Cellulose
Cellulose [9004-34-6] deserves a special position among the industrially used raw
materials for two general reasons. First, cellulose belongs to the natural products which,
when used carefully, are inexhaustible since it is regularly regenerated by nature in
relatively short time periods. As long as we ensure that the primary sources of cellulose,
forests and cotton plantations, are not damaged by destructive lumbering or over
cropping, we can expect regular and significant natural annual reproduction.
According to reference [8], the annual yield of cellulosic matter resulting from photo
initiated biosynthesis amounts to approximately 1.3×109 metric tons. A tree produces an
average of 13.7 g of cellulose daily. If they were lined up, the cellulose chain molecules
formed each day would result in a string of 2.62×1010 km in length, or 175 times the
distance between the sun and the earth.
In wood, cellulose is part of an ingeniously constructed fiber-reinforced composite in
which long, stiff cellulose chain molecules organized in thin fibrils constitute the plant
reticulum material held together and protected by hydrophobic lignin acting as binder
and encasement.
To isolate cellulose from wood for industrial applications, the wooden composite must be
broken up by so-called pulping processes. In these treatments, other wood constituents,
such as lignin and hemicelluloses, are to a large extent degraded and dissolved. Thus
far, these byproducts have found only limited use. In most cases, wood pulp
manufacturers concentrate the waste pulping liquors to concentrates consisting of ca. 50
% solids. The organic matter is used as fuel to produce steam and electric power, while
the inorganic pulping chemicals (soda, magnesium, or ammonium base and sulfur
dioxide) are simultaneously recovered. These recovery processes have practically
solved the long-standing environmental problems of the wood pulp industry.
Both cellulose and lignin are biologically degradable and, thus, ecologically beneficial.
They will decompose in the open. Cellulose products, such as paper or cellulosic
textiles, will decompose and eventually form valuable humus. In industrial use,
environmental problems are not caused by cellulose or lignin but by the chemicals used
in the isolation or in subsequent chemical processing and transformation into cellulose
derivatives, films, or fibers. Therefore, the long-term task of modern cellulose research
will be the development of novel processes which yield no or only a few ecologically
harmful byproducts. If these efforts are successful, cellulose will surely maintain and
strengthen its position as a renewable and environmentally beneficial, industrially
important raw material competing with synthetically produced polymers.
1.1. Properties
1.1.1. Molecular Structure
Cellulose is an isotactic b-1,4-polyacetal of cellobiose (4-O-b-D-glucopyranosyl-Dglucose). The actual base unit, the cellobiose, consists of two molecules of glucose. For
this reason, cellulose can also be considered as a (syndiotactic) polyacetal of glucose.
Basic Structure. The basic chemical formula of cellulose is the following:
C6PH10P+2O5P+1 » (C6H10O5)P or (C6H10O5)n
where P = the degree of polymerization; n = the number of units in the chain.
The elemental composition of 44.4 % C, 6.2 % H, and 49.4 % O was already known to
PAYEN in 1842 [9]. The molecular mass of the glucose base unit is m0 = 162, and the
molecular mass of the cellulose polymer is
Constitutional Formula. HAWORTH [10] first discovered the covalent bonds inside and
between the glucose units while STAUDINGER [11] found the final proof for the
macromolecular nature of the cellulose molecule.
Conformational Formula. The glucopyranosic ring adopts a 4C1 chair conformation, as
revealed by recent X-ray crystallography and nuclear magnetic resonance studies [12],
[13] with glucose. The chair formation in comparison to the tray conformation exhibits a
free stabilization enthalpy of GS = 20.05 kJ/mol [14]. In this conformation, the three
hydroxyl groups are positioned in the ring plane while the hydrogen atoms are in a
vertical position. It seems only natural to assume that the same conformation also exists
in the cellulose molecule.
Structural Anomalies. As a naturally occurring polymer, cellulose always contains
small amounts of other constituents in addition to glucose (over 99 %). These may
already be partially built into or onto the cellulose molecules during biosynthesis, such as
lignin–cellulose complexes [15]. Most of the changes in the molecular structure,
however, result from secondary reactions, i.e., hydrolysis or oxidation, during isolation
from natural sources. For morphological reasons, such chemical changes occur
preferably in the accessible interlinking regions between the crystallites of the
elementary fibrils or their aggregations. The glucosidic links in these accessible areas,
especially if oxidized sites are also present, split 1000 – 5000 times faster than
glucosidic linkages inside the well-ordered crystallites. The existence of weak links, as
proposed in reference [16], is hard to determine. In homogeneous acid hydrolysis, all
glucosidic linkages split at the same rate [17].
Cellulose always contains carboxyl groups: In wood pulp, one –COOH group per 100 –
1000 anhydroglucose units (AHG) exists; in cotton, one –COOH group per 100 – 500
AHG units.
Molecular Size. The molecular size of a polymer can be defined by its average
molecular mass (Mr) or its average degree of polymerization (P); whereby Mr = P m0 (m0
= molecular mass of the base unit, i.e., of glucose in the case of cellulose).
By investigation of certain physical properties of cellulose or polyhomologous cellulose
derivative solutions, the average degree of polymerization can be determined. Table (1)
lists the number average degree of polymerization of a number of celluloses of various
origin.
Table 1
For such physical investigations, cellulose solutions in aqueous copper(II)tetrammonium
hydroxide (Schweitzer's reagent; Cuoxam), copper(II)ethylenediamine hydroxide (Cuen),
alkaline solutions of the ethylenediamine complexes of cadmium or nickel can be used.
Cellulose trinitrate (CTN) or cellulose tricarbanilate (CTC) solutions in appropriate
solvents are also suitable for such studies (see Table (2)). In the latter, it should be kept
in mind that chain degradation often occurs in substitution reactions performed under
unfavorable conditions.
Table 2
Light scattering studies performed on dilute solutions of cellulose or cellulose derivatives
will yield the weight average (Mw) and osmotic measurements the number average (Mn)
of the molecular mass (or the corresponding average degrees of polymerization: PW or
Pn). Sedimentation experiments in an ultracentrifuge enable the determination of a
higher order average molecular mass, the so-called "Z-average" (Mz). These various
quantities are defined as follows:
The simplest and most widely applied practical method for the determination of the
degree of polymerization is based on measuring the "intrinsic viscosity " (Staudinger
index). The intrinsic viscosity expresses the reduced viscosity of a solution at an infinitely
small concentration. The latter can be derived from the relative viscosity, which is the
ratio of the flow time of the dilute polymer solution of a given concentration (tps) and that
of the solvent (ts) in a capillary viscometer:
wherein wi = weight fraction of a molecularly uniform fraction with a degree of
polymerization of Pi ; Pv = viscosity average of the degree of polymerization which for
cellulose or cellulose derivative solutions closely resembles the weight average Pw; and
Mv = viscosity average of the molecular mass.
Table (2) summarizes some of the more important a, Km , and Kp values. These values
have been obtained by calibration to one of the above-mentioned absolute methods for
the determination of the degree of polymerization.
Quite often one speaks simply of the "average degree of polymerization" (DP), which is
misleading, however, unless the method of determination is properly stated (Pn , Pw , Pz ,
or Pv).
Polymolecularity. Cellulose isolated from its native sources is always polydisperse; i.e.,
it consists, as do all polymers, of a mixture of molecules with the same basic
composition and chemical constitution but differing widely in their chain length or degree
of polymerization, respectively. The relative amounts of molecules of various lengths
present in a given cellulose substrate can be specified by the so-called differential mass
distribution curves. Unlike most synthetic polymers, cellulose substrates have
complicated mass distribution functions. Figure (1) shows typical mass distributions of
various cellulose samples.
Figure 1: Differential mass distribution curves of various celluloses [18]
a) Cotton; b) Cotton; c) China grass (ramine); d) Flax; e) Ramie; f)Balsam; g) White fir;
h) Birch
Knowledge of the molecular mass distribution is important for many applications.
However, most of the methods applied in the past, such as fractionation by precipitation
or selected dissolution, consume a great deal of time and are often subject to objection.
In recent years, gelpermeation chromatography, especially in combination with lowangle laser light scattering, was developed to such a state as to provide a fast and highly
reproducible method for this purpose [21].
As a convenient measure of the broadness of the mass distribution, the so-called
nonuniformity factor (NU) is often used:
A polymer with a normal (most probable) molecular mass distribution will have a
nonuniformity factor of 1. In most cases, cellulose substrates show much higher
molecular nonuniformities.
Secondary Structure. In solution, the cellulose molecules exist in form of largely
expanded coils. In addition to isolated and solvated molecules, cellulose and cellulose
derivative solutions frequently also contain supermolecular gel particles, so-called
micelles [22].
In solid cellulose, high-order microcrystalline structures ("crystalline regions") alternate
with those of a distinctly lower order ("amorphous regions"). Cellulose is polymorph; i.e.,
depending on the origin or the conditions during isolation or conversion, cellulose will
have or can adopt various crystal lattice structures.
Lattice Structure. Native celluloses all show the so-called cellulose I lattice structure.
Each unit cell houses two countercurrently arranged cellulose molecules. The lattice of
cellulose I is of the monoclinic sphenolitic type. The cellulose chains are in line with the b
axis of the unit cell. Figure 2 shows a schematic of the unit cell of the cellulose I
modification.
Figure 2
The cellulose molecules are aligned in the fibrillar axis and form the b axis of the unit
cell. The length of the b axis is 1.03 nm (10.3 Å), somewhat shorter than the extended
length of a cellobiose unit, which suggests a slight helical twist in the cellulose chains
along the b axis [23][24][25]. This twist is caused by intramolecular hydrogen bonding
primarily between the hydroxyl groups on the carbon atom C–3 of one glucose unit and
the pyranose ring oxygen in the adjacent glucose unit of the same chain molecule. This
intramolecular secondary valence bond is also responsible for the relative rigidity of the
cellulose molecule [26]. Some authors [27] suggest a second intramolecular hydrogen
bond involving the hydroxyl groups on C–6 and C–2 of adjacent glucose units in the
same molecule.
In more recent years, some researchers suggested a unit cell for cellulose I in which the
a and c axis of the Meyer–Mark–Misch model are doubled [28], [29]. However, these
newer interpretations of X-ray and electron diffraction results are more or less closely
related to the Meyer–Mark–Misch lattice structure. In reference [30], it is claimed that
there is a closer agreement with the observed diffraction intensity data for parallel
arrangement of the cellulose molecules in the unit cell. However, this is still controversial
[31].
The internal cohesion of the cellulose molecules in the unit cells and crystalline domains
is due to intermolecular secondary valences — partly hydrogen bonds and partly van der
Waal's forces. These bonds can act either between molecules situated in the same
crystal lattice plane (intraplanar bonds) or between molecules located in neighboring
lattice planes (interplanar bonds). The intraplanar hydrogen bonds are formed primarily
between adjacent cellulose molecules in the same 002 lattice planes giving a sheetlike
structure. The 002 sheets are then bonded to one another by hydrogen bridges involving
the hydroxyl groups on C–6 and the glucosidic ring oxygen atoms of cellulose molecules
favorably located in neighboring 002 planes, or by van der Waal's forces acting between
neighboring glucopyranose rings.
The unit cells of the other polymorphic structures of cellulose — the most important one
being the so-called cellulose II — differ basically in the lengths of their a and c axis and
the angle of inclination b. The cellulose II modification is formed as the
thermodynamically most stable polymorph when cellulose fibers are treated with
concentrated sodium hydroxide solution (> 14 %) or precipitated (regenerated) from
solution.
In addition to the cellulose I and cellulose II modifications, two other polymorphic lattice
structure are known, the cellulose III and cellulose IV crystal modifications. The cellulose
III structure is formed when the reaction product of native cellulose fibers is decomposed
with liquid ammonia. This modification has a lattice structure closely related to that of
cellulose II. The cellulose IV modification is obtained by treating regenerated cellulose
fibers in hot baths under stretch. The lattice of this polymorph is closely related to that of
cellulose I. Some distinct differences in their infrared absorption spectra seem to indicate
their existence. Some researchers however doubt their actual existence [32].
Table (3) lists the lattice parameters of the unit cells of these four polymorphic crystal
structures.
Table 3
Crystallites. The ability of hydroxyl groups to form secondary valence hydrogen bonds is
– together with the stiff and straight chain nature of the cellulose molecule – the cause
for the high tendency to organize into crystallites in parallel arrangement and crystallite
strands (elementary fibrils), the basic elements of the supermolecular structure of
cellulose fibers.
The dimensions of the elementary crystallites differ only slightly for native or regenerated
cellulose fibers. Their length ranges between 12 and 20 nm (= 24 – 40 glucose units)
and their width between 2.5 and 4.0 nm. The often observed larger "micro- or
macrofibrils" (or fragments thereof) are aggregations of elementary fibrils.
Two questions concerning the crystal structure are still under dispute. The first deals
with the antiparallel or parallel arrangement of the cellulose molecules in the crystal
lattice as previously mentioned. The second question (still open) concerns the existence
or nonexistence of folded chains in the lattice [33][34][35]. While a folded cellulose chain
position in the lattice seems unlikely to most experts, the parallel molecule arrangement
in the cellulose I lattice is principally acceptable, under the condition that two cellulose II
lattice structures exist, one for heterogeneously mercerized native celluloses with
parallel arrangement of the molecules and the other for regenerated cellulose substrates
with antiparallel molecule arrangement.
1.1.2. Supermolecular Structure (Texture)
The basic structural element of cellulose fibers is the so-called elementary fibril. It can be
seen with the electron microscope, as illustrated in Figure (3)
Figure 3
The cross-dimensions of the elementary fibrils correspond with those of the elementary
crystallites. The elementary fibril is a strand of elementary crystals linked together by
segments of long cellulose molecules. The lateral order in the interlinking regions is
distinctly less pronounced (amorphous). This structure is schematically shown in
Figure 4 [36][37][38][39].
Figure 4
Several elementary fibrils associate to form larger aggregations of so-called microfibrils
and macrofibrils, which can also be seen with a light microscope.
The elementary fibrils and their aggregations are determined by nature in such native
fibers as cotton or wood pulp fibers and are laid down in various cell wall layers in a
typical manner [18], [19]. Figure 5 shows the structural organisation of wood pulp and
cotton fibers.
FIGURE 5
Figure 5
Synthetic cellulose fibers, such as viscose, do not have a native morphology. Their
supermolecular structure can be described as a network of elementary fibrils and their
more or less random associations. This is called a "fringe fibrillar" structure [40], which is
shown in Figure (6).
Figure 6
Structure Characterization. The methods used to characterize the molecular and fine
structure of native and synthetic cellulose fibers include the following [41]:
1. determination of the average degree of polymerization (Pn) by the osmotic
method;
2. determination of the average crystallite length by meridional X-ray low-angle
scattering on slightly hydrolyzed fiber samples or by measurement of the band
width of the meridional 040 X-ray wide-angle reflection at half-maximum
intensity;
3. determination of the degree of order ("crystallinity," CrI) with a method for
separating overlapping equatorial X-ray diffractions [42] and deriving from the
band width at half-maximum intensity the average cross-dimensions of the
crystalline regions; furthermore, this analysis yields information on the lattice
structure, polymorphic composition, and accessibility;
4. determination of the degree of orientation by measuring the azimuthal intensity
distribution of major equatorial X-ray diffraction arcs or by IR dichroism.
Structure and Properties. Physicomechanical properties of cellulose fibers such as
tenacities, elongations, or moduli in the conditioned or wet state are determined by the
following structural parameters [43]:
1. the average length of the fiber-forming molecules (Pn);
2. the average length of the elementary crystallites (PnL = number average "limiting"
degree of polymerization);
3. the degree of lateral order (crystallinity, CrI);
4. the degree of orientation (fr) with respect to the fiber axis; and
5. the presence of heterogeneities (natural defects, incorporated gel or sand
particles, etc.).
This may be illustrated by the following examples: As shown by Figure 7, the tenacity of
the conditioned fibers is determined by the length of the molecules in relation to the
length of the elementary crystallites building the elementary fibrils (1/PnL – 1/Pn), by the
degree of order (CrI), and by the degree of orientation (fr).
Figure 7
The elongation at break in the conditioned state is mainly dependent on the degree of
– 1) in which
orientation. Simple geometric considerations give the parameter (1/cos
the angle a derived from the orientation factor (fr) is the mean angle of deviation of the
basic structure elements from the fiber axis. Figure 8 illustrates the relation of this
parameter to the breaking elongation of a number of cellulosic fibers.
Figure 8
The wet moduli of the various fibers show a close relation to the product of the length of
the elementary crystallites (PnL), the degree of order (CrI), and the square of the
orientation factor (fr 2). This is demonstrated in Figure (9).
Figure 9
1.1.3. Physical Properties
Cellulose is relatively hygroscopic. Under normal atmospheric conditions (20 °C, 60 %
relative humidity), it adsorbs ca. 8 – 14 % water. Cellulose swells in water (see Table
(4)). It is, however, insoluble in water or dilute acids. In concentrated acids, solution can
be achieved under severe degradation. Caustic solutions cause extensive swelling and
dissolution of low molecular mass portions (P <= 200). Solvents for cellulose are listed in
Table (2)).
Table 2
Cellulose is nonmelting; thermal decomposition starts at 180 °C; the ignition point is
> 290 °C. With chlorine and zinc iodide, cellulose takes on a red-violet to blue color; with
phloroglucinol–hydrochloric acid, pure cellulose should not take on a red color (test for
residual lignin). Additional data:
Density: 1.52 – 1.59 g/cm3
Refractive index:
1.62 parallel to the fiber axis
1.54 perpendicular to the fiber axis
Dielectric constant: 2.2 – 7.2 (at 50 Hz). Highly dependent on humidity conditions.
Insulation resistance: 1014 – 1017 W cm. Highly dependent on humidity conditions.
Electric strength: 500 kV/cm
Heat of combustion: 17.46 J/g
Heat of crystallization: 18.7 – 21.8 kJ/mol of glucose
Specific heat: 1.00 – 1.21 J g–1 K–1
Coefficient of thermal conductivity:
0.255 kJ m–1 h–1 K–1 (loosely packed) to
0.920 kJ m–1 h–1 K–1 (compressed)
Specific internal surface: 10 – 200 m2/g
1.1.4. Chemical Properties
The chemical reactivity of cellulose is determined to a large extent by the
supermolecular structure of its solid state. Most of the reactions on cellulose fibers are
heterogeneous in nature. The reaction medium acts on a two-phase solid system: (a) the
less-ordered (amorphous) regions which are mainly located on the surface of the
elementary fibrils or their aggregations and in the interlinking regions between the
elementary crystallites in the fibrils, and (b) the well-ordered elementary crystallites or
fused associations of the elementary fibrils. Any reaction will first start on the lessordered surface of the elementary fibrils or their aggregations (topochemical reaction)
and then, under favorable conditions, proceed into the interlinking regions between the
elementary crystallites to penetrate from both ends into the crystallites. Therefore, as
long as the reaction is limited to the accessible surface of the fibrils or fibrillar
aggregations and the regions interlinking the elementary crystallites (i.e., up to degrees
of substitution (DS) of 1.3 – 1.7), there is no visible effect in the crystalline structure. At
increased degrees of substitution (to ca. DS = 2.5), the X-ray diffractogram shows
overlapping diffraction bands of the original cellulose I structure and the cellulose
derivative. At still higher degrees of substitution, the pure diffraction pattern of the
derivative will finally result. This course of reaction implies that partially substituted
cellulose derivatives are always a mixture of completely substituted cellulose, partially
substituted portions (of block-polymer nature), and unsubstituted cellulose.
A quasi-homogeneous reaction can be achieved when the fiber structure is loosened by
swelling treatments to such an extent that all cellulose molecules can react
simultaneously. A real homogeneous reaction can, however, only be achieved by
bringing the cellulose into a molecularly dispersed solution.
The reactivity of cellulose substrates can be greatly enhanced by activation treatments,
such as swelling, solvent exchange, inclusion of structure-loosening additives,
degradation, or mechanical grinding, which enlarge accessible surfaces by opening
fibrillar aggregations. These treatments also restore, in most cases, the loss of reactivity
due to so-called hornification, which occurs when water is removed from cellulose by
drying under severe conditions.
Swelling with water or other polar liquids is the most frequently applied activation
treatment. It exclusively opens the interfibrillar interstices and swells the less-ordered
surface and interlinking regions of the fibrillar elements. The solvent exchange technique
is a special kind of activation from the water-swollen state. It allows the introduction of
media being inert in subsequent reactions that are unable to swell the cellulose
substrate thus maintaining the reactive water-swollen state. An interesting variation of
the solvent exchange treatment is the so-called inclusion technique [44]. Inert liquids,
such as cyclohexane or benzene, are introduced into the cellulose substrate by solvent
exchange from the water-swollen state. During drying, they are permanently
incorporated into the interfibrillar interstices or voids, thus preventing fusion of fibrils, i.e.,
the hornification. Such inclusion celluloses are very reactive, as shown in Table (5).
Table 5
Another very effective way of activating cellulose fibers is to enhance the accessibility of
fibrillar surfaces and to open the less-ordered regions interlinking the crystallites in the
fibrils by treatment with systems causing not only interfibrillar, but also intracrystalline,
swelling. Some inorganic acids, various salt solutions, and especially certain inorganic
and organic bases achieve this at distinct concentrations. They apparently penetrate the
fiber through existing capillaries and pores by opening the fibrillar interstices and
entering the interlinking regions between the crystallites. From there, they enter the
elementary crystallites from both ends and force them open. At suitable concentrations
and temperatures, they ultimately cause crystal lattice transfer, particularly with respect
to opening the 101-plane distances (see Table (6)).
Table 6
So-called mercerization is a frequently used practical method of activation, i.e., the
treatment of native cellulose substrates with 10 – 20 % sodium hydroxide solutions at
moderate temperature (< 20 °C). In this treatment, sodium cellulose I is formed in which
the 101-plane distance is increased from 0.61 nm (6.1 Å) in native cellulose to 1.22 nm
(12.2 Å). In this lattice transition, the glucopyranose rings are dislocated and aligned into
the 101 lattice plane. The hydroxyl groups on the C–2 and C–6 carbon atoms are thus
freely exposed and jut into the widened space between the 101 lattice planes, making
them accessible for reactions.
The chemical character of cellulose is determined by the sensitivity of the b-glucosidic
linkages between the glucose repeating units to hydrolytic attack and the presence of
one primary and two secondary reactive hydroxyl groups in each of the glucopyranose
units. These reactive hydroxyl groups are able to undergo exchange, oxidation, and
substitution reactions, such as esterification and etherification.
Sorption and Exchange Reactions. Cellulose undergoes sorption and exchange
reactions with water and deuterium oxide. These reactions are of special interest since
they give a good indication of the accessibility of the cellulose substrate.
The cross-dimensions of the well-ordered regions can be derived from the width of the
equatorial X-ray wide-angle diffractions. If it is assumed that the crystallites are
surrounded by one layer of disturbed ("amorphous") unit cells and also that the
molecules in the next outer layer of the well-ordered crystalline core of the crystallites
are in addition accessible for reactants such as water or deuterium oxide, an explanation
for the extent of water adsorption and deuterium exchange should be possible, see
Figures 10A & 10B [46].
Figure 10 A
Figure 10B
Table 7 compares the experimentally obtained water sorption and deuterium exchange
values with those predicted from structure investigations according to the above outlined
concept.
Table 7
The internal surface data calculated from crystallite cross-dimensions are also in good
agreement with those determined by gas adsorption.
Degradation by Acid Hydrolysis. Degradation in acidic medium is based on the
hydrolysis of the b-glucosidic linkages between the glucose base units. The reaction
depends strongly on pH and already proceeds at a remarkable rate at low pH and
temperatures well under 100 °C. Initially the acetal oxygen of the glucosidic linkage is
protonated. Through heterolysis, an intermediate carbonium ion is formed, causing
chain-splitting. The carbonium ion finally reacts with water, which reforms the proton.
The following reaction scheme illustrates the course of this reaction.
Homogeneous and heterogeneous hydrolysis of cellulose are both first-order reactions.
The reaction speed is strongly dependent on the acid and the cellulosic material. The
course of homogeneous and heterogeneous hydrolysis also reveals a basic difference.
In homogeneous hydrolysis degradation proceeds at a constant rate until all of the
cellulose is degraded to cellobiose or glucose, respectively. In heterogeneous
hydrolysis, the rate decreases continuously and degradation stops almost completely
when the number average degree of polymerization reaches 25 – 100. This corresponds
to the length of the elementary crystallites ("level-off degree of polymerization").
Hydrolysis proceeds thereafter at a very slow rate. It is interesting to note that during the
course of homogeneous hydrolysis, the degraded and isolated residue adopts an
increasingly normal molecular mass distribution as indicated when the ratio of its weight
to the number average degree of polymerization approaches a value of 2. This is
indicative of a statistical degradation (see Table (8)).
Table 8
In contrast, the residue in heterogeneous hydrolysis tends toward a value of 1 for the
ratio of its weight to number average molecular mass. This is a strong indication that the
course of heterogeneous degradation is not determined solely by the sensitivity of the bglucosidic linkages, but primarily by morphological aspects [49]. Hydrolytic attack is
almost completely limited to the molecules situated on the surface of the fibrillar strands
and the accessible molecule segments connecting the crystallites. It is also important to
note that mass loss in heterogeneous hydrolysis performed under moderate conditions
is relatively small in the initial fast reaction.
Degradation in Alkaline Media. Hydrolysis of the b-glucosidic linkages in alkaline
media occurs at a significant rate only at temperatures above 150 °C. It is most probable
that chain-splitting proceeds by way of the 1,2-anhydro configuration [50].
Acidic as well as alkaline hydrolysis of the glucosidic bond is remarkably enhanced (belimination) by oxidative changes at the C–2, C–3, or C–6 carbons leading to carbonyl
groups. An example of this chain-splitting reaction, which can even occur at moderate
temperatures, is [51]:
An additional degradation reaction taking place in alkaline media is the so-called peeling
reaction. In the course of this reaction, which even takes place at temperatures well
below 100 °C, the cellulose chain molecules are degraded step-by-step, beginning at the
reducing end and proceeding in a "zipper-like" reaction. The terminal glucose unit is first
transformed into the 1,2-enediol, which isomerizes to the corresponding ketose and
splits off the chain. The ketose is transformed further into the alkali-stable isosaccharinic
acid [52].
The newly formed aldehydic end of the cellulosic chain will repeat the same reaction.
When ca. 50 – 60 glucose units are split off under these conditions, the reaction
normally stops due to the interference of a chain-stopping reaction. This termination
reaction leads by way of the 2,3-enediol to an alkali-stable metasaccharinic acid end
group, which stabilizes the cellulose molecules against further degradation.
Termination Reaction
The degradation in acidic as well as in alkaline media is of great importance in the
manufacture of pulp from wood and other plants and in the processing of cellulose
derivatives, regenerated fibers, and films.
The microbiological degradation of cellulose should also be mentioned in this
connection. This degradation takes place through enzymatic hydrolysis of the bglucosidic linkages and has recently found increasing interest especially in connection
with the use of plant biomass [53].
Oxidation Reactions. The hydroxyl groups and the aldehydic end groups participate in
the oxidation reactions of cellulose. These reactions form aldehyde, ketone, and
carboxyl groups. Extensively oxidized and degraded products are designated as
oxycelluloses. Some oxidizing agents show specific action. They attack only specific
functional groups, forming defined oxidation products. Other oxidants react
nonspecifically with all types of oxidizable groups in the cellulose molecules. Under
special conditions hypoiodite and chlorite attack only the aldehyde end group on C–1,
oxidizing it to form a carboxylic group. Another oxidant with specific action is periodate,
which attacks the glycol configuration on the carbon atoms C–2 and C–3, thus causing
ring-splitting and forming a di-aldehyde structure.
Nitrogen dioxide (dinitrogen tetroxide) reacts not quite as specifically. It oxidizes,
however, with a certain preference the hydroxyl group on C–6 to a carboxyl group and to
a lesser extent hydroxyl groups on C–2 and C–3 to ketone groups.
Nonspecific oxidants are chlorine, hypochlorite and chromic acid. They also oxidize all
accessible hydroxyl groups to aldehyde, ketone, and carboxyl groups. The oxidative
action of chlorine and hypochlorite is extensively used as a bleaching agent in the pulp
industry. However, one must keep in mind that the introduction of carbonyl groups on C–
2, C–3, and C–6 causes alkali instability of glucosidic linkages in the b-position which
initiates degradation under alkaline conditions.
Esterification and Etherification. Both of these substitution reactions are used in
industry for the manufacture of widely used products (Cellulose Esters, Cellulose
Ethers). The acetate and nitrate esters and the methyl and carboxymethyl ethers of
cellulose, and to a lesser extent the ethyl and hydroxyethyl ethers, have acquired
practical significance.
To perform the esterification or the etherification reaction properly, the hydroxyl groups
in the cellulose substrate must be made accessible for the reaction. The supermolecular
structure of the cellulose must be activated before or in the course of the substitution
reaction. For this purpose the cellulose substrate is treated with strong acids or alkali. In
these treatments addition compounds are formed between the cellulose, acid, or alkali
and the water present in the system. During the opening of hydrogen bonds between the
molecules in the cellulose substrate, more or less defined addition compounds are
formed in the crystalline regions. Their formation is often accompanied by changes in the
lattice structure, resulting mostly in an increased distance between the 101 planes. This
exposes the hydroxyl groups on C–2 and C–6, making them accessible. Hydrated
cations or anions of the reactant, respectively, are incorporated into the widened
interplanar space, where they initiate and facilitate the reactions. The addition
compounds thus formed are only stable in equilibrium with the reactant and will
decompose readily when the system is diluted with water.
The basic principle of the esterification and etherification reaction is quite similar. The
first step is a nucleophilic substitution or addition and the formation of an oxonium ion on
the carbon atom carrying the reactive hydroxyl group. A surplus of esterification or
etherification reactant will lead to the formation of the corresponding ester or ether:
1. General esterification mechanism (formation of an oxonium ion; inorganic acid
esterification)
2. Esterification with organic acids (nucleophilic addition)
3. Acid-catalyzed esterification
4. General etherification mechanism (formation of an oxonium ion; alcohol excess
leads to ether formation)
5. Etherification of cellulose with alkali consumption
6. Etherification of cellulose without alkali consumption
The esterification reaction is promoted by water-binding catalysts. In etherification, prior
swelling of the cellulose substrate with alkali or the transfer to alkaline cellulose is
essential for the reaction.
The presence of three hydroxyl groups in each glucose unit allows the formation of
mono-, di-, and triesters or ethers, respectively. Contrary to the reaction behavior of
primary and secondary hydroxyl groups in low molecular mass alcohols, where the
primary hydroxyl group always shows a higher reactivity, the secondary hydroxyl group
on C–2 quite often shows preferred reactivity in heterogeneous esterification or
etherification of cellulose. Table (9) illustrates this for the relative reaction rate of the
hydroxyl groups on C–2, C–3, and C–6 in various etherification reactions [54].
Table 9
The normally expected preferred reactivity of primary hydroxyl groups on C–6 can be
observed only in substitution reactions on cellulose in solution. The fact that most
substitution reactions on cellulose are performed in heterogeneous systems also has the
consequence that most partially substituted cellulose derivatives are actually mixtures of
fully substituted, irregularly substituted (block substitution), and unsubstituted cellulose
molecules.
The extent of substitution in cellulose derivatives is described by the so-called "degree of
substitution" (DS). This value states the average number of substituents linked to each
glucose base unit in the cellulose molecules. Since substitution normally occurs
irregularly along the cellulose chains, the DS can assume any value between 0 and 3.
Industrially produced cellulose esters and ethers find practical use as fibers, films,
lacquers, explosives, adhesives, and as auxiliaries in paper, textile, and food industries.
Their properties and, consequently, their applications are primarily dependent on the
nature of the substituents, the degree of substitution, the distribution of substituents
along the cellulose molecules and from molecule to molecule, and their degree of
polymerization.
In particular, the xanthation of cellulose is of great economic importance, i.e., the
esterification with dithiocarbonic acid. This reaction is the basis for the so-called viscose
process, which is widely used for the manufacture of regenerated cellulose fibers and
films. The cellulose xanthate is obtained in the reaction of alkali cellulose with carbon
disulfide and is soluble even in dilute sodium hydroxide solutions at relatively low
degrees of substitution (DS < 1).
Graft Copolymerization. During the last two decades, another way of modifying
cellulose, the grafting reaction, has found substantial interest. This method allows the
attachment of chemically different side-chains to a given polymer molecule. The reaction
mechanisms are principally the same as in the synthesis of polymers [55], [56]. The
most frequently used method is to initiate grafting with radical catalysts. To achieve graft
copolymerization, the cellulose substrate must be in the presence of peroxides or redox
systems (i.e., hydrogen peroxide, hydrogen peroxide – iron(III) ions, Ce(IV) ions) and as
such is able to undergo polymerization with such compounds as styrene, acrylic acid,
acrylic ester, etc. The introduction of substituents that are able to enhance radical
transfer, such as thiol or xanthate groups, promotes the grafting reaction. Industrial
applications for graft-modified cellulose fibers are thus far very limited.