Cellulose and some cellulose derivatives can play vital roles in the enhancement of the performance of absorbent products. Cellulose itself, in the form of cellulosic fibers or nano-fibers, can provide structure, bulk, water-holding capacity, and channeling of fluids over a wide dimensional range. Likewise, cellulose derivatives such as carboxymethylcellulose (CMC) have been widely studied as components in superabsorbent polymer (SAP) formulations. The present review focuses on strategies and mechanisms in which inclusion of cellulose – in its various forms – can enhance either the capacity or the rate of aqueous fluid absorption in various potential applications.
Enhanced Absorbent Products Incorporating Cellulose and Its Derivatives: A Review
Martin A. Hubbe, Ali Ayoub, Jesse S. Daystar, Richard A. Venditti, and Joel J. Pawlak
Cellulose and some cellulose derivatives can play vital roles in the enhancement of the performance of absorbent products. Cellulose itself, in the form of cellulosic fibers or nano-fibers, can provide structure, bulk, water-holding capacity, and channeling of fluids over a wide dimensional range. Likewise, cellulose derivatives such as carboxymethylcellulose (CMC) have been widely studied as components in superabsorbent polymer (SAP) formulations. The present review focuses on strategies and mechanisms in which inclusion of cellulose – in its various forms – can enhance either the capacity or the rate of aqueous fluid absorption in various potential applications.
Keywords: Absorbency; Cellulosic materials; Fluff pulp; Superabsorbents; Hydrogels; Carboxymethylcellulose (CMC); Disposable; Rate of absorption; Life cycle analysis
Contact information: Department of Forest Biomaterials, College of Natural Resources, North Carolina State University, Campus Box 8005, Raleigh, NC 27695-8005
Cellulose-based products have been used for the absorption of water and other aqueous fluids throughout recent history (Buchholz and Peppas 1994; Chatterjee and Gupta 2002). Among such products, paper towels and tissue papers are some of the most widely used absorbents (Gigac and Fiserova 2008; Beuther et al. 2010). Fluff pulps, which are generally based on kraft pulping and optimized for high bulk and absorbency products, are in widespread use, especially for product niches such as disposable diapers (Lund et al. 2012). In recent decades the amount of cellulosic fiber employed in making an average disposable diaper, as well as in related products such as incontinence products and feminine hygiene products, has declined; this trend has been attributed to increased use of superabsorbent polymer (SAP) products (Nagorski 1994; Buchholz and Graham 2005; Lund et al. 2012). Thus, the goal of this review is to highlight published literature regarding enhanced absorbent materials and products that incorporate various forms of cellulose or its derivatives. Main sections are shown in the following table of contents:
Table of Contents
Although SAPs have a much greater ability to absorb fluids in comparison to cellulosic fibers alone, they have not displaced cellulose completely, nor is such a complete displacement expected. Cellulose, in its various forms, can contribute many positive attributes to absorbent products, including enhancement of SAP performance. Also, there is a widespread perspective that cellulose-based products are beneficial in terms of the environment and personal health.
Certain terms will be used repeatedly in this article, e.g. hydrogel and superabsorbent. The term “absorbent” will be defined here by the phrase “capable of imbibing and holding onto aqueous fluids”. “Hydrogel” will be defined as a hydrophilic polymer structure capable of absorbing a lot of water or saline solution. According to Kabiri et al. (2011), an ordinary absorbent hydrogel is able to absorb as much as 10 g of pure water per gram of dry-based material. Hydrogels that can accommodate more water than 10 g/g are called superabsorbents. As will become clear from tabulated information in this review, absorbency levels in the thousands of grams of water per gram of SAP have been frequently reported. Laftah et al. (2011) suggested a possible further distinction between hydrogels and ordinary absorbents such as paper towels: only the hydrogel-based sorbents are able to provide significant resistance to the release of fluid due to squeezing. To paraphrase Lund et al. (2011), hydrogel-based absorbents “swell and restrain” liquids to a remarkable extent. The amount of aqueous fluid held within a gel at equilibrium will be governed by a balance between osmotic pressure, the restraining pressure exerted by the material itself, and any externally applied pressure. The term “osmotic pressure” can be defined as a chemically-induced pressure differential between the inside and the outside of a hydrogel in the presence of a liquid phase. An additional aspect of the definition of a superabsorbent hydrogel is that it needs to be sufficiently crosslinked so that it does not dissolve in the fluid that it is intended to absorb (Zohuriaan-Mehr and Kabiri 2008).
Motivations to Incorporate Cellulose-Based Materials
The most widely used SAP products, to be described in greater detail later in this document, are not readily biodegradable (Dutkiewics 2002; Sannino and Nicolais 2005). Additionally, they are primarily derived from non-renewable resources including petroleum-derived acrylamide (AAm) and acrylic acid (AA) monomers, together with crosslinking agents, most of which can be classified as non-renewable resources (Buchholz and Graham 2005). By contrast, the renewable and biodegradable nature of cellulose is well known (Cooke 1990). Certain derivatives of cellulose such as carboxymethylcellulose (CMC) and hydroxypropyl cellulose (HPC) not only are highly biodegradable, but also, as will be discussed in this article, can form the basis for production of high-performing superabsorbent hydrogels. Various cellulose-based SAPs have been intentionally designed or tested to achieve a suitable degree of biodegradability (Wach et al. 2001; Lionetto et al. 2003; Sahoo et al. 2005; Yoshimura et al. 2005, 2006, 2009; Wang et al. 2008; Sannino et al. 2009; Feng et al. 2010a; Kono and Fujita 2012). Kosemund et al. (2009) emphasized safety and health issues, which are additional factors supporting the use of cellulose-based polymers rather than the use of petroleum-based SAPs.
Costs of materials have been mentioned as a key factor governing the types and compositions of available absorbent products. For instance, Kabiri et al. (2011) reviewed various research studies showing that hybrid superabsorbent structures incorporating montmorillonite “nano-clays” or cellulose-based nanomaterials sometimes achieve superior mechanical properties and can be manufactured at lower costs relative to conventional acrylic-based SAPs. Kono and Fujita (2012) expressed a different view and proposed that a relatively high cost of superabsorbents based on CMC have constituted a key barrier to their widespread use. However, viewed from another perspective, Liang et al. (2009) suggest that the relatively high cost of even the acrylic-based SAPs, in addition to their lack of biodegradability, offers motivation for further development of high-performing cellulose-based SAPs, as well as hybrid products that combine favorable aspects of cellulose and acrylamide-based chemistry.
Since the perspective of the present review article is specialized – focusing on the incorporation of cellulose-based materials in absorbent products – readers are encouraged also to refer to related review articles emphasizing various different aspects of absorbency and absorbents. Information related to absorbent products in general has been amply discussed (Buchholz and Peppas 1994; Chatterjee and Gupta 2002; Buchholz and Graham 2005). Likewise, the somewhat narrower topic of superabsorbents also has been well reviewed (Matushek 2008; Zohuriaan-Mehr and Kabiri 2008; Kabiri et al. 2011; Laftah et al. 2011). The topic of porous hydrogels also has been reviewed (Omidian et al. 2005; Lu et al. 2007).
Regarding the absorbency of products made mainly from cellulosic fibers, Parham and Hergert (1980) provide a review of fluff pulp, including the history of its development. Gupta (1988) reviewed structural factors affecting the performance of nonwoven absorbents. Fink et al. (2001) reviewed the absorbent qualities of certain regenerated cellulose products.
Regarding gel materials, cellulose-based hydrogels have been reviewed by Sannino et al. (2009) and by Chang and Zhang (2011), who emphasized non-superabsorbent materials. Athawale and Lele (2001) reviewed starch-based SAPs. Chitosan, which has a molecular structure that is analogous to that of cellulose, has been reviewed in the context of superabsorbent products (Dutkiewicz 2002). Medical applications of superabsorbent materials have been described and reviewed (Wichterle and Lim 1960; Bourges et al. 2002). Agricultural applications of hydrogels have been reviewed (Kazanskii and Dubrovskii 1992). Theories to account for the capacities and rates of water absorption into hydrogels have also been reviewed (Scallan 1983; Brannon-Peppas and Peppas 1990; Schuchardt and Berg 1991; Wiryana and Berg 1991; Kazanskii and Dubrovskii 1992; Ganji et al. 2010; Zohariaan-Mehr et al. 2010).
The present review article is motivated in part by issues of sustainability, considering factors that may contribute to the minimization of adverse environmental impacts. In principle, resource sustainability is advanced when renewable, bio-based materials are employed in the manufacture of a product; thus the next section of this article will consider a variety of different absorbent products that have been reported to contain cellulose or its derivatives. Product sustainability is often improved when product goals are achieved with a reduction in material requirements. Thus, separate sections of the article will examine chemical and structural effects on the performance of cellulose-based materials that have been used in absorbent products. Aspects related to the composition of the liquid to be absorbed will also be considered. Theoretical aspects will be considered in sections related to absorption capacity and the corresponding rates of absorption. Finally, the life cycle of absorbent products will be considered, asking such questions as “Why can’t a highly absorbent product be designed so that it empties itself upon immersion in hot water, allowing it to be easily laundered and dried”? Alternatively, one might ask, “Will the next generation of renewable and biodegradable superabsorbents based on cellulose eventually compete successfully with and displace acrylamide-based SAPs?”
CLASSIFICATION OF ABSORBENTS INCORPORATING CELLULOSE
Since the same set of tabulated data will need to be referenced throughout the article, key information is given in Table A, which due to its length is placed in the appendix. The main categories are organized by absorbent class, fiber, hydrogel, “other,” fluid update, “key findings”, and a final column for the literature reference in abbreviated form. Within the “fiber” category, separate columns are provided to show the main fibrous components described in a given cited article and any chemical treatments of that fiber. Within the “Hydrogel” category, there are separate columns for the main chemical components of the hydrogel(s) described in a given article, the method of forming the hydrogels, and the type of crosslinking system employed. Under “Fluid uptake” there are two columns corresponding, respectively, to uptake of pure water and uptake of saline solution roughly corresponding to the ionic concentration of bodily fluids. Such concentrations, e.g. 0.9% NaCl, are often studied due to the widespread availability of physiological saline solutions.
Cellulose-based Materials and Aqueous Absorbency
Mechanically pulped wood
Among wood-based materials, one of the least expensive that could be considered for application as an absorbent would be mechanical pulp, a material derived by the compression and shearing of wood chips in the water-filled gap between the rotor and stator of a refiner. Eriksson et al. (1991) showed that such material can absorb approximately an equal mass of water, even without any specialized treatment. However, the same study revealed a key problem associated with such pulp: The rigid nature of the lignin in mechanically pulped wood resists the swelling of such fibers. One way to overcome such rigidity was demonstrated by removing up to 60% of the lignin. Following such treatment the material was able to take up two to four grams of water per dry gram of fiber. Water uptake also was increased if the temperature of the water was raised above about 70 oC, which is above the softening point of lignin under water-saturated conditions (Back and Salmén 1982).
Chemically pulped wood
According to Beuther et al. (2010), typical paper towels and tissue products that are composed primarily of bleached kraft fibers are capable of absorbing between five and ten times their mass of water without dripping. It was noted that the relative void volume fraction in such paper products can be greater than 0.9 in typical cases. Zohuriaan-Mehr and Kabiri (2008) reported the following values for the absorbency of water by lignin-free products: 1.8 g/g for filter paper; 4 g/g for facial tissue; 12 g/g for bleached kraft fluff pulp; and 19 g/g for cotton balls. The absorptive nature of kraft fibers can be attributed in part to the removal of the more hydrophobic lignin component of the fibers. In addition, the pores left behind by the breakdown and dissolution of lignin during pulping and bleaching gives rise to extensive porosity, though some of the pores close irreversibly when the fibers are dried (Stone and Scallan 1966, 1968; Berthold and Salmén 1997).
Despite their substantially greater absorbency in comparison to mechanical pulp fibers, kraft fibers tend to suffer from one key deficiency: Compared to the mechanically pulped fibers, the wet kraft fibers are not as capable of maintaining a bulky structure, especially when pressure is being applied. Lund et al. (2011) showed that such a deficiency can be overcome by crosslinking kraft fibers with a di-epoxide; the treatment makes the fibers remain stiffer when wet, such that greater bulk is achieved in the paper.
Another approach to enhancing the absorptive ability of kraft fibers has been through the use of rewetting agents (Shepherd and Xiao 1999). The surfactants employed significantly reduced the time required to take up a given amount of water. In addition to enhancing fiber wettability, the treatment often increased the size of inter-fiber pores formed by the structure of the paper. Ström and Carlsson (1992) took a different approach, making the kraft fibers more wettable by oxygen plasma treatment; it was found that such treatment could completely overcome the adverse effect of wood extractives, which otherwise tended to decrease the ability of the fibers to take up water.
Schuster et al. (2006) reviewed the water-absorbing properties of regenerated cellulose materials. Lyocell fibers, which are formed from a solution of -cellulose in N-methylmorpholine N-oxide (NMMO), were reported to absorb about 5 g/g water. The absorptive ability was attributed to fine-scale porosity, which was extensively docu-mented by means of scanning electron micrographs. The absorption behavior of regenerated cellulose products has been quantified in various studies (Fink et al. 2001; Cai and Zhang 2006; Schuster et al. 2006). Cellulose regeneration from solvent also can be used as a starting point for preparation of hydrogels (Nishio and Manley 1988; Lim et al. 2001, 2003; Saito et al. 2003; Chen et al. 2008; Chang et al.2010; Lin et al. 2010; Chang and Zhang 2011; Kono and Fujita 2012; Kono et al. 2013). This topic will be addressed in later sections
New insights regarding the wettability of regenerated cellulose fibers were reported by Yamane et al. (2006). These authors showed that the wettability of the resulting fibers was greatly affected by the liquid medium present during regeneration. Also, the wettability could be affected, to some extent, by post-equilibration of cellulose films with either a nonpolar solvent, or a polar liquid such as ammonia or hot glycerol. Such effects were attributed to preferential exposure of different crystal faces of cellulose at the surface. Based on molecular simulation, it was predicted that the equatorial vs. the axial sides of glucopyranose rings in cellulose crystallites should exhibit hydrophilic vs. hydrophobic character, respectively. In agreement with the work cited above, Quddus et al. (2012) carried out molecular simulations of the spreading of pure water onto three crystal planes of cellulose; significant differences in contact angle were predicted based on the analysis.
Aerogels and Foams Prepared from Cellulose
Porous superabsorbent materials, i.e. foams, have emerged as an important category of SAPs due to their faster sorption in comparison to conventional SAPs (Yoshinobu et al. 1992; Kabra et al. 1998; Omidian et al. 2005; Lu et al. 2007; Chang et al. 2009a; Zamani et al. 2010; Salam et al. 2011a). However, even without employing SAP technologies (see later), it has been shown that substantial absorbency can be achieved by suitable preparation of highly porous cellulose or its chemical derivatives. Brodin et al. (2012) showed that TEMPO oxidation of kraft pulp followed by freeze drying yielded porous foams that absorbed 21 to 65 g/g of water. Other authors (Pääkkö et al. 2008; Sehaqui et al. 2010, 2011; Theliander et al. 2011) showed that highly porous aerogels could be formed by freeze-drying of nanofibrillated cellulose.
Chemically derivatized fibers
Slight derivatization or grafting of the surface of cellulose is a known approach to changing the wettability and other attributes of cellulosic fibers. For instance, oxidation of fibers by (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO) has been found to increase the water uptake of kraft pulp (Brodin and Theliander 2012; Brodin et al. 2012). The effect was attributed to the generation of new carboxylate and aldehyde groups.
More profound effects have been achieved by grafting of polymeric chains by various means onto cellulosic fibers. For instance, improved absorption has been achieved by treating cellulose fiber surfaces with acrylic monomers (Lepoutre et al. 1973; Vitta et al. 1986; Deo and Gotmare 1999; Gürdağ et al. 2001; Sahoo et al. 2005; Aliouche et al. 2006; Jain et al.2007; Jain and Ni 2008). Warner and Rezai (1997) showed that such treatments can be expected to change the composition throughout the fiber structure, not just at its surface.
In cases where a similar grafting approach is carried out more extensively, rather than ending up mainly with surface-derivatized fibers, the cellulose material becomes extensively transformed (Fanta et al. 1987). As shown by Goetz et al. (2009, 2011), such an approach can lead to absorbencies in the range of 30 to 200 g/g, especially if a cross-linking agent is employed. Such results show that it is possible to blur the distinction between what is a surface-modified fiber and what is a true hydrogel. Indeed, a unique class of highly absorbent materials has been achieved by basically skipping the conversion of cellulose to CMC, etc., and just grafting cellulose directly, often in combination with crosslinking (Fanta et al. 1987; Yoshinobu et al. 1992; Gürdağ et al. 2001; Farag and Al-Afaleq 2002; Ibrahim and El-Zawawy 2004; Kim and Mun 2009; Yoshimura et al. 2006; Liu et al. 2009, 2011; Feng et al. 2010a; Zhang et al. 2010; Goetz et al. 2011; Shu et al. 2011; Kono and Fujita 2012; Pan and Ragauskas 2012; Wu et al. 2012; Xie et al. 2012; Kono et al. 2013). Thus, hydrogels will be discussed next.
As mentioned in the Introduction, an absorbent hydrogel can be defined by its ability to hold a relatively large amount of water (up to 10 g/g for ordinary absorbent hydrogels, or more in the case of superabsorbents), by its resistance to the release of aqueous fluids, depending on the applied pressure, and by its resistance to substantial dissolution, despite being chemically compatible with the fluid being absorbed. Absorbent hydrogels are most often prepared from hydrophilic polymers of relatively high molecular mass (Sannino et al. 2009; Chang and Zhang 2011). A key step in the formation of any hydrogel is “gelation,” which can be interpreted as the inter-connecting of the macromolecular chains in some manner, such that essentially the whole structure becomes linked together. Many different means of gelation have been demonstrated. For instance, one or more types of polymer chains may be precipitated from a suitable solvent by changing the conditions to reduce the solubility (Nishio and Manley 1988; Tang et al. 2010). In some cases progressive gelation can be induced by subjecting a polymer solution to freezing and thawing (El Salmawi 2007). Alternatively, the polymer can be precipitated from the solution by changing the temperature (Joshi et al. 2008). This gelation can be attributed to coalescence of polymeric chains due to thermal break-down of the organized cage-like structure of water surrounding each segment of a molecular chain as the temperature is raised through a critical range (Joshi et al. 2008). In other cases gelation may develop gradually by means of self-segregation of parts of adjacent chains that have differing hydrophobic or hydrophilic tendency (Sammon et al. 2006; Silva et al. 2008). Such a mechanism was supported by shifts in the infrared absorption spectra at the temperatures corresponding to gelation (Sammon et al. 2006). Further support for the coalescence of polysaccharide chain segments is provided by results of X-ray analysis, which revealed considerable crystalline character in certain hydrogels prepared from hemicellulose mixtures with chitosan (Gabrielli et al. 2000).
More aggressive gelation can be achieved by the addition of a suitable agent to form linkages among the chains. One of the simplest approaches, most suitable for the highly soluble polyacrylates and related copolymers, involves the addition of a well-chosen amount of multivalent cation (Kulicke and Nottelmann 1989; Yoshimura et al. 2009). Weerawarna and Su (2010) patented a system in which a mixture of kraft fiber and CMC was treated with aluminum sulfate to precipitate the CMC onto the fiber surface; the inventors characterized this approach as a form of crosslinking. In the present context, however, addition of multivalent cations, e.g. acidic aluminum compounds, is typically considered undesirable due to a strong suppression of swelling (Kulicke and Nottelmann 1989; Feng et al. 2010a).
Covalent crosslinking has emerged as the preferred means of converting suitable hydrophilic polymers into absorbent hydrogels, and many variants of this approach have shown excellent potential for the preparation of superabsorbents (Kabra et al. 1998; Liu et al. 2002; Sannino et al. 2003; Buchholz and Graham 2005; Chang et al. 2008; Zohuriaan-Mehr and Kabiri 2008; Laftah et al. 2011; Pohjanlehto et al. 2011; Kono and Fujita 2012). Notably, the highest levels of swelling are often detected when the extent of crosslinking is just sufficient to cause phase separation (Wach et al. 2001). Optimization of the crosslinking thus involves a balance between achieving high absorbency while also achieving suitable mechanical strength and minimizing the amount of material that dissolves into the bulk of the liquid when a highly absorbent hydrogel is in use (Wach et al. 2001; Guilherme et al. 2005).
Types of Superabsorbent Hydrogels
To help understand the context of superabsorbent hydrogels that contain cellulosic components, it should be noted that the commercially dominant SAPs do not contain cellulose-based polymers. Acrylic-based and synthetic hydrogels are by far the dominant products manufactured and sold (Askari et al. 1993; Buchholz and Peppas 1994; Matushek 2008; Zohuriaan-Mehr and Kabiri 2008; Laftah et al. 2011). In general terms, acrylic-based SAPs are most often prepared from aqueous solutions containing acrylic acid (AA) and acrylamide (Am) monomers. These are typically crosslinked with N,N’-methylenebisacryamide (MBA) or divinylsulfone (DVS).
Carboxymethylcellulose (CMC)-based hydrogels
Among cellulose-derived materials used in SAPs, CMC holds a lead position (Oppermann 1995; Barbucci et al. 2000; Bin et al. 2000; Fei et al. 2000; Wach et al. 2001; Liu et al. 2002; 2005a; Bao et al. 2012). Crosslinking of the homopolymer can be achieved by gamma irradiation (Bin et al. 2000; Fei et al. 2000; Wach et al. 2001; Liu et al. 2002, 2005a), by chloromethylene iodide (Barbucci et al. 2000), or by polyvalent carboxylic acids such as citric acid (Demitri et al. 2008). Hydrogels formed by mutual gelation or covalent crosslinking of CMC and other water-soluble polymers have been studied (El Salmawi 2007). For instance, superabsorbents have been prepared with combinations of CMC and polyvinyl alcohol (El Salmawi 2007). Alternatively, CMC-based hydrogels have been prepared with the use of highly platy montmorillonite clay (“nanoclay”) to achieve a crosslinking effect (El Salmawi and Ibrahim 2011).
Hydroxyethylcellulose (HEC) and other cellulose-derivative-based hydrogels
In an analogous manner to CMC, other cellulose ethers or esters have been used as a basis for forming hydrogels, either alone or in combination with other polymers (Harsh and Gehrke 1991; Oppermann 1995; Kabra et al. 1998; Bourges et al. 2002; Wach et al. 2002; Lionetto et al. 2003; Demitri et al. 2008; Joshi et al. 2008; Dai and Kadla 2009; Bao et al. 2012). Bao et al. (2012) observed higher absorbency in superabsorbents made from graft copolymers of CMC, in comparison with graft copolymers from hydroxypropyl methyl cellulose, methyl cellulose, and hydroxyethyl cellulose. Like CMC, other derivatives of cellulose and the hydrogels produced from them have been shown to be biodegradable (Lionetto et al. 2003).
Copolymers of CMC by reaction with monomers
Considerable research has been devoted to the development of copolymers in which CMC is reacted with acrylic monomers and various crosslinking agents to form SAPs (Kuwabara and Kubota 1996; Pourjavadi and Mahdavinia 2005; Pourjavadi and Ghasemzadeh 2006; Abd El-Mohdy 2007; Ibrahim et al. 2007; Pourjavadi et al. 2007a-b, 2008, 2009, 2010; Suo et al.2007; Ali et al. 2008; Chen et al. 2008; Yang et al. 2009; El-Din et al. 2010; Feng et al. 2010a; Lin et al. 2010; Wang and Wang 2010; Bao et al. 2011; Eldin et al. 2011; Huang et al. 2011; Liu et al. 2011; Shu et al. 2011; Varaprasad et al. 2011; Wang et al. 2011c). Water absorbency values generally in the range of 40 to 1400 g/g have been reported in such studies, with the highest value reported by Pourjavadi et al. (2008). Copolymers incorporating CMC have also been prepared with polyvinyl alcohol (El Salmawi 2007), hydroxyethyl cellulose (Oppermann 1995; Lionetto et al. 2003; Sannino and Nicolais 2005; Demitri et al. 2008; Dai and Kadla 2009), and with hydroxypropylethyl cellulose (Faroongwarng and Sukonrat 2008).
Cellulose-synthetic graft copolymer hydrogels
Although underivatized cellulose is not nearly as hydrophilic as CMC, several investigators have prepared hydrogels by extensive grafting of other monomers onto cellulose, generally employing suitable organic solvent systems (Yoshinobu et al. 1992; Wu et al. 2008; Chang et al. 2009b; Kim and Mun 2009; Buyanov et al. 2010; Feng and et al. 2010a; Anirudhan et al.2011; Shu and Zhao 2011; Wu et al. 2012). Generally, water absorbency values in the range of 60 to 1800 have been reported for such hydrogels, though Kim and Mun (2009) reported a high value of 2500 g/g. As an alternative approach, Kono and Fujita (2012) showed that it is possible to form a hydrogel from a solution of native cellulose in an ionic liquid, employing a widely used crosslinking agent.
Hydrogels based on cellulose with other natural-based polymers
As one might expect based on environmental concerns, many researchers have investigated the preparation of “all-natural-based” hydrogels incorporating cellulose or its derivatives (Harsh and Gehrke 1991; Barbucci et al. 2000; Bin et al. 2000; Fei et al. 2000; Wach et al. 2001, 2002; Hirsh and Spontak 2002; Liu et al. 2002, 2005a; Rodriguez et al. 2003; Sannino et al.2003, 2004a,b; Sannino and Nicolais 2005; Yoshimura et al. 2006; Demitri et al. 2008; Faroongwang and Sukonrat 2008; Pourjavadi et al. 2008, 2010; Chang et al. 2009, 2010; Dai and Kadla 2009; El Salmawi and Ibrahim 2011; Salam et al. 2011a,b; Wang et al. 2011b; Kono and Fujita 2012; Pan and Ragauskas 2012; Kono et al. 2013). For instance, a hydrogel based on a combination of cellulose and CMC can be prepared from a solution of the ingredients in a NaOH/urea solvent system (Chang et al. 2010).
Major Covalent Crosslinking Agents
Superabsorbent hydrogels that incorporate cellulose-based components, such as CMC, require crosslinking, and many of the crosslinking agents are the same as those used in SAPs not containing any cellulose-based component. Table 1 provides an overview of the most important cross-linking systems that have been employed in studies of superabsorbent hydrogels that have incorporated cellulose-based compounds.
In addition to the agents listed in Table 1, crosslinking also can be achieved by use of dialdehyde cellulose (Han et al. 2010), making it possible to achieve crosslinking with multifunctional amines on lyocel fabric. Swelling was decreased by such treatment, which had utility for imparting crease resistance to clothing.
One type of hydrogel that merits special mention employs pairs of non-reacting polymer networks, i.e. interpenetrating networks. In such systems each network relies on a different crosslinking mechanism, and the resulting hydrogels are formed in-situ in such a way that, at least in principle, the two networks can remain intensively entangled. Several reported examples have exhibited high absorption capacity for water (Myung et al. 2008; Chang et al. 2009b; Chang and Zhang 2011; Varaprasad et al. 2011; Liu et al. 2013).
There is reason to doubt, however, that an interpenetrating structure is essential in achieving the improved performances reported for such systems. None of the cited articles provides a convincing explanation of why interpenetrating networks would be expected to absorb more fluid. Logic would suggest that such complex, entangled structures would result in more resistance to swelling in comparison to having just one network, in which the crosslinking degree and molecular mass had been suitably optimized. It seems likely that the cellulose-based components of such systems may act as “spacers”, partly suppressing the crosslinking within the acrylamide-based component. Further research is also needed to find out whether both networks are truly continuous in such systems.
Table 1. Major Crosslinking Agents Used in Studies of SAP Preparation
A unique type of hydrogel can be formed through the use of pairs of oppositely charged polyelectrolytes via the formation of polyelectrolyte complexes (PECs). At least one of the polyelectrolytes could be a cellulosic material such as carboxymethyl cellulose (CMC) (Feng et al. 2006; Feng and Pelton 2007) or hemicellulose (Salam et al. 2011b). Though PECs generally do not show promise for the absorption of large amounts of pure water, various studies have shown favorable results for the absorption of saline solutions, possibly representing urine (Paula et al. 2002; Feng et al. 2006; Feng and Pelton 2007). Despite the fact that under many circumstances the ionic interactions are sufficient to maintain the integrity of such gels, some researchers have reported enhancement of hydrogel capabilities when covalent crosslinkers have been applied in the preparation of such PEC hydrogels (Chen and Tan 2006; Shang et al. 2008; Salam et al. 2011b). As will be discussed in more detail later, the ability of some such hydrogels to increase their absorbency with increasing salinity can be attributed to a salt-induced weakening of the ionic pair-wise interactions that hold PECs together.
Products Combining Cellulosic Fibers with SAPs
A cursory inspection of currently available disposable absorbent products reveals a combination of cellulosic fibers and superabsorbent hydrogel materials, along with various barrier layers, permeable layers, elastic elements, and closures (Nagorski 1994). Within such products, conventional cellulosic fibers, i.e. the fluff pulp, play a variety of roles. Though, as mentioned earlier, bleached kraft fibers can absorb about 12 g water per g or solid (Zohuriaan-Mehr and Kabiri 2008), this is not enough to compete with the much higher water-holding capacity of the SAPs. Rather, the fluff pulp is used primarily to provide structure and wicking (Schuchardt and Berg 1991). In particular, the addition the fluff pulp can prevent gel blocking, which is the blocking of liquids by a complete wall of saturated SAP particles that are not separated sufficiently by the fibers. Masoodi et al. (2012) carried out studies directed at examining the combined effects of cellulose fiber and CMC hydrogel powder with respect to the wicking and uptake of water. Numerical simulation was used to model the phenomenon observed in the experiment. Though conventional assumptions of flow through capillary structures appear to be valid, the permeability was found to decrease with time (Schuchardt and Berg 1991; Wiryana and Berg 1991; Mosoodi et al. 2012). It was proposed in the cited studies that swelling of the hydrogel particles tended to seal off channels within the composite structure.
Weerawarna and Bing (2006) claimed a process in which superabsorbent gels were chemically attached to suitable fibers. Schuchardt and Berg (1991) describe various practical ways in which SAPs have been incorporated into cellulosic pad structures, including the various shapes and sizes of hydrogel particles. The distribution of SAPs within the structure is clearly an important consideration that will merit further study.
Integration of Cellulosic Solids into SAPs
An alternative way in which cellulose can be incorporated into highly absorbent products is by including solid cellulose in the mixture during gelation of the SAP. This approach has been applied by Larsson et al. (2010, 2011), with a focus on nanocellulose product applications. These authors observed that the effects of adding nanocellulose, in terms of both swelling and mechanical properties, was approximately the same as the addition of the same mass of conventional crosslinking agent. This suggests that the cellulosic fibrils, by being relatively long, tended to act as reinforcement, forming a nanocomposite structure (Hubbe et al. 2008; Pandey et al. 2013).
The effects of adding nanocellulose during the preparation of SAPs, as just described, are strikingly in contrast to various results that have been obtained when alkali montmorillonite, i.e.“nanoclay”, or other finely divided minerals have been added to the formulation during the preparation of SAPs (Pourjavadi and Ghasemzadeh 2006; Pourjavadi et al. 2008; Dai and Kadla 2009; Zhang et al. 2008b; 2009; Wang and Wang 2010; Wang et al. 2010, 2011a,b,d; Anirudhan et al. 2011; Bao et al. 2011; Chang and Zhang 2011; El Salmawi and Ibrahim 2011; Kabiri et al. 2011). In some cases the inclusion of nano-sized mineral particles during hydrogel preparation yielded a substantial increase in water absorption capacity (Pourjavadi et al. 2008; Zhang et al. 2008b; 2009; Wang and Wang 2010; Wang et al. 2010, 2011a,b,d; El Salmawi and Ibrahim 2011), whereas in some other cases the nanominerals tended to suppress absorbency (Pourjavadi and Ghasemzadeh 2006; Dai and Kadla 2009). A possible mechanism to account for the enhancement of absorption, in the cases indicated, could be a debonding effect, allowing pools of fluid to occupy openings that are made available by poor bonding of the adjacent SAP material to a suitably sized mineral surface. In addition, progressive exfoliation of montmorillonite may provide additional capacity for the holding of fluid (Knudson 1993). Alternatively, in cases where the minerals happen to be better bonded to the matrix, they could be expected to restrain swelling and imbibition.
Various published studies have addressed the question, “In what ways can the chemical composition of absorbent material affect its absorbency?” This section first considers publications dealing with the absorbency of cellulose and related biomaterials, followed by various copolymer derivatives, hydrogels, and combined structures.
Cellulose as an Absorbent Material
The chemistry of cellulose itself clearly can contribute to absorbency on account of the numerous hydrophilic –OH groups throughout its structure. The most widely used tissue and towel products are typically made from delignified wood fibers; in other words they are mainly comprised of cellulose and hemicellulose (Gigac and Fiserova 2008). As documented in Table A, the absorbency of ordinary delignified cellulosic fibers often lies in the range between 2 and 10 g/g, based on tests with pure water (Shepherd and Xaio 1999; Gigac and Fiserova 2008; Zhang et al. 2008a; Beuther et al. 2010). Mechanical pulp fibers, which contain essentially all of the relatively hydrophobic and rigid lignin that was originally present in the wood, tend to absorb less water in comparison to kraft or sulfite (delignified) fibers (Eriksson et al. 1991). Breier (1997) reported corresponding values of 0.5 g/g for cotton fiber and 0.6 to 1 g/g for various regenerated cellulose products.
The somewhat higher reported performance of wood-derived fibers, especially after delignification, probably can be attributed to their mesoporous nature and/or internal delamination as a result of mechanical treatments. Also, as will be discussed in detail later in this article, drying of cellulosic material can make a great difference in terms of its pore structure and capacity to absorb water.
Though the outermost surfaces of cellulose crystalline regions are able to interact with water, the crystalline domains within the cellulosic material are not expected to imbibe any water molecules (Bergenstrahle et al. 2008). Rather, it is the amorphous regions of cellulose, along with spaces occupied by the non-crystalline hemicelluloses, that are expected to swell and accommodate water (Heiner and Teleman 1997; Heiner et al. 1998; Bergenstrahle et al. 2008; Kocherbitov et al. 2008). These regions are represented schematically in Fig. 1. In support of this type of simplified model, it has been shown that the treatment of cellulose with ZnCl2 solution is able to disrupt the crystallization to some extent, an effect that can greatly increase the amount of water absorption (Williams and Stannet 1979; Vitta et al. 1986).
Fig. 1. Representation of crystalline regions and amorphous regions within the nanostructure of delignified fibers from wood
Absorption in the amorphous regions of cellulose, as well as within the domains containing hemicellulose, can be expected to involve strongly bound water, i.e. a form of hydration (Joubert et al. 1959; Gert 1996). In support of this statement, a strong exotherm has been observed accompanying the first increment of water uptake onto completely dry cellulose (Kocherbitov et al. 2008). Water that is strongly bound to cellulose has also been characterized as “hard to remove water” as measured by thermogravimetric analysis or as freezing or non-freezing bound water as measured using differential scanning calorimetry (Park et al. 2006, 2007; Sen et al. 2012).
In the case of conventional cellulosic fiber materials (not considering hydrogels), a majority of the retained water within a pad or sheet product is likely to reside between fibers rather than within them (Jerkeman 1970). Issues relative to the bulkiness of cellulosic absorbent materials will be discussed in a later section.
Hemicellulose and Absorbency
The swelling of cellulosic fibers has been found to correlate positively with the proportion of hemicellulose (Eriksson et al. 1991). The positive contribution of hemicel-lulose to swelling can be partly attributed to its irregular structure, with branching and side groups (e.g. as shown in Fig. 1). The irregularity of hemicellulose tends to discourage its crystallization (Billmeyer 1971; Brydson 1975). Peura et al. (2008) observed levels of crystallinity only in the range 16 to 19% in films cast from purified xylan. Likewise, Gabrielii et al. (2000) and Gröndahl et al.(2004) observed somewhat higher levels of crystallinity in films from aspen wood xylan prepared under conditions optimized to promote crystallization. A second factor that contributes to the hydrophilic nature of hemicelluloses is the content of carboxyl groups, which can dissociate to the carboxylate form at intermediate to high conditions of pH (Lindström and Carlsson 1982). Indeed, Salam et al. (2011a) showed that hemicellulose could be rendered even more hydrophilic by derivatizing it to further increase the content of carboxyl groups.
Extraction of hemicellulose from kraft fibers with alkaline solutions has been found to increase the water absorptivity of the fibers (Lund et al. 2012). This is somewhat surprising, given the fact that hemicelluloses have a distinctly hydrophilic nature. The explanation appears to lie in the conversion of other ionic forms of the remaining hemicellulose to the sodium form, which is known to enhance swelling (Lindström and Carlsson 1982). The importance of hemicellulose was also shown by Laine and Stenius (1997), who demonstrated a positive relationship between fiber ionic charge and swelling with water. Recently several researchers have attempted to take advantage of the hydrophilic nature of hemicelluloses by using them as a key component of highly absorbent hydrogels (Gabrielli and Gatenholm 1998; Gabrielli et al. 2000; Meena et al. 2011; Peng et al. 2011; Pohjanlehto et al. 2011; Salam et al. 2011a; Yang et al. 2011; Karaaslan et al. 2010, 2011, 2012).
Lignin and Absorbency
Compared to the polysaccharides in wood, lignin is more hydrophobic. Obataya et al. (2007) found that delignification of wood increased its tendency to swell in water. Likewise, as reported by Parham and Hergert (1980), delignified kraft pulps can be expected to absorb significantly more water than mechanical pulps, in which most of the native lignin still remains after pulping. Further demonstration of the relative hydrophobicity of lignin has been shown by studies in which lignin or its derivatives were precipitated onto cellulose, thus decreasing the substrate’s wettability by water (Sipponen et al. 2010; Li et al. 2011; Payne et al. 2012). One way to overcome the adverse effect of lignin on swelling is to heat lignified fibers above approximately 70 oC (Eriksson et al. 1991). Such heating, in the presence of water, was observed to soften lignin sufficiently so that the water uptake was increased.
Extractives and Absorbency
Because the term “extractives” usually implies that a substance can be removed from cellulosic material with an organic solvent (Back and Allen 2000), it follows that most such substances are likely to resist water wetting and penetration. When microfibrillated cellulose materials of different lignin and extractives contents were exposed to an organic solvent to remove any extractives, the water-cellulose interactions as measured by water contact angle all increased, with the largest increases observed for the material with the most extractives initially (Spence et al. 2010). However, a search of the literature revealed a paucity of studies attempting to correlate levels of extractives and extents of swelling or absorbency. As already noted, Lund et al. (2012) observed an increased swelling of softwood kraft fibers after their extraction with NaOH. Though the authors attributed this effect to the removal of hemicelluloses, it seems likely that at least part of the effect may have been due to removal of certain extractive materials, e.g. the salts of fatty and resin acids.
Ström and Carlsson (1992) found that an adverse effect of extractives on the water-wettability of kraft pulps could be overcome by treatment with oxygen plasma. The effect was attributed to oxidation of the exposed lignin. Related results have been published by Uner (2009), who found that corona discharge treatment of paper resulted in increased absorbency. Another way to increase wetting and absorbency involves use of suitable surfactants, i.e. rewetting agents (Shepherd and Xiao 1999).
Hydrophilic Substituent Groups and Absorbency
The presence and increasing level of hydrophilic groups in a cellulose-based material can contribute to water uptake in a number of ways, e.g. through hydrogen bonding with the water, by electrostatic effects (if the groups are charged), and by somehow interfering with the densified structure of the biomaterial (Lepoutre et al. 1973). Effects of several kinds of water-loving ionic groups are considered below.
The relationship between two key types of hydrophilic groups and the transient hydrogen bonds within liquid water is illustrated in Fig. 2. Hydroxyl groups exposed at the surface of an absorbent material can be expected to form hydrogen bonds with adjacent water molecules, as shown in the left frame of the figure. The hydrogen bonding extends, in a continuously changing pattern, throughout the whole liquid phase (Drost-Hansen 1969). The right-hand frame of the figure emphasizes the role of bound ionic groups, which can be expected to have a strong affinity for the polar molecules of water. As shown, the bound negatively charged groups can be expected to interact strongly with the positively charged pole of adjacent water molecules. The water molecules tend to orient themselves in a transient pattern, consistent with their polar nature.
Fig. 2. Representation of hydrogen bonding among the molecules of liquid water and also associated with hydrophilic groups. Left: Hydrogen bonding associated with hydroxyl groups bound to a substrate. Right: Interaction and orientation of polar H2O molecules with ionized groups such as carboxylate or sulphonate groups, all associated with the transient hydrogen bonded structure of liquid water
By low levels of oxidative treatment it is possible to increase the proportion of carboxylic acid and/or aldehyde groups on the surfaces of cellulosic fibers. Such oxidative treatment (e.g.TEMPO-mediated oxidation) has been shown to greatly increase water absorbency of softwood kraft pulps (Brodin et al. 2012; Brodin and Theliander 2012). Periodate oxidation has been shown to offer an alternative means of achieving similar effects (Zhang et al. 2008a).
Carboxymethylation and succinylation
Chemical derivatization is another way to increase the number of carboxyl groups on cellulosic surfaces. The idea is to covalently attach groups containing carboxyl functions to the cellulose. In particular, it has been found that carboxymethylation of fibers can greatly increase absorbency even when the treatment level is low enough to avoid solubilization of the source cellulosic material (Schuchardt and Berg 1991). As already noted, however, it is more common to employ extensive carboxymethylation of cellulose, an approach that allows use of fully soluble CMC as a basis for forming superabsorbent hydrogels through crosslinking (Zohuriaan-Mehr and Kabiri 2008; Sannino et al. 2009; Chang and Zhang 2011). Carboxylic acid groups also can be attached to cellulosic fibers by means of reaction with succinic anhydride or maleic anhydride; such treatment likewise has been shown to increase absorbency (Yoshimura et al. 2005, 2006).
Zhang et al. (2008a) showed that sulfonate groups can be especially effective to increase the water retention values of bleached hardwood kraft fibers. It was noted that because sulfonic acid is a strong acid, a high proportion of such groups tend to retain their negative charge at any given moment, even when the charge density of a fiber surface is relatively high. The acid groups remain dissociated throughout the pH range where such fibers are likely to be used. Hou et al. (2008) also showed that sulfonation improved absorbency.
Sulfonate groups also have been shown to be advantageous in the preparation of high-performing superabsorbent hydrogels. For instance, sulfated starch has been used as a polyelectrolyte component in the preparation of high-performing SAPs (Lim et al. 2001). The improved performance, relative to similar hydrogels containing just carboxylate groups, again was attributed to a tendency of the strong acid groups to remain fully dissociated despite the relatively high levels of substitution in the starch-based polyelectrolyte.
Saito et al. (1994a, 1995) showed improvements in absorbency analogous to those of the sulfate group when using phosphorylated lignocellulosic materials. The phosphate was characterized as “a water-attracting group” (Saito et al. 1995). However, in their earlier article the authors had concluded that effects of preparation conditions on surface area and porous structure were more important relative to the content of ionizable groups (Saito et al. 1994a). In a third study the same authors attributed the increased absorbency to inter-crystalline swelling (Saito et al. 1994b).
Hydrophobic Substituent Groups and Absorbency
Cellulose-based materials are in general hydrophilic, and it has been found that a de-swelling effect can be achieved by derivatization with hydrophobic groups. Such an effect was shown most clearly by Koroskenyi and McCarthy (2001), who used acetylation to systematically decrease the swelling behavior of glucomannan. Obataya et al. (2007) showed that acetylation increased the ability of lignin-containing fibers to absorb non-polar solvents. One exception to the rule has been observed when cellulose fibers are derivatized at a low level, e.g. 5% substitution of the hydroxyl groups by acetyl groups; in such cases Ehrnrooth et al. (1977) reported increased swellability, which was attributed to the reduced crystallinity of the partially substituted cellulose.
Hydrogel Attributes and Absorbency
Considerable attention has been devoted to the absorbency of superabsorbent hydrogels in light of their ability to take up 10 to over 1000 times their mass in water. Various factors involving the structure of hydrogel composition and molecular structure appear to control absorbency. Some of the key factors, as described below, are the molecular mass of the component polymers, the density of crosslinking, and the nature of the chemical species used in the crosslinking. In addition, the presence and density of ionizable groups on polymer chains, which are physically unable to diffuse out of the gel, provide a driving force for osmotic effects, inducing more water to enter the gel.
High molecular mass has been proposed as being a key requirement of polymers that are intended as the main components of superabsorbent hydrogels (Yoshimura et al. 2006). Thus, Kono and Fujita (2012) observed increased absorbency with increasing molecular mass of the cellulose used to prepare a crosslinked CMC hydrogel from an ionic liquid solution. In principle, if the molecular mass is relatively low, then there will be a greater likelihood of dissolution and loss of the polyelectrolyte from the hydrogel during its use. Though a higher degree of crosslinking could be used to overcome such deficiencies, such an approach would not be expected to yield high levels of absorbency (see next).
In the absence of any crosslinking, many of the most widely used superabsorbent products would simply dissolve when placed in water. Though the highest swelling performance is often observed at the lowest levels of crosslinking (Bin et al. 2000; Wach et al. 2001, 2002; Liu et al. 2005a; Demitri et al. 2008; Han et al. 2010; Pourjavadi et al. 2010; Kono and Fujita 2012), it is generally observed that higher levels of crosslinking (or other reinforcement) are needed to achieve the needed mechanical resilience and resistance to dissolution of superabsorbent products (Guilherme et al. 2005; Myung et al. 2008; Tanaka et al. 2005; Schexnailder and Schmidt 2009; Zohuriaan-Mehr et al. 2010).
Somewhat contrary results were reported by Barbucci et al. (2000). Although these authors observed a monotonic decrease in swelling with increase in crosslinking up to a 50% level, they observed a higher swelling at “100% crosslinking”. The anomalous behavior was attributed to disruption of hydrogen bonding by the crosslinking agent. Anomalous increases in swelling with increasing crosslinking also have been observed when the crosslinking agent is much more hydrophilic than the polymer chain. Thus, Oliveira et al. (2013) showed that increasing crosslinking of cellulose acetate (a hydrophobic polymer) with a di-anhydride increased water sorption. It was noted that the curing of a dianhydride can form two ester bonds and two free carboxyl groups, the latter of which increase the ionic character of the hydrogel at intermediate pH values (Kono and Zakimi 2013; Oliveira et al. 2013).
Numerous studies have addressed the goal of finding an optimum level of crosslinking for various types of superabsorbent gels (Ali et al. 2008; Pohjanlehto et al. 2011; Kono and Fujita 2012). Kono et al. (2013) observed that the balance between simple grafting and crosslinking reactions is sometimes sensitive to the choice of solvent system. As one means to control this balance, Lionetto et al. (2005) showed that the degree of crosslinking can be monitored during the process by evaluation of ultrasonic wave propagation.
Chang et al. (2008) showed that the type of crosslinking, i.e. covalent-type vs. physical crosslinking, also can make a big difference relative to results. For the observed system it was found that chemical crosslinking yielded higher absorbency. In the cited case the chemical crosslinking approach made it possible to avoid use of multivalent cationic agents. As will be discussed later, such species generally tend to reduce the swelling of typical hydrogels.
Several innovative studies have shown that, at least in principle, higher absorbency of hydrogels can be achieved by using longer, flexible crosslinking agents (Xiao and Meshitsuka 2001; Sannino et al. 2003; Chang and Zhang 2011; Kabiri et al. 2011; Pohjantehto et al. 2011). In practice this is achieved by inserting a “spacer” between the two reactive groups of a suitable crosslinking agent (Chang and Zhang 2011). The spacer within the crosslinked structure of a SAP can be expected to allow further expansion and therefore greater uptake of aqueous fluid. It is notable that analogous results were obtained when paper was treated with high- vs. low-mass polyvalent carboxylic acid species; the longer molecules gave rise to greater tensile strength and folding endurance (Yang and Xu 1998; Xu and Yang 1999).
There is an interesting parallel between the effects of spacers and the effects of certain very fine mineral products, which were mentioned earlier (Pourjavadi et al. 2008; Zhang et al.2008b; 2009; Wang and Wang 2010; Wang et al. 2010, 2011a,b,d; El Salmawi and Ibrahim 2011). Such nano-minerals, if poorly adhered to the adjacent polymeric material, can be expected to allow the opening up of fluid-filled spaces within the hydrogel. The presence of void spaces in such nanocomposite structures is supported by observations of relatively high opacity in some cases (Dai and Kadla 2009). It seems likely that some of the high levels of water uptake reported for SAPs based on interpenetrating networks (e.g. Çaykara et al. 2006; Chang et al. 2009b) may be due to one network interfering with the crosslinking of the other network during its preparation.
Ionizable Groups and Absorbency
Hydrogels that contain carboxylic groups generally exhibit increasing absorbency with increasing pH, especially within the pH range of about 3 to 7; the following examples are from studies involving cellulose-based materials in SAPs (Barbucci et al. 2000; Harsh and Gehrke 1991; Sannino and Nicolais 2005; Pourjavadi et al. 2008; Anirudhan et al. 2011; Bao et al.2011; Wang et al. 2011a). The effect has been attributed to dissociation of the groups to the carboxylate form (Barbucci et al. 2000). As noted by Harsh and Gehrke (1991), such effects of pH are consistent with a principle that absorbency increases with increasing charge density. While most such studies have employed pH as the main independent variable to manipulate the charge density, an alternative approach can be achieved by adjusting the composition of the polyelectrolytes used in forming the hydrogel. Thus, Wu et al. (2012) determined that absorbency increases with increasing level of charged monomer groups up to a 70% of acrylic acid in a cellulose graft copolymer hydrogel with acrylamide monomers and a crosslinker. Considering the importance of the topic, more research is recommended.
Particle Size of Hydrogel
Based on limited available data, the size of hydrogel particles can affect the rate of absorption of water, but the effect on absorption capacity appears to be weak (Tanaka and Fillmore 1979). The cited study dealt with acryalamide gels formed at a 5% concentration of polymer in water. The cited authors observed that the rate of swelling was inversely proportional to particle size, as would be expected according to a diffusion-limited process. Bao et al. (2011) found only a weak and inconsistent relationship between equilibrium absorption and particle size in the case of superabsorbent hydrogels based on CMC, acrylamide copolymer, and montmorillonite nanoparticles.
In principle, structural aspects of various hydrogels can affect not only the absorption capacity but also the rate of filling. The focus here is on the effects of various cellulosic components on absorbent structure and performance in different situations. One can consider three size ranges in which important effects may be manifested, i.e. due to the presence of cellulosic fibers, due to nanocellulose, and due to the inherent differences in stiffness of different types of polymer chains within hydrogels.
Macroscopic Structural Effects
Cellulosic fiber use
Fluff pulp, which is usually provided as bleached softwood kraft fibers, is typically a main component of highly absorbent products (Jerkeman 1970; Parham and Hergert 1980; Nagorski 1994; Buchholz et al. 2005; Lund et al. 2012). Although, as noted earlier, the fibers themselves can accommodate some liquid (Zohuriaan-Mehr and Kabiri 2008), most of the absorption capacity of such products can be attributed to the inclusion of SAPs (Aliouche et al. 2001). The fibers are used in order to provide integrity to the product (Brodin and Theliander 2012) as well as to accommodate more rapid wicking of fluid throughout the absorbent and to prevent blocking (Buchholz et al. 2005).
According to Buchholz et al. (2005) a network of cellulosic fibers intertwined through a loose matrix of SAPs in an absorbent product can be expected to restrict the swelling of the SAPs. A constraining pressure of about 2 kPa was estimated in the cited work. It follows that the fibers within a SAP structure may decrease the absorption capacity; however, the fibers might be needed to maintain structural integrity and to allow quicker distribution of liquid through the product. A typical composition for a disposable diaper has been cited as 43% fluff pulp, 27% SAP material, and the remainder being various plastic materials and adhesives (Edana 2005).
Refining of fibers and absorbency
Refining refers to the repeated mechanical compression and shearing of a slurry of cellulosic fibers in the gap between the raised bars of a stator and a rotor (Smook 1992). The usual intention of refining is to cause internal delamination of the cell wall structure and thereby to increase the flexibility and conformability of the wet fibers. Refined fibers can be expected to develop a greater relative bonded area, hence achieving higher strength of paper products (Page 1969; Lowe et al. 2007). However, when high absorbency is the key goal, studies have shown that the extent of refining should be kept low (Gigac and Fiserova 2008). The cited authors observed that the manner of refining, in addition to the applied energy, can be important; a high specific edge load (generally obtained by using refiner plates with a relatively coarse pattern) resulted in higher absorbency of the resulting tissue paper. These effects can be attributed to a more voluminous packing of fibers that have become bent as result of harsh but limited refining treatment. Due to larger void volumes, more water was able to be absorbed into the tissue paper.
Consistent with the preceding discussion, factors that tend to reduce the packing density of a structure can be expected to favor more rapid flow into products made from cellulosic fibers. Though mechanical pulp fibers individually tend to have lower absorbency than bleached kraft fibers, they also tend to be stiffer and less compressible, often yielding a bulkier structure of paper. The relationship between fiber stiffness and the resulting paper’s apparent density was demonstrated in a fundamental manner by Steadman and Luner (1985). Accordingly, Brodin et al. (2012) observed higher absorbency for pads composed of chemithermomechanical pulp (CTMP) fibers in comparison with kraft fibers. In the case of kraft pulps, Lund et al. (2011) showed that increased bulk could be achieved by crosslinking the fibers with a diepoxide. Such treatment would find application in products that need to resist wet compaction, as might happen in the case of a wet diaper holding up the weight of a child.
Another way to increase the bulkiness of a dry paper structure is to treat the fiber suspension with a suitable cationic surfactant before the drying process. Fatehi et al. (2010) showed that treatment with such “debonding agents” rendered the resulting tissue paper less dense. Likewise, Shepherd and Xiao (1999) showed that surfactant treatment of wet pulp resulted in an increased effective pore size and hydrophilicity of fibers once the paper had been dried. Cationic surfactants yielded the largest increase in the rate of wetting.
Another factor that is sometimes overlooked when considering the absorption capacity of a cellulose-based product is the swelling of the fibers themselves upon wetting, an effect that can change the dimensions of the structure as a whole. Bristow (1972) introduced this concept as a means of accounting for the absorption capacities of fiberboard products. Factors affecting the swelling of cellulosic fibers have been considered in various studies (Jayme and Büttel 1964, 1968; Scallan and Carles 1972; Gigac and Fiserova 2008; Zhang et al. 2008a; Cheng et al.2010; Lund et al. 2011).
Cellulosic fiber webs, especially if the fibers have a relatively coarse and bulky nature, appear well suited to achieving a relatively rapid uptake of water (Beuther et al. 2010). Thus, as noted by Rezai and Warner (1997), combinations of SAPs and cellulosic fibers can be used in an attempt to achieve the best results from each component – relatively rapid wicking by the fibers and high absorbency contributed by the hydrogel. Schuchardt and Berg (1991) showed that, to a first approximation, well-established equations of flow through packed beds can be used to model the flow of aqueous fluids through mixtures of cellulosic fibers and superabsorbent hydrogel particles. However, it was apparent that swelling of the SAPs tended to progressively restrict the flow as a function of time. A correction term, i.e. a permeability factor, was defined to account for this effect. To make it less likely that swollen SAP particles can seal the passages adjacent to a cellulosic fiber, one could select fibers having an irregular cross-section (Pramanick and Crouse 1998). For instance, certain viscose fibers have been observed to be multi-lobal in cross-section (Schuster et al. 2006). The use of twisted, curly fibers for absorbent products has been patented (Sun and Lindsay 2003; Hu and Ko 2006).
As an alternative strategy to achieve high flow rate within a relatively large volume of superabsorbent material, certain authors have prepared superabsorbent fibers. These fibers tend to pack less densely than the same mass of rounded particles. For example, Feng et al. (2010b) spun blends of cross-linked polyacryonitrile and cellulose acetate as a means of producing highly absorbent fibers. Related results were achieved by Goetz et al. (2011) by extensive grafting of softwood kraft fibers with poly(methyl vinyl ether-co-maleic acid) and polyethylene glycol. Both studies achieved SAP-like levels of water absorbency, while retaining the bulky bed structure inherent in the use of suitably stiff fibers.
Another way in which to consider aspects related to the bulky structure of cellulose-based sorbents is to examine the effects of pore spaces. Such spaces can vary greatly in size, size distribution, and connectivity. Pores also can be quantified in terms of their net volume relative to the solid content. For instance, Beuther et al. (2010) estimated that a typical sheet of tissue paper can have a relative void volume greater than 0.9, consistent with the ability to pick up as much as ten times its mass in aqueous fluids. Brodin et al. (2012) used porosimetry to quantify the pore size distribution within an absorbent foam; by this means it was shown that inclusion of relatively stiff CTMP fibers resulted in greater preservation of pore volume when a mixed absorbent structure was compressed. Studies by Sannino et al. (2004b, 2005) showed that the porous nature of hydrogel foams could be systematically adjusted by varying the conditions of precipita-tion and foam formation.
At the nano-scale, as noted already, it has been proposed that a significant amount of water contained in a hydrogel may be present as small “lakes” rather than as chemically bound water (Gelin et al. 2007). In other words, the distribution of water may be uneven, when considered on a nano-scale. Hydrogels formed in the presence of nano-sized micelles of surfactant molecules have been shown to have enhanced porosity; such effects have been shown to nearly double the absorption capacity for water in some cases (Shi et al. 2013). In the case of chemically pulped fibers, effects related to porosity also extend to the nano-scale. A mesoporous structure within various kinds of cellulose material has been well documented (Stone and Scallan 1968; Schuster et al. 2006). Also, it has been proposed that grafting of such fibers with hydrophilic polymer chains may serve to brace adjacent microfibrils apart from each other, thus contributing to the swellability and ultimately to the volume available for sorption of fluids (Lepoutre et al. 1973).
At a molecular scale it has been proposed that the inherent stiffness of the cellulose polymer chain can lead to a firm hydrogel structure that is less susceptible to collapse of foam structures (Chang et al. 2009a; Zhang et al. 2010). As a likely explanation, the apparent stiffness may be due to cellulose’s tendency to self-associate into microfibrils, i.e. semi-crystalline domains, during the process of gelation (Chen et al. 2006; Kondo et al. 2008). A substantial presence of crystalline domains has been detected even in the case of hydrogels based on hemicellulose (Karaaslan et al. 2010, 2012). Nishio and Manley (1988) showed that the presence of polyvinyl alcohol tended to suppress the formation of crystallinity in cellulose when it was regenerated from an ionic liquid solution. It is reasonable to expect that such microfibrils and partially crystalline character play a role, especially when hydrogels are prepared from solutions of either cellulose itself or of derivatives such as CMC having a low enough degree of substitution as to still permit local, well-organized self-association of chain segments (Thuresson and Lindman 1999; Saito et al. 1994a,b; Bhandari et al. 2012). Bhandari et al. (2012) found an inverse relationship between the swelling tendency and the degree of crystallinity within cellulose films regenerated from ethanol-water-NaOH solutions.
Nanocellulose and Absorbency
During recent years there has been a high level of research activity devoted to the preparation and characterization of very finely separated cellulosic fibrillar or elongated crystalline particles (Klemm et al. 2011). Likewise, there has been related patenting activity (Duran et al. 2012; Charreau et al. 2013). Cheng et al. (2010) showed that certain nanocellulose materials by themselves achieved water retention values in the range of 3.4 to 3.5 g/g. Hua et al. (2011) patented a production system for nanocellulose and described their use in absorbent products.
Nanocellulose materials can be formed into highly porous, absorbent structures by freeze-drying of their aqueous suspensions. Thus, Pääkkö et al. (2008) showed that native cellulose nanofibrils, obtained by enzymatic and mechanical processing of bleached sulfite softwood fibers, yielded strong and voluminous aerogels upon freeze-drying. Immersion of an aqueous suspension of nanofibrillated cellulose in liquid propane, followed by vacuum pumping, yielded an aerogel structure that shrunk by only 7%. A practical, short-cut method was developed in which aerogels were formed by direct vacuum pumping (and induced freezing) of the initial aqueous suspensions. The latter method caused 20% shrinkage. Brodin et al. (2012) explored the use of microfibrillated cellulose in combination with macroscopic cellulose fibers in a freeze-dried structure; absorbencies in the range of 21 to 65 g/g were observed. The composite structure displayed better mechanical stability of its foam structure, as well as higher absorbency. Oxidation was used in the cited work to further increase the absorbency of the fiber component of the structure. Kettunen et al. (2001) reported an absorbency of 35 g/g in the case of freeze-dried nanocellulose aerogels.
Nanocellulose also has been incorporated into hydrogel products (Abbas et al. 2011; Karaaslan et al. 2011; Gane et al. 2012). The mechanical integrity of such hydrogels is generally improved by such addition (Karaaslan et al. 2011), suggesting that the elongated nanocellulose particles can have a function similar to that of a cross-linking agent, or probably more precisely as nano-reinforcement of the gel structure. A tight, relatively defect-free attachment between cellulose nanofillers and a hydrogel matrix has been demonstrated by a high transparency of such nanocomposite hydrogels (Dai and Kadla 2009). This was in contrast with the effect of nanoclay usage, which resulted in opaque hydrogels.
Maintaining meso-pore structure
Cellulosic nanomaterials as part of a hydrogel structure may help to alter or maintain the structure of pores. Spagnol et al. (2012a) found that incorporation of cellulose nano-whiskers (i.e.long crystals) tended to increase the pore size of starch-acrylic hydrogels. Relatively low levels of nanocellulose addition tended to create defects in the structure, thus decreasing the elastic modulus. However, amounts greater than 5% yielded a net increase in strength, suggesting the presence of a contiguous network of cellulosic elements within the matrix (Favier et al. 1997; Chakroborty et al. 2006a,b; Hubbe et al. 2008).
Effects on swelling
Effects of the presence of nanocellulose on the swelling of SAP composites has been considered in several studies (Buyanov et al. 2010; Larsson et al. 2010, 2011; Laftah et al. 2011; Spagnol et al. 2012a). In the review article by Laftah et al. (2011), the findings of such studies were summarized by stating that nanofibers can either increase or decrease absorbency in different cases. The hydrophilic nature of cellulosic fibers contributes to the absorbency of the structure as a whole, yet the more rigid nature of the resulting network structure can contribute to decreased absorbency. This may explain an observed complexity of the relationship between nanocellulose content and absorbency (Larsson et al. 2011). Anisotropic swelling was observed by Buyanov et al. (2010) in nanocellulose-filled hydrogel structures that had been pre-compressed during their formation. Spagnol et al. (2012a) reported that although nanocellulose inclusion at relatively high levels decreased absorbent capacity, it also resulted in greater pore size. The increase in pore size was judged to be an important advantage in terms of the absorption rate and the overall performance of the absorbents. The cited authors found, however, that at a low content of nanocellulose (5 to 10%) the water content was the highest. This trend was confirmed in subsequent work (Spagnol et al. 2012b,c).
EFFECTS OF FLUID COMPOSITION ON ABSORBENCY
In addition to the chemical and structural aspects of the sorbent material, as discussed in the previous two sections, the absorbency characteristics of cellulose-based materials can also be profoundly affected by the nature of the aqueous fluid to be absorbed. In addition to providing practical information, the goal of this section is to summarize evidence that can be used to support various theoretical concepts, which are addressed in later sections.
Effects of Salinity
Inspection of the last two data columns in Table A reveals a strong contrast between the amounts of pure water taken up by various cellulose-based absorbents vs. the corresponding amounts of saline solution (often evaluated with 0.9% NaCl). In typical cases the latter value is less than half of the absorption capacity in the case of pure water. Especially sharp contrasts in capacity, when comparing deionized vs. saline water, have been reported in the case of superabsorbent hydrogels (Lim et al. 2001; Suo et al. 2007; Zhang et al. 2008b, 2009; Kim and Mun 2009; Liang et al. 2009; Pourjavadi et al. 2008, 2009; Chang et al. 2010; Bao et al. 2011; Liu et al. 2011; Wang et al. 2011a,c,d; Wu et al. 2012).
For certain end-uses it would be highly desirable to maintain a high level of absorbency even in the presence of intermediate levels of salt ions. A number of authors have reported progress in development of somewhat “salt-tolerant” hydrogels (Kulicke and Nottelmann 1989; Yang et al. 2009; Wu et al. 2012). Yang et al. (2009) attributed the observed salt-tolerance to graft polymerization between CMC and polyacrylamide. Rezai and Warner (1997) reported that even though salt greatly decreased the absorbency of acrylonitrile-grafted wood pulp fibers, such fibers still absorbed three times as much water as their ungrafted counterparts. Also, various hydrogels comprised of hemi-cellulose in combination with chitosan have shown favorable performance in their relative ability to absorb saline solutions (Salam et al. 2011b).
Fält and Wågberg (2003) as well as Li and Dai (2004) observed an interesting effect of salt on the swelling of highly purified cellulosic fibers. The samples had been prepared with the complete exclusion of salt ions. Swelling actually tended to increase with the first added amounts of such salts as Na2SO4 and NaCl. The effect was attributed to an increase in effective pH within the mesopore spaces within the fibers. Related mechanisms have been further considered (Hubbe 2006; Hubbe et al. 2007a), as a way to account for effects of salt concentrations on the streaming potential of cellulosic fibers.
Effect of pH
Conventional kraft fibers used in papermaking are known to exhibit increased water retention with increasing pH (Jayme and Büttel 1964; Lindström and Carlsson 1982). Such effects have been observed most clearly in the case of relatively high-yield pulps, i.e. fibers that contain substantial amounts of hemicellulose and other species bearing carboxylic acid groups (Lindström and Carlsson 1982). In comparison, such effects in the case of bleached kraft pulps tend to be small (Lindström and Carlsson 1982; Hubbe and Panczyk 2007a).
In the case of hydrogels having cellulosic content, numerous studies have reported increased swelling with increasing pH, at least up to the neutral point (Vitta et al. 1986; Barbucci et al.2000; Sannino and Nicolais 2005; Chen and Tan 2006; Ali et al. 2008; Wang and Wang 2010; Anirudhan et al. 2011; Bao et al. 2011; Eldin et al. 2011; Wang et al. 2011b,d). Certain studies (Liu et al. 2005a; Ibrahim et al. 2007; Pourjavadi et al. 2007b; Akar et al. 2012; Shi et al. 2013; Wang et al. 2008, 2013) showed a distinct maximum swelling at neutral pH; a possible explanation for this finding is that increasing amounts of ions must be added to such a system to achieve pH values significantly different from 7; such ions would be expected to have the same effect corresponding to the previous subsection (Effects of Salinity). Odd results, roughly opposite to those just cited, were obtained by Varaprasad et al. (2011), though no explanation was given. Wang et al. (2010) developed a system based on hydroxyethylcellulose-g-poly(acrylic acid) with very finely divided attapulgite mineral that achieved consistent high swelling within the broad pH range between 4 and 10.
Evidence of an additional contribution to de-swelling has been observed in some hydrogel systems at the low end of the pH range investigated. Barbucci et al. (2000) observed a strong contraction of hydrogels at pH values well below the pKa value of the constituent carboxyl groups. The effect was attributed to the formation of hydrogen bonded pairs between adjacent protonated carboxylic acid groups.
Shang et al. (2008) developed a creative system in which the bending of a gel body could be used as an indication of the strength and direction of an imposed electric field. According to the authors, the bending of the gel was attributable to slight pH differences on either side of the hydrogel, resulting from the half-reactions at the electrodes used to generate the electric field.
Effects of Valency and Identity of Positive Ions
Numerous studies have shown greater decreases in swelling with increasing valency at a constant concentration of cations in the aqueous solution to which a hydrogel is exposed (Vitta et al. 1986; Kulicke and Nottelmann 1989; Feng and Wang 2010a; Bao et al. 2011; Akar et al. 2013; Chang et al. 2012; Li et al. 2012; Raafat et al. 2012). The effect has been attributed to reversible complexation between the cationic species and one or more carboxylate groups in the hydrogel (Bao et al. 2011; Raafat et al. 2012). In addition, multivalent cationic species can be expected to have a relatively large effect in reducing ionic double layer forces of repulsion associated with the bound carboxylate groups of a typical hydrogel (Hiemenz 1977). As noted by Vitta et al. (1986), by acting as a bridge between more than one carboxylate group in a hydrogel, a multivalent cation can act as a crosslinking agent, thus decreasing swelling.
Though not as obvious or important as the effect of valency, many studies have detected differences in the effects on swelling between cations of the same valency. Two studies dealing with SAPs showed a greater suppression of swelling in the presence of the divalent magnesium ion, in comparison with the same concentration of calcium ion (Bao et al. 2011; Li et al.2012). The same authors showed that monovalent sodium ion generally suppressed swelling to a greater degree than the potassium ion. In the case of acrylic-grafted sulfite fibers, Vitta et al. (1986) reported the following sequence of increasing water retention values at a fixed concentration of the following ions:
Al3+ < Cu2+ < Zn2+ < Ca2+ < Mg2+ < H+ < Na+ < K+ < Li+
It should be stressed that all of the effects related to different cations of the same valency were small relative to the effect of valency. It is interesting, though, that within the individual studies cited, the results tended to follow regular progressions according to the tabulated size of the respective ionic species.
Surfactants and Absorbency
Surfactants have been shown to affect absorbency in three ways. First, in the case of ordinary paper, addition of various surfactants has been shown to facilitate more rapid wetting in some cases (Hodgson and Berg 1988; Shepherd and Xiao 1999). Such treatments were able to overcome a “wetting delay” phenomenon that is often observed when water is spread onto initially-dry paper. Increased absorbency was observed by Shepherd and Xiao (1999) following treatment of the paper with surfactant.
Second, addition of a cationic surfactant before preparation of paper has been found to yield a bulkier structure (Mann and Leah 1987; Fatehi et al 2010). The debonding effect of such treatment resulted in dry-strength decreases in the range of 14 to 19%. Poffenberger et al. (2000) showed that, due to the hydrophobic nature of typical cationic surfactants, the net effect of debonding treatment during the production of tissue paper may be a reduction in the uptake of water. However, by the careful selection of debonding agents having a relatively hydrophilic character, such decreases can be avoided.
Third, in the case of SAPs, addition of cationic surfactant has been shown to result in a marked decrease of absorbency (Wang and Wang 2010; Wang et al. 2011c). Such an effect is consistent not only with the partially hydrophobic nature of the surfactants employed, but also with the expected electrostatic interactions. Wang and Wang (2010) noted that changes in swelling can be affected by screening of the negative charges within the SAP due to the cationic charge of the surfactants, changes in hydrogen bonding interactions, and a hydrophobic interaction between the alkyl tails of the surfactant molecules. In related work, Gawel et al. (2012) showed that the nature of a conventional anionic hydrogel was transformed upon the addition of a cationic polyelectrolyte. The surfactant treatment caused the hydrogel to have absorption behavior similar to that of a polyelectrolyte complex type of hydrogel, as was discussed earlier. In the case of ordinary kraft fibers, Swerin et al. (1990) showed that addition of cationic polymers caused a loss of swelling. Effects of cationic polymers on the swelling of cellulosic materials and anionic hydrogels will be considered in a later section dealing with theoretic aspects of swelling capacity.
The term “adsorption” can be defined as a process by which a chemical species (such as a polyelectrolyte) becomes attached to a surface – either on the apparent external surface of a material or on surfaces within a porous structure. A study by Ström and Kunnas (1991) demonstrated that polyelectrolytes adsorbed onto papermaking fibers can make a significant contribution to their water retention value. Cationic polyelectrolytes of high charge density were shown to decrease the amount of water held within the pulp fibers after their centrifugation to remove free water. The effect was attributed to the ability of highly-charged cationic polymers to accelerate dewatering of paper pulp; this was a surprising finding, since water retention value tests are usually intended to provide information about equilibrium conditions. A related study by Hubbe and Panczyk (2007b) confirmed the effect of low- to moderate-mass, high-charge-density cationic polymers on reducing the water retention of kraft pulps. By contrast, sufficiently high levels of very-high-mass cationic polymers tended to increase water retention. The latter effect was attributed to the water associated with the polyelectrolytes themselves in their adsorbed state.
Effects of temperature are of both theoretical and practical interest (Miyamoto et al. 1995). Some authors have called thermal changes in absorbency a “smart” feature, since they may allow specialized applications of a thermo-sensitive hydrogel (Pourjavadi et al. 2007b; Yang et al. 2011). Typical temperature-responsive hydrogels exhibit markedly decreased absorbency with increasing temperature (Harsh and Gehrke 1991; Kuwabara and Kubota 1996; Çaykara et al 2006; Chang et al. 2009b). Decreases in absorbency as high as a factor of five (Kuwabara and Kubota 1996) or ten (Çaykara et al 2006) have been reported when comparing absorbency at about 20 oC to the correspond-ding value at about 40 oC. The strongest thermal effects are for SAPs incorporating relatively high proportions of poly(N-isopropylacrylamide); these generally exhibit a strong decrease in absorbency at temperatures near to physiological conditions (Kuwabara and Kubota 1996; Liu et al. 2004; Chang et al. 2009b; Yang et al. 2011). For example, Kuwabara and Kubota (1996) used a two-step procedure to graft a 51/49% acrylic acid/N-isopropylacyrlamide onto CMC at a grafting level of 205%; water absor-bency fell from about 14 g/g at 10 oC to 3 g/g at 50 oC. Çaykara et al. (2006) prepared semi-interpenetrating networks of poly-[(N-tert-butylacrylamide)-co-acrylamide] with hydroxypropylcellulose; water sorption was observed to fall from as high as 19 g/g at 5 oC down to about 2 g/g at 40 oC and above. Chang et al. (2009b) reported a decline in absorbency from 18 g/g to 5 g/g over a temperature rise from 10 oC to 60 oC in the case of an interpenetrating network of one part cellulose to three parts of poly-NiPAM. Likewise, in the case of SAPs comprising carboxymethylchitosan and poly-NiPAM, Ma et al. (2007) observed a decrease in buffer solution absorption from 19 g/g at 25 oC to 2 g/g at 35 oC. An opposite thermal effect, with moderately increased swelling of hydrogels with increasing temperature, has been reported less frequently (Liu et al. 2005a; El-Din et al. 2010). Hirsh and Spontak (2002) showed that the thermo-responsive characteristics can be attributed to changes in the elastic properties of such hydrogels.
External Pressure and Absorbency
In various applications of absorbent materials it may be important to resist the release of water upon compression. In fact, one of the strategies in the design of highly absorbent products is to employ a relatively stiff fiber matrix such that compressive forces acting on SAPs in the product are minimized (Buchholz et al. 2005). According to the same source, as has been noted, the presence of fibers also can restrict swelling. Guilherme et al. (2005) evaluated the stress-strain behavior of hydrogels as a function of temperature. It was shown that higher temperatures led to increased stress values in the cases considered. This was associated with the absorption of water, i.e. osmotic pressure. Sannino et al. (2004b) described the development of a test device capable of comparing the effects of compression on the absorption capacity of SAPs. Based on the results shown, it was clear that even a moderate applied pressure of about 300 kPa was sufficient to reduce the swelling with water by 90% in the case of CMC/HEC hydrogels that had been equilibrated (at different levels of compression) for 24 hours.
ACCOUNTING FOR OBSERVED SORPTION CAPACITIES
As has been shown in previous sections, the amount of aqueous fluid taken up by cellulose-based absorbents can depend strongly on the chemical composition and structure of the absorbent, and also on the characteristics of the liquid. Theoretical explanations that have been proposed to account for absorption capacities will be considered here. Some notable review articles that emphasize theoretical issues relevant to the extent of swelling are available (Scallan 1983; Brannon-Peppas and Peppas 1990; Kazanskii and Dubrovskii 1992; Ganji et al.2010).
Quantification of Sorption Capacities
As a first step in reviewing theories to account for absorption capacities and rates, it is important to consider how absorbency is tested. Some notable advances in measure-ment procedures have been described, as listed in Table 2.
Table 2. Methods for Quantification of Absorption Capacity or Rate
When testing the water sorption capacity of cellulosic fibers, Welo et al. (1952) observed that a simple weight-loss method, with free evaporation under ambient conditions, was easy and repeatable. The materials were first fully swollen with an excess of water, then weighed periodically. It was found, for each specimen, that weight loss was linear with temperature up to a critical point, and thereafter slowed down. The critical point was identified with the loss of all external water, leaving behind only water of swelling. The water at that critical point was used as an approximation of the water-holding capacity of the fibers. More recently, a similar concept was developed using isothermal drying in a thermogravimetric analyzer; the non-external water was named hard-to-remove water (Park et al. 2006).
The Donnan Equilibrium and Osmotic Pressure
When an absorbent material takes up a liquid, either that liquid needs to displace another fluid (such as air), or the material needs to swell to accommodate the new liquid. This section will consider osmotic pressure and various studies showing how osmotic pressure can account for such swelling, especially when a cellulosic material or a hydrogel is placed in water having few or no salt ions (Grignon and Scallan 1980; Vitta et al. 1986; Oppermann 1995; Sannino et al. 2003; Buchholz et al. 2005; Ganji et al. 2010). The same principles also apply to cellulosic fibers, though usually to a lesser extent. Part of the justification for considering cellulosic materials in the context of hydrogels involves the presence of ionizable bound groups, mainly carboxyl groups, within their structure (Herrington and Petzold 1992). Such groups contribute to swelling (Grignon and Scallan 1980; Lindström and Carlsson 1982) and ion exchange phenomena (Rudie et al. 2006; Hubbe et al. 2011) within cellulosic fibers.
One of the most important principles underlying swelling is that, under equilibrium conditions, there will be a fixed relationship between the concentrations of ions inside and outside of a swellable, gel-like material (Grignon and Scallan 1980). In other words a Donnan equilibrium condition will become established between the ion concentrations inside and outside of the hydrogel. If one assumes, for instance, that an aqueous system containing a suspended anionic hydrogel phase contains the ions OH–, H3O+, Cl–, and Na+, then the Donnan equilibrium can be expressed in the form,
where the subscript “g” indicates location within the gel phase and the subscript “s” indicates location within the solution phase, respectively.
A characteristic feature of a typical hydrogel is that ionizable groups are bound to polymer chains within the gel phase. The first analysis of the relationship between such charged groups and swelling was based on ordinary gelatin and the effects of acidity in a surrounding bulk solution (Proctor 1914). Grignon and Scallan (1980) were the first to extend the same concept to the swelling of pulp fibers; they noted that the amorphous polysaccharide regions within pulp fibers can be regarded as being hydrogels. In either case, the tendency of the solid to swell upon placement in deionized water can be visualized as a process whereby water rushes into the hydrogel zones in order to make the value of become closer to one. As water progressively enters the gel phase, the ionic concentration associated with the bound ionizable groups becomes diluted; however, in salt-free systems, such dilution is never enough to eliminate the contrast in ionic strength between the inside and the outside of hydrogel zones.
Once equilibrium has been established, the osmotic pressure tending to swell the hydrogel phases within the mixture can be calculated based on the following equation (Ganji et al. 2010),
where is the osmotic coefficient of the gel phase and is the corresponding quantity for the external aqueous solution. Likewise, is the concentration of the ith species in the gel phase and is the corresponding value in the external solution phase. At equilibrium it is understood that the osmotic pressure will be in balance with elastic forces tending to limit swelling in the material (Grignon and Scallan 1980). The following equation can be used to estimate the osmotic coefficient of the gel phase in Eq. 2 (Ganji et al. 2010),
where np is the molarity of the polymer, nm is the molarity of the monomer, and is the degree of ionization as a fraction.
Entropy of Mixing
A second key factor accounting for the swelling of a hydrogel with water involves the entropy of mixing (Sannino et al. 2003; Ganji et al. 2010). Such effects can be understood based on Eq. 4, for which a negative value of the change in Gibbs Free Energy (G) of a system indicates that a process will occur spontaneously:
G = H – T S (4)
In this expression, H is the change in heat energy (enthalpy), T is the absolute temperature, and S is the change in entropy. The term S is sometimes called a measure of the disorderliness of a system. In principle, energy is required to separate a mixture into pure subphases; by contrast, miscible materials will spontaneously form a single, disorganized phase with the release of energy when placed together.
Entropy can play a significant role in the swelling of hydrogels because a system in which the gel material is infused with water has greater randomness (higher entropy) than the separated (dry) hydrogel and water phases. The contribution to swelling pressure due to the entropy of mixing can be estimated from Eq. 5 (Ganji et al. 2010),
where is the polymer volume fraction, V1 is the molar volume of the solvent, and is the interaction parameter.
Figure 3 illustrates another way in which changes in entropy tend to contribute to the swelling of a typical hydrogel material when it is placed in water (Khokholv and Nyrkova 1992). Here it is assumed that the gel material contains ionizable acid-type groups, such as –COO– or –SO4–, etc. The left-hand frame of the figure represents the gel material in its dry form as the sodium salt of the poly-anionic gel. In such form, the component ions have very limited freedom of motion, being bound by their ionic bonds and the contracted, solidified nature of the gel. As illustrated by the right-hand frame of the figure, the addition of water, along with the swelling of the gel, implies more degrees of freedom. For instance, sodium ions may effectively “switch places” as counter-ions for the bound anionic groups. As indicated in the upper right of the figure, such an exchange even can occur with salt ions from the bulk. Further contributions to disorder and freedom of motion within such a system are associated with chain mobility and the relatively long (in atomic terms) average distance between oppositely charged ions in an aqueous system having very low electrical conductivity (Hiemenz 1977).
Fig. 3. Illustration of increased disorder and freedom of motion of a counter-ion (e.g. sodium) in a typical hydrogel that contains acid groups (e.g. carboxylate). Left: Dry form as sodium salt of polycarboxylate hydrogel; Right: Water-swollen form, in which the counter-ions are free to exchange with each other, to diffuse within the field of electrostatic forces, and also for the polyelectrolyte chain segments to undergo some constrained motion.
Hydration and Hydrogen Bonding Effects
A third key element to account for the swelling of hydrogels, in addition to the osmotic effects and entropy contributions just described, is hydrogen bonding. The relatively strong interactions between water molecules and hydrogel materials promote hydration and ionization, such that the osmotic and entropic effects can operate. The importance of hydrogen bonding relative to hydrogel swelling was perhaps best demonstrated by Koroskenyi and McCarthy (2001), who studied the highly hydrophilic polymer konjac glucomannan. In its native state the gelatinous material absorbed over 105 times its mass of water. By contrast, after full acetylation, removing all of the –OH groups, the same material had a water absorbency of only 1 g/g. Another clue that hydrogen bonding is intimately related to gel behavior comes from infra-red analysis. Sammon et al. (2006) detected distinct shifts in the wavelengths of infrared absorbance maxima that were coincident with gelation of hydroxylpropylmethylcellulose (HPMC) as a function of temperature.
Regarding the state of water within a typical hydrogel, various views have been expressed in the literature. According to Karaaslan et al. (2012), most of the water within a swollen hydrogel should be regarded as bound water. In other words, its properties can be expected to differ from those of bulk water in terms of such behaviors as freezing tendencies and temperatures. Accordingly, Faroongwarng and Sukonrat (2008) observed that the freezing point of water was depressed within hydrogels. Gelin et al. (2007) estimated that only about 10% of water present in a typical hydrogel acts like bulk water. Their work with measurement of ion diffusion rates, in addition to freeze-fracturing of hydrogels, led to the conclusion that the bulk-like water within hydrogels tended to be confined to isolated “lakes” within the pore structure. This finding suggests that various reported porous hydrogels or “foams” might be expected to contain substantial bulk-like water in addition to chemically bound water (Kabra et al. 1998; Omidian et al. 2005; Lu et al. 2007; Chang et al. 2009a,b; Sehaqui et al. 2010; Salam et al. 2011a). Joshi et al. (2008) proposed that the relatively sharp gelation transition that takes place upon heating of methylcellulose solutions can be attributed to a thermal breakdown of the water structure, allowing coalescence of the polymer chains. Venditti and coworkers have demonstrated that freezing and non-freezing bound water and bulk water are present in cellulosic fibers and can be measured with differential scanning calorimetry (Park et al. 2007). The heat of vaporization of the bound water can be almost 30% higher than bulk water. For fibers that cannot participate in hydrogen bonding, such as polypropylene, bound water and increased heat of vaporization of waters do not exist (Park et al. 2007).
Progress in understanding the nature of water within cellulosic materials has been achieved through advances in spectroscopic methods. Froix and Nelson (1975), using nuclear magnetic resonance relaxation times, concluded that free water, in addition to chemically bound water, existed within cellulosic fibers even in “dry” paper in equilibrium with relative humidity values above 0.1. Using NMR and ESR methods, Capitani et al. (1990) came to the conclusion that water domains within cellulose somehow permeate cellulose domains, such that the two kinds of domains interact by hydrogen bonding. Using calorimetric methods, Maloney et al. (1998) found evidence that bulk-like water was present within the larger mesopores of never-dried kraft fibers after their placement in water; by contrast, only bound water was present within the cell walls of previously dried cotton. Park et al. (2006) showed that such considerations can be used to estimate the dimensions of nano-sized water-filled pores within cellulosic materials.
As proposed by Proctor (1914), the extent of swelling of a typical hydrogel in pure water or various electrolyte conditions can be modeled in terms of a balance between osmotic forces and restraining elastic forces due to the swollen material. Scallan and coworkers extended these concepts to cellulosic fibers (Grignon and Scallan 1980; Scallan and Tigerström 1992). In either case, pure water can be expected to diffuse into a hydrogel until a combination of partial dilution of the effective ionic concentration within the gel, plus elastic stretching of the gel achieves a balance of forces (Vitta et al. 1986; Ganji et al. 2010). The viscoelastic properties of various hydrogels, sometimes as a function of their state of swelling, have been studied (Guilherme et al. 2005; Feng et al. 2006; Chang et al. 2009a; Dai and Kadla 2009; Larsson et al. 2010). Notably, Larsson et al. (2011) and Spagnol et al. (2012a) showed that the inclusion of microfibrillated cellulose or cellulose nanocrystals within hydrogels had a strong reinforcing effect, essentially increasing the modulus and limiting the amount of swelling. Yoshinobu et al. (1992) and Lionetto et al. (2003) showed that there was a relationship between the degree of crosslinking and the elastic modulus of various hydrogels. As already noted, Oppermann (1995) proposed that inclusion of cellulose derivatives within a hydrogel can result in stiffening of the gel. Such an effect follows from the inherent stiffness of cellulose, which may be related to local association of cellulose segments.
An elastic contribution to pressure within a hydrogel, resulting from the configurational entropy of polymer chains, can be estimated if one makes various assumptions concerning the nature of the hydrogel. The stretching of polymer segments associated with the swelling of the gel implies a restriction in the number of motions of individual groups and segments along the chain, i.e. a loss of configurational entropy when the hydrogel material is stretched. The lead terms in an equation to estimate the resulting contribution to pressure (tending to resist expansion of the gel) are as follows (Ganji et al. 2010),
where is the concentration of the polymer chains at the onset of gel formation.
Network Volume Constraints
As was noted earlier, the highest recorded values of swelling tend to be associated with the lowest practical (and sometimes impractical) extents of crosslinking (Bin et al. 2000; Wach et al. 2001, 2002; Liu et al. 2005a; Demitri et al. 2008; Han et al. 2010; Pourjavadi et al. 2010; Kono and Fujita 2012). The basis for this relationship is illustrated in Fig. 4. As discussed by Ganji et al. (2010), it is possible to estimate the maximum extent of swelling of a hydrogel based on the density of crosslinking, and more specifically the lengths of the characteristic polymer segments between adjacent crosslink points. The quantity , represented by a typical segment in Fig. 4, is defined as the average or typical molecular mass between adjacent crosslink points. The quantity represents the linear correlation distance between the crosslinks. The volume fraction of the polymer in a swollen gel is then given by the following expression,
v2,s = Vp /Vg = (1/2) / [ (Qm /1) + (1/2) ] (7)
where Vp is the polymer volume within the material, Vg is the volume of the swollen gel phase as a whole, 2 is the density of the polymer, 1 is the density of the solution, and Qm is the swollen volume on a mass basis. The degree of crosslinking, X, can be defined as
where Mo is the characteristic molecular mass of the initial polymer chains before crosslinking. In cases where the polymer-solvent interactions have been tabulated, the typical dimensions of spaces between crosslink points can be estimated from (Ganji et al. 2010),
where l is the bond length between carbon atoms, and Cn is the tabulated ratio for the system, as derived by Flory (1953). It should be noted that while the cited work did not take electrostatic effects into account, such interactions can be expected to profoundly affect polymer conformations (Fleer et al. 1993) and hence swelling.
Fig. 4. Representation of a hydrogel in its somewhat condensed (left) and highly swollen (right) states. Solid dots indicate cross-linked points of the polyelectrolyte chains (shown as coiled or wavy lines). The characteristic molecular weight of segments between crosslinks () and the correlation distance between crosslink points () are illustrated by considering one segment.
Cellulosic Fibers and Hydrogel Swelling
Various authors have suggested that a network of cellulosic fibers, i.e. a paper-like structure or a loose network of nanofibrillated pulp or fluff pulp fibers filled with hydrogel particles, may play a role in maintaining the structural integrity in a combined absorbent structure that contains SAPs (Buchholz et al. 2005; Brodin et al. 2012). It follows that such fibers may also restrict the swelling of the SAP material upon introduction of aqueous fluid. The concept is illustrated schematically in Fig. 5.
Fig. 5. Illustration of the concept that a network of macroscopic cellulosic fibers in an absorbent product might tend to constrain the swelling of hydrogel particles. Left: Representation of dry condition; Right:Swollen condition
As was noted earlier, when nanocellulose elements are incorporated into hydrogels, the observed result is generally a marked decrease in absorbency, often accompanied by an increase in the mechanical strength of the gel material (Dai et al. 2009; Larsson et al. 2010, 2011; Spagnol et al. 2012a). In an analogous manner to the mechanism just described to account for the restraining effects of fluff pulp fibers on SAP swelling, some authors have attributed the effects of nanocellulose on SAP absorbency to a “reinforcement” effect (Larsson et al. 2011). Indeed, when cellulosic nanofibrils or crystallites have been compounded with various plastics, the resulting “nanocomposites” are typically much stiffer than the matrix plastic material by itself (Azizi Samir et al. 2005; Hubbe et al. 2008). However, before placing too much weight on any such analogy, it is important to emphasize the huge size difference between a softwood kraft fiber and the smallest classes of cellulosic nanomaterials. Figure 6 provides a schematic comparison. The lengths of the nanocellulose entities, in some cases, may be as much as 10,000 times smaller than a typical fluff pulp fiber. As shown, one of the likely roles of the nanocellulose may be to function as cross-linking points. It is assumed here that polymer segments within the hydrogel may associate locally with the cellulosic surfaces. Such behavior would be consistent with observations that the effects of adding a crosslinking agent or adding nanocellulose are sometimes nearly equivalent (Dai and Kadla 2009; Larsson et al. 2010).
Fig. 6. Illustration emphasizing the great size difference between fluff pulp fibers vs. cellulosic nanocrystal, a difference that suggests that their roles with respect to reinforcement, crosslinking, or other effects might also be different
Drying and Absorbency
A profound loss of swelling capacity is known to occur when delignified pulp fibers are dried (Stone and Scallan 1966; Weise and Paulapuro 1999). Loss of fiber absorbency and the relative effects of loss of conformability of the wetted fibers, especially when fibers are dried multiple times, is a major contribution to reduced paper strength when the material is recycled (Weise and Paulapuro 1996; Hubbe et al. 2007b).
A unique example of loss of absorbency upon drying was documented by Mizutani et al. (1999), who studied the water absorbency of never-dried cotton fibers starting very soon after their biosynthesis in the bowl of the living plant. At the earliest measurable stage, the never-dried cotton fibers were associated with 2.5 times their mass in water. Upon drying of such fibers, they were then only able to absorb water to an extent of about 0.8 times their mass. The effect was attributed to the fact that in the continuously wet condition, while sealed within the fresh cotton bowl, the water present between the adjacent cellulosic microfibrils in the young cotton prevented those fibrils from coming together and hydrogen bonding directly with each other. As shown by Stone and Scallan (1966, 1968), the loss of absorbency upon drying of cellulose fibers generally can be attributed to a partly irreversible closure of submicroscopic pores. Strong capillary forces acting during the drying process (Campbell 1959) can pull the walls of some pores into close contact, and the regularity of the cellulose structure can result in a zipper-like formation of multiple, repeating hydrogen bonds, essentially causing the adjacent surfaces to coalesce into a single cellulosic entity (Hult et al. 2001; Pönni et al. 2012). Though additional refining has been shown to restore much of the bonding strength that is lost when kraft fibers are dried, the corresponding effects related to restoration of porosity tend to be more limited (Klungness and Caulfield 1982).
Efforts to suppress the loss of absorbency (and other valuable traits such as bonding potential) when kraft fibers are dried have been limited in success (Zhang et al. 2002). Early work by Higgins and McKenzie (1963) showed only one kind of treatment – the drying of cellulose fibers in the presence of high concentrations of sugar – that was effective in avoiding the irreversible loss of swelling potential. Work by Lindström and Carlsson (1982) showed that some of the damage could be avoided by drying under neutral to alkaline pH conditions – but only if the pulp also had a relatively high level of carboxylate groups, i.e. a high hemicellulose content. Subsequent work has confirmed the favorable effects of hemicellulose (Cao et al.1998), which may be attributed not only to its content of carboxyl groups, but also to its somewhat irregular chain structure that tends to resist crystallization. The importance of hemicellulose for swelling of birch kraft pulps was shown dramatically by Moss and Pere (2006). These researchers showed that intentional removal of most of the xylan component from the fibers almost eliminated the ability of such fibers to swell when they were placed back into water.
To avoid the usual deswelling effect of drying, freeze-drying is one strategy that has been successfully employed. Such an approach has been used to prepare cellulose-based “aerogels” that maintain very high porosity, bulk, and absorption capacities (Saito et al. 1995; Pääkkö et al. 2008; Kettunen et al. 2011; Theliander et al. 2011; Brodin et al. 2012). Another successful approach has been to transfer the swollen cellulosic material to an acetone solution before it is dried in air. Sannino et al. (2010) used such an approach to preserve the swollen nature of nanoporous crosslinked cellulose gels when they were dried.
Polyelectrolyte Complexes as a Smart Restraint on Hydrogel Swelling
In view of recent success in achieving highly favorable levels of absorption of saline solution through the use of polyelectrolyte complex hydrogels (Salam et al. 2011b), coupled with the tendency of such complexation to restrain the absorption of pure water (Dogu et al. 2009), this section will consider theoretical aspects of such absorbency. Wood-derived polysaccharides and their derivatives such as xylans and carboxymethylcellulose (CMC) can be readily incorporated into such systems, in combination with various cationic polymers such as chitosan (Salam et al. 2011a,b; Ayoub et al. 2013).
The tendency of polyelectrolyte complexes (PECs) to contain relatively little water within their core structure, under conditions where the amounts of oppositely charged bound ionizable groups are in approximate balance, can be understood based on the mechanism shown in Fig. 7. As shown in the figure, there are bound anionic groups on one set of polyelectrolyte chains and bound cationic groups on the other set. In each case the bound charges are balanced by an equal and opposite number of counter-ions. (Note that this is in addition to the various other monomeric ions of each charge in the system. Such ions have to sum up to equal normality of each charge). Upon mixing the two polyelectrolyte solutions, a complex is formed (Michaels 1965). As shown in the figure, the cooperative interactions involved in multiple ionic interactions between the oppositely charged polyelectrolytes tends to drive the complexation to completion. It has been proposed that complexation might occur in a zipper-like fashion or in a disorganized manner (Michaels 1965). In either case, the resulting ion pairs within the PEC can remain stable, such that the charges within the system are essentially self-neutralized. As illustrated in the figure, one of the results, therefore, is that the counter-ions can diffuse away into the bulk of solution. Some authors have credited the resulting increase in entropy as being the most important factor in promoting PEC formation (Michaels et al. 1965; Cundall et al.1979; Argüelles-Monal et al. 1990; Baker et al. 1995).
Fig. 7. Mechanism of polyelectrolyte complex (PEC) formation in which ion pairing between bound groups on the respected polyacid and polybase chains make it possible for the monomeric counter-ions to diffuse away, with a large increase in entropy of the system.
The tendency of polyelectrolyte complexes and polyampholyte systems to swell with water upon addition of salt has been called the “antipolyelectrolyte effect” (Valencia and Pierola 2007; Dogu et al. 2009; Wang et al. 2010). Because this effect appears to be consistent with the behaviour of the hemicellulose-based PECs just described, some background will be given here. As an underlying explanation for the effect, it has been shown that the addition of salt tends to weaken the electrostatic interactions within a PEC (Dragan and Cristea 2001). Increased absorbency with increasing salinity of aqueous solutions has been observed in several cases in which oppositely charged polyelectrolytes were employed in the preparation of cellulose- or hemicellulose-based hydrogels (Paula et al. 2002; Feng and Pelton 2007; Valencia and Pierola 2007; Wang et al. 2010; Shang et al. 2008; Salam et al. 2011a,b; Gawel et al.2012). Such weakening of the pair-wise associations of ionic groups on the polymer chains allows a partial reversal of the process illustrated in Fig. 7, thus drawing more nomomeric ions, along with some associated water, into the interior of the hydrogel phase.
Fascinating behavior related to the sorption of monovalent salt ions was observed in the course of recent studies with PEC-type hydrogels formed from biopolymers having relatively high charge densities (Ayoub et al. 2013). The ability of such hydrogels to remove sodium and chloride ions from solution, beyond what would have been predicted from the concentration of those ions in solution, was attributed to the process described in Eq. 10,
RCOO–—+NH4R’ + NaCl RCOO– Na+ + R’NH4+ Cl– (10)
where the dashed line is used to indicate ion-pair interactions within the hydrogel before introduction of the saline solution. The “R” and “ R’ ” indicate the two respective polyelectrolyte chains. As shown in Eq. 10, addition of NaCl to the material is expected to result in weakening of the complex such that the salt ions behave as counter-ions to the charged polyelectrolyte chains. The effectiveness of the PEC-type hydrogels to enable the removal of NaCl from solution is notable because most reported applications of cellulosic-based materials for ion exchange have involved removal of divalent metal ions (Rudie et al. 2006; Hubbe et al. 2011; Kumar et al. 2011). In such applications it is typical for the divalent ions to displace monovalent ions within the substrate due to their greater binding constants. The process represented by Eq. 10 is unique due to the fact that PEC-based hydrogels can be readily obtained in almost monomer-free condition; upon precipitation of the PEC during its formation, it is possible to decant the salt-containing solution, optionally replacing it with deionized water. Subsequently, upon introduction of a sufficiently concentrated solution of monovalent salt, the PEC interactions becomes weakened to such an extent that Eq. 10 becomes favorable.
ACCOUNTING FOR OBSERVED SORPTION RATES
In addition to the absorption capacity, another main concern of users of absorbent products is the rate of absorption. Several reviews dealing with absorption rates have appeared. In the case of ordinary paper products (not superabsorbent), several key studies have focused on theoretical issues (Bristow 1967; Aspler et al. 1987; Schuchardt and Berg 1991). Related numerical simulation work has been reported (Masoodi et al. 2012). The following articles have considered rates of absorption in the case of hydrogels (Kabra and Gehrke 1994; Zohariaan-Mehr et al. 2010; Wang et al. 2011c). Finally, Berg and coauthors have considered the rates of wicking into structures that are contained by fibers and hydrogel materials (Schuchardt and Berg 1991; Wiryana and Berg 1991).
Tests to Measure Absorption Rates
Practical tests of rates of absorbency of cellulose-based materials often involve rapid immersion of material packed into a cage, followed by evaluation of weight gain (Aberson 1969). The cited author noted that the results of such tests can be difficult to interpret. To overcome this problem, a new test was developed in which dry fluff pulp is exposed only on one side (Aberson 1969). As was noted in Table 2 (in previous section), rates of sorption can also be studied by dielectric spectroscopy (Gelin et al. 2007), wicking rates (Beuther et al. 2010), and the time for a water drop to be absorbed (Ström and Carlsson 1992).
For ordinary paper and paperboard products, various TAPPI methods are available (TAPPI 1996). For instance, TAPPI Method T411 describes the gravimetric Cobb test in which the sample is exposed to the liquid from one side for a set period (e.g. 120 s), after which the liquid is poured off and a blotter is used to remove excess liquid. TAPPI Method 491 is a related test in which the whole sample is submerged in the test liquid for a specified period (e.g. 10 min), and the sample is blotted with a standard roller before being reweighed. TAPPI Method T432 is intended for highly absorbent paper products such as blotter paper, tissue, and towel. A specified volume of liquid is gently placed on the sample, and one measures the amount of time that passes before the disappearance of specular reflection from the smooth surface of the liquid. TAPPI Method 561 is a dynamic test in which one continuously monitors the mass of liquid remaining in a reservoir attached to a siphon tube in contact with the paper specimen. Related tests have been adopted for evaluation of textiles (ASTM 1983). For instance, test TS-017 is based on the time in second it takes distilled water to climb either 20 mm or 30 mm up a vertical strip of cloth that has been dipped into a beaker. Test TS-018 measures the time for complete sorption and disappearance of a drop of water having a specified size.
For evaluation of events during very short exposure times of paper samples and aqueous solutions, two main procedures have been developed. According to a test developed by Bristow (1967), a narrow strip of paper is mounted on a slowly rotating wheel. A “headbox” with a precisely measured volume of colored fluid and a narrow slit opening at its base is lowered until it touches the sample. The length of the colored streak transferred to the paper can be used to calculate the volumetric uptake per unit of contact time. Related results have been obtained more recently with a device that measures the speed of transmission of ultrasonic vibrations through a water bath as a function of the time after sudden placement of a sheet of paper perpendicular to those waves (Pan et al. 1988; Leks-Stepien and Khadzhynova 2005).
Viscous Resistance Limiting Rates of Flow through Porous Material
A key factor that tends to reduce the rate with which an aqueous fluid passes into cellulosic-based materials, including hydrogels, is the fluid’s viscosity. The magnitude of viscous resistance also depends on the sizes and lengths of the passages through which the fluid needs to flow. These relationships have been captured in Darcy’s law and its various refinements by subsequent investigators (Darcy 1856; Masoodi and Pillai 2010). In its simple form, Darcy’s law can be expressed as in Eq. 11,
where Q is the flow rate through a porous packed bed of uniform cross-section, k is a permeability constant, A is the cross-sectional area of the bed, Pb is the outlet pressure, Pa is the inlet pressure, is the fluid’s dynamic viscosity, and L is the linear distance from the inlet to the outlet.
When applying Darcy’s law to predict the rate of wicking in paper, Masoodi and Pillai (2010) noted that the model tended to work best if it was assumed that the material swelled in the course of wetting in order to fully make room for the liquid. The permeability of initially dry paper appears to vary as a function of time during the wicking of aqueous fluid.
Various approaches can be used to estimate the value of the coefficient k in the case of paper-like structures (Kozeny 1927; Carman 1938, 1939; Carrier 2002; Chapuis and Auberton 2003; Hubbe and Heitmann 2007). For instance, a simple laminar hydrodynamic model was used to derive the following expression,
where k is the permeability (distance of penetration per unit of time), is a unit mass of the fluid, is the fluid’s dynamic viscosity, CK-C is the Kozeny-Carman coefficient (usually taken to be about 5), So is the specific surface area per unit displacement volume of particulate material, and e is the fractional void volume (Carrier 2002; Chapuis and Auberton 2003). Ingmanson (1952, 1953) applied similar concepts to the case of cellulosic fibers, taking into consideration the bulky structures formed by fibers, but also considering the compressibility of cellulose. In general, all such models imply that a given applied pressure is likely to result in greater flow with increasing pore size and/or decreasing effective surface area of the packing material. This general prediction has been confirmed, for instance, by studies in which the effective pore size within paper structures was increased by forming the paper in the presence of surfactants (Shepherd and Xiao 1999).
Another approach that has been used to model the kinetics of swelling of hydrogels is based on a visco-elastic model (Omidian et al. 1998). Variation of some parameters that can improve the superabsorbent properties can lead to considerable differences in swelling kinetics, which can be fitted to a Voigt-cell (spring and dashpot) model, as
St= p(1 – e-1/r) (13)
where St is the amount of swelling (g/g) at any moment, p is the power parameter (g/g), t is time (s), and r is the rate parameter. Swelling at different times can be measured in distilled water. Then, the results obtained at equilibrium and the swelling rate of the individual samples can be fitted into the above equation to find the values of the rate and power parameters. Another useful parameter, based on the same model, is the time required to reach 63% of full capacity (Omidian et al. 1998).
Capillarity and the Lucas-Washburn Equation
Capillary forces, due to the surface tension of water, are known to play a major role in the absorption of water and other fluids into porous materials (Buchholz et al. 2005). Lucas (1918) and Washburn (1921) showed that capillary forces can account for the rise of liquids within straws and related porous systems. It has been suggested that the lumens of cellulosic fibers can play a similar role in the wicking of water (Mizutani et al. 1999). Capillary forces also have been credited with explaining the rapid absorption of aqueous fluids into some highly porous hydrogels (Omidian et al. 2005). Sannino et al. (2003) proposed that capillary forces also can increase the amount of water held within swollen hydrogels. In addition, it is reasonable to expect also that capillary forces are involved in cases where fibers having irregular cross-sections are designed to provide channels for wicking within SAP media (Schuchardt and Berg 1991; Pramanick and Crouse 1998).
Figure 8 illustrates some essential features of the Lucas-Washburn model of capillary absorption. As shown, the capillaries are represented as smooth, cylindrical pores of a selected size. In the simplest case one assumes that the material is comprised of open-cell pores, such that the pressure of “Air” shown in the figure is always equal to the ambient pressure of the bulk liquid. As indicated by Eq. 14, the capillary pressure exerted by the interfacial tension at the meniscus is balanced by a viscous force due to the flow within tight passages as water continues to progress into the pore.
The integrated form of the Lucas-Washburn equation, Eq. 15, indicates that the distance of penetration into the porous material can be expected to be proportional to the square-root of time. Another implication of Eq. 14 is that the rate of absorption is expected to increase with increasing pore size. This prediction is reflected in the more rapid absorption of water into coarse, bulky paper (Aberson 1969).