Polymerdynamics of cellulose and other polysaccharides in solid state-secondary dielectric relaxation processes

Abstract

Dielectric relaxation spectroscopy (DRS) separates different molecular groups of a repeating unit of a polymer with respect to the rate of its orientational dynamics. In the case of dry solid polysaccharides, four modes of relaxation processes can be observed in the sub-T_g range, which we interpret in the following way. The local main chain motion forms the β-relaxation, and the side groups motion in the repeating unit generates the γ-relaxation. Additionally, the so-called δ-relaxation can be observed in the low frequency side of the β-relaxation for well dried samples and a further β_wet-relaxation occurs only in wet samples in the room temperature range, but the origins of this last process are not clear up to now. In the high temperature range (T>80°C), the σ-relaxation can be measured which is associated with the hopping motion of ions in the disordered structure of the biopolymeric material. For all these processes, we give experimental evidence. In addition, further relaxation processes are detected in the electrical inhomogeneous polysaccharide samples, which are associated with internal interfaces and the interface to the electrode and are well known as Maxwell–Wagner–Sillars and the electrode polarisation. The influence of the type of the glucosidic linkage to the β-relaxation is discussed by comparing the dynamic dielectric behaviour of different polysaccharides. Small amounts of water or other swelling solvents in the sample modify the relaxation processes in a characteristic manner and increase the activation energy and the cooperativity of the local chain motion. The morphological structure of the cellulose affects the dielectric spectra in the low frequency range below the β-loss peak. This spectral range in the DR spectra correlates with the chemical accessibility and the water retention capacity of chemical pulps. In the case of derivatives of cellulose or starch, we can show that the relaxation of side groups can be separated depending on the type of the side group and its position in the anhydro-glucose unit (AGU). Results are presented both in the form of dielectric spectra and as fit parameters calculated with the help of the Havriliak–Negami function and also in the form of the activation energies and the pre-exponential factors resulting from the Arrhenius representation. Essential literature concerning relaxation processes in polysaccharides is reviewed and the results given are compared with our findings.

Introduction

Dielectric relaxation time spectroscopy (DRS) is a well-established tool in the investigation of synthetic polymer material. Many review articles [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32] deal with this subject, presenting theoretical and experimental aspects and summarise many experimental data of this substance class. However, in all these reviews, the biopolymers only play a subordinate role. The books of Pething [17] or Takashima [19] on the (di-)electric properties of biopolymers concentrate on proteins and DNAs, whereas the chapters on polysaccharides are very small and their content is not representative. On the other hand, there are spare discussions about the dielectric and dynamic properties in modern books and review articles about cellulose or polysaccharides [33], [34], [35], [36], [37], [38], [39], [40], [41], [42]. In general, the dielectric spectroscopy of polysaccharides has been considered controversial by many scientists, up to now.

Cellulose and other polysaccharides, which are the subject of this review, consist of anhydroglucose units (AGU) (Fig. 1) carrying two hydroxyl groups (-OH) and one methylol group (-CH₂–OH).

The AGUs are linked in the polymer chain via acetal oxygens in the equatorial- (β-form) or in the axial directions (α-form). This O-bridge forms the so-called glucosidic linkage.

The dielectric relaxation studies of cellulose, which date back to the measurements made by von Schweidler [43] and Wagner [1], [162], show a marked progress in the last 15 years [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70], [71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82], [83], [84], [85], [86], [87], [88], [89], [90], [91], [92], [93], [94], [95], [96], [97], [98], [99], [100], [101], [102], [103], [104], [105], [106], [107], [108], [109], [110], [111], [112], [113], [114], [115], [116] caused by the commercial availability of dielectric broadband spectrometers [117] in the frequency range of 1 mHz–1 GHz [29] and, additionally, by the new research activities in the field of polysaccharide chemistry and physics, in general.

Cellulose and other polysaccharides are technically very interesting, renewable, biocompatible and owning excellent biodegradable resources and versatile chemical and physical properties.

The technical application and technological processes in the cellulose industry are not without problems, which are related to the complex structure of these types of biopolymers. For an extended commercial application and a better understanding of these biopolymers, the investigation of their dynamics at the molecular level is a useful new contribution to complete structural and energetic investigations.

DRS is a method which can investigate molecular motions in the extended time scale of 0.1 ns–100 s if the moving sites of the repeating unit and the attached side groups own a permanent dipolar moment.

In each case, the morphology of cellulose and other polysaccharides strongly depends on the preparation conditions of the sample before the measurements and on the kind of chemical functionalisation [118], [119], [120], [121], [122]. The properties of polysaccharides, on the one hand, are very sensitive to low water content and its distribution in the sample and, on the other hand, to the complex supramolecular structure of polysaccharides (often characterised as secondary, tertiary or quaternary structure). This structure is characterised, for instance, in the case of cellulose by elementary fibrils, which associate by hydrogen bonds to micro- and macrofibrils forming a morphology with a specific system of holes, pores, micro crevices and capillaries.

These hydrogen bonds are a dominant aspect of the structure of cellulose and also the other polysaccharides. Because of the polarity of the hydroxyl groups on the cellulose chain, strong hydrogen bonds are not only found inside the cellulose chain, but also between chains and between larger agglomerates (see Fig. 2).

Hydrogen bonds are abundant in the native structure of cellulose and some are formed additionally in pulps when water is removed from the fibres and from the interfaces between the cellulose chains. The formation of some bonds is irreversible, of other ones it is reversible. The physical result reveals itself in a densification of the micro- and macrostructure, an irreversible closing of pores, lower re-swelling ability and changes in the properties (mechanical, chemical and also electrical) [41]. The accessibility of cellulose for a reactant or a solvent molecule is strongly dependent on the preparation conditions of the pulp and especially on the drying procedure of the pulps [123], [124]. The irreversible structural changes in swollen, microporous pulp after drying are also designated in an unspecific way as ‘hornification’. The details of this phenomenon are not yet elucidated. It is known, however, that the mechanisms of collapse of pores, slight recrystallisation of domains adjacent to micro-crystallites and the formation of additional H bonds within the less ordered regions will contribute to it. Some authors have already investigated this effect of hornification and its influence on the chemical accessibility and reactivity of pulps [122], [123], [124], [125]. All these processes also have an effect on the dielectric properties of these materials.

In the case of starch, which is a blend of amylose and amylopectine with a helical polymer chain structure, the same dominance of supermolecular structure effects can be observed. Starch is characterised by a granular supramolecular structure depending on the origin and the treatment of the sample. The granules are formed by layers and within them the amylopectine forms a branched polymer with a very high molecular weight and the shorter amylose molecules are embedded in the amylopectine matrix forming crystalline or amorphous regions. In starches, a similar effect as in cellulose can be observed. The starch is restructured irreversibly in comparison with the native starch by drying and regeneration. These complex structure elements (different types of fibrils or granules, crystalline or amorphous regions in a fibre forming noncrystalline spheres in which also a preorientation exists and no complete disorder) exist in different length scales and are typical for each special polysaccharide. This structure forming complicates the interpretation of dielectric spectra in a significant manner. In this context, the description of polysaccharides by a two phase model with a crystalline and an amorphous phase is an inadequate simplification to interpret the results of dynamic measurements. That also means, polysaccharide samples in most cases are not homogeneous dielectric solids and their characterisation by electrical material parameters such as the dielectric function or the electrical conductivity is a fundamental problem.

Considering all these aspects, the preparation of the samples for dielectric spectrometric measurements is an important point for the successful application of this analytical technique. Many controversial results and discussions in literature can be put down to the problem of the insufficient physico-chemical characterisation of the polysaccharides tested.

Therefore, we hope to show in this review, that, today, practical DRS of this class of biopolymers has reached such a level that it can work as new and useful analytical method to investigate different problems in present polysaccharide research.

This paper is structured in the following way. Firstly, we give a short overview of a few experimental and preparatory aspects of practical DRS. Next, we present a short introduction into the phenomenological theory and the description of the dielectric spectra with the help of well-tried model functions and relaxation parameters. Furthermore, we create the prerequisite for the evaluation of dielectric spectra. After that, we summarise and discuss the basic molecular models for the interpretation of the relaxation modes measured in dielectric spectroscopy of polysaccharides and present new experimental proofs of the assignment of the different dielectric loss processes to molecular motions.

The main part of this review focuses on experimental data confined to the sub-T_g region and their interpretation. The so-called primary or α-relaxation, which is associated to the glass transition temperature (T_g), should not be discussed within this review, because in all our experiments we have not found any evidence for a dynamics with a Vogel–Fulcher temperature dependence which is typical for the glass-transition dynamics [126], [127], [128]. We subdivide this main chapter into the following classes of substances: (1) celluloses in native form and regenerative cellulose and pulps; (2) other pure polysaccharides in comparison with cellulose; (3) water influence on the dielectric properties of cellulose; (4) cellulose derivatives; and (5) starch derivatives.

Concluding remarks give an outlook for investigations under preparation and to specific basic problems which have to be solved in the future to use all possibilities of this analytical tool for polysaccharide research.