**3.1.1 Cellulose water/moisture interactions**

One of the most important features of cellulosic substrates is their propensity to absorb moisture from ambient air, expressed in terms of either moisture regain or moisture content. Water absorption causes swelling of the substrate, which alters the dimensions of the fibre, and this, in turn, will cause both changes in physical properties such as the size, shape, stiffness, and permeability of yarns and fabrics,(Morton & Hearle, 1993) as well as sorption/desorption characteristics (Široká et al., 2008), and in mechanical properties such as tensile modulus and breaking stress,(Kongdee et al., 2004) and, therefore, interaction between cellulose and water plays an important part in the chemistry, physics and technology of cellulose isolation and processing.

Cellulose accessibility largely depends on the available inner surface, supramolecular order, fibrillar architecture, and also fibre pore structure. In most cases, there is interaction with water which consequently destroys weak hydrogen-bonds, but cannot penetrate into the region of high order, in contrast to, for example, aqueous solutions of sodium hydroxide, and therefore cellulose is not dissolvable in water. In Fig. 8, an overview is given about the depth of fibre reorganization/dissolution and also indicates the fields of its application; sorbed water on such a polymer can cause structural changes predominantly in the amorphous or intermediate phases. However, there is also a significant role associated with pores, capillaries, and the network of voids, which do not have uniform size and shape.

Fig. 8. Different stages of cellulose fibre reorganisation at different level: a) solvent, b) yarn, and c) fabric level.

From the chemistry perspective, cellulose-water interactions are basically a competition of hydrogen-bond formation between hydroxyl groups in the polymer and hydrogen-bond formation between one hydroxyl group of a cellulose chain and a water molecule or a water cluster.(Klemm et al., 1998) Differential scanning calorimetry (DSC) may be used as valuable technique to explore the interaction of either water or moisture with various natural and synthetic polymers with hydrophilic groups, and thermal properties of polymers and water.

Alkali Treatments of Woven Lyocell Fabrics 189

surface and the liquid properties are determining factors of fibre wetting properties. The geometric configurations and the pore structure of the porous medium (*e.g.* woven fabric) determine its liquid transport properties and hence, capillary water differs in various types of fabric construction (plain-, twill- or sateen-woven fabrics). Fabric geometry and fabric construction influence properties of macro-pores (pore size, pore size distribution, pore connectivity, total pore volume).(Hsieh et al., 1992) Pores in woven fabrics can be intra-fibre, inter-fibre and inter-yarn;(Hsieh, 1995) intra-fibre pores are smallest and are discontinuous, not being merged with neighboring fibres. Structure and dimensions of inter-fibre and interyarn pores are influenced by yarn structure and yarn density in woven fabrics and thus, the

When cellulose fibres are dried there is a subsequent reduction in the extent of swelling when the fibres are re-wet; the loss of swelling resulting from a drying and re-wetting cycle is so-called "*hornification*" or "*zip up of voids*".(Crawshaw & Cameron, 2000) Basically, some voids "*zip up*" after drying of swollen substrate by the formation of additional hydrogenbonds between cellulose molecules, and this may lead to re-crystallization or to spot-wise bridging, which produces non-crystalline, brittle, inaccessible domains on a larger scale. According to this model, during wetting, cellulose chains are hydrated with water molecules, re-arrangement of cellulose chains takes place due to opening of closed pores and joining up of small pores. When drying, the water molecules are partially removed and hydrogen-bonds between cellulose molecules are generated. This interaction induces rearrangement of fibre structure, loss of fibre pores and collapse of large pores, which is not recovered by re-wetting. Furthermore, amorphous cellulose may become re-crystalline on drying. This change results in lowered pore volume and surface area that is especially found in non-crosslinked type fibres. The limited re-expansions can also be attributed to covalent interactions, such as the formation of lactone bridges between hydroxyl and carboxylic groups in the polymer chains (Diniz et al., 2004). There is also some evidence of dryinginduced crystallization in cellulosics, to which is also attributed the phenomenon of hornification (Brancato, 2008). Accessibility and reactivity of cellulosic fibres are reduced due to the already mentioned *hornification* phenomena. In accordance with Crawshaw et al. (2000) findings on drying, the void fraction falls and the void size rises, suggesting that a large number of voids "*zip up*", essentially, some voids merge and creates larger ones. These authors proposed a schematic illustration of water-swollen and dried Tencel fibres. Another widely accepted explanation is so-called fibrillar 'fusion', the formation of hydrogen bonds between surface hydroxyls, which causes the reduction of interfibrillar interstices and hence, the accessible and reactive internal surface is decreased.(Krässig, 1993b) This fusion is enhanced the lower the crystalline order and the higher the degree of swelling prior to drying of cellulose substrate from the water wet state.(Krässig, 1993b) The same reference states that the effect of hornification can be avoided by appropriate swelling treatment (*e.g.* ammonia pre-treatment) or by drying after solvent exchange to more hydrophilic liquids.

An additional effect can be expected due to differences in cross-section type of fibres used. These differences will influence pick-up, sorption properties, and fibre swelling in respect to structural differences and also macroscopic properties like stiffness and handle. Usually, fibres of different cross-section are used in specific applications, *e.g.* hygiene and personal care or technical applications and thus, systematic scientific comparisons on fabric level are

rare.

observed differences are mainly determined by fabric construction.

Tatsuko and Hyoe Hatakeyama's classified different fractions of water in cellulose structures.(Hatakeyama & Hatakeyama, 1998) The first-order phase transition of water fractions closely associated with the polymer matrix is usually impossible to observe; such fractions are termed *non-freezing water* and it can directly interact with hydroxyl groups of cellulose. Another fraction of water, *freezing bound water*, is less closely associated and in contrast, it exhibits melting/crystallisation; it is able to interact with either hydroxyl groups of cellulose or water already absorbed to cellulose. The sum of the non-freezing and freezing bound water fraction is *the bound water content*. Water, whose melting/crystallisation temperature and enthalpy are not significantly different from those of normal (bulk) water, is designated as *freezing water*. *Bound water* in the water-insoluble hydrophilic polymers, such as cellulose, lignin, *etc.*, breaks hydrogen-bonding between the hydroxyl groups of the polymers, and its content depends on the chemical and supramolecular structure of each polymer. Schematic illustration of all kinds of water within the fibre structure is proposed in Fig. 9, where from it is obvious that the freezing (Wf) and non-freezing (Wnf) water provides a dry touch and dry perception respectively. However, with further water absorption within fibre structure, the bulk (Wb) and capillary water (Wc) provides a wet touch and hence, it imparts wet perception respectively.

Fig. 9. Water in cellulosic fibre.

Transport of liquid or moisture within the porous solid such as textile fibre or fabric occurs by external forces or capillary forces only. If a liquid is spontaneously transported and it is driven inside porous solid, it is known as wicking. The capillary force increases as the capillaries; gaps between the individual fibres, become thinner – the finer the fibres, the smaller the gaps are, and the better the transport. Wicking is a result of spontaneous wetting in a capillary system due to fact that the capillary forces are caused by wetting. Wetting of fibres or fabric usually involves three phases which may be characterized the best as the displacement from a surface of a fibre-air interface with a fibre-liquid interface. It needs to be noted that the wicking and wetting are not two different processes, but wetting is a prerequisite for wicking.(Kissa, 1996; Rosen, 2004)

Surface wettability (wetting characteristics) of any fabric containing a single fibre type is identical to the wettability of its constituent single fibres.(Hsieh, 1995) Therefore, the fibre

Tatsuko and Hyoe Hatakeyama's classified different fractions of water in cellulose structures.(Hatakeyama & Hatakeyama, 1998) The first-order phase transition of water fractions closely associated with the polymer matrix is usually impossible to observe; such fractions are termed *non-freezing water* and it can directly interact with hydroxyl groups of cellulose. Another fraction of water, *freezing bound water*, is less closely associated and in contrast, it exhibits melting/crystallisation; it is able to interact with either hydroxyl groups of cellulose or water already absorbed to cellulose. The sum of the non-freezing and freezing bound water fraction is *the bound water content*. Water, whose melting/crystallisation temperature and enthalpy are not significantly different from those of normal (bulk) water, is designated as *freezing water*. *Bound water* in the water-insoluble hydrophilic polymers, such as cellulose, lignin, *etc.*, breaks hydrogen-bonding between the hydroxyl groups of the polymers, and its content depends on the chemical and supramolecular structure of each polymer. Schematic illustration of all kinds of water within the fibre structure is proposed in Fig. 9, where from it is obvious that the freezing (Wf) and non-freezing (Wnf) water provides a dry touch and dry perception respectively. However, with further water absorption within fibre structure, the bulk (Wb) and capillary water (Wc) provides a wet touch and hence, it

Transport of liquid or moisture within the porous solid such as textile fibre or fabric occurs by external forces or capillary forces only. If a liquid is spontaneously transported and it is driven inside porous solid, it is known as wicking. The capillary force increases as the capillaries; gaps between the individual fibres, become thinner – the finer the fibres, the smaller the gaps are, and the better the transport. Wicking is a result of spontaneous wetting in a capillary system due to fact that the capillary forces are caused by wetting. Wetting of fibres or fabric usually involves three phases which may be characterized the best as the displacement from a surface of a fibre-air interface with a fibre-liquid interface. It needs to be noted that the wicking and wetting are not two different processes, but wetting is a

Surface wettability (wetting characteristics) of any fabric containing a single fibre type is identical to the wettability of its constituent single fibres.(Hsieh, 1995) Therefore, the fibre

imparts wet perception respectively.

Fig. 9. Water in cellulosic fibre.

prerequisite for wicking.(Kissa, 1996; Rosen, 2004)

surface and the liquid properties are determining factors of fibre wetting properties. The geometric configurations and the pore structure of the porous medium (*e.g.* woven fabric) determine its liquid transport properties and hence, capillary water differs in various types of fabric construction (plain-, twill- or sateen-woven fabrics). Fabric geometry and fabric construction influence properties of macro-pores (pore size, pore size distribution, pore connectivity, total pore volume).(Hsieh et al., 1992) Pores in woven fabrics can be intra-fibre, inter-fibre and inter-yarn;(Hsieh, 1995) intra-fibre pores are smallest and are discontinuous, not being merged with neighboring fibres. Structure and dimensions of inter-fibre and interyarn pores are influenced by yarn structure and yarn density in woven fabrics and thus, the observed differences are mainly determined by fabric construction.

When cellulose fibres are dried there is a subsequent reduction in the extent of swelling when the fibres are re-wet; the loss of swelling resulting from a drying and re-wetting cycle is so-called "*hornification*" or "*zip up of voids*".(Crawshaw & Cameron, 2000) Basically, some voids "*zip up*" after drying of swollen substrate by the formation of additional hydrogenbonds between cellulose molecules, and this may lead to re-crystallization or to spot-wise bridging, which produces non-crystalline, brittle, inaccessible domains on a larger scale. According to this model, during wetting, cellulose chains are hydrated with water molecules, re-arrangement of cellulose chains takes place due to opening of closed pores and joining up of small pores. When drying, the water molecules are partially removed and hydrogen-bonds between cellulose molecules are generated. This interaction induces rearrangement of fibre structure, loss of fibre pores and collapse of large pores, which is not recovered by re-wetting. Furthermore, amorphous cellulose may become re-crystalline on drying. This change results in lowered pore volume and surface area that is especially found in non-crosslinked type fibres. The limited re-expansions can also be attributed to covalent interactions, such as the formation of lactone bridges between hydroxyl and carboxylic groups in the polymer chains (Diniz et al., 2004). There is also some evidence of dryinginduced crystallization in cellulosics, to which is also attributed the phenomenon of hornification (Brancato, 2008). Accessibility and reactivity of cellulosic fibres are reduced due to the already mentioned *hornification* phenomena. In accordance with Crawshaw et al. (2000) findings on drying, the void fraction falls and the void size rises, suggesting that a large number of voids "*zip up*", essentially, some voids merge and creates larger ones. These authors proposed a schematic illustration of water-swollen and dried Tencel fibres. Another widely accepted explanation is so-called fibrillar 'fusion', the formation of hydrogen bonds between surface hydroxyls, which causes the reduction of interfibrillar interstices and hence, the accessible and reactive internal surface is decreased.(Krässig, 1993b) This fusion is enhanced the lower the crystalline order and the higher the degree of swelling prior to drying of cellulose substrate from the water wet state.(Krässig, 1993b) The same reference states that the effect of hornification can be avoided by appropriate swelling treatment (*e.g.* ammonia pre-treatment) or by drying after solvent exchange to more hydrophilic liquids.

An additional effect can be expected due to differences in cross-section type of fibres used. These differences will influence pick-up, sorption properties, and fibre swelling in respect to structural differences and also macroscopic properties like stiffness and handle. Usually, fibres of different cross-section are used in specific applications, *e.g.* hygiene and personal care or technical applications and thus, systematic scientific comparisons on fabric level are rare.

Alkali Treatments of Woven Lyocell Fabrics 191

mercerization under tension does not cause such a large increase,(Morton & Hearle, 1993) and generally causes an increase in fibrillar (crystalline) orientation.(Krässig, 1993a) Moreover, cellulose is susceptible to limited degrees of dissolution in alkali solutions, leading to weight loss in substrates. For instance, lyocell fibre mass reduces by 2-10% with increasing NaOH concentration up to 10 mol dm-3.(Jaturapiree, 2007; Zhang et al., 2005)

The ability of cellulose to adsorb alkali hydroxide ions was originally considered to be a chemical process. In 1907, Vieweg observed two ranges of caustic soda concentrations at which the uptake of sodium hydroxide reached a plateau by applying so-called "change-intiter" method;(Krässig, 1993a) the principle of this method was to measure the drop in alkali concentration of a given caustic soda solution by a known weight of immersed cellulose, and, at the time, had been use for a decade as a convincing observation for the occurrence of a defined chemical reaction following stoichiometric rules. Nevertheless, Vieweg's findings (and those of other research groups) were subjected to criticism by Leighton (1916) who introduced the centrifugation method; he mainly criticized the assumption made by Vieweg that no water is taken up by solid phase and he showed that "his method" allowed the separation of the uptake of water and the uptake of sodium hydroxide. Moreover, Leighton's method revealed that there is initially a preferential water uptake, which has a maximum at medium concentrations, whereas the uptake of sodium hydroxide steadily increases over the whole concentration range, except at the lowest concentrations. The balance of the two separate uptakes explains the plateau in Vieweg's curve of apparent

From these observations, it was concluded that the uptake of alkali hydroxides by cellulose is an adsorption phenomena based on the Donnan equilibrium theory,(Donnan, 1924; Neale, 1929) rather than a chemical process. It is assumed that cellulose, behaving as a very weak monobasic acid, forms a sodium salt to an extent increasing with the concentration of the alkali. Excess alkali diffuses into the mechanically separable cellulose phase. It results in an unequal distribution of ions what causes an osmosis or movement of water which distends the cellulose until the osmotic pressure is balanced by the forces arising from the cohesion of the gel and therefore, the cellulose swells. In addition, Neale's theoretical osmotic curves showed that as the temperature falls the maximum osmotic pressure rises, and the maximum occurred at a lower concentration of alkali. One of the difficulties with the absorption theory was based on the fact that uptake by native cellulose substrates is not smooth and reversible as might be expected for an adsorption process.(Krässig, 1993a) Another issue with the earlier investigations and results interpretation is that the effect of the morphology, fine structure, and accessibility of the cellulosic fibre substrate was, in most

cases, ignored and it was assumed that the fibres were equally accessible throughout.

A comprehensive study of parameters, such as *sodium hydroxide concentration* and *temperature*, which influence the formation of various sodium-celluloses was conducted by Sobue et al. (1939) using an X-ray technique. They proposed a phase diagram of ramie cellulose and its various Na-cellulose crystalline complexes. They also revealed a new highly swollen and poorly crystalline allomorph, namely *Na-cellulose Q*, occurring at NaOH concentrations between 7% to 10% (1.75 to 2.50 mol dm-3 NaOH) and temperatures between –10 and 1°C. More recently, it was found that if an adequate pre-treatment of cellulose was applied, within the *Q phase*, total dissolution of cellulose can occur;(Kamide et al., 1984) solutions of 5% (w/v) cellulose could be obtained at a sodium hydroxide concentration of

sodium hydroxide uptake.
