http://www.e-polymers.org
e-Polymers 2008, no. 022
ISSN 1618-7229
Sisal, Sugarcane Bagasse and Microcrystalline Celluloses:
Influence of the Composition of the Solvent System N,NDimethylacetamide /Lithium Chloride on the Solubility and
Acetylation of these Polysaccharides
Gabriela T. Ciacco, Beatriz A. P. Ass, Ludmila A. Ramos, Elisabete Frollini*
*Instituto de Química de São Carlos, Universidade de São Paulo, C. P. 780, CEP
13560-970, São Carlos, SP, Brazil; fax: +55 16 3373 9952; e-mail:
[email protected].
(Received: 11 January, 2007; published: 7 February, 2008)
Abstract: The present work describes an investigation concerning the acetylation
of celluloses extracted from short-life-cycle plant sources (i.e. sugarcane bagasse
and sisal fiber) as well as microcrystalline cellulose. The acetylation was carried
out under homogeneous conditions using the solvent system N,Ndimethylacetamide/lithium chloride. The celluloses were characterized, and the
characterizations included an evaluation of the amount of hemicellulose present in
the materials obtained from lignocellulosics sources (sugarcane and sisal). The
amount of LiCl was varied and its influence on the degree of acetate substitution
was analyzed. It was found that the solvent system composition and the nature of
the cellulose influenced both the state of chain dissolution and the product
characteristics. The obtained results demonstrated the importance of developing
specific studies on the dissolution process as well as on the derivatization of
celluloses from various sources.
Introduction
In recent years, natural polymers have aroused a growing interest among
researchers around the world [1]. However, efforts are required in order to increase
the amount of natural polymers and their derivatives, as compared to synthetic
polymers, in the plastic market, but this implies improving their competitiveness in
terms of cost, and mainly properties. Nonetheless, it should be pointed out that,
considering how the environmental impact legislation has already been consolidated
in some countries, the benefit attained by the substitution of a biodegradable material
for a non-biodegradable one in some cases overcomes the effect of an eventual
added cost [2].
In the case of cellulose, it would clearly be convenient to be able to extract the raw
material in large-scale from the fibers of rapidly-growing plants, such as sisal and
sugarcane as opposed to wood, considering the long life-cycle of trees. One of the
fundamental aspects related to the derivatization of cellulose involves the choice of
reaction medium, which can be either homogeneous or heterogeneous. For
derivatization in a homogeneous medium, i.e., dissolution of cellulose followed by its
derivatization, the reaction occurs uniformly along the polymer chain, since, in this
case, the cellulose chains are dispersed in the reaction medium [3].
1
Among cellulose esters, acetates are one of the most important, due to their multiple
applications [4]. The properties and practical applications of acetates depend on the
degree of substitution, which determines their compatibility with other plastics and
resins, as well as their solubility in many solvents. Normally, cellulose acetates are
prepared in heterogeneous medium, using excess of acetic anhydride (10-40%
above the amount necessary to form cellulose triacetate) in the presence of sulfuric
acid and perchloric acid which act as catalysts [4]. Acetates with a degree of
substitution lower than three can be obtained by triacetate hydrolysis.
The interest in the derivatization of cellulose under homogeneous conditions resides
mainly in the fact that it allows a more efficient control of the degree of substitution
(DS) of the cellulose derivative through adjustment of the molar ratio of the
derivatizing agent and cellulose. Moreover, the substituting groups can be introduced
more regularly along the polymer chain and materials with well-defined
characteristics can be produced using methods with good reproducibility [5-7].
Derivatization of cellulose under homogeneous conditions generally involves three
steps, i.e. activation, dissolution, and subsequent reaction with the derivatizing agent.
The aim of the activation step is to reduce the cellulose crystallinity by breaking the
strong hydrogen bonds between the polymeric chains with a subsequent penetration
of the solvent into the fiber. After activation, the dissolution of cellulose is generally
reached by stirring the mixture for a certain time [6, 7]. The maximum degree of
substitution attainable for cellulose derivatives, such as cellulose acetates, obtained
in homogeneous solution, depend both on properties of the particular raw material,
e.g. crystallinity, molar mass and percentage of α-cellulose, and on the reaction
conditions, i.e. the time and temperature of each stage (dissolution and acetylation)
as well as the composition of the solvent mixture. Earlier work [8, 9] has shown that
the characteristics of given cellulose greatly affect the efficiency of its dissolution and
the degree of substitution of the obtained product. 1,
The progress of cellulose chemistry has been stimulated by the proposal of using
non-derivatizing and non-aqueous solvent systems for the functionalization of
cellulose in order to obtain products with homogeneous distributions of functional
groups. Among these systems, solutions consisting of amide N,N-disubstituted
compounds, such as N,N-dimethylacetamide (DMAc), and lithium halogen salts,
normally LiCl, stand out [8]. This solvent system has proved to be a suitable medium
for synthesis of a wide variety of cellulose derivatives [3-14] in solution. However, it is
important to realize that the cellulose chains may form aggregates in such solutions,
thus reducing the number of hydroxyl groups freely accessible to the reagent and
ultimately decreasing the reactivity and extent of modification of the polymer [5, 8, 15,
16].
A variety of supramolecular structures of cellulose chains in solution have been
found, including both smaller and larger aggregates, depending on the dissolution
process conditions, e.g. the salt and cellulose concentrations as well as the cellulose
source, due to the varying degrees of crystallinity, hemicellulose content, etc that can
be found [15, 16]. When studying the process of dissolution and derivatization of
celluloses obtained from lignocellulosics sources, the possibility that hemicellulose
may be present is not always considered, even though this can have a large
influence in both mentioned processes.
The present work was carried out on celluloses from rapidly-growing plants, i.e.
sugarcane and sisal. The celluloses were characterized, including the evaluation of
2
the amount of hemicelluloses present. They were subsequently derivatized in a
homogeneous medium, using N,N- dimethylacetamide/lithium chloride as the solvent
system. Microcrystalline cellulose was also investigated in an attempt to compare its
behavior with those of the celluloses obtained from sisal and sugarcane. The
influence of the LiCl percentage on the degree of substitution obtained for the
cellulose acetates was considered. In previous studies where DMAc/LiCl was also
employed as a solvent system, various process-related aspects were taken into
account, including dissolution conditions [5-9], and acetate synthesis using linters,
sisal, sugarcane bagasse, and microcrystalline cellulose as cellulose sources [5-7,
9].
Results and discussion
Cellulose Characterization
Celluloses from sisal, sugarcane bagasse and microcrystalline cellulose were
characterized with regard to the degree of polymerization, the α-cellulose content
(which corresponded to the percentage of pure cellulose in the material), the degree
of crystallinity, and morphological structures. The SEM images in Figure 1 show the
surface morphologies of microcrystalline, sisal, and sugarcane bagasse celluloses.
Fig. 1. SEM images of: A) microcrystalline cellulose (X 1000); B) sisal cellulose (X
500); C) sugarcane bagasse cellulose (X 500).
As can be observed from the SEM images (Figure 1), the microcrystalline cellulose
consisted of crystalline particles of cellulose with pores (clearly visible) of
approximately 2 μm in diameter. In contrast, the sugarcane bagasse and sisal
celluloses, whose pores could not be observed by this technique, formed fiber
bundles approx. 10-25 μm thick.
3
In the X–ray diffraction curve of microcrystalline cellulose shown in Figure 2, the
typical diffraction of the polymorphic form attributed to cellulose I, i.e. Bragg’s angles
of 23° (002 lattice), 21° (021 lattice), 17° (10 1 lattice), and 15° (101 lattice) can be
observed. These X–ray diffraction patterns also indicated the maximum diffraction
intensity (002 lattice) attributed to the crystalline regions of the samples, Imax (2θ ≈
22-230), and the minimum intensity, attributed to the non-crystalline regions, Imin (2θ
≈ 18-190). These parameters were used for calculating the degree of crystallinity, Ic.
The other celluloses presented X–ray diffraction curves (not shown) with similar
profile to the one displayed in Figure 2, and these were used to determine the
respective Ic-values of each material (Table 1).[17]
250
Imáx
Intensity (cps)
200
150
100
Imin
50
0
0
10
20
30
40
50
60
Bragg angle, 2θ (degree)
Fig. 2. X–ray diffraction patterns of the microcrystalline cellulose.
Table 1 gives the experimental values of Ic, DP, and α-cellulose content of the three
cellulose types.
Tab. 1. Average degree of polymerization (DPv), % α-cellulose, crystallinity index (Ic)
of celluloses.
Cellulose
Sisal
Sugarcane bagasse
microcrystalline
Ic (%)
77
67
80
DPv
648
716
140
α-cellulose (%)
86
89
---
Microcrystalline cellulose, while possessing the highest crystallinity, had the lowest
degree of polymerization and thus exhibited domains in the order of 100 nm in size
(approximately the chain length of 200 AGUs) [22], as well as pores nearly 2 μm in
diameter (Figure 1). Thus, this cellulose dissolved readily in DMAc/LiCl, as this
solvent was able to penetrate the crystalline regions and rupture the hydrogen bonds
between neighboring chains that stabilize the previous structure. However, the low
4
molar mass may have facilitated an aggregation of cellulose chains in solution,
resulting in not all of the reactive OH groups being free to react, as will be discussed
later.
The two celluloses obtained from lignocellulosic fibers (sisal and sugarcane bagasse)
displayed nextser values of DP and α-cellulose content (Table 2) as compared to
microcrystalline cellulose. Nonetheless, as will be discussed next, they behaved
differently during dissolution and acetylation in DMAc/LiCl.
Celluloses acetylation
Figure 3 shows a typical 1H NMR spectrum of cellulose acetate. Peaks related to the
resonance of methyl protons of the acetate groups (1.70-2.20 ppm) and the integrals
corresponding to the resonance of glucose ring protons (2.90-5.10 ppm) can be
observed.
methyl
AGU
5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5
ppm
Fig. 3. 1H-MNR spectrum of cellulose acetate (DS = 1.55).
Spectra with similar profile to that in Figure 3 (figures not shown) were obtained for
the other acetates prepared in the present work and were used for the calculations of
DS.
It is well-known that LiCl must be present in the solvent system in order for the
cellulose to go into solution, since this is the first step in the synthesis of derivatives
by the homogeneous route. However, for each type of cellulose used, it is important
to study the effects obtained when varying the amount of this salt in the solvent,
since celluloses from different sources do not behave identically during dissolution
and derivative formation, even under equivalent conditions. This is due to differences
in chain lengths, crystallinity, α-cellulose content, etc., as was mentioned earlier. The
mechanism of dissolution of cellulose in the solvent system DMAc/LiCl is described
in literature by different models, as illustrated in Figure 4.
Basically, these structures (Figure 4) differ in the role exerted by lithium and chloride
ions (A-C), whether LiCl is dissociated (A-C) or not (D,E) in this medium, and whether
5
chlorine participates (A-D) or not (E) in the dissolution process [3]. Although it was
not the aim of the present work to investigate the dissolution mechanism, we
confirmed the influence of amount of salt on the chain dissolution state, which
influences the properties of the products formed in this medium. Figure 5 shows the
DS as a function of the percentage of LiCl used in the dissolution of sugarcane
bagasse, sisal, and microcrystalline celluloses.
[ ( CH3)2
O Li
N C CH3
O
]
(+)
Cl
(-)
( CH3)2
HO Cell
Li
(+)
N C
O
H3 C
Cell
(A)
H
Cl
(-)
(B)
(+)
(-)
O
Li
O H
H3 C C
N
Cell
O
Li
O
N C
Cl
H
Cl
( CH3)2
H3 C
(CH3)2
(D)
(C)
( CH3)2
Cell
O
Li
Cl
N C
O
H
H3C
Cell
O
C
N
( CH3)2
H3 C
(E)
Fig. 4. Mechanism of dissolution of cellulose in the solvent system DMAc/LiCl [3].
When using cellulose from sugarcane bagasse, a large increase in the DS of the
acetates could be observed as the content of LiCl was raised from 4% to 5%. At even
higher concentrations (6, 7 and 8%), however, the value of DS did not vary
significantly from that at 5% (Figure 5). Apparently, 4% LiCl was insufficient to
overcome the intermolecular hydrogen bonding and thus disperse the cellulose
chains in solution. This led to aggregation and consequently the DS of the obtained
product was low (Figure 5). On the other hand, 5% LiCl gave rise to a much better
separation of the chains, leaving the OH groups of the cellulose available for
substitution reactions, and the DS was significantly increased. At concentrations
above 5%, the DS did not vary much, indicating that there was no further effect on
the state of aggregation of the cellulose chains and/or that the higher salt
concentration hindered the access of acetic anhydride to the hydroxyls, since they
were involved in complexes with LiCl.
6
In the case of sisal cellulose, there was a tendency for the DS of acetates to
decrease as the LiCl content was increased. This behavior was the opposite of that
observed for sugarcane bagasse cellulose (Figure 5). The latter had a lower
crystallinity than sisal cellulose (Table 1) and should thus dissolve more readily, as a
result of the solvent molecules penetrating the non-crystalline region of the cellulose
with greater ease. This is believed to be an important factor when explaining the
observed differences between the two celluloses. The α-cellulose content was close
in the two cases and the type of hemicellulose found in sisal [23] is similar to that in
bagasse [24], i.e. it is mainly composed of xylans. The presence of hemicellulose
exerts a strong influence on the state of aggregation of the cellulose chains. It must
be pointed out that it is very difficult to eliminate all hemicellulose present in the sisal
and sugarcane bagasse raw material, and the influence exerted by hemicellulose on
the material properties must thus be considered in all discussions of the properties of
these two cellulose types. The frequency of side-chains on the xylan rings affects the
intensity of the interaction with the cellulose. Moreover, the degree of branching of
the hemicellulose molecules can influence the dissolution behavior in DMAc/LiCl and
hence the aggregation, which in turn affects the DS of the formed cellulose acetates.
Probably, both aspects (frequency of side chains and degree of branching) are
different for sisal and sugarcane bagasse, which can contribute also for the different
behaviour observed in Figure 5.
2,0
sugarcane bagasse
sisal
microcrystalline
1,8
1,6
DS
1,4
1,2
1,0
0,8
0,6
3
4
5
6
7
8
9
% LiCl
Fig. 5. DS of cellulose acetates as a function of %LiCl in DMAc. The data points are
joined by lines to assist identification.
In the case of microcrystalline cellulose, despite a rather small variation in DS
between 5% and 9% LiCl, the results displayed in Figure 5 were very similar to those
obtained in a previous study [9]. Lower concentrations of the salt (5% and 6%)
seemed to be insufficient in separating the cellulose chains to such an extent that the
largest possible number of hydroxyl groups was free to react. As the concentration
increased to 7% and 8%, the DS became higher. However, a further increase (9%)
led to a drop in DS, probably because of the greater number of LiCl forming
complexes with the hydroxyl groups and thus blocking their accessibility to the
reagent (Figure 5). Although a higher molar ratio of the anhydride was used in the
reactions with microcrystalline cellulose, the acetylation achieved a lower conversion
degree (50 %) in this case (9% LiCl). According to recent results in light-scattering
experiments (to be published soon), the shorter chains of microcrystalline cellulose
favored aggregation, thus reducing the number of available hydroxyls.
7
Table 2 reports on the reaction conditions yielding the highest DS values for each
cellulose type considered in the present work.
The data in Table 2 displays the differences in behavior between fibrous and
microcrystalline cellulose. The best reaction conditions for each cellulose type
corresponded to different LiCl concentrations in the solvent system. These results
confirmed the necessity of individual studies on celluloses with different
characteristics [25].
Tab. 2. Reaction conditions that led to the highest values of degree of substitution
(DS) for each cellulose.
Cellulose
sisal
Sugarcane
bagasse
microcrystalline
Percentage of LiCl
(%)
3.5
6.0
1.40
1.40
Acetylating efficiency *
(%)
70.0
70.0
7.0
1.60
77.5
DS
* acetylating efficiency = (DSobtained/DS stoichiometryc) 100
In none of the considered reactions did the obtained DS values correspond to what
was expected, considering the molar ratio of acetic anhydride/AGU (stoichiometric
DS). For microcrystalline cellulose, the expected DS was 3.0, and for the sisal and
sugarcane bagasse celluloses, it was 2.0 (see Table 3). It must however be pointed
out that, in a previous study, the obtained DS for the Ac2O/AGU ratio also differed
from what was expected. However, a correlation between DS and the Ac2O/AGU
ratio, allowing the control of the reaction stoichiometry was found [26]. This finding is
deemed important in order to render this kind of process application viable in large
scale.
Conclusions
The results reported herein have shown that the optimum salt quantity for acetylating
cellulose in a LiCI/DMAc solvent system differed for each type of cellulose
considered. This was probably a result of the state of aggregation of the polymer
chains, which in turn depended on the characteristics of the cellulose (e.g. DP, Ic,
hemicellulose content) and also on the composition of the solvent system. Hence, the
degree of substitution of the synthesized cellulose acetates depended on a fine
adjustment of the dissolution and acetylation conditions for each cellulose.
Solutions of cellulose in the DMAc/LiCl solvent system are generally also prepared
with the aim of analyzing the polysaccharide by techniques such as light-scattering,
size-exclusion chromatography, etc. The present results show that an exploratory
study should be performed for each type of cellulose in order to find the best solvent
system composition for cellulose dispersion. This would allow the interpretation of the
analysis results as real properties of the individual chains, as expected in this kind of
analysis.
The behavioral contrast between the three studied celluloses, sugarcane bagasse,
sisal, and microcrystalline cellulose demonstrated that: a) the source and
characteristics of the raw material (Ic, DP, α-cellulose content) used to prepare the
derivatives in a homogeneous medium must be detailed in related literature as they
8
can be of fundamental importance for understanding the properties of the obtained
derivatives; b) it is necessary to refine the search for the best dissolution and
synthesis conditions of celluloses, considering celluloses from different sources
separately.
Experimental part
Materials
Sisal (kindly supplied by Lwarcell- Lençóis Paulista, SP, Brazil), sugarcane bagasse
(obtained after soda/anthraquinone treatment of the fiber) and Avicel PH-101
microcrystalline celluloses (FMC Inc., Philadelphia, USA) were used.
The fibrous celluloses (sisal and linters) were milled and then dried in a vacuum oven
at 100 °C for 14 h. N,N-Dimethylacetamide (Riedel-deHaën) was distilled from CaH2
and kept over activated type 4 Å molecular sieves. Before use, the lithium chloride
(Vetec) was dried in an oven at 200 °C for 3 h, cooled under reduced pressure, and
kept in a desiccator. The acetylating agent, acetic anhydride (Synth) was distilled
from P2O5 prior to use.
Characterization
-Celluloses
Morphological surface analysis: The cellulose surfaces were examined by Scanning
Electron Microscopy (SEM) using a LEO 440 ZEISS/LEICA model. All samples were
gold-sputtered prior to analysis.
Crystallinity index (Ic): The crystallinity index was determined by X-ray diffraction,
using a VEB CARL ZEISS-JENA URD-6 Universal diffractometer operating at 40
kV/20 mA and λ(Cu Kα) = 1.5406Å. Ic was calculated according to the equation
Ic = 1 – I1/I2,
where I1 is the intensity at the minimum of the crystalline peak (18o <2θ<19°) and I2 is
the intensity at its maximum (22o <2θ< 23o) [17].
Degree of polymerization (DP): The degree of polymerization was determined by
measuring the viscosity (at 25 ºC) of a cellulose solution in Cuen (copper:ethylene
diamine, 1:1 v/v) with an Ostwald viscometer. The average molar mass was given by
Mv = DP x 162, according to the TAPPI standard T230 om – 89 [18].
Content of α-cellulose: The α-cellulose content was given by the loss of mass after
treatment of cellulose with a 17.5% NaOH solution (at 25 ºC). It is defined as % α-cell
= (mass of dry α-cell/mass of dry cell) x 100 [19].
-Cellulose acetates
Degree of substitution
The degree of substitution (DS) of cellulose acetates was determined by proton
NMR. Spectra were recorded with a Brucker AC-200 spectrometer running at 200
MHz and 80 oC; the number of scans was 392. Samples were dissolved in DMSO-d6
(10 mg/mL). A drop of trifluoro-acetic acid was added to the sample solution in order
to shift the signal of residual water and hydroxyl protons to a lower field, outside the
9
spectral region of interest, without affecting the chemical shifts of the glucose ring
protons [20, 21].
Dissolution of the celluloses
The dissolution and acetylation steps were similar to those used by Edgar [20] with
some modifications. A mixture of cellulose (2.00 g) and 100 mL of DMAc (see Table
2) was heated to 150 oC and stirred for 1.0 h, in a glass reactor equipped with a
mechanical stirrer and reflux condenser, under a flow of N2. Then, lithium chloride
was added (see Table 1) and the reflux condenser was substituted by a short-path
distillation apparatus. The mixture was heated to 170 oC and 10% of the solvent
volume was distilled off with the aim of eliminating most of the residual water present
in both solvent and cellulose. The reaction mixture was then cooled to room
temperature and allowed to stir overnight.
Acetylation of the celluloses
A clear cellulose solution was obtained and the temperature was raised to 110 oC,
under reflux and a flow of N2. Acetic anhydride (Ac2O) was added drop wise, in
appropriate molar ratios to the anhydroglucose units (AGU), as given in Table 3. The
system was kept at 110 oC for 4.0 h in case of sisal and sugarcane bagasse
celluloses and for 1.0 h for microcrystalline cellulose, as summarized in Table 3, and
then cooled to room temperature.
Tab. 3. Acetylation conditions.
Cellulose
Molar ratio Ac2O/AGU
Microcrystalline
Sisal
Sugarcane bagasse
3:1
2:1
2:1
Reaction time
(h)
1
4
4
% LiCl
5.0-9.0
3.5-8.0
4.0-8.0
The product was precipitated with methanol and purified by Soxhlet extraction in
methanol. Finally, the cellulose derivative was dried at 50 oC under vacuum, until
constant weight. The reactions were carried out at least twice and the DS values
obtained were very close, indicating good reproducibility of the reactions.
Acknowledgements
E. Frollini is grateful to CNPq (National Research Council, Brazil) for research
productivity fellowship and financial support, and FAPESP (The State of São Paulo
Research Foundation, Brazil) for financial support.
References
[1] Kamide, K. Cellulose and cellulose derivatives: molecular characterization and its
applications, Elsevier, 2005.
[2] Toyoshima, I. Highly advanced utilization and functionalization of cellulose from
the aspects of both chemistry and industry, in : J. F. Kennedy, G. O. Phillips, P. A.
Willians (eds.). Cellulosic: Chemical, Biochemical and Material Aspects. Chichester,
Ellis Horwood, 1993. p.125-140.
[3] Dawsey, T. R.; McCormick, C. L. J. Mol. Struct., C30 1990, 405.
10
[4] Klemm, D.; Philipp, B.; Heinze, T.; Heinze, U.; Wagenknecht, W. Comprehensive
cellulose chemistry, fundamentals and analytical methods. Wiley-VCH, 2001, 1,
p.107-110.
[5] Ass, B. A. P.; Frollini, E. Anais Assoc. Bras. Quím. 2001, 50, 76.
[6] El Seoud, O. A.; Regiani, A. M.; Frollini, E. Derivatization of celluloses in
homogeneous conditions: a brief review, in: Natural Polymers and Agrofiber-Based
Composites. E. Frollini, A. Leão, L. H. C. Mattoso (eds.), São Carlos, Brazil, USPEMBRAPA-UNESP, 2000. p.73-89.
[7] Marson, G. A.; Ciacco, G. T.; Frollini, E.; El Seoud, O. A. Macromol. Chem. Phys.
2001, 201, 882.
[8] Ramos, L .A.; Assaf, J.M.; El Seoud, O. A.; Frollini, E. Biomacromolecules 2005,
6, 2485.
[9] Ciacco, G.T.; Ass, B.A.P.; Ramos, L. A.; Frollini, E. Acetylation of celluloses under
homogeneous reaction conditions, in: L. H. C. Mattoso, A. Leão, E. Frollini (eds).
Natural polymers and composites, Sao Paulo, Brazil: EMBRAPA-USP- UNESP,
2000, p. 139-145.
[10] Potthast, A.; Rosenau, T.; Sartori, J.; Sixta, H.; Kosma, P. Polymer 2003, 44, 7.
[11] El-Kafrawy, A. J. Appl. Polym. Sci. 1982, 27, 2435.
[12] Spange, S.; Reuter, A.; Vilsmeier, E.; Heinze, T.; Keutel, D.; Linert; W. J. Polym.
Sci.: Part A: Polym. Chem. 1998, 36, 1945.
[13] Bianchi, E.; Ciferri, A.; Conio, G.; Cosani, A.; Terbojevich, M. Macromolecules
1985, 18, 646.
[14] Hult, E.-L.; Larsson, P.T.; Iversen, T. Polymer 2001, 42, 3309.
[15] Terbojevich, M.; Cosani, A.; Conio, G.; Ciferri, A.; Bianchi, E. Macromolecules
1985, 18, 640.
[16] Röder, T.; Morgenstern, B.; Schelosky, N.; Glatter, O. Polymer 2001, 42, 6765.
[17] Buschle-Diller, G.; Zeronian, S.H. J. Appl. Polym. Sci. 1992, 45, 967.
[18] Tappi press – Viscosity of pulp (capillary viscosimeter method) T230 om –89. In:
TAPPI test methods 1991. Atlanta, 1990; 1:1.
[19] Browning, B. L. Methods of wood chemistry, John Wiley, 1967; 2: 499.
[20] Edgar, K.J.; Arnold, K.M.; Blount, W.W.; Lawniczak, J.E.; Lowman, D.W.
Macromolecules 1995, 28, 4122.
[21] Buchanan, C.M.; Edgar, K. J.; Hyatt, J.A.; Wilson, A.K. Macromolecules 1981,
13, 127.
[22] Isogai, A.; Atalla, R.H. Cellulose 1998, 5, 309.
[23] Ciacco, G.T.; Liebert, T.; Frollini, E.; Heinze, T. Cellulose 2003, 10,125.
[24] Sun, J. X.; Sun, R.; Sun, X.F.; Su, Y. Carbohydr. Res. 2004, 339, 291.
[25] Ass, B. A. P.; Ciacco, G.T.; Frollini, E. Biores. Technol. 2006, 97, 1696.
[26] Ass, B. A. P.; Belgacem, N.; Frollini, E. Carbohydr. Polym. 2006, 63, 19.
11