Cellulose, considered as one of the most abundant renewable polymeric material (Brinchi et al. 2013), is naturally organized as microfibrils linked together to form cellulose fibres, in which every single filament consists of several cells of cellulose- based crystalline microfibrils connected by lignin and hemicellulose (Siqueira et al. 2010) . Cellulose consists of p-D — glucopyranose units linked together by (3-1-4- linkages, containing hydroxyl groups able to form hydrogen bonds, playing there­fore a major role in directing the crystalline packing and also governing the physical properties of cellulose (John and Thomas 2008).

Cellulose cannot be considered uniformly crystalline, but ordered regions dis­tributed throughout the material, called crystallites, can be found in its structure. The linear association of these components is a microfibril, and can be considered the elementary unit of the plant cell wall. These microfibrils are found to be 10-30 nm wide, indefinitely long, and they can contain approximately 2-30,000 cellulose molecules in cross section. Individual cellulose nanocrystals (CNC) and cellulose microfibres (Fig. 11.4) can be isolated from crystalline cellulose core breaking down these crystalline regions (Oke 2010; Kalia et al. 2011).

There is a wide range of cellulose-based particle that can be considered different depending on cellulose source and extraction processes. Two main families of nano­sized cellulosic particles can be found: the first one consists of CNC or nanowhis­kers (CNW) and the second one is the microfibrillated cellulose (MFC) (Belgacem and Gandini 2008; Lu et al. 2008; Brinchi et al. 2013). In this section, our attention

Fig. 11.4 Scheme of main steps needed to prepare CNC from natural sources

will be given to the extraction of CNC from natural sources, their specific properties and final application in polymeric matrices. CNC can be extracted from a variety of cellulosic sources, such as plants, bacteria and algae. The extended literature on this subject confirms that commonly studied source materials are represented by wood, plants, microcrystalline cellulose or bleached pulp because of their extensive avail­ability and high content of cellulose. Essentially, CNC can be easily prepared from commercial microcrystalline cellulose or from filter paper, because of their purity and availability in laboratories (Klemm et al. 2011) and they can be isolated using a two — stage procedure, as shown in Fig. 11.4. The first one is a pretreatment of the source material, in which complete or partial removal of hemicelluloses and lignin, with consequent isolation of the cellulosic fibres, is obtained. The second treatment— generally a chemical hydrolysis—is able to remove the amorphous regions of the cellulose polymer, thus yielding a highly crystalline structure.

Azizi Samir et al. (2005) reported that even high-purity cellulose crystals starting from crystalline domains can be obtained, and together with these a non-crystalline state (amorphous) of the cellulose can be found. These cellulose amorphous regions are randomly oriented in a spaghetti-like arrangement having a lower den­sity if compared to nanocrystalline regions (de Souza Lima and Borsali 2004; Saxena and Brown 2005). These amorphous regions are easily subjected to acid attack and they can be removed leaving intact crystalline regions (de Souza Lima and Borsali 2004) . De Souza Lima and Borsali (2004) described how cellulose amorphous regions can be disrupted for the production of CNC. Going more into detail, hydronium ions can easily penetrate these amorphous domains, promoting the hydrolytic cleavage of the glycosidic bonds and isolating individual crystallites. Dong et al. (1998) firstly studied the effect of hydrolysis conditions on the proper­ties of resulting CNC. They have shown that longer hydrolysis times lead to shorter monocrystals with increased surface charge. Indeed, Beck-Candanedo et al. (2005) studied the properties of CNC obtained by hydrolysis of softwood and hardwood pulps. They considered the influence of hydrolysis time and acid-to-pulp ratio in order to obtain CNC. From their analysis, they explained how reaction time is cer­tainly one of the most important parameters to be considered for the acid hydrolysis of wood pulp, since too long reaction times can completely depolymerize the cel­lulose, up to yielding its component sugar molecules.

In contrast, lower reaction times will only yield large fibres and aggregates that cannot be easily dispersed. Araki et al. (1998) compared the effects of using both sulfuric acid and hydrochloric acid cellulosic nanocrystals (CNC), demonstrating that sulfuric acid is able to provide more stable aqueous suspensions compared to hydrochloric acid. Recently, CNC with an acicular structure ranged from 100 to 200 nm in length and 15 nm in width were extracted from Phormium tenax leaf fibres by acid hydrolysis (Fortunati et al. 2012d): in the cited paper, CNC extraction pro­cess was studied in terms of yield, thermal and chemical properties of the obtained nanocrystals and the results coming from the analysis of Phormium fibres were com­pared with those obtained extracting crystals from two reference materials, Flax of the Belinka variety and commercial microcrystalline cellulose (Fortunati et al. 2013a, b). Morphological, thermal and chemical characterization of the obtained CNC from different plant sources confirmed that natural fibres offered high levels of extraction efficiency if compared with commercial sources of nanocellulose. On the basis of these results, Fortunati et al. (2013c) extracted from the bast, for the first time in the case of okra (Abelmoschus Esculentus) natural fibre, cellulose micro — and nano­fibres, with a view to obtaining cellulose structures with a high crystallinity and thermal stability. Previously obtained results on thermal and mechanical behaviour of okra fibre indicated it as a possible candidate for use in the production of biode­gradable composites (De Rosa et al. 2010a); however, the use of okra bast fibres in textiles presents a number of drawbacks such as limited rub resistance, scarce colour fastness, sensitivity to wear and it is very much prone to creasing, possibly because of high degree of orientation of cellulose in the fibre. Furthermore, these studies demonstrated that commonly applied chemical treatments, such as bleaching and alkalization, do not substantially generate any improvement to okra fibre properties, indicating that high variability of their mechanical properties limits their use in com­posites (Moniruzzaman et al. 2009; De Rosa et al. 2011).

The possibility of using okra fibres as a source of CNC may be a viable alter­native to employing them in the form of technical fibres. Fortunati et al. (2013d) demonstrated that hydrolysis parameters already applied for extraction of cellulosic fraction starting from microcrystalline material (Fortunati et al. 2012a, b, c) are adequately suitable for hydrolysis starting from a macrofibre, such as okra. In the specific case, the extraction of cellulose was carried out in a two-step procedure in which the holocellulose produced by the action of a first chemical treatment was then exposed to the action of sulphuric acid for further hydrolysis, allowing obtaining CNC in an aqueous suspension.

The use of cellulose nanostructures as a reinforcing phase in nanocomposites has numerous well-known advantages, e. g. low density, renewable nature, a wide variety of filler available through the world, low energy consumption, high specific properties, modest abrasion during processing, biodegradability, relatively reactive surface, useful for the grafting of specific groups (Siqueira et al. 2010). However, cellulose nanoparticles present some disadvantages, for instance, high moisture absorption, poor wettability and incompatibility with most polymeric matrices and limitation in the processing temperature. In fact, lignocellulosic materials start to degrade near 220 °C restricting the matrix types that can be used in association with natural fillers (Wambua et al. 2003). Several reviews have been written on these topics, demonstrating that CNC can be successfully used as filler in nanocomposites, improving mechanical and barrier properties of the matrix (Klemm et al. 2005, 2011; Hamad 2006; Dufresne 2008, 2010; Hubbe et al. 2008; Eichhorn et al. 2010; Habibi et al. 2010; Visakh and Thomas 2010; Siqueira et al. 2010; Duran et al. 2012).

New biopolymer nanocomposites using poly(vinyl alcohol) (PVA) as matrix and reinforced with CNC extracted from okra bast fibres were produced for the first time by Fortunati et al. (2013d). The partial aggregation of cellulose in PVA appears to demonstrate that a good level of compatibility between the hydrophilic crystalline nanocellulose and the polymer matrix was obtained. It has been demonstrated that CNC are able to increase the degree of crystallinity of PVA matrix, specifically the nanocomposites containing 5 wt% of cellulose appeared the most suitable formula­tion, with an increase of 40 % in crystallinity value and of 150 % in the elongation at break with respect to the PVA matrix. The result can be justified considering that this content may lead the available surface area of cellulose to its optimal dimension in the nanocomposite. Moreover, the obtained results showed how okra fibres, and in general bast herbaceous systems, can be applied in the CNC form in nanocom­posite formulations to be used in some industrial areas, such as packaging.