`Rheological Behavior of Polysaccharides Aqueous Systems
`
`Jacques Lefebvre and Jean-Louis Doublier
`INRA-Laboratoire de Physico-Chimie des Macromole´cules, Nantes, France
`
`I.
`
`INTRODUCTION
`
`The relevance of rheology to polysaccharide studies con-
`cerns both their practical and their fundamental aspects.
`The capacity of polysaccharides to extensively modify
`the rheology of aqueous media into which they are intro-
`duced, even at fairly low concentrations, is the basis of their
`‘‘functional properties’’ as thickening and gelling agents. It
`is also involved in many other types of applications, such as
`encapsulation, controlled release, etc. Some degree of
`rheological characterization is essential in particular to
`evaluate the potential uses of a polysaccharide as extracted
`from a natural source or subsequently modified.
`On the other hand, rheology provides precious tools to
`explore and understand the properties of polysaccharides in
`aqueous systems. The rheological behavior of polymer
`systems manifests the underlying structure of the systems.
`In the simplest case, that of polymer solutions, viscosity is
`directly related to fundamental molecular properties (mo-
`lecular conformation, molecular weight and molecular
`weight distribution, intramolecular and intermolecular inter-
`actions). In the case of more structured polymer systems,
`gels, for example, their viscoelastic properties are related to
`supramolecular organization. Although the relation between
`structure and rheological properties subsequently becomes
`more complex and less direct, rheology allows us to probe
`the structure of the systems at different scales, in conditions
`where other physical methods are impossible or difficult to
`use. Rheological techniques are especially useful to monitor
`and to probe structural changes in the systems, such as
`gelation or phase separation processes.
`Polymer rheology, which has been thoroughly studied
`and set up on firm theoretical foundations, provides a
`general frame for the investigation and the interpretation
`of the rheological behavior of polysaccharide systems.
`However, usually the parallel cannot be drawn very far
`into the details, particularly on quantitative grounds, as the
`following short discussion will make clear.
`
`A remarkable variety of physical chemical properties
`reflect the structural diversity of polysaccharides. Even
`chemically related polysaccharides can behave quite differ-
`ently in aqueous media, and, furthermore, in a way which
`strongly depends on solvent conditions. The same applies
`by way of consequence to their rheological properties.
`Polysaccharides are seldom homopolymers; in most
`cases, their backbones comprise several types of sugar
`monomers linked in sequences which, when indeed not
`totally unknown, are not characterized in detail. The
`‘‘heteropolymeric’’ character is responsible for the capacity
`of many polysaccharides to form gels in certain conditions.
`Moreover, polysaccharides are often branched polymers;
`the degree and pattern of branching, the lengths of the side
`chains, their composition itself in many cases, are generally
`ill defined. Among the polysaccharides that behave on the
`whole as linear homopolymers from a rheological point of
`view, many can be nevertheless grafted with short side
`chains, in which the number, length, and distribution along
`the backbone do play an important role in their properties.
`On the other hand, the presence of a few of sugar hetero-
`monomers inserted along an otherwise uniform linear
`sequence has a strong effect on the conformation of the
`chain and its physical chemical behavior. Such limited
`structural differences often result in a large variety of
`rheological properties, which can be observed among
`polysaccharides as belonging to the same chemical class
`but arising from different biological origins.
`Extraction, purification, and fractionation processes
`are another source of structural diversity. First, some
`polysaccharide or protein impurities, the nature and quan-
`tity of which depend on the purification procedure, are
`likely to remain in the sample. Their presence has been
`found to deeply affect the properties of the polysaccharide
`under consideration, or can be suspected to do so. Sup-
`posing this problem is solved, because in the initial material
`the polysaccharide of interest certainly presents some
`degree of structural heterogeneity of natural origin, the
`
`Copyright 2005 by Marcel Dekker
`
`357
`
`ALL 2032
`PROLLENIUM V. ALLERGAN
`IPR2019-01505 et al.
`
`
`
`358
`
`Lefebvre and Doublier
`
`exact composition and structure of the purified sample will
`depend on the way it has been prepared. This is the very
`consequence of the selectivity of the extraction and frac-
`tionation operations themselves. Finally, chemical modifi-
`cations of the polysaccharide are likely to occur during the
`preparation of the sample, giving rise to an artifactual
`heterogeneity; an example is backbone hydrolysis and
`partial demethylation of pectins during their extraction
`and purification. Indeed, many different samples may be
`obtained with as many preparation procedures! As a result,
`purified polysaccharide samples generally show at the same
`time very large polydispersity, i.e., they contain macro-
`molecules with widely different molecular weights and
`polymolecularity—they contain chains with qualitatively
`similar compositions, but differing in their quantitative
`compositions and in their structures. With much work,
`polysaccharide fractions with relatively narrow polydis-
`persity can be obtained (to be used as molecular weight
`standards, for example), but in most cases, it is practically
`impossible to avoid some degree of polymolecularity.
`One more source of complexity in the behavior of
`polysaccharide systems is the special nature of the sol-
`vent—water—and the delicate balance between chain–
`chain and chain–solvent interactions (hydrogen bonding,
`hydrophobic interaction,
`ionic interactions). Small
`changes in ionic strength or ionic composition, and limited
`shifts of temperature can induce a change in the physical
`nature of the system, e.g., a transition from the state of a
`macromolecular solution to that of a gel, causing a drastic
`modification of its rheological behavior.
`The polydisperse and polymolecular character of poly-
`saccharides, as they are available in practice, and the fact
`that their macromolecular structure is generally only quite
`vaguely known put severe limitations to the quantitative
`application of polymer rheology theories, whereas their
`structural diversity and complexity give rise to an extremely
`rich phenomenology, which largely gives rise to their
`functional versatility. However, within a defined physical
`state, the rheological behaviors of polysaccharide/(aque-
`ous) solvent systems are amenable to a general pattern, and
`they essentially differ on quantitative grounds. Structural
`differences mainly show in the conditions controlling the
`shift from one physical state to the other (from the solution
`to the gel and conversely, or from the solution to the
`dispersion or to the precipitate in phase separating sys-
`tems). Accordingly, on one hand, theoretical considera-
`tions will be kept to a minimum in this chapter. On the
`other hand, the rheology-relevant specific features perti-
`nent to the different classes of polysaccharides will not be
`surveyed. Our intention is neither to scan the different
`classes of polysaccharides, nor to draw a panorama of the
`application of different rheological techniques in this field;
`several valuable books and extensive reviews, written along
`these lines, are available and will be quoted in due place. We
`shall instead consider the main types of polysaccharide
`aqueous systems, and show schematically how rheological
`studies can provide an insight into their organization,
`focusing more on the common features than on the specific
`properties of each polysaccharide. The main types of
`polysaccharide systems that are encountered in the appli-
`
`cations can be distributed schematically in three classes:
`solutions, gels, and polysaccharide/polysaccharide (or
`polysaccharide/protein) mixtures in aqueous media. The
`last class comprises an extremely broad spectrum of sys-
`tems ranging from mixed solutions to complex structured
`multiphase systems. Biopolymers, even differing little in
`composition or structure, are in effect generally incompat-
`ible in aqueous media. As a consequence, the simultaneous
`presence of two polysaccharides (or of a polysaccharide
`and a protein) in the system results in a variety of micro-
`structures through phase separation processes. These pro-
`cesses can interfere or combine with gelation processes in a
`way that is highly dependent on the details of the structure
`of the biopolymers and on the experimental conditions.
`This gives rise to a fascinating variety of morphologies at
`different spatial scales, and, among other applications,
`allows ‘‘tuning’’ the rheological behavior of the systems
`to suit specific requirements. In spite of the theoretical and
`practical interest of polysaccharide mixtures, we shall
`restrict this chapter to simple solutions and gels, the
`rheology of which marks out the field of the behaviors of
`polysaccharide aqueous systems. Only shear deformation
`will be considered. Finally, the discussion of viscoelasticity
`will be limited to the linear domain.
`
`II. POLYSACCHARIDE SOLUTIONS:
`GENERAL REMARKS
`
`True polymer solutions are defined by the conjunction of
`the two following characteristics: (1) they are thermody-
`namically stable systems; (2) only physical interactions
`exist between the coils—hydrodynamic interactions and
`topological constrains which develop above a critical con-
`centration (coil overlap concentration).
`Rheology of polymer solutions has been extensively
`studied and thorough theoretical treatments are available,
`at least for linear neutral chains. This provides a frame to
`understand the rheology of polysaccharide solutions.
`However, because of polydispersity, polymolecularity,
`and molecular interactions, departures from polymer laws
`are often observed.
`In particular, obtaining true polysaccharide solutions
`is often not trivial. Polysaccharides are well-known to
`manifest a strong propensity to associate via hydrogen
`bonds, because of the abundance of hydroxyl groups. This
`is indeed the basis of the gelling properties of polysaccha-
`rides such as amylose and amylopectin, agarose, etc. but
`reversible and/or irreversible association occurs also in
`many other cases, although it does not lead to gel forma-
`tion in usual observation conditions. In any event, solubi-
`lization of polysaccharides is always difficult when the
`concentration is not low. Aggregation and incomplete
`solubilization result in the presence of microgels of more
`or less swollen particles in the system. The system is then
`actually no longer a solution, but a suspension in a polymer
`solution. These problems increase with polysaccharide
`concentration, with the consequence that polysaccharide
`solutions can be prepared in practice only up to rather
`limited concentrations. In many cases, the range of con-
`
`Copyright 2005 by Marcel Dekker
`
`358
`
`
`
`Rheological Behavior of Polysaccharides
`
`359
`
`distant along the chain cannot occupy the same volume
`element at the same time when approaching each other, as a
`consequence of their finite volume. This is the so-called
`‘‘excluded volume interaction.’’ On the other hand, there
`are always attractive and repulsive monomer/monomer
`and monomer/solvent interactions. The balance between
`these interactions will result in coil contraction or coil
`expansion. Whereas backbone rigidity and molecular
`weight are intrinsic characteristics of the polymer chain
`considered, this balance depends on solvent ‘‘quality’’ (the
`chemical nature of the solvent and the temperature). The
`effective ‘‘excluded volume’’ will therefore depend on
`solvent and temperature. In ‘‘good’’ solvent conditions,
`the chain configuration is more expanded, because solva-
`tion of chain segments increases excluded volume; then, its
`2 and can
`average square end-to-end distance will beL 2>Lo
`be written as:
`
`2 ¼ b2N2m; with m
`L
`
`z
`
`0:5
`
`ð1Þ
`
`Parameter m is the exponent of the radius of gyration–
`molecular weight relationship of the coil (Rg~Mm), its
`value depending on chain flexibility: obviously, m=1 for
`fully extended rigid chains, and m=1/3 for compact
`spheres. For polymer coils, m lies of course in between these
`two limiting values. However, it depends not only on the
`more or less flexible character of the backbone, but also on
`polymer–solvent interactions, which govern coil expan-
`sion. Exponent m has the value of 0.5 for Gaussian chains,
`as we have seen; for typical flexible polymers in good
`solvents, it is close to 0.6. In ‘‘bad’’ solvent conditions,
`attraction between chain segments dominates and causes
`coil collapse: the coil contracts, forming a dense particle
`(m!1/3), which eventually precipitates from the solution.
`Somewhere in between these two situations, repulsion
`between chain segments can be exactly compensated for
`by attraction; such solvent conditions are called ‘‘Q con-
`ditions’’ or ‘‘Q solvent.’’ In Q conditions, the polymer coil
`behaves indeed as if it were actually Gaussian and has its
`2=b2N (m=0.5).
`‘‘unperturbed’’ dimensionLo
`It is classical to express coil dimension as:
`
`ð1bÞ
`
`Þo
`
`2 g
`
`
`
`¼ a2gðR
`
`2 g
`
`; or R
`
`2 o
`
`2 ¼ a2
`L
`LL
`
`2, called the expansion factor of the polymer, is equalaL2 or ag
`
`
`to 1 in Q conditions and >1 in good solvents. Because the
`probability that one segment comes to take the place already
`occupied by another increases with N, the expansion factor
`increases with the molecular weight of the polymer. It is a
`complicated function of L, b, m, and of the second virial
`coefficient A2 of the polymer in the solution [1].
`
`B. Solutions of Noncharged Chains at Finite
`Concentrations: The Three Concentration
`Regimes
`
`Three concentration domains can be distinguished in so-
`lutions of polymers with molecular weights above the
`critical value Mc.
`
`centrations over which the rheological behavior can be
`studied is moreover limited on the lower side because of
`relatively low MW, hence low viscosity and low viscoelas-
`ticity that would require high sensitivity instruments to be
`measured. Few systematic rheological studies have been
`carried out on polysaccharide solutions (with the exception
`of cellulose derivatives) for the above reasons, and also
`because of the tedious extraction, purification, and frac-
`tionation operations necessary to obtain polysaccharide
`samples suitable for quantitative characterization.
`
`III. SOLUTIONS OF NONCHARGED CHAINS
`
`A. The Isolated Polymer Coil
`
`A typical polymer molecule is a long, flexible or semiflex-
`ible chain that can adopt in solution any configuration
`compatible with its fixed bond lengths and angles and other
`possible steric restrictions. The set of these spatially and
`temporally fluctuating configurations defines the equilibri-
`um statistical conformation of the polymer coil in the
`solvent. The polymer coil is a swollen structure; the volume
`occupied by the monomers accounts only for a small
`fraction of the total volume pervaded by the coil; monomer
`density decreases from the center of gravity of the coil to its
`periphery. If the volume of the chain itself is neglected, and
`if sterical and physical chemical interactions between
`monomers or between monomers and solvent molecules
`are absent, the spatial distribution of the monomers (or of
`the chain segments) within the coil would be Gaussian
`(‘‘random coil’’). The volume of the Gaussian coil depends
`only on the length (or the molecular weight M) of the chain,
`and on the size bo of the monomer. In real polymer
`molecules, bond angles are fixed and there are steric and
`energetic restrictions to rotations, resulting in a less-flexible
`chain (nonfreely jointed chain). Nevertheless, polymer
`molecules can still be treated as an equivalent Gaussian
`chain by replacing as statistical units the monomers by
`segments comprising n monomers; the rigidity of the
`backbone will be reflected in the length b=nbo of the
`statistical elements the articulation of which forms the
`chain. This length b (twice the ‘‘persistence length’’ Lp)
`can be directly determined from small angle neutron or
`small-angle X-ray scattering experiments. An appropriate
`chain stiffness parameter is the ratio of the contour length L
`of the chain to the length b of the statistical element: for a
`given polymer, chain flexibility increases with the molecu-
`lar weight. A ratio L/b > 10 would be required for the
`polymer conformation to be regarded as a coil; this corre-
`sponds to M higher than some limiting value Mc. Most
`polysaccharides are relatively stiff chains, with statistical
`f3 104.
`element length of f10 nm and Mc
`ffiffiffiffiffiffi
`N, being the number of statistical elements of the
`chain, the average square end-to-end distance of the equiv-
`¼ b2N ¼ kM: The diameter of
`p
`alent Gaussian chain is L
`
`
`1/2, its radius of gyration (Rg2)o
`the coil will be equal to b
`N
`2/6; the average monomer (or segment)such as (R g2)o=L o
`
`
`density within the coil will decrease as N 1/2.
`The coil volume is actually somewhat larger than that
`of the equivalent Gaussian coil, because two segments
`
`2 o
`
`Copyright 2005 by Marcel Dekker
`
`359
`
`
`
`Lefebvre and Doublier
`
`zones exhibit lifetimes much larger than that of entangle-
`ments. The system then has shifted from the state of an
`entangled solution to that of a ‘‘physical gel’’ (cf. Section
`VII). The difference between the characteristics of the two
`types of systems is primarily a difference in degree, and not
`in nature. However, the establishment of such junction
`zones generally involves a change in the conformation of
`the chains, which lose their character of ‘‘random’’ coils.
`(cf. Section VII).
`
`3. Concentrated Regime (c > c**)
`At certain concentrations c**> c *, the coils reach their Q
`dimension; the excluded volume interaction is completely
`screened off. Above c**, coils will shrink no more; the
`polymer solution becomes an entanglement network where
`the chains have completely lost their individual character.
`The only characteristic length in the system is now the mesh
`size f of the network, which continues to decrease as
`concentration increases, tending toward its limit value b
`in the melt.
`
`IV. THE CASE OF POLYELECTROLYTES
`
`A polyelectrolyte is a flexible polymer electrically charged
`because its structure includes monomers bearing ionizable
`groups with charges of the same sign. Many polysaccha-
`rides, such as alginates, low-methoxyl pectins, carra-
`geenans, etc., are anionic polyelectrolytes, negatively
`charged at pH values above the pK of ionization of their
`acid groups. The only commonly found cationic polysac-
`charide is chitosan.
`The distinctive feature of polyelectrolytes is that the
`conformation of the macromolecule depends sharply on
`the ionic strength of the solvent, because the range of the
`electrostatic interaction decreases as ionic concentration
`increases. The Debye screening length j 1~I 1/2, where I
`is the ionic strength, classically measures the range of the
`electrostatic interaction in simple electrolyte solutions. In
`the case of polyelectrolyte solutions, the polymer itself,
`because of its proper charge and of the counterions sur-
`rounding its charged groups, as well as the salt dissolved in
`the solvent, both contribute to electrostatic screening. The
`
`Debye length is now j2 ¼ j2p þ j2s, where the indices p and s
`
`refer to the polymer and to the small ions contributions
`(including the counterions of the polyion), respectively.
`Thus, the conformation depends on both the polymer and
`~fcp, where cp is the concentra-
`the salt concentrations; j2
`p
`tion of the polyelectrolyte and f is the fraction of charged
`monomers in its chain. However, the effective contribution
`of the polymer is generally lower than expected from its
`theoretical charge density, as a result of the binding of
`counterions on the macroion. Ion binding (‘‘condensa-
`tion’’) results from strong attraction of counterions by
`the polyelectrolyte when its charge density is high; the ionic
`atmosphere surrounding the fixed charges can then differ
`widely from that of the Debye–Hu¨ ckel approximation.
`Obviously, electrostatic repulsion between the charged
`segments of the polyelectrolyte will favor more expanded
`conformations of the chain than if excluded volume effect
`were the only long-range interaction existing between chain
`
`360
`
`1. Dilute Regime (c < c*)
`
`In a very dilute solution, the volume available to each
`polymer molecule is much higher than that of the individ-
`ual coil. The coils remain statistically far from each other,
`and encounters are infrequent. The coils maintain the
`dimensions of the isolated chain. This situation prevails
`up to the critical overlap concentration c*, at which the
`coils fill the volume of the solution. The volume of each coil
`
`Þ3=2 ¼ ð1=6Þ3=2ðL2Þ3=2,
`being proportional to ðR
`
`c ~M=ðL2Þ3=2~N1 3m
`
`ð2Þ
`
`2 g
`
`2. Semidilute Regime (c* < c < c**)
`
`When polymer concentration is increased above c*, there is
`a progressive interpenetration of the coils, concomitant
`with a contraction of their individual volume. Coil con-
`traction is a result of the progressive screening off of the
`‘‘excluded volume interaction’’: as a result of interpenetra-
`tion, segments of ‘‘foreign’’ chains interpose themselves
`between segments belonging to the same chain. The solu-
`tion becomes a transient network of entangled chains, an
`entanglement being the topological constraint correspond-
`ing to a point of contact between two chains, and being due
`to the fact that the chains cannot cross each other. A given
`polymer with M>Mc statistically contracts a determined
`number of entanglements at a given concentration c>c*,
`but, because of chain conformation fluctuations, these
`entanglements continuously unfasten to be reformed on
`other points along the chain contour; their lifetime is very
`short. If M<Mc, the length of the chain is shorter than the
`minimum distance required between entanglement points.
`An alternative view [2] is to consider the system as equiv-
`alent to a cage between the rails of which one given chain
`has to reptate along its own contour in order to move away;
`the chain is confined within a tube, the diameter of which is
`the mesh size of the temporary network.
`In the semidilute domain, the mesh size f of the network
`(average distance between entanglement coupling) and the
`size of the coil decrease as concentration increases. f2
`measures the mean square end-to-end distance of the chain
`segment comprised between two successive entanglement
`points along one chain. The chain segment thus defined,
`made up of g statistical elements, constitutes a ‘‘blob,’’ and a
`chain of N/g blobs forms the polymer molecule. The
`excluded volume interaction between two adjacent blobs
`along the chain being screened off by an interposed foreign
`chain, the chain of blobs is Gaussian, so that the mean
`square end-to-end distance a polymer chain can be written:
`2 ¼ f2N=g. Within the blob, on the contrary, the excluded
`L
`volume interaction is effective and f2=b2g2m. Scaling laws
`give [3]: f~cm/1 3m andL
`2~c2m 1=1 3m.
`In the semidilute regime, there are two characteristic
`lengths in the system: the size of the coil—as the coils still
`retain some degree of individuality—and the entanglement
`spacing (or size of the blob).
`Looking at the system as a network can be readily
`extended to the case where low-energy physical chemical
`interactions develop between chains in the regions of
`entanglements, giving rise to junction zones. Junction
`
`Copyright 2005 by Marcel Dekker
`
`360
`
`
`
`Rheological Behavior of Polysaccharides
`
`361
`
`segments. Thus, the ionized polyelectrolyte will exhibit
`larger mean-square end-to-end length and radius of gyra-
`tion values than the uncharged chain would have in good
`solvent. This can be accounted for by introducing an
`electrostatic contribution to the persistence length of the
`chain (see, for example, Ref. [4]). In very dilute salt-free
`solutions, the macromolecule tends to adopt an extended
`rodlike conformation in order to minimize the electrostatic
`2cL~
`contribution to the free energy of the chain; then,L
`N. As I increases, electrostatic interaction between charged
`segments is progressively screened off. At moderate values
`of I the chain conformation resumes a spherical symmetry,
`but with a larger radius of gyration than the equivalent
`uncharged chain in the same solvent, and at higher values
`of I, the dimension of the coil approaches that of the
`uncharged chain.
`Even a very schematic analysis of the effect of cou-
`lombic repulsion of ionized groups on polymer conforma-
`tion would be beyond the scope of this chapter. The reader
`can refer to Refs. [5,6] for a thorough theoretical treatment.
`Many points about the behavior of polyelectrolytes remain
`indeed unclear. Therefore, we shall just point out here a few
`remarks of practical importance.
`Since its conformation strongly depends on its own
`concentration c, as well as on salt concentration cs, the
`phase diagram of a polyelectrolyte is more complex than
`that of neutral polymers. An example of such a diagram is
`shown on Fig. 1. Because the macromolecule is highly
`expanded, the coil overlap concentration c* is extremely
`low at low and intermediate salt concentrations. Below c*,
`the situation is in fact complex. At low salt concentra-
`tions, the polyelectrolyte is in its extended conformation
`(DR regime of Fig. 1), far different from that of a neutral
`polymer, and furthermore, there are strong intermolecular
`interactions. However, for large enough salt concentra-
`
`Figure 1 Theoretical phase diagram for an aqueous poly-
`electrolyte solution (molar concentration c) in presence of
`added salt (molar concentration cs). DR: dilute rodlike regime.
`DF: dilute regime of flexible conformation. SF: semi-dilute
`regime. The diagram has been calculated for a chain with
`N=3350 monomers, an average number of monomers between
`charges A=5, and Nb/A=3, where A is the actual extended
`length of the chain. Reproduced from Dobrynin et al. [5].
`
`tion, the chain becomes flexible at all polyelectrolyte
`concentrations (DF regime). The conformation is then
`analogous to that of the neutral chain in good solvent, but
`the expansion factor is controlled by the electrostatic
`screening length, which depends on both c and cs. Above
`c* (SF regime), the chain is likewise flexible, but now the
`classical excluded volume interaction screening effect of
`neutral polymers combines with the electrostatic screening
`effect in governing the expansion factor. Whereas both
`effects depend on c, the contribution of the latter
`decreases as cs increases, and at high enough salt concen-
`trations the situation becomes similar to that for the
`neutral chain. Because at low and intermediate ionic
`strength values, polyelectrolyte dimension is strongly
`dependent on its concentration, the semidilute regime is
`very wide;
`it can extend over three or four polymer
`concentration decades.
`The extremely low values of c* and the existence of
`strong intermolecular interactions for c < c* make the
`experimental characterization of the isolated macromole-
`cule difficult at low and intermediate cs values. At large salt
`concentrations, experimental problems also arise, because
`the added salts affect solvent quality, independently of their
`electrostatic screening effect; actually, many flexible neu-
`tral water-soluble polymers approach unperturbed (Q)
`dimensions as salt concentration increases; they eventually
`even precipitate (‘‘salting out’’). The density of charges
`along the chain, i.e., the relative number of monomers
`bearing ionizable groups and the degree of ionization are
`the primary intrinsic characteristics of the chain affecting
`polyelectrolyte expansion. Theories take into account an
`average value for charge density; they introduce, e.g., an
`average number of monomers between charges. But the
`distribution of the charged groups along the chain also
`plays a role: the repulsion between a pair of charges is likely
`to have a larger effect on the overall conformation when the
`charges are distant along the chain than when they are
`adjacent. In the case of polysaccharides, charge distribu-
`tion is neither even nor random along the chain, and is
`generally completely unknown; it is susceptible to consid-
`erable variation for a given polysaccharide with a given
`average charge density.
`
`V. FLOW BEHAVIOR OF POLYSACCHARIDE
`SOLUTIONS
`
`A. Origin of Rheological Properties of
`Polymer Solutions
`
`In the dilute regime, Newtonian flow behavior and absence
`of viscoelasticity are generally observable in practical
`conditions, at least for noncharged polymers,* because
`statistically, macromolecules are spatially and temporally
`noncorrelated. Nevertheless, in principle, polymer coils are
`able to deform when submitted to velocity gradients, and to
`recover their equilibrium conformation after cessation of
`
`* As we shall see later, this is not the case for dilute polyelectrolyte
`solutions in low added salt conditions, because of long-range
`electrostatic interactions.
`
`Copyright 2005 by Marcel Dekker
`
`361
`
`
`
`362
`
`Lefebvre and Doublier
`
`great theoretical interest, it has little practical importance
`and the topic will not be considered here. In this section,
`viscosity and intrinsic viscosity will refer implicitly to
`measurements performed within the Newtonian domain,
`i.e., at shear rates low enough for the dilute solutions to
`display shear rate-independent viscosity. The condition of
`Newtonian behavior, in most cases, easy to achieve for di-
`luted polymer solutions, can be in some others (when
`working on polymers with very high molecular weight, or
`with rather stiff chains, and on polyelectrolytes at low salt con-
`centrations) more difficult to accomplish experimentally.
`
`1. The Intrinsic Viscosity of Noncharged Polymers
`
`the mechanical excitation, hence intrinsic non-Newtonian
`and viscoelastic properties. However, these properties are
`very faint and are exhibited in practice only at fairly high
`deformation rates. They are, in consequence, of little direct
`concern for applications, as far as polysaccharide dilute so-
`lutions are considered. However, internal coil deformation
`modes are responsible for the short-time or high-frequency
`viscoelastic response of polymer nondilute solutions and
`gels. On the other hand, the characteristics of the polymer
`that are reflected in the viscosity of its dilute solutions
`govern, to a large extent, its viscous behavior in nondilute
`solutions. In consequence, no characterization of the semi-
`dilute and concentrated regimes of polysaccharides, which
`are the relevant issues regarding applications, is possible
`without the knowledge of the behavior of the polysacchar-
`ides in dilute solutions. This is why dilute solution proper-
`ties have to be given some development.
`In practice, non-Newtonian flow behavior and visco-
`elasticity of polymer solutions develop for c > c*. In the
`case of flexible or semiflexible chains, these properties
`originate from the disentanglement/re-entanglement pro-
`cesses resulting from the opposite actions of flow and of
`thermal agitation. They are manifestations of the transient
`network structure. Dispersions of rigid macromolecules or
`particles exhibit similar properties above some critical
`concentration; but they are mainly due, in this case, to
`the spontaneous establishment of a local order at rest, as a
`consequence of crowding; strain perturbs this order and
`thermal agitation tends to restore it. When macromolecules
`or particles are asymmetrical, orientation in the direction
`of flow plays, in addition, an important role (liquid crystal
`structures). For solutions of flexible chains as well as for
`dispersions of rigid particles, non-Newtonian flow and
`viscoelastic properties are therefore of entropic origin.
`
`B. Viscosity of Dilute Solutions
`
`Only a very schematic and short account will be provided
`here on viscosity of dilute solutions of polymers in general,
`and of polysaccharides in particular. The matter, which has
`direct bearing on macromolecular conformation studies,
`has been the subject of an enormous amount of theoretical
`and of experimental work, and books or extensive reviews
`are available on it (for example, Refs. [7–9]). The main
`information that can be extracted from viscosity measure-
`ments on dilute macromolecular solutions is contained in
`the intrinsic viscosity of the macromolecule. Intrinsic vis-
`cosity is not a viscosity at all, but actually a measure of the
`hydrodynamic volume of the coil in the case of noncharged
`polymeric chains, or of the asymmetry of the particle in the
`case of rigid macromolecules. The case of polyelectrolytes
`is specific, and remains poorly understood.
`As already alluded to, for most polyme