GS, Marques GSM, Sampaio A, Fraga IS, Rodrigues MAM, et
al. Enzymatic saccharification of biologically pre-treated wheat straw with white-
rot fungi. Bioresour Technol 2010;101(15):6045-50.
219
Figure captions
Figure 1. Block diagram of assayes performed: First, selection of the most efficient
steam explosion pre-treatment (1); and then, study of the effect of xylanase and laccase
on the enzymatic hydrolysis of the solid fraction, and evaluation of acid and enzymatic
hydrolysis of the liquid fraction (2).
Figure 2. Percentage of each wood component remaining in the solid fraction after each
steam explosion treatment, compared to the original amount of these components in the
raw material (EG).
Figure 3. Glucose and total sugar yields during enzymatic hydrolysis of the solid
fractions obtained after steam explosion treatments.
Figure 4. Glucose and total sugar yields during enzymatic hydrolysis of the solid
fractions obtained after steam explosion and laccase/xylanase treatments.
Figure 5. Glucose and xylose yields during enzymatic hydrolysis (EH) and acid
hydrolysis (AH) of the two liquid fractions obtained from the SE8 pre-treatment (SE8.1
and SE8.2). The total glucose and xylose yield was calculated from the sum of
monomers present in the two liquid fractions.
220
Figure 1
1
Enzymatic
Solid fraction Hydrolysis rate
E. globulus Steam explosion hydrolysis Selection of steam
wood pre-treatments explosion pretreatment
Liquid fraction
2 Xylanase
Enzymatic
Solid fraction Laccase Hydrolysis rate
hydrolysis
Xylanase Laccase
E. globulus Selected steam
wood explosion
Figure 2
100
90
80
% remained in solid
70
60 lignin
Glucan
50
Xylan
40
Arabinan
30
Acetyl groups
20
10
0
EG SE0 SE1 SE2 SE3 SE4 SE5 SE6 SE7 SE8
Figure 3
40
35
Enzymatic hydrolysis yield (%)
30
25
20 Glucose
Total sugars
15
10
0
EG SE 0 SE 1 SE 2 SE 3 SE 4 SE 5 SE 6 SE 7 SE 8
221
Figure 4
40
Enzymatic hydrolysis yield (%)
35
30
25
20
15
10
0
SE 8 SE8-L SE8-X SE8-XL
Figure 5
120
100
Hydrolysis yield (%)
80
60
40
20
0
SE8.1-EH SE8.2-EH Total-EH SE8.1-AH SE8.2-AH Total-AH
Glucose Xylose
222
Table 1. Steam Explosion Conditions
Table 2. Chemical composition of the original Eucalyptus globulus (EG) and solid
fractions after steam explosion pre-treatments, expressed as g per 100 g of sample.
223
Table 3. Chemical characterization of the solid fractions from steam explosion pre-
treatments, after the enzymatic hydrolysis, expressed as g per 100 g of sample.
224
Table 5. Chemical characterization before and after the enzymatic hydrolysis of the
SE8 solid fraction treated with laccase (L) and/or xylanase (X), expressed as g per 100 g
of sample.
225
Table 7. Chemical characterization of the liquid fractions (g/l) obtained during the two
SE8 cycles of steam explosion and percentage of sugar recovered in the liquid respect to
the sugar content in the raw material (EG). [A] and [AO] indicate the monomer and
oligomer concentrations respectively.
226
PUBLICACIÓN IV:
Integration of Kraft pulping on a forest biorefiney by the addition of a steam
explosion pretreatment.
Martín-Sampedro, R., Eugenio, M.E., Revilla, E., Martín, J.A., Villar, J.C.
227
PEER-REVIEWED ARTICLE bioresources.com
Keywords: Eucalyptus globulus; Forest biorefinery; Kraft pulping; Steam explosion; Pulp properties
INTRODUCTION
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 513
PEER-REVIEWED ARTICLE bioresources.com
The steam explosion process consists of treatment of lignocellulosic material with
high-pressure steam, followed by a rapid decompression, which forces the fibrous
material to “explode” into separated fibers and fiber bundles. When no other chemicals
are added to the process, the high-temperature steam (due to the high rate of dissociation
of water) leads to the release of acids from the acetylated wood components; the released
acids then catalyze hydrolytic reactions in the wood polymers. These autohydrolysis
reactions result in a loss of hemicellulose and a decrease in the content of β-O-4
structures in lignin (Robert et al. 1988; Jakobsons et al. 1995; Josefsson et al. 2001,
2002). These chemical changes depend greatly on the temperature used and the treatment
time (Chua and Wayman 1979).
The earliest application of this process was developed by Mason in 1928. This
author proposed using a steam treatment followed by an explosive decompression for the
production of Masonite board stock (Mason 1928). Another application was suggested by
Babcock (1932), who used steam explosion as a pretreatment method for the production
of ethanol. Steam explosion has also been proposed as an alternative to improve
conventional mechanical or chemo-mechanical pulping methods (Delong 1987). During
this process, lignin is softened to allow higher defibration without excessive damage
and/or degradation to individual fibers. However, the process has not been industrially
implemented (Heitner et al. 1993) because pulps so produced have not shown any
superior properties over those from chemimechanical pulp (CMP) or chemithermo-
mechanical pulp (CTMP).
As has been mentioned before, steam explosion could be also used to remove
hemicelluloses, since during the steam treatment, hemicelluloses are depolymerized,
partially degraded, and solubilized in hot water, yielding a mixture of monomeric and
oligomeric sugars (Puls et al. 1985). Removed hemicelluloses can be a valuable source of
hexose and pentose sugars, and they can be further converted into value-added products
such as ethanol, polymers, and other chemicals (Li et al. 2010).
Removal of hemicelluloses from wood as a pretreatment step is being practiced
commercially in the production of dissolving pulps, where the hemicelluloses are
removed to allow the production of pure cellulose (Wafa Al-Dajani and Tschirner 2008).
The most common commercial procedure for extracting hemicelluloses is pre-steaming to
release natural wood acids (autohydrolysis) followed by water extraction or acid
hydrolysis with small amounts of mineral acids (sulfuric acid or hydrochloric acid).
Compared to these processes, steam explosion not only allows the recovery of wood
components without their degradation (Avellar and Glasser 1998; Britt 1970), but also
leaves a solid residue (exploded chips) that is expected to be pulped more easily than
those obtained conventionally due to the open structure of the exploded chips that might
encourage an efficient diffusion of cooking liquor into the fibers (Ahvazi et al. 2007).
Therefore, taking into account the advantages of the steam explosion pretreatment, it can
be concluded that this process could be used as an efficient pretreatment in pulping,
increasing the total revenue stream for the pulp industry and converting the pulp mill into
an integrated forest biorefinery (Ragauskas et al. 2006; Van Heiningen 2006).
Although some studies have assessed the pre-extraction of hemicelluloses before
kraft pulping (Li et al. 2010; Wafa Al-Dajani et al. 2009; San Martin et al. 1995; Ahvazi
et al. 2007), few of them have evaluated the potential of steam explosion to this end.
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 514
PEER-REVIEWED ARTICLE bioresources.com
Therefore, deeper research would be needed to evaluate not only the effect of steam
explosion pretreatment on kraft pulping, but also the potential recovery of sugars from
the liquid fraction generated during this pretreatment.
The aim of this work was to evaluate the possibility of integrating a kraft pulping
into a forest biorefinery, studying the sugar recovery and the effect of steam explosion on
kraft pulping and pulp properties. We selected Eucalyptus globulus because it is the raw
material most commonly used to produce kraft pulp in Spain, and also because it has a
high content of hemicelluloses (pentosan content 17.9 % on o.d. wood).
EXPERIMENTAL
in which T is the temperature (ºC) and t the duration of the treatment (min).
The treated chips were then removed from the blowing tank, washed with cold
water, air-dried, and finally weighted. The water content was measured to determine the
total yield of the steam explosion process.
The water retention, or hydration capacity, of the chips was the weight of water
absorbed by the chips after being immersed for 6 hours, and it was expressed as grams of
water per 100 grams of dry wood.
The pretreatment that provided the highest water retention was selected for further
studies. The pretreated chips were analyzed to determine: acetone extractives (UNE-EN
ISO 14453), hot water extractives (UNE 57-013), lignin content (TAPPI T 222 om-88;
UNE 57-100), and holocellulose content (Wise et al. 1946). Chips from control A were
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 515
PEER-REVIEWED ARTICLE bioresources.com
also analyzed as reference. Also, the liquid fraction of the selected steam explosion was
recovered to determine the sugar content. An aliquot (1 mL) was filtered through 0.45
μm membranes and used for direct HPLC determination of monosaccharides and acetic
acid. A second aliquot of 25 mL was subjected to quantitative posthydrolysis with 4%
H2SO4, at 120ºC, for 60 min., before HPLC analysis. The increase in the concentrations
of monosaccharides and acetic acid caused by posthydrolysis was used as a measure of
the concentrations of oligomers and acetyl groups bound to oligosaccharides, respectively
(Garrote et al. 1999).
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 516
PEER-REVIEWED ARTICLE bioresources.com
pulping, whether steam-exploded or not. Moreover, the overall yield has been calculated
taking into account both the steam explosion and the kraft processes. Black liquor was
titrated with HCl to determine the chemical reagent consumption. The kappa number
(TAPPI T-236/UNE 57034), viscosity (UNE 57039-1), and hexenuronic acid content
(Gellersted and Li. 1996) were determined in the pulp samples. Also, the pulp quality
was assessed in an optical microscope (Vanox AH3 Olympus connected to a Color View
III digital camera).
The cooking process that produced the best pulp of those that were assayed was
scaled up, changing the liquid-wood ratio and the active alkali because of an equipment
limitation, as it will be explained below. Kraft cooking was done in a 26 L batch reactor
furnished with a system for recirculation and heating of the cooking liquor. The cooking
temperature was controlled by a computer running specially developed software. Due to
the volume of the liquor recirculation circuit and to the characteristics of the steam
exploded chips, the liquid-wood ratio had to be increased to 6 L/Kg in order to assure a
complete immersion of the chips. The active alkali was also raised to 18% to approach
the concentration in the small reactors, since a greater volume of water was used. Other
cooking conditions were fixed as follows: 1000 g of dry chips, 20% sulfidity, 160 ºC
cooking temperature, 40 minutes to maximum temperature, and 27 minutes at cooking
temperature (H-Factor of 200).
After the cooking process was completed, the chips were discharged into the
blowing tank, washed, disintegrated, and screened to determine their reject content,
screened and total yields, kappa number (ISO 302), and pulp viscosity (ISO 5351-1).
Control pulp, obtained from chips that were not subjected to the steam explosion
pretreatment, had a high lignin content, needing further delignification to be comparable
with the pulp from pretreated chips. Therefore, oxygen delignification was applied to this
pulp in a 20 L rotatory reactor. Conditions of this treatment were: 0.6 MPa oxygen
pressure, 10% consistency, 5% NaOH (on dry pulp), 0.5% MgSO4 (on dry pulp), and 60
minutes of treatment at 110 ºC. Control pulp characteristics were evaluated as previously
described for the other pulps.
Accelerated Ageing
Exploded, control, and commercial pulps were subjected to accelerated ageing in
order to study the evolution of their optical and mechanical properties and also the effect
of the steam explosion pretreatment on this evolution. The accelerated ageing consisted
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 517
PEER-REVIEWED ARTICLE bioresources.com
of a moist heat treatment at 80ºC and 65% relative humidity during 6 days, according to
the standard UNE 57092-4.
a) b)
100% 300
Water / Wood (%)
80% 250
200
60%
Yield
Taking into account these results, the SE8 treatment, consisting of water pre-
extraction followed by two steam explosion cycles of 10 and 3 minutes, was selected as
the most appropriate pretreatment for kraft pulping. It provided chips with a more open
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 518
PEER-REVIEWED ARTICLE bioresources.com
structure that could be pulped more easily. The low yield obtained under these conditions
would reduce the overall yield after the kraft process, but it would remove unaltered
hemicelluloses that could be converted into value-added products, polymers, or ethanol
for fuel (Li et al. 2010), increasing the total revenue stream for the pulp mill (Ragauskas
et al. 2006; Van Heiningen 2006).
The chemical composition of the SE8 and the Control A samples, as well as the
sugar content of the liquid fraction from the SE8 treatment are shown in Tables 2 and 3,
respectively. It can be observed that there was an increase in the extractives content
(water soluble and acetone soluble) and a reduction in the holocellulose content of the
wood chips after steam explosion. The increase in extractives could be due to the fact
that, during the steam explosion treatment, some macromolecules, such as lignin and
polysaccharides, are partially degraded, becoming extractible in a subsequent solvent
extraction. The holocellulose reduction was due to the hemicellulose degradation during
the steam explosion treatment, as has been mentioned above. The sugar content in the
liquid fraction obtained during this treatment also confirmed this hemicellulose
degradation, since xylose was the main sugar present in this fraction. Eight percent of the
original solid material (E. globulus) was recovered as xylose, after the SE8 treatment, and
this could be used to produce value-added products.
Control A SE8
a a
Acetone extractives 1.1% 4.9%
a a
Hot water extractives 2.8% 7.5%
a a
Lignin 18.0% 18.4%
a a
Holocellulose 73.9% 64.7%
a a
Pentosans 17.9% 10.4%
Holocellulose / lignin ratio 4.1 3.5
a
Referred to weight/weight over dry wood
Table 3. Sugar Contents of the Liquid Fraction from the Steam Explosion
Treatment SE8
a b
g/L % extracted
Acetyl groups 1.7 4.6%
Glucose 0.2 0.5%
Xylose 3.0 8.0%
Arabinose 0.04 0.1%
Total 4.96 13.2%
a
Including monomers and polymers
b
Weight of extracted sugar respect to weight of original material
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 519
PEER-REVIEWED ARTICLE bioresources.com
Total and screened kraft yields (without taking into account steam explosion
yield; in other words, the yield calculation was based on the solids content of the material
submitted to kraft pulping, whether steam-exploded or not) have been plotted versus the
H-Factor in Fig. 2a and 2b, respectively. The total yield of both control and exploded
samples decreased with the cooking time (higher H-Factor) because more lignin, and also
more carbohydrates, were dissolved in the cooking liquor. Total and screened yields
obtained with control chips were higher than those obtained with exploded chips in all
cases, except from screened yields when the H-Factor was lower than 200. In order to
explain this phenomenon, it has to be taken into account that when short pulpings were
performed, control chips were not cooked long enough and a high amount of rejects was
obtained; therefore a higher total yield and a lower screened yield were achieved. On the
other hand, when longer pulping was carried out, control chips were cooked long enough,
but exploded chips were overcooked, dissolving a high amount of carbohydrates in the
cooking liquor and obtaining lower yields. However, the last comparison was made for
the same H-factor, and the exploded and control chips did not need the same pulping
intensity. Therefore, a comparison focusing on the kappa number of the pulps would be
more appropriate.
In Fig. 2c and 2d it can be observed that the steam explosion pretreatment
increased the consumption of both Na2S and NaOH during kraft pulping, in contrast to
what has been reported by San Martin et al. (1995). However, this higher chemical
consumption produced pulp with a lower kappa number (as it will be explained below),
suggesting a higher efficiency of the chemical in delignification.
a) b)
70% 70%
Screened Yield
60% 60%
Total Yield
50% 50%
40% 40%
30% 30%
20% 20%
150 250 350 450 150 250 350 450
ControlA ControlA
H-Factor H-Factor
SE8 SE8
c) d)
100% 100%
NaOH Consumption
Na2S Consumption
80% 80%
60% 60%
40% 40%
20% 20%
0% 0%
150 250 350 450 150 250 350 450
ControlA ControlA
H-Factor H-Factor
SE8 SE8
Fig. 2. Yields and chemical consumption during kraft pulping of control (Control A ) and
exploded chips (SE8 ): total yield (a), screened yield (b), Na2S consumption (c), and NaOH
consumption (d)
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 520
PEER-REVIEWED ARTICLE bioresources.com
In Fig. 3a and 3b, the kappa number and viscosity of exploded and control pulps
are plotted versus the H-factor. Kappa numbers of exploded pulps were much lower than
those obtained for control pulps due to the open structure of exploded chips, which
increases the reagent accessibility. A maximum reduction of 70% in the kappa number
(from 49 to 16) was observed when exploded chips were cooked under the mildest
conditions (H-factor of 180). These results are similar to those found by San Martin et al.
(1995), although they obtained a lower decrease in the kappa number (44% from control
pulp). Moreover, the kappa number decreased, as it was expected, when cooking was
more intensive (higher H-factor) because more lignin was removed. This drop in the H-
factor was lower in the cases of exploded pulps because the kappa number obtained with
an H-factor of 180 was already low, and removing the residual lignin was more difficult
and caused the degradation of more carbohydrates (providing a lower screened yield with
longer pulping processes).
With regard to pulp viscosity, similar values were found in most of the samples
and were comparable with the values reported for commercial pulps. However, the
viscosity of the exploded pulp decreased dramatically when the H-Factor was higher than
360 due to overcooking of the chips, causing cellulose degradation (not observed in
control chips). For an H-Factor of 270, the same viscosity was found for exploded and
control pulps although the kappa number was much lower for the former (11.2 versus
30.4). Therefore, higher delignification could be obtained with exploded pulp without
degrading carbohydrates.
Moreover, the content of hexenuronic acids (HexA) was found to be lower in
exploded pulp than in control pulp (Fig. 3c). Taking into account that hemicelluloses are
dissolved during the steam explosion pretreatment and that HexA are formed from xylans
during kraft pulping, exploded chips gave pulps with lower HexA content. This could be
a great advantage of the steam explosion pretreatment because the presence of HexA is
detrimental to bleaching, since it increases the consumption of bleaching chemicals
(Vuorinen et al. 1999; Ragnar 2001). On the other hand, the loss of hemicellulose in the
steam explosion pretreatment could be detrimental to the refining and mechanical
properties of the handsheets. This effect will be considered later in this article.
a) b) c)
μmol HexA / g dry pulp
60 1400 40
Viscosity (ml/g)
Kappa Number
50 1200 35
1000 30
40 25
800
30 20
600 15
20
400 10
10 200 5
0 0 0
150 250 350 450 150 250 350 450 150 250 350 450
ControlA ControlA ControlA H-Factor
SE8
H-Factor SE8 H-Factor SE8
Fig. 3. Kappa number (a), viscosity (b) and hexenuronic acids content (c) of kraft pulps obtained
from control (Control A ) and exploded chips (SE8 )
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 521
PEER-REVIEWED ARTICLE bioresources.com
In Fig. 4, fibers from exploded and control pulps can be observed. These pictures
show that, after the 24 minute pulping process (H-factor of 190), control chips had not
cooked long enough, presenting a high number of fiber bundles (shives). On the contrary,
neither shives nor broken or degraded fibers were found in pulp produced from exploded
chips.
a) b)
Fig. 4. Pulp fibers obtained from (a) control chips (Control A) and (b) steam exploded chips (SE8)
From all these results, an H-factor of 200 (27 minutes at maximum temperature)
was selected, because a more intensive cooking would overcook the steam exploded
chips, as has been indicated above. This process was scaled up in a 26-L digester in order
to evaluate the effect of the scale change as a first step prior to the industrial scale-up.
The kappa number, the viscosity, and the total and screened kraft yield obtained for steam
exploded (SE8) and untreated chips (control A) are shown in Table 4.
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 522
PEER-REVIEWED ARTICLE bioresources.com
viscosity could be simultaneously obtained. These results corroborate that steam
explosion followed by a short kraft pulping pretreatment is a more efficient process than
short kraft pulping followed by oxygen delignification.
From all these results, it can be concluded that, under the same cooking
conditions, the steam exploded chips cooked faster and more efficiently than the
untreated chips, or alternatively, they could be pulped at a lower temperature and/or with
lower active alkali, while maintaining the same production rate (Ahvazi et al. 2007; San
Martin 1995; Wafa Al-Dajani et al. 2009). With the first option, the cooking time could
be reduced, increasing the production rate of the pulp mill. Moreover, as the
hemicelluloses have been removed before pulping, the charge of solids in the recovery
furnace has been reduced also and an increase of pulp production could be possible.
In these trials, the cooking time was reduced from 50 to 20 minutes to obtain a
pulp of similar characteristics (kappa number ~16 and viscosity ~1000 ml/g) with a good
yield (+50%). Even though the overall pulp yield of the process (including steam
explosion and kraft pulping) was lower (~40%), the pulp production of a hypothetic mill
could be maintained at a similar level because of the increase in pulp production.
Moreover, most of the hemicelluloses were removed before pulping without excessive
degradation and converted into value-added products. Finally, since the exploded pulp
had a very low HexA content, the consumption of chemicals in a subsequent bleaching
process is expected to be reduced also.
60
50
40
ºSR
30
20
10
0
0 1000 2000 3000
Nº Rev. PFI
SE Control Indust.
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 523
PEER-REVIEWED ARTICLE bioresources.com
The response of exploded pulp to refining was evaluated and compared with the
control pulp response. As has been mentioned, before refining, the control pulp was
subjected to oxygen delignification to obtain a pulp with lignin content similar to that of
the steam explosion pulp. Moreover, an industrial pulp (kindly provided by Torraspapel
group) was also refined and used as reference. The variation of the refining degree
(measured as Schopper–Riegler) with the intensity of the PFI refining is plotted in Fig. 5.
It can be observed that the exploded pulp exhibited a slower increase of ºSR with
PFI revolutions than control and industrial pulps did. This means that the exploded pulp
requires more energy to obtain a certain ºSR. This could be due to the removal of
hemicelluloses during the steam explosion pretreatment, as has been mentioned above.
Moreover, the control pulp showed a higher ºSR for a given PFI revolution number
compared to the industrial pulp studied.
Handsheets were formed from unrefined and refined control, industrial, and
exploded pulps, and their mechanical and optical properties were determined. Results are
shown in Fig. 6.
The unrefined steam exploded pulp showed a ºSR of 17, similar to that obtained
for unrefined control and industrial pulps (15 and 16 ºSR respectively). Also, the
mechanical properties were found to be similar to those of unrefined industrial pulp, apart
from the tensile index. This index is related to the linking capacity between fibers, so the
low value obtained for the exploded pulp could be due to the loss of hemicelluloses
during the steam explosion pretreatment. Tear index was found to be similar for
unrefined exploded and industrial pulps, which would indicate a similar degree of fiber
degradation, consistent with what San Martin et al. (1995) have reported.
It can be observed also in Fig. 6 that the apparent bulk density increased with
refining in a similar way for exploded, control, and industrial pulps.
The mechanical properties also increased with refining in all pulp samples, as was
expected. Tensile and burst indexes were found similar for control and industrial pulps
but lower for the exploded pulp. As has been mentioned above, this could be due to the
removal of hemicelluloses during steam explosion.
Although control and industrial pulp tear indexes were higher than the exploded
pulp tear index, the differences between them were lower than those observed in tensile
and burst indexes. Unrefined exploded and industrial pulps gave similar values of tear
index; however when refining was performed, the industrial pulp showed a better
evolution, providing higher values of tear index.
In order to compare optical properties, the brightness, yellowness index, and
opacity of dried pulp sheets were measured. The steam exploded pulp presented the best
optical properties, since its yellowness index was found to be much lower than that of the
control and industrial pulps, its opacity was higher than that of the control pulp, and its
brightness was higher than that of the industrial pulp. Considering the evolution of the
optical properties with refining, it can be observed that yellowness index and opacity did
not change, but brightness decreased when the ºSR increased. This could be caused by
the homogenization of the pulp resulting from the split of the microscopic fiber bundles
present in the pulps.
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 524
PEER-REVIEWED ARTICLE bioresources.com
a) b)
0,9 100
90
0,8
80
0,7 70
(g/cm 3)
0,6 60
50
0,5 40
0,4 30
20
0,3 10
0,2 0
10 20 30 40 50 60 10 20 30 40 50 60
ºSR ºSR
SE Control Indust. SE Control Indust.
c) d)
10
5
9
8 4
7
3
6
5 2
4
1
3
2 0
10 20 30 40 50 60 10 20 30 40 50 60
ºSR ºSR
SE Control Indust. SE Control Indust.
e) f)
45 50
Yellowness (%)
45
Bightness (%)
40
40
35
35
30
30
25
25
20 20
10 20 30 40 50 60 10 20 30 40 50 60
ºSR ºSR
SE Control Indust. SE Control Indust.
Fig. 6. Mechanical and optical properties of handsheets from exploded (SE ), control (Control
) and industrial (Indust. ) pulps: a) apparent bulk density, b) tensile index, c) tear index,
d) burst index, e) brightness, and f) yellowness index.
Based on the last results, it can be concluded that although exploded pulp showed
slightly lower mechanical properties, optical properties were better than those of control
and industrial pulps. Therefore, depending on the final paper use, exploded kraft pulp
could be a good alternative to conventional kraft pulp.
Accelerated Ageing
The exploded, control, and industrial pulps were subjected to accelerated ageing
in order to study the evolution of their optical and mechanical properties. Moreover, the
effect of the steam explosion pretreatment on this evolution was evaluated.
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 525
PEER-REVIEWED ARTICLE bioresources.com
Table 5 shows the changes occurring in mechanical and optical properties with
ageing. The reduction percentage for each mechanical or optical property was calculated
as an average of the reductions (positive values in Table 5) or increases (negative values
in Table 5) observed at different ºSR in each pulp before and after the accelerated ageing.
After six days of accelerated ageing, the tensile index did not decrease
significantly, while the tear and burst indexes were more affected. The steam exploded
pulp showed the lowest decrease in brightness with ageing of all samples, but in the
control pulp, the decrease was more pronounced than in the other pulps (9.2%). However,
the steam exploded pulp presented the biggest increase in yellowness index (-16.7%), but
even after ageing, the exploded pulp exhibited lower values of yellowness index than the
control and industrial pulps.
From all these results, the steam explosion pulps exhibited a more favorable
evolution with time, experiencing a lower decrease in their mechanical and optical
properties after the accelerated ageing.
CONCLUSIONS
ACKNOWLEDGMENTS
The authors wish to thank the Spanish Ministry of Science and Innovation for
funding this study via Project CTQ 2006-00385, CTQ 2009-11268, the Ramón y Cajal
Programme, and the FPI fellowship (BES2007 – 14528).
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 526
PEER-REVIEWED ARTICLE bioresources.com
REFERENCES CITED
Ahvazi, B., Radiotis, T., Bouchard, J., and Goel, K. (2007). “Chemical pulping of steam-
exploded mixed hardwood chips,” J. Wood Chem. Technol. 27(2), 49-63.
Avellar, B. K., and Glasser, W. G. (1998). “Steam-assisted biomass fractionation. I.
Process considerations and economic evaluation,” Biomass Bioenerg. 14 (3), 205-
218.
Babcock, L.W. (1932). “Method of producing fermentable sugars and alcohol from
wood,” U.S. Patent No. 1855464.
Britt, K. W. (1970). Handbook of Pulp and Paper Technology, 2nd edn. Van Nostrand,
New York.
Chua, M. G. S., and Wayman, M. (1979). “Characterization of autohydrolysis aspen
(P.tremuloides) lignin. Part 1. Composition and molecular weight distribution of
extracted autohydrolysis lignin,” Can. J. Chem. 57, 1141-1149.
Delong, E. A. (1987). “Method of rendering lignin separable from cellulose and
hemicelluloses and the product so produced,” Canadian Pat. 1217765.
Garrote, G., Dominguez, H., and Parajo, J. C. (1999). “Mild autohydrolysis: An
environmentally friendly technology for xylooligosaccharide production from wood,”
J. Chem. Technol. Biotechnol. 74, 1101-1109.
Gellersted, G., and Li, J. (1996). “An HPLC method for the quantitative determination of
hexeneuronic acid groups in chemical pulps,” Carbohydr. Res. 294, 41-51.
Heitner, C., Argyropoulos, D. S., Miles, K.D., Karnis, A., and Kerr, R. D. (1993).
“Alkaline sulphite ultra-yield pulping of Aspen chips – A comparison of steam-
explosion and conventional chemimechanical pulping,” J. Pulp & Paper Sci. 19(2),
58-70.
Jakobsons, J., Hortling, B., Erins, P., and Sundquist, J. (1995). “Characterization of alkali
soluble fraction of steam exploded birch wood,” Holzforschung 49, 51-59.
Josefsson, T., Lennholm, H., and Gellerstedt, G. (2001). “Changes in cellulose
supramolecular structure and molecular weight distribution during steam explosion of
aspen wood,” Cellulose 8, 289-296.
Josefsson, T., Lennholm, H., and Gellerstedt, G. (2002). “Steam explosion of aspen
wood. Characterization of reaction products,” Holzforschung 53(3), 289-297.
Li, H., Saeed, A., Jahan, M.S., Ni, Y., and Van Heiningen, A. (2010). “Hemicellulose
removal from hardwood chips in the pre-hydrolysis step of the kraft-based dissolving
pulp production process,” J. Wood Chem. Technol. 30, 48-60.
Manson, W. H. (1928). “Process and apparatus for disintegration of wood and the like,”
U.S. Patent No. 1655618.
Puls, J., Poutanen, K., Körner, H.U., and Viikari, L. (1985). “Biotechnical utilization of
wood carbohydrates after steaming pretreatment,” Appl. Microbiol. Biotechnol. 22,
416-423.
Overend, R. P., and Chornet, E. (1987). “Fractionation of lignocellulosics by
steamaqueous pretreatments,” Philos. Trans. R. Soc. London. A321, 523-536.
Ragauskas, A. J., Nagy, M., Kim, D. H., Eckert, C.A., Hallett, J. P., and Liotta, C. L.
(2006). “From wood to fuels: integrating biofuels and pulp production,” Indus.
Biotech. 2(1), 55-65.
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 527
PEER-REVIEWED ARTICLE bioresources.com
Ragnar, M. (2001). “On the importance of the structural composition of pulp for the
selectivity of ozone and chlorine dioxide bleaching,” Nordic Pulp Pap. Res. J. 16, 72.
Robert, D., Bardet, M., Lapierre, D., and Gellerstedt, G. (1988). “Structural changes in
aspen lignin during steam explosion treatment,” Cell. Chem. Technol. 22, 221-230.
San Martin, R., Perez, D., and Briones, R. (1995). “Simultaneous production of etanol
and kraft pulp from pine (Pinus radiata) using steam explosion,” Bioresource
Technol. 53, 217-223.
Van Heiningen, A. (2006). “Converting a kraft pulp mill into an integrated biorefinery,”
Pulp Pap. Can. 107(6), 141-146.
Vuorinen, T., Fagerström, P., Buchert, J., Tenkanen M., and Teleman, A. (1999).
“Selective hydrolysis of hexenuronic acid groups and its application in ECF and TCF
bleaching of kraft pulps,” Pulp Pap. Sci. 25(5), 155-162.
Wafa Al-Dajani, W., and Tschirner, U.W. (2008). “Pre-extraction of hemicelluloses and
subsequent kraft pulping. Part I: alkali extraction,” Tappi J. 7(6), 3-8.
Wafa Al-Dajani, W., Tschirner, U.W., and Jense, T. (2009). “Pre-extraction of
hemicelluloses and subsequent kraft pulping. Part II: Acid- and autochydrolysis,”
Tappi J. 30-37.
Wise, L. E., Murphy, M., and Daddieco, A. A. (1946). “Chlorite holocellulose, its
fractionation and bearing on summative wood analysis and on studies on the
hemicelluloses,” Tech. Assoc. Pap. 29, 210-218.
Article submitted: November 11, 2010; Peer review completed: December 18, 2010;
Revised version received and accepted: December 23, 2010; Published: December 28,
2010.
Martin-Sampedro et al. (2011). “Steam explosion kraft,” BioResources 6(1), 513-528. 528
PUBLICACIÓN V:
Biobleaching of Eucalyptus globulus kraft pulps: Comparison between pulps
obtained from exploded and non-exploded chips
245
Bioresource Technology 102 (2011) 4530–4535
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
a r t i c l e i n f o a b s t r a c t
Article history: The aim of this work was to evaluate the response to biobleaching of steam exploded kraft pulps and to
Received 25 November 2010 compare the results with the controls. For this end, a laccase-mediator treatment using commercial
Received in revised form 22 December 2010 laccase (Novozyme 51003) and a natural mediator (acetosyringone) were assayed, followed by alkaline
Accepted 22 December 2010
extraction and hydrogen peroxide stages.
Available online 30 December 2010
Our approach resulted in exploded biobleached pulps with lower kappa number and improved optical
properties compared to controls, even after subjecting pulps to accelerated ageing. Additionally, use of
Keywords:
hydrogen peroxide was reduced. The LMS (laccase-mediator system) had a smaller impact on the prop-
Steam explosion
Biobleaching
erties of the bleached pulps and on hydrogen peroxide consumption than the steam explosion process
Laccase did.
Acetosyringone Ó 2010 Elsevier Ltd. All rights reserved.
Eucalyptus globulus
1. Introduction exploded pulps have showed better optical properties and a lower
hexenuronic acid content than control pulps (non-exploded) (Mar-
The steam explosion process consists in the treatment of tin-Sampedro et al., 2010). Bearing in mind all these findings,
lignocellulosic materials with high-pressure steam followed by a improving the efficiency of the bleaching process by using ex-
rapid decompression which forces the fibrous material to ‘‘ex- ploded pulps seems a reasonable expectation and the evaluation
plode’’ into separated fibers and fiber bundles. If no chemicals of the steam exploded pulps response to bleaching could provide
are added to the process, the high-temperature of the steam re- interesting results.
leases acids from the acetylated wood components which catalyze Elemental chlorine free (ECF) and totally chlorine free (TCF) se-
auto-hydrolytic reactions in the wood polymers with the resulting quences are the main bleaching processes used at the pulp and pa-
loss of hemicellulose and decreased content of b-O-4 structures in per industry. However, TCF sequences normally use reagents such
lignin (Robert et al., 1988; Jakobsons et al., 1995; Josefsson et al., as oxygen, ozone or hydrogen peroxide which, because of their low
2002, 2001). selectivity for lignin, degrade cellulose and yield bleached pulps of
Steam explosion has been proposed for a wide range of lignocel- lower quality which account only for a limited part of the market.
lulosic applications, such as biomass fractionation, pre-treatment The application of biotechnology to this process has attracted
in bioethanol production or alternatives to conventional chemi- considerable interest. Many researchers have pretreated pulps
mechanical pulping (CMP) and chemi-thermo-mechanical pulping with enzymes, mainly with the laccase-mediator system (LMS), be-
(CTMP). Also, recent reports have shown that the cooking liquor fore applying the standard bleaching sequences (Viikari et al.,
diffuses more freely into the open structure of exploded chips 1986; Moldes et al., 2008; Eugenio et al., 2010a). These biobleach-
which leads to easier and shorter kraft pulping (Ahvazi et al., ing studies have shown that enzymatically pretreated pulps re-
2007; Martin-Sampedro et al., 2010). Moreover, hemicelluloses re- quire a smaller amount of chemicals in subsequent bleaching
moved during the steam explosion pre-treatment can be further processes, that the quality of the pulp is maintained, if not im-
converted into value-added products such as ethanol, polymers proved, and that the pollutant load of the process effluents is re-
and other chemicals (Li et al., 2010). As a result of all this, the total duced. However, these biobleaching processes have never been
revenue streams for the pulp and paper industry could be in- applied to exploded pulps. The improvement of pulp properties
creased, converting the pulp mill into an integrated forest biorefin- with steam explosion, combined with the advantages resulting
ery (Ragauskas et al., 2006; Van Heiningen, 2006). In addition, from the use of enzymes, could increase the effectiveness of
conventional industrial bleaching processes.
⇑ Corresponding author. Tel.: +34 913476761; fax: +34 913476767. All these considerations have led our team to evaluate the
E-mail address: [email protected] (J.C. Villar). response to biobleaching of steam exploded kraft pulps and to
0960-8524/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2010.12.090
R. Martín-Sampedro et al. / Bioresource Technology 102 (2011) 4530–4535 4531
compare the results with those obtained using non-exploded kraft were set at 10% and 17.5 U/g (o.d.p.), respectively. Here 1 U of lac-
pulps. case is defined as the amount of laccase required to convert
1 lmol/min of ABTS to its cationic radical (0.1 M phosphate buffer,
pH 6, 24 °C). Acetosyringone was the natural mediator used in all
2. Methods
enzymatic treatments at a concentration of 1.5% o.d.p. A few drops
of 0.05% Tween 80 were added. We designated this enzymatic
2.1. Raw materials and chemicals
treatment as LMO2.
Controls (No-LMO2) were included in our experimental design.
Eucalyptus globulus chips and E. globulus kraft pulp were kindly
Although the laccase and the mediator were omitted here, the pulp
provided by La Montañanesa pulpmill (Torraspapel group, Spain).
was treated with an aqueous solution (buffer) with the same pH,
This industrial pulp was obtained using standard kraft pulping
time and temperature values as those used in other laccase
without steam explosion pre-treatment, nor oxygen delignifica-
pre-treatments, but without adding oxygen.
tion. Its initial kappa number, brightness and viscosity were 14,
The importance of oxygen presence during the enzymatic
34.4 °ISO and 1136 ml/g, respectively.
bleaching stage is well established; nevertheless, in order to study
2,2́-Azino-bis-3-ethylbenzthiazoline-6-sulphonate (ABTS) was
its degree of influence, we carried out another enzymatic experi-
purchased from Roche (Madrid, Spain). All other chemicals: aceto-
ment (called LM-NoO2) using the same conditions as those of the
syringone (ASG), sodium hydroxide (NaOH), magnesium sulfate
LMO2 treatment and without the addition of oxygen.
(MgSO4), diethylene triamine pentaacetic acid (DTPA), sodium sul-
fide (Na2S) and hydrogen peroxide (H2O2) were reagent-grade and
supplied by Merck (Barcelona, Spain) or Sigma–Aldrich (Madrid, 2.4. Alkaline extraction (E) and hydrogen peroxide (P) treatments
Spain).
Conditions for the alkaline extraction subsequent to the enzy-
2.2. Pulps production matic treatment were: 1.5% NaOH, 5% consistency and 90 °C tem-
perature for 120 min. Thereafter, pulps were washed, and a
The steam explosion treatment of the E. globulus chips consists hydrogen peroxide bleaching stage was applied under the follow-
of cold water pre-extraction followed by two steam explosion cy- ing conditions: 2% H2O2, 1.5% NaOH, 1% DTPA, 0.2% Mg2SO4, 5%
cles of 10 and 3 min each. In each cycle, chips were treated with consistency and 90 °C for 90 min. Residual hydrogen peroxide
steam at 183 °C (10 kg/cm2) and discharged at 6 kg/cm2 into a was analyzed in the bleaching effluent by standard titration.
blowing tank. The process yield was 80%, and 50% of the hemicel-
luloses were extracted. The composition of the exploded chips 2.5. Accelerated ageing
was: lignin content, 18.4%; holocellulose, 64.7%; acetone extrac-
tives, 4.9%; and hot water extractives, 12.4%. Next after the bleaching sequence (LEP), pulps were subjected
Steam exploded chips and control chips (no steam explosion) to accelerated ageing in order to study the influence of steam
were subjected to kraft cooking in a 26-L digester connected to a explosion and enzymatic pre-treatment on optical properties.
blowing tank by a pneumatic valve. The cooking liquor was heated Accelerated ageing consisted of a moist heat treatment conducted
indirectly with steam in an external tubular heat exchanger, and at 80 °C and 65% relative humidity for 6 days, in accordance to the
the cooking temperature was controlled by computer software. standard ISO 5630-3.
Operational conditions were: 18% active alkali, 20% sulfidity,
40 min to maximum temperature, 27 min at maximum tempera- 2.6. Pulps characterization
ture (160 °C) and a 6 L/kg liquor-to-wood ratio.
The pulp obtained from control chips was treated with oxygen Unbleached pulps and pulps obtained after each step of the
for further delignification in a 20-liter rotatory reactor with a jack- bleaching sequence (LEP) were characterized in terms of their kap-
et-type electrical heater. Conditions of oxygen delignification were pa number, viscosity, brightness, yellowness index and CIE L⁄a⁄b⁄
110 °C, 60 min, 0.6 MPa of oxygen pressure, 5% NaOH o.d.p. (over and CIE L⁄C⁄ coordinates, in conformity with the standards ISO
dry pulp), 0.5% MgSO4 o.d.p. and 10% consistency. 302, ISO 5351-2, ISO 2470, T 1216 and T 527, respectively.
Fig. 1. Pulp kappa numbers for each stage in LEP sequences. L: laccase stage; E: alkaline extraction stage; P: hydrogen peroxide stage; Ind: industrial pulp; NoSE: control pulp
without steam explosion pre-treatment; SE: steam exploded kraft pulp; No-LMO2: no laccase, no mediator, no oxygen; LMO2: laccase, mediator and oxygen; LM-NoO2:
laccase, mediator and no oxygen.
stage (L), the alkaline extraction (E) and the bleaching by hydrogen end of the LEP sequence, a pulp with a slightly lower reduction
peroxide (P). of the kappa number (52.3%) than that observed with 6 kg/cm2 of
Fig. 1 shows that the kappa number of the unbleached exploded oxygen pressure (SE:LMO2) (54.5%). These findings suggested that
kraft pulp was lower than that found for the unbleached non- oxygen presence is not required during the enzymatic stage to im-
exploded kraft pulp. This means that a steam explosion pre-treat- prove delignification, and similar results have been reported else-
ment, followed by short kraft pulping, caused higher delignifica- where (Eugenio et al., 2010b).
tion than short kraft pulping followed by oxygen delignification. The viscosity is an indirect indicator of the length of the poly-
On the other hand, the unbleached industrial pulp had a slightly saccharide chains in cellulose pulps. As Fig. 2 shows, non-exploded
lower kappa number than the exploded kraft pulp, suggesting sim- pulps had slightly less viscosity than exploded pulps. This can be
ilar delignification in both pulping processes. However, the steam explained by the elimination of short chain polysaccharides, such
explosion pre-treatment had the advantage of reducing the cook- as hemicelluloses, in the steam explosion treatment, resulting in
ing time, which can potentially increase the pulpmill productivity an increase in the average length of the polysaccharides. Hydrogen
(Martin-Sampedro et al., 2010). peroxide bleaching was found the most influential stage for viscos-
In most cases, each stage in the biobleaching sequence led to a ity reduction in the non-exploded pulps, whereas viscosity was
reduction in the kappa number. Moreover, using the LMS resulted mainly affected by the laccase treatment in the exploded pulps.
in more pronounced reductions of the kappa number after each The industrial and the exploded pulps had similar viscosity. How-
stage, compared to control treatments, an outcome also reported ever, hydrogen peroxide bleaching was found to be the viscosity
by other authors (Fillat et al., 2010; Valls et al., 2010). Thus, in determining stage for industrial pulps, as it was the case with
non-exploded pulps, the total reductions of the kappa number at the non-exploded pulps.
the end of the LEP sequence were 27.7% and 14.5% for biotreated Despite viscosity reductions being small, there were not big dif-
pulp (NoSE:LMO2) and its control (NoSE:No-LMO2), respectively. ferences in viscosity across all the bleaching processes. The same
Other authors have reported a greater reduction of the kappa num- result has been observed by Oudia et al. (2008), who have demon-
ber. However, these authors included more stages in the TCF strated that high biodelignification can be achieved without
bleaching sequence, used longer enzymatic and hydrogen peroxide decreasing pulp viscosity.
treatments and added chemicals in higher concentrations (Moldes Fig. 3 shows that the brightness of the unbleached exploded
et al., 2010). kraft pulp was very similar to the brightness of the unbleached
Kappa number reductions in exploded kraft pulps were greater non-exploded kraft pulp, in spite of the higher lignin content found
than in non-exploded pulps (54.5% vs 41.7% for the SE LMO2 and in the non-exploded unbleached pulp. In contrast to this, the
SE:No-LMO2 treatments, respectively). These last results proved unbleached industrial pulp showed lower brightness than the
that the steam explosion pre-treatment significantly increased aforementioned pulps, probably because of the effect of a longer
the reduction of the kappa number not only in laccase treated sam- pulping process.
ples but also in control pulps. When exploded pulps were com- Also in Fig. 3, it can be observed that the highest increase in
pared with an industrial pulp, a lower delignification was brightness took place during the hydrogen peroxide stage, as it
observed in the industrial pulp after the LEP bleaching (47.0% in was expected, whereas the laccase stage resulted in a reduction
biobleached industrial pulp [Ind:LMO2] and 36.6% in the control of brightness, probably as a result of the formation of chromoph-
pulp [Ind:No-LMO2]). These results also suggested that a steam ores during the oxidation of lignin by the laccase-mediator system
explosion treatment previous to kraft pulping could improve any (Fillat et al., 2010). This reduction was found more pronounced
subsequent bleaching process. This statement is in agreement with when the LMS was present, being these results consistent with
Josefsson et al. (2002), who reported that the stronger the steam others reported elsewhere (Moldes et al., 2008; Eugenio et al.,
explosion pre-treatment, the lower is the lignin content in pulps 2010a). However, at the end of the LEP sequence, the effect was re-
after a chemical bleaching. versed, and higher brightness was obtained when using LMS as a
To evaluate the effect of oxygen pressure in the LMS treatment, pre-treatment. Therefore, it can be concluded that the LMS treat-
a control without addition of oxygen was performed on the pulp ment improved the effectiveness in the subsequent bleaching
which showed the most efficient LEP bleaching (highest delignifi- stages, as other authors have also showed (Vivekanand et al.,
cation degree). This experiment (SE:LM-NoO2) produced, at the 2008; Moldes et al., 2010).
R. Martín-Sampedro et al. / Bioresource Technology 102 (2011) 4530–4535 4533
Fig. 2. Pulp viscosity for each stage in LEP sequences. L: laccase stage; E: alkaline extraction stage; P: hydrogen peroxide stage; Ind: industrial pulp; NoSE: control pulp
without steam explosion pre-treatment; SE: steam exploded kraft pulp; No-LMO2: no laccase, no mediator, no oxygen; LMO2: laccase, mediator and oxygen; LM-NoO2:
laccase, mediator and no oxygen.
Fig. 3. Pulp brightness for each stage in LEP sequences. L: laccase stage; E: alkaline extraction stage; P: hydrogen peroxide stage; Ind: industrial pulp; NoSE: control pulp
without steam explosion pre-treatment; SE: steam exploded kraft pulp; No-LMO2: no laccase, no mediator, no oxygen; LMO2: laccase, mediator and oxygen; LM-NoO2:
laccase, mediator and no oxygen.
If the brightness of exploded pulps at the end of the LEP se- stage (L), being this increase more pronounced in the presence of
quence is compared to that of non-exploded pulps at the same LMS, and even more in the exploded kraft pulp. After the E stage,
point in the process, it is obvious that the steam explosion pre- the yellowness index decreased in most cases to values near to
treatment favored an increase in brightness during bleaching; spe- those of the unbleached pulp. However, it was in the P stage that
cifically 25.2% over 16.3% ISO, for SE:LMO2 and NoSE:LMO2 respec- the biggest reductions took place. On the other hand, if the whole
tively. The industrial pulp (Ind:LMO2) reached a final brightness LEP sequence is taken into consideration, a reduction in the yellow-
higher than the non-exploded pulp but lower than the exploded ness index between 7 and 15 was found in bleached pulps
pulp, which leads to conclude again (as when the kappa number compared to the unbleached ones. Furthermore, the lowest yellow-
was discussed above) that introducing the steam explosion before ness index was that of the steam exploded pulps, suggesting that
pulping improves the subsequent bleaching process. the steam explosion treatment contributes to yield pulps with less
Fig. 3 shows that, at the end of the LEP sequence, the control color and therefore closer to the ideal white.
pulp, SE:LM-NoO2, had higher brightness than SE:LMO2 (26.4% vs CIE L⁄a⁄b⁄ and CIE L⁄C⁄ color coordinates were measured after
25.2% ISO). Taking into account that this difference was not highly the P stage in pulps obtained by the most efficient bleaching se-
significant, and that the addition of oxygen did not improve the quence, corresponding in all cases to those in which LMS was used.
delignification considerably (as mentioned earlier in regards with These results were compared with those of unbleached pulps, in or-
the kappa number), it can be concluded that the addition of oxygen der to evaluate color changes undergone during the biobleaching
during the enzymatic bleaching was not a requisite to improve the process, since the color appearance of paper is an important param-
bleaching process. eter in many different applications. Therefore, a numerical defini-
Fig. 4 shows the yellowness index of pulps before and after each tion of color is essential for good quality control and for a good
bleaching stage. It can be observed that the unbleached steam ex- customer–producer relationship. Furthermore, good white base col-
ploded kraft pulp has a lower yellowness index than the non-ex- ors are necessary where products are to be dyed, printed, or other-
ploded and the industrial kraft pulps. During the LEP sequence, wise colored. The brightness and full range of color in color printing
an increase in the yellowness index occurred after the enzymatic depend to a large extent upon the whiteness of the base paper.
4534 R. Martín-Sampedro et al. / Bioresource Technology 102 (2011) 4530–4535
Fig. 4. Pulp yellowness for each stage in LEP sequences. L: laccase stage; E: alkaline extraction stage; P: hydrogen peroxide stage; Ind: industrial pulp; NoSE: control pulp
without steam explosion pre-treatment; SE: steam exploded kraft pulp; No-LMO2: no laccase, no mediator, no oxygen; LMO2: laccase, mediator and oxygen; LM-NoO2:
laccase, mediator and no oxygen.
After a LEP bleaching sequence, all pulps increased L⁄ and de- tion of steam explosion and LEP bleaching provided the most effi-
creased C⁄ (i.e. lightness increased and color decreased) compared cient bleaching process, yielding pulp with better colorimetric and
to the unbleached pulps. However, although the decrease in color optical properties.
was lower for the SE:LMO2 pulp (DC⁄, 0.8 vs 2.9 and 4.7, for No-
SE:LMO2 and Ind:LMO2, respectively), this pulp showed the lowest 3.2. Hydrogen peroxide consumption
color (C⁄, 11.2 vs 16.3 and 12.6 for NoSE:LMO2 and Ind:LMO2,
respectively) and also the highest lightness (L⁄, 91.9 vs 90.5 and Hydrogen peroxide consumption was assessed in all our exper-
90.6 for NoSE:LMO2 and Ind:LMO2, respectively), suggesting that iments. Samples subjected to a steam explosion pre-treatment
the steam explosion helped to produce pulps with better colori- showed lower hydrogen peroxide consumption during the P stage
metric coordinates. The LEP bleaching also decreased the a⁄ and (55.7–54.0%) than control pulps (77.6–66.8%) and industrial pulps
b⁄ coordinates for the steam exploded pulp (Da⁄ = 4.1 and (79.5–59.1%). Keeping in mind that hemicelluloses dissolve during
Db⁄ = 0.3), the control pulp (Da⁄ = 2.1 and Db⁄ = 2.9) and the indus- the steam explosion pre-treatment and that acidic groups on xylan
trial pulp (Da⁄ = 2.9 and Db⁄ = 4.5). turn into hexenuronic acid during kraft pulping, there is reason to
In Fig. 5, reflectance is plotted vs wavelength. The LEP bleaching expect that exploded chips will yield pulps with less hexenuronic
sequence gets the unbleached pulps closest to the ideal white, i.e. acid. In our experience, the hexenuronic acid content of the ex-
100% reflection of all wavelengths within the visible spectrum. Fur- ploded pulp was 4 lmol/g, vs 30 and 37 lmol/g in non-exploded
thermore, the steam explosion pre-treatment resulted in higher and industrial pulps, respectively. According to Vuorinen et al.
reflectance for the bleached pulp (SE:LMO2), in spite of the low (1999) and Ragnar (2001), consumption of bleaching chemicals is
reflectance observed with wavelengths above 440 nm before the minimum when the hexenuronic acid content of pulps is the low-
bleaching process. Moreover, for wavelengths close to 457 nm est. Thus, a steam explosion pre-treatment, prior to kraft pulping,
(brightness measurement), the differences in reflectance between can be expected to reduce the chemical consumption in subse-
SE:LMO2 and the other bleached pulps (NoSE:LMO2 and Ind:LMO2) quent bleaching.
were more pronounced, as the brightness values showed (Fig. 3). A logical consequence of lignin removal (kappa number reduc-
Based on these last results, it can be concluded that the combina- tion) in the LMS pre-treatment is less hydrogen peroxide use in a
further bleaching stage: the lignin remaining in the pulp has been
partially oxidized by the LMS and, as a consequence, the need for
hydrogen peroxide is reduced. Consequently, when LMS was not
used during the enzymatic treatment in our experiments, demands
for hydrogen peroxide were higher (55.7% vs 54.0% for exploded
pulps, 77.6% vs 66.8% for non-exploded pulps and 79.5% vs 59.1%
for industrial pulps).
However, if no oxygen was added during this enzymatic treat-
ment, hydrogen peroxide consumption decreased (51.2%). This
finding is consistent with the outcomes in kappa number and opti-
cal properties, suggesting that the addition of oxygen is not essen-
tial since no saving on chemical reagent nor improvement in pulp
quality was observed.
Ahvazi, B., Radiotis, T., Bouchard, J., Goel, K., 2007. Chemical pulping of steam-
ageing (from light exposure and/or high temperature and humid- exploded mixed hardwood chips. J. Wood Chem. Technol. 27 (2), 49–63.
ity) could be measured by assessing changes in the yellowness Cadena, E.M., Vidal, T., Torres, A.L., 2010. Influence of the hexenuronic acid content
on refining and ageing in eucalyptus TCF pulp. Bioresour. Technol. 101, 3554–
index. 3560.
Table 1 lists the yellowness index and the brightness before and Eugenio, M.E., Santos, S.M., Carbajo, J.M., Martín, J.A., Martín-Sampedro, R.,
after accelerated ageing of industrial, non-exploded and exploded González, A.E., Villar, J.C., 2010a. Kraft pulp biobleaching using an
extracellular enzymatic fluid produced by Pycnoporus sanguineus. Bioresour.
bleached kraft pulps. Brightness reductions and yellowness index
Technol. 101, 1866–1870.
increments were observed in all bleached pulps after accelerated Eugenio, M.E., Miranda, J., Martin-Sampedro, R., Villar, J.C., 2010b. Influence of
ageing, with a greatest impact on the industrial pulp treated with laccase bleaching process variables on bleached pulps properties. In:
Proceeding of VI Iberoamerican Congress on Pulp and Paper Research
LMS.
(CIADICYP), Lisboa, Portugal, pp. 160–161.
Although, overall, the optical properties were more stable in ex- Fillat, A., Colom, J.F., Vidal, T., 2010. A new approach to the biobleaching of flax pulp
ploded and non-exploded pulps than in the industrial pulp, the with laccase using natural mediators. Bioresour. Technol. 101, 4104–4110.
non-exploded pulp showed the lowest increase in yellowness. Jakobsons, J., Hortling, B., Erins, P., Sundquist, J., 1995. Characterization of alkali
soluble fraction of steam exploded birch wood. Holzforschung 49, 51–59.
Nevertheless, optical properties were best in the exploded pulp, Josefsson, T., Lennholm, H., Gellerstedt, G., 2001. Changes in cellulose
even after undergoing accelerated ageing, with the highest value supramolecular structure and molecular weight distribution during steam
in brightness and the lowest yellowness index. The different re- explosion of aspen wood. Cellulose 8, 289–296.
Josefsson, T., Lennholm, H., Gellerstedt, G., 2002. Steam explosion of aspen wood.
sponses to accelerated ageing showed by pulps in our study could Characterization of reaction products. Holzforschung 53 (3), 289–297.
result from the interaction of multiple factors, such as the lignin, Li, H., Saeed, A., Jahan, M.S., Ni, Y., Van Heiningen, A., 2010. Hemicellulose removal
hexenuronic acid and hemicellulose contents (Vuorinen et al., from hardwood chips in the pre-hydrolysis step of the kraft-based dissolving
pulp production process. J. Wood Chem. Technol. 30, 48–60.
1999; Sevastyanova et al., 2005). Martin-Sampedro, R., Revilla, E., Martin, J.A, Eugenio, M.E., Villar, J.C., 2010.
The effect of the LMS in the bleaching sequence can also be ob- Evaluation of steam explosion as a pretreatment prior to kraft pulping of
served in Table 1. With the only exception of the industrial pulp, Eucalyptus globulus wood. In: Proceeding of VI Iberoamerican Congress on Pulp
and Paper Research (CIADICYP), Lisboa, Portugal, pp. 62–63.
the enzymatic treatment provided pulps with more stable optical
Moldes, D., Díaz, M., Tzanov, T., Vidal, T., 2008. Comparative study of the efficiency
properties, reducing variations of brightness and yellowness. This of synthetic and natural mediators in laccase-assisted bleaching of eucalyptus
effect was more pronounced in the exploded pulp, suggesting that kraft pulp. Bioresour. Technol. 99, 7959–7965.
Moldes, D., Cadena, E.M., Vidal, T., 2010. Biobleaching of eucalypt kraft pulp with a
the combination of both treatments could be more advantageous.
two laccase-mediator stages sequence. Bioresour. Technol. 101, 6924–6929.
Other authors have reported that the enzymatic treatment in- Oudia, A., Queiroz, J., Simoes, R., 2008. Potential and limitation of Trametes
creases the brightness reversion (Cadena et al., 2010). However, versicolor laccase on biodegradation of Eucalyptus globulus and Pinus pinaster
it should be noted that their results were assessed after the alka- kraft pulp. Enzyme Microb. Technol. 43, 144–148.
Ragauskas, A.J., Nagy, M., Kim, D.H., Eckert, C.A., Hallett, J.P., Liotta, C.L., 2006. From
line extraction and not after the hydrogen peroxide stage, which wood to fuels: integrating biofuels and pulp production. Indus. Biotech. 2 (1),
could remove compounds that contribute to increase the bright- 55–65.
ness reversion. Ragnar, M., 2001. On the importance of the structural composition of pulp for the
selectivity of ozone and chlorine dioxide bleaching. Nordic Pulp Pap. Res. J. 16,
Finally, when a control was run with no addition of oxygen in 72.
the laccase treatment (SE:LM-NoO2), both the brightness and the Robert, D., Bardet, M., Lapierre, D., Gellerstedt, G., 1988. Structural changes in aspen
yellowness index were less stable than those of pulps obtained lignin during steam explosion treatment. Cell. Chem. Technol. 22, 221–230.
Sevastyanova, O., Lindström, M.E., Gellerstedt, G., 2005. Influence of a bleaching
with oxygen addition (SE:LM-O2). Therefore, although oxygen did sequence on the brightness stability of eucalyptus kraft pulp. Appita 251, 255.
not improve pulp properties before ageing, as indicated above, Valls, C., Vidal, T., Roncero, M.B., 2010. The role of xylanases and laccases on
there is a possibility that it could provide pulps with more stable hexenuronic acid and lignin removal. Process Biochem. 45, 425–430.
Van Heiningen, A., 2006. Converting a kraft pulp mill into an integrated biorefinery.
properties.
Pulp Pap. Can. 107 (6), 141–146.
Viikari, L., Ranua, M., Kantelinen, A., Linko, M., Sundquist, J., 1986. Biobleaching with
4. Conclusions enzymes. In: Proceedings of the Third International Conference on
Biotechnology in the Pulp and Paper Industry, Stockholm, pp. 67–69.
Vivekanand, V., Wived, P., Sharma, A., Sabharwal, N., Singh, R.P., 2008. Enhanced
Our results show that the LMS treatment boosted the delignification of mixed wood pulp by Aspergillus fumigates laccase mediator
subsequent chemical bleaching. Also, steam explosion prior to system. World J. Microbiol. Biotechnol. 24, 2799–2804.
Vuorinen, T., Fagerström, P., Buchert, J., Tenkanen, M., Teleman, A., 1999. Selective
kraft pulping further improved bleaching. The combination of hydrolysis of hexenuronic acid groups and its application in ECF and TCF
LMS and steam explosion took on the advantages of both pre-treat- bleaching of kraft pulps. Pulp Pap. Sci. 25 (5), 155.
ments and provided pulps with minimum kappa number and high-
PUBLICACIÓN VI:
Combination of steam explosion and laccase-mediator treatments prior to
Eucalyptus globulus kraft pulping.
253
Bioresource Technology 102 (2011) 7183–7189
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
a r t i c l e i n f o a b s t r a c t
Article history: The effect of a pretreatment consisting of steam explosion (SE) followed by a laccase mediator system
Received 1 February 2011 (LMS) stage on Eucalyptus globulus kraft pulping has been evaluated and compared with fungal pretreat-
Received in revised form 14 April 2011 ments. Pretreatment with SE and LMS was more efficient than pretreatments using Pycnoporus sanguin-
Accepted 18 April 2011
eus and Trametes sp. I-62. Steam explosion not only improved the enzyme penetration into the wood
Available online 23 April 2011
chips and shortened the pulping process by 60%, but also extracted around 50% of the hemicelluloses
which could be converted into value-added products. The optimal conditions for the LMS treatment were
Keywords:
3 h, 3 UA/g and 40 °C. Compared to SE, the SE/LMS treatment yielded an increase in delignification of
Steam explosion
Biopulping
13.9% without affecting pulp properties, provided a similar screened kraft yield, and reduced consump-
Laccase tion of chemical reagents Na2S and NaOH by 11.5% and 6.3%, respectively. Therefore, SE/LMS is a prom-
White-rot fungi ising pretreatment for converting the pulp mill into a forest bio-refinery.
Eucalyptus globulus Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction consumption (Maijala et al., 2008; Ramos et al., 2004) and in-
creased the extent of delignification (Dyer and Ragauskas, 2004;
Biopulping is the biological pre-treatment of wood chips to Jacobs-Young et al., 1998; Zhao et al., 2002).
modify wood composition such as lignin content, with the aim of One disadvantage of enzymes for biopulping is their low effec-
facilitating fiber separation or lignin removal in a subsequent tiveness due to difficulties in penetrating the chips (Grethlein,
mechanical or chemical pulping process. The concept of biopulping 1985). Therefore, enzymes have been applied to raw wood material
is based on the ability of some white-rot fungi to degrade lignin which has previously been defibrated to a certain extent (Burton,
selectively in the initial period of wood decay, leaving cellulose rel- 2001), or to chemical pulp in the bleaching process (Eugenio
atively intact (Bajpai et al., 2001; Ferraz et al., 2008; Mendonça et al., 2010). Alternatively, a vacuum (Jacobs-Young et al., 1998)
et al., 2008). White-rot fungi are not only capable of producing or pressure (Maijala et al., 2008; Pere et al., 2005) can be applied
lignin-degrading enzymes, but are also able to penetrate into the to facilitate enzyme penetration into the wood chips.
substrate to transport these enzymes into materials such as wood In the current study, steam explosion is being explored as a
chips (Messner and Srebotnik, 1994). means to facilitate enzymes penetration. This process consists in
Biotreatment could produce energy savings during defibration the treatment of lignocellulosic materials with high-pressure
or refining steps, increase in delignification rate, decrease in alkali steam followed by rapid decompression. The sudden release of
consumption, improve of pulp strength properties and reduce pul- pressure in the ‘‘explosion’’ results in fiber separation and physical
ping time and bleaching chemical consumption (Bajpai et al., 2001; rupture of the fiber walls. This open structure of the exploded chips
Islam et al., 2008; Mendonça et al., 2008; Messner and Srebotnik, not only increases the enzymatic hydrolysis of biomass for the pro-
1994; Oriaran et al., 1990; Wolfaardt et al., 2004), but fungal duction of fermentable sugars (Jurado et al., 2009; Ruiz et al., 2007;
growth can be slow, maintenance of aseptic and optimal condi- Wu et al., 1999), but also leads to easier and shorter kraft pulping,
tions during the long treatment time is required and yield loss owing to the efficient diffusion of cooking liquor into the fibers
due to fungal degradation of polysaccharides can occur. Therefore, (Ahvazi et al., 2007; Martin-Sampedro et al., 2011).
the direct application of enzymes on wood chips is being explored. The enzymes used in biopulping include cellulases, hemicellu-
It has been shown that enzymatic pre-treatments enhanced con- lases (xylanases), pectinases and ligninolytic enzymes such as lig-
ventional mechanical and chemical pulping, reduced energy nin peroxidases and manganese peroxidases. The laccase-mediator
system (LMS), which is used for biobleaching, has usually only
been used in combination with other enzymes for biopulping
⇑ Corresponding author. Tel.: +34 913476761; fax: +34 913476767. applications (Dyer and Ragauskas, 2004; Maijala et al., 2008;
E-mail address: [email protected] (J.C. Villar). Petit-Conil et al., 2002). In the current study, LMS treatment
0960-8524/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2011.04.053
7184 R. Martín-Sampedro et al. / Bioresource Technology 102 (2011) 7183–7189
following steam explosion was carried out and results were com- to convert 1 lmol/min of ABTS to its cationic radical (k = 436 nm,
pared with those of treatment with Pycnoporus sanguineus and Tra- eo = 49600 M1 cm1, 0.1 M phosphate buffer, pH 6 at 24 °C).
metes sp. I-62 prior to kraft pulping of Eucalyptus globulus wood Assays were performed using steam exploded chips prepared as
chips. it was described in Section 2.2.1. The chips were placed into
500 mL pressurized reactors, 0.2 mL of 0.05% Tween 80 was added,
and oxygen was introduced until reaching a pressure of 6 kg/cm2.
2. Methods
The reactors were submerged into a thermostatic bath to assure
a constant temperature of 40 °C.
2.1. Raw materials and chemicals
Riegler) was determined according to ISO 5267-1 and handsheets ing kraft pulping if the laccase or the mediator have been joined to
were formed from these unrefined pulps according to ISO 5269- the fibers. Thus, when medium and high laccase doses were as-
2. Apparent bulk density, tensile index and tear index were deter- sayed, the increase in the chemical consumption was higher than
mined according to ISO 5270; brightness and opacity according to the reduction due to the enzymatic oxidation of lignin, resulting
ISO 2470 and ISO 2471. in an increase in chemical consumption.
A higher delignification after kraft pulping was found in LMS
pretreated samples (Fig. 1e) without a strong degradation of the fi-
3. Results and discussion
bers, since differences in viscosity lower than 10% were observed
when treated and exploded pulp are compared (Fig. 1f). The high-
3.1. Enzymatic treatment
est delignification was achieved when the pre-treatment was car-
ried out for 3 h with a laccase dose of 3 UA/g of dry wood,
3.1.1. Steam explosion treatment
obtaining an unbleached pulp with a kappa number of 13.7, which
To facilitate the enzymes penetration into the wood chips, a
means an increase in delignification of 13.9% (compared to kappa
steam explosion (SE) pre-treatment was carried out. The water
number of exploded control pulp: 16.0). When a longer treatment
retention capacity of the chips increased after SE treatment from
was carried out with the same laccase dose, lower but significant
67 to 272 g of water per 100 g of dry wood. This finding suggests
kappa number reduction was observed (14.8 vs 16.0). Significant
that the accessibility of the exploded chips has been increased,
delignification was also found when laccase dose was 16.5 UA/g,
which could also improve the diffusion of enzymes during a subse-
independently of the treatment time, obtaining pulps with kappa
quent enzymatic treatment.
number between 14.2 and 14.4 (increase in delignification of
The effect of this pre-treatment on kraft pulping was evaluated
10.0–11.1% compared to exploded control pulps). However, for
comparing the average kappa number, viscosity, screened kraft
high laccase dose (30 UA/g) and treatment times of 3 and 5 h, the
yield and chemical consumption of the exploded samples with
kappa number of the resulting kraft pulps was higher than those
those of controls without steam explosion pre-treatment. A reduc-
of exploded control pulps. This observation indicates possible
tion in kappa number of 67.8% (15.7 vs 48.8 in control pulp) was
grafting of the LMS onto the pulp fiber (Dyer and Ragauskas,
observed, while the viscosity and the NaOH consumption were in-
2004). Thus use of a low laccase dose requires an enzymatic treat-
creased by 23.8% and 9.1% respectively, in agreement with other
ment of at least 3 h; whereas with a medium dose, 1 h of treatment
works (Ahvazi et al., 2007; Martin-Sampedro et al., 2011). Further-
was enough to increase delignification after kraft pulping. Treat-
more, screened kraft pulping yield was found higher for exploded
ments in the presence of higher laccase for more than 1 h of treat-
sample (49.6%) than for control pulp (28.9%), even if the overall
ment had a negative effect on kappa number.
yield of the process (including steam explosion) is considered
Fig. 2 shows the mechanical and optical properties of the hand-
(39.3%). Control pulp has lower screened yield because non-ex-
sheets formed from the unrefined pulps obtained using the steam
ploded chips would need a longer pulping to reduce the amount
explosion and enzymatic pre-treatment plus kraft pulping. All
of uncooked material (rejects), due to their less open structure
pulps presented a degree of refining of 15° Schopper–Riegler and
compared to exploded chips.
no big differences (lower than 4%) were observed in apparent bulk
density (around 0.39 g/cm3). Tensile and tear indexes were similar
3.1.2. Experimental design to study laccase-mediator treatment to those of exploded control pulps in most of the cases, but slightly
The influence of the laccase dose and the treatment time in total higher variation for 1 h treatments was observed. Contrarily,
and screened kraft yield, NaOH and Na2S consumptions and kappa Petit-Conil et al. (2002) observed an increase in mechanical prop-
number and viscosity of the obtained pulps is shown in Fig. 1. erties when spruce chips were treated with LMS prior to thermo-
In Fig. 1a and b it can be observed that the highest total and mechanical pulping (TMP), likely due to the different behavior of
screened kraft yields were found for longer treatment (5 h), softwood and hardwood and the different pulping process used
although no clear correlation was observed between the treatment by these authors.
time and the kraft yields. Screened kraft yield decreased when the Regarding to optical properties, a change in opacity often relates
laccase dose increased, except for 1 h of treatment at the highest to a change in the brightness of the paper (darker paper shows
laccase dose. Therefore, when a LMS treatment was carried out higher opacity), but opacity is also related to the light scattering
for 5 h with a laccase dose of 3 UA/g the highest increase in ability of paper (Maijala et al., 2008). Although treated samples
screened kraft yield was found (2.8%) followed by the treatment presented optical property values close to those of exploded con-
with 16 UA/g of laccase during 5 h (2.3%). From these results it trol samples, opacity tended to decrease when laccase doses and
can be pointed out that to increase kraft yield, at least 5 h of enzy- treatment times were increased, except for 5 h treatments. In con-
matic treatment were needed and that, even in this case, high lac- trast, brightness showed a slight increase with treatment time in
case doses were detrimental. most of the cases, but no with laccase dose, as the highest bright-
NaOH consumptions during kraft pulping of the different lac- ness (one point higher than control) for a medium laccase dose
case treated samples were lower than those of control samples (16.5 UA/g) at 5 h was obtained. Therefore, despite the highest del-
(around 67% in most of the cases; Fig. 1c). When 5 h treatments ignification after 3 h in the presence of 3 UA/g, these conditions did
were carried out with a laccase dose of 30 UA/g, the lowest NaOH not cause a change in brightness compared to that of the control.
consumption was observed (58%). However, Na2S consumptions This finding is in concordance with that reported by Dyer and
were found higher than those of control samples in most of the Ragauskas (2004) who found that LMS treated chips were darker
cases when high and medium laccase doses were used (Fig. 1d). than control chips before pulping, but these color differences were
For low laccase dose (3 UA/g), reductions in Na2S consumption no longer apparent in most of the resulting pulps after kraft
were observed. Therefore, laccase pre-treatment could reduce both pulping.
NaOH and Na2S consumptions during kraft pulping when a low Therefore, when LMS treatment with 3 UA/g at 40 °C for 3 h was
laccase dose was used, but an increase in Na2S consumption could carried out after a steam explosion treatment and prior to kraft
take place for high laccase doses. An explanation for the reduction pulping of E. globulus, pulps with a lower kappa number than that
in chemicals needed after LMS treatment could be that lignin par- of exploded control pulp (reduction of 2.3 units) could be obtained.
tially oxidized by the LMS would needed less chemical input for This treatment showed no negative effects on mechanical and
solubilization whereas some chemicals have to be consumed dur- optical properties, resulted in screened kraft yields similar to those
7186 R. Martín-Sampedro et al. / Bioresource Technology 102 (2011) 7183–7189
50%
50%
48%
48%
46%
46%
44% 44%
42% 42%
1h 3h 5h 1h 3h 5h
3 UA/g 16 5 UA/g
16.5 30 UA/g Control 3 UA/g 16 5 UA/g
16.5 30 UA/g Control
% Na2S Consumption
70% 70%
60% 60%
50% 50%
40% 40%
30% 30%
20% 20%
10% 10%
0% 0%
1h 3h 5h 1h 3h 5h
3 UA/
UA/g 16 5 UA/g
16.5 UA/ 30 UA/g
UA/ C t l
Control 3 UA/g 16 5 UA/g
16.5 30 UA/g Control
19
(e) (f) 1400
18
1200
17
1000
Viscosity (ml/g)
Kappa Number
16
15 800
14 600
13
400
12
200
11
10 0
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
Fig. 1. Total and screened kraft yield, NaOH and Na2S consumptions and kappa number and viscosity of exploded kraft pulps pretreated with laccase-mediator system at
different laccase doses and treatment times. Exploded kraft pulp without any enzymatic treatment was used as control. Error bars show the standard deviation of three
replicas.
obtained with exploded control sample, and reduced Na2S and could be shortened from 60 to 25 min and a kraft pulp with even
NaOH consumption by 11.5% and 6.3%, respectively. The observed lower kappa number than that of control pulp was obtained.
increase in delignification (13.9%) was higher than the 8.7% re- Therefore, the pulp productivity increased and the subsequent
ported by Dyer and Ragauskas (2004) when laccase-HBT treatment bleaching process might consume fewer chemical reagents. As a
at 45 °C was carried out for 2 h prior to kraft pulping of Pinus taeda. result, steam explosion followed by LMS treatment seems to be a
This higher delignification could be due to small differences in the promising concept for selective delignification and chemical
conditions of the laccase treatment and/or to different behavior of savings.
hardwood and softwood, but more likely, it may be due to the
addition, in our case, of the steam explosion pre-treatment prior 3.2. Fungal treatment
to the enzymatic treatment. Such a treatment improves reagent
diffusion into the chips in kraft pulping (Ahvazi et al., 2007; In order to compare the results obtained with steam explosion
Martin-Sampedro et al., 2011) and enhances enzymatic hydrolysis plus LMS pre-treatment with those obtained using a more common
of biomass (Jurado et al., 2009; Ruiz et al., 2007; Wu et al., 1999). biopulping process, E. globulus chips that had not undergone steam
Furthermore, due to the reduction in kappa number as a conse- explosion treatment, but were extracted with acetone and/or
quence of only using a steam explosion treatment prior to kraft water were treated with the ligninolytic fungi, P. sanguineus and
pulping (67.8%), the total improvement in delignification including Trametes sp. I-62.
the enzymatic treatment (13.9%) rose by up to 72.4%. In others The effects of extractions without fungal inoculation on the
words, when both pre-treatment were combined, pulping time kraft process and pulp quality are shown in Table 2. When an ace-
R. Martín-Sampedro et al. / Bioresource Technology 102 (2011) 7183–7189 7187
14 0.39
12 0.37
10 0.35
8 0.33
6 0.31
4 0.29
2 0.27
0 0.25
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
14.0
2.0
12.0
Tensile Index (Nm/g)
6.0 1.0
4.0
0.5
2.0
0.0 0.0
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
99
42
Brightness (%ISO)
98
40
97
Opacity
38 96
95
36
94
34
93
32 92
1h 3h 5h 1h 3h 5h
3 UA/g 16.5 UA/g 30 UA/g Control 3 UA/g 16.5 UA/g 30 UA/g Control
Fig. 2. Mechanical and optical properties of exploded kraft pulps pretreated with laccase-mediator system at different laccase doses and treatment times: (a) Schopper–
Riegler degree, (b) apparent bulk density, (c) tensile index, (d) tear index, (e) brightness and (f) opacity. Exploded kraft pulp without any enzymatic treatment was used as
control. Error bars show the standard deviation of three replicas.
Table 2
Influence of different extractions on the pulp quality and on the kraft process.
Kappa number DKappa (%) DViscosity (%) Kraft yielda (%) DKraft yield Chemical Consumption
D%Na2S D%NaOH
NoE 15.6 49.9
AC 15.0 4.1b +12.1b 55.9 +6.0b 3.7b 1.9b
W 11.0 29.4b +5.9b 51.8 +1.9b +7.0b +0.3b
AC + W 11.3 27.6b +13.1b 51.9 +2.1b 0.4b 0.5b
a
Screened kraft yield in pulping process, without including pre-treatment.
b
Compared to control without any extraction (NoE).
tone extraction was carried out, the yield in kraft pulping increased after the kraft process was obtained. When an acetone extraction
while both kappa number and chemical consumption decreased. was carried out before water extraction, the reduction observed
However, the reduction in kappa number was higher when a water in kappa number was almost the same as in the case of only water
extraction was performed, as an increase in delignification of 29.4% extraction. These findings could be explained by improved liquor
7188 R. Martín-Sampedro et al. / Bioresource Technology 102 (2011) 7183–7189
Table 3
Influence of fungal treatment on the pulp quality and on the kraft process compared with control with the same extraction but without fungal inoculation.
Kappa number DKappa (%) DViscosity (%) Kraft yielda (%) DKraft yield Chemical Consumption
D%Na2S D%NaOH
NoE –P 15.8 +0.8b +2.1b 49.8 0.1b +2.6b +0.2b
NoE – T 14.2 9.3b +3.4b 48.4 1.5b +5.9b +0.5b
AC – P 14.5 3.7b 5.1b 51.9 4.0b +4.3b +1.7b
AC – T 13.6 9.5b 15.0b 53.3 2.6b +5.6b +2.4b
W–P 11.9 +7.9b +1.6b 50.0 1.8b 3.4b +0.3b
W–T 10.8 2.0b 1.8b 50.4 1.4b +0.5b +0.9b
AC + W – P 12.4 +9.0b 20.4b 51.6 0.3b +5.7b +1.2b
AC + W – T 12.5 +10.2b 3.0b 52.4 +0.5b +4.9b +1.4b
a
Screened kraft yield in pulping process, without including pre-treatment.
b
Compared to control with the same extraction but without fungal inoculation.
penetration and increased delignification during kraft pulping as a Phanerochaete chrysosporium with direct fungus treatment prior
result of the extractions (Ferraz et al., 2008). Furthermore, hydro- to thermomechanical pulping (TMP) and chemithermomechanical
lysis of polysaccharides, which takes place during hot water pulping (CTMP).
extraction, could facilitate delignification. Taking into account that during steam explosion pre-treatment,
A comparison of the outcomes of 28 days fungal treatments of most of the hemicelluloses are removed before pulping, the overall
extracted and non-extracted chips is provided in Table 3. Small dif- pulp yield of the process (including pre-treatments and kraft pul-
ferences were found in screened kraft yield during the pulping of ping) would be reduced by about 40%, but pulp production of a
biotreated chips compared to their controls; however, when hypothetic mill could be maintained at a similar level because of
weight losses during the fungal treatment were taking into ac- the reduction in pulping time. Moreover, the hemicelluloses are re-
count, the overall yields for biotreated pulps were around 5% lower moved without excessive degradation and could be converted into
than those of control pulps due to fungal degradation of polysac- value-added products, increasing the total revenue stream for the
charides. Nevertheless, in most cases, biotreated pulps showed a pulp industry and converting the pulp mill into an integrated forest
lower kappa number than their controls. bio-refinery (Martin-Sampedro et al., 2011; Van Heiningen, 2006).
NaOH and particularly Na2S consumptions were higher for bio- This potential could be one of the main advantages of a combined
treated samples compared to their controls. Despite these disad- steam explosion and LMS treatment compared to fungal treatment,
vantages, biotreatment with Trametes sp. I-62 promoted the in which polysaccharides are degraded.
pulping process, since this pre-treatment produces pulps with a
lower kappa number than those of control pulps in all cases, apart
4. Conclusions
from the AC + W samples. In contrast, P. sanguineus treatment did
not result in an increase in delignification under the assay condi-
Steam explosion followed by LMS treatment seems to be a
tions, contrary to what was reported by Wolfaardt et al. (2004).
promising pretreatment for increasing the delignification rate in
The reason could be differences in the P. sanguineus strains. These
the biopulping process since pulping times could be shortened
differences have been observed previously in other white rot fungi
by 60%, and most of the hemicelluloses could be extracted before
(Akhtar et al., 1992; Blanchette et al., 1992). Biotreated pulps
pulping and converted into a value-added product. The optimal
showed a viscosity similar to that of their controls in most of the
conditions assayed for LMS were 3 UA/g and 3 h of treatment at
cases.
40 °C. Compared with fungal treatment, lower kappa number and
When comparing biotreated and controls pulps that had under-
chemical consumptions were observed, and less polysaccharides
gone the same extraction process, the highest reductions in kappa
degradation took places as shown by better kraft yields.
number due to fungal treatment were observed in samples sub-
jected to acetone extraction and without any extractions. In these
samples, fungal colonization and laccase activity were also higher Acknowledgements
(data no shown). Overall, the lowest kappa number was observed
in pulp obtained from chips pre-extracted with hot water and trea- The authors wish to thank the Spanish Ministry of Science and
ted with Trametes sp. I-62. Innovation for funding this study via the Projects CTQ 2006-00385
and CTQ 2009-11268 and the FPI fellowship (BES2007 – 14528).
Ahvazi, B., Radiotis, T., Bouchard, J., Goel, K., 2007. Chemical pulping of steam-
The most efficient biotreatment assayed in this study consisted exploded mixed hardwood chips. J. Wood Chem. Technol. 27 (2), 49–63.
of a steam explosion treatment followed by a LMS treatment at Akhtar, M., Attridge, M.C., Myers, G.C., Kirk, T.K., Blanchette, R.A., 1992.
40 °C for 3 h with a laccase dose of 3 UA/g, as can be concluded Biomechanical pulping of loblolly pine with different strains of the white-rot
fungus, Ceriporiopsis subvermispora. Tappi J. 75 (2), 105–109.
from data presented in Section 3.1.2. This enzymatic pretreatment Bajpai, P., Bajpai, P.K., Akhtar, M., Jauhari, M.B., 2001. Biokraft pulping of Eucalyptus
with LMS resulted in a lower kappa number and chemical con- with selected lignin-degrading fungi. J. Pulp Paper Sci. 27 (7), 235–239.
sumption than steam explosion and xylanase pre-treatment in pre- Blanchette, R.A., Burnes, T.A., Eerdmans, M.M., Akhtar, M., 1992. Evaluating isolates
of Phanerochaete chrysosporium and Ceriporiopsis subvermispora for use in
liminary studies and in studies reported by Zhao et al. (2002) after
biological pulping processes. Holzforschung 46, 109–115.
soda pulping of xylanase pretreated wheat straw. Burton, S.W., 2001. Low energy thermomechanical pulpin process using an enzyme
Steam explosion plus enzymatic treatment was more efficient treatment between refining zones. Patent US 6267841.
in term of treatment times, delignification, chemical savings and Dyer, T.J., Ragauskas, A.J., 2004. Laccase: a harbinger to kraft pulping. ACS Sym. Ser.
889, 339–362.
pulp yield. Similar results were found by Ramos et al. (2004) Eugenio, M.E., Santos, S.M., Carbajo, J.M., Martín, J.A., Martín-Sampedro, R.,
who compared an enzymatic treatment using crude extracts of González, A.E., Villar, J.C., 2010. Kraft pulp biobleaching using an extracellular
R. Martín-Sampedro et al. / Bioresource Technology 102 (2011) 7183–7189 7189
enzymatic fluid produced by Pycnoporus sanguineus. Bioresour. Technol. 101, Messner, K., Srebotnik, E., 1994. Biopulping: an overview of developments in an
1866–1870. environmentally safe paper-making technology. FEMS Microbiol. Rev. 13, 351–362.
Ferraz, A., Guerra, A., Mendonça, R., Masarin, F., Vicentim, M.P., Aguir, A., Pavan, P.C., Oriaran, T.P., Labosky, P., Blankenhorn, P.R., 1990. Kraft pulp and papermaking
2008. Technological advances and mechanistic basis for fungal biopulping. properties of Phanerochaete chrysosporium degraded aspen. Tappi J. 73,
Enzyme Microb. Tech. 43, 178–185. 147–152.
Grethlein, H.E., 1985. The effect of pore size distribution on the rate of enzymatic Pere, J., Ellmen, J., Viikari, L., 2005. Process for preparing mechanical pulp. Patent
hydrolysis of cellulosic substrates. Nat. Biotechnol. 3, 155–160. WO/2005/056915.
Islam, M.N., Karim, M.R., Malinen, R.O., 2008. Beneficial effects of fungal treatment Petit-Conil, M., Semar, S., Niku-Paavola, M.-L., Sigoillot, J.C., Asther, M., Anke, H.,
before pulping and bleaching of Acacia mangium and Eucalyptus camaldulensis. Viikari, L., 2002. Potential of laccases in softwood-hardwood high-yield pulping
Turk. J. Agric. For. 32, 331–338. and bleaching. Prog. Biotechnol. 21, 61–71.
Jacobs-Young, C., Heitmann, J.A., Venditti, R.A., 1998. Conventional kraft pulping Ramos, J., Rojas, T., Navarro, F., Davalos, F., Sanjuan, R., Rutiaga, J., Young, R.A., 2004.
using enzyme pre-treatment technology: role of chip thickness, specie and Enzymatic and fungal treatments on sugarcane bagasse for the production of
enzyme combinations. AIChE Annual Meeting on the Innovative Advances in mechanical pulps. J. Agri. Food Chem. 52, 5057–5062.
the Forset Products Industries. Los Angeles (EEUU). 94 (319), 1–15. Ruiz, E., Cara, C., Manzanares, P., Ballesteros, M., Castro, E., 2007. Evaluation of
Jurado, M., Prieto, A., Martínez-Alcalá, A., Martínez, A.T., Martínez, M.J., 2009. steam explosion pre-treatment for enzymatic hydrolysis of sunflower stalks.
Laccase detoxification of steam-exploded wheat Straw for second generation Enzyme Microb. Tech. 42, 160–166.
bioethanol. Bioresour. Technol. 100, 6378–6384. Van Heiningen, A., 2006. Converting a kraft pulp mill into an integrated biorefinery.
Maijala, P., Kleen, M., Westin, C., Poppius-Levlin, K., Herranen, K., Lehto, J.H., Pulp Pap. Can. 107 (6), 141–146.
Reponen, P., Mäentausta, O., Mettälä, A., Hatakka, A., 2008. Biomechanical Wolfaardt, F., Taljaard, J.L., Jacobs, A., Male, J.R., Rabie, C.J., 2004. Assessment of
pulping of softwood with enzymes and white-rot fungus Physisporinus rivulosus. wood-inhabiting Basidiomycetes for biokraft pulping of softwood chips.
Enzyme Microb. Tech. 43, 169–177. Bioresour. Technol. 95, 25–30.
Martin-Sampedro, R., Eugenio, M.E., Revilla, E., Martin, J.A., Villar, J.C., 2011. Wu, M.M., Chang, K., Gregg, D.J., Boussaid, A., Beatson, R.P., Saddler, J.N., 1999.
Integration of kraft pulping on a forest biorefinery by the addition of a steam Optimization of steam explosion to enhance hemicelluloses recovery and
explosion pretreatment. Bioresources 6 (1), 513–528. enzymatic hydrolysis of cellulose in softwoods. Appl. Microbiol. Biot. 77–79,
Mendonça, R., Jara, J.F., González, V., Elissetche, J.P., Freer, J., 2008. Evaluation of the 47–54.
White-rot fungi Ganoderma australe and Ceriporiopsis subvermispora in Zhao, J., Li, X., Qu, Y., Gao, P., 2002. Xylanase pretreatment leads to enhanced soda
biotechnological applications. J. Ind. Microbiol. Biotechnol. 35, 1323–1330. pulping of wheat Straw. Enzyme Microb. Tech. 30, 734–740.
PUBLICACIÓN VII:
Effect of steam explosion and enzymatic pre-treatments on pulping and bleaching
of Hesperaloe funifera.
Manuscrito
263
a
Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria INIA.
Carretera de la Coruña, km 7.5, 28040, Madrid (Spain)
Abstract
A non-wood raw material with high potential for pulp and paper applications
(Hesperaloe funifera) was subjected to a steam explosion pre-treatment, and the
subsequent effect of this pretreatment on biopulping and biobleaching was studied. An
increase in the delignification rate, bigger than that reported for autohydrolysis and acid
hydrolysis pre-treatments, and a reduction in chemical consumption were found during
kraft pulping of the exploded samples. However, biopulping with the laccase-mediator
system (LMS) did not lead to a reduction in the kappa number in either non-exploded or
exploded unbleached pulps. On the other hand, the steam explosion pretreatment
boosted the advantages of the LMS pre-treatment (decrease in kappa number and
increase in brightness) favored biobleaching, with a 53.1% delignification rate and a
final brightness of 67 % ISO. Finally, the steam explosion pre-treatment also improved
the color properties of the bleached pulp and reduced the hydrogen peroxide
consumption by 24.6%.
Keywords:
265
1. Introduction
One of the main goals of research in pulp and paper industry is to achieve a
sustainable development by using new eco-friendly technologies that improve pulping
and bleaching processes, optimize energy and chemical consumption and ensure an
appropriate use of resources. This objective could be reached by using biotechnology
(biopulping and biobleaching) and new raw materials such as agricultural residues and
non-woody plants rather than wood fiber. The advantages of the use of non-wood raw
materials include: 1) the added value given to agrifood crops by exploiting their
residues; 2) the availability of these materials in countries with limited forest resources;
3) the ability to obtain large amounts of special paper by virtue of the wide variability in
morphology and chemical composition of non-wood fibers; and 4) the reduced
consumption of chemicals and energy in cooking and bleaching operations by effect of
the more porous and readily accessible tissue structure of non-wood raw materials
(Alaejos et al. 2004; Sanchez et al. 2011). These advantages have brought about a
14.6% increase in the production of cellulose pulp from non-wood raw material in the
period 2000-2009, while the production of cellulose pulp from wood has decreased by
6.9% in the same period. Despite this fact, in 2009, the pulp production from non-wood
materials represented only an 8.4% of the total production in the same year (FAO
2011), probably due to the high ash and extractive contents of these materials, which
contribute to a low pulping yield. Other disadvantages of non-wood raw materials are:
low density, which complicates transportation; the need of specific environmental
growing conditions; and problems derived from a very short harvest period.
Some authors have explored the potential of H. funifera as raw material for pulp
production (Fairbank and Detrick 2000; McLaughlin 2000; Revilla et al. 2010; Sanchez
et al. 2010a; Sanchez et al. 2010b; Wong and McLaughlin 2000), however only
266
Sanchez et al. (2011) have evaluated the effect of some pretreatments, such as acid
hydrolysis and autohydrolysis, on pulping of this non-wood raw material.
On the other hand, the use of biological pre-treatments to boost pulping and
bleaching processes has attracted considerable interest in the last two decades. Most
studies of biopulping have examined the direct application of fungi to chips. However,
this approach has many disadvantages for industrial use, such as slow fungal growth,
the difficulty of keeping aseptic and optimal conditions during the long treatment
required, and yield loss due to the degradation of polysaccharides simultaneous to lignin
degradation. In an attempt to overcome these issues, the direct application of enzymes,
such a xylanases, laccases or manganese peroxidases, to wood chips has been suggested
as a more attractive alternative for industrial purposes. Nevertheless, in view of the fact
that penetration of the wood chips by these enzymes is rather difficult, researchers have
recommended pre-treating the chips by means of defribration (Burton 2001), vacuum
(Jacobs-Young et al. 1998), compression (Maijala et al. 2008; Pere et al. 2005) or steam
explosion (Martin-Sampedro 2011c) in order to open their structure. An alternative
option is to use a raw material more readily accessible, with a more porous structure,
such as non-wood materials (Qin et al. 2006; Ramos et al. 2004; Zhao et al. 2002).
Although, as mentioned above, some studies have tested the biopulping of a non-wood
267
raw material, we found no studies reporting on the use of H. funifera with (or without) a
steam explosion pre-treatment.
In order to boost the bleaching process, many researchers have pretreated pulps
with enzymes, mainly with the laccase-mediator system (LMS), before applying the
standard bleaching sequences (Viikari et al. 1986; Moldes et al. 2008; Eugenio et al.
2010). These biobleaching studies have shown that enzymatically pretreated pulps
require a smaller amount of chemicals in subsequent bleaching processes, maintain or
increase the pulp quality, and reduce the pollutant load of the effluents. According to
Martin-Sampedro et al. (2011b), the combination of LMS and steam explosion takes on
the advantages of both pretreatments when Eucalyptus globulus is used as raw material,
providing pulps with minimum kappa number and optimum optical properties and
consuming 20% less hydrogen peroxide compared to control pulps.
Building on this base of evidence, and keeping in mind that our overall goal as
researchers is to develop new eco-friendly technologies that save energy (reducing the
pulping time) and chemicals and ensure an appropriate utilization of resources, our
study was designed to evaluate how steam explosion impacts both the biopulping and
the biobleaching of H. funifera. In this work we used a non-wood raw material with
potential pulp and paper applications and combined traditional pulp processes with
enzymatic treatments that, boosted by the steam explosion pre-treatment, could increase
their benefits and the viability of their industrial application.
All chemicals used were reagent-grade and were supplied by Merck (Barcelona,
Spain), Panreac (Barcelona, Spain) or Sigma–Aldrich (Madrid, Spain), except for 2,2´-
Azino-bis-3-ethylbenzthiazoline-6-sulphonate (ABTS), which was purchased from
Roche (Madrid, Spain).
268
The severity factor of each treatment was calculated using this equation inferred
by Overend and Chornet (1987), in which T is the temperature (ºC) and t the duration of
the treatment (min).
. (1)
The water retention, or hydration capacity, of the chips was calculated as the
weight of water absorbed by the chips after being immersed for 6 hours, and it was
expressed as grams of water per 100 grams of dry sample.
269
The laccase used was Novozym 51003. The enzyme dose was expressed as units
of laccase activity per gram of dry wood (UA/g), where 1 U of laccase activity is
defined as the amount of laccase required to convert 1μmol / min of ABTS to its
cationic radical ( =436nm, o=49600 M-1cm-1, 0.1M phosphate buffer, pH 6 at 24 ºC).
After each batch, the reactors were removed from the vessel, and the cooked
chips were washed, disintegrated and screened to determine their reject content and
270
screened and total yields. Black liquors were titrated with HCl to determine the
chemical reagents consumption. The kappa number (ISO 302) and viscosity (ISO 5351-
2) were determined in the pulp samples.
Novozym 51003 was the laccase used in these assays and the natural mediator
was acetosyringone. This natural mediator was selected for the biobleaching pre-
treatment based on Eugenio et al. (2010), who evaluated acetosyringone and HBT as
mediators in biobleaching of E. globulus and determined acetosyringone to be the most
efficient.
All the enzymatic assays were performed in a similar way to that described in
the 2.3.1. section but using the following operational conditions: 50 g of non-exploded
or exploded kraft pulps, 40ºC, 2 hours, pH 6 (phosphate buffer, 100 mM), 10%
consistency, 17.5 UA/ g dry pulp, 1.5% acetosyringone, 6 Kg/cm2 of oxygen pressure
and a few drops of 0.05% Tween 80. Controls without the addition of laccase nor
mediator in the enzymatic treatment were performed (control-buffer, named as -Lc).
2.4.2. Alkaline extraction (E) and hydrogen peroxide (P) bleaching stages
Conditions for the alkaline extraction were: 1.5 % NaOH, 5% consistency and
90 ºC temperature for 120 min. Thereafter, pulps were washed, and a hydrogen peroxide
bleaching stage was applied under the following conditions: 2 % H2O2, 1.5 % NaOH, 1
% DTPA, 0.2 % Mg2SO4, 5 % consistency and 90 ºC for 90 min. Residual hydrogen
peroxide was titrated in the bleaching effluent.
271
Unbleached pulps and pulps obtained after each stage of the bleaching sequence
(LEP) were characterized in terms of their kappa number, viscosity, brightness and CIE
L*a*b* and CIE L*C* coordinates, in conformity with the ISO 302, ISO 5351-2, ISO
2470 and T 527 standards, respectively.
High contents of ash, ethanol extractives and hot water extractives were
measured in H. funifera (5.2%, 3.1% and 15.9% respectively). These large amounts of
extractives indicate a reduction in pulping yield. However, the lower lignin content
detected for this material (9.4% klason lignin and 3.6% acid soluble lignin), compared
to other non-wood and wood materials, could make the pulping easier and compensate
for the expected low yield.
The biometric analysis shows that H. funifera has longer and thinner fibers (4.2
mm long and 7.2 μm wide) than other non-wood materials, such as cotton linters (3.5
mm and 21 μm), sisal (3.0 mm and 17 μm) or wheat straw (1.5 mm and 13 μm)
(Fairbank and Detrick 2000), and also compared to woody species such as Picea abies
(3.5 mm and 36 μm), Pinus taeda (4.0 mm and 43 μm) or Eucalyptus saligna (1.0 mm
and 19 μm) (McLaughlin and Schuck 1991). H. funifera’s relatively significant
slenderness (L/D ratio), compared to other raw materials, could provide more inter-fiber
bonding and increased paper strength.
H. funifera samples were steam exploded under two different sets of conditions;
and the most severe treatment resulted in a reduction of the yield (Table 1). These
results are consistent with those found by other authors (Al-Dajani et al. 2009; Li et al.
2010), who report that the amount of hemicelluloses dissolved in the liquid fraction was
larger with longer treatments, decreasing the yield of the process. The yield obtained for
H. funifera when the SE2 treatment was carried out was lower than that obtained when
272
E. globulus was subjected to the same steam explosion treatment (62.5% vs 79.3%)
(Martin-Sampedro et al. 2011a). This finding is likely to be the result of a higher
content of hot water extractives when using H. funifera (15.9% vs 1.8% with E.
globulus).
The water retention capacity could provide information about changes in the
internal structure of the material and correlates well with its accessibility to chemicals.
As it can be observed in Table 1, the non-exploded H. funifera sample showed a high
water retention capacity (267g of water per 100g of dry sample). If this value is
compared with those reported for E. globulus in a previous study (Martin-Sampedro et
al. 2011a), a big difference is evident: 65g of water per 100g of non-exploded chips and
272g of water per 100 g of wood after the same steam explosion treatment as the one
used in SE2. This finding suggests that the accessibility of non-exploded H. funifera is
similar to that of exploded E. globulus after a pre-extraction and 2 cycles of steam
explosion, each 10 and 3 minutes long (183ºC). These results can be attributed to the
more porous and readily accessible tissue structure of non-wood raw materials (Alaejos
et al. 2004; Sanchez et al. 2011).
3.3. Biopulping
Non-exploded (HP) and exploded (SE1 and SE2) H. funifera samples were
subjected to a biopulping process consisting of a laccase mediator system (LMS)
treatment followed by kraft pulping.
Figure 1 shows the total and the screened yield obtained during the kraft pulping
of each sample. Both yields were similar because all pulps had low reject contents
(around 1%). Thus, the removal of extractives during the steam explosion pre-treatment
increased the relative cellulose content before pulping and brought about an increase in
both yields in the exploded samples (SE1 and SE2). However, if the yield of the steam
explosion pre-treatment is taken into account, the overall yield for HP, SE1 and SE2
273
was 37%, 36% and 29%, respectively (including both the kraft yield showed in figure 1
and the steam explosion yield). No other significant differences have been observed in
kraft yields.
The quality of the obtained pulps can be evaluated from Figure 2, which shows
the kappa number and the viscosity of the pulps. A reduction in the kappa number,
compared to non-exploded pulps, was evident when the steam explosion was performed
prior to kraft pulping. Moreover, this reduction was accentuated when the severity of
the treatment increased (17.8% and 28.4% for SE1 and SE2 respectively). Therefore,
the exploded material cooked faster and more efficiently than the untreated material, so
that the cooking time or the cooking temperature could be reduced obtaining a pulp of
similar characteristics (kappa number and viscosity) and reducing the energy
consumption. However, the kappa number reduction with the SE2 pre-treatment using
E. globulus (67.8%) (Martin-Sampedro et al. 2011c), was far greater than that obtained
for H. funifera. This finding could be explained considering that non-exploded H.
funifera already has a readily accessible tissue structure (as the water retention capacity
showed) and an increase in this accessibility did not mean an improvement in reagent
diffusion as that of E. globulus. Nevertheless, pre-treatment of H. funifera by means of
steam explosion was associated to more delignification during kraft pulping than
reported by Sanchez et al. (2011), who tested other pre-treatments, such as
autohydrolysis and acid hydrolysis, preceding soda-anthraquinone pulping.
On the other hand, a similar reduction in the kappa number was observed when
non-exploded H. funifera was subjected to enzymatic treatment. This reduction was a
consequence of the buffer treatment at 40ºC and not a result of the LMS effect, since no
significant differences in delignification were observed when the LMS was
incorporated. Therefore, the increase in delignification that accompanied the use of
steam explosion or enzymatic pre-treatments is likely to be caused by the removal of
extractives during steam/water treatment at high or moderate temperatures (183ºC
during few minutes in SE or 40ºC during 5 hours in enzymatic treatment). When both
treatments were performed sequentially, the delignification observed after kraft pulping
reached even higher levels, with the more intensive steam explosion leading to more
delignification. However, as it has been mentioned above, delignification levels with the
addition of LMS in the enzymatic treatment were not higher than those observed when
only buffer was added. Furthermore, the addition of LMS in non-exploded samples
274
(HP) caused an increase in the kappa number, versus control-buffer. This observation
could be evidence of grafting of the LMS onto the pulp fiber (Dyer and Ragauskas
2004). Improving the delignification rate with the LMS pre-treatment requires further
experiments to determine the optimal operational conditions (laccase dose, treatment
time, etc.)
The chemical consumption during kraft pulping was evaluated by measuring the
residual active alkali (AA) and sulfidity (S) in the black liquors (Table 2). The use of
the steam explosion pretreatment was associated to an increase in the residual AA (4.0,
4.6 and 5.4 for HP, SE1 and SE2 control samples respectively), which means that
chemical consumption decreases (NaOH and Na2S) as the severity of the steam
explosion increases. This is likely the result of the removal of hemicelluloses, which
during the kraft pulping process are degraded to carboxylic acids that consume
significant amounts of alkali (San Martin et al. 1995). On the other hand, the addition of
a LMS treatment increased the consumption of Na2S (reduction in residual sulfidity: 17,
17.4 and 18.0 for HP, SE1 and SE2 laccase samples, compared to around 23 for control
and control-buffer samples). A similar effect has been observed by Martin-Sampedro et
al. (2011c), who introduced SE and LMS pre-treatments with a high laccase dose (30
UA/g) before kraft pulping of E. globulus. However, the residual AA showed that the
total chemical consumption did not varied significantly with the addition of the LMS
treatment (Table 2).
3.4. Biobleaching
275
peroxide bleaching), was applied to the exploded and non-exploded H. funifera kraft
pulps.
In Figure 3, the kappa number and viscosity of the pulps after each stage of the
LEP bleaching sequence are showed. For all pulps assayed, a reduction of the kappa
number was observed after each stage, while the viscosity remained approximately
constant during the whole bleaching process. Furthermore, in all pulps, the LMS
treatment resulted in an even higher delignification rate without decreasing the pulp
viscosity. This finding reflects a significant biodelignification of H. funifera without
degradation of carbohydrate chains, which is consistent with results reported by other
authors that used eucalyptus (Eugenio et al. 2010; Oudia et al. 2008) and flax fibers
(Fillat et al. 2010).
276
all the H. funifera pulps assayed. Similar advantages of LMS treatments have been
found by other authors using different raw materials (Eugenio et al. 2010; Fillat et al.
2010; Vivekanand et al. 2008).
277
68.1% for HP-L and SE2-L respectively). A possible explanation of this effect is that
the steam explosion removes substances that compete with lignin for the hydrogen
peroxide. This effect has already been observed by Martin-Sampedro et al. (2011b) in
the biobleaching of E. globulus kraft pulp and it has been attributed to the low
hexenuronic acid content of exploded pulps. However, when a LMS pretreatment was
added to the bleaching sequence, slightly more hydrogen peroxide was consumed with
all H. funifera pulps, in contrast to what has been reported by other authors who have
used wood raw materials (Eugenio et al. 2010; Martin-Sampedro et al. 2011b; Valls et
al. 2009). Nevertheless, it should be taken into account that, in these studies, the LMS
pre-treated pulps had lower kappa numbers and higher brightness, and therefore, if
compared at the same delignification and brightness level, the chemical consumption of
the LMS pre-treated samples would probably be lower than that of the control pulps.
4. Conclusions
The steam explosion pre-treatment increased the delignification rate during the
subsequent kraft pulping of H. funifera, providing kraft pulps with lower kappa
numbers (or alternatively reducing the cooking time) and higher viscosity than those
reported with autohydrolysis and acid hydrolysis pre-treatments. Furthermore, a
reduction in the chemical consumption was observed. The addition of a LMS pre-
treatment before kraft pulping did not improve delignification in the assayed conditions.
However, the addition of a LMS pre-treatment as the first stage in the bleaching
sequence resulted in smaller kappa numbers and an increase in brightness in bleached
pulps. These improvements in biobleaching were boosted when a steam explosion pre-
treatment preceded the kraft pulping, reducing also the color of the pulp and
consumption of hydrogen peroxide during the bleaching process.
Acknowledgements
The authors wish to thank the Spanish Ministry of Science and Innovation for
funding this study via the Projects CTQ 2006-00385 and CTQ 2009-11268 and the FPI
fellowship (BES2007 – 14528).
278
References
Alaejos J, López F, Eugenio ME, Fernández M (2004) Non-wood raw materials for
pulp and paper making. A review. Afinidad LXI, 513:400-410
Eugenio ME, Santos SM, Carbajo JM, Martín JA, Martín-Sampedro R, González AE,
Villar JC (2010) Kraft pulp biobleaching using an extracellular enzymatic fluid
produced by Pycnoporus sanguineus. Bioresour Technol 101:1866–1870
Dyer TJ, Ragauskas AJ (2004) Laccase: a harbinger to kraft pulping. ACS Sym. Ser.
889:339-362
Fillat A, Colom JF, Vidal T (2010) A new approach to the biobleaching of flax pulp
with laccase using natural mediators. Bioresour Technol 101:4104-4110
279
Pere J, Ellmen J, Viikari L (2005) Process for preparing mechanical pulp. Patent
WO/2005/056915
280
Ragauskas AJ, Nagy M, Kim DH, Eckert CA, Hallett JP, Liotta CL (2006) From wood
to fuels: integrating biofuels and pulp production. Indus Biotech 2(1): 55-65
Revilla E, Eugenio ME, Martin-Sampedro R, Villar JC (2010) Using wood and non-
wood fibers as a reinforcement to produce recycled paper from newspapers.
Proceedings of VI Iberoamerican Congress on Pulp and Paper Research
(CIADICYP), 329
San Martin R, Perez D, Briones R (1995) Simultaneous production of etanol and kraft
pulp from pine (Pinus radiata) using steam explosion. Bioresource Technol
53:217-223
Sanchez R, Rodriguez A, Requejo A, Ferrer A, Navarro E (2010b) Soda pulp and fuel
gases synthesis from Hesperatloe funifera. Bioresour Technol 101:7032-7040
Valls C, Molina S, Vidal T, del Río JC, Colom JF, Martínez AT, Gutiérrez A, Roncero
MB (2009) Influence of operation conditions on laccase-mediator removal of
sterols from eucalypt pulp. Process Biochem 44:1032–1038
Van Heiningen A (2006) Converting a kraft pulp mill into an integrated biorefinery.
Pulp Pap Can. 107(6):141-146
281
282
Figure Caption
Fig. 1 (a) Total and (b) screened kraft yields for non-exploded (HP) and exploded (SE1
and SE2) pulps after LMS pre-treatment (laccase); compared with a control without the
addition of LMS in the enzymatic treatment (control-buffer) and a control without the
enzymatic treatment step (control)
Fig. 2 (a) Kappa number and (b) viscosity of non-exploded (HP) and exploded (SE1 and
SE2) pulps after LMS pre-treatment (laccase); compared with a control without the
addition of LMS in the enzymatic treatment (control-buffer) and a control without the
enzymatic treatment step (control)
Fig. 3 (a) Kappa number and (b) viscosity after biobleaching of non-exploded (HP) and
exploded (SE1 and SE2) pulps. Controls without addition of LMS (-Lc) were included.
The biobleaching sequence consisted of laccase stage (L), alkaline extraction stage (E)
and hydrogen peroxide stage (P)
Fig. 4 Optical properties after biobleaching of non-exploded (HP) and exploded (SE1
and SE2) pulps: (a) brightness evolution during the bleaching sequence; where L:
laccase stage, E: alkaline extraction stage and P: hydrogen peroxide stage; (b) CIE
L*a*b* coordinates and (c) CIE L*C*coordinates for unbleached (black symbols) and
biobleached (grey symbols) pulps
283
Figure 1
a) b)
55% 50% 46%
45% 45%
50% 47% 47% 44% 44% 44%
46% 45%
45% 45% 40%
44%
45%
40% 40%
36% 36%
40% 37% 37%
35%
35%
30%
30%
25% 25%
20% 20%
HP SE1 SE2 HP SE1 SE2
Figure 2
a) b)
16 1400
14,1
1200 113411141124
14 1064
10101020
11,6
Kappa Number
1000
11,1
798 832 817
Viscosity (ml/g)
12
10,3 10,4
10,1 800
9,6
10
600
7,9 8,2
8
400
6 200
4 0
HP SE1 SE2 HP SE1 SE2
284
Figure 3
a)
16
14,1 14,1
14
12,5
12,1 11,6
12 11,611,2
11,2
10,2 9,8 10,1 10,1
9,6 9,5
Kappa Number
10 8,9
8,1 8,3 8,3
7,7
8 7,1
6,4 6,3
6 5,2
4,7
0
HP-Lc HP-L SE1-Lc SE1-L SE2-Lc SE2-L
Unbleached L LE LEP
b)
1400
600
400
200
0
HP-Lc HP-L SE1-Lc SE1-L SE2-Lc SE2-L
Unbleached L LE LEP
285
Figure 4
a)
80
67 68
70
60 61 62
58
60
53
Brightness (% ISO)
51
50 45 46 46
45 43 43 44 45 44 43
40 42
40 36 36 36 36
30
20
10
0
HP-Lc HP-L SE1-Lc SE1-L SE2-Lc SE2-L
Unbleached L LE LEP
b) c)
13 95
90
12
HP HP
85
HP-LEP HP-LEP
b* 11 L*
SE1 SE1
SE1-LEP 80 SE1-LEP
SE2 SE2
10
SE2-LEP 75 SE2-LEP
9 70
-1 0 1 2 3 4 10 11 12 13
a* C*
286
Table 1. Steam explosion conditions, process yields and water retention capacity of the
samples.
HP SE1 SE2
Pre-extraction No Yes Yes
1st cycle (min) - 3 10
2nd cycle (min) - - 3
Severity S0 - 2.92 3.56
Yield (%) - 81.9 62.5
Water retentiona 267 359 434
a
Water retention capacity as grams of water per 100 grams of dry sample
Table 2. Residual active alkali (AA) and sulfidity (S) in the black liquors.
Sample AA S
HP - Control 4.0 23.1
HP - Control-Buffer 4.9 24.4
HP - Laccase 4.7 17.0
SE1 - Control 4.6 25.0
SE1 - Control-Buffer 3.9 20.3
SE1 - Laccase 3.8 17.4
SE2 - Control 5.4 19.8
SE2 - Control-Buffer 4.5 22.1
SE2 - Laccase 5.1 18.0
287
288