nanobuilding blocks ( NBB ) : current applications and future opportunities in the food packaging sector

I M I T t b 2 E e t T ( g shelf life studies based on modie gies. She has been doing her PhD Packaging lab under the guidance o deals with the design and develo nanocomposite lms and coatings f From July 2013 till May 2014 sh fellow at the University of Warwi Research Group of WMG led by supervision of Dr Chaoying Wan. DeFENS, Department of Food, Environment Division, University of Milan, Via Celoria 2 farris@unimi.it; Fax: +39 0250316672; Tel: Department of Natural and Territorial Scien 4, 07100 Sassari, Italy


Introduction
From the 60's to date, the concept of nanotechnology has developed into a multidisciplinary eld of fundamental and applied science and technology.Over the last decade, in particular, nanotechnology has become one of the most appealing topics within the research programs of both public agencies (e.g., universities) and private companies.At least by researchers, nanotechnology is generally deemed as an exciting area in which to work, allowing creativity in obtaining new materials with deliberately induced nano-sized architectures.Eventually, this eld will yield new and innovative features and functions that can be protably exploited in everyday life.Some authors, however, have expressed some skepticism about nanotechnology, the main question being whether it is a real opportunity or just a hype to extract funds. 1 Although nanomaterials and nanoparticles have been consumed by humans for centuries, consumers' perceptions toward nanotechnology are not totally agreeable, mainly due to the still-existing uncertainties concerning the potential risks for humans and the environments associated with the use of nanomaterials for sensitive applications, such as many areas of the food industry, including packaging, nutrient delivery, and food quality. 2 Nevertheless, in 2008 nanotechnology involved over $15 billion in worldwide research and development money (public and private) and employed over 400 000 researchers across the globe. 3Moreover, nanotechnologies will likely impact at least $3 trillion across the global economy by 2020, and worldwide nanotechnology industries may require at least 6 million workers to support them by the end of the decade. 3espite the still open debate about nanotechnology, it is not rash to say that the world is in the middle of a nano-revolution.
The food industry seems to receive the largest benets from nanotechnology, with potential uses already identied in virtually every segment of the sector (i.e., agriculture, food processing, food packaging, and nutrient supplements). 4oreover, a number of the world's largest food companies are actively exploring the potential of nanomaterials for use in food applications. 5Within the food industry, however, packaging applications form the largest share of the current and shortterm predicted market for nano-enabled products, 6 with an estimated annual growth rate of $12% by 2014. 7One of the reasons for this trend is the consumers' willingness to embrace nanotechnology in "out-of-food" applications more than the direct addition of nanoparticles into the food. 8,9This is also why most nanotechnology applications for food and beverages are lagging behind applications for food packaging, which are already becoming a commercial reality. 10ne of the most successful applications of nanotechnology in the eld of packaging concerns the development of "nanocomposites". 11In polymer science, while the term "composites" generally refers to mixtures of polymers with inorganic or organic additives having micron-length scale and certain geometries (bers, akes, spheres, and particulates), the use of nano-length-scale entities can be more specically referred to as "nanocomposites."These entities are well-dened nanobuilding blocks (NBB), which consist of preformed objects that keep their integrity in the nal nanocomposite materials. 12hese NBB can be clusters, nano-core shells, organically pre-or post-functionalized nanoparticles (e.g., metal oxides), and layered compounds (e.g., clays). 13The preferential route to produce nanocomposites lies in the "top-down" approach, which progressively reduces the size of nanocomposite bulk materials using grinding, milling, etching, lithography, or precision engineering. 14][17][18] A new subclass of nanocomposite materials has recently stemmed from the increasing endeavor to replace oil-based polymers with polymers partially or totally obtained from renewable resources.The term "bionanocomposites" refers to those materials in which the polymer matrix carrying the nanosized llers is a biopolymer (e.g., polysaccharides and proteins) or by a polymer of natural origin obtained by synthetic (e.g.polylactic acid, PLA) or biotechnological (pullulan, poly-hydroxyalkanoates-PHAs) routes.The use of NBB for the generation of bionanocomposites has enormous potential for overcoming the drawbacks exhibited by natural polymers, such as poor mechanical and thermal properties, sensitivity to moist environments, and inadequate barrier properties to gas and vapors; the development of bionanocomposites has thus drawn considerable attention in recent years.However, most examples concern the incorporation of the inorganic phase directly into the bulky biopolymer.][21][22] We conceived this review as guidance for tomorrow's research in the eld of nanocomposite materials, especially in the form of coatings specically intended for food packaging applications, with special emphasis on those solutions envisaging the use of biopolymer matrices.First, we provide a brief introduction, including key denitions and classications.Second, we present an in-depth description of inorganic llers with special focus to the following: inorganic clays (e.g.montmorillonite), which are the most widely used for food packaging purposes; metals and metal oxides, especially those obtained by sol-gel routes that have gained much attention for food packaging applications due to their high versatility; zeolites, which despite having many appealing features, are still under-exploited llers; and graphene, which is one of the "llers of the future", owing to its promising and attractive properties for the next generation of materials-not only for packaging applications.Besides discussing the most relevant physicochemical properties, a detailed description of current developments for the food packaging sector is also reported for each class of ller.Among other aspects, in the nal "technical" section, we rst thoroughly discuss the processing of inorganic llers within (bio)polymer matrices to obtain (bio)nanocomposite coatings.Successively, we describe the most important aspects that may impact the ultimate performance of (bio)nanocomposite lms and coatings in light of their end use as food packaging materials.We also provide an outlook on future trends with the goal of prompting new research on (bio)nanocomposite coatings as potential alternatives to the currently used solutions.[25][26][27][28][29][30][31][32][33]

Classification of fillers
A rst distinction within the huge eld of llers can be done according to their chemical composition.Accordingly, it is useful to distinguish between organic llers, namely those containing carbon atoms and most oen directly derived from biomass, and inorganic llers, which may be also of natural origin but as a general rule do not consist of any carbon atoms.Graphene is an exception to this rule, as it is arbitrarily included within the inorganic ller class although it is only consists of carbon atoms.As both classes (organic and inorganic) encompass a wide assortment of examples, each including many different applications, it would be impossible to provide an exhaustive overview that takes into account all of them.Therefore, our focus will be only on the inorganic llers.

Inorganic llers: physicochemical properties and main developments for food packaging applications
The use of inorganic nano-sized building blocks as llers in the preparation of polymer/inorganic nanocomposites has attracted increasing interest in recent years, owing to their unique properties that nd numerous applications in many industrial elds.As shown in Chart 1, a rst arbitrary classication of inorganic llers, primarily based on chemical composition and crystal structure, is between silicate and nonsilicate minerals.Both families include several classes and many groups that, in turn, comprise different mineral species potentially suitable to produce nanocomposites.However, only few of these minerals have so far been exploited in practice.If, on one hand developers have thus far used layered silicates (e.g.clays) the most extensively as llers to obtain nanocomposites, 4,[34][35][36] on the other hand, developers have introduced carbon-based nanollers such as carbon black, [37][38][39] expanded graphite, [40][41][42] carbon nanotubes (CNTs), [43][44][45] and carbon nano-bers [46][47][48] in the preparation of polymer nanocomposites to achieve properties otherwise unattainable, such as electrical and thermal conductivity of the nal nanocomposite material.0][51][52] The only drawback of CNTs as a nanoller is their relatively high production cost. 43,51,53,54Therefore, as Nicholos said, 55 "When carbon bers just won't do, but nanotubes are too expensive, where can a cost-conscious materials scientist go to nd a practical conductive composite?The answer could lie with graphene sheets".
In this section, we rst review the most widely used inorganic llers, i.e. clays, with also a brief mention of other layered minerals of interest, i.e. micas and layered double hydroxides.Next, we discuss the current and potential use of zeolites, while the last part of this section will be dedicated to the rise of graphene as a promising candidate for the development of new advanced materials.We will place special emphasis on the structure-property relationships that eventually lead to the enhancement of material properties of interest for food packaging applications, such as barrier, mechanical and thermal properties.
2.1.1Clay minerals.Clay minerals used in the manufacturing of nanocomposite materials are natural or synthetic layered hydrated aluminum phyllosilicates, oen hydrated, with neutral or negative charged layers. 56The building blocks of clays include tetrahedral and octahedral sheets.In the tetrahedral sheet each tetrahedron shares three corners with the adjacent tetrahedrons, ideally dening a surface with hexagonal mesh.The tetrahedrons are centered by one silicon atom (sometime substituted by aluminum and/or iron), which coordinates four oxygen atoms located at the corners.A plane of octahedrons sharing edges, arranged in a hexagonal pattern, denes the octahedral sheet.Depending on the number of octahedral sheets, two different structures can be eventually encountered, i.e. the gibbsite-like structure (dioctahedral) and the brucite-like structure (trioctahedral).The center of the octahedron is occupied by aluminum or magnesium, whereas six oxygen atoms, some of them also linked to a hydrogen atom, are located at the corners.Fe 3+ and Fe 2+ can also be found at the center of the octahedron.The superimposition of tetrahedral (T) and octahedral (O) sheets may originate both 2 : 1 (T-O-T) and 1 : 1 (T-O) layered structures, where the sum of each sheet originates a layer (known as a platelet). 57In 2 : 1 layered structures, if isomorphic substitution of silicon and/or aluminum with a lower-valence atom takes place (either in the octahedral or tetrahedral sheets), the mineral is characterized by a negative surface charge, which, in some clay groups, will be naturally counterbalanced by positive inorganic ions (e.g.Li + , Na + , Ca 2+ , K + , Mg 2+ , etc.) in the interlayer.In these cases, the nature of the cations plays a crucial role in determining the swelling tendency of the clay upon water adsorption.As a general rule, the smaller is the cation and the lower is its charge, the more the clay swells.For example, in montmorillonite (a clay belonging to the smectite group), the swelling increases according with the following trend: Li + > Na + > Ca 2+ > Fe 2+ > K + . 58he K + ion provides the lowest swelling capability because it stabilizes the structure due to the fact that, more than the other ions, its size well ts with the hosting size represented by the center of the hexagon dened by six Si-centered tetrahedrons.
The capability of layered silicates to exchange cations is quantied by the cation-exchange capacity (CEC), usually expressed as meq per 100 g.Ideally, the CEC of a clay mineral is equivalent to its (negative) layer charge.This parameter can greatly vary depending on the type and origin of the clays, as well as within the same clay, because the surface charge is locally randomly distributed (i.e. it may vary from layer to layer or within the same layer). 59The layers of mineral clays generally organize to form stacks with a regular van der Walls gap in between them called the "interlayer distance" or "gallery", while the "d-spacing" (d 001 ), also called the "basal spacing", accounts for the sum of the interlayer and the platelet thickness (Fig. 1).When the hydrated cations are ion-exchanged with organic cations (e.g. more bulky alkyammoniums), which is normally done to increase the affinity towards more hydrophobic polymers such as polyolens or PLA, the nal result is a larger interlayer spacing, thus a larger "d" value.Three organization levels can be recognized within clays: (i) the layer (or platelet), with thickness of approximately 1 nm and length varying from a few hundreds of Å to several microns, depending on the type of clay; (ii) primary particle, composed of ve to ten platelets stacked on top of one another; (iii) micro-aggregates (or tactoids), structures of approximately 0.1 to 10 mm formed through the association of randomly oriented several primary particles. 60ne common feature of clays is the high aspect ratio (i.e., the ratio between length and thickness).Although this parameter can vary greatly depending on the type and origin of clays, it may be of the order of 1000. 15ased on our bibliographic search, the four main groups of clay minerals of technological interest for the fabrication of nanocomposite materials are-in order of relevance for this eld-smectite, kaolin, palygorskite-sepiolite, and vermiculite.
In addition to "single polymer matrix" nanocomposites, researchers have also proposed blends of two or more polymeric materials (i.e.binary or ternary blends) for the development of MMT-based nanocomposites with enhanced physicochemical properties ensuing from synergistic effects.Blends of modied PA and nylon-6, 130 PET and aromatic PA, 131 PP and PA 6,6, 132 nylon-6 and ethylenevinylalcohol (EVOH) 133 were reinforced with MMT for the improvement of barrier, thermal and mechanical properties.Developers have also used MMT to build multifunctional thin lms by alternating two or more layers of anionic and cationic polymer aqueous solutions or dispersions; this process could theoretically include an innite number of layers.For instance, one study described the lm growth and structure of assemblies made with cationic PEI and anionic MMT clay and polyacrylic acid (PAA), referring to one deposition sequence of PEI/PAA/PEI/MMT as a quadlayer.The study asserted the resulting polymer nanocomposite thin lms have unprecedented barrier performance, especially in terms of oxygen permeability. 134ue to the increasing concerns about environmental issues linked to the waste disposal of plastics at the end of the service, research has focused on the possibility of extending the technical benets of nanocomposites to polymers of natural origin, with the main goal of overcoming the traditional shortcomings of biopolymers, such as sensitivity to water or humid environments, as well as poor barrier and mechanical properties.More specically, because the food industry is among the sectors that takes advantage of poorly biodegradable petroleum-based plastic materials, the development of eco-friendly bionanocomposite lms and coatings is one of the greatest challenges that has become imperative to this industry.Within the food packaging sector, several studies have reported on the improvement of the mechanical, thermal, barrier, active, optical, and water-resistance properties of several bionanocomposites based on MMT, as reported in Table 1.Although MMT clays have attracted much attention due to their undeniable advantages over other type of clays, researchers have also explored the use of other clays belonging to the smectite group to fabricate bionanocomposites for food packaging applications.
2.1.1.1.2Rectorite.Rectorite (REC) (Fig. 2b) is a yellowishwhite layered silicate consisting of a regular interstratication of 2 : 1 dioctahedral mica layers (see paragraph 2.2.2.2) and 2 : 1 dioctahedral smectite (MMT-like) layers.The structure and properties of REC thus fall between completely expanded MMT and completely collapsed mica.The proportion of smectite layers in a mixed-layer clay determines the clay's expandability.In REC, since the distribution of smectite and mica layers in a clay crystallite is ordered (or regular), the nal properties are virtually in between those of smectite and mica. 1847 The Brunnuer, Emmet and Teller (BET) surface area was $30 m 2 g À1 , 188 whereas the length and thickness of REC layer were $100-200 nm and 2 nm, respectively, yielding an aspect ratio of 50-100, lower than MMT. 189Examples of bionanocomposite coatings based on REC are reported in Table 1.
2.1.1.1.3Saponite.Saponite (SAP) (Fig. 2c) is another mineral clay belonging to the smectite group, with a chemical formula (M x + $nH 2 O)Mg 3 (Si 4Àx Al x )O 10 (OH) 2 , 67 surface area $200 m 2 g À1 , 193 CEC of $80-120 meq per 100 g, aspect ratio of 50-60, 17,[194][195][196][197] and basal spacing between 12 Å (ref.198) and 14.1 Å. 197 Unlike MMT, which is a dioctahedral smectite, SAP has Mg 2+ in the trioctahedral brucite-like sheet.Substitution of  Chitosan + chitin whisker 192 Chitosan + chitin whisker 192 Chitosan + chitin whisker 192 Chitosan + chitin whisker 192 Chitosan 189 Chitosan + PVOH blend 191 Saponite PBS 202 PBS 202 Laponite BC 207 BC 207 Starch-PVOH blend 218 Mg 2+ with Al 3+ gives rise to a positive excess charge in the octahedral sheets, 67,199 which is, however, over-compensated by the positive charge deciency in the tetrahedral sheets due to substitutions of Si 4+ with Al 3+ (tetrahedral Al reported to be as high as 90% of the total Al in the sample). 193This eventually leads to an overall negative charge of SAP clays.The excess of positive charges in the octahedral sheet of SAP gives it a weaker aptitude to hydration compared with MMT, similar to vermiculite. 194However, the charge deciency in the tetrahedral sheet of SAP has an important implication on the state of the interlayer water.In minerals in which the greatest part of the charge deciency comes from tetrahedral substitutions (as in SAP), water molecules arrange in a three-dimensional organization mediated by the formation of hydrogen bonds between water and some negatively charged surface oxygen. 200Relatively rigid bridges are thus formed from one layer to another, the exchangeable cations acting as intermediaries. 194Therefore, with respect to water, SAP has a higher water affinity, but a lower hydration tendency than MMT.Another difference between SAP and MMT concerns the morphology.Whereas the structure of MMT is in the form of hexagonal lamellae, SAP shows a structure in the form of ribbons and laths. 201evelopers have used SAP in combination with biodegradable polymers such as poly(butylene succinate) (PBS), in which the addition of the ller yielded better mechanical and oxygen permeability properties 202 (Table 1).SAP used in combination with poly(vinyl alcohol) (PVOH) yielded PVOH/SAP nanocomposites with remarkable mechanical properties, thermal stability, 197,203 water resistance, 197,204 optical properties, 203 and oxygen permeability. 203.1.1.1.4Laponite.Laponite (LAP) (Fig. 2d) is a fully synthetic smectite clay similar in structure and composition to natural hectorite (see next section).In this clay part of the Mg 2+ in the trioctahedral sheet is replaced by Li + , resulting in a net negative charge of the layer, which is counterbalanced by Na + cations located between adjacent layers in a stack. 205LAP is currently available in the market under two main types, with the following empirical formulae: [Si 8 (Mg 5.5 Li 0.3 )O 20 (OH) 2.5 F 1.5 ]-Na 0.7 , (type 1) and [Si 8 (Mg 5.5 Li 0.3 )O 20 (OH) 4 ]Na 0.7 (type 2). 206LAP has a plate-like shape with dimensions typically around 20-30 nm in length and 0.92 nm in width, basal spacing of $1.3 nm, [207][208][209] estimated surface area of over 900 m 2 g À1 (206) and CEC typically 90 to 100 meq per 100 g for type 1 products and 50 to 60 meq per 100 g for type 2 products. 210One of the main technological advantages of LAP lies in the absence of silica and iron oxides in its composition, which allows producing clear and transparent colloidal suspensions of clay particles in water. 211Commercial application of LAP have thus far mainly involved household products, personal care items, horticulture, and the preparation of paper and polymer lms. 212he materials science eld has used LAP in combination with both hydrophobic and hydrophilic synthetic polymers, such as PE, 213 PVC 214 and PVOH, 215,216 respectively.Developers have also used LAP in the production of bionanocomposites with claimed potential application in the food sector as biodegradable packaging lms, as summarized in Table 1.
2.1.1.1.5Hectorite.Hectorite (HEC) (Fig. 2e 220 Sinha Ray and Okamoto reported a CEC of $120 meq per 100 g. 17 A more recent study reported a CEC of 43.9 meq per 100 g, a surface area of $65 m 2 g À1 and basal spacing of 14.4 Å for Ca 2+ -HEC. 66Other research conrmed a surface area of 63 m 2 g À1 . 221HEC generally exhibits an aspect ratio of 196 which is higher than MMT and SAP.To the best of our knowledge, only two works have been published on HEC-based bionanocomposites with potential application for the food packaging sector (Table 1).
2.1.1.2Kaolin clays.The term kaolin refers to a rock, white or near white in color, mainly constituted by kaolinite (i.e., KAO), which is the most important mineral within the group of kaolin clays.KAO is a 1 : 1 (or T-O) layered silicate with a chemical formula Al 2 Si 2 O 5 (OH) 4 consisting of one gibbsite-like dioctahedral sheet and one tetrahedral sheet layer (Fig. 2f). 56he ultimate structure is an asymmetric conguration with very little substitution in the structural lattice; thus it has a minimal layer charge and a low exchange capacity ($3-15 meq per 100 g). 148This asymmetric structure allows the formation of hydrogen bonds between consecutive layers, providing a large cohesive energy. 223In turn, this makes intercalation of polymers more difficult, which is probably the main reason for the fewer occurrences of kaolin in the preparation of nanocomposite polymer materials, despite it being inexpensive, abundant, readily available, and environmental friendly.Another important technological property of KAO is the low absorption capacity, which relates to the minimal layer charge and the low surface area (5-20 g m À2 ) in comparison to smectite clays. 148lso, KAO exhibits a lower aspect ratio compared to smectite clays, usually around 10 but can vary widely from very low values $2 for laminated stacks (or "books") to more than 11 up to 36 for very thin platelets in the 4-22 range. 224Although the largest use of kaolin is for coating paper, 148 kaolinite has also been used to produce bionanocomposite materials especially in combination with corn starch derived from different sources (Table 1).
Within the kaolin group, halloysite (HAL) has recently gained researchers' attention.HAL is a 1 : 1 two-layered aluminum hydrated phyllosilicate with a predominantly hollow tubular structure in the submicron range. 230HAL may be regarded as a hydrated kaolinite Al 2 Si 2 O 5 (OH) 4 $2H 2 O (or even 4H 2 O) with curved layers and a basal spacing of 1 nm that decreases to about 0.7 nm upon dehydration. 67The neighboring alumina and silica layers and their waters of hydration, curve and form multilayer tubes with average diameters of 50 nm and inner lumen diameters of 15 nm, while the length may vary between 500 and 1000 nm. 231This peculiar tubular shape gives rise to the commonly used term "halloysite nanotubes" (HNTs) (Fig. 2g). 232NTs have CEC of 5-10 meq per 100 g, 233 whereas the typical surface area is 65 m 2 g À1 .The basal space of HNTs is 0.73 nm. 234NTs are able to entrap a range of molecules with specic sizes and can serve as a viable nano-cage for active molecules, owing to the empty space inside the nanotubes (pores volumes of $1.25 mL g À1 ). 2357][238][239] HNTs have also served as supports to immobilize catalyst molecules such as metallocomplexes. 2402][243] However, we found very few works on this topic (Table 1).The results reported support the concept of bionanocomposite technology as a valuable tool to improve the water barrier, mechanical, and thermal properties of biopolymer lms, with high potential for food packaging purposes.
2.1.1.3Palygorskite and sepiolite clays.Palygorskite and sepiolite contain a continuous two-dimensional tetrahedral sheet (that is why they are considered phyllosilicates), whereas the octahedral sheet lacks of continuity.Their framework contains ribbons with a T-O-T structure, and the apices of tetrahedra belonging to adjacent ribbons point in opposite directions. 67This periodic inversion of the SiO 4 tetrahedra determines channels or tunnels (extended along X-axis 67 ) of rectangular section, which is the reason of the brous structure of these two non-swelling clays.The channels (wider in sepiolite) contain water and, due to the low CEC of both minerals, a small amount of exchangeable cations.Palygorskite (oen called attapulgite in industrial and trade applications) is intermediate between di-and trioctahedral, whereas sepiolite is essentially trioctahedral.As far as their cell parameters, the major difference concerns b, respectively 1.78-1.80nm and 2.68-2.72 nm. 67epiolite (SEP -Fig.2h), with a theoretical half-unit-cell formula Mg 8 Si 12 O 30 (OH,F) 4 (OH 2 ) 4 $8H 2 O, 251 is a clay of relevance for the nanocomposites manufacturing.It shows a pore volume of $0.4 cm 3 g À1 , 252-256 a surface area of $200-400 m 2 g À1 (257,258) and a CEC generally about 30 to 40 meq per 100 g. 148 SEP is a "high silanols carrying ller", with silanols spaced every 5 Å along the length of the SEP needle. 259This very high density of silanol groups (2.2 groups per 100 Å (260) ) suggests that forces such as hydrogen bonding and van der Waals interactions are largely involved in coupling reactions with both polymers and organic surfactants as well as in adsorption phenomena at the interface. 261Both the excellent affinity with polymeric matrices and the strong anisotropy of this mineral account for the outstanding reinforcing effect on polymers, which results in a marked increase of the elastic modulus of the nal compounds.However, the high concentration of hydroxyl groups on SEP surface is responsible for the strong catalytic effect of SEP on the thermal degradation of polyolens. 262Compared to other inorganic llers, SEP has found wide use in obtaining bionanocomposites with potential for food packaging applications (Table 1).
2.1.1.4Other clay minerals 2.1.1.4.1 Vermiculite.Vermiculite (VMT) (Fig. 2i) is a 2 : 1 clay, generally trioctahedral.As in smectite, hydrated cations are located in the interlayer, although in VMT the negative layer charge, larger than that of smectite, arises mostly from substitution of Al 3+ with Si 4+ in the tetrahedral sheets. 67Occasionally isomorphic substitutions of Mg 2+ with Al 3+ determine an excess of positive charge in the brucite-like sheet (+0.62 electrostatic valence units), which is overcompensated by the excess of negative charge in the tetrahedral sheet (À1.81 electrostatic valence units). 271The location of the layer charge determines the strength of the Lewis base: VMT behaves as a hard base, whereas MMT, which presents relevant octahedral substitutions, behaves as a so base. 271][273][274] Besides, VMT has a slight higher CEC compared with MMT, generally between 100 and 150 meq per 100 g. 72 Brigatti et al. reported CEC ¼ 84.4 meq per 100 g for MMT and 142 meq per 100 g for VMT, 271 whereas Wang et al. reported CEC ¼ 86.5 meq per 100 g for Na + -VMT. 272One study measured the basal spacing (d 001 ) of vermiculite at 11.2 Å. 272 Considering that the thickness of the Na + -VMT platelet is reported between 9.6 Å and 10 Å, 273,274 it follows that the VMT interlayer distance (layer thickness subtracted from the basal spacing) ranges between 0.16 and 0.12 nm.Compared to MMT, VMT shows slightly lower surface area (500-700 m 2 g À1 ), whereas Xu et al. reported VMT layers with thickness of $2-5 nm, and length of $200 nm. 72VMT is not as widely used as, for example, MMT for the generation of bionanocomposites.Few examples of VMT-based bionanocomposites are reported in Table 1.
2.1.1.4.2 Imogolite.Imogolite (IMO) is a clay-size hydrous alumino-silicate of short-range order with the chemical formula (OH) 3 Al 2 O 3 SiOH. 67IMO forms slender hollow tubules (Fig. 2j).The SiOH groups are located on the inner surface of the nanotube, while AlOH groups are on the outer surface.The hydroxyl groups on the walls and rims make the nanotube hydrophilic.1][282] Synthetic IMO has a monodisperse nanotube length of 100 nm in solution, as determined by dynamic light scattering. 283These dimensions are considerably lower than HAL (halloysite); therefore IMO can provide higher surface areas than HAL for the same volume.
In IMO the tetrahedral sheets are located in the interior of the nanotubes, while the dioctahedral gibbsite-like sheets are on the exterior, with the hydroxyl groups of the Al 2 OH groups pointing outward, i.e. to the surrounding medium.As a consequence, the surface charge (and thus the exchange capacity) varies with the pH of the medium, which greatly affects the dissociation of hydroxyls.In this respect, the high pH at pointof-zero charge (pH > 11) on the outer parts of the tube and the anomalously high chloride adsorption of IMO may originate a weak positive charge on the outer tube walls, while a negative charge develops in the tubular pores. 281Therefore, IMO can have a positive charge under acidic conditions. 284However, chloride adsorption is largely equimolar with sodium adsorption when NaCl is added over a wide pH range, 285,286 thus indicating the simultaneous presence of a weak positive charge and a weak negative charge along the outer and inner IMO nanotubes, respectively.
The surface area of IMO as measured by N 2 -BET adsorption is $300-400 m 2 g À1 . 281The repeating distance along the tubule axis is 0.84 nm. 67The formation of IMO occurs in a gibbsite sheet through the displacement of three hydroxyl groups surrounding a vacant octahedral site by an orthosilicate anion.This requires a considerable shortening of the O-O distances around the site (to <3 Å), which causes the gibbsite sheet to curl and form a tube.Doubly coordinated -Al 2 OH groups are found on the outer tube walls, whereas the inner tube walls carry SiOH groups. 281Although IMO does not swell under ordinary conditions, it retains 1.5 times more water than Na + -MMT due to micropores formed by intertwining brous particles. 287MO is a "green" nanomaterial, although natural IMO is not sufficient to fulll current demands.Recent advances thus concern IMO synthesis and new methods for large quantity production, which developers have performed without using toxic reagents or aggressive solvents, making this nanomaterial easily available.284 2.1.2Other layered minerals of interest 2.1.2.1 Layered double hydroxides. Layed double hydroxides (LDHs) (Fig. 2k), also called "non-silicate oxides and hydroxides", have many physical and chemical properties that are surprisingly similar to those of clay minerals.288  Li + , Mg 2+ , Ni 2+ , Cu 2+ , Ca 2+ , or Zn 2+ ; M III is a trivalent metal ion such as Al 3+ , Fe 3+ , Cr 3+ , or Ga 3+ ; y is the value of the layer charge for an octahedral unit; and A mÀ is an exchangeable anion with valence m like NO 3 À , Cl À , CO 3 2À , or SO 4 2À .Therefore y is equal to 2x À 1 or x for M is a univalent or a divalent metal ion.289 LDHs consist of stacked brucite [Mg(OH 2 )]-like trioctahedral sheets that are positively charged due to the replacement of Mg atoms by Al atoms, which is why LDHs are called "anionic clays" (indeed LDHs encase anionic exchangeable species intercalated between the layers).Hydrotalcite, Mg 0.75 Al 0.25 (OH) 2 -(CO 3 ) 0.5 $0.5H 2 O (formula oen abbreviated as Mg-Al-CO 3 ) is one of the most representative mineral of the group.LDHs possess a high anion exchange capacity (AEC) of 200-500 meq per 100 g, 290 whereas the BET surface area for Mg-Al-CO 3 LDH can be as high as 21.6 m 2 g À1 .291 The ion exchange or coprecipitation methods can introduce inorganic or organic anions between the octahedral hydroxide layers.292 Because many biomolecules (e.g.polysaccharides) are negatively charged, LDHs can nd advantageous use in enhanced biopolymer/ller complexes.
Studies have reported the basal spacing of Mg-Al-CO 3 (given by the sum of a brucite-like octahedral sheet and the interlayer spacing) as 7.8 Å (ref.293) and 7.9 Å. 289 Assuming the brucitelike layer of 4.8 Å, 289,294 the interlayer spacing can be deducted as $3 Å thick, indicating the presence of strong interaction between the water molecules and the lattice OH groups of hydroxide layers. 295While modication of LDHs can be carried out to modify the surface properties, delamination of LDHs has been proven to be very difficult. 296One study derived a nominal aspect ratio of Mg-Al-NO 3 LDH incorporated in PLA by the Nielsen model and calculated it as 50, while the same parameter increased to 85 for the same LDH aer organo-modication with sodium g-polyglutamate (g-PGA). 295Several works have reported on the fabrication of LDHs-based bionanocomposites, indicating the potential application of these complexes for the food packaging eld (Table 1).
2.1.2.2 Micas.The mica group (Fig. 2l) includes sheet silicates that are ubiquitous in nature, and oen readily available at low cost, like muscovite.These minerals (natural or synthetic), are T-O-T phyllosilicates, and both di-and trioctahedral micas exist.In some naturally occurring micas, Si 4+ nearly lls all of the tetrahedral sites, whereas in the most common species (muscovite and phlogopite) Al 3+ substitutes for Si 4+ in a ratio close to 1 : 3. 67 However, the Al 3+ for Si 4+ substitution can reach even a 1 : 1 ratio.This mineral group is subdivided in exible (or true) micas, where the negative charge of the tetrahedral sheet is balanced by an interlayer monovalent cation (usually potassium), and brittle micas, where a bivalent cation (generally calcium) compensates a higher negative charge of the tetrahedral sheet.As a rule, the cations hosted in the mica interlayer are anhydrous.A typical example is muscovite (ideal formula KAl 2 [AlSi 3 O 10 ](OH) 2 ), which has layers well held together because the size of the K + matches the hexagonal hole present in the tetrahedral mesh.Therefore, micas do not swell in water and, like pyrophyllite (a 2 : 1 dioctahedral phyllosilicate without any substitution of atoms), have no internal surface area. 303,304However, micas able to expand with very high aspect ratios (up to 1000) have been obtained by synthesis. 305Micas, in combination with different biopolymers, could serve as potential ller for the production of bionanocomposites for food packaging applications (Table 1).
Despite the wide range of nanollers currently available on the market, selecting the most appropriate solution for a specic (bio)polymer system and for a given application may be not easy as it might seem.This choice requires accounting for a number of parameters with care, all of them generally relying on the maximization of the affinity between polymer matrix and ller, with the goal of making the latter an integrative part of the former.The chemical compatibility between the two phases (polymer and ller) is the rst aspect to discriminate between natural and organo-modied llers.The overall charge in its multifaceted aspects, namely type (negative or positive), amount (density of the charge), and location (tetrahedral or octahedral layers, bulk charge or edge charge), is another important parameter to consider especially if polyelectrolyte polymers (e.g. the negatively charged pectins, carrageenans, hyaluronic acid, or positively charged chitosan) have to be associated with the ller, or if active molecules (e.g.drugs, antimicrobials, antioxidants) must be hosted within the inorganic lattice.The water absorption and swelling capability of the ller may also be critical.While most oen these properties coincide, they do not necessarily appear in the same ller; physical (e.g.porosity) and chemical (e.g.interlayer forces, type of counterions) characteristics may contribute in different ways.Thus, a "zero-point-charge" porous ller may absorb water but not swell and vice versa.Another important aspect is the shape of the ller and thus its aspect ratio, which becomes of utmost importance in light of the specic application.For example, selecting a needle-like shape or a platy shape can make a difference if the ultimate goal is to improve the mechanical properties or the barrier performance of the nal nanocomposite material.
We report a summary of the information gathered during our literature search in an "all-inclusive" table (Table 2) that has never been proposed before.The aim is to provide a useful (although not exhaustive) tool for scientists working in the design and development of nanocomposite materials by offering the possibility to compare the most relevant properties or features of well-known silicate and non-silicate minerals (with the exclusion of metals and metal oxides) in one resource.
2.1.3Metals and metal oxides.Silver is by far the most widely used metal for the generation of functional (bio)nanocomposite materials, due to its long time known antimicrobial properties. 306Compared to other antimicrobial agents (e.g., molecular antimicrobial), silver is effective against a wide spectrum of bacteria as well as fungi, algae, and possibly some viruses. 307In addition, being an element, silver is shelf stable for an extended temporal window, which is a pivotal aspect for several applications.More recent uses of silver as an antimicrobial agent concern silver nanoparticles (AgNPs).It is believed that the profound activity of AgNPs is due to two main mechanisms: (i) AgNPs act as a source of Ag + ions, which detach from the surfaces of AgNPs and interact with negatively charged biomacromolecular components (disulde or sulydryl groups of enzymes) and nucleic acids, causing structural changes and deformation in bacterial cell walls and membranes that lead to a disruption of metabolic processes, followed by cell death; 308 and (ii) AgNPs bind membrane surfaces, causing morphological changes (pitting) with subsequent loss of structural integrity and impermeability. 309Besides medical and pharmaceutical applications, AgNPs are increasingly used in the food industry; for example, to produce (bio)polymer nanocomposites for food packaging applications.Compared to silver zeolites (which will be discussed shortly), AgNPs-based nanocomposites provide slower silver ion release rates into stored foods, but also lower acute antimicrobial responses. 310From a practical point of view, this means that while a zeolite-based material might offer a superior immediate effect, the sustained antimicrobial activity of the nanocomposite would be better suited for the packaging of foods that require long transportation distances or storage times. 4ilver nanoparticles based on silver salts or metallic silver have been demonstrated to be readily incorporated into thermoplastic packaging polymeric materials such as PE, PP, PS, and nylon. 311Sánchez-Valdes et al. coated a ve layer (PE/tie/PA-6/tie/PE) plastic lm with an AgNPs/polyethylene nanocomposite layer and found antimicrobial activity against the fungus A. niger, a common food contaminant. 312From a practical point of view, the same authors interestingly observed that the activity was dependent on the coating method: methods that gave rise to a rougher surface (and hence more surface area for silver-ion release) had higher activity than those that resulted in a smoother surface.Colloidal silver particles have also been coated 90-150 nm thick onto paper using ultrasonic radiation, and this coated paper was shown to manifest excellent antimicrobial activity against E. coli and S. aureus,  suggesting its potential application as a food packing material for longer shelf life. 3139][320] It has been pointed out that to make the Ag-based (bio)nanocomposite materials fully effective it is of utmost importance to consider various factors such as particle size, size distribution, degree of particle agglomeration, silver content, and the interaction of silver surface with the base polymer. 321This is because the antimicrobial efficacy of AgNPs, as any other nanocomposite antimicrobial system, strongly relies on the high surface-to-volume ratio and enhanced surface reactivity of the nano-sized antimicrobial particles, making them able to inactivate microorganisms more effectively than their micro-or macro-scale counterparts. 322nother metal with well-recognized antimicrobial properties is copper, although its potential biocidal activity is lower compared to Ag ions (about 10 mg Cu 2+ kg À1 in water is necessary to kill 106 cells of Saccharomyces cerevisiae). 323The activity of copper ions has been exploited only in few polymer nanocomposites, such as Cu-PE nanocomposite lms 324 and Cu-chitosan bionanocomposite lms intended for food packaging applications. 325The main reason for this restricted use is probably because copper is regarded as toxic in contact with food, and it would accelerate biochemical deterioration with foods due to its catalytic action of oxidation. 326etal oxides also exhibit excellent antimicrobial properties against a wide spectrum of microorganisms.For this reason, (bio)polymer nanocomposites based especially on TiO 2 , ZnO, and MgO have been developed for food packaging applications.Among them, TiO 2 nanocomposites are the most widely investigated.The antimicrobial properties of TiO 2 lie in its photocatalytic activity, which is intimately linked to its crystal structure.There are three different forms of titanium dioxide (rutile or anatase, tetragonal, brookite, and orthorhombic) with different reactivity depending on their characteristic band gap.The irradiation of TiO 2 at higher energies than the band gap generates electron-hole pairs, while giving rise to redox reactions.Negative electrons generate O 2À , and positive electric holes generate hydroxyl radicals.Reactive oxygen species oxidize organic molecules, and kill bacteria and viruses. 323TiO 2 coated polypropylene lms illuminated with UV light sources, were effective in decreasing the counts of E. coli in in vitro experiments up to 3 log 10 CFU g À1 , but also during the storage of lettuce a reduction over 1 log 10 CFU g À1 was observed. 327VOH lms loaded with TiO 2 nanoparticles were produced by Cerrada et al., specically for food packaging applications. 328he authors observed the effective photo-activated biocidal properties of the lms against the nine microorganisms (bacteria and yeasts) cited to be involved in food poisoning and spoilage.From a practical point of view, the main limitation of TiO 2 nanoparticles for the food packaging sector is due to its low photon utilization efficiency and necessity of the ultraviolet (UV) as an excitation source (i.e., TiO 2 -based antimicrobials are only active in the presence of UV light). 329Besides the antimicrobial activity, TiO 2 nanoparticles may provide new features to the advantage of the overall performance of the nal packaging material.In principle, food packaging lms incorporating TiO 2 nanoparticles may have the additional benet of protecting food content from the oxidizing effects of UV irradiation while maintaining good optical clarity, as TiO 2 nanoparticles are efficient short-wavelength light absorbers with high photostability. 4As for AgNPs and TiO 2 , their photoactivity can oxidize ethylene to H 2 O and CO 2 . 323TiO 2 nanoparticles have also been used to produce oxygen scavenger lms. 330In addition, incorporation of TiO 2 into synthetic plastic matrix has shown to increase the biodegradability. 331Another interesting development concerns the 'intelligent' packaging nanocomposite lms, i.e., those intended to monitor the condition of packaged food or the environment surrounding the food.Lee et al. developed an UV-activated colorimetric oxygen indicator, which uses nanoparticles of TiO 2 to photosensitize the reduction of methylene blue (MB) by triethanolamine in a polymer encapsulation medium using UVA light.Aer UV irradiation, the sensor bleaches and remains colorless until it is exposed by oxygen, when its original blue color is restored.The rate of color recovery is proportional to the level of oxygen exposure. 332iO 2 thin lms can be prepared on several substrates by various techniques such as chemical vapor deposition, evaporation, magnetron sputtering, ion beam technique, chemical spray pyrolysis, electro-deposition, and sol-gel method.The sol-gel process is suitable for producing composite materials of high purity without multiple steps. 333inc oxide (ZnO) particles are also known for their antimicrobial activity.Compared to AgNPs, ZnO have some advantages: rst of all, the lower cost, which is related to the abundance of zinc as ubiquitous trace metal; and second, ZnO nanoparticles have also claimed to possess UV-blocking properties. 334Different ZnO-based nanocomposite lms specically intended for food packaging applications have been developed in recent years.For example, PVC lms coated with ZnO nanoparticles were reported to have antimicrobial activities against E. coli and S. aureus. 335In a more recent work, Li et al. also veried the potential of the nano-packaging containing ZnO nanoparticles during the storage of Fuji apple cuts, observing a better preservation of quality indicators such as ascorbic acid and polyphenol content, and lower counts of typical altering microorganisms. 336Nanocomposite lms of LDPE containing AgNPs and ZnO were claimed to exhibit antimicrobial activity, showing an impressive impact of the proposed nano-packaging on the shelf life of orange juice. 337It was also demonstrated that combinations of allylisothiocyanate, nisin, and zinc oxide nanoparticles coated on glass jars were able to effectively inactivate Salmonella in liquid egg albumen. 338It was observed that the antibacterial activity of ZnO nanoparticles increases with decreasing particle size. 339In addition, this activity does not require the presence of UV light (unlike TiO 2 ), but it is stimulated by visible light. 340.1.4Zeolites.Zeolites (Fig. 3a), also known as "framework silicates", are the only aluminosilicates in the form of hydrated crystals within the tectosilicates subclass, with a three-dimensional channel system giving rise to a highly porous structure, with pore size ranging from 2.5 to 15 Å (341) (Fig. 3b).Zeolites have the following general formula M x D y [A x+2y Si nÀ(x+2y) O 2n ]$ mH 2 O]; M and D represent univalent and divalent cations, respectively, that balance the excess of negative charge due to Al 3+ for Si 4+ substitutions in the tetrahedral framework.Water molecules and cations are extra-framework components, which are located inside the channel system and can migrate throughout its length.This explains the cation exchange capacity and dehydration/rehydration ability of zeolites.There is a large variety of naturally occurring as well as synthetic zeolites with different pore structures, crystal sizes, and chemical compositions.Common features of zeolitic materials include the following: adsorption capacity; catalytic sites with different strengths; sizes of cages and channels very similar to those of many molecules of interest; and high cation exchange capacity and selectivity. 342Concerning synthetic zeolites, adjusting the size, shape, and polarity allow enhancing or depressing the above properties according to the targeted application.On the other hand, also some natural zeolites are currently used in industrial applications.
With regard to zeolitic membranes, despite many announcements in the last 15 years that have claimed the release of new commercial products for industrial applications, 342 solvent dehydration remains to date the only industrial application able to exploit the multifaceted potential of zeolites protably. 343This represents quite a surprising circumstance, in light of several excellent results in laboratory studies on zeolite membranes specically intended for gas-and liquid-phase separations. 343Incorporation of zeolites into polymer gave rise to the fabrication of mixed-matrix membranes, 344 which developers claimed were superior in selectivity compared to the traditional synthetic polymeric membranes used in the petroleum and petrochemical industries for separation applications [345][346][347] such as pervaporation, e.g. for separation of volatile organic compounds (VOCs)-water mixtures, [348][349][350][351][352] liquid separation by reverse osmosis, 353 forward osmosis techniques, 354 and ultraltration. 355Zeolites have also shown great potential for the catalysis of a variety of hydrocarbon reactions, including cracking, hydrocracking, alkylation, and isomerization, based on their combination of high thermal stability, high acidity and shape selectivity. 356eolite nanoparticles are of particular interest also for their adsorption capability.For example, zeolites successfully removed the pungent smell components identied in a coffee aroma-containing gas, 357,358 reduced the odor and VOCs emissions of natural-our-lled PP composites, 359 and removed toxic and carcinogenic compounds (e.g.1][362][363][364][365] Present and possible future applications of natural and synthetic zeolites also concern the biomedical eld and the cosmetic sector. 366ilver ions (Ag + )-loaded zeolite micro-and nanocomposites are among the most widely used materials that include zeolites.In these composites, ionic exchange mechanisms replace the alkaline and alkaline earth metal ions naturally present within the zeolite with silver. 323Ag + ions, due to their inherent antimicrobial activity, allow the production of potentially protable new materials with antimicrobial features for use in many applications, including food contact and active food packaging scopes.For example, developers in Japan have used Agsubstituted zeolites as the most common antimicrobial agent associated with plastics to preserve packaged food products, in particular for those systems in which the use of volatile agents is not feasible, i.e.where the bioactivity is based on the migration of the antimicrobial molecule by direct contact with food. 367ccording with the Food and Drug Administration (FDA) Food Contact Substance Notication, in the USA, zeolite-based materials may be used in all types of polymers intended to come in contact with food, 368 while the European Food Safety Agency expressed a positive opinion of in 2005 about the use of two different zeolites containing silver ions in food contact surfaces. 369However, the use of Ag + -zeolites for food packaging applications is limited to 0.05 mg Ag + kg À1 of food, which represents the no observable adverse effect level for humans. 370he extensive research on the development of Ag + -zeolite antimicrobial food packaging systems has led to manifold applications over the last two decades, as reported in recent reviews. 29,323However, the use of Ag 2+ -zeolites nanoparticles for food packaging applications is still highly debated, due to some evidences pointing to potential risks associated with these new types of antimicrobial materials. 323ost recent uses of zeolites-based composites for food packaging applications concern membranes and lms for fresh or minimally processed fruits and vegetables, which require selective permeation of carbon dioxide, oxygen, ethylene, and water vapor.Most research in this eld involves ethylene, a plant growth regulator that plays a key role in physiological processes and during postharvest.Controlling the presence of ethylene in packages and storage environments could extend the shelf life of many fresh fruits and vegetables.Monprasit et al. incorporated a commercial hydrophobic zeolite (5 wt%) into low-density polyethylene (LDPE), achieving much higher ethylene permeability and selectivity, with concomitant appropriate tensile properties. 371In another study, 372 the quality of broccoli packaged in LDPE-containing zeolite (as an ethylene absorber) resulted as the most similar to fresh broccoli.Boonruang et al. compared the effects of the use of various types of packaging lms in maintaining the quality and extending the shelf life of mangoes fruit. 373The authors demonstrated that an ethylene-removing mechanism based on the use of zeolites incorporated into PE lms extended the shelf life of mangoes up to 40 days, as compared with 20 days for the control (i.e.nonpackaging treatment) and 5 days with common non-perforated PE lm.Another study described the use of zeolite as CO 2 absorber in combination with Na 2 CO 3 to develop a suitable packaging system for kimchi, with controlled pressure and volume conditions. 374Metal-doped (e.g.iron, palladium, and platinum) zeolites are suitable oxygen scavengers for products sensitive to oxidation, 323 as has also been demonstrated for zeolites with adsorbed terpenes or phenol derivatives loaded in packaging bags. 375Another study investigated the possibility of using zeolite modied with a series of organosilane compounds to control the gas permeability properties of membranes containing the inorganic ller. 346The addition of zeolite into PP lms made PP a water-and water vapor-sensitive material. 376evelopers have also produced PP/zeolite composite lms coated with aluminum to obtain a packaging material impermeable to light, impermeable to moisture from the outer surface, and permeable to water and adsorbing moisture from the inner surface. 377he use of zeolites as inorganic ller may also improve the mechanical and thermal properties of plastics commonly used in food packaging.Biswas et al. reported that increasing zeolite 4A content up to 50 wt% increased the Young's modulus and yield stress of linear low density polyethylene (LLDPE) from 127 to 179 MPa and 9.4 to 11.1 MPa, respectively. 378Yuzay et al. reported that increasing zeolite 4A concentration up to 5 wt% resulted in increased Young's modulus of PLA from 1295 to 1465 MPa and tensile strength from 62.5 to 67.6 MPa, although elongation at break slightly decreased from 6.9 to 6.6%. 379Other studies attributed the poor mechanical properties of zeoliteloaded PP to the weak interfacial interactions between polymer and inorganic ller. 376,380To address this issue, developers have improved the ller compatibility and mechanical properties of the composites by surface treatments of natural zeolite. 381imilarly, studies have reported that an activated natural zeolite has a compatibilizing effect on a recycled tetra-component polymer blend consisting of PP (40 wt%), LDPE, high density polyethylene (HDPE) (15 and 40 wt%, respectively), and polystyrene (PS) (5 wt%). 382eolites have also improved the thermal properties of oilbased polymers such as PP, 359,376,383 polymers of natural origin such as PLA, 384 sodium alginate, 385 and blends of oil-based and natural-origins polymers, such as PE/thermoplastic starch (PE/ TPS). 386.1.5Graphene.Graphene is the building unit of graphite, a three-dimensional layered mineral allotrope of carbon composed of several stacked layers of graphene 387 (Fig. 4a).More specically, graphene is a two-dimensional material composed of a single planar sheet of sp 2 -bonded carbon atoms packed in a honeycomb crystal lattice with large specic surface area (Fig. 4b).Although the rst reported method for production of graphene nanosheets dates back to 1970, 388 its individual layered form was rst discovered in 2004 through a micromechanical cleavage method using ake graphite as a starting material. 389Graphene nanosheets have extremely high Young's modulus values ($1000 GPa), 390 fracture strength ($125 GPa), 390 thermal conductivity ($5000 W m À1 K À1 ), 391 mobility of charge carriers ($200 000 cm 2 V À1 s À1 ), 392 specic surface area (calculated value, $2630 m 2 g À1 ), 393 fascinating transport phenomena (e.g., the quantum Hall effect 394 ), and gas impermeability. 395These properties make graphene one of today's most exciting materials, as it represents the best candidate for the enhancement of electrical, mechanical, thermal and gas barrier properties for advantageous exploitation in many areas, such as photovoltaics, biosensors, supercapacitors, super adsorbents and fuel cells, 396 just to provide some examples.Despite the broad spectrum of potential applications, there are still many challenges for graphene to reach its full potential.Among others, production costs represent the main limitation to large scale utilization, mainly due to the highly expensive and low-yielding methods and procedures to obtain graphite monolayers.
The rst attempt to produce graphene was based on the chemical conversion of graphite into graphite oxide by chemical oxidation, with subsequent dispersion and exfoliation in water or suitable organic solvents.Several researchers, such as Staudenmaier, 397 Hummers and Offeman, 398 developed oxidation methods to produce graphene oxide (GO) from graphite akes.][400] The new functional groups arising from chemical oxidation are mostly in the form of hydroxyl and epoxy groups on the basal plane, with smaller amounts of carboxyl, carbonyl, phenol, lactone, and quinone at the sheet edges. 401,402These highly reactive oxygen functionalities, besides making possible the graing of both small molecules and polymers onto GO sheets, are responsible for the strongly hydrophilic character of GO, such that intercalation of water molecules between the layers readily occurs, leading to an interlayer spacing between 6 and 10 Å, depending on the water content. 403ost recent methods to produce graphene fall within either the bottom-up or the top-down strategies.The bottom-up approach, based on the organic synthesis starting from small precursors (e.g.atoms, molecules, and so on) includes chemical vapor deposition (CVD), 404,405 epitaxial growth on silicon carbide [406][407][408] and metals, 409,410 reduction of CO, 411 and conversion of CO 2 . 412In particular, CVD and epitaxial growth produce relatively good-quality large graphene sheets with fewer defects, but are not suitable for the fabrication of polymer nanocomposites, because these require large quantities of graphene sheets.
Top-down methods are based on the exfoliation of graphite into few or single-layered graphene sheets by breaking the van der Waals-like forces between graphite layers.Top-down methods account for two main routes.The rst route, which includes a micromechanical cleavage based on "Scotch tape" 389 method and mechano-chemical methods such as ball milling 413 and roll milling, 414 consists of the separation of strongly layered graphite into individual graphene sheets without basal plane distortion.The second route involves the basal plane distortion of graphene sheets and includes sono-chemical methods such as sonication 415 and ultrasonication. 416Micromechanical cleavage of graphite, referred to as "Scotch tape" method, gave birth to renewed interest in graphene because it disclosed the huge potential of the material. 389Although the Scotch tape method provides large and high-quality sheets, it enables only limited production, which makes it suitable exclusively for fundamental studies. 389By contrast, sonication and milling techniques allow industrial-scale outputs that make them effective and low-cost strategies for mass production of highquality graphene sheets.
The use of graphene even at low loading, makes it possible to obtain nanocomposite polymer systems with highly improved properties, such as tensile strength and elastic modulus, electrical and thermal conductivity, thermal stability, gas barrier, and ame retardance.Due to this broad multifunctionality, graphene/polymer nanocomposites can nd application in various elds.These include the following: advanced biochemical and electrochemical applications with exible and electrically conductive properties; 417,418 sensors, including electrochemical biosensors 419,420 and prototype gas sensors; 421 thermal interface applications with high thermal conductivity, low electrical conductivity, and highly compliant nature (e.g.thermal pads and thermal adhesives); 422,423 electric and electronic engineering materials, such as touch panels or exible displays, where high optical transparency, good barrier properties, thermal stability, and mechanical strength are required; 404,[424][425][426][427] automotive engineering, e.g.parts where thermal, mechanical, and ame-retardant properties are necessary; 428 highly efficient visible-light plasmonic photocatalysts; 429,430 biomedical applications, where controlled release behavior, long term biocompatibility, better mechanical properties, faster tissue regeneration, and less thrombogenicity are sought-aer properties; 431,432 transportation and space technologies with high thermal conductivity; 433 piezoelectric applications; 434 aerospace and radar evasion applications with electromagnetic wave shielding effects at microwave frequencies; 435,436 electronic, energy storage, and photovoltaic devices with enhanced electrical conductivity; 405,437,438 package-free exible photovoltaic devices; 432,439 and anticorrosion coatings with outstanding barrier properties against both O 2 and H 2 O. 440 Potential use of graphene-based polymer nanocomposites also includes food packaging applications due to the expected enhancement of mechanical, [441][442][443][444][445][446][447][448][449][450] thermal, 442,444,445,[447][448][449]451,452,454 and thermo-mechanical properties 454,455 of the nal materials. Another bent arising from the use of graphene that can boost its application in the food packaging eld relies on its gasimpermeable atomic membrane, which would allow producing barrier materials against permeation of O 2 , 445,455,456 N 2 , 444,455 and He.444 Graphene exhibits a unique tunable water-permeable atomic membrane, which is of paramount importance for food packaging applications. For exaple, Geim et al. recently reported the tunable permeation rate of water across a GO membrane that is completely impermeable to other liquids, vapors, and gases.457 Yoo et al. have recently published a comprehensive review on the potential applications of graphene and GO to produce barrier polymers.458 In addition, a recent study tested the transparency and water resistance of graphene-based polymers, 450,456 making this the preferential ller for exploring new packaging applications.Although only few works have clearly addressed the potential impact of graphene/polymer nanocomposites in the food packaging area (Table 3), this number is increasing considerably, demonstrating the interest in graphene-based nanocomposites also for this sector.
The impact at biological level of the newly developed graphene-based nanocomposites is not yet a major focus of research.Because a thorough understanding of the biological behavior of nanomaterials would guarantee the sustainability of nanotechnology, [473][474][475][476][477][478] future research toward this topic must ll in the current lack of this important information.
In light of the recently obtained exciting results, which may make the production costs of graphene accessible, the use of graphene, even though it is in its early stages, may be the most promising approach to guide future advancement in the development of new, revolutionary packaging materials.

Fabrication of (bio) nanocomposites
To take full advantage of the great potential of (bio)nanocomposites for real applications, the integration of the selected nano building blocks in the polymer matrices is of primary importance.Many factors, including the exfoliation degree of the llers, their spatial arrangement (wrinkling or stretched), the morphology of the composite, and the dispersion state in the polymer matrix (stacked or agglomerated)-hence the extent of the interfacial interactions between ller and polymer-may dictate the nal performance of the nanocomposite packaging material. 479Accordingly, for example, the ultimate  barrier properties of nanocomposite polymers will be affected by nanosized llers in two specic ways: (i) a more tortuous path for diffusion of the permeant and (ii) local changes in the polymer matrix properties (e.g., molecular mobility) at the interfacial (polymer nanoparticle) regions. 4istorically, the nanoller incorporation into the polymer matrix takes place through three main methods, i.e. (i) the in situ polymerization, (ii) the solution casting or (iii) the melt processing. 15,60,480,481Most recently, sonication and high-shear mixing have been proposed as alternative techniques to prepare (bio)nanocomposite materials.

In situ polymerization
This process involves the polymerization of monomer species in the presence of the layered materials.In this process, the nanoclays are rst swollen within the liquid monomer or monomer solution, which is followed by its polymerization in between the intercalated sheets (Fig. 5a).Polymerization can be initiated either by heat or radiation, by the diffusion of a suitable initiator or by an organic initiator or catalyst xed through cationic exchange inside the interlayer and before the engorgement step when required. 17The macromolecules molecular weight increases, leading to a d 001 increase and sometimes to an almost fully exfoliated morphology for some studied systems. 17One of the main drawbacks of this method lies in the tendency of inorganic particles to phase separate and sediment quickly from the organic polymer.To enhance the interaction at the solvent/ller interface, specic groups have to be linked onto their surface to stabilize the nanoparticle dispersions. 482Another relevant aspect concerns the unsuitableness of this method for biomacromolecules such as proteins and polysaccharides (that are already extracted as 'polymerized' entities), which indeed represent target polymers for the generation of bionanocomposites. 60

Melt processing
In the melt processing technique, the nanoparticles are mixed with the polymer in the molten state (Fig. 5b).More specically, the process involves mixing the particles with the polymer and heating the mixture above the soening point of the polymer, statically or under shear interlayers. 483,484The main advantages of the melt processing method are the absence of any solvent throughout the process and its compatibility with current industrial processes, such as extrusion and injection molding.Several factors may affect the extent of exfoliation/intercalation by melt processing, such as the thermodynamic interaction between the polymer and the nanoparticle and the transport/ diffusion of polymer chains from the bulk melt into the silicate interlayers. 483,484To increase the compatibility between polymer and nanoparticles to ensure proper dispersion, two main factors have to be taken into consideration, namely the favorable enthalpic interaction between the polymer and the nanoparticle (which can be manipulated by chemical modication of the ller and/or the polymer) and proper processing conditions. 481In this regard, shearing is necessary to induce the platelets delamination from tactoids, while an extended residence time is needed to allow the polymer chains to diffuse into the inter-layer gallery and then obtain an exfoliated morphology.However, in the case of most bionanocomposites (e.g., PLA-, chitosan-, and pullulan-based nanocomposites), the melt processing technique is not adequate in principle because these polymers tend to degrade due to either the mechanical shearing force or the temperature applied during processing.For example, PLA experiences processing instability (i.e., thermal, oxidative, and hydrolytic degradations) may occur during processing, which leads to the cleavage of polymer chains and consequently to a decrease in molecular weight.This degradation could even be accelerated in nanocomposites that have organically modied clays, which could initiate the degradation. 481Therefore, careful optimization of the processing parameters is required for the majority of thermally sensitive biopolymers. 60

Solution casting
The solution casting method is based on a solvent system in which the polymer (or pre-polymer, in case of insoluble polymers) and any other component of the mixture (e.g., surfactants) is soluble.The polymer is usually dissolved in a suitable solvent while the nano-llers are dispersed in the same or a different solvent before the two are mixed together to generate a homogeneous dispersion (Fig. 5c).The main advantage of this method is the relatively rapid exfoliation of the stacked layers by the use of an appropriate solvent. 17,484The successive addition of polymer solution to the dispersion of the complete delaminated nanoparticles (e.g., platelets) leads to the strong interaction between macromolecules and individual layers.The driving force for the intercalation of the biopolymer into the clay galleries from solution is the entropy gained from the desorption of the solvent molecules, which compensates for the decreased entropy of the conned, intercalated chains.When the solvent is evaporated, the intercalated structure remains, which results in the nal nanocomposites. 481Due to the large amount of the solvent required, this method is perceived as unsafe and non-environmentally benign when organic solvents are required (e.g., for non-polar or highly hydrophobic polymers). 486Conversely, this method has gained increasing attention for water-soluble polymers such as PVOH, especially in the form of thin coatings, which reduces the amount of water used throughout the process.More recently, the solution casting method has been adopted for the generation of bionanocomposites, for which both in situ polymerization and melt intercalation are oen unsuitable due to the inherent characteristics of most biopolymers, as discussed before.From a practical point of view, the fabrication of (bio)nanocomposite lms and coatings through the solution casting method requires special attention during the removal of the solvent (evaporation) step.Indeed, if a small amount of solvent remains entrapped in the nal product, a lower interfacial interaction between the polymer and the ller can arise. 487For this reason, coupling infra-red lamps with high performance air ovens is the best strategy to prevent this potential drawback. 488

Sonication
Sonication is increasingly used in the top-down generation of nanoparticles.This is achieved through deagglomeration and reduction of microsized particles (e.g., tactoids) by means of sound waves (more frequently, ultrasound waves) as a result of the mechanical effects of the phenomenon called cavitation, which refers to the formation, growth, and implosive collapse of bubbles in a liquid (Fig. 6). 489The speed of sound in a typical liquid is 1000 to 1500 m s À1 , and ultrasonic wavelengths will vary from roughly 10 cm down to 100 mm over a frequency range of 20 kHz to 15 MHz, much larger than the molecular size scale.4][495] The extreme, transient conditions produced during acoustic cavitation allow the formation of unique materials and also permit syntheses on the benchtop in a room-temperature liquid that would otherwise require high temperatures, high pressures, or long reaction times.The most important physical phenomena for the preparation or modication of nanomaterials are microjets and shock waves.Microjets occur when bubbles collapse near an extended surface (i.e., surfaces several times larger than the bubble radius). 496In addition, shock waves can accelerate solid particles suspended in the liquid.][499] Ultrasonication, rst adopted to develop nanocomposites based on oil-derived polymer/inorganic clay systems, has Fig. 6 Schematic illustration of the ultrasonication method based on the acoustic cavitation.The sonotrode of an ultrasonicator generates high intensity ultrasound waves of alternating expansive and compressive cycles.This causes the formation, growth, and collapse of bubbles, which in turn yields intense shockwaves that promote interparticle collisions and, eventually, the exfoliation of the layered filler.
1][502][503] More recently, ultrasonication has been suggested as a valuable tool for the preparation of graphene-based nanocomposites starting from graphite akes or particles dispersed in a liquid medium, both non-aqueous and aqueous systems.Due to the signicant advantages of aqueous systems over non-aqueous systems (e.g., lower costs and fewer potential health risks and environmental issues), the former approach (i.e., using water as a solvent) has been attracting much attention over the recent years.Lotya et al. 504 rstly reported the ultrasonication-assisted exfoliation of graphite in an aqueous system using a cationic surfactant, i.e., sodium dodecyl benzene sulfonate (SDBS).A later development accounted for a 'one-pot' procedure to obtain graphene-based bionanocomposite materials, which involves the use of biopolymers directly in the ultrasonication reactor, without the necessity to recover the nanoparticles in a second step.According to this method, it is believed that biopolymer works similarly to surfactants (i.e., as a spacer), but without the formation of micelles.More specically, the biopolymer adsorbs onto the exposed surfaces of the graphene layers, thus preventing any possible aggregation or restacking phenomenon, which would represent the simplest, most efficient, and environmentally friendly strategy for the preparation of bionanocomposite materials. 505Examples of direct exfoliation of graphite into few layers graphene sheets by (bio)polymerassisted ultrasonication concerned poly(vinyl pyrrolidone) (PVP) and PVOH, 506 a pyrene-functionalized amphiphilic block copolymer, 507 pyrene polymers, 508 gum arabic, [509][510][511] and acrylate polymers. 512

High-shear mixing
High-shear mixing methods for the exfoliation of layered materials have been developed more recently compared to other routes in an attempt to provide an alternative approach that is able to boost the progress from the laboratory to commercial applications.4][515] A recent study has demonstrated that graphite can be exfoliated to give graphene by generating shear in a very thin liquid layer in a rapidly rotating tube. 516However, this method gives very small quantities of graphene and is inherently un-scalable.Recently, Paton et al. demonstrated a method to exfoliate graphite into graphene nanosheets by high-shear mixing using a closely spaced rotor/stator combination. 517This method is claimed to be many times more efficient than sonication in terms of yield, enabling large quantities of defect-free, unoxidized graphene to be obtained.In addition, the same method can be scaled-up to an industrial level.The same authors developed a simple model showing that exfoliation occurs once the local shear rate exceeds 10 4 s À1 in liquid volumes from hundreds of milliliters, up to hundreds of liters and beyond.
Ball milling is another commonly used high-shear mixing technique, which involves a solid-state mixing at room temperature. 518It is a simple and efficient technique based on the pure shear among balls of various diameters, which is able to break the van der Waals interactions between layers. 519Clay dispersion is thus promoted by the energy transfer between milling tools (generally balls) and polymer/clay mixture, which in turns results grinded and intimately mixed. 513At the same time, the newly obtained layer would be wrapped up by polymer matrix, not only preventing the layers from sticking together, but also keeping the inherent structure of the layers. 519Among the available techniques used for nanocomposites preparation, ball milling has the considerable advantage of not requiring the use of high temperature or solvent treatments, which makes the composites preparation more green, convenient and effective. 520he ball milling method was introduced to fabricate bionanocomposites such as MMT/pectin 513 and LDH/PCL. 514 Technological aspects linked to the manufacture of (bio)nanocomposite coatings The success of (bio)nanocomposite materials strongly depends on some important factors during the design and development steps.Besides signicant costs, technological aspects may act as the "go-no go" gate before market applications.These aspects concern the coating system both before (e.g., the coating suspension) and aer (i.e., the dry (bio)nanocomposite coating) the deposition on the selected substrate.

Compatibility between polymer and ller
The affinity between organic and inorganic components at the hybrid interface plays a key role in determining the nal performance of (bio)nanocomposite polymer systems.Maximizing the advantages of the interface effect between polymer and ller therefore requires selecting the best possible starting materials.In this respect, a signicant difference applies between petrol-based polymers and biopolymers.Most plastic polymers derived from oil are inherently hydrophobic (e.g., polyolens), while inorganic llers (e.g., clays) are naturally hydrophilic, thereby resulting in a marked incompatibility between phases.
To enhance the compatibility with oil-based polymers, production methods generally replace the alkali or alkalineearth interlayer cations by means of ion exchange reactions with organic ammonium and phosphonium cations to reduce the surface energy of the inorganic phase. 13In other circumstances, an alternative approach consists in the tailored functionalization of the ller by graing new moieties with high affinity for the polymer.For example, researchers have achieved the interfacial adhesion between graphene and the polymer matrix by chemical functionalization of graphene surface via "graing from", 521-523 "graing on", 399,524,525 and "graing to" 526 techniques.Unlike common plastics, most biopolymers are hydrophilic, which makes their miscibility easier with natural llers such as cellulose-derivatives, protein-derived whiskers, and especially clays.
In practice, enhancing the compatibility between organic and inorganic components allows the minimization of phase separation during processing as well as successful intercalation, swelling, and exfoliation of nanoclays.In turn, this will yield an even dispersion and distribution of the ller within the polymer matrix, avoiding the formation of traditional lled polymers with fewer applications due to the depression of the "interface effect".Eventually, efficient dispersion of the ller in the hosting matrices should enable low-cost solutions [527][528][529][530] for sustainable and lightweight nanocomposite materials.

Rheological properties
Manufacturers of nanocomposite polymer systems widely use rheological measurements, as they represent an indirect method to assess the state of dispersion of the nano-objects (i.e., llers), such as the degree of intercalation, exfoliation, and dispersion in the polymer matrix. 531Because most synthetic routes adopted to obtain the nal nanocomposite polymer system (i.e., solution processing, melt processing, in situ polymerization, and mesophase mediated processing) involve ow, rheological properties of polymer nanocomposites may help to achieve optimum process conditions 17,532,533 while an understanding of the rheological behavior can provide insightful information on the structure-property relationship in the nal materials. 17lthough melt intercalation is the most widely adopted approach for the preparation of nanocomposite materials, the use of solvent-based systems is predominant within the coating technology, whereby water is the main (sometimes the unique) solvent when biopolymers are used as the organic matrix.The study of the rheology of nanoclay suspensions has a relatively long history because of the application of nanoclays as thickening and thixotropic agents in different elds than food packaging, such as cosmetics, inks, and paints.Even ller concentrations as low as 4-5 wt% can drastically affect the rheological properties of these suspensions, particularly the viscoelastic behavior.5][536] Van Olphen (1964) rst described different nanoplatelet associations: (i) face-to-face, (ii) edge-to-face, and (iii) edge-to-edge aggregations (Fig. 7). 537ace-to-face association leads to thicker and larger tactoids, whereas edge-to-face and edge-to-edge lead to three-dimensional "house of cards" structures.Aggregation is described by the face-to-face association, whereas the continuous gel-like structure is a consequence of edge-to-face and edge-to-edge association.Moreover, the nanoclay dispersion could be destabilized by salts (coagulation) or by polymers (occulation).One of these systems has to be realized as a function of the nal application.To achieve a true nanocomposite polymer system with improved properties compared to conventional microcomposites, the ller must be uniformly dispersed and not aggregated.This will rst allow developing the highest proportion of an interfacial zone, thus maximizing the adsorption of the polymer chains onto the sheets. 535Secondly, efficient dispersion of the ller is the key to obtaining suspensions that are stable over time.9][540] In fact, during rest periods, all solid particles connect and form a gel characterized by a perceptible limiting shear stress, or yield stress, at which the material begins to ow: if the system is subjected to a constant shear rate, viscosity decreases with time as the gel structure is broken down, until equilibrium viscosity is reached.Most gelling colloidal systems follow Herschel Bulkley's rheological model; in contrast, most non-gelling colloidal dispersions follow Ostwald's rheological model (or power law). 536[536][537][538][539][540][541]

Optical properties
Optical properties of materials are particularly important in certain sectors, where they can impact either the performance of the nal material or the consumer's choice.In the food packaging eld, both aspects are relevant and worth high consideration when designing a new (bio)nanocomposite coating.On one hand, UV radiation (wavelengths below $340 nm) should be prevented, because it may cause photooxidation of photosensitive foods such as meat, beer, and milk, resulting in changes in color, avor, and taste.On the other hand, high transmittance of visible light (wavelengths between $340 nm and $800 nm) should be guaranteed at the same time, as it allows consumers to see through the package (visual inspection of the packaged food).
Researchers have developed (bio)nanocomposite systems (e.g., coatings) with UV-shielding/absorbing properties based on the use of metal oxides such as TiO 2 (ref.542 and 543) and ZnO 544 or clays. 167However, aer coating deposition and solvent removal by evaporation, the optical properties (e.g., transparency) of the substrate beneath can dramatically be impaired due to the presence of the ller.This can be due to three main reasons: (i) unsuccessful dispersion and distribution of the ller within the polymer matrix, generally caused by either erroneous selection of the (bio)polymer/ller system (incompatibility) or inefficient deagglomeration procedure/set up; (ii) high concentration of the ller-increasing the ller volume fraction leads to a proportional increase in haze due to an increase in scattering centers, 19 and reaggregation phenomena-for the same ller volume fraction, the reaggregation of previously-exfoliated platelets may occur as a consequence of high energy input to the system (e.g., prolonged ultrasonication time), with subsequent restoration of the original macro-sized tactoids.In this case, also the size of the scattering centers play a role in decreasing the transparency of the (bio)nanocomposite coating. 19herefore, the effect of the (bio)nanocomposite coatings on the optical properties of the substrate (e.g., plastic lms) can be controlled by two main routes: (i) selecting suitable methods and procedures throughout the overall manufacturing process, and (ii) selecting the most appropriate ller type.Regarding procedure selection for example, in the exfoliation step of the ller, physical, chemical, or mechanical methods can be more or less efficient depending on the ller.Moreover, regardless of the specic methods, pinpointing the best process conditions (e.g., time, temperature, energy input, amount of additives, solvents and co-solvents, and so on) may dictate the nal size of the ller (macro-sized aggregates or nano-sized entities).In selecting the most appropriate ller type, depending on the nal application, the nal choice will affect the optical properties of the nal materials differently.For example, if the transparency of the nal material is mandatory, the synthetic clay LAP offers higher performance over sheet-like and needlelike clays such as MMT and SEP, respectively, due to both its well-dened chemical composition and dimensions and lack of common inorganic components (e.g.silica and iron oxides) generally carried in natural clays, which allows the production of clear and transparent colloidal water suspensions.

Surface properties and the scalping effect
The addition of llers to the formulation of coatings may also have an impact on the surface properties of the nal material.In food packaging, the surface properties play a major role in several converting operations, such as printing, lamination, and co-extrusion.As a general rule, the addition of an inorganic ller, especially for a high ller volume fraction, is rst reected by changes in the surface topography, i.e. an increase in the surface roughness, which is linked to the spatial distribution of the llers within the coating thickness (Fig. 8). 19,167,545For example, platy clays with large surface areas oen display a "house-of-cards" or "cell-like" conguration aer solvent removal. 19An increase in roughness, besides leading to an increase in the haze of the nal material, 546 may have great impact on the wettability attribute of the surface, as described by both Wenzel's and Cassie-Baxter's theories. 547,548In addition, highly porous llers (e.g., zeolites) as well as llers with high affinity towards water molecules (e.g., natural MMT) might boost absorption phenomena at the solid/liquid interface. 167ot only the surface properties, but also mechanical properties can be inuenced by an increase in roughness.][551] Whether this change is positive or negative strongly depends on the nal application.For example, low coefficient of friction values aid the unwinding operations of plastic lms on industrial lines, to avoid the "blocking effect" on the reels and decreasing overall throughput.Conversely, lower coefficient of friction values are problematic if the nal packages (e.g., plastic bags) have to be stacked on top one of another; high coefficients of friction can keep the stack from collapsing.
Finally, but not less important, is the adverse effect that can potentially arise from the use of (bio)nanocomposite coatings on the overall quality of packaged food."Scalping" is the sorption of food constituents, especially aroma compounds, by the packaging materials.Both porosity and chemical affinity of the ller for aromas and volatile molecules must be taken into account in the design of a (bio)nanocomposite coating, especially if it will face the inner side of the package and be close to the food.Scalping phenomena have received considerable attention within the food packaging industry as they may negatively inuence consumer acceptance of the food product due to loss of aroma intensity or the development of an unbalanced avor prole. 552

Advantages of bionanocomposite coatings in food packaging
From an industrial perspective, the development of (bio)nanocomposite coatings has aimed at improving three main properties of the base resin: mechanical (stiffness, impact, and wear resistance), ame-retardance (anti-ammable materials), and barrier properties (against gases, vapors, and radiation).In the food and beverage packaging sector, however, considerable industrial and research developments of nanocomposite coatings have focused on improving the barrier properties, which has led to several applications in recent years.The excellent barrier properties of (bio)nanocomposite coatings may be explained by the "tortuous path" theory, 25 i.e. the formation of a labyrinth ensuing from the spatial dispersion of nano-clay platelets within the polymeric structure, which physically decreases the diffusion of the permeant (e.g., gases and vapors) across the thickness of the material.
Studies have reported examples of applications of modied atmosphere packaging for food (meats, cheese, confectionery, cereals and "boil-in-the-bag" foods), extrusion-coating applications in association with paperboard for fruit juice and dairy products, and in combination with co-extrusion processes for the manufacture of beer and carbonated drink bottles. 553everal nanocomposite materials intended for food packaging applications are available in the market.Nanocor and Mitsubishi Gas Chemical Company developed the MXD6 high barrier semi-aromatic nylons in nanocomposite form. 554The latest development of the resin MDX6 led to the Imperm® products, a family of superior gas barrier resins particularly useful in extending package shelf life. 555Honeywell has developed a polyamide-based nanocomposite with passive barrier properties against oxygen under the Aegis® trade name.Passive barriers use clay particles, which render the oxygen transmission inside the composite more difficult. 556The U.S. military and NASA, in conjunction with Triton Systems, Inc., worked together on nanoclays as barrier enhancers for ethylene vinyl alcohol (EVOH) in long-shelf life packaging (i.e., non-refrigerated food).This collaboration led to an experimental thermoformed food tray from EVOH loaded with 3% of Southern Clay's Cloisite® in a layer sandwiched between two PP layers.It reportedly imparts a three-to ve-years shelf life without refrigeration, plus good clarity, processability, and recyclability. 557InMat Inc. developed Nanolok™, a high barrier, waterbased, environmentally friendly nanocomposite barrier coating for transparent packaging applications.They claim the nanocomposite coating provides an excellent oxygen barrier up to 80% relative humidity and is more cost effective than EVOH. 558inally, NanoPack Inc. has developed a water-based coating made of PVOH and VMT.This coating, sold under the Nano-Seal™ trade name, is specically intended for food packaging applications where protection of food against oxygen and aromas is required. 559esides advantages directly linked to their functional properties, the use of (bio)nanocomposite coatings may indirectly provide additional benets, oen underestimated or neglected.For example, the weight advantage could present a signicant impact on environmental concerns.According to the "packaging optimization" principle, 560 the use of high-performance thin layers may be a valuable approach to down-gauging current packaging structures, e.g.laminates and/or co-extruded materials (Fig. 9).(Bio)nanocomposite coatings thus have the great potential to allow reducing the thickness of the package without jeopardizing (rather improving) the overall performance, at  reasonable cost.In turn, this would mean an upstream reduction of plastics, hence a less amount of plastics dumped into the environment, which would help in facing the waste disposal issue.From a strict economical point of view, the use of (bio) nanocomposite coatings to obtain lighter packages has the additional advantage to reduce the energy inputs for production, transport, and storage.Of course, the higher the degree of multi-functionality integrated in the (bio)nanocomposite coatings, the wider the environmental and economic advantages.

Concluding remarks
The development of new packaging materials offering new functionalities, less environmental impact, and economical benets is now an urgent necessity.The driving forces behind this pressing task are manifold.On one hand, the shelf life extension of packaged products can be an answer to the exponential increase of worldwide demand for food.On the other hand, the increase in the price of crude oil, together with the uncertainty related to its durability, has imposed the necessity to manufacture new structures to replace oil-derived polymers partially or totally in the future.Finally, consumers' awareness towards environmental issues increasingly pushes industries to look with renewed interest to "green" solutions.
Coating technology is one of the most efficient and convenient approaches to exploit toward this goal.This versatile technique enables manipulating matrices of completely different origins and compositions (e.g., water-based/organic solvent-based, biopolymer/oil-derived polymers, solutions/ dispersions) to yield thin layers that ennoble the substrate beneath (e.g., plastic webs) by conferring new (sometimes unexpected) functional properties.Within a shelf life extension perspective, the generation of biopolymer nanocomposite coatings is one of the most striking and promising outcomes of coating technology for tomorrow's challenges.

Stefano
Farris received his M.S. in Agricultural Science & Technology from the University of Sassari, Italy, in 2004.In 2007 he earned his PhD in Food and Microbial Biotechnology.From 2007 to 2008 he was a postdoctoral fellow at Rutgers University (NJ) in the Food Packaging lab led by Prof. K. Yam, where he worked on the development of lms and coatings from renewable resources.In 2011 he joined Prof. M. Hedenqvist's group at KTH, the Royal Institute of Technology in Stockholm, in the Department of Fibre and Polymer Technology.There, he worked on the development of hybrid materials by means of in situ polymerization through sol-gel approaches.At present he is Assistant Professor in the Department of Food, Environmental and Nutritional Sciences (DeFENS) at the University of Milan, Packaging Lab.His current research activity is mainly devoted to the development and deposition of high-performance coatings.

Chart 1
Classification of inorganic fillers based on chemical composition and crystal structure.nanocompositematerials.Although rectorite does not belong to the smectite group, it has arbitrarily been included in this section because it is partially constituted by montmorillonite layers.2.1.1.1.1Montmorillonite.Montmorillonite (MMT) (Fig. 2a) is the most widely exploited clay for the fabrication of polymer/ clay nanocomposites. 68In MMT's structure, with a general formula (M y + $nH 2 O)(Al 2Ày Mg y )Si 4 O 10 (OH) 2 (M y + refers to a generic monovalent interlayer cation), 67 the negative surface charge coming from the partial substitution of the trivalent Al-cation in the octahedral gibbsite layer by the divalent Mgcation is balanced by sodium and calcium ions, which exist hydrated in the interlayer. 69Because the layers are held together by relatively weak forces, water and other polar molecules can enter between the layers, causing the lattice to expand. 70A typical sodic MMT (Na + -MMT) is characterized by a d 001 of 11.7 Å (at relative humidity values $50%), aspect ratio of 100-200, surface area of 700-800 m 2 g À1 , and CEC ranging between 70 and 120 meq per 100 g, depending on their extraction site. 71,72
LDHs have a general chemical formula represented by [M 1Àx M x III (OH) 2 ] y+ [A y/ m mÀ $nH 2 O]; M is an univalent or a divalent metal ion such as

a
Summary of main physicochemical features of most common clay and clay-likefillers used in the food packaging sector This is due to the co-existence of both smectite-like and mica-like layers.b Once dehydrated.c Depending on the pH.d At pH ¼ 7.

Fig. 4
Fig. 4 (a) TEM image of graphene sheets (dark areas denote overlapping zones); (b) HRTEM image of 3 overlapping zigzag-armchair edges obtained by applying Joule heating to a graphitic nanoribbon inside a HRTEM; hexagonal models are depicted for clarity (reproduced with permission from Elsevier Ltd 387 ).

463 T 3 ( 4 C
Representative polymer-graphene/graphene derivatives nanocomposite systems with potential application in food packaging a,b Transmittance at 0.5% concentration is over 82% 453 Transmittance sharply decreases at 1.0% concentration Oxygen permeability (23 C, 0% RH) at 5% concentration decreases by 20.7% of neat PDN Young's modulus and tensile strength at 5% concentration increase by 49.1% and 63%, respectively; elongation at break decreases by 58% of PDN lm PS/PIFG Hot pressing 0.02-2.27vol% Transmittance $40% at 0.02% concentration; full opacity at 0.24% concentration 459 Oxygen permeability (23 C, 0% RH) at 0.02% concentration $80% of neat PS PLA/GO or PLA/GNP Doctor blade 0.2-0.8wt% GO Transmittance $72% and $23% at 0.6% concentration GO and GNP, respectively 460 0.2-1.0wt% GNP The largest decreases in oxygen and nitrogen permeabilities at 0.4% GO and GNP concentration are 70% and 60% of neat PLA, respectively Young's modulus and yield strength at 0.3% GO concentration increases by 115% and 95% compared to the neat PLA lm Young's modulus and yield strength at 0.4% GNP concentration increases by 156% and 129% compared to the neat PLA lm PPC/GO Compression molding 0.1-1.0wt% Young's modulus and tensile strength at 1% concentration increase by 1800% and 1000%, respectively; elongation at break decreases by $50% compared to neat PPC lm, though still remaining as high as 700% 461 Storage modulus at 1% concentration at 30 C increases up to 1600 MPa (storage modulus of neat PPC $30 MPa) T g of PPC increases from 29 C to 39.5 C aer loading 1% GO Chitosan/GNP or starch/GNP Solvent casting 0.2-3.0wt% Young's modulus and tensile strength at 1.5% concentration increase, further concentration increase results in decrease in both chitosan and starch lms; elongation at break decreases as concentration increases compared to the neat chitosan and starch lms 462 The T g values do not exhibit a signicant effect attributable to GNP (T g $ 72 C) WVTR of starch and chitosan (25 C, 50% RH) at 3% concentration decreases by 24% and 17%, respectively, of neat lms The maximum increase in tensile strength is at 2% EFG by 50%; elongation at break decreases by 89% as concentration increases compared to the neat lm g of PPC increases from 48.3 C to 52.1 C aer loading 5% EFG OTR (23 C, 0% RH) and WVTR (37.5 C, 90% RH) at 5% concentration decreases by 45% and 18%, respectively, of neat PPC Oxygen and CO 2 permeability (room temperature, 50% RH) at 1.37% concentration decreases by 45% and 68%, respectively PLA/GO Ultrasonication/ compression molding 0.17-1.37vol%T g is increased by 15.6 C at 1.37% retarding the degradation of the PLA 464 Transmittance of neat PLA lm is about 60%, loading is above 0.35%, and nearly no UV light is penetrated EVOH/TRG Ultrasonication/solvent casting 0.5-4.0wt% Young's modulus and tensile strength at 4% concentration increase by 400% and 280%, respectively, elongation at break decreases by $87% compared to neat EVOH lm 465 Only 0.5 wt% TRG loading is signicantly decreased oxygen permeability (room temperature, 37% RH to 8.517 lms increases from 65.31 C to 73.81 C aer loading 1.5% TRG and then gradually shis to high temperatures with increasing TRG loading PU/HD-GNR Sonication/solvent casting 0.1-0.5 wt% Nitrogen gas effective diffusivity of TPU was decreased by 3 orders magnitude with only 0.5 wt% HD-GNRs 466 The modulus increased and peaked at 0.1 wt% HD-GNRs, and then gradually decreased upon further ller additions This journal is © The Royal Society of Chemistry 2014 Table RH) at 1.5% concentration decreases by more than 99% compared to neat PET 467 T g does not exhibit a signicant GNP effect.GNP stabilizes the PET-manifested in increasing its char residue by 5 wt% at 1.5 wt% GNP content at 600 Young's modulus, tensile strength, elongation at break, and fracture energy of PBS at 2% concentration increase by 70, 53, 12, and 100%, respectively 468 Tm does not exhibit a signicant GO effect while T g increased by 5 C with addition of 2.0% GO (PEI/GO) n multilayer PLA Sonication/dipping 0.02-0.06wt% Compared with the pure PLA lm (the transmittance of 92.3%), the multilayer lm coated with (PEI/GO) 6 still remained a high visible light transmittance (84.4%) 469 OTR (23 C, 0% RH) at 0.06% GO of PLA lm coated with 4 layer decreases by 98.do not exhibit a signicant EFG effect.Ti is increased by 14.at 0.5% retarding the degradation of the EVOH and further loading results in decreasing 470 OTR at 23 C, 0% RH and at 23 C, 80% RH at 2% concentration decreases by 97% and 77%, respectively, compared to neat EVOH HDPE/DA-RGO or HDPE/DA-E 0 of the DA-GO/HDPE and DA-RGO/HDPE nanocomposites with only 0.1 wt% ller content increase by 7.1% and 7.4% from, respectively, whereas the E 0 of the nanocomposites with 1.0 wt% ller increases by 17.3% 471 With only 0.1 wt% DA-GO loading the P O 2 of neat HDPE decrease by 62.3%.Only 66.7% decrease of P O 2 can be obtained by addition of 1.0 wt% DA-GO.With 0.1 wt% and 1.0 wt% DA-RGO contents, the P O 2 of DA-RGO/HDPE nanocomposites decreased by 63.6% and 67.0%, respectively The temperature at 50% weight loss of HDPE only increase by 7.2 and 6.4 C at 1% DA-GO and DA-RGO nanocomposite, respectively HPMC/GO Ultrasonication/solvent casting 0.02-1.3wt% Young's modulus, tensile strength, elongation at break, and fracture energy of PBS at 1.3% concentration increase by 201, 122, 120, respectively 472 The temperature at 50% weight loss of HPMC increase by 8 C at 1.3% GO WVTR (75% RH) at 1.3% concentration decreases by 44% compared to neat HPMC a Please refer to the main text for abbreviations.b Ti: initial degradation temperature, TRG: thermally reduced graphene, EFG: exfoliated graphite, HD-GNR: hexadecyl-functionalized low-defect graphene nanoribbons; DA-GO: dodecyl amine (DA) functionalized graphene oxide and DA-RGO: dodecyl amine functionalized reduced graphene oxide; E 0 : storage modulus.

Fig. 9
Fig. 9 (a) Schematic illustration of the 'packaging lightweighting' concept: the use of multifunctional nanocomposite coatings makes possible avoiding multiple layers packages; (b) optical microscope cross-section of a multi-layer (total thickness 90 mm) package (left) and SEM of a nanocomposite coating ($0.7 mm) on a 12 mm PET substrate (right).
Yang et al. reported for Na + -REC a CEC $45 meq per 100 g and d-spacing of 22.1 Å, 186 while Jinhong et al. calculated a CEC of 32.2 meq per 100 g and d-spacing of 19.47 Å.

Table 1
Effect of layered silicate and non-silicate mineral incorporation on the properties of bionanocomposites for potential applications in the food packaging sector a ) is another 2 : 1 trioctahedral smectite with a theoretical chemical formula (M y + $nH 2 O)(Mg 3Ày Li y )Si 4 O 10 (OH) 2 ; 67 one study reported the CEC and surface area of HEC are 70-80 meq per 100 g and $200-300 m 2 g À1 (measured by N 2 adsorption at liquid nitrogen temperatures), respectively. 219Sànchez et al. measured a BET surface area of $220 m 2 g À1 and basal spacing between 13.4 Å and 14.2 Å.