This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
AbstractMicrowave Assisted Extraction of Xylan Doctor of Philosophy, 2014 Fathimathul Suhara Panthapulakkal Graduate Department of Chemical Engineering and Applied Chemistry University of Toronto
Xylan is one of the major hemicelluloses present in plant cell wall matrix, where it is closely
associated with other cell wall components, cellulose and lignin. Xylan has enormous potential
as a renewable biopolymer and recently, research in the direction of isolation and utilization of
xylan is gaining lot of research attention. Extraction of xylan from the plant cell walls involves
the hydrolysis of xylan and its transfer from the plant cell wall matrix to the hydrolyzing media.
Current process of extraction involves prolonged heating of the biomass with the hydrolysis
media at high temperature and/or pressure that leads to molecular degradation of xylan and
limits its high potential polymeric applications. In this research, microwave assisted alkaline
extraction of polymeric xylan from birch wood is investigated as an alternative to the time
intensive conventional extraction. The hypothesis to be tested is that the microwave’s selective
heating ability leads to the generation of hot spots through its interaction with the alkali present
in the fibers and the resulting "explosion effect" loosen the recalcitrant fiber structure network
thereby facilitating the hydrolysis of xylan and its dissolution before undergoing significant
degradation. Effect of microwave extraction on the yield of xylan and wood solubilization,
physico-chemical properties of wood fibers and of isolated xylan were investigated in
comparison with conventional extraction. Low power input microwave (110 W) alkaline
extraction was found to be an efficient alternative to the conventional extraction. FTIR and
chemical composition of wood fibers after extraction demonstrated an increased removal of
xylan from the wood fibre using microwave extraction. SEM, X-ray microtomography, and X-
iii
ray crystallinity studies of wood fibers demonstrated a porous and loosened fibre structure after
microwave extraction confirming the hypothesis. Molecular weight of the isolated xylan using
microwave extraction was found to be higher compared to the xylan isolated using conventional
extraction indicating less molecular degradation. About 75% of xylan present in birch wood
could be extracted using a low power input microwave heating under optimized extraction
conditions of 8wt% NaOH solution, 1:8 (g:mL) solid to liquid ratio, and 25 minutes of
extraction time.
iv
ACKNOWLEDGEMENT
It is my utmost pleasure to thank the many people who made this thesis possible.
I would like to express my sincere gratitude to my advisor Prof. Mohini M Sain for his
continuous guidance, generous support and valuable advice throughout my research.
I am deeply grateful to Prof. Donald Kirk, Department of Chemical Engineering and Applied
Chemistry, University of Toronto, and Professor D. N. Roy, Faculty of Forestry, University of
Toronto for their valuable suggestions, and constructive support throughout this work.
I wish to express my warm and sincere thanks to my fellow students and other department staff
at the Faculty of Forestry and Department of Chemical Engineering and Applied Chemistry,
with a special mention to Mr. S. Law, Dr. S. Kunar, and Dr. O. Faruk for their help and
support during this research.
A special note to my husband Sreekumar and my kids Abhinav, Tejas and Rithik - Thank you
for your constant support and encouragement and with so much of happiness let me say that
this effort is dedicated to you all.
Finally, to my parents, your love and thoughts have been my strength and inspiration and would
always remain.
Thank you all.
v
Table of Content
Abstract ......................................................................................................................................................... ii
List of Tables.............................................................................................................................................viii
List of Figures..............................................................................................................................................x
2.2 Research Objectives ............................................................................................................................ 30
Chapter 3 Materials and Methods ....................................................................................................... 31
6.2 Material ............................................................................................................................................ 109
Table 1.1 Cellulose, hemicellulose and lignin content of woody biomass compiled from
different authors
Table 1.2 Wood hemicellulose characteristics
Table 1.3 Available literature on microwave assisted extraction of polysaccharides from
various biomass
Table 4.1 Chemical composition of extractive-free birch wood fibers
Table 4.2 Experimental conditions used for microwave extraction
Table 4.3 Material balance analysis of wood fibers after microwave assisted extraction
Table 4.4 Chemical composition of xylan extracted using different experimental conditions
Table 4.5 Molecular mass distribution of xylan extracted using different experimental
conditions
Table 5.1 Experimental conditions used for microwave and conventional extraction
Table 5.2 Time-temperature combinations used in the microwave and conventional
extractions
Table 5.3 Temperature of the slurry after different duration of microwave extractions
Table 5.4 Mass balance analysis of wood fibers after xylan extraction
Table 5.5 Effect of alkaline extraction on the chemical composition of birch wood fibers
Table 5.6 Peak assignment of wood fibers before and after different extraction processes
Table 5.7 Crystallinity index of wood fibers before and after different extraction processes
Table 5.8 Chemical composition of xylan isolated under different extraction conditions
Table 5.9 Intrinsic viscosity, viscosity average molecular weight and number average
degree of polymerization of xylan obtained by microwave convetnional extraction
Table 6.1 Experimental conditions used for extraction
Table 6.2 Porosity of wood samples after different extraction methods (sample size > 2 mm)
Table 6.3 Calculated porosities of wood samples after sodium hydroxide extraction using
conventional heating and microwave irradiation process ( sample size < 2 mm)
ix
Table 6.4 Comparison of the calculated porosities of wood samples after 10 minutes of
sodium hydroxide extraction using conventional heating and microwave
irradiation process ( sample size < 2 mm)
Table 7.1 Independent variables studied in the CCD with their coded and uncoded levels
Table 7.2 DOE design matrix and the results
Table 7.3 Polynomial equations for the quadratic model and the regression coefficients
Table 7.4 Analysis of variance (ANOVA) for the RSM model
Table 7.5 Experimental and predicted values of wood dissolution, yield of xylan, and
temperature at the optimum extraction conditions used for the alkaline extraction
of xylan
x
ListofFigures
Figure 1.1 Typical structures of hemicelluloses in wood
Figure 1.2 Structure of hardwood xylan
Figure 1.3 Polysaccharide network in lignocellulosic matrix
Figure 1.4 Suggested lignin-carbohydrate bonds in lignocellulosic matrix
Figure 1.5 Representation of peeling and stopping reactions of polysaccharides
Figure 1.6 Schematic of a sinusoidal electric field to an ideal dielectric and the out-of-phase
displacement current which is induced
Figure 1.7 (a) Phase diagram for an ideal dielectric where the energy is transmitted without
loss; (b) phase diagram where there is a phase displacement and (c) the
relationship between ’ and ’’
Figure 3.1 Protocol for the xylan extraction and characterization
Figure 3.2 Experimental protocol for characterization of wood
Figure 3.3 HPLC calibration graphs for sugars
Figure 3.4 Calibration graphs for total sugar content determination
Figure 3.5 Calibration graph for molar mass determination
Figure 4.1. Effect of NaOH concentration on the wood dissolution and xylan yield (Power
level 110W)
Figure 4.2. Mechanism of alkaline hydrolysis and dissolution of hemicelluloses
Figure 4.3. Effect of irradiation time and microwave power on the solubilization of wood (a)
110W (b) 330W, and 550 W (c) 770W and 1100W
Figure 4.4. Temperature of the wood fiber slurry after different microwave irradiation time
Figure 4.5. Effect of irradiation time and microwave power on the yield of xylan (a) 110W
(b) 330W and 550 W (c) 770W and 1100W
Figure 4.6. Effect of microwave power and irradiation time on the dissolved sugar content of
the liquid phase after the precipitation of xylan
Figure 4.7. Effect of microwave power and irradiation time on dissolution of dissolved lignin
at 110 W, 330 W, 550 W, 770 W, and 1100 W
xi
Figure 4.8. X-ray crystallographs of wood fiber before and after different duration of
microwave extraction
Figure 4.9. FTIR spectra of xylans obtained at different extraction conditions
Figure 4.10. Typical SEC signal for xylan obtained by microwave assisted method of
extraction
Figure 5.1 Percentage of wood solubilized and yield of xylan after microwave and
conventional extraction
Figure 5.2 Comparison of the amount of wood solubilized during microwave and
conventional extraction
Figure 5. 3 SEM photomicrographs of wood fiber before and after extraction
Figure 5. 4 Effect of microwave and conventional extraction of xylan on the solubilization of
wood
Figure 5.5 Comparison of the yield of xylan (based on the total xylan) during microwave and
conventional extraction
Figure 5.6 Sugar content of the supernatant after precipitating the extracted xylan using two
different processes
Figure 5.7 Effect of microwave and conventional extraction of xylan on the yield of xylan
Figure 5.8 Effect of temperature on the yield of xylan
Figure 5.9 Effect of microwave and conventional extraction of xylan on the chemical
composition of wood
Figure 5.10 FTIR spectra of wood fibers before and after microwave and conventional
extraction
Figure 5.11 XRD crystallographs of birch wood fibers before and after microwave and
conventional extraction
Figure 5.12 SEM photomicrographs of birch wood fibers before and after microwave and
conventional extraction
Figure 5.13 FTIR spectra of xylan isolated using microwave and conventional extractions at
two different experimental conditions
Figure 5.14 ηsp/C vs. Concentration of xylan solutions in CED
Figure 6.1 Binary images of birch wood and the 3D image of the vessels extracted from the
binary images
xii
Figure 6.2 Typical X-ray microtomographic binary images of birch wood obtained at
different resolution
Figure 6.3 Average porosity of wood samples using different resolutions
Figure 6.4 Average porosity of wood samples using different volume sizes
Figure 6.5 Typical 3D images of wood chips after different extraction conditions (sample
size >2mm)
Figure 6.6 Typical 3D tomographic images of wood samples, samples after conventional and
microwave assisted extraction process (sample size < 2 mm)
Figure 6.7 Typical 3D images of wood chips after 10 minutes of extraction at 100oC, 120oC,
and 140oC
Figure 7.1 Actual temperature vs. predicted temperature
Figure 7.2 Response surface plots showing the interaction between the variables affecting the
temperature of wood slurry
Figure 7.3 Actual wood dissolution vs. predicted wood dissolution
Figure 7.4 Response surface plots showing the interaction between the variables affecting the
wood dissolution
Figure 7.5 Actual yield of xylan vs. predicted yield of xylan
Figure 7.6 Response surface plots showing the interaction between the variables affecting the
yield of xylan
Figure 7.7 ηsp/C vs. Concentration of xylan obtained at the optimal conditions of microwave
extraction
1
Chapter1 ScientificBackground
1.1 Introduction
During the last decade, an increased interest has been observed in the research of biopolymers
from renewable sources. The main driving forces behind the research activities in the area of
polysaccharides, including hemicelluloses are (i) increased awareness of the future shortage of
natural energy sources, (ii) replacement of petroleum-based products as a solution of worldwide
environmental problems, and (iii) demands for healthy food and alternative medicines. In the
past, research activities in the field of hemicellulose were designed mainly for utilizing plant
biomass by conversion into sugars, chemicals, fuel, and as sources of heat energy. Nevertheless,
these polysaccharides can also be used as biopolymers in their native or modified forms
(Ebringerova et al., 1994; Gabrielii et al., 2000; Sun et al., 1999; Jain et al., 2000, Lindblad et al.,
2005). Structural varieties and diversity of these biopolymers provides enormous application
potentials in the fields of food, medical, and industrial applications.
Forest and agricultural biomass are the inexpensive resources for hemicellulosic polysaccharides.
Canadian forests sector is the world’s second largest supplier of woody biomass, behind United
States and as such, an annual supply of more than 200 million m3 of biomass through
commercial operations is reported (FAO, 2003; NRCAN, 2003). Pulp and paper industry is the
major end user of the forest biomass and has been the major contributor to the North American
economy for many years. Recently, the pulp and paper sector has been experiencing a down turn
due to simultaneous impact factors such as immense worldwide competition from paper
2
industries producing fibers from fast growing species, the high cost of the energy, and the
reduced demand for the news print and pulp (Helmerius et al., 2010). The pulp and paper
industry needs to identify additional economic strategies in the coming year to revitalize the
forest sector and to strengthen its competitiveness in the current global market. Authors
including Mabee at al. (2005), Towers et al. (2007), Thorpe (2005), Carvalheiro (2008), and Mao
et al. (2008 ), have explored the concept of integrated forest biorefinery as an opportunity for the
forest product industry to increase revenues and improve environmental sustainability. In this
concept, all the components of biomass can be fractionated to utilize in a most profitable manner
to make high value chemicals, fuels, materials, heat, and power in addition to the traditional core
products. (van Heiningen, 2006, Huang et al., 2008).
The woody biomass is a highly complex material with three major components such as cellulose,
hemicellulose and lignin in different proportions along with minor amounts of extractives and
ash. In general, about 70 % of all wood is polysaccharides (cellulose and hemicelluloses)
(Sjostrom, 1993) and the composition of these components vary species to species, age of wood
and ecological factors. Compositions of different wood species are summarized in the Table 1.1.
Current practice in the pulp and paper industry to fractionate the woody biomass into cellulose
fibers is a “destructive” technology, where other components such as lignin and hemicelluloses
undergo degradation during the pulping process. Majority of hemicelluloses (approximately
20% of the dry weight of wood) are degraded during the chemical pulping process to simple
sugars or isosacharinic acid along with lignin (Casebier and Hamilton, 1965; Kleppe, 1970;
Roberts and El-Karim, 1983; Gustavsson and Al-Dajani, 2000). The sugars and lignin extracted
3
Table 1.1. Cellulose, hemicellulose and lignin content of woody biomass compiled from different authors (Sjostrom, 1993; Fengel and Wegener, 1984, and Garrote et al., 1999)
Raw material Species Cellulose Hemicellulose Lignin Hardwoods Red maple Sugar Maple Trembling aspen Common beech Silver birch Paper Birch Blue gum Softwoods Balsam fir Douglas fir Eastern hemlock Monterey pine Norway Spruce White spruce
from the severe degradation of wood and agro-residues during steam explosion treatment is a
bottleneck of the process (Glasser et al., 2000)
18
1.6.3 Alkalineextraction
Alkaline extraction at moderate temperatures is another common method for extracting
hemicelluloses from biomass (Dashek, 1997; Vuorinen and Alen, 1998; Ebringerova and Heinze,
2000; Glasser et al., 2000; Gabrieli et al., 2000). Hemicelluloses are partly alkali soluble and
highly accessible under hot alkali and acid conditions (Puis and Saake, 2004). Alkali hydrolyses
the ester linkages between plant polysaccharides and lignin, which releases polysaccharides and
increases the solubility of polysaccharides, without reducing their molar mass, under moderate
reaction conditions. The very common procedure to extract xylan is the alkaline extraction of
delignified wood. (Ebringerova and Heinze, 2000; Gustavasson et al., 2001). However,
delignification process modifies the chemistry of polysaccharides. (Sjostrom, 1993). Unlike hot
water or steam explosion process, alkaline treatment deacetylates the polysaccharide (van
Hazendonk et al., 1996).
In alkaline hydrolysis, several physical and chemical changes occur on hemicelluloses (Teleman
et al., 1995; Alen, 2000; Lai 2001; Sun and Cheng, 2002; Sun and Sun, 2002): alkali induces
swelling, leading to an increased surface area, a decrease in degree of polymerization, and
crystallinity occurs with a consequent separation of structural linkages between lignin and
carbohydrates. This results in the disruption of the lignin structure followed by saponification of
intermolecular ester bonds that cross link hemicelluloses and other components leading to an
increased porosity. In addition, removal of acetyl and uronic acid substitutions of hemicelluloses
also occurs.
Xylan is a hydrophilic polysaccharide shown to contribute to fiber wall swelling and flexibility
(Eriksson et al., 1991, Laine and Stenius, 1997). Swelling is the major factor in achieving a good
19
fractionation of hemicelluloses, where increasing concentration of alkali used. It was reported
that there is an optimum swelling concentration above which the swelling of the cellulose
inhibits the removal of hemicelluloses (Nelson and Schuerch, 1957). To improve the efficiency
of extraction of hemicelluloses using a given extractant, the structure must be as accessible as
possible to the extractant, and this can only be achieved by complete swelling. The importance of
swelling to remove hemicelluloses of different types of delignified wood in different extractants
was also emphasized by Hamilton and Thompson (1959), Dutton and Hunt (1958), and Morak
and Ward (1960;1961). The results of extraction depend on the type of alkali used and its
concentration and by developing a suitable extraction it is possible to attain a considerable
fractionation of xylan.
It was reported that hardwood xylans can be extracted in significant amounts using aqueous
alkaline solutions. The extractive power of lithium, sodium and potassium hydroxide solutions,
for hemicelluloses removal from Western Hemlock wood and its holocellulose was investigated
by Hamilton and Quimby (1957). They reported that sodium and lithium hydroxides had a
greater extractive ability than potassium hydroxide, for the isolation of glucomannans. However,
for xylose polymers containing arabinose and galactose, the extractive nature of all three alkali
salts was similar when used at concentrations equal to or greater than 2.8 N. Koshijima et al.
(1965) succeeded in removing 90% of xylan from trembling aspen wood using a 24% aqueous
potassium hydroxide solution. For the isolation of xylans from hardwood, in particular, two-step
procedures with a NaOH/H2O2 delignification step were shown to be more acceptable in practice
than the hazardous delignification with sodium chlorite. In a detailed study, Gabrielii et al.
(2000) described the isolation procedure and material properties of the methyl glucuronoxylan
(MGX) from aspen wood by an alkali extraction followed by a hydrogen peroxide treatment to
20
remove lignin, ultra filtration, and spray drying. The approach removed 55% of hemicelluloses
and the recovered xylan was reported to contain 65.2%xylan, 6.4% lignin, and 2.6% ash. A
comparison of the mild steam extraction and alkaline extraction of poplar wood and agricultural
residues showed that alkaline extraction has the advantage of providing polymeric xylan
compared to steam extraction (Glasser et al., 2000). The study also compares the efficiency of
various methods for the removal of associated lignin with the extraction and found that bleaching
with hydrogen peroxide followed by ultra filtration is a good procedure for isolation of xylan in
its polymeric form. A thermo-mechanical chemical fractionation of poplar hemicelluloses using
a twin screw extruder equipped with a filtration module in alkaline pulping was reported by
N’Diaye et al. (1996) and N’Diaye and Rigal (2000). Westbye et al. (2008) used a solid –liquid
extraction followed by a liquid-liquid extraction using pyridine/acetic acid mixture in an attempt
to produce xylan in a very pure form and was able to reduce the lignin content from 12% to 3%.
End-wise depolymerization (peeling) is one of the most important reactions of carbohydrate in
the alkaline reaction media that restricts the molecular mass of the extracted polysaccharides.
Hemicelluloses, like cellulose, possess a non-reducing end at one end of the chain and a reducing
end at the other. The peeling reaction removes one unit at a time from the reducing end of the
hemicelluloses chain until the chain end transforms to a stable oxidized carboxyl containing end
group, through a competing stopping reaction mechanism that hinders the peeling reaction
(Fengel and Wegener, 1984) (Figure 5). The rate of these reactions largely depends on the pH
and temperature of the alkaline medium. The activation energies for the peeling and stopping
reactions have been estimated to be 103 kJ/mol and 135 kJ/mol respectively (Haas et al., 1967).
21
Figure 1.5. Representation of peeling and stopping reactions of polysaccharides
The peeling reaction is nearly 65 times as fast as the stopping reaction (Franzon and Samuelsson,
1957; Lai and Sarkanen, 1967). Under alkaline conditions, deacetylated oligomeric xylan is more
stable due to the 4-O-methylglucuronic acid side chains and so alkaline extraction is suitable for
extraction of xylan from hardwood. However, solubilized polysaccharides can completely
degraded to sacharinic and hydroxyl acids under drastic conditions (Bhaskaran and von Koepen,
1970; Niemela and Alen 1999). Thus, it is possible to isolate xylan in its high molecular weight
before undergoing a complete peeling reaction by proper designing of the extraction method.
Selection of method of fractionation or pretreatments depends on the final requirement of the
product and the best pretreatment options are those which combine elements of both physical and
22
chemical methods (Ghosh and Singh, 1993; Saddler et al., 1993). In this regard, a fast and mild
alkaline extraction utilizing mild microwave energy would be an ideal choice for extracting
xylan in its polymeric form.
1.7 MicrowaveTechnology
Microwave radiation in the electromagnetic spectrum lies between infra red radiation and
radiofrequency waves. Frequencies of microwaves range from 300 MHz to 30 GHz
corresponding to the wavelength of 1cm-1m. The frequencies allotted for microwave heating
applications are 915 MHz and 2.45 GHz corresponding to a wavelength of 33.3 and 12.2 cm and
the latter is being used most often. Microwave chemistry has an edge over the conventional
methods for conducting chemical experiments and has the potential to achieve cleaner and
efficient reactions over conventional methods. Over the past three decades, this technology has
laid the foundation in the field of science with its applications extend from analytical
applications such as ashing, digestion, extraction, and fat analysis to synthetic organic and
polymer chemistry as synthesis of fine chemicals, organometallic intercalation, coordination, and
polymer curing (Bond et al., 1993; Caddick, 1995; Thostenson and Chou, 1999; Jones et al.,
2002; Hoz et al., 2005; Strauss and Varma, 2006; Zhang and Hayward, 2006). The enhanced
chemistry is attributed to the direct interaction between microwave energy and the materials of
interest which is explained in the following section.
The interaction between microwave and dielectric materials occurs through the dipoles or
induced dipoles present in the materials. The polarization of the dielectrics arises from the finite
displacement of or rotation of dipoles in an electric field. At the molecular level, this is due to the
distribution of the electron cloud within a molecule or the physical rotation of dipoles in the
23
presence of an alternating electric field and the latter being responsible for microwave radiation
heating.
Figure 1.6. Schematic of a sinusoidal electric field to an ideal dielectric (top) and the out-of-phase displacement current which is induced (bottom)
Microwaves consist of electric and magnetic field components and the electric field component
of microwaves interacts with the dipoles in the materials resulting in the rotation of the
molecules. In the microwave frequency range, the electric field component of microwave
radiation changes its direction 2.4 x 109 times per second (Metaxas and Meridith, 1983; Datta,
2001). The re-orientation of the dipoles and the resultant displacement of charge is equivalent to
an electric current, known as the Maxwell displacement current. For materials where the
molecules can keep pace with the field changes (ideal dielectrics), there is no lag between the
orientation of the molecules and the variations of the alternating voltage and the resultant
displacement current is 90° out-of-phase with the oscillating electric field as shown in Figure
24
1.6. There is no component of current in phase with the electric field. Rotation of polar
molecules lags behind the electric field oscillation and the resulting phase displacement
acquires a current component in-phase (I sin) with the electric field leading to resistive heating
mechanism in the medium and is described as dielectric loss (Figure1.7). If the phase angle
significantly differs from 90o, the materials can act both as a dielectric and conductor. The total
permittivity of the material - the parameter showing the materials ability to interact with the
radiation – is thus have a complex character and is expressed mathematically as
where ’ is the real part of the relative permittivity (the dielectric constant) and ’’ is the loss
factor which reflects the conductance of the material.
The phase diagrams representing ideal behavior and phase lag displacement and the relationship
between ’ and ’’ are shown in the Figure 1.7. In short, the interaction of microwaves with any
material depends on its dielectric properties: dielectric constant and dielectric loss factor. The
dielectric constant is a measure of the ability of a material to store electromagnetic energy and
the dielectric loss factor is a measure of the ability of a material to convert electromagnetic
energy into heat (Kumar et al., 2007). In the phase diagram, ''/' = tan, and is described as
energy dissipation factor (loss tangent) and is the parameter generally used to describe the
overall efficiency of a material to utilize energy from microwave radiation and convert to
thermal energy within the dielectric (Newham et al., 1991; Gabriel et al., 1998; Datta, 2001).
25
Figure 1.7. (a) Phase diagram for an ideal dielectric where the energy is transmitted without
loss; (b) phase diagram where there is a phase displacement and (c) the
relationship between ’ and ’’
1.7.1 Microwaveassistedsolventextraction
Microwave assisted extraction is a process where microwave energy is used to partition the
material of interest from the sample to the surrounding liquid. This process has been significantly
exploited in the pharmaceutical, fine chemicals, and environmental applications. Unlike
conventional thermal processing, where energy is transferred to the material through convection,
conduction, and radiation of heat from the surfaces of the material, microwave energy is
delivered directly to materials through molecular interaction with the electromagnetic field and
deposit energy within the material leading to heat generation throughout the material (which is
26
called volumetric heating). Moreover, when materials with different dielectric properties exposed
to microwave radiation microwaves will selectively couple with the higher lossy material. In
general, the higher the dielectric constant, the more efficient the molecules absorb energy and
hence are heated more efficiently. Molecules with low dielectric constant and loss factor cannot
absorb energy as they cannot couple efficiently with the microwave radiation. The special
heating mechanism and the fact that different chemical substances absorb microwaves at
different levels can make microwave assisted extraction an efficient and selective method of
extraction of target compounds.
1.7.2 Microwaveassistedextractionofplantmaterials
Extraction of plant compounds using microwave assisted process is highly dependent on the
solvents used. Selection of solvent depends on the solubility of the target compound and the
dielectric properties of the solvent. When a polar solvent with a relatively high dielectric
constant and loss factor is used, the solvent will be heated through dipole rotation by interaction
with microwave energy. The heated solvents will accelerate the process of desorption of the
matrix-solvent interface and the diffusion of the target compounds into the solvent (Hawthorne et
al., 1995). The special extraction mechanism of microwave assisted extraction can be better
interpreted when a non-polar solvent is used for extraction, where the material interacts with the
microwave radiation. Studies on the solvent extraction of leaves and seeds with highly absorbing
solvents and non-polar solvents have been demonstrated to selectively extract the compound of
interest. It was found that in the case of non-polar solvent extraction, a significant fraction of
microwave energy is absorbed by the sample mainly by the water in the glandular and vascular
systems and the volumetric heating effect enhanced the extraction process and reduced the time
of extraction (Pare and Belanger, 1994; Chen and Spiro, 1994; 1995; Chemat et al., 2005).
27
Ganzler et al. (1986) reported that for extraction of crude fat, and other components from oil
seeds, the duration of microwave assisted extraction was almost 100 times less than the
traditional methods. Pare (1995) also reported similar observation to extract plant components
from cedar, garlic, parsley and mint leaves. The reduction in the time of extraction compared to
conventional extraction is the unique feature in all these extraction studies.
The earliest known study in the lignocellulosic materials on microwave pretreatment examined
the effect of microwave radiation on rice straw and bagasse immersed in water and reported an
improvement in total reducing sugar production by a factor of 1.6 for rice straw and 3.2 for
bagasse in comparison to untreated biomass (Ooshima et al., 1984). Microwave pretreatment of
sugar cane bagasse and rice hulls soaked in water followed by lignin extraction was reported to
yield 77-84% of total available reducing sugars (Azuma et al., 1984). A similar study involving
microwave pretreatment of rice straw and sugarcane bagasse followed by lignin extraction
reported a yield of 43-55% of total available reducing sugars (Kitchaiya et al., 2003). Microwave
pretreatment of rice straw soaked in dilute alkali resulted in a glucose yield of 65% and total
carbohydrate conversion of 78% (Zhu et al., 2005). A recent study on microwave based alkali
pretreatment of switch grass investigated low power levels for extended pretreatment time and
reported 70-90% sugar yields (Hu and Wen, 2008). Application of microwave technology in the
extraction of hemicelluloses from various biomasses including flax shives (Jacobs et al., 2003;
Burnov and Mazza, 2010), wood (Teleman et al., 2000; Palm et al., 2003; Jacobs et al., 2002;
Lundquist et al., 2002), corn pericarp and fiber (Yoshida et al., 2010; Benko et al., 2007), and
barley husk (Roos et al., 2009) were reported. The details of the studied parameters such as yield
of xylan, molecular weight distribution, and lignin content are cited in the Table 1.3.
28
There is great fundamental and tremendous environmental and economic importance entailed to
achieve the extraction of hemicellulose, xylan, with minimum molecular alteration. But, the
studies reported so far either performed at a high temperature (>120oC) using water, acid, or base
catalyzed reaction for extraction or a pressurized solvent assisted extraction. The severe
conditions of extraction leads to degradation of the xylan and further, such processes require
costly experimental set up which has a negative impact on the cost efficiency. Also, these
studies reported different yields (lower and higher) and different characteristics of the xylan
extracted compared to conventional methods. Moreover, there is no available literature on
systematic analysis to differentiate the microwave assisted extraction from that of conventional
extractions. Hence, a fundamental study focused on the mechanism of extraction of xylan from
the plant materials using microwave energy will be justified for understanding and developing an
efficient physico-chemical extraction process for extracting xylan from the plant biomass. The
study would provide not only an innovative knowledge on the fundamentals of microwave
assisted extraction but also, an environmentally friendly method for the isolation of
hemicellulose (xylan) with minimum degradation, which is highly valuable to the bio-refinery
industry.
29
Table 1.3: Available literature on microwave assisted extraction of polysaccharides from various biomass Authors Raw material Solvent used Yield & Lignin
content Molecular weight
Burnov and Mazza (2010) Teleman et al., (2000) Palm and Zacchi (2003) Benko et al., (2007) Jacobs et al., (2002) Alexandra et al., 2009 Lundquist et al. 2002 Jacobs et al., (2003) Yoshida et al., (2010)
Water; ethanol Water Water; ethanol Sulfuric acid, sodium hydroxide and water water Water, dil. Alkali, Dil.acid (NaOH, H2SO4) Water, NaOH, KOH and H2SO4 Acid and base catalysed hydrothermal microwave treatment Alkaline Hydrothermal treatment
xylooligosaccharides; Lignin 6.4% Predominant compound in the extract were Xylo-oligomers No data on lignin Yield: Water (20%), Ethanol (6%) No data on lignin No data 11-30% NaOH-4-11% H2SO4 -2-11 Water-20% No data Water- 50% Acid -35% Base – 9% Galactoglucomannan oligopolysaccharide yield- 70-60%
No data No data Mw-water 32251 Mw-alcohol 3965 Average degree of polymerization 2-35 24000-870 g/mol (DP-133) Mw 136000-172000 13000-18000 127000 DP- 6-20 Mw-13,000 Mw-13,000 Mw-52,000 Mol.Wt., 800-80000 Mw-low 1700-4500
30
Chapter2 ResearchHypothesisandObjectives
2.1 Hypothesis
Selective interaction of microwave energy with the more lossy component (alkali) in the wood
fiber generates hot-spots and the resulting "explosion effect" rupture the recalcitrant
lignocellulosic structure and increase porosity thereby enhancing the mass transfer of hydrolyzed
components to the solution and reduce the time of extraction.
2.2 ResearchObjectives
1. Investigate and establish an efficient physico-chemical process using microwave energy
for the extraction of polymeric xylan from lignocellulosics for developing value added
polymers.
2. Investigate the effect of interaction of microwave energy with the wood slurry on the
extraction of xylan in comparison with conventional heating process
3. Elucidate the mechanism involved in the microwave assisted alkaline extraction and
study the effect of microwave irradiation on the physico-chemical properties of xylan
4. Optimize microwave assisted extraction of xylan from birch wood.
31
Chapter3MaterialsandMethods
3.1 Materials
3.1.1 Birchwoodfibers
Birch wood fibers were prepared using chips obtained from St. Mary’s Paper Company, Ontario,
Canada, and from birch wood log collected near the University of Toronto’s premises. The bark
of the logs was removed and was cut into smaller blocks. The chips and small wood blocks were
dried at 40oC for 5 days, and were separately ground to 2 mm size using a Wiley mill. The
ground wood fibers were sieved using a 0.42 mm sieves to remove the fines. Extractives were
removed from these fibers before the xylan extraction. Preparation of extractive-free wood fibers
involved the following procedure. The fibers were treated with 0.05 M hydrochloric acid (solid
to liquid ratio of 1:10 g/mL) at 70oC for 2 hours. After cooling the suspension, ammonium
hydroxide (14 M) was added to a pH of 9-10 and the fibers were allowed to swell, in the
suspension, overnight to remove pectins, starch and fat (Gabrielii et al., 2000). The suspension
was then filtered through 0.42 mm sized screen and washed thoroughly with water until the
washings were neutral. The extractive-free wood fibers were then dried at 40oC for 72-96 hours
and stored in air tight containers until used.
3.1.2 Chemicals
All the chemicals used in this study were of analytical or reagent grade and are listed below:
0.05 M hydrochloric acid, ammonium hydroxide, sodium hydroxide, sulfuric acid (72% and
96%), and dilute acetic acid were of reagent grade; HPLC grade de-ionized water for HPLC
used as a reference for FTIR characterization, sugars used as standards for chromatographic
analysis (glucose, xylose, mannose, arabinose, and galactose), and dextran standards with
different molecular weights used for size exclusion chromatography were of analytical reagents
and were obtained from Sigma Aldrich.
3.2 Methods
3.2.1 Isolationofxylanfrombirch
The protocol used for the isolation of xylan from birch fibers is given in the Figure 3.1. Slurry of
the wood fibers in sodium hydroxide solution was subjected to conventional and microwave
assisted extraction. Xylan was precipitated from the extract and the wood residue was
characterized to investigate the structural changes occurred during the extraction process. The
extract after precipitation of xylan was used to quantify the molecular degradation of the xylan
during the extraction process.
33
Figure 3.1. Protocol for the xylan extraction and characterization
3.2.1.1 Extractionofxylanfrombirchusingmicrowave
A general purpose microwave oven (Sanyac Model) operating at 2450MHz and with adjustable
microwave power input (110 to 1100W) was used for microwave assisted extraction. Wood
fiber slurry was prepared by adding required amount of sodium hydroxide (NaOH) solution
(concentration of NaOH solution: 1 wt% to 4 wt%) to the weighed extractive-free wood fibers
taken in an Erlenmeyer flask. The slurry was kept at room temperature for 5 minutes to
Conventional method of
extraction
Filtrate
Characterization
Degraded sugar content
Microwave assisted
extraction
Extractive free
wood sample
Filtrate
Xylan
Characterization
Sugar composition, Lignin
content, Molar mass, &
FTIR,
Precipitation of xylan
Residue Residue
Characterization
SEM & X‐ray tomography
FTIR, X‐ray crystallography
34
completely wet the wood fibers and placed in the middle of the oven over the rotating plate and
xylan was extracted by exposing the slurry to microwave radiation according to the experimental
conditions explained in the following chapters. At the end of heating, temperature of the mixture
was noted. Immediately after, the mixture was filtered through a previously weighed crucible
and the residue was washed with distilled water (50-100 mL depending on the conditions of
extraction) and combined the washings with the filtrate. The filtrate was then neutralized with
acetic acid to a pH of 4.6 to precipitate dissolved xylan. The precipitate was allowed to settle
overnight and separated from the liquid by centrifugation at room temperature. An aliquot of
liquid phase after the separation of precipitated xylan was collected to determine the dissolved
sugar content. The separated xylan was washed with water to remove the salt, 95% ethanol,
centrifuged, and freeze-dried or oven dried at 40oC. In some cases, the precipitated xylan was
re-dissolved in minimum amount of sodium hydroxide and the solution was poured into 3
volumes of ethanol to re-precipitate the xylan. The precipitated xylan was then washed with
water, centrifuged and freeze dried. The wood fibers after each extraction was washed until the
washings were neutral to pH paper and were dried at 40oC for 72-96 hours. All the extractions
were carried out at least three times to get an average result.
3.2.1.2 Extractionofxylanfrombirchusingconventionalheating Conventional extraction was performed at different temperatures according to the experimental
conditions described in the respective chapters using a water bath or oil bath depending on the
temperature used for the study. The wood fiber slurry was extracted isothermally using sodium
hydroxide solution at different temperatures. The protocol used for separation of wood residue
35
from the liquid phase and precipitation of xylan after extraction were the same as those used in
microwave extraction.
3.2.2 CharacterizationofBirchwood
Figure 3.2. Experimental protocol for characterization of wood
Extractive-free wood fibers were characterized for moisture content, chemical composition, and
ash content. A sample of the extractive-free wood fiber was ground to 0.25 mm size using a
Wiley mill and used for the determination of chemical composition. The experimental protocol is
demonstrated in Figure 3.2. For all the quantitative analysis, three samples were used and the
results expressed are the average of three measurements.
3.2.2.1 Holocellulose Holocellulose represents the entire polysaccharide, cellulose and hemicelluloses, portion of
wood. Determination of holocellulose followed the method described by Zobel and McElvee
(1966) by dissolving the lignin in an acidic medium with chlorine based solution. The procedure
used was as follows: About 0.3-0.5 g of air dried extractive-free wood was accurately weighed
Moisture content
Holocellulose
Extractive free Birch woodAsh content
‐ cellulose Lignin Sugar composition
36
into a 125 ml Erlenmeyer flask and added 32 ml of distilled water, followed by 0.1 ml glacial
acetic acid and 0.3 g reagent grade sodium chlorite. The flask was capped with a loose fitting
inverted 10 ml Erlenmeyer flask and heated in a temperature controlled water bath at 70oC. The
contents were heated for 1 hour with occasional swirling to ensure a uniform mixing of the
reaction mixture. Without cooling, an additional 0.1 ml of glacial acetic acid and 0.3 g sodium
chlorite were added successively after one hour for two times. After three hours of heating at
70oC, the flasks were placed in an ice bath and cooled the reaction mixture below 10oC. The
contents of the flask were filtered through a previously weighed coarse fritted glass crucible
using a minimum quantity (25 ml) of ice distilled water to transfer all the holocellulose and
remove the color and odor of the chlorine dioxide. The contents were then washed with 100 ml
of hot distilled water under suction, washed with acetone without suction, dried by suction and
dried in an oven at 105°C for 24 hours. The crucible with the sample was allowed to cool in a
desiccator before weighing. Holocellulose was determined using the equation 3.1.
%
100
3.1
3.2.2.2 ‐Celluloseandhemicelluloses
-Cellulose was determined from holocellulose after hydrolyzing the heterogeneous
polysaccharides for 2 hours using 17.5% alkali, as per the TAPPI procedure T203 om-93.
Holocellulose used for - Cellulose determinations were not oven dried, but air dried for
overnight in a conditioning chamber. -Cellulose in the sample was calculated using the
equation 3.2.
37
%
100
3.2
The difference between the holocellulose and -cellulose was accounted as hemicelluloses
content.
3.2.2.3 AshContent
Ash content of the samples was carried out using a modified Tappi procedure T413 pm 95.
Representative samples were weighed accurately in previously weighed silica crucibles, and
heated at 600oC for 4 hours to remove the carbonaceous materials. The samples were then cooled
to room temperature in desiccator before weighing. Ash content was calculated based on the
oven dry weight of the wood sample using the equation 3.3.
%
100
3.3.
3.2.2.4 SugarCompositionandLigninContent
The constituent sugars and lignin content of the wood fibers were determined using a two-step
hydrolysis with 72 and 4% sulfuric acid (Sluiter et al., 2008). Samples were hydrolyzed using
72% sulfuric acid, followed by a second hydrolysis of the samples with 4% sulfuric acid for 1
hour at 121oC in an autoclave. The procedure involved is as follows: About 0.3 g of the wood
sample was measured in to a tared glass tube followed by 3 mL of 72% sulfuric acid and mixed
the reaction mixture thoroughly with a stirring rod. The glass tubes were incubated in a water
bath set at a temperature of 30oC with frequent stirring, without removing the sample from the
water bath. After 1 hour of incubation, the samples were completely transferred to pressure tubes
38
using 84 g of deionized water, using a balance, in order to dilute the acid concentration to 4%.
The samples were then autoclaved at 121oC for 1 hour. A set of sugar recovery standards with
4% sulfuric acid was also autoclaved to determine the loss of sugar during the hydrolysis.
The hydrolyzed samples were then filtered through a previously weighed medium fritted glass
crucible, and an aliquot of the filtrate was kept aside for the determination of sugar composition
and acid soluble lignin. The solid residue from the flask was completely transferred
quantitatively to the crucible using deionized water. The residue was then washed with hot
deionized water until the washings were free from acid. The crucibles with the acid insoluble
residue were dried at 105oC for 24 hours and were then cooled to room temperature before
weighing. Acid insoluble lignin was determined using the equation 3.4.
100
3.4
For determining acid soluble lignin, absorbance of the collected filtrate was measured using a
UV-visible spectrophotometer at a wavelength of 240 nm. An aliquot of the sample was diluted
using deionized water to get the absorbance reading in the range of 0.7-1. The amount of acid
soluble lignin was calculated using the equation 3.5.
∈100
3.5
where, UVabs = average UV-visible absorbance for the sample at 240 nm;
Volume filtrate = volume of the filtrate (87 mL)
39
3.6
where, ε = absorptivity of biomass at specific wavelength ( at 240 nm, ε is 25 L/g cm).
For the monosaccharide composition of the hydrolyzed samples, an aliquot (30 mL) of the
filtrate sample was transferred to an Erlenmeyer flask, and neutralized using calcium carbonate
to a pH of 5-6. The sample was allowed to settle, and decanted the supernatant. The neutralized
samples were analyzed using a high performance anion exchange chromatography (HPAEC)
(Dionex system) equipped with a Carbopac PA10 column, a pulsed amperometric detector (ED-
40) and an auto sampler. Sugars were separated using 3 mM NaOH solution under isocratic
conditions for 25 minutes. Flow rate of eluent was 1 mL/minute and the detection was performed
by pulsed amperometry (ED-40, Dionex). After each isocratic elution, the column was cleaned
with 250 mM NaOH for 10 minutes and then equilibrated with the eluent for further 20 minutes.
The total time for the analysis of each sample was 57 minutes. Calibration was performed with
five sets of standard solutions containing L(+) arabinose, D(+) galactose, D(+) glucose, D(+)
xylose, and D(+) mannose in appropriate concentrations. Calibration plots obtained for different
sugars are shown in Figure 3.3. Two replicates of each sample prepared were analyzed and for
each sample two replicates of hydrolysis were performed. Chromeleon software was used for
the quantification of the monosaccharides using the calibration curves. Content of individual
polysaccharides was calculated by multiplying the content of corresponding mono sugar with the
correlation factors; 0.88 for xylose and arabinose, and 0.9 for glucose, mannose and galactose
(Sluiter et al., 2008). The results reported were the average of four analysis results.
40
Figure 3.3. HPLC calibration graphs for sugars
3.2.2.5 ScanningElectronMicroscopy(SEM)
The wood fibers before and after both extraction processes, were subjected to SEM (Hitachi S-
2500, Tokyo, Japan) to study the changes in the microstructure during extraction. The dried
fibers were mounted on metal stubs and were sputter coated with a thin layer of gold to avoid
electrostatic charge during examination. Images were taken using an accelerating voltage of
15kV and a working distance of 10 mm.
3.2.2.6 X‐raydiffraction
XRD- measurements were conducted on a Bruker AXS D8 Discovery diffraction system (Bruker
AXS Inc., Madison, WI, USA) equipped with a high power point focus (1x1mm) Cu-kα target,
graphite monochromator (26.53o) for elimination of Cu-kβ lines, and a Hi-Star GADDS area
y = 425.37xR² = 0.9956
y = 505.37xR² = 0.996
y = 534.71xR² = 0.9959
y = 620.45xR² = 0.9964
y = 501.5xR² = 0.9979
0
4
8
12
16
20
0 0.005 0.01 0.015 0.02 0.025 0.03 0.035
Are
a o
f th
e p
ea
k
Concentration (mg/mL)
Arabinose Galactose Glucose Xylose Mannose
41
detector for 2-D images. The powdered samples having a size of 0.25 mm were pressed into the
rectangular sample holder with dimensions of 15 x 20 mm and with a thickness of 1mm. The
mass of the sample in the sample holder was approximately the same for all the experiments. The
X-ray diffractometer was operated at a voltage of 40 kV with a current density of 40 mA. The
scanning range was from 2 - 5o to 40o at a scan speed of 2o/s. The data was collected using a
continuous mode with angular intervals of 0.02o. The percentage of crystalline material in the
wood fibers was evaluated as a crystallinity index (CrI) using Segal's method (Segal et al.,1959)
as shown in the equation 3.7.
%
100 (3.7)
where I002 is the maximum intensity (in arbitrary units) of the 002 diffraction peak at 2 = 22.8o
and Iam is the intensity of diffraction in the same units at 2 =18o.
3.2.2.7 X‐raymicrotomography
X-ray microtomographic technique was used to quantify the porosity of the wood samples. A
Skyscan 1172 high resolution desktop X-ray micro CT system (Skyscan, Artselaar, Belgium:)
was used to obtain the microtomographical images of the samples. The system consists of a
microfocus sealed X-ray tube with high voltage power supply, an object stage with a precision
manipulator, a 2D X-ray CCD detector and a computer with tomographic reconstruction
software (NRecon from Skyscan). The X-ray CCD is based on a medium resolution of cooled
CCD sensor with a fiber optic coupling to an X-ray scintillator. The samples were scanned using
a medium resolution of 2014x1024 at a source voltage of 46 kV and a current of 180 µA for
thicker samples and 30 kV and 122 µA for thinner samples. The X-ray radiographs were
recorded over the interval of 0 to 360 degrees using a stepwise rotation of 0.4 degree. The X-ray
42
source and detector were remained at a fixed position. In order to improve the signal to noise
ratio, and to get a high quality radiograph images a 6 frame averaging, (each radiograph image
obtained is the average of 6 shadow projections) was used. After the image acquisition, the 2D
cross sectional images of the samples were reconstructed using Skyscan's reconstruction
software (NRECON_version 1.4.4). The 2D binary images of the samples obtained were
processed for 3D construction and porosity analysis using image analysis software Image Pro
(version 6) and the method is described in Chapter 6.
3.2.3 Analysisofliquidextracts
Lignin content of the liquid extract after extraction processes was determined by measuring the
UV absorbance at a wavelength of 240 nm (Sluiter et al., 2008) using a similar equation for the
soluble lignin as described in section 3.2.3.4.
The total sugar content of the liquid phase after precipitating the xylan was determined by
phenol-sulfuric acid method using glucose as the standard (Dubois et al., 1979; Fournier, 2001).
A series of 10 standard solutions with different concentrations were prepared using a glucose
standard solution (1 mg/mL). A definite volume of the standard sugar solutions (5 µL-50 µL)
were mixed with 500 µL of 4% phenol and 2.5 mL of concentrated sulfuric acid (96%) to
develop the color and the absorbance of the solutions was then measured using UV-visible
spectrophotometer at 490 nm to prepare the calibration curve. Similar procedure was used to
develop color for the samples of liquid phase after the precipitation of xylan. Volume of the
sample taken was adjusted to obtain the absorbance within the range of the absorbance of the
standard solutions. Concentration of the sugar (mg/mL) was determined from the absorbance of
the sample solution, using the calibration curve (Figure 3.4).
43
The percentage of total sugar based on oven dried wood was calculated using the equation 3.8.
% .
100 3.8
Figure 3.4. Calibration graphs for total sugar content determination
3.2.4 Characterizationofxylan
3.2.4.1 Sugarcompositionandlignincontent
The sugar composition and lignin content of the extracted xylan was determined by a two-step
acid hydrolysis using the same procedure as described in the wood characterization.
3.2.4.2 Molecularweightdetermination
The molecular weight of the extracted xylan was determined using an HPLC equipped with a
In order to study the effect of microwave power input and irradiation time on the efficiency of
xylan extraction, wood fibers were extracted with 4 wt% sodium hydroxide solution (solid to
liquid ratio: 1: 10 g/mL) at different microwave power inputs (110 W – 1100 W) for different
periods of time (10 sec-10 minutes) (Table 4.2). The results were studied in terms of
solubilization of wood and the yield of precipitated xylan.
4.2.2.1 Solubilizationofwood
Solubilization of wood components and the resultant wood loss are caused by the hydrolysis of
the esterified linkages of hemicelluloses and lignin and the transfer of these wood components
into the alkali. The mechanism of transport phenomena occurs during the alkaline reactive
extraction of hemicelluloses from wood consists of the following steps as shown in Figure 4.2:
(i) diffusion of NaOH solution (axially) from the bulk phase to the vessels or central canal in the
cells (ii) diffusion of NaOH from the voids (radially) to the reaction sites into the cell walls
(primary and secondary cell walls) (iii) hydrolysis of the hemicelluloses-lignin linkages,
liberation and solubilization of the hemicelluloses (iv) radial diffusion of hemicelluloses from the
walls to voids, and (v) axial diffusion of hemicelluloses from the central voids to the bulk phase.
Dissolution of lignin along with the hemicelluloses is also expected under alkaline extraction
conditions.
52
Figure 4.3 shows the amount of solubilized wood after each extraction for different microwave
power inputs and time of irradiation. It is obvious that dissolution of wood depends on the
extraction conditions such as irradiation time and microwave power as these are the two main
factors affecting the radiation intensity on the reaction mixture and the resulting temperature rise.
At all the power levels studied, wood solubilization increased with the time of irradiation. This
is attributed to the temperarure increase of the wood slurry as a result of longer irradiation time
(Figure 4.4). As the temperature increases, diffusion of NaOH solution as well as the hydrolysed
components (hemicelluloses and lignin) increases that respectively leads to the enahnced
Figure 4.2. Mechanism of alkaline hydrolysis and dissolution of hemicelluloses (Prat et al., 2002)
hydrolysis of hemicelluloses-lignin linkages and their transfer to the solution. For example, at a
power input of 110W, solubilization of wood increases from 9% at an irradiation time of 1
minute (temperature was 43oC) to 19.7% for 10 minutes of radiation (temperature was 96oC),
(Figure 4.3a) indicating a high dissolution at severe conditions of extraction. A similar trend in
53
wood dissolution with time was observed at all the power inputs studied.
Figure 4.3. Effect of irradiation time and microwave power on the solubilization of wood (a) 110W (b) 330W, and 550 W (c) 770W and 1100W
0
5
10
15
20
25
0 200 400 600 800
Wo
od
so
lub
iliza
tio
n (
% O
D b
as
is)
Time (s)
0
5
10
15
20
25
0 50 100 150 200 250
Wo
od
so
lub
iliza
tio
n (
% O
D b
as
is)
Time (s)
330W 550W
0
5
10
15
20
25
0 20 40 60 80 100
Wo
od
so
lub
iliza
tio
n (
% O
D b
as
is)
Time (s)
770 W 1100W
(a)
(b)
(c)
110 W
54
The effect of power input on the solubilization of wood was also demonstrated in the Figure 4.3.
As the power input increased, the time required for approximately the same amout of wood
dissolution decreased significantly. Maximum wood solubilization obtained at the experimental
conditions studied was about 20%. Time required to obtain this solubilization was found to
decrease from 600 seconds to 180 sec to 60 seconds when the power level increased respectively
from 110 W to 550 W to 1100W. Similarly, the amount of wood solubilized after 60 seconds of
extraction at each power level studied are found to increase and are reported as 8.9%, 13.7%,
15.0%, 15.9%, and 18.6% respectively for 110 W, 330W, 550W, 770W and 1100 W. The
increasd solubilization for the same duration of irradiation or decreased time of irradiation for the
same amount of dissolution was attributed to the rapid increase in the temperature of the reaction
slurry with the increased power level as seen in the Figure 4.4.
Figure 4.4. Temperature of the wood fiber slurry after different microwave irradiation time
0
20
40
60
80
100
120
0 100 200 300 400 500 600 700
Te
mp
era
ture
(oC
)
Time (s)
330W 550W 770 W
1100W 110 W
55
4.2.2.2 Yieldofxylan
It was demonstrated from the wood solubilization study that longer irradiation time at lower
power input as well as shorter irradiation time at higher power input leads to higher dissolution
of wood. A similar trend was expected in the yield of xylan, as more xylan dissolves with the
increase in the wood solubilization. Figure 4.5 shows the effect of irradiation time as well as
power input on the yield of xylan obtained during different extraction conditions. At all the
power levels studied, similar to wood solubilization, yield of the precipitated xylan increased
with time of irradiation (Figure 4.5). However, at higher power inputs, yield of xylan obtained
was found to be leveling out at longer duration of extraction. Moreover, the maximum yield
obtained at higher power input was lower than the lower power input extractions. During
microwave heating the temperature of the reaction media increases due to the direct interaction
of the microwaves with the media. Figure 4.4 shows that a rapid rise in temperature occured
wihin very short time at the higher power inputs compared to the lower power input. It is already
known that wood hemicelluloses are relatively easily hydrolyzed by alkali, and at the same time
undergoes degradation or peeling of the dissolved carbohydrates in the presence of alkali
(Kleppe, 1970). It is anticipated that the in-situ heat generation and the increase in the
temperature of the reaction slurry during microwave assisted extraction leads to these two
opposing phenomena. The rise in temperature increases the diffusion of NaOH and the
hydrolyzed xylan leads to an increase in the hydrolysis rate as well as mass transfer of the
hydrolyzed xylan to the solution. On the other hand, the increase in temperature may also leads
to the degradation of the already dissolved xylan. Yield of the xylan obtained depends on these
two opposing phenomena. At the lowest power level studied, (110 W), yield of xylan increased
56
Figure 4.5. Effect of irradiation time and microwave power on the yield of xylan (a) 110W
(b) 330W and 550 W (c) 770W and 1100W
0
10
20
30
40
50
60
70
0 200 400 600 800
Yie
ld o
f x
yla
n (
%)
Time (s)
0
10
20
30
40
50
60
70
0 50 100 150 200 250
Yie
ld o
f x
yla
n (
%)
Time (s)
330W 550W
0
10
20
30
40
50
60
0 20 40 60 80 100
Yie
ld o
f x
yla
n (
%)
Time (s)
770 W 1100W
(a)
(b)
(c)
110 W
57
steadily with time of irradiation, from 33% at the mildest condition (60 seconds) to 60% at the
severe conditions (600 seconds). The amount of wood solubilized at these conditions was 9%
and 20% respectively and the temperature of the slurry was 43oC and 96oC respectively. Though
there is degradation of the carbohydrates due to increase in temperature of the alkaline reaction
media (43oC - 96oC), the results indicates that increased hydrolysis rate and transfer of the
dissolved components to the solution is the major process at this low power extraction.
At higher power levels, even though the yield of xylan increased with time of radiation, the
maximum yield obtained was significantly lower than the xylan obtained at the lowest power
level (110W) studied. It was also observed that the yield of xylan obtained decreased with the
increase in the power input. The yield of xylan obtained after maximum duration of irradiation at
each power levels studied are 45, 46 wt% respectively for the power levels of 330 and 550 W
(for 180 seconds), and 38 and 35 wt% respectively for 770 and 1100 W (for 60 seconds). The
amount of wood solubilized at these extraction conditions were 17.4 %, 19.5 %, 16 %, and
18.6% respectively. A comparison of the yield obtained at the lowest power extraction studied
(110 W) indicated that for a wood dissolution of about 15% to 20%, the yield of xylan obtained
varied from 44 % to 60%. Further, comparison of the yield of xylan obtained for the similar
extraction duration (Figure 4.5 b, and 4.5.c) indicates that, the yield obtained was leveled out
faster at the higher power level extractions (550 W for 3 minutes and 1100 W for 1 minute)
compared to the respective lower power inputs (330 W for 3minutes and 770 W for 1 minute).
These results indicate that during extractions using high power input, degradation of the
carbohydrate (xylan) becomes significant compared to the similar duration at the low power
extraction. This can be attributed to the rapid increase in the temperature of the wood slurry as a
result of high intensity of microwave irradiation during the higher power input extraction. As the
58
power input increased, the temperature reaches rapidly to about 95-97oC ( 1-3 minutes) and this
might enhance the degradation of the dissolved xylan lowering the yield.
It is expected that the supernatant after the precipitation of xylan contains lower molecular
weight hemicelluloses and/or monomeric sugars and/or other degraded products. Hence, to study
the degradation of the carbohydrates, total sugar content of the supernatant after the precipitation
of xylan was determined using phenol- sulfuric acid method. Figure 4.6 shows the amount of
total sugar in the supernatant obtained after different duration of extractions using different
power inputs. Earlier studies reported that under mild conditions of alkali hydrolysis (below
100oC) dissolution of cellulose is negligible compared to hemicelluloses (Kleppe, 1970) and
hence the dissolved sugar is assumed to be derived from the hemicelluloses present in the wood.
In the figure, the sugar obtained was presented in terms of the total sugar content in the wood as
well as the sugar content in terms of the hemicellulose present in the wood. At all the power
levels studied, total sugar content in the supernatant increased with the time of irradiation (Figure
4.6). The total sugar content in the supernatant at the conditions of maximum wood
solubilization at each power level studied is about 2.9 to 3.3 % of the total sugar content, which
is about 7.6% to 9% of the total hemicelluloses. The results demonstrated that the amount of
total sugar in the supernatant after 600 seconds of irradiation at a microwave power input of 110
W is about the same as the sugar obtained after 180 seconds of irradiation at a power input of
550 W and 60s of irradiation at a power input of 1100 W. Similarly, for the same duration of
irradiation at each power input, the amount of degraded sugar increased with the power level
indicating more degradation of the carbohydrates at higher power input for the same duration.
The degradation rate of xylan increases with the increase of the power input, which is due to the
59
rapid rise of temperature and this leads to the lower yield of xylan at higher power microwave
extraction. These results demonstrated that lower microwave power input and longer duration of
irradiation will be the most suitable processing parameters for the efficient extraction of xylan.
Figure 4.6. Effect of microwave power and irradiation time on the dissolved sugar content of
the liquid phase after the precipitation of xylan (A and B represents sugar content based on the total sugar content; A1 and B1represents sugar content based on the hemicelluloses)
60
4.2.2.3 Dissolvedlignin
Dissolution of lignin along with the hemicelluloses is also expected under alkaline extraction
conditions. Percentage of dissolved lignin was calculated on the basis of the lignin originally
present in the wood. As expected, at all the power levels studied, amount of dissolved lignin
increased with the increase of irradiation time (Figure 4.7). At a power output of 110 W,
amount of soluble lignin showed an increase with the time (5% at 2 minute vs. 10% at 10
minutes) of irradiation indicating the dependency of hydrolysis on the time and temperature. At
all power levels, amount of dissolved lignin increased at a particular irradiation time and the
results are in accordance with the high dissolution of wood as explained in the previous section.
Figure 4.7. Effect of microwave power and irradiation time on dissolution of dissolved lignin at 110 W, 330 W, 550 W, 770 W, and 1100 W (Secondary axis are for 110 W as shown by the arrows)
0 100 200 300 400 500 600 700
0
2
4
6
8
10
12
0
2
4
6
8
10
0 50 100 150 200
Time, (s)
Dis
solv
ed li
gnin
con
ten
t, %
Dis
solv
ed l
ign
in c
onte
nt,
%
Time, (s)330W 550W 770W 1100W 110 W
61
4.2.3 MaterialBalance
A material balance analysis of the wood fiber extracted at different power inputs for the
maximum duration at each power input has also been performed to demonstrate the difference in
the yield of xylan obtained at different power levels and is given in table 4.3. It is noticed from
the table that as the power input or energy increased, the difference in the mass balance is also
increased and this may be due to the formation of other degradation products, such as sacharinic
acid from sugar degradation, formed as a result of the rapid generation of high temperature.
Higher amount of solubilized wood, lower yield of precipitated xylan and higher amount of
degraded products for shorter duration of microwave irradiation at higher power levels indicates
degradation of the carbohydrates.
Table 4.3. Material balance analysis of wood fibers after microwave assisted extraction Material 110W, 10min 330W, 3 min 550W, 3 min 770W, 1min 1100W, 1min Initial Wood fiber (g) Solid residue after extraction, (g) High mol.wt. xylan (g) Soluble lignin (g) Total soluble sugar after precipitating xylan (g) Total (g)
Time temperature combination of the microwave assisted and conventional extraction used in the
second part of the study are given in table 5.2. The temperatures selected for each conventional
extraction performed was the final temperature of the slurry obtained after the same duration of
microwave assisted extraction. Since conventional heating has a long thermal lag time, NaOH
solution was preheated to the required temperature before being mixed with the wood fibers
(which is previously weighed and kept at 40oC). The protocol used for the separation of wood
fibers and precipitation of xylan was same as the procedure described in Chapter 3. In this study,
5 g of the wood fibers (OD basis), prepared from birch wood logs obtained from premises near to
the University of Toronto, were extracted using 50 mL of 4 wt% NaOH solution ((solid to liquid
ratio of 1:10 g:mL). A higher amount of material (volume of solution) was used in this study to
prevent evaporation of the solution due to longer duration of extraction.
Table 5.2. Time-temperature combinations used in the microwave and conventional extractions *Microwave Conventional (isothermal) Temperature , (oC) 37,60, 80, 95, 98,100 37, 60, 80, 95, 98, 100 Time, (min) 1, 5, 10, 20, 30, 40 1, 5, 10, 20, 30, 40 * final temperature of the slurry after extraction
microwave extraction performed is given in table 5.3. The slurry temperature varied from 56oC
to 100oC during the second stage of wood dissolution in the microwave extraction. The
conventional extraction performed is an isothermal extraction at a temperature of 90oC, which is
closer to the temperature generated after 8 minutes of microwave extraction. Considerably
higher rate of dissolution observed at this stage during microwave extraction can be attributed to
the rapid increase in temperature of the fiber slurry which increase the hydrolysis and the
dissolution of the wood components. The maximum amount of wood dissolved during extraction
is about 29-32% under the experimental conditions studied using conventional and microwave
extraction. The time required to obtain the maximum dissolution is 240 minutes in conventional
extraction whereas it took only 18 minutes in microwave extraction indicating significant
reduction (about 13 times) in the extraction duration using microwave heating. It is anticipated
that about 10 degree rise in temperature double or triple the rate of the reaction, and by changing
the temperature from 90oC to 100oC, the rate of dissolution can be doubled or tripled. However,
the rate of dissolution of wood in the second stage of microwave extraction is significantly
77
higher (19 times) than conventional extraction. In most of the microwave heating applications
spectacular accelerations was observed, due to the efficient internal heating produced by the
direct coupling of microwave energy with the molecules present in the reaction mixture
(Kingston and Haswell, 1997; Kappe, 2008). Lewis et al. (1992) observed enhanced reaction
kinetics during microwave assisted chemical reactions and the authors attributed this
phenomenon to the non-uniform energy distribution at the molecular level during the chemical
reactions and are reported as the major microwave effects produced. It was assumed that,
microwave heating can induce “hot spots”, in the biomass due to the selective interaction of the
microwaves with the inhomogeneous lignocellulosic materials, and disrupt the structure of the
materials. For example, Hu and Wen, (2008) found that microwave treatment enhances surface
disruption and breaking of lignin structures in alkaline treated switch grass, whereas
conventional heating the fibers remain intact. They also reported that microwave treated switch
grass undergoes easy enzymatic saccharification to produce sugars compared to the conventional
treated one. These studies leads to the speculation that the significant enhancement observed in
the rate of wood dissolution or reduction in the time of extraction could be due to the combined
effect of thermal and the "microwave effects” caused during the microwave extraction. It is
hypothesized that the selective heating feature of microwaves with the lossier component in the
reaction system leads to direct interaction of the microwaves with the alkali present in the fibers,
and the temperature rise inside the fibers produces an “explosion” effect in the wood fibers. This
leads to the rupture of the fiber structure and facilitates the hydrolysis and diffusion of the
hydrolyzed components to the solution. Morphological study of the fibers before and after
extraction (discussed below) supported this hypothesis.
78
Figure 5. 3. SEM photomicrographs of wood fiber before and after extraction. (a) before extraction (b) after conventional extraction (90 minutes at 90oC) under two different magnifications (c) after microwave extraction (10 minutes at a power level of 110 W) under two different magnifications
SEM photomicrographs of the wood fibers before and after both extraction processes are shown
in Figure 5.3. Figure 5.3a represents the fiber before extraction. Fiber surface becomes smooth
after both extractions indicating the dissolution of debris present on the fibers. After
conventional extraction, no significant change is noticed in the overall fiber structure (Figure
5.3b1 and b2), whereas the fibers after microwave extraction becomes more porous (Figure 5.3c1
(a)
(b1) (b2)
(c1) (c2)
79
and c2.). The more porous structure leads to enhanced hydrolysis, faster diffusion and dissolution
of the hydrolyzed components. An attempt to quantify the micro structural changes has also been
performed and will be discussed in Chapter 6.
Though there is difference observed in the rate of wood solubilization during conventional and
microwave assisted extraction it is essential to have similar experimental conditions for both
processes to accurately differentiate the two processes. So, another set of conventional
experiments were performed at different temperatures for the same duration as that of microwave
extraction to separate the temperature effects from the microwave effects. In this part of the
study different microwave extractions were performed for different extraction duration (time of
extraction vary from 1 minute to 40 minutes) as given in the table 5.2. Temperature of the wood
fiber slurry was noted after each microwave extraction. Conventional experiments were
performed for the same duration as that of microwave extraction, in an isothermal extraction at a
particular temperature, which is the final temperature of the slurry after the microwave extraction
for that particular time of extraction (Experimental conditions are given in table 5.2). Further, to
decrease the temperature gradient occurred during the conventional heating of the wood slurry,
and for faster attainment of the temperature, the weighed wood fibers were kept at 40oC and
NaOH solution was preheated to the required temperature before mixing with the wood fiber.
The amount of wood solubilization, calculated using the equation 4.1, during microwave assisted
and conventional alkaline extraction of wood fibers at different duration of extraction is shown in
Figure 5.4. The results indicated the wood dissolution behavior is similar in both cases. Again,
for shorter duration the difference in the wood dissolution is not separable from each other and
this confirms that early stage wood dissolution was mainly due to the dissolution of the surface
80
Figure 5. 4. Effect of microwave and conventional extraction of xylan on the solubilization of
wood
debris on the fiber. As the duration of extraction increased, temperature of the slurry increased
and the rate of wood dissolution and amount of solubilized wood increased in both extractions.
Higher wood dissolution was expected during conventional extraction, as the temperature for the
extraction was kept constant at that particular temperature, whereas in microwave assisted
extraction, the temperature of the slurry increased fast and reached to the final temperature. The
rate of wood solubilization in microwave extraction [(Y/t)Microwave = 0.45] is greater than the
rate of wood dissolution in conventional extraction [(Y/t)Conventional = 0.39] . The rate of wood
dissolution and the amount of solubilized wood in microwave extraction would have been the
same or lower than the conventional extraction if the wood dissolution is only because of the
increase in the temperature. However, it was found that the rate of dissolution and the amount of
wood dissolved for particular duration is higher in the microwave extraction. This confirms the
hypothesis that the synergic effect observed during microwave extraction is not only due to the
y = 0.4508x + 16.37R² = 0.9911
y = 0.3888x + 16.09R² = 0.9608
0
5
10
15
20
25
30
35
40
0 10 20 30 40 50
Wo
od
dis
so
luti
on
(%
)
Time (minutes)
Microwave Conventional
81
temperature effect on the hydrolysis and wood solubilization, but also due to the structural
changes caused by the "explosion effect" that favors hydrolysis and mass transfer of the
hydrolyzed components to the solution. The hypothesis will be verified later by analyzing the
physical structural changes of the wood fibers after the two different extraction processes.
The yield of precipitated xylan calculated using the equation 4.2 for different duration of
extraction under two different experimental conditions (table 5.1) is shown in the Figure 5.1. It is
clear from the figure that under these experimental conditions, time required to extract a definite
amount of xylan during microwave assisted extraction is significantly lower compared to the
conventional extraction. In order to understand how conventional and microwave extraction
affects the yield of xylan, the graphs were re-plotted with time as done in the case of wood
dissolution and is shown in Figure 5.5. The figure pointed out that the yield obtained during
both extractions followed a similar path; increased very fast initially (stage 1), then a slow and
steady increase (stage 2), which is followed by a decrease in yield (stage 3). Yield of xylan
during alkaline extraction depend on the following simultaneous and multiple reactions and or
processes involved in the system; (i) hydrolysis of easily hydrolysable fraction of xylan and their
dissolution to the solution (ii) hydrolysis of xylan from the relatively tougher zone of the fibers
(iii) degradation (peeling) of the xylan to low molecular weight fractions and sugars, and (iv)
further degradation of the sugars to other products such as sacharinic acid. The yield of xylan
obtained during both the processes under the experimental conditions can be explained with
these different reactions involved and are discussed below.
82
Figure 5.5. Comparison of the yield of xylan (based on the total xylan) obtained during microwave and conventional extraction. (a) Microwave extraction; (b) Conventional extraction
y = 33xR² = 0.9555
y = 3.0537x + 29.956R² = 0.9834
y = ‐0.5949x + 64.917R² = 0.899
0
10
20
30
40
50
60
70
0 5 10 15 20 25 30
Yield of xylan (%
)
Time (Minutes)
y = 3.312xR² = 0.9071
y = 0.3045x + 30.637R² = 0.9027
y = ‐0.1045x + 66.303R² = 0.951
0
10
20
30
40
50
60
70
0 50 100 150 200 250 300
Yie
ld o
f x
yla
n (
%)
Time (Minutes)
(a)
(b)
83
It is clear from the Figure 5.5 that a portion of xylan can be extracted easily within a short
duration of extraction (within 1-2 minutes for microwave extraction and within 10 minutes
during conventional extraction). This represents the yield obtained from the susceptible or easily
hydrolysable fraction of xylan present in the birch wood. This easily accessible fraction of the
xylan obtained during the initial stages in both processes of extraction is about the same and is
found to be around 32%. After this period of extraction, yield of xylan obtained increased with
the duration of extraction up to about 10 minutes in microwave heating (Figure 5.5. a) and
decreased thereafter. It is expected that after the removal of easily accessible xylan, hydrolysis
and dissolution of the relatively inaccessible xylan and the degradation of the already dissolved
xylan occur simultaneously and the yield obtained is possibly due to the result of these two
processes. The observed increase in the yield of xylan up to 10 minutes of microwave extraction
indicates that at this stage, hydrolysis and dissolution of the xylan is higher compared to
degradation of the dissolved xylan. Temperature of the wood slurry increased from 56oC to about
96oC during this period. A similar observation was found during 10 to 90 minutes of
conventional extraction. Lower yield of xylan after 10 minutes of microwave extraction
demonstrates that during this period, degradation of xylan is more compared to the hydrolysis
and dissolution. This could be due to the longer duration of exposure of the xylan at high
temperature (> 95oC). Such a decrease was observed in the conventional extraction at about 120
minutes of extraction (Figure 5.5.b).
Similar to the wood solubilization, a comparison of the slope of the different stages of extraction
was analyzed to investigate the effect of heating on the extraction of xylan. It was found that
rate of extraction of xylan in microwave assisted process is significantly higher than the
84
conventional extraction. The rate of xylan extraction in the first and second stages in the
microwave and conventional extraction is as follows:
where Y is the percentage yield of xylan calculated based on the amount of xylan present in the
raw wood, and t is the time of extraction in minutes. At both stages the rate of extraction of xylan
is almost 10 times higher than the conventional extraction. The increased rate of the xylan
extraction in the microwave assisted process can be explained using the selective heating ability
of the microwave energy and the resulted exploded structure as in the wood dissolution.
However, the sudden increase in temperature may also leads to the degradation of the already
dissolved xylan and hence the yield of high molar fraction of xylan. The degradation becomes
more prominent for longer duration, as the temperature of the slurry is higher and leading to a
lower yield of xylan. It is found that maximum yield is obtained at about 10 minutes and
thereafter, the yield is decreased during microwave assisted extraction. Similarly in conventional
extraction, maximum yield is obtained at about 90 minutes and thereafter the yield of xylan is
decreased.
The degradation of the dissolved xylan was confirmed by quantifying the total sugar content of
the supernatant after precipitating the xylan (Figure 5.6). The Figure shows that as the duration
of the extraction increased the sugar content left in the supernatant also increased and is
attributed to the degradation of the high molecular weight xylan. However, the difference in the
dissolved sugar content obtained in both processes was found to be more or less similar. As a
85
result of longer duration extraction at 90oC, there may be degradation of the carbohydrates that
not only leads to sugar but also to other products such as sacharinic acid, which are not
quantified in the process, and this might be the reason for a low xylan yield.
Figure 5.6. Sugar content of the supernatant after precipitating the extracted xylan using two different processes
Mass balance analysis of the wood fibers after the two extraction processes, where the maximum
yield of xylan obtained has also been performed (table 4.3) to demonstrate the degradation of
xylan to other products. The wood residue left after both extractions was found to be more or
less similar, indicating a similar amount of wood dissolution. However, the mass balance
difference between the original fiber and the total amount after extraction is higher in the
conventional extraction (4.2 g) compared to microwave extraction (1.22 g). The increased
0
1
2
3
4
Su
ga
r Co
nte
nt,
%
Time, minutes
90C110W
10 30 60 90 120110W: 2 4 6 8 10
86
difference observed in the mass balance of the fibers after conventional extraction could be due
to the degradation of the xylan to other products as explained above.
Table 5.4. Mass balance analysis of wood fibers after xylan extraction
Material Microwave (10 minutes)
Conventional (90 minutes)
Original wood fiber (g) Wood residue left after extraction(g) Xylan Precipitated (g) Lignin dissolved (g) Total sugar left in the supernatant (g) Total (g)
I – Microwave extraction, II- conventional extraction
The percentage removal of xylan on the basis of the amount of xylan present in the original
fibers are shown in the Figure 5.9.A, which also confirmed that the microwave heating resulted
in faster xylan removal compared to the conventional extraction. Similar results were reported
earlier in the pre-treatment of corn stover using water and acid treatment (Shi et al., 2011),
alkaline pre-treatment of switch grass (Hu and Wen, 2008), and FeCl3 treatment of rice straw (Lu
and Zhou, 2011).
91
Figure 5.9. Effect of microwave and conventional extraction of xylan on the chemical
composition of wood (A) xylan removal (B) lignin removal (c) Glucan left in the fiber
0
10
20
30
40
50
60
70
5 /60 10 /80 20 /98 30 /98
% r
em
ov
al o
f x
yla
n f
rom
wo
od
(b
as
ed
on
ori
gin
al
xy
lan
co
nte
nt)
Time (minutes)/ Temperature (oC)
microwave convetnional
0
5
10
15
20
25
5 /60 10 /80 20 /98 30 /98
% r
em
ov
al
of
lign
in w
oo
d, (
ba
se
d
on
ori
gia
nl
lign
in c
on
ten
t)
Time (minutes)/ Temperature (oC)
microwave convetnional
(A)
(B)
0
10
20
30
40
50
60
5 /60 10 /80 20 /98 30 /98
% in
cre
as
e o
f G
luc
an
in t
he
wo
od
Time (minutes)/ Temperature (oC)
microwave convetnional (C)
92
The lignin left in the alkaline extracted wood decreased from 21.3 g per 100 g of oven-dry wood
fiber to 19 - 17.4 g lignin per 100 g oven-dry wood fiber over the duration of 5-30 minutes of
microwave assisted extraction, and a similar decrease in lignin content was observed in the
conventional extraction for the same duration of extraction. The percentage removal of lignin
on the basis of the original lignin content in the wood fibers (Figure 5.9B) showed that about 11
to 18 % of lignin removed during the period of 5-30 minutes of microwave assisted extraction
and no significant difference was observed during the conventional extraction.
Unlike xylan and lignin reduction after alkaline extraction, glucan content increased as a result of
both microwave and conventional extraction. This is because of the removal of xylan and lignin
from the lignocellulose matrix which makes the fiber richer in cellulose. The glucan left in the
extracted fiber using microwave extraction increased from 42.3 g glucan per 100 g of oven-dry
wood fibers to 54.3-60 g of glucan per 100g of oven-dry wood fibers. The increase during
conventional extraction from the similar experimental conditions varied from 52.7-58.6 g. The
increase in the glucan content on basis of the original glucan present in wood fibers with the
time/temperature conditions are shown in Figure 5.9C, confirming that the increase in glucan in
the microwave extracted fibers are higher and is due to the higher removal of xylan and lignin
from the wood fibers during microwave assisted extraction. Comparatively higher amount of
xylan, and lignin removal (lower xylan and lignin, and higher glucan) is expected in the fibers
after conventional extraction, as the conventional performed isothermally at the final temperature
of the microwave assisted extraction. However, the lower removal of xylan and lignin from the
fibers after conventional process compared to microwave assisted extraction again indicates the
role of “microwave explosion effects" in enhancing the hydrolysis and dissolution of xylan.
93
5.3.3.2 FT‐IRspectraofwoodfibers
Fourier transform infrared spectroscopy (FTIR) is a very powerful tool for obtaining rapid
information about the chemistry of wood, wood constituents and chemical changes taking place
during weathering, decay and chemical treatments, thermal treatments or natural aging (Faix et
al., 1993; Korner et al., 1992; Rodriguez et al., 1998; Popescue et al., 2007, 2009,2010). FTIR is
used in this study to see the chemical structural changes happened during the alkaline extraction
and to evaluate if there is any difference in the structure of wood after microwave extraction
compared to conventional extraction. To simplify the complexity of the spectra, they were
separated into two regions, namely the OH and CH stretching vibrations in the 3900 -
2700cm-1 and the finger print region in 1900-800 cm-1 (shown in Figure 5.10). The major peaks
assigned for wood taken from different references (Korner et al., 1992; Faix et al., 1993; 1998;
Pandey, 1999; Moore and Owen, 2001; Kubo and Kadla, 2005; Popescue et al., 2007, 2010) and
that obtained for wood after both extraction processes are given in the table 5.6. The broad band
at 3400 cm-1 represents the hydrogen bonded O-H stretching due to the presence of aliphatic
hydroxyl groups of cellulose, hemicelluloses and lignin. The peaks at 2936 cm-1, 2898 cm-1, and
2847 cm-1 represents C-H symmetric stretching in methyl and methylene groups, asymmetric
methoxyl stretching and symmetric CH2 stretching respectively. The enhancement in the
intensity of these peaks is more in the microwave assisted extracted fibers compared to the
conventionally extracted fibers. There are many well defined peaks in the finger print region of
1900-400 cm-1. The bands at 1743 cm-1 assigned to C=O stretching of acetyl and carbonyl
groups of hemicelluloses, and the disappearance of this peak suggests the susceptibility of the
removal of xylan during alkaline hydrolysis. The peak at 1645 cm-1 represents the absorbed OH
94
Table 5.6. Peak assignment of wood fibers before and after different extraction processes
Band assignment Wave number, cm-1
Wood before extraction
Wood after conventional extraction
Wood after microwave extraction
OH and CH stretching region Multiple formation of intra-molecular hydrogen bonds between phenolic groups and that of alcoholic groups
3421
3352
3354-3310
C-H stretching in methyl and methylene groups 2936
2936
2937
Asymmetric methoxyl C-H stretching 2898 2899 2899
Symmetric CH2 stretching 2847 2844 2844
Finger print region Unconjugated C=O in xylans
1737 (1743)
-
-
Absorbed O-H and conjugated C=O; Xyloglucan C=O vibration of carboxylic acids
1645 Decreased intensity; peak observed at 1658
Decreased intensity; peak observed at 1658
C=C of aromatic skeletal (lignin)
1596, 1506
1596, 1506
1595, 1506
C-H deformation in lignin and carbohydrates
1464,1426
1464,1426
1464,1426
C-H deformation in cellulose and hemicelluloses
1372
Decreased intensity
Decreased intensity
C-H vibration in cellulose and C1-O vibration in syringyl derivatives
1329
1329
1331
Guaiacyl ring breathing, C-O stretch in lignin and for C-O linkage in guaiacyl aromatic methoxyl groups
1269
Decreased intensity
Decreased intensity
Syringyl ring and C-O stretch in lignin and xylan 1242
1232 1235
C-O-C vibration in cellulose and hemicelluloses
1162
1162
1162
Aromatic skeletal and C-O stretch
1115
1121
1123
C-O stretch mainly from C3 and C5 in cellulose
1055
1055
1056
C-O and C-C stretching ring in celluloses and hemicelluloses
1034
1034
1035
Pyran ring stretching 897 (898) 898 897
95
2600280030003200340036003800
wood
M-5
M-10
M-20
M-30
(A1)
26002800300032003400360038004000
wood
C-5
C-10
C-30
C-20
Ab
sorb
an
ce U
nits
Wave number, cm-1
(B1)
96
Figure 5.10. FTIR spectra of wood fibers before and after microwave and conventional extraction (A1 and A2 : microwave assisted extraction and B1and B2 conventional extraction). M5-M30 and C5-C30 represents 5 - 30 minutes of extraction using microwave and conventional heating respectively.
80010001200140016001800
wood
M-5
M-10
M-20
M-30
Ab
sorb
ance
Un
its
Wave number, cm-1
(A2)
80010001200140016001800
wood C-5
C-10
C-30
C-20
Absorbance Units
Wave number, cm‐1
(B2)
97
and conjugated C= O (Ph-C=O) absorption and the intensity of this peak is decreased with the
time of extraction. After 5 minutes of microwave assisted extraction the intensity of this peak
decreased considerably, whereas in conventional extracted fibers the peak intensity decreased
only after 10 minutes of extraction, indicating faster removal of ester linkages between
hemicelluloses and lignin. The observed peak at 1658cm-1 in the fiber after 5 -10 minutes of
extraction is due to the absorbed OH peak in the fibers and may be due to the increase in the OH
concentration as a result of the hemicellulose/lignin hydrolysis. The peaks at 1596 and 1506 cm-1
represents the C=C stretching of lignin skeletal vibration, and C-H deformation in lignin
respectively. Intensity of the peak at 1329 cm-1 represents C-H vibrations in cellulose and C1-O
vibrations in syringyl derivatives of the lignin moieties, whereas the peak at 1269 cm-1 represents
guaiacyl ring breathing. The decrease in the intensity of the peak at 1329 cm-1 indicates the
susceptibility of syringyl groups for the alkaline hydrolysis. The intensity of the peak at 1260 cm-
1 (1269 - 1242 cm-1) decreased significantly in the extracted fibers indicating the ether bonds
(aromatic-C-O-C-aliphatic) also undergoes changes in the alkaline extraction. The intensity
decrease is higher in microwave assisted extraction compared to conventional extraction. This
could be due to the increased breakage of the ether linkages (a higher energy reaction compared
to ester bond hydrolysis) between lignin and hemicelluloses caused by the focused temperature
rise in the fiber matrix. The peaks at 1055 and 1034 cm-1 represents the C-O and C-C stretching
in cellulose and hemicelluloses respectively. The sharp band at 898 cm-1 is characteristic of β-
(1,4)-glycosidic linkages between the sugar units in cellulose and hemicelluloses (characteristic
peak of pyran ring stretching). However, it has to be noted that many of the peaks in this region
is the overlap of frequencies of lignin and carbohydrates.
98
5.3.3.3 Crystallinityofwoodfibers
Figure 5.11. XRD crystallographs of birch wood fibers before and after microwave and
conventional extraction
0
400
800
1200
0 10 20 30 40 50
Inte
nsi
ty, A
rbit
rary
un
its
Angle, 2, degree
Birch wood
0
400
800
1200
0 10 20 30 40 50
Inte
ns
ity,
Arb
itra
ry u
nit
s
Angle, 2, degree
5 minutes 10 minutes 20 minutes 30 minutes
0
400
800
1200
0 10 20 30 40 50
Inte
ns
ity,
Arb
itra
ry u
nit
s
Angle, 2, degree
5 minutes 10 minutes 20 minutes 30 minutes
Conventional
Microwave
99
Crystallinity of wood fibers is important as it determines the downstream application of the
fibers after xylan extraction including enzymatic hydrolysis, pulping to produce cellulose fibers,
and for using reinforcements in bio-composites. The removal of lignocellulosic components such
as hemicellulose (xylan) and lignin during the extraction process can change the crystal structure
of cellulose by altering inter- and intra- hydrogen bonds present in the lignocellulosic matrix. X-
ray diffraction patterns of the wood fibers before and after microwave and conventional alkaline
extraction of wood fibers are shown in Figure 5.11 and the crystallinity index, which is a
measure of crystallinity, calculated using the equation 3.7 is given in table 5.7. The crystallinity
index of original wood fibers is 41.8% and this value increased after both extraction processes,
indicating the removal of amorphous portion of the lignocellulosic matrix. Crystallinity index
increased to about 61-62% after 10 minutes of extraction and thereafter decreased in both
extraction processes. The reduction in the crystallinity at high temperature and for longer
duration may be due to the changes in the crystal structure of the cellulose during extraction. The
Table 5.7. Crystallinity index of wood fibers before and after different extraction processes Sample ID Time
(Minutes) Intensity of I002 Intensity at
2=18o
Crystallinity index (%)
Birch wood fibers Wood fibers after microwave extraction Wood fibers after conventional extraction
- 5 10 20 30 5 10 20 30
870 1260 1136 1296 1246 1149 1024 1129 1183
506 576 433 586 586 475 396 454 502
41.8 54.3 61.9 54.8 54.7 58.7 61.3 59.8 57.6
100
reduction in crystallinity of the fibers after microwave assisted extraction is higher compared to
the conventional extraction, confirming the structural explosion that affects the crystal structure
of cellulose. The structural explosion was further confirmed by the SEM analysis of the fibers in
the following section. The high crystallinity of these extracted fibers makes them suitable for the
production of cellulose fibers/ cellulose nano fibers and also as the reinforcements for the
plastics, as the strength of the fibers increases with the increase in crystallinity.
5.3.3.4 Scanningelectronmicroscopy(SEM)
SEM images of the wood fibers before and after microwave assisted and conventional alkaline
extraction are given in the Figure 5.12. The fiber morphology has changed after both extraction
processes. However the changes observed are different from each other. After conventional
extraction, for all the time and temperature combinations, the fiber structure becomes smooth
and intact and as the time of extraction increased the fiber become striated and became thinner as
a result of the amorphous hemicellulose and lignin removal from the matrix. Unlike conventional
extraction at all duration of extraction, the fiber structure is fibrillated and a more loosened or
porous structure was observed in fibers after microwave extraction, and as the time increased
more fibrillar structure can be seen with thinner fiber surfaces. This porous structure of the wood
after the microwave assisted extraction explained the higher dissolution of the wood components
and the decreased crystallinity for longer duration of extraction. Further, this confirms the
hypothesis that the explosion effect caused by the interaction of microwave and alkali in the
fibre increase the porosity and loosens the recalcitrant fibre structure thus enhancing the
hydrolysis and the mass transfer of the dissolved components to the solution.
101
Figure 5.12. SEM photomicrographs of birch wood fibers before and after microwave and
conventional extraction (A) wood fiber before extraction ; M -microwave extraction and C- conventional extraction
(A)
102
5.3.4 XylanCharacterization
5.3.4.1 Chemical composition of xylan
Chemical composition of xylan isolated using microwave and conventional extraction for
different duration are given in table 5.8. Xylan contained about 78-93% xylose, and 0.3-1.1%
glucose, and 2-3.8% lignin, indicating the polymer extracted was a polymer with xylopyranose
backbone.
Table 5.8. Chemical composition of xylan isolated under different extraction conditions Sample ID Glucose (%) Xylose (%) Lignin (%) XM- 5minutes XM- 10 minutes XM- 20 minutes XM- 30 minutes XC- 5minutes XC- 10 minutes XC - 20 minutes XC- 30 minutes
average degree of polymerization is more informative rather than the viscosity average molecular
weight. Hence, conversion of intrinsic viscosity value into number average degree of
polymerization was carried out according to the folowing equation [η]CED = DPn x 4.7 x10-3
(Lebel et al., 1963), where DPn is the number average degree of polymerization. The intrinsic
viscosity, viscosity average molecular weight, and number average degree of polymerization of
xylan obatined at different experimental conditions are given in the table 5.9. It is clear from the
table that intrinsic viscosity decreased (22.5%) during 5 - 30 minuts of conventional
extraction,whereas viscosity values remain relatively constant during microwave assisted
extraction until about 20 minutes of extraction and then decreased rapidly (15.6 %). Further, in
both extractions, the decrease in the degree of polymerization is significant after 20 minutes of
extraction. This again confirms that higher temperture above 95oC enhanced the degrdation of
xylan in alkali. However, the percentage decrease in the intrinsic visocisty is lower (15.6%)
compared to the xylan obtained by convetnional method implying that molecular degradation is
higher during convetnional extraction. Calculated viscosity average molecular weight of the
xylan is in the range of 18000-19000 and the corresponding degree of polymerization is about
150. Such a degree of polymerization of birch xylan was reported earlier by Lebel et al. (1963).
Similar to viscosity, degree of polymerization is decreased (from 149-115) during convetnional
extraction, but remains almost constant during microwave extraction until 20 minutes of
extraction and then decreased to 126. The degradation of the xylan during convetnional
extraction may be due to the prolonged exposure of the dissolved xylan to high temperature.
These findings support the degradation of xylan and hence the lower yield of high molecular
weight xylan during convetnional extraction due to prolonged extraction duration.
105
Figure 5.14. ηsp/C vs. Concentration of xylan solutions in CED (a) Xylan obtained using microwave extraction (b) Xylan obtained using conventional extraction (C and M represents conventional and microwave extraction respectively; 5,10,20 and 30 represents the duration of time in each process)
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
0 0.2 0.4 0.6 0.8
ηs
p/C
(d
L/g
)
Concentration of xylan solution (g/dL)
M‐5 M‐10 M‐20 M‐30
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.2 0.4 0.6 0.8
(ηs
p/C
) (d
L/g
)
Concentration of xylan solution (g/dL)
C-5 C-10 C-20 C-30
106
Table 5.9 Intrinsic viscosity, viscosity average molecular weight and number average degree of polymerization of xylan obtained by microwave convetnional extraction Sample ID Temperature
X-ray computed microtomography (micro CT) has been used for qualitative as well as
quantitative analysis in a number of research applications including soil science, geology,
hydrology, and material science (Thovert et al., 1993; Perret eta l., 1999; Wildenschild et al.,
2002; Muler et al., 2002; Salvo et al., 2003). In this technique, a series of radiographic
projections at different viewing angles of the samples are reconstructed mathematically to
visualize and analyse the architecture of the materials with a micrometer resolution. The
visualization of the internal structure of the materials depends on the travel of the X-rays through
the structure which in turn depends on the mass density and absorption coefficient of the
materials of interest. Literatures are available on the use of X-ray CT for the analysis of the pore
structures of fruits and vegetables such as internal quality changes of peaches and cucumber
during storage, microstructure of cereal products, and pore space analysis of apples (Harker and
Hallet, 1992; Barcelon et al., 1999; Mendoza et al., 2007). The advantage of this process over the
conventional microtomy is that the analysis can be done without any special sample preparation
or chemical fixation which usually a time and effort intensive process in the conventional
microtomy analysis. Moreover, X-ray micro CT enables the study of 3- dimensional internal
Results of this chapter have been published in Journal of Wood Chemistry and Technology.
Volume 33, Issue 2, June 2013, pages 92-102
109
microstructure of the materials of interest. The method has been recently used for the
quantification of the plant microstructure (Fromm et al., 2001; Stuppy et al., 2003; Steppe et al.,
2004; Mayo et al., 2009). It has been reported that the micro CT method can be used effectively
to quantify the microstructure of the wood; particularly the xylem anatomy and porosity
provided a proper image analysis method is used for the processing of the images (Steppe et al.,
2004). In this study, we tried to make an attempt to quantify the structural changes of the wood
during conventional and microwave assisted alkaline extraction using X-ray computed
microtomography to understand the difference in the mechanism of microwave and conventional
method of extraction. Further, we used this technique as an indirect means for measuring the
temperature generated inside the wood fibers during microwave extraction.
6.2 Material
The birch wood chips used in this study were obtained from St. Mary's Paper, Ontario, Canada.
The chips were oven dried at 40oC for 72 hours. Rectangular samples with approximate
thickness of 1-2 mm and 3-5 mm were cut from the dried wood chips. The samples were soaked
in a 4 wt% solution of sodium hydroxide (with a sample to liquid ratio of 1:10) for 2 hours to
completely wet the chips. The samples with the solution were then subjected to microwave
irradiation at a power level of 110 W for different times.
Another set of samples were extracted using the conventional method of heating using an oil
bath for different duration at different temperatures. For temperatures higher than 100oC,
extractions were performed using small laboratory digestion cylinders. Extraction time was
determined after the temperature of the solution reaches the set temperature. Extraction
conditions used in this study are given in the table 6.1. After extraction, the samples were
110
washed thoroughly and air dried overnight. Three different samples from each set were used for
the X-ray CT imaging to accommodate the variability of the wood samples.
Table 6.1. Experimental conditions used for extraction
Extraction conditions
Conventional Microwave assisted
Temperature Time (minutes)
Power output Time (minutes)
70oC 90oC 100oC 120oC 140oC
30 60 120 120 10 10 10
110W
1 6 10
6.3 X‐raycomputedmicrotomographyimageprocessing
The procedure used for the 3D construction and quantitative analysis of porosity is demonstrated
in Figure 6.1. The scanned images were reconstructed to grey scale image using the NRecon
software from Skyscan. The grey scale images were converted to black and white binary images
to specifically select the voids from the matrix, and were done by the segmentation technique.
The simplest method for segmentation is thresholding. In the figure (Figure 6.1) the bright areas
represent the xylem vessels whereas the dark areas represent the remaining wood matrix.
The threshold value should define the boundary between the objects in the area of interest and
should be selected in between the intensity of both the void and the solid phases, as the selection
of threshold highly influences the analysis. A similar procedure used by Steppe et al. (2004) was
111
Figure 6.1. Binary images of birch wood and the 3D image of the vessels extracted from the binary images (the 3D structure consists of a finite number of voxels with a length of voxel side of 7.4 um)
used to reduce artifacts from the image. A global thresholding followed by a morpho-
mathematical filter with a rounding structural element (available in Image Pro software) was
used to extract the vessel elements. The 2D images were then converted to 3D images. After
processing, the images were labeled and performed a quantitative measurement of each
individual in the volume of interest (VOI) to get the volume of the voids. Porosity of the wood
To determine the correlation between the thickness of the samples and the porosity of the fiber
structure during alkaline extraction, samples of thickness less than 2 mm were used for the
extraction and computation of porosity. Typical 3-dimensional tomographic images are shown in
the Figure 6.6 and the calculated porosities are given in the table 6.3. It is clear from the figure
119
Figure 6.6. Typical 3D tomographic images of wood samples, samples after conventional and
microwave assisted extraction process (sample size < 2 mm) (a) wood (b) wood after 2 hours of extraction @90oC (c) wood after 10 minutes of microwave assisted extraction
and table that porosities are different for samples after conventional and microwave extraction.
Porosity is increased with temperature in the case of conventional extraction; however, there is
no significant difference observed between thicker and thinner samples (table 6.2 & 6.3). On the
other hand, after microwave extraction thinner samples exhibit a higher porosity. As the sample
(a)
(b) (c)
120
Table 6.3. Calculated porosities of wood samples after sodium hydroxide extraction using conventional heating and microwave irradiation process (sample size < 2 mm)
Sample Porosity (%) Wood Wood after 2 hours of extraction @70oC Wood after 2 hours of extraction @90oC Wood after 10 minutes of microwave assisted extraction
18.2 ± 0.4 20.1 ± 0.9 22.9 ± 1.2 25.9 ± 0.8
Table 6.4. Comparison of the calculated porosities of wood samples after 10 minutes of sodium hydroxide extraction using conventional heating and microwave irradiation process ( sample size < 2 mm)
Sample Porosity (%)
Wood Wood after extraction @100oC Wood after extraction @120oC Wood after extraction @140oC After Microwave extraction (10minutes)
significance of the model developed. Analysis of variance (ANOVA) of the model for each
response was also performed to study the significance of the regression coefficient of
determinations for each effect (linear, quadratic, and interaction) in the model, and to assess
response model fit to the data (the lack of fit parameters), and hence to evaluate the significance
of the models for further rationalization. The competence of the model was also evaluated using
adjusted R2 and predicted R2, rather than R2 values, as the value of R2 increase with the addition
of variables despite the significance of the added variables.
130
7.4.2 Optimizationoftheprocessingvariables
The microwave extraction process was optimized using the numerical optimization approach
available in the software with the objective of maximizing the yield of xylan. Four additional
experiments were performed using the optimal conditions selected and the results (responses)
were compared with the predicted responses to verify the validity of the surface response model
developed.
7.5 ResultsandDiscussion
The results of the optimization experimentation of the four variables (extraction time,
concentration of alkali, solid to liquid ratio, and sample size) on the responses (wood dissolution,
yield of xylan, and temperature generated) are given in the table 7.2. The surface response
Table 7.3. Polynomial equations for the quadratic model and the regression coefficients Quadratic model equations (A= extraction time, B= NaOH concentration, C= solid to liquid ratio, and D= sample size)
Figure 7.2. Response surface plots showing the interaction between the variables affecting the temperature of wood slurry (A) interaction between time and NaOH solution concentration (B) interaction between time and solid to liquid ratio (C) interaction between time and sample size (D) interaction between sample size and NaOH concentration and (E) interaction between sample size and solid to liquid ratio
between NaOH solution concentration and time (at a constant level of sample size and solid to
liquid ratio: 7.5 g, and 1:14 g/mL), and that of NaOH solution concentration and sample size (at
a constant level of time and solid to liquid ratio; 17.5 minutes and 1:14 g/mL) on temperature.
The NaOH concentration under these experimental conditions did not affect the temperature of
the slurry. Increase in temperature leads to increased hydrolysis and leads to increased wood
dissolution, and so a high yield of xylan is expected at these conditions.
Wood dissolution during alkaline extraction occurs due to the hydrolysis of the lignin-
carbohydrate linkages. The amount of wood dissolved during the experimentation (Table 7.2)
varied from about 17 wt% (run #14) to 35% (run#19). The highest amount of wood dissolution
during the microwave assisted extraction was observed at high levels of extraction time (+1, 25
minutes) and concentration of sodium hydroxide (+1, 8 wt %), and low levels of sample size (-1,
5 g) and solid to liquid ratio (-1, 1:8 g/mL). On the other hand, the lowest amount of wood
dissolution occurred at a low level of sodium hydroxide (-2, 2 wt %) and mid levels of time (0,
14 minutes), solid to liquid ratio (0, 1:14 g/mL) and sample size (0, 7.5 g). It is clear that higher
concentration of alkali and longer duration of extraction increased the wood dissolution. Longer
duration of irradiation increases the temperature of the system (respective temperatures for the
low and high dissolution of wood were 93.5oC to 98.5oC) as described in the previous section
and this enhances the wood dissolution.
The simultaneous effect of the independent variables on the dissolution of wood is established by
the surface response quadratic model using the CCD (Table 7.3). The relatively high values of R2
(0.9962) and adjusted R2 (0.9927) of the response surface model indicate that the model
considers only the significant terms and so a good agreement between the experimental and the
predicted responses. The predicted R2 (0.9804) was also found to be closer to the adjusted R2
values indicating the adequacy of the model. Relatively smaller standard deviations of the
model (0.36) used for the prediction of wood dissolution indicated the better reproducibility of
the results. Further, adequate precision for the wood dissolution model was reported to be 69.30,
which is well above the required value of 4. The higher values of R2, smaller standard deviation
137
and high value of adequate precision of the model demonstrated the suitability of the model for
the prediction of the real relationship among the variables studied to the wood dissolution. Figure
7.3 demonstrates the strong correlation between the predicted and observed wood dissolution
during microwave assisted extraction of xylan.
Figure 7.3. Actual wood dissolution vs. predicted wood dissolution
The ANOVA for the response surface quadratic models for wood dissolution are given in the
table 7.4. The p-value for the quadratic model for the wood dissolution (<0.0001) and high F-
value (281.50) implied that this model is significant. From the regression coefficients of the
variables, it is clear that the linear terms of all the variables studied, interaction terms between
variables except the two interaction terms (time and sample size; and NaOH solution
concentration and sample size) and the quadratic terms were significant. Significance of the
2
15.00
20.00
25.00
30.00
35.00
15.00 20.00 25.00 30.00 35.00
Actual wood dissolution (%)
Pre
dict
ed w
ood
diss
olu
tion
(%) (a)
138
interaction terms between time and NaOH solution concentration (F value, 8.76, p-value 0.0097),
and that of solid to liquid ratio and sample size (F-value 13.48, p-value 0.0023) are less
significant than the interaction terms between time and solid to liquid ratio (F-value 24.15, p-
value 0.0002), and NaOH concentration and solid to liquid ratio (F-value 39.99, p-value
<0.0001).
The response surfaces of the significant interaction effects on the wood dissolution are shown in
Figure 7.4. The percentage of wood dissolution affected by different NaOH concentration and
extraction time was shown in Figure 7.4A, where solid to liquid ratio and sample size were kept
constant at 1:14 (g/mL) and 7.5 g respectively. Extraction time and NaOH concentration time
exhibited a positive impact on the wood dissolution and the wood dissolution increased with the
increase of time of extraction and sodium hydroxide concentration. Maximum wood dissolution
was obtained at an irradiation time of 25 minutes with a 8wt% NaOH solution. Figure 7.4B
depicts the interaction of time and solid to liquid ratio on the wood dissolution at a constant level
of NaOH solution (6 wt %) and sample size (7.5 g). The figure indicated that maximum wood
dissolution occurred for 25 minutes of extraction at a solid to liquid ratio of 1:8 g/mL. Effect of
interaction terms of NaOH solution concentration and solid to liquid ratio on the wood
dissolution is shown in Figure 7.4C. The extractions were performed using 7.5 g of wood sample
for 17.5 minutes. The wood dissolution increased as the level of NaOH concentration increased
139
Figure 7.4. Response surface plots showing the interaction between the variables affecting the wood dissolution (A) interaction between time and NaOH solution concentration (B) interaction between time and solid to liquid ratio (C) interaction between solid to liquid ratio and NaOH concentration (D) interaction between solid to liquid ratio and sample size
and a maximum of about 30 wt% was observed when the NaOH concentration was at 8 wt%, and
the solid to liquid ratio was 1:8 g/mL. Increase in the sample size and solid to liquid ratio has a
negative impact on the wood dissolution (Figure 7.4D), when the extraction was performed for
17.5 minutes using 6 wt% NaOH. Hence, 25 minutes of microwave irradiation of the slurry
using 8 wt% of NaOH solution at a solid to liquid ratio of 1:8 (g/mL) is expected to provide a
maximum amount of wood dissolution.
A B
C
D
5.00
6.00
7.00
8.00
9.00
10.00
8.00
11.00
14.00
17.00
20.00
25
26
27
28
29
Woo
d di
ssol
utio
n
C: Solid to liquid ratio
D: Sample size 8.00
11.00
14.00
17.00
20.00
4.00
5.00
6.00
7.00
8.00
22
24
26
28
30
32
Woo
d di
ssol
utio
n
B: NaOH C
C: Solid to liquid ratio
8.00
11.00
14.00
17.00
20.00
10.00
13.00
16.00
19.00
22.00
25.00
22
24
26
28
30
32
Woo
d di
ssol
utio
n
A: Time
C: Solid to liquid ratio 4.00
5.00
6.00
7.00
8.00
10.00
13.00
16.00
19.00
22.00
25.00
20
22
24
26
28
30
32
34
Woo
d di
ssol
utio
n
A: Time
B: NaOH Concentration
140
7.5.3 Effectofextractionvariablesonyieldofxylan
The yield of xylan (based on oven dry wood) obtained in the CCD design are given in the table
7.2. The lowest and highest yields of xylan obtained in the design space were 4.33% and 22.49%
respectively (run #14 and #15). Lowest yield of xylan observed for the experiment corresponds
to the lowest amount of wood dissolution where the levels of independent variables were:
sodium hydroxide (-2, 2 wt %), time (0, 14 minutes), solid to liquid ratio (0, 1:14 g/mL) and
sample size (0, 7.5 g). Contrary to the maximum wood dissolution (run# 19), the highest yield of
xylan (run #15) was obtained when all the independent variables are at the higher levels;
extraction time (+1,25 minutes), concentration of sodium hydroxide (+1, 8 wt%), solid to liquid
ratio (+1, 1:20 g/mL), and sample size (+1, 10 g). The yield of xylan obtained for run #19, where
maximum wood dissolution was observed, was 14.9%. The variation in yield can be attributed to
the rise in temperature of the wood slurry. The temperature of the slurry after run #19 was
98.5oC, whereas that of run #15 was 83.5oC. High wood dissolution is expected at higher
temperature, as the hydrolysis rate and dissolution of the hydrolyzed components (hemicelluloses
and lignin) increase with the temperature. However, the high temperature generated lead to the
degradation of the hemicelluloses resulting in a low yield of xylan, which is in agreement with
the earlier results discussed in Chapter 5.
The effect of independent variables on the yield of xylan was studied using the surface response
quadratic model given in the table 7.3. The model was found to be significant from the values of
regression coefficients (R2 = 0.9715 and adjusted R2 = 0.9450). The predicted R2 (0.8722) was
closer to the adjusted R2 and would give a good fit to the statistical model used. The observed
small standard deviation of the model (1.08) and the high adequate precision value (22.91) also
indicates the reproducibility and applicability of the quadratic model for the prediction of yield
141
of xylan. Figure 7.5 shows the predicted and actual yield of xylan obtained during the
experimental design studied and the strong correlation observed (R2=0.9856) indicates the
accuracy of the model.
Figure 7.5. Actual yield of xylan vs. predicted yield of xylan
Analysis of variance (ANOVA) of the regression model is given in the table 7.4 and was used to
evaluate the statistical significance of the model. The F-value for the quadratic model for the
yield of xylan is 36.59 with a p-value <0.0001 indicating the model is significant. The effect of
linear terms of extraction time, NaOH solution concentration, and sample size, interaction terms
between time and solid to liquid concentration and between solid to liquid ratio and sample size,
2
0.00
5.00
10.00
15.00
20.00
25.00
0.00 5.00 10.00 15.00 20.00 25.00
Actual yield of xylan (%)
Pre
dict
ed y
ield
of x
ylan
(%) (b)
142
and the quadratic terms of time, NaOH concentration, and solid to liquid ratio are found be
significant on the yield of xylan. The Lack of Fit F-value of 1.20 (p-value 0.4435) imply the
lack of fit is not significant relative to the pure error and hence the model can be used to predict
the yield of xylan in the present study.
Figure 7.6. Response surface plots showing the interaction between the variables affecting the
yield of xylan (A) interaction between time and solid to liquid ratio (B) interaction
between solid to liquid ratio and sample size
The response surface contour plots of the significant interaction terms are shown in Figure 7.6.
Figure 7.6A depicts effect of the interaction between time and solid to liquid ratio on the yield of
xylan (NaOH concentration and sample size was kept at 6 wt% and 7.5 g respectively), whereas
7.6B shows the effect of interaction between solid to liquid ratio and sample size (NaOH
concentration and time of extraction was kept at a constant level of 6 wt% and 17.5 minutes
respectively) on the yield of xylan. Larger sample size and smaller solid to liquid ratio resulted in
a higher yield of xylan (Figure 7.6B), whereas longer extraction time and larger solid to liquid
8.00
11.00
14.00
17.00
20.00
10.00
13.00
16.00
19.00
22.00
25.00
0
5
10
15
20
25
Yie
ld
A: Time C: Solid to liquid ratio 5.00
6.00
7.00
8.00
9.00
10.00
8.00
11.00
14.00
17.00
20.00
0
5
10
15
20
25
Yie
ld
C: Solid to liquid ratio D: Sample size
A B
143
ratio resulted comparatively higher yield. It is clear from all these results that larger sample size,
higher concentration of NaOH, longer irradiation time, and smaller solid to liquid ratio would
lead to a higher yield of xylan.
7.5.4 Optimizationofmicrowaveassistedextractionofxylanandvalidationofthe model
The objective of the study was to find the microwave assisted extraction conditions that result in
the highest yield of xylan from birch wood. The optimal extraction conditions for obtaining the
highest yield of xylan were extracted by the Design Expert software using a numerical
optimization. Several potential solutions were provided by the software, and based on the
experimental feasibility, the following optimum operating settings were selected for obtaining a
higher yield of xylan: time (A) = 25 minutes, NaOH concentration (B) = 8 wt%, solid to liquid
ratio (C) = 1:8 g/mL, and sample size (D) = 10 g. Four replicates of the extraction were
performed at this optimal extraction conditions. The results of the extraction with the respective
predicted values are given in the table 7.5. Based on the numerical optimization condition
selected, the maximum yield predicted was 22.66% (OD wood basis) with a standard deviation
of 1.08. The predicted wood dissolution, and temperature at this point were 33.53% (standard
deviation 0.36) and 96.46oC (standard deviation 0.69) respectively. The yield of xylan, wood
dissolution, and temperature of the slurry obtained were 21.27±2.65%, 33.78±0.21%, and
97.67±0.47 respectively. The experimental values obtained were found to be in good agreement
with the values calculated from the models suggesting that the quadratic surface models were
adequate for the optimization of microwave assisted extraction under study. Wood used for this
study contained 29.4% of xylan. Conversion of the yield of xylan obtained based on the oven dry
144
Table 7.5. Experimental and predicted values of wood dissolution, yield of xylan, and temperature at the optimum extraction conditions used for the alkaline extraction of xylan
Optimized extraction conditions
Wood dissolution (%) Yield of xylan (%) Temperature (oC) Experimental Predicted Experimental Predicted Experimental Predicted
Time, (min) NaOH concentration (wt%) Solid to liquid ratio (g:mL) Sample size (g)
25 8 8 10
33.78 ± 0.21
33.53 ± 0.36
21.27 ±2.6
22.66 ±1.08
97.63 ±0.5
96.46 ±0.69
basis of wood to that of the original xylan present in wood indicate that about 72.5 % of the
xylan present in birch wood can be extracted by 25 minutes of microwave assisted extraction .
Though there is no available literature for direct comparison, Sun et al. (1995) reported that a
low temperature alkaline extraction of wheat provided 76.4% of xylan, when wheat straw was
treated with 1.5% NaOH at solid loading of 2.5% dry matter for 144 hours at room temperature.
It is clear that great reduction in the extraction time can be obtained using microwave assisted
extraction for a similar yield of xylan. In a recent study (Obermeier et al., 2012) using
microwave assisted alkaline extraction, a similar type of xylan yield (73.6%) was reported in the
supernatant obtained from wheat straw extraction. The respective reaction conditions reported
were 5% NaOH solution at a temperature of 140oC with 10 minutes preheating time and 10
minutes of residence time for 3.5 g of wheat straw with a solid to liquid ratio of 1:10 using an
input power of 265 W. A total of 14 g of samples were used for extraction in a single batch. The
energy input for the extraction was calculated as the multiple of the power input into the reaction
chamber and the sum of the preheating and residence time in seconds. The energy calculated for
this extraction was 318 kJ and the energy input per g of wheat straw calculated as 22.7 kJ/g. In
our study, the energy input at the optimized extraction conditions for getting maximum yield of
145
xylan was calculated as 66 kJ and the energy input per g of wood fibers was 6.6 kJ/g. Though it
is not appropriate to compare two different raw materials, it is clear that optimization of the
reaction conditions are necessary to obtain maximum yield of xylan with lower input of energy.
Figure 7.7. ηsp/C vs. Concentration of xylan obtained at the optimal conditions of microwave extraction
The high temperature generated at the optimized conditions of extraction might degrade the
degree of polymerization of the polymer. As expected, viscosity measurement of the polymer in
CED solution showed a low intrinsic viscosity of 0.605 (Figure 7.7) and the corresponding
molecular weight and degree of polymerization calculated was 16564 and 128 respectively.
Comparison of the results discussed in section 5.3.4.3, the xylan obtained at the optimal
conditions is similar to the xylan obtained after 30 minutes of irradiation using 4 wt% NaOH
solution for 30 minutes.
y = 0.7311x + 0.6046R² = 0.9764
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
0 0.2 0.4 0.6 0.8
ηsp
/C (
dL
/g)
Concentration of xylan solution (g/dL)
146
7.6 Conclusions
Central composite design and surface response methodology were employed to optimize the low-
temperature microwave assisted alkaline extraction of xylan from birch wood. The effect of four
extraction parameters (time, alkali concentration, solid to liquid ratio and sample size) on the
temperature of the wood slurry, wood dissolution, and yield of xylan were studied. Three
quadratic polynomial models were developed to correlate the extraction variables with the three
responses. Analysis of variance (ANOVA) of the models developed indicated the statistical
significance of the models for the prediction of each response studied. Temperature of the slurry,
and wood dissolution were significantly influenced by all the four extraction parameters,
whereas the yield of extraction was not affected by the linear effect of solid to liquid ratio.
Process optimization was conducted to maximize the yield of xylan and the experimental values
obtained for the respective yield of xylan, wood dissolution and temperatures were found to
agree satisfactorily with the quadratic model used for xylan extraction. Optimum conditions used
for the maximum extraction of xylan from birch wood under microwave irradiation were: 10 g of
wood fibers, 8 wt% of NaOH solution, 1:10 solid to liquid ratio (g: mL) and 25 minutes of
irradiation time.22 wt% of xylan (OD) was obtained from about 34% of the dissolved wood at
the optimal extraction conditions.
147
Chapter8 SummaryandConclusions
The following conclusions were drawn from the various results discussed in this thesis.
a) Direct interaction of microwaves with the wood slurry reduce the time required for the
extraction of xylan from wood fibers compared to conventional extraction
b) Ten minutes of microwave assisted extraction of wood using a power input of 110 W offered
about 60% of high molecular weight xylan polymers with degree of polymerization of 150.
c) Xylan obtained from microwave assisted extraction provided a higher number average degree
of polymerization compared to conventional process under similar conditions (150 vs. 125 for 20
minutes of extraction)
c) Interaction of microwaves with the alkali in the fibers resulted in a rapid increase in the
temperature and produced an "explosion effect” that loosened the recalcitrant lignocellulosic
structure and increased the porosity of the fiber thus enhancing the mass transfer of the
hydrolyzed components to the solution
d) During microwave assisted extraction of xylan, temperature as well as the fibre structure
rupture leads to the enhanced hydrolysis and dissolution of wood.
e) SEM analysis of the fibers after extraction demonstrated the "explosion effect" induced during
the microwave extraction and this was supported by the internal structure exploration by X-ray
tomography. Porosity of the wood fibers after microwave extraction was higher than the
conventional extraction.
148
f) Wood dissolution during the extraction process consisted of two stages, whereas the yield
obtained was assumed to be the consequence of three stages including the dissolution of easily
accessible xylan, followed by the steady increase in the yield where hydrolysis and dissolution of
xylan is higher, which is then followed by the decrease in yield, where degradation of the
dissolved polymer is higher than the dissolution of the hydrolyzed polymer.
g) Temperature of the extraction was found to be related with the yield of xylan as well as the
degree of polymerization, the increase in temperature and time of extraction decreased the yield
of xylan as a result of the molecular breakdown and hence the degree of polymerization.
h) Degradation of xylan in alkali was found to be significant at temperatures above 95oC.
i) Decrease in the crystallinity of the fibers after microwave extraction also supported the
hypothesis of the "explosion effect" and the structural changes produced by the interaction of
microwave with the alkali in the fiber slurry.
j) X-ray microtomography can be used as technique to determine the internal temperature
generated inside the fibers, and demonstrated that a 10 minute of microwave extraction using a
power input of 110 W makes structural changes equivalent to the temperature of 120oC.
k) Optimization of the microwave extraction process demonstrated that 25 minutes of microwave
extraction using 10g of the fibers at a solid to liquid ratio of 1:10 with 8 wt% NaOH solution can
yield up to 75% of the xylan present in the fibers.
149
Chapter9 RecommendationsforFutureResearch
Extraction of xylan from hard wood species birch has demonstrated that microwave assisted
extraction can considerably reduce the time of the conventional extraction. This research
established a synergetic microwave effect produced by the hot spot or explosion effect of the
fibers accelerated the extraction of the xylan polymers from the woody biomass. The study also
established that a low power input microwave can be more efficient for the extraction of the
polymeric xylan compared to high energy input and this method can be used as an effective
alternate to the conventional extraction of xylan polymers. In this section, recommendations for
future research in terms of the extension of the current research and the future necessary research
needed for the implementation and application of the microwave assisted extraction.
The dominant factors governing the extraction of the hemicelluloses from the lignocellulosic
matrix by microwave assisted extraction are the solubility of the hemicelluloses in the solvent,
the strength of hemicellulose/matrix interaction and the mass transfer kinetics of the
hemicelluloses from the matrix to the solution phase.
Interaction of hemicelluloses with other structural components such as cellulose and lignin
highly depends on the origin or type of the lignocellulosics under study. Extension of the
process of microwave extraction to other lignocelluloses such as other hard wood species, soft
wood species, as well as agro-residues such as wheat straw, bagasse, or corn stover would be
needed to investigate the effect of structural differences of the biomaterials on the microwave
150
extraction of the hemicelluloses and also to establish the effectiveness of the method. Effect of
pre-treatments of the lignocellulosic materials such as delignification and size reduction are also
to be considered respectively for further elucidation of the structural effects on the microwave
extraction, and to provide additional knowledge on the optimum size of the materials used for an
efficient extraction of polymers using microwave heating.
An important component of the engineering aspect of the process development is the
understanding of the process kinetics. Extraction process kinetics becomes difficult when there is
mass transfer along with chemical reactions. Understanding the mass transfer mechanisms and
kinetics is important for the proper designing of the equipment as well as optimizing the process
conditions during the scale-up of the process. Hence, investigation and establishment of the
extraction kinetics of the microwave assisted extraction has to be explored. This study can also
focus on the development of models to predict the mass transfer kinetics as well as the validation
of the developed models.
For the successful demonstration and implementation of the microwave extraction technology,
energy efficiency and economic viability (cost analysis) of the process has to be studied
compared to the conventional method of extraction processes.
Application potential of the lignocellulosic residue after hemicelluloses extraction and
modification of the polymeric hemicelluloses are the two other areas of research that would be
highly beneficial for the biorefinery industry.
151
References
Al-Dajani, W. F., Tschirner, U.W., Jenson, T. (2008). Pre extraction of hemicelluloses and
subsequent kraft pulping. Part II, Acid and autohydrolysis. Tappi Journal, 3-8
Alén, R., 2000. Structure and chemical composition of wood. In: Stenius, P. (Ed.), Forest
Products Chemistry. Chapter 1, pp. 11–57
Allen, S. G., Schulman, D., Lichwa, J., et al. (2001). A Comparison of Aqueous and Dilute-
Acid Single-Temperature Pretreatment of Yellow Poplar Sawdust. Industrial and Engineering
Chemistry Research. 40(10): 2352-2361
Amash, A.; Zugenmaier, P. (1998). Study on cellulose and xylan filled polypropylene