Packaging plays an essential role in the food supply chain, protecting and containing food from the processing and manufacturing stages, along with distribution, handling and storage, until it reaches the final consumer. Currently, food packaging accounts for the largest share in the total packaging sector (85%). The global packaging market revenues increased from $
42.5 billion in 2014 to nearly $
48.3 billion by 2020 [1
]. Plastics are the most used packaging materials, bearing lightweight, with good processability, low production cost and outstanding mechanical and barrier properties [2
]. The plastic packaging market has been expanding with a growth rate of 20–25% per year [1
]. However, there is an increasing concern regarding the massive use of petroleum-based plastics, which are used for a short period of time but then take centuries to degrade. Most synthetic plastics used in packaging are recyclable. In the scope of the European Strategy for Plastics in a Circular Economy, the Directive (EU) 2019/904 was published to prevent and reduce the impact of single use plastic products on the environment, in particular the aquatic environment, and on human health, as well as to promote the transition to a circular economy [3
]. Additionally, the Directive (EU) 2018/852 sets a minimum of 70% by weight of all packaging waste as target for recycling and a minimum of 55% for plastic, by the end of 2030 [4
]. However, in some countries, the recycling rates are still rather low: in 2017, the European Union had a plastics packaging recycling rate of 41.7% [5
]. Furthermore, mechanical recycling, the most common process used, impacts the final properties of the plastics [6
]. Since 1950, about 6300 million tons of plastic waste have been generated, of which 4977 million accumulated in landfills and waterways [7
]. Another emerging problem are the so called microplastics (< 5.0 mm) that are present in the air, water and soil [6
], with harmful effects to both terrestrial and marine ecosystems.
Therefore, industry and academia have been working on the development of renewable and sustainable alternatives with competitive features [8
]. Biopolymers, while biodegradable and highly available, often have inferior performance than their petroleum-based counterparts. To improve their performance, composite technology has emerged as an approach to blending biopolymers with different properties, leveraging on the best properties of each individual component. In the global biodegradable polymer market, the revenues are expected to grow from $
3.1 billion in 2016 to $
7.1 billion by 2021, an annual growth rate of 18% [1
Vegetable cellulose and its derivatives are widely used for paper production, pharmaceutical compounds, textiles and packaging [9
]. Paper, paperboard and CellophaneTM
are examples of cellulose-based materials used traditionally in food packaging. More recently, plant nanocellulose (NC) (which includes nanocrystals (CNCs) and nanofibrilated cellulose (NFCs)) became industrially available, offering unique characteristics such as high specific surface area, and consequently high concentration of active groups for surface modification, and the ability to improve the mechanical performance of the nanocomposites [9
]. Nanocellulose has been widely studied for the application in textile [2
], optical electronic devices [15
], food industry [9
], biomedicine [8
] and to a lower extent in papermaking [17
]. Lately, research and development communities have sought the development of nanocellulose-based materials for food packaging applications [18
], focusing mainly on improving their mechanical and barrier properties, by using different nanocelluloses, production processes and matrix biopolymers [20
]. Nanocellulose is also a trending material as a support in active and intelligent packaging for additional functionalities such antimicrobial and antioxidant properties, as well as a component in sensoring systems [21
]. However, like any other food contact material, nanocellulose in food packaging applications raises potential safety concerns.
2. Food Packaging Requirements
Packaging has become essential in modern life and its use has increased over time. The main goal of packaging is to contain food in a cost-effective way, while satisfying industry and regulatory requirements and consumer expectations, keeping food products protected from the external environment [2
]. For the majority of food products, the protection afforded by the package is an essential part of the preservation process. The requirements of a packaging system intended for fresh, frozen, dehydrated, thermal or aseptic processed products, depend on the: (i) intrinsic properties of the food product, such as water activity and oxidation potential that determine their perishability; (ii) extrinsic factors, namely storage temperature, relative humidity and exposure to light, and finally (iii) the required shelf-life. All these factors must be taken in account when specifying the required barrier ability to water vapour, oxygen and other gases, including aromas and light. The physical and mechanical properties are also important [24
] during processing, packaging operations and handling through the supply chain. Packaging geometry (surface area to volume) and the material thickness are variables that largely affect both the barrier and the physical performance of the package. Table 1
presents the requirements regarding moisture and oxygen barrier for several foods (information adapted from [25
]) for materials commonly used for this application.
It is recognised that petroleum-based plastics have great thermomechanical and barrier properties, as well as light weight and low production cost [26
], yielding a hard to beat overall performance (Figure 1
); they are commonly used in food packaging as indicated in Table 1
The environmental impact of non-biodegradable plastic materials and the increasing need for a more sustainable use of packaging and plastics in particular, are ever-growing global concerns. Solutions to reduce and in some cases to replace those materials are on top of research efforts. Biopolymers such poly lactic acid (PLA), polyhydroxyalkanoates (PHA) and thermoplastic starch (TPS) have been sought as alternative solutions [17
]. The most widely exploited one is PLA [18
], which is mainly used in packaging applications. It is used as films or thermoformed or injected packages for relative short-term and mild temperature contact conditions, such as fresh salads and beverage drinks, because of its low resistance to temperature. One major limitation commonly referred is the high price and commercial shortage, as compared to conventional plastics.
Corn is presently the major raw material used in the production of PLA but a second-generation feedstock is under development. Although generally considered a biodegradable material, PLA is actually poorly degradable under simulated ocean and soil conditions, only being compostable at high temperatures [28
]. PHAs, and in particular poly-(3-hydroxybutyrate) (PHB) is one of most widely studied biopolymers, the easiest to produce, and is regarded as an alternative to polypropylene (PP) for food packaging [29
], although with much less commercial applications. Starch, is the second most abundant biopolymer in nature [30
]. Thermoplastic starch (TPS) in particular has potential for food packaging application, in spite of its poor mechanical and barrier properties (Figure 1
). An important commercial application is that of bags for non-packaged fruits and vegetables and as shopping bags, in blends with fossil-based polymers. However, the bio-based materials available lack the performance to fulfil the most demanding specific requirements (Table 1
) for food packaging [25
] and apart a few exceptions, commercial applications are yet not volume comparable. In Figure 1
, the mechanical (tensile strength, elongation at break) and barrier properties (water vapour permeability and oxygen permeability of conventional plastics and biopolymers are compared. In general, biopolymers have lower elongation at break values and higher water vapour permeability than conventional plastics (Figure 1
NC has received an exponential interest as a component for food packaging purposes and its properties include a high stiffness (comparable to polyethylene terephthalate (PET)) and a low oxygen permeability (comparable to ethylene vinyl alcohol (EVOH)), as depicted in Figure 1
. Despite the good mechanical and oxygen transfer properties in dry conditions, the moisture barrier ability of nanocellulose is one of the poorest when compared to petroleum-based plastics and biopolymers (Figure 1
In order to improve the performance of the mentioned biopolymers, up to a level compatible with food packaging applications, nanocomposite technology has been regarded as an option, consisting on the combination of two or more materials, to enhance the overall properties of the composite. However, the outcome of NC mixtures with other polymers is not straightforward, as it depends on the ability of NC to interact with the other materials, in particular through hydrogen bonding.
4. Safety of Nanocellulose Based Composites
In the current European legislation, all materials in contact with food (FCMs) must meet the requirements of the framework Regulation (EC) No 1935/2004. The regulation states that “Materials (…), shall be manufactured in compliance with good manufacturing practice and so that, (…) they do not transfer their constituents to food (…) in quantities which could: endanger human health; or bring about an unacceptable change in the composition of the food; or bring about a deterioration in the organoleptic characteristics thereof”. These requirements also apply to nanocellulose based composites, although this legislation does not address specific provisions for nanomaterials used in food contact materials. The regulation applicable to plastic materials—Regulation (EU) No. 10/201 as well as the Regulation (EC) No. 450/2009 on active and intelligent packaging materials indicate that a specific evaluation is required for substances in nanoform. Nanoparticles mechanisms of mass transfer and interaction with the host materials and with the food may be different from those known at the conventional particle size scale. Therefore, nanoparticles may lead to different exposure and different toxicological properties. Consequently, the pre-market authorisations which are based on the risk assessment of a substance with conventional particle size do not cover the use of the same substance in its nano-dimension which shall only be used if explicitly authorised and mentioned in the positive lists of the above-mentioned regulations [133
The information relevant for the risk assessment of FCMs containing nanoparticles include three relevant aspects: (i) the characteristics of the nanomaterial used to produce the material; (ii) the characteristics of the material once it is incorporated into the FCM, as these may differ from the original characteristics, being influenced by the FCM matrix and/or the manufacturing conditions; and, (iii) the characteristics of any nanomaterial that migrates into the food matrix which is influenced by the food environment [134
The nanoparticles size, shape and aggregation properties, among other factors, can affect the interactions of NC with living cells [135
]. The safety of NC in food packaging depends on its transfer into food, which in turn determines the human exposure and on its toxicological profile. Different toxicity testing approaches are recommended to be applicable to nanomaterials depending on their migration behaviour and persistence in the nanoform, namely depending on if it occurs, or not, nanoparticle transformation into the non-nanoform in the food matrix before ingestion, or in the gastrointestinal tract following ingestion [134
Cellulose and several derivatives are recognized as safe and are already authorised under the European Regulation (EC) No 10/2011 for plastics for food packaging, for use as polymer additive, production aid and other starting substances. Additionally, for food packaging, cellulose and cellulose acetate butyrate, different alkyl and hydroxyalkyl celluloses are authorised as additives and polymer production aids; and nitrocellulose and lignocellulose are authorised as monomers or other starting substances (Reg 10/2011). However, NC is not specifically listed and therefore it is not currently authorised for food contact applications.
There are few recent studies addressing the toxicity of NC. The endpoints recognised as most important regarding a nanomaterial toxicity are cytotoxicity, (pro-) inflammatory response, potential to generate reactive oxygen species (ROS) and oxidative stress, genotoxicity and integrity of the gastrointestinal barrier [134
]. Endes et al. [136
] reviewed key studies in vitro, in vivo and with ecosystem models regarding the biological impact of NC, addressing some of these endpoints. The compiled information revealed some inconsistency in the results achieved, as some studies showed low or no toxicity, while others stress the potential of NC for adverse effects. This was attributed to the variability of the biological systems, test conditions and source of the cellulosic material used, as well as to the lack of thorough characterization of the administered CNCs [136
]. The physical-chemical characteristics, especially the aggregation level was considered to have a major impact on the results for inflammation. Some of the studies considered in this review, focused the inhalation route [137
] which is of upmost relevance for occupational exposures as nano-sized particles (such NCs) may be inhaled during composite processing, but not for the use of NC-based composite food packaging. Therefore, studies with ingestion as the main route, focusing in relevant existing data gaps, such as migration and resulting exposure, uptake and fate in the gastrointestinal system and organ distribution, are still missing.
The effect of NC in the gastrointestinal tract was addressed in very few studies. A recent study evaluated the in vitro biological effect of unmodified and modified NC (carboxymethylation, phosphorylation, sulfoethylation and substitution with hydroxypropyltrimethylammonium) on human gut bacteria and gastrointestinal cells [142
]. The metabolic activity and cell membrane integrity of intestinal cells Caco-2 and the growth of representative of microbiota bacteria were measured. Results indicated no cytotoxicity after exposure to unmodified NFC and to surface functionalized NFC. Furthermore, a bacteriostatic effect on E. coli
was observed but not on L. reuteri
Low in vitro (cell viability higher than 70%) cytotoxic effects and no mutagenic effect of TEMPO-oxidised NFC were reported in a recent study [143
]. The cytotoxicity of the TEMPO-oxidised NFC was evaluated through the MTT test. However, the mutagenic activity was evaluated through the Ames test which is not considered suitable for nanomaterials owing to the inability of bacterial cells to internalise particles. Instead, mammalian cells are recommended to address genotoxicity endpoints [134
]. The toxicological effects of ingested NC in in vitro intestinal epithelium and in vivo rat models were recently reported [144
]. NFC and CNC, at 0.75% and 1.5% w
were tested for the effect on cell layer integrity, cytotoxicity and oxidative stress. Micron-scale cellulose and TiO2
were used as controls. A 10% increase in ROS for 1.5% w
CNC was reported, but no other significant changes in cytotoxicity, ROS or monolayer integrity were observed. Results from in vivo toxicity suggest that ingested nanocellulose has little acute toxicity and is likely non-hazardous when ingested in small quantities [144
]. The authors highlight the need for chronic studies to assess long term effects, and potential detrimental effects of NC on the gut microbiome and absorbance of essential micronutrients. These studies would be of major interest for the application of nanocellulose in food packaging systems, where the chronic exposure is expected once the widely commercial use of NC will occur.
As with the synthetic counterparts, the migration to foods of intentionally added substances from the adjacent polymers and contaminants potentially present on the nanocellulose-based composite, must also be addressed. Besides contaminants and impurities, substances formed during nanocomposite processing, can appear throughout the production chain, when fossil-based and/or biopolymers are used. The processing of these materials may provide a source of non-intentionally added substances, with potential to migrate to foods upon contact. Potential contaminants from bio-based polymers include: heavy metals, persistent organic chemical contaminants, residues (e.g., pesticides, veterinary medicines), allergens and natural toxins [20
]. Chemicals used in the pre-treatment of vegetable NC, by-products and culture media residues of BNC are also possible sources for chemical contaminants.
Studies reporting the impact of nanocellulose in composites on limiting the migration of components from the polymeric phase have been reported. The effect of incorporating CNC in several plastic films in the overall migration has been studied, particularly for biopolymers, such as PHB, PHBV and PLA. CNC-PHB films with different CNC loadings were tested for overall migration into ethanol 10% (v/v
) and isooctane, respectively at 40 °C for 10 days and at 20 °C for 2 days. At low concentrations (1–2%) of CNC dispersed in the PHB matrix, the overall migration decreased in comparison to neat PHB, for both simulants. However, CNC loadings above 3% yield higher overall migration values, possibly due to a decrease in the adhesion between the hydrophilic CNC and hydrophobic PHB phases. The migration increased when the PHB was loaded with 5% CNC, from 20 to 40 µg/kg into isooctane and from 90 to 180 µg/kg into ethanol. The contact with ethanol simulant was reported to cause a change in the film appearance, in the thermal properties and also a decrease in the molecular weight due to the degradation of PHB chains into small oligomeric fractions [79
]. The migration levels from a PHBV film into isooctane and 10% ethanol were also reduced by ca 50%, after incorporation of CNC into the plastic matrix [145
The incorporation of NFC in PLA (ratio NFC:PLA 1:19) decreased by 20% the overall migration levels into ethanol 10%. However, no significant differences were found on the migration into the nonpolar simulant isooctane [146
]. The overall migration values reported were low, in the order of 10 to 100 µg/kg. The results reported show that the chemical modification of the nanocellulose and the amount incorporated largely influence the migration behaviour. Both factors have an impact on the adhesion between the nanocellulose and the plastic, which may affect the simulant penetration and consequently the migration.
The CNC modification through esterification improved the compatibility with PLA, which reduced the migration from the PLA phase into both simulants [146
]. This effect was observed for low CNC loading. However, increasing the CNC concentration (either modified and unmodified) in the composite led to an increase in the overall migration, particularly into isooctane [146
]. As indicated, these studies focused on overall migration, and as such, characterisation of the migrants was not addressed. Studies focusing on the migration of specific substances have not been reported, apart from a few exceptions, and the influence of nanocellulose on the migration of substances of different chemical nature is not known. The migration of several hydrocarbons, as surrogate compounds for mineral oil, into the dry-food simulant Tenax®
was studied. A decrease of more than 90% was achieved by coating a HDPE film of 48 µm with a TEMPO-NFC 6 µm layer [148
Clearly, there is a need for further research concerning the migration from nanocellulosic composites, particularly on screening throughout the production chain of nanocellulose based composites and traceability of possible contaminants throughout the production chain.
5. Technology Readiness Level of Nanocellulosic Composites
Parallel to the academic studies, an increasing growth has been noticed on the industrial production of NC and its composites. In fact, since 2010, 7636 patents are related to NC, 2827 patents related to nanocellulosic composites, 169 patents related to NC for food packaging applications and 29 patents related to NC-based composites for food packaging applications. This means that the interest in NC has been growing in the last decade, although not focusing particularly on the production of NC-based composites for food packaging applications. Additionally, most of the publications on NC-based composites for food packaging patents are from research institutions and universities. Nevertheless, several companies are currently commercializing vegetable NC (in the form of NFC and CNC) for a wide variety of applications (Table 2
The high interest in NC, the advances in technology and scale up production can explain the advanced level of development. However, the price of NC is still volatile due to the production cost (related to the production method adopted), the market place and the specific format and characteristics [172
]. The estimated production costs of CNC, NFC and BNC (Figure 4
) are, respectively (in dry equivalent); 15, 2.15 and 22.11€/Kg, although the current market prices are higher [172
]. It must be remarked that the cost of NCs in Figure 4
does not include hydrophobization, which may significantly add to the production cost taking in account the large surface area. On the other hand, this increase in cost may be somewhat mitigated by an increase in the production scale. Despite the good features of NC, the manufacturing cost (as that of PLA, PHA and starch) is still not competitive with petroleum-based plastics [156
]. Nevertheless, there is margin to improve the market price of NC by increasing the production capacity (dry basis) (surrounding 1.6 thousand tonnes per year). The global production of NC was roughly estimated from the data collected from [177
]. Achieving higher investments to improve technology processing and optimise the processes of NC production (especially for biotech BNC, still not available in large scale) is on demand. NFC has the lowest production cost since there is a higher investment by companies.
The TRL of thermoplastic biopolymers reinforced with natural fibres is at level 5 [156
] demonstrating that there is an interest in technology development and validation. For the most promising bio-based materials, there is a need to lower the production cost and promote higher yields, through scalable processes for these materials become competitive in the market. Among the above-mentioned biopolymers (PLA, PHA and starch), combinations of starch and NFC are most likely to succeed, provided that the mixture is properly hydrophobized (e.g., treated with alkenylsuccinic anhydride). This combination turns out to be simple to process, as both are hydrophilic; and the alkenylsuccinic anhydride treatment will hydrophobize the composite to ensure enhanced water vapour permeability and water uptake, a known problem within NCs-TPS composites [82
]. Another successful strategy to produce composites with two materials with low compatibility (as NCs and PLA) is the preparation of a master batch. This pre-mixing will assure a better dispersion within the matrix during melt mixing. A highly homogeneous composite is obtained with interesting mechanical and barrier properties [106
Regarding specific applications to food contact materials and packaging there are already in the market some commercial solutions with biopolymer thermoplastics reinforced with natural fibre (Table 3
], although other applications, such as construction and automotive are more common. A few examples include FuturaMat which develops composites with BioFibra®
used to compostable coffee pods [185
] and Kareline Oy that develops tableware based in Kareline®
]. In both cases the composites are based in PLA and wood fibre but do not incorporate NC in their product composition. Specific cases using NC in fibre based products are: (i) CelluComp (UK), which extract cellulose from root vegetables and produces Curran®
cellulose nano-fibres that can be used in coatings and paper and board for packaging [190
], (ii) Stora Enso (Finland), which develops biocomposite and paperboard grades containing microfibrilated cellulose [191
], claimed to be used in Elopak beverage cartons for milk [192
]. Currently, the conventional cellulose based materials have a stronger presence (with NatureFlex and Cellophane from Futamura and PaperWise with Bio4pack) in the market (Table 3
). So, NC related applications in general are on different TRL levels, and specifically regarding food packaging applications the TRL seems to be lower, calling for additional research investment.
6. Final Considerations
The purpose of this review was to assess the potential of using nanocellulose in the food packaging industry. Great amount of research was reported regarding the development of composites with different types of nanocellulose (NFC, CNC and BNC). Regardless of the matrix phase, a positive effect of nanocellulose was generally observed, namely improved the mechanical and oxygen properties, particularly at low humidity. On the other hand, low water resistance (vapour and liquid) is observed, which is an important feature for food packaging applications. Several approaches were reported to modify nanocellulose in order to counteract the poor water resistance, to promote better dispersion in hydrophobic matrices and to provide functionality, specifically for the application on active and intelligent packaging. The technologies of solvent casting, melt processing, impregnation, coating, layer by layer deposition, electrospinning and all-cellulose composites may be adopted for the production of NC composites, allowing the application of nanocellulose on several fronts of food packaging, as a reinforcing agent, as a barrier agent and for smart packaging.
Studies regarding the toxicity of nanocellulose and the impact of nanocellulose on the overall migration of composites, should be considered before its application in food contact materials, the available results indicating lack of or low toxicity. However, further research should be addressed to demonstrate the safety of nanocellulose, since it is not currently authorised for food contact applications. NC is produced industrially and commercially available, but its main applications lie on the composites and automotive industries.
The use of NC-based materials in replacement of petroleum-based plastics is still a major challenge, due to existing technological weaknesses regarding rheological behaviour, scalability and overall properties of the final material in regard to the requirements of today’s packaging systems. The sustainability of the global process of handling the biomatter and converting into packaging, such as the energy input and the use of solvents for example, should be considered and balanced with the impact of using reusable and recyclable, although non-renewable materials, together with safety and quality food requirements. Nevertheless, the potential for specific applications of functionalised nanocellulose is well recognised for specific and tailor-made high-valuable applications and these is expected to drive the research efforts in near future. Further research and investment on nanocellulose-based materials should be made in order to fill the narrowing regulatory, economic and technologic gap between the sustainable and the conventional packaging used in the food industry.