26.5.1 Lignocellulosic biomass

Pectins

Pectins are water-soluble anionic heteropolysaccharides composed mainly of (1 → 4)-α-D-galactopyranosyluronic acid units, in which some carboxyl groups of galacturonic acid are esterified with methanol (Fig. 26.5). They are extracted from citrus peels and apple pomace by hot dilute mineral acid. Short hydrolysis times produce pectinic acids and high-methoxyl pectins (HMP), while extended acid treatment de-esterifies the methyl esters to pectic acids and generates low-methoxyl pectins (LMP). Commercial pectins are categorized according to their degree of esterification (DE), defined as the ratio of esterified to total galacturonic acid groups (Sriamornsak, 2003). HMP have a DE > 50 (usually > 69), whereas LMP have a DE < 50 (Farris et al., 2009). The ratio of esterified to non-esterified galacturonic acid determines the behaviour of pectin in food applications, since it affects solubility and gelation properties of pectin. HMP form gels with sugar and acid, whereas LMP form gels in the presence of divalent cations such as Ca^2 +, which links adjacent LMP chains via ionic interactions, forming a tridimensional network (Janjarasskul and Krochta, 2010).

Pectin films, similarly to other polysaccharide films, have poor water resistance, and have been proposed for potential industrial uses where water binding either is not a problem or can provide specific advantages, such as edible bags for soup ingredients (Fishman et al., 2000). LMP films, on the other hand, when crosslinked with calcium ions, have not only improved water resistance, but also improved mechanical and barrier properties (Kang et al., 2005). Moreover, the presence of carboxyl groups carrying a negative charge at pH > pK_a enables exploiting electrostatic interactions of pectin with positively charged counterparts (Farris et al., 2009).

Lignin

Lignin is the second most abundant terrestrial biopolymer after cellulose, accounting for approximately 30% of the organic carbon in the biosphere (Boerjan et al., 2003). It is associated with cellulose and hemicelluloses in plant cell walls, and is an abundant waste product in the pulp and paper industry.

Lignins are complex aromatic heteropolymers derived mainly from three hydroxycinnamyl alcohol monomers (monolignols) differing in their degree of methoxylation (Fig. 26.6): p-coumaryl, coniferyl and sinapyl alcohols (Boerjan et al., 2003; Buranov and Mazza, 2008).

Lignin has some interesting properties to be used for packaging films, such as its small particle size, hydrophobicity and ability to form stable mixtures (Park et al., 2008). Moreover, lignins have been shown to have efficient antibacterial and antioxidant properties (Ugartondo et al., 2009). Lignin has been used as a film component in composites with gelatin (Núñez-Flores et al., 2013; Ojagh et al., 2011; Vengal and Srikumar, 2005), starch (Baumberger et al., 1998; Vengal and Srikumar, 2005) and chitosan (Chen et al., 2009). Baumberger et al. (1998) observed that starch films incorporated with up to 30% lignin presented higher elongation and water resistance than control starch films. On the other hand, lignin impaired the tensile strength of the films at high relative humidity (71%), reflecting the incompatibility between the hydrophilic starch and the hydrophobic lignin, which was bolstered by the water. Indeed, microscopic observations confirmed that the material consisted of two phases – a hydrophilic starch matrix filled with hydrophobic lignin aggregates. On the other hand, Chen et al. (2009) reported a good dispersion of lignin (up to 20%) in a chitosan matrix, evidenced by SEM, which was corroborated by FTIR results, indicating the existence of hydrogen bonding between chitosan and lignin. An FTIR spectroscopy study on gelatin-lignin films revealed strong protein conformational changes induced by lignin, producing a plasticizing effect, which was reflected in the mechanical and thermal properties (Núñez-Flores et al., 2013).

A drawback from incorporating lignin in films is that they acquire a brownish colour (Mishra et al., 2007; Núñez-Flores et al., 2013). On the other hand, they acquire better light barrier properties (Núñez-Flores et al., 2013), which could be of interest in food applications when ultraviolet-induced lipid oxidation is a problem.

26.5.2 Polysaccharides (other than cell wall polysaccharides)

Polysaccharides are hydrophilic polymers and therefore exhibit very low moisture barrier properties. They are of prime interest as matrices for biodegradable film formation because of their availability and rather low cost.

Most polysaccharides are neutral, although some gums are negatively charged. As a consequence of the large number of hydroxyl and other polar groups in their structure, hydrogen bonds play important roles in film formation and characteristics. Negatively charged gums, such as alginate, pectin and carboxymethyl cellulose, tend to present some different properties depending on the pH (Han and Gennadios, 2005).

Polysaccharide films are usually formed by disrupting interactions among polymer segments during coacervation and forming new inter-molecular hydrophilic and hydrogen bonds upon evaporation of the solvent (Janjarasskul and Krochta, 2010). Because of their hydrophilicity, polysaccharide films provide a good barrier to CO₂ and O₂, hence they retard respiration and ripening of fruits (Cha and Chinnan, 2004). On the other hand, similarly to other hydrophilic materials, their high polarity determines their poor barrier to water vapour (Park and Chinnan, 1995) as well as their sensitivity to moisture, which may affect their functional properties (Janjarasskul and Krochta, 2010).

Starches

Starches are polymers of D-glucopyranosyl, consisting of a mixture of the predominantly linear amylose and the highly branched amylopectin (Fig. 26.7). Native starch molecules arrange themselves in semi-crystalline granules in which amylose and amylopectin are linked by hydrogen bonding. When heat is applied to native starch in the presence of plasticizers such as water and glycerol, the granules swell and hydrate, which triggers the gelatinization process, characterized by loss of crystallinity and molecular order, followed by a dramatic increase in viscosity (Kramer, 2009). This transformation is named gelatinization and leads to the so-called thermo-plastic starch (TPS) (Huneault and Li, 2007).

Amylose responds to the film-forming capacity of starches, since linear chains of amylose in solution tend to interact by hydrogen bonds and, consequently, amylose gels and films are stiff, cohesive and relatively strong. On the other hand, amylopectin films are brittle and non-continuous, since branched amylopectin chains in solution present little tendency to interact (Peressini et al., 2003).

The first studies using starch in biodegradable food packaging were focused on substituting part of the synthetic matrix (usually polyethylene) by starch, but there were difficulties ascribed to chemical incompatibility between the polymers. Recently, studies on pure starch-based materials have predominated, usually focusing on two major drawbacks related to application of starch films. The first one is that native starch commonly exists as granules with about 15–45% crystallinity, and starch-based materials are susceptible to ageing and starch re-crystallization (retrogradation) (Forssell et al., 1999; Ma et al., 2006), which makes starch rigid and brittle during long-term storage, restricting its applications (Huang et al., 2005). The second is their high hydrophilicity, causing its barrier properties to decrease with increasing relative humidity. Plasticizers are used to overcome the first drawback and improve material flexibility (Moates et al., 2001; Peressini et al., 2003; Mali et al., 2004), but, since they are usually highly hydrophilic, presenting hydroxyl groups capable of interacting with water by hydrogen bonds, they tend to increase the moisture affinity of the films (Mali et al., 2005).

Some companies have commercially produced starch-based packaging materials, such as Eco-Go (Thailand), Plantic (Australia) in a joint venture with Du Pont (USA), JMP (Australia), StarchTech (USA), Biome (UK), and BASF (Germany).

Alginates

Alginates, which are extracted from brown seaweeds, are salts of alginic acid, a linear co-polymer of D-mannuronic and L-guluronic acid monomers (Fig. 26.8), containing homogeneous poly-mannuronic and poly-guluronic acid blocks (M and G blocks, respectively) and MG blocks containing both uronic acids. The presence of carboxyl groups in each constituent residue (Ikeda et al., 2000) enables sodium alginate to crosslink with di- or trivalent metal cations, especially calcium ions (Ca^2 +), to produce strong gels or films (Cha and Chinnan, 2004). Calcium ions pull alginate chains together via ionic interactions, after which interchain hydrogen bonding occurs (Kester and Fennema, 1986). Films can be formed either from evaporating water from an alginate gel or by a two-step procedure involving drying of alginate solution followed by treatment with a calcium salt solution to induce crosslinking (Janjarasskul and Krochta, 2010). The strength and permeability of films may be altered by changing calcium concentration and temperature, among other factors (Kester and Fennema, 1986). Alginate films have been studied as edible coatings to be applied to a variety of foods such as fruits/vegetables (Fan et al., 2009; Fayaz et al., 2009) and meat products (Marcos et al., 2008; Chidanandaiah et al., 2009).

Chitosan

Chitosan is a linear polysaccharide consisting of β-(1 → 4)-linked residues of N-acetyl-2-amino-2-deoxy-D-glucose (glucosamine) and 2-amino-2-deoxy-D-glucose (N-acetyl-glucosamine) (Fig. 26.9). It is produced from partial deacetylation of chitin, which is considered as the second most abundant polysaccharide in nature after cellulose (Dutta et al., 2009; Aranaz et al., 2010). Chitin is present in the exoskeleton of crustacea and insects, and can also be found in the cell wall of certain groups of fungi, particularly zygomycetes (Chatterjee et al., 2005). It is usually extracted from crab and shrimp shells as a byproduct of the seafood industry. Since the deacetylation of chitin is usually incomplete, chitosan is a copolymer comprising D-glucosamine and N-acetyl-D-glucosamine with various fractions of acetylated units (Aranaz et al., 2010).

Chitosan is soluble in diluted aqueous acidic solutions due to the protonation of –NH₂ groups at the C2 position (Aranaz et al., 2010). The cationic character confers unique properties to the polymer, such as antimicrobial activity and the ability to carry and slow-release functional ingredients (Coma et al., 2002). The charge density depends on the degree of deacetylation as well as the pH. Quaternization of the nitrogen atoms of amino groups has been a usual chitosan modification, whose objective is to introduce permanent positive charges along the polymer chains, providing the molecule with a cationic character independent of the aqueous medium pH (Curti et al., 2003; Aranaz et al., 2010).

Chitosan films have been proven effective in extending the shelf life of fruits (Hernández-Muñoz et al., 2006; Chien et al., 2007; Lin et al., 2011; Ali et al., 2011) and to have retarded microbial growth on fruit surfaces (Hernández-Muñoz et al., 2006; Chien et al., 2007; Campaniello et al., 2008). The polycationic structure of chitosan probably interacts with the predominantly anionic components (lipopolysaccharides, proteins) of microbial cell membranes, especially Gram-negative bacteria (Helander et al., 2001).

26.5.3 Proteins

Proteins are widely available as biomass byproducts from plants (wheat gluten, maize zein, soybean proteins) and animals (collagen, gelatin, keratin, casein, whey proteins).

Proteins are linear, random copolymers built from up to 20 different monomers. The main mechanism of formation of protein films involves denaturation of the protein initiated by heat, solvents, or change in pH, followed by association of peptide chains through new intermolecular interactions (Janjarasskul and Krochta, 2010).

Proteins are distinguished from polysaccharides in that they have approximately 20 different amino acid monomers, rather than just a few or even one monomer, such as glucose in cellulose and starch. The amino acids are similar in that they contain an amino group (–NH₂) and a carboxyl group (–COOH) attached to a central carbon atom, but each amino acid has unique properties conferred by a different side group attached to the central carbon, which can be non-polar, polar uncharged, or polar (positively or negatively) charged at pH 7 (Cheftel et al., 1985; Krochta, 2002). While hydroxyl is the only reactive group in polysaccharides, proteins may be involved in several possible interactions and chemical reactions (Hernandez-Izquierdo and Krochta, 2008), such as chemical reactions through covalent (peptide and disulfide) bonds and non-covalent (ionic, hydrogen, and van der Waals) interactions. Moreover, hydrophobic interactions may occur between non-polar groups of amino acid chains (Kokini et al., 1994). In addition, protein-based films are considered to have high UV barrier properties, owing to their high content of aromatic amino acids which absorb UV light (Mu et al., 2012).

The most unique properties of proteins compared to other film-forming materials are heat denaturation, electrostatic charges, and amphiphilic character. Protein conformation can be affected by many factors, such as charge density and hydrophilic–hydrophobic balance (Han and Gennadios, 2005). Proteins have good film-forming properties and good adherence to hydrophilic surfaces. Protein-derived films provide good barriers to O₂ and CO₂ but not to water (Cha and Chinnan, 2004). Their barrier and mechanical properties are impaired by moisture owing to their inherent hydrophilic nature (Janjarasskul and Krochta, 2010).

The processing of protein films or other protein-based materials mostly requires three main steps: breaking of intermolecular bonds (non-covalent and covalent, if necessary) by chemical or physical rupturing agents; arranging and orienting polymer chains in the desired conformation; and allowing the formation of new intermolecular bonds and interactions stabilizing the film network (Jerez et al., 2007). Globular proteins are required to unfold and realign before a new three-dimensional network can be formed, and stabilized by new inter- and intra-molecular interactions (Verbeek and van den Berg, 2010).

A material to be extruded receives a considerable amount of mechanical energy, which may affect the characteristics of the final products. Extrusion requires the formation of a melt, which implies processing the protein above its softening point. Proteins have many different functional groups, and consequently a great variety of possible chain interactions that reduce molecular mobility and increase viscosity, resulting in a softening temperature which is often above decomposition temperature. Plasticizers can be useful to reduce the softening temperature, but protein extrusion is usually only possible within a limited range of processing conditions (Verbeek and van den Berg, 2010).

26.5.4 Lipids

Unlike other macromolecules, lipids are not biopolymers, being unable to form cohesive, self-supporting films for packaging. So they are used either as coatings directly applied to food to provide a moisture barrier, or as components in stand-alone composite emulsion films, in which proteins or polysaccharides provide structural integrity and lipids respond for hydrophobic character, improving the water vapour barrier (Krochta, 2002). Moreover, the presence of lipids in composite formulations provides an appealing glassy finish to the material.

Polarity of lipids depends on the distribution of chemical groups, the length of aliphatic chains and presence and degree of unsaturation (Morillon et al., 2002). Lipids with longer chains and lower unsaturation and branching degrees have better barrier against water vapour, because of their lower polarity (Rhim and Shellhammer, 2005; Janjarasskul and Krochta, 2010). On the other hand, more polar lipids have more affinity for proteins and polysaccharides when in emulsions, producing more homogeneous films and avoiding the separation of a lipid phase (Fabra et al., 2011b).

Several types of lipids can be used for food coating or as film components, such as waxes, triglycerides, as well as di- and monoglycerides. Waxes, which are esters of long-chain aliphatic acids with long-chain aliphatic alcohols (Rhim and Shellhammer, 2005), are especially resistant to water diffusion, because of their very low content of polar groups (Kester and Fennema, 1986) and their high content in long-chain fatty alcohols and alkanes (Morillon et al., 2002). There are a variety of naturally occurring waxes, derived from vegetables (e.g., carnauba, candelilla and sugar cane waxes), minerals (e.g., paraffin and microcrystalline waxes), or animals – including insects (e.g., beeswax, lanolin and wool grease) (Rhim and Shellhammer, 2005). Some studies reported extension of shelf life of fruits resulting from application of wax-based coatings, including carnauba wax (Dang et al., 2008; Gonçalves et al., 2010) and candelilla wax (Saucedo-Pompa et al., 2009).

Most commonly, lipids have been used as hydrophobic components of emulsion films. Lipid-polysaccharide or lipid-protein emulsion films combine the complementary advantages of each component. Polysaccharides and proteins act as matrices, since they have the mechanical properties to form self-supporting films, and have good barrier against gases such as O₂ and CO₂, while lipids are useful to reduce water vapour permeability. García et al. (2000) demonstrated that lipid addition to starch films decreased their crystalline-amorphous ratio, which is expected to increase film diffusivity and consequently permeability. On the other hand, lipid addition also reduces the hydrophilic–hydrophobic ratio of films, which decreases their water solubility and therefore water vapour permeability. Emulsion films have been reported to enhance stability of fresh-cut (Chiumarelli and Hubinger, 2012) and whole fruits (Bosquez-Molina et al., 2003; Maftoonazad et al., 2007; Navarro-Tarazaga et al., 2011), vegetables (Conforti and Zinck, 2002), nuts (Mehyar et al., 2012), eggs (Wardy et al., 2011) and bakery products (Bravin et al., 2006).

26.6.1 Polyols

Glycerol can be produced either by microbial fermentation or synthesized chemically from petrochemical feedstock. Additionally, glycerol is a major byproduct of the increasing biodiesel production, thus creating a significant surplus resulting in a sharp decrease in glycerol prices. In this context, application of glycerol for value-added products is a necessity as well as an opportunity for the biodiesel industry (Johnson and Taconi, 2007; Yang et al., 2012). The use of glycerol as plasticizer in biopolymer-based films can be a way to help solve the existing surplus of this byproduct from biodiesel production.

Glycerol as well as other polyols, including sorbitol, propylene glycol, and polypropylene glycol, have great affinity for polysaccharide and protein films, because of their hydrophilicity. Several studies have demonstrated the effectiveness of polyols as plasticizers for polysaccharide-based and protein-based films and coatings (Bergo and Sobral, 2007; Jouki et al., 2013; Ramos et al., 2013;Tapia-Blácido et al., 2013). Qiao et al. (2011) used polyol mixtures including mixtures of glycerol and higher molecular weight polyols (HP) such as xylitol, sorbitol and maltitol. The increase of the molecular weight and the content of HP in the polyol mixture enhanced the thermal stability and mechanical strength of the resulting materials.

26.6.2 Mono- and disaccharides

It is not only molecular size, configuration and total number of functional hydroxyl groups that are important characteristics to be considered for an effective plasticizer, but also its compatibility with the film-forming polymer. Polymer–plasticizer compatibility is necessary to generate a homogeneous mixture without phase separation. It has been suggested that some monosaccharides work as plasticizers in polysaccharide films more effectively than polyols, because of the structure similarity between monosaccharides and polysaccharides (Zhang and Han, 2006). Indeed, Zhang and Han (2006) observed that starch films plasticized with monosaccharides (glucose, mannose and fructose) presented better overall physical properties when compared to starch films plasticized with polyols (glycerol and sorbitol). On the other hand, polyols (especially glycerol) presented better ability to lower T_g of the films, indicating that polyols are more effective in terms of thermomechanical properties. Other studies have indicated monosaccharides and disaccharides as effective plasticizers in different polysaccharide-based films (Olivas and Barbosa-Cánovas, 2008; Piermaria et al., 2011).Whey protein films plasticized with sucrose presented excellent oxygen barrier properties, but sucrose tended to crystallize with time (Dangaran and Krochta, 2007).

26.6.3 Lipids

Generally, the purpose of adding lipids to films is to reduce their water vapour permeability and/or to provide an attractive gloss. Moreover, incorporating lipids in protein- or polysaccharide-based films may interfere with polymer chain-to-chain interactions and provide flexible domains within the film. Fabra et al. (2008) observed that oleic acid apparently interacted with the protein (sodium caseinate) matrix forming bonds through polar groups, modifiying the interaction balances in the protein network. The result can be a plasticizing effect, including reduction of film strength and increase of film flexibility, as described for whey protein films plasticized with beeswax (Talens and Krochta, 2005), caseinate films with oleic acid (Fabra et al., 2008), and gelatin films with stearic and oleic acids (Limpisophon et al., 2010).

The major drawback of using lipids as plasticizers is their low compatibility with most biopolymer matrices, because of their hydrophobicity. On the other hand, this same characteristic of lipids represents an advantage in which they reduce the water vapour permeability and moisture sensitivity of biopolymer materials.

26.7.1 Phenolic compounds

Several natural phenolic compounds derived from plants have been used as crosslinkers to modify biopolymer films. Several potential interactions may be involved, such as hydrogen bonding, ionic, hydrophobic interactions, and covalent bonding, although covalent bonds are more rigid and thermally stable than other interactions (Zhang et al., 2010a). The postulated chemical pathway involves oxidization of diphenol moieties of phenolic acids or other polyphenols, under alkaline conditions, producing quinone intermediates which react with nucleophiles from reactive amino acid groups such as sulfhydryl groups of cystine, amine groups of lysine and arginine, and amide groups of asparagine and glutamine, forming covalent C–N or C–S bonds between the phenolic ring and proteins. The thus regenerated hydroquinone can be reoxidized and bind a second protein chain, resulting in a crosslink (Strauss and Gibson, 2004; Zhang et al., 2010a).

Caffeic and tannic acids were used as crosslinkers for gelatin (Zhang et al., 2010a), and the use of high-resolution NMR technique confirmed the occurrence of chemical reactions between the phenolic groups in phenolic compounds and amino groups in gelatin, forming covalent C–N bonds; the crosslinking decreased the mobility of the gelatin matrix. A similar study by the same group (Zhang et al., 2010b) indicated that the structure of a gelatin film crosslinked by tannic acid (3 wt%) was stable even under boiling, and that the crosslinking modification enhanced the mechanical properties of the protein.

Several other studies have reported beneficial effects from crosslinking protein matrices with phenolic compounds, such as tannic acid and sorghum condensed tannins (Emmambux et al., 2004), procyanidin (He et al., 2011) and ferulic acid (Ou et al., 2005; Fabra et al., 2011a).

Mathew and Abraham (2008) and Cao et al. (2007) reported significant increases in tensile strength and decreases in elongation at break resulting from adding ferulic acid to starch/chitosan films and gelatin-based films, respectively, which was ascribed to the crosslinking between ferulic acid and the polysaccharides or protein used. Ferulic acid is found as a naturally occurring component in cell walls, crosslinking polysaccharides (especially hemicelluloses) with each other and with other cell wall components such as lignin (Ng et al., 1997; Parker et al., 2005).

26.7.2 Genipin

Genipin (Fig. 26.12) is a hydrolysis product from geniposide, which is a component of traditional Chinese medicine, isolated from gardenia fruits (Gardenial jasminoides Ellis). According to studies by Sung et al. (1999), genipin is about 10,000 times less cytotoxic than glutaraldehyde.

Genipin reacts with nucleophilic groups such as amino groups, being an adequate crosslinking agent for protein films as well as chitosan. Mi et al. (2005) studied the reaction mechanism of chitosan with genipin, and found that genipin undertakes a ring-opening reaction to form an intermediate aldehyde group resulting from the nucleophilic attack by chitosan amino groups. The genipin molecules reacting with a nucleophilic reagent may further undergo polymerization (Fernandes et al., 2013; Jin et al., 2004; Yuan et al., 2007). A dark-blue coloration appears in crosslinked materials exposed to air, which is associated with the oxygen radical-induced polymerization of genipin as well as its reaction with amino groups (Muzzarelli, 2009). Mi et al. (2005) observed that the colour of genipin crosslinked chitosan membranes varied from original transparent to bluish or brownish, depending on the pH value upon crosslinking. These authors ascribed the colour changes to establishment of different structures of crosslinked chitosan resulting from reaction of original genipin or polymerized genipin with primary amino groups on chitosan. The degrees of crosslinking of the genipin-crosslinked chitosan membranes depended significantly on their crosslinking pH values, being higher around pH 7.

Crosslinking with genipin has been reported to improve mechanical properties, water resistance of chitosan films, although they have turned dark bluish (Jin et al., 2004), which may be a market hindrance to many food packaging applications. Bigi et al. (2002) reported that gelatin films crosslinked with genipin presented higher Young’s modulus, better thermal stability (reflected in higher denaturation temperature), and better water resistance than uncrosslinked films. After 1 month of storage in buffer solution, a small gelatin amount (about 2%) was lost from the films, but their mechanical, thermal and swelling properties were very close to those of gelatin films previously crosslinked by glutaraldehyde (Bigi et al., 2001).

Another geniposide derivative, aglycone geniposidic acid, has been used to crosslink chitosan films, improving their tensile strength and water vapour barrier, although reducing their elongation (Mi et al., 2006).

26.7.3 Aldehydes

Aldehydes such as glutaraldehyde, glyoxal or formaldehyde have been used as crosslinking agents to improve mechanical and barrier properties of protein films. However, due to concerns about possible toxic effects of such aldehydes, naturally occurring, less toxic aldehydes and other crosslinking agents have been explored for use in food packaging.

Cinnamaldehyde is an aromatic unsaturated aldehyde derived from cinnamon, consisting of a phenyl group attached to an unsaturated aldehyde (Fig. 26.13). It has been used as an antimicrobial agent in active packaging applications (Becerril et al., 2007), and it can also act as a crosslinking agent for proteins. Balaguer et al. (2011b) demonstrated the crosslinking effect of cinnamaldehyde for gliadin films, which was ascribed to the formation of intermolecular covalent bonds between polypeptide chains, polymerizing gliadins and reticulating the protein matrix. However, it is still uncertain which functional groups of proteins or other macromolecules have more potential to react with cinnamaldehyde, but Balaguer et al. (2011a) proposed a crosslinking mechanism involving the amino groups of proteins, although not ruling out the participation of other reactive groups.

Gliadin films presented significant improvements in their tensile strength and elastic modulus upon crosslinking with cinnamaldehyde. Such effects, as well as water resistance, were reported to be proportional to the cinnamaldehyde concentration (Balaguer et al., 2011b). The crosslinked films did not disintegrate upon a 5-month immersion in water, although a weight loss was reported, indicating that part of the material was solubilized (Balaguer et al., 2011a).

Dialdehyde polysaccharides have received attention as crosslinking agents of protein films. The oxidation of polysaccharides by periodate is characterized by the cleavage of the C2–C3 bond of glucose residues, resulting in the formation of two aldehyde groups per glucose unit, forming 2,3-dialdehyde polysaccharides (Li et al., 2011a). The aldehyde groups can crosslink with ε-amino groups by C = N bonds, as in lysine or hydroxylysine side groups of gelatin (Fig. 26.14) to improve the properties of protein films (Dawlee et al., 2005; Mu et al., 2012). Dialdehyde starch (DAS) was used as a crosslinking agent for gelatin films (Martucci and Ruseckaite, 2009). Crosslinking with DAS up to 10 wt% enhanced moisture resistance and barrier properties of the films, but higher amounts of DAS conducted to phase separation, impairing transparency and tensile properties. Mu et al. (2012) reported that the addition of dialdehyde carboxymethylcellulose (DCMC) to gelatin films increased their tensile strength and thermal stability and reduced their water sensitivity, while keeping their transparency.

26.8.1 From micro to nanoscale

The change of filler dimensions from micro to nanoscale brings about important advantages concerning the resulting composite materials. Because of their size, nanoparticles have larger surface area-to-volume ratio than their microscale counterparts. A uniform dispersion of nanoparticles leads to a very large matrix/filler interfacial area, which changes the molecular mobility, improving thermal, barrier and mechanical properties of the material. Fillers with a high ratio of the largest to the smallest dimension (i.e., aspect ratio), such as nanofibres, are particularly interesting because of their high specific surface area, providing better reinforcing effects (Azizi Samir et al., 2005; Dalmas et al., 2007). In addition, an interphase region of altered mobility surrounding each nanoparticle is induced by well-dispersed nanoparticles, resulting in a percolating interphase network playing an important role in improving the nanocomposite properties (Qiao and Brinson, 2009). According to Jordan et al. (2005), for a constant filler content, a reduction in particle size increases the number of filler particles, bringing them closer to one another, and causing the interface layers from adjacent particles to overlap, altering the bulk properties significantly.

De Moura et al. (2009) observed that the water vapour permeability of hydroxypropyl methylcellulose (HPMC) films reinforced with chitosan nanoparticles was affected by the size of the CsN – the smaller the particles, the lower the permeability. According to the authors, CsN tended to occupy the empty spaces in the pores of the HPMC matrix, thereby improving tensile and barrier properties.

26.8.2 Cellulosic reinforcements

Cellulose fibres are built up by smaller and mechanically stronger entities (cellulose nanoparticles) which can be extracted under proper conditions. Cellulose nanoparticles (i.e., cellulose elements having at least one dimension in the 1–100 nm range, here referred to as nanocellulose), are inherently a low cost and widely available material. Moreover, they are environmentally friendly, easy to recycle by combustion, and require low energy consumption in manufacturing (Klemm et al., 2005; Charreau et al., 2013).

A uniform dispersion of cellulose nanoparticles on a polymer matrix reduces the molecular mobility, changes the relaxation behaviour, and improves the overall thermal and mechanical properties of the material (Azizi Samir et al., 2005; Charreau et al., 2013; Qiu and Hu, 2013). In this context, nanocellulose has been presented as a promising reinforcing and barrier component for elaboration of low cost, lightweight, and high-strength bionanocomposites for food packaging purposes mainly due to its compatibility with the biopolymers (Helbert et al., 1996; Podsiadlo et al., 2005; Moon et al., 2011). Basically two different classes of nanocellulose can be obtained – whiskers and nanofibrils (Azizi Samir et al., 2005). The term ‘whiskers’ (or ‘nanocrystals’) is used to designate elongated crystalline rod-like nanoparticles, whereas the designation ‘nanofibrils’ is used for long flexible nanoparticles consisting of alternating crystalline and amorphous strings (Abdul Khalil et al., 2012).

Different approaches have been introduced to produce nanocellulose: top-down (nanometric structures are obtained by size reduction of bulk materials) and bottom-up approaches (nanostructures are built from individual atoms or molecules capable of self-assembling) (Yousefi et al., 2013). Any cellulosic material can be virtually considered as a potential source for top-down approaches, including crop residues and agroindustry non-food feedstock. However, variations in cellulose source and its preparation conditions lead to a broad spectrum of structures, properties and applicability, which affect performance of cellulose nanoreinforcements (Kvien and Oksman, 2007; Azeredo, 2009; Dufresne, 2012).

Acid hydrolysis has been the primary method for isolating nanocellulose, consisting basically in removing the amorphous regions present in the fibrils leaving the crystalline regions intact; the dimensions of the whiskers after hydrolysis depend on the percentage of amorphous regions in the bulk fibrils, which varies for each organism (Gardner et al., 2008). The morphology of the obtained nanowhiskers is influenced by acid-to-pulp ratio, reaction time, temperature and cellulose source. In spite of the widely varied dimensions (of 3–70 nm widths and 35–3000 nm lengths) reported from different cellulose sources and hydrolysis conditions, cellulose nanowhiskers typically consist of structures 200–400 nm in length and with an aspect ratio of about 10 (Beck-Candanedo et al., 2005; Elazzouzi-Hafraoui et al., 2008; Rosa et al., 2010). Other methods can be used to extract nanocellulose from the lignocellulosic sources, usually based on successive chemical and mechanical treatments, including high-pressure homogenization (Zimmermann et al., 2010), electrospinning (Konwarh et al., 2013), enzymatic hydrolysis (de Campos et al., 2013), TEMPO-mediated oxidation (Isogai et al., 2011), solvent-based isolation (Yousefi et al., 2011), chemi-mechanical forces (Yousefi et al., 2013), ultrasonication (Chen et al., 2011; de Campos et al., 2013), cryo crushing (Alemdar and Sain, 2008) and steam explosion (Deepa et al., 2011). Such processes usually produce longer nanostructures (typically several micrometers in length), but with less uniform width (5–100 nm) (Siró and Plackett, 2010) and lower crystallinity (Iwamoto et al., 2007).

Although the most important industrial source of cellulosic fibres is wood, crops such as flax, cotton, hemp, sisal and others, especially from by-products of these different plants (corn, wheat, rice, sorghum, barley, sugar cane, pineapple, bananas and coconut crops) are likely to become of increasing interest as sources of nanocellulose. These non-wood plants generally contain less lignin than wood and therefore pre-treatment processes are less demanding (Siró and Plackett, 2010; Rosa et al., 2010).

In contrast to celluloses from plants, which require mechanical or chemomechanical processes to produce nanosized structures, the aforementioned bacterial cellulose (BC) is produced already as a nanomaterial by bacteria through cellulose biosynthesis and the building up of bundles of microfibrils (Nakagaito and Yano, 2005). In addition, BC is produced as a highly hydrated and relatively pure cellulose membrane and therefore no chemical treatments are needed to remove lignin and hemicelluloses, as is the case for plant celluloses (Siró and Plackett, 2010).

Nanocomposites for food packaging purposes have been developed by adding nanocellulose to polymers to enhance their physical and mechanical properties (Paula et al., 2011; Azeredo et al., 2012b; Abdollahi et al., 2013; Zainuddin et al., 2013). Nanocellulose has been reported to have a great effect in improving tensile strength and elastic modulus of polymers, especially at temperatures above the T_g of the matrix polymer (Wu et al., 2007; Azeredo et al., 2012b; Zainuddin et al., 2013; Abdollahi et al., 2013). This effect is ascribed not only to the geometry and stiffness of the nanocellulose, but also to the formation of a fibril network within the polymer matrix, the cellulose fibres probably being linked through hydrogen bonds. Barrier properties of polymer films have also been observed to be improved by cellulose nanostructures (Sanchez-Garcia et al., 2008; Paula et al., 2011; Azeredo et al., 2012b; Abdollahi et al., 2013). The presence of crystalline fibres is thought to increase the tortuosity in the materials leading to slower diffusion processes and, hence, to lower permeability (Sanchez-Garcia et al., 2008). Nanocellulose has also been reported to improve thermal properties of polymers (Petersson et al., 2007; Paula et al., 2011). The performance of nanocellulose has been reported to be strongly related to the content, dimensions and consequent aspect ratios of the nanostructures, as well as to the degree of matrix–cellulose interaction and percolation effects (Petersson and Oksman, 2006; Hubbe et al., 2008; Tang and Liu, 2008; Kim et al., 2009).

Since 2011, pilot plants have been opened for the production of nanocellulose in Sweden, Canada and the United States. These facilities make it possible to produce nanocellulose on a large scale for the first time, and it was an important step towards the industrialization of this technology.

26.8.3 Starch nanoreinforcement

Starch granules can be submitted to an extended-time hydrolysis at temperatures below that of gelatinization, when the amorphous regions are hydrolysed before the crystalline lamellae, which are more resistant to hydrolysis. The nanocrystals thus separated show platelet morphology with thicknesses of 6–8 nm (Kristo and Biliaderis, 2007). To prepare starch nanoparticles instead of nanocrystals, Ma et al. (2008) precipitated a starch solution within ethanol as the precipitant.

Similarly to cellulose nanocrystals, the reinforcing effect of starch nanocrystals (SNC) is usually ascribed to the formation of a percolating network maintained by hydrogen bonds above a given filler content (the percolation threshold) (Le Corre et al., 2010). Although not proven, this phenomenon was evidenced from experiments which indicated a changing behaviour above certain filler concentration (Angellier et al., 2005). It has been suggested that starch nanocrystals, similarly to nanoclays (which have a platelet morphology as well), create a tortuous diffusion pathway for permeant molecules through the nanocomposite materials, improving their barrier properties (Le Corre et al., 2010).

Kristo and Biliaderis (2007) reported that the addition of SNC to pullulan films resulted in improved tensile strength and modulus and decreased water vapour permeability. Moreover, the SNC promoted an increase in T_g, probably because of strong interactions of nanocrystals with one another and with the matrix, restricting chain mobility. Chen et al. (2008) observed that the tensile strength and elongation of polyvinyl alcohol (PVA) were only slightly improved by addition of SNC up to 10 wt% and, above this content, such properties were impaired by SNC. On the other hand, the properties of PVA nanocomposite with SNC were better than those of the composites with native starch, indicating that SNC presented a better dispersion and stronger interactions with the matrix than native starch granules.

26.8.4 Chitin/chitosan nanostructures

Chitin whiskers can be prepared by acid hydrolysis of chitin (Lu et al., 2004; Sriupayo et al., 2005), and have been successfully prepared from different chitin sources such as squid pens (Paillet and Dufresne, 2001), crab shells (Nair and Dufresne, 2003), and shrimp shells (Sriupayo et al., 2005). When the acid hydrolysis is followed by mechanical ultrasonication/disruption, chitin nanoparticles (nanospheres) can be formed rather than chitin whiskers (Chang et al., 2010b). Chitosan nanoparticles can be obtained by ionic gelation of chitosan, where the cationic amino groups of chitosan form electrostatic interactions with polyanionic crosslinking agents, such as tripolyphosphate (López-León et al., 2005).

Some authors reported beneficial effects from adding chitin whiskers to biopolymer films. The chitin whiskers added by Lu et al. (2004) to soy protein isolate (SPI) greatly improved the tensile properties and water resistance of the matrix. Sriupayo et al. (2005) reported that the whiskers improved the water resistance of chitosan films, and enhanced their tensile strength until a content of 2.96%, but impaired the strength when at higher contents. Similarly, Chang et al. (2010b) reported that chitin nanoparticles were uniformly dispersed and presented good interaction with a starch matrix when at low loading levels (up to 5%), improving its mechanical, thermal and barrier properties. However, aggregation of nanoparticles occurred at higher loading, impairing the performance of the matrix.

Chitosan nanoparticles have also been proven to be effective as nanoreinforcement to bio-based films. The incorporation of chitosan-tripolyphosphate (CS-TPP) nanoparticles significantly improved mechanical and barrier properties of hydroxypropyl methylcellulose (HPMC) films (De Moura et al., 2009). The authors attributed such effects to the nanoparticles filling discontinuities in the HPMC matrix. Chang et al. (2010a) added chitosan nanoparticles to a starch matrix. The tensile, barrier and thermal properties of the matrix were improved by low contents of the nanoparticles, when they were well dispersed in the matrix. Such effects were ascribed to close interactions between the nanoparticles and the matrix, due to their chemical similarities. However, higher nanoparticle loads (8% w/w) resulted in their aggregation in the nanocomposites, impairing the physical properties of the materials.

26.9.1 Polymer surface modifications

Since most biopolymers have limitations for their applications as packaging materials, such as water sensitivity, brittleness or poor mechanical performance, chemical modification techniques can sometimes be used to generate new biomaterials with improved properties. Shi et al. (2011) grafted lauryl chloride groups onto zein molecule through an acylation reaction, and obtained seven-fold increase in elongation at break of modified zein. Moreover, the modification increased the zein hydrophobicity, suggesting that the end material presented probably decreased moisture sensitivity and water vapour permeability.

The combination between ultraviolet light and ozone (UV-O treatment) has been suggested as an effective method to modify polymer surfaces, leading to oxidation reactions at the polymer surface, while the bulk properties, such as thermal, barrier and mechanical properties of the polymer, may not be altered (Shi et al., 2009). Shi et al. (2009) used UV-O treatment to control hydrophilicity of zein films. The treatment converted some of the surface methyl groups mainly to carbonyl groups, decreasing the water contact angles and increasing the surface hydrophilicity of zein films. Moreover, the authors suggested that, once the surface has been modified, several active compounds could have been linked to the functional groups formed, and the polymer would be not only a barrier material but a carrier to active compounds, constituting an active packaging material.

26.9.2 Active and bioactive biopackaging

Conventional food packaging systems are supposed to passively protect the food, acting as barriers between the food and the surrounding environment. On the other hand, an active food packaging may be defined as a system that not only acts as a barrier but also interacts with the food in some desirable way, for example by releasing desirable compounds (such as antimicrobial or antioxidant agents), or by removing a detrimental factor (such as oxygen or water vapour), usually to improve food stability, that is to say, to better maintain food quality and safety. The compounds added more frequently are antimicrobials, such as chitosan (Shen et al., 2010), acids (Guillard et al., 2009), phenolic compounds (Cerisuelo et al., 2012; Arrieta et al., 2013) and antimicrobial peptides (Sanjurjo et al., 2006; Gómez-Guillén et al., 2011).

More recently, the concept of bioactive packaging, in which a food packaging or coating has a potential to enhance food impact over consumer health, has been proposed by Lopez-Rubio et al. (2006). Enclosing bioactive compounds within packaging instead of directly incorporating them into food presents some industrial benefits, such as increasing retention of bioactives, reducing some incompatibility problems between the bioactive and the food matrix, and reducing changes to food sensory properties (Lopez-Rubio et al., 2006). Antioxidant packaging materials could be included in both active and bioactive packaging concepts, since antioxidant compounds usually have alleged benefits both for food stability and for consumer health. Some studies have described the incorporation of antioxidant compounds or extracts to biopolymer films and coatings, such as phenolic compounds to zein films (Arcan and Yemenicioglu, 2011), α-tocopherol to chitosan films (Martins et al., 2012), curcumin and ascorbyl dipalmitate to cellulose-based films (Sonkaew et al., 2012), and ferulic acid and α-tocopherol to sodium caseinate films (Fabra et al., 2011a). Other studies have described the development of probiotic films in coatings, based on the incorporation of lactic acid bacteria, such as Bifidobacterium lactis incorporated to alginate and gellan films for coating of fresh fruits (Tapia et al., 2007), Lactobacillus sakei added to caseinate films to control Listeria monocytogenes in fresh beef (Gialamas et al.,2010),Lactobacillus acidophilus in starch-based coatings for breads (Altamirano-Fortoul et al., 2012) and L. acidophilus and Bifidobacterium bifidum incorporated to gelatin coatings for fish (López de Lacey et al., 2012).

The active and bioactive compounds are most frequently simply added to the film-forming formulation, constituting one of the formulation components, but not being chemically linked to the matrix polymer. However, some studies have suggested the covalent immobilization of active compounds onto functionalized polymer surfaces, with many possible applications for food packaging purposes. Biopolymer films may be functionalized in two basic steps, namely, the treatment of the polymer surface in order to produce reactive functional groups, and the reaction of such groups with an active compound (Kugel et al., 2011). Sometimes the functionalization can be favoured by an intermediary step, such as grafting a polyfunctional agent onto the polymer surface, increasing the density of available functional groups, or by a spacer molecule to reduce steric hindrances (Goddard and Hotchkiss, 2007). An active compound which is covalently immobilized onto the packaging material is not supposed to be released, but becomes effective when in contact with the food surface (Han, 2003). The active compounds to be tethered can be enzymes, antimicrobials, biosensors, bioreactors, etc. Some studies have described the covalent attachment of active compounds onto the surface of biopolymer materials to be applied as food packaging or coatings, such as lysozyme onto chitosan coatings (Lian et al., 2012) and N-halamine onto chitosan films (Li et al., 2013).