Turning Food Protein Waste into Sustainable Technologies

Abstract

For each kilogram of food protein wasted, between 15 and 750 kg of CO₂ end up in the atmosphere. With this alarming carbon footprint, food protein waste not only contributes to climate change but also significantly impacts other environmental boundaries, such as nitrogen and phosphorus cycles, global freshwater use, change in land composition, chemical pollution, and biodiversity loss. This contrasts sharply with both the high nutritional value of proteins, as well as their unique chemical and physical versatility, which enable their use in new materials and innovative technologies. In this review, we discuss how food protein waste can be efficiently valorized not only by reintroduction into the food chain supply but also as a template for the development of sustainable technologies by allowing it to exit the food-value chain, thus alleviating some of the most urgent global challenges. We showcase three technologies of immediate significance and environmental impact: biodegradable plastics, water purification, and renewable energy. We discuss, by carefully reviewing the current state of the art, how proteins extracted from food waste can be valorized into key players to facilitate these technologies. We furthermore support analysis of the extant literature by original life cycle assessment (LCA) examples run ad hoc on both plant and animal waste proteins in the context of the technologies considered, and against realistic benchmarks, to quantitatively demonstrate their efficacy and potential. We finally conclude the review with an outlook on how such a comprehensive management of food protein waste is anticipated to transform its carbon footprint from positive to negative and, more generally, have a favorable impact on several other important planetary boundaries.

1. Introduction

Maintaining our planet Earth as the perfect home for subsequent generations is the most urgent scientific, economic, social, and ethical challenge that humanity is facing today. Yet, according to the planetary boundaries concept put forward by Rockström et al. over a decade ago,¹ humanity has already crossed the boundaries for a safe operating space in three of the nine main planetary systems, i.e., climate change, nitrogen cycle, and biodiversity loss. It appears obvious that, with a steadily growing world population, the origins of these planet threats are mostly of anthropogenic nature. In fact, the world population has more than doubled in only the last 50 years,² and such growth can only be sustained by an efficient global food production system and, equally importantly, by achieving zero food waste. Food waste is a primary contributor to climate change, with approximately 1.3 billion tons of food wasted annually. This amount corresponds to 30% of the total food produced for human consumption, and to 8–10% of global greenhouse gas emissions (GHG) associated with food waste.^3,4 Food protein waste, as its most valuable component, plays a major role in the overall situation. For example, the intensive production of animal proteins has a direct impact on climate change, as it results in 12% of total GHG emissions and 30% of all human-induced terrestrial biodiversity loss.⁵ Plant protein production, on the other hand, has severe and immediate implications for both nitrogen and phosphorus cycles,^1,6 and also impacts the rate of biodiversity loss via conversion of pristine forest into exploitable lands for agriculture.

Other planetary boundaries are also indirectly affected by food protein production. Indeed, both classes of proteins increase pressure, via both crop and livestock, on global freshwater use, change in land use, and chemical pollution. Therefore, whether originating from animal or plant, once produced, protein-rich foods will result not only in a positive carbon footprint but also have a severe impact on most of the other planetary boundary systems in general. Because decreasing the production of foodstuffs is not an option in a growing world population, it is only by carefully reducing overall food protein waste that the carbon footprint can be reduced and the impact on other environmental pillars can be mitigated.

Additionally, reducing food protein waste has implications which extend beyond even the environmental impact. Proteins are the most valuable of the macronutrients from both an economic and human diet perspective; furthermore, they can be used as templates for biosynthesis of artificial materials and thus are key players in emerging and consolidated green technologies.⁷⁻⁹ In short, we no longer have the luxury of discarding food protein waste as an end-product of a linear economy, and efficient valorization into a circular economy has become critical from different perspectives.¹⁰

This review discusses recent efforts undertaken to revalue food protein waste by reintroduction into a circular economy via two different pathways: (i) via innovative food and nutritional applications within the food-value chain, and (ii) by exiting the food-value chain to be utilized as new building blocks and templates for the development of sustainable technologies, thus alleviating some of the most urgent global challenges. In the context of this second pillar, the benchmark technologies of interest to this review are biodegradable bioplastics, water purification, and renewable energy.

Figure 1 provides a graphic summary of the state of the art of food protein waste valorization, organized across the four main pillars discussed in this review: (i) revalued food, (ii) bioplastics, (iii) water purification, and (iv) renewable energy. Pillar (i) is the only one considered within the food-value chain and implies that food protein waste, or part of it, re-enters the food chain by being reutilized for human consumption (as animal feed, or ReFeed, which is widely considered as food waste, although this differs somewhat between Europe and the U.S.).¹¹

Figure 1.

General state of the art of food protein waste valorization within (revalued food) and beyond (bioplastics, water purification, energy) the food-value chain. Production of bioethanol is not considered because this is not regarded here as a sustainable technology because it still directly impacts several of the planetary boundaries.

The other pillars (ii) to (iv) consider alternative uses of food protein waste beyond the food-value chain for sustainable technologies of immediate significance: with ∼10 billion tonnes of annual plastic production expected 50 years from now,¹² plastic pollution is another major global issue causing devastating effects on the planet’s ecosystem, with petrol-based plastic production accelerating climate change and directly compromising global biodiversity.¹³⁻¹⁵ Water depletion and fossil energy consumption complete a dire understanding of the anthropogenic impact on our planet: it is estimated that water purification, as a solution to alleviate pressure on global freshwater use, contributes globally to 5% of overall GHG emissions,¹⁶ with an estimated increase in global demand of water of 20–30% by 2050,¹⁷ whereas, it is commonly accepted that the pursuit of renewable energies is mandatory as an alternative to fossil energy consumption, which alone contributes to up 89% of climate change.¹⁸

The review is structured as follows: after briefly reviewing the main composition and sources of food protein waste, we discuss recent progress across the four pillars highlighted above. We then enrich the discussion by carrying out a detailed life cycle assessment (LCA) in the fields of bioplastics and water purification, for which life cycle inventory data are sufficiently well established, by considering in each field of application one example of animal protein, one example of plant protein, and a corresponding relevant benchmark. By doing so, we demonstrate the superior LCA performance of protein-enabled technologies, and illuminate the potential of food protein waste as a key factor in establishing sustainable technologies capable of solving some of the most pressing global challenges of our time. We finally conclude the review by providing some projections on the possible impacts of implementing this circular approach on a global scale, with tangible effects on several of the compromised planetary boundaries.

2. Food Protein Waste Sources and Composition

Food production makes a major contribution to the environmental impact of modern society, contributing to climate change, resource scarcity, and pollution. As previously mentioned, it is estimated that up to one-third of the total food produced each year is wasted or lost.^19,20 Food loss and waste occur along each sector of the food supply chain. For example, food waste can be generated by inefficient harvesting procedures, by inadequate handling conditions during transport and storage, or during processing in the food industries and at the household level. In industrialized countries, approximately 40% of the food waste occurs at the retail and customer level, and up to 40% of it is generated during the manufacturing processes in the food industries.^19,21,22 Most food waste that is generated by the food industries is inevitable, as it is produced during the transformation of raw ingredients into the final products. This portion of food waste is much easier to valorize, as it is readily available, easy to collect, and produced continuously and in a predictable manner.

Proteins are essential for the human diet, and animal- or plant-based protein-rich foods provide essential amino acids and macronutrients with good digestibility.²³ However, in the past few years, several studies demonstrated a great environmental impact of intensive meat and plant-protein production, and confirmed that protein supply is directly connected with most of the planetary boundaries.^5,24,25 At the same time, the demand for protein-rich foods is increasing with the growing population.^26,27 As a consequence, reducing protein waste is the ideal solution to mitigate the environmental and social consequences, especially considering that almost one billion people are still undernourished.²⁸ Moreover, the disposal management of protein-rich side streams is particularly difficult and expensive because these materials are characterized by high biological and chemical oxygen demands (BOD and COD, respectively) and can contain a wide range of alkaline or acidic detergents, pesticides, or sterilizing agents that contribute in aggregate to air, soil, and water pollution.^26,29−31

2.1. Animal-Based Proteins

2.1.1. Dairy

Every year, approximately 150 million tonnes of milk are produced in Europe alone, with almost 50% of it being processed by the dairy industry for cheese production.³² During this transformation, fats and caseins are extracted and 80–90% of the processed milk volume is transformed into liquid whey, which represents the main byproduct of the dairy industry. The composition of whey varies based on the type of cheese produced, and generally contains 94% water and 6% solid mass, of which approximately 55% are milk nutrients, including 20% of its proteins.³³ Depending on the processing techniques, two classes of dairy whey exist:^34,35 sweet whey, the side-stream of the production of the majority of cheese products via rennet or enzymatic coagulation, and acid whey, originating from the production of acid-coagulated cheeses, such as cottage, quark, and Greek yogurt. As suggested by their names, the main difference between these two classes of whey is their pH, ranging from 3.5 to 4.5 for acid whey and 6 to 6.5 for sweet whey. Both of their dry-mass compositions are markedly similar, comprising 71–75 wt % lactose, 10–12 wt % proteins, 8–9 wt % fats, and 4–10% minerals.³⁵⁻³⁷ With global production of approximately 240 million tonnes of whey annually (Figure 2a),³⁸⁻⁴⁰ which corresponds to almost 2 million tonnes of high nutritional value proteins (approximately 55% β-lactoglobulin, 25% α-lactalbumin, 10% immunoglobulins, and 10% bovine serum albumin (BSA)),³³ significant effort has been dedicated to the valorization of this valuable and abundant byproduct in the past decades.^31,35,41 Because of their high organic load, with a BOD of 48 g/L and a COD of 95 g/L, dairy wastewater requires expensive processing prior to disposal, which is estimated to be in the range of 0.05–2.97 €/kg⁴² depending on processes’ efficiency. The most common solutions for valorizing whey for nutritional purposes include the production of concentrated whey powder, whey protein concentrates, whey protein isolates (WPI), and the extraction of powdered lactose.^33,35,38,41 β-lactoglobulin, the most abundant whey protein, belongs to the lipocalin family involved in the transport of hydrophobic molecules through its hydrophobic β-barrel structure.⁴³ It is rich in essential amino acids and offers excellent nutritional value, such that it is typically exploited in athletic and high-protein diets while also being commonly used in the food industry as an emulsifying, foaming, or gelling agent.³⁵ α-lactalbumin, the second major whey protein, is involved in the synthesis of lactose in mammalian milk production.⁴⁴ Possessing good nutritive properties and commonly used as an additive in infant foods, it can also be employed in biomedical applications due to its antihypertensive, antioxidant, antiobesity, and antitumoral properties.³⁵ BSA, which functions as a drug and nutrient transporter in the plasma,⁴⁵ has found use not only in food and used in food processing as a gelation and emulsification agent but also in therapeutic applications as an antioxidant and in cancer prevention.³⁵ However, up to 50% of the total whey produced globally unfortunately remains unprocessed and is used as fertilizer, in irrigation systems, as animal feed, or regarded as waste and discharged into the environment. Each of these uses of unprocessed whey presents disadvantages that are even more limiting, such as affecting soil conditions and pH when using acid whey as fertilizer or in irrigation and causing salt toxicity or animal digestive disorders when used as animal feed as a consequence of the high lactose content.^35,41,46

Figure 2.

Global production of industrial protein-rich food waste and their protein content. (a) Global production of the different food waste sources described in this review. (b) Protein content in wt % for the main byproducts that are generated globally.

2.1.2. Meat Industry

Although the ethics and health benefits of meat-rich diets have been widely considered to be dubious, meat consumption is increasing with the growing global population, and animal products remain the most abundant source of proteins in the human diet.⁵ As previously mentioned, the intensive production of animal proteins has a direct impact on climate change and biodiversity loss.⁵ In addition, the meat industry generates consistent volumes of inevitable byproducts, such as blood, bones, skin, hair, feathers, viscera, horns, hoofs, etc., and wastewater, which is often heavily contaminated with chemicals and organic matter.^47,48 Overall, it is reported that approximately 52.6 million tons of meat byproducts are generated globally, representing 20% of yearly meat production (Figure 2a).⁴⁹ Although some of these byproducts are edible and even rich in nutrients, such as proteins, essential amino acids, vitamins, and minerals, their consumption for human nutrition is declining and remains limited to a few localized areas.⁴⁷ The disposal of these byproducts is particularly challenging and expensive as it is highly regulated because their unappropriated disposal can severely contaminate the environment, posing a serious threat to human health such as the spreading of encephalopathies.⁵⁰⁻⁵² For these reasons, significant effort has focused on developing processes to efficiently revalue some of these byproducts,^52,53 with particular attention given to those that are rich in proteins, such as skin, bones, cartilages, and blood. These processes, discussed in detail in section 3, resulted in the development of a broad range of products for human nutrition, pet food, animal feed, and fertilizers, as well as for chemical, biomedical, and pharmaceutical applications.⁵²⁻⁵⁴

Collagen is the most abundant protein in mammals and also the main constituent of many meat byproducts, including skin, cartilages, and bones. Consisting of bundles of fibrillar structures, collagen imparts unique mechanical strength and stability to different body parts of living organisms.⁵⁵ However, because of collagen’s low nutritional profile and lack of essential amino acids, which are primarily composed of glycine, proline, and hydroxyproline, collagen-rich byproducts are not valorized for nutritional purposes but rather for the extraction of bioactive peptides.^{47,52−54,56} However, several purification techniques have been developed to efficiently isolate this protein from many byproducts, both in its native form or partially hydrolyzed form as in gelatin,⁵⁷ to exploit its technofunctional properties in food applications, cosmetics, and biomedicine.^52−54,57

Although global blood production is not reported, it is estimated that approximately 2.5 billion liters of this byproduct are produced annually,⁵⁸ which in total contains enough protein to fulfill the annual protein requirement of approximately 17 million adults. This waste is not revalued to its full potential, as only less than 4% of these proteins are valorized for human consumption.⁵⁸ Depending on processing level, industrial uses of blood for low-added value products include applications as a colorant in textiles, as a spray adjuvant, as fertilizer, and can also be used for animal and human nutrition, and pharmaceutical applications with further processing.⁵⁸ The main components of blood, red blood cells (40%) and plasma (60%), are rich in proteins, corresponding to 35% and 8% of their volumes, respectively.⁵⁸ Albumin is the most abundant protein in plasma, accounting for up to 60% of its protein content. In its purified form, albumin is a valuable component in medical applications, as it is utilized as a stabilizer in vaccines, in antibiotic sensitivity tests, and in Rh factor testing.⁵⁸

Most of the byproducts that are rarely considered as food ingredients due to their low digestibility and nutritional values, such hair, nails, hooves, or feathers, are extremely rich in keratin (Figure 2b). The filamentous structure of keratin, analogous to collagen⁵⁹ and characterized by high mechanical and chemical stability, endows with extreme resistance most physical and biological agents. Every year, 40 million tonnes of keratin-rich waste is generated,⁶⁰ which raises challenges with regard to waste management due to a lack of effective disposal procedures.⁶¹ Conventional incineration of keratin waste produces toxic gases that are rich in sulfur due to the high content of cysteine, while high degrees of disulfide cross-linking and compact structures result in slow biodegradation rates,⁶² leading to accumulation of these slow-degrading byproducts and therefore presenting an uncontrollable issue for the environment and human health globally.⁶³⁻⁶⁵

Because of the low digestibility of keratinous materials, hydrolysis techniques are necessary to produce soluble amino acids or peptides for animal and human consumption, or as usable fertilizer.⁶³⁻⁶⁵ After treatment, pure keratin extracts have multiple applications in biotechnology, cosmetic, and medical sectors as biocompatible materials, such as fibers, sponges, and microcapsules.⁶³⁻⁶⁵ However, the methods discussed above to recover the protein are still costly and inefficient, and thus sustainable alternatives to valorize these byproducts are urgently needed.⁶³⁻⁶⁵

Different glands and organs are considered edible in several regions of the world, but their use for human consumption remains limited.⁶⁶ In a few cases, however, they provide valuable pharmaceutical compounds, such as melatonin from the pineal gland, heparin from the liver, insulin from the pancreas, and estrogen and progesterone from ovaries.^52,67

2.1.3. Seafood Industry

Each year, approximately 160 million tonnes of seafood are harvested and processed globally, of which an average of only 20–50% of it is recovered as edible portions from processing, while the remaining parts are considered as byproducts or waste. This seafood waste contains valuable components, including lipids, enzymes, oils, vitamins, flavors, and an average protein content of 60%⁶⁷ (Figure 2b), and are therefore used as animal or fish feed. Their valorization also remains limited as a result of their rapid deterioration.⁶⁸

Like most of the byproducts of the meat industry, seafood waste is also highly rich in collagen, which provides structural stability in most marine animals and is the primary component of fish skin, scales, fins, bones, and swim bladder, and of invertebrates, such as jellyfish, mollusks, and sponges. The most common pathway to valorize these byproducts is the extraction of this protein in its native form for its bioactive peptides, or hydrolyzed in the form of gelatin.⁶⁸⁻⁷⁴ The production of collagen from seafood waste for nutritional or high-value applications is rising and has attracted increasing attention in past decades, due to the relatively lower risks of disease outbreaks compared to those of land animals and also social-religious aspects related to the use of animal collagen in the production of kosher and halal products.⁷² Other valuable components that can be extracted from seafood waste include the polysaccharides chitin and chitosan that can be extracted from the exoskeleton of crabs, shrimp, and krill and have multiple pharmaceuticals, nutritional, and medical applications.^75,76

More recently, a new family of structural proteins found in marine animals, called “suckerins”, is increasingly attracting attention.⁷⁷ This protein isoform family makes up almost entirely the sucker-ring teeth of squids and cuttlefish which are used for grappling prey. Assembled into an extremely robust supramolecular network characterized by outstanding mechanical properties, these proteinaceous parts are tough materials even when compared to strong synthetic polymers.^78,79 Because of its excellent mechanical performance, suckerins have been used to develop functional and biomimetic materials, such as adhesives, gels, nano capsules, and microneedles.⁷⁷⁻⁸¹

2.2. Plant-Based Proteins

2.2.1. Soy Industry

Soybeans, with a global production of approximately 400 million tonnes each year, is the world’s most abundant cultivated crop and one of the most popular plant protein sources.⁸² Soybeans are consumed throughout the world either in their raw form or processed into several different products, such as soymilk, soy sauce, tofu, and miso. The major byproducts of the soy industry are soy whey and soybean pulp, also known as okara. Although these byproducts contain substantial amounts of nutrients, they possess very high moisture levels, which require several pretreatment steps prior to disposal, contributing significantly to costs and environmental impact of soy production.

Soy whey is the wastewater generated by soaking and boiling soybeans, or during tofu curd draining and soy protein isolate production.⁸³ Soy whey is produced in abundant quantities, as approximately 9 kg of soy whey is generated from each kg of tofu produced.^42,83−85 Depending on the source and on the processing conditions, soy whey contains proteins in the range of 0.3–8.2 g/L, has a pH in the range of 5.4–6.6, and a high concentration of organic compounds.^42,83 This byproduct has a markedly limited shelf life (approximately 1 day), and its disposal remains challenging due to its organic load.⁸³ In particular, the treatment of soy industry wastewater is expensive as a result of the high BOD and COD (6.8 and 12 g/L, respectively), and it is estimated to be on the order of 0.13 $/kg.⁴² Currently, two main pathways for soy whey valorization exist: the recovery of nutrients, such as proteins, oligosaccharides, or magnesium; and biological/enzymatic biotransformation.⁸³

Soybean pulp/curd residue (okara) is another major byproduct of the soy industry. Annually, 14 million metric tonnes of okara are obtained from soymilk and tofu production worldwide.⁸⁶ For instance, one metric ton of soybeans generates seven metric tonnes of soymilk and two metric tonnes of okara as byproducts.⁸⁷ Okara is a white or yellowish pulp with a moisture content of 80–85%, which on a dry-matter-basis contains crude protein (20.9–39.1%), fiber (12.2–61.3%), oil (4.9–21.5%), and ash (3.4–5.3%).⁸⁷ Although almost half of the currently produced okara is used as livestock feed or fish food, its rapid putrefaction and high drying cost result in the majority of it being directly disposed of by incineration or dumped in landfills.⁸⁶ More recently, some techniques have been proposed to transform okara into higher-value products. In particular, some authors employed okara as biomass for ethanol production,⁸⁸ and to produce soil supplements⁸⁹ and functional aerogels.⁹⁰

2.2.2. Oilseed Industry

Plant oils are macromolecules with high-efficiency energy storage properties, and therefore many plants species store energy in the form of seed oil to provide their progeny with sufficient energy and nutrients until they can autonomously undergo photosynthesis.⁹¹ For thousands of years, seeds have been the main source of vegetable oil for human and animal consumption, and more recently, they provide valuable materials for industrial and fuel purposes, leading to a steadily increasing demand for vegetable oils worldwide.^92,93 During the refining of vegetable oils, approximately 364 million tons of byproducts are generated in the form of meals or cakes (Figure 2a),⁹⁴ whose composition includes valuable components, such as proteins, polysaccharides, phenolic compounds, cellulose, etc. These byproducts therefore constitute promising materials to be valorized for their nutritional value as feed, the extraction of bioactive compounds, such as proteins, fibers, or antioxidants, and as substrates for the production of biofuels enzymes, antibiotics, mushrooms, and surfactants.^95,96 Depending on the type of plant wastes, residual cakes can reach up to 50% protein content. The constituent proteins primarily consist of seed storage proteins that function as sources of nutrients such as nitrogen, oxygen, carbon, and minerals during growth and germination.⁹⁷ From a nutritional perspective, however, because of the presence of several different antinutritional compounds commonly found in oilseed cakes and meals, it rarely finds use in human nutrition and when it does, it requires proper pretreatment and purification steps before consumption.⁹⁵ Although several valorization pathways have been proposed for these byproducts, they are still considered to be low-value materials that are mainly used as livestock feed, or directly disposed in landfills or incinerated.

Rapeseed, with a production of approximately 24 Mt in Europe and 74 Mt globally, is the second most cultivated oilseed crop in the world after soybeans.^92,93 For the production of oil, rapeseed is defatted through mechanical extraction, e.g., screw extrusion, and because the seeds contain 42% oil, approximately 40 Mt of rapeseed byproducts are produced each year globally.⁹⁸⁻¹⁰¹ Depending on the processing conditions, these byproducts exist in the form of defatted cold-pressed cake, with a protein content of 28–31%, and hot-pressed meal, with a protein content of 38–45% (Figure 2b). As briefly mentioned above, although this byproduct is rich in proteins, its utilization for human nutrition remains limited due to the presence of antinutritional elements, such as glycosylates and phenolic compounds, fibers, and phytic acid, and therefore it is primarily utilized as feed for poultry, pigs, and cattle. Because of the high production scale of this byproduct and its limited practical use, it is a cheap protein-rich material that can be purchased for approximately 100–300 $/t.¹⁰² The proteins contained in rapeseed cake can be extracted and used for different technological applications, such as adhesives, detergents, and cosmetics, as well as stabilizers in emulsions and building blocks for nanomaterials.¹⁰³

After soybeans and rapeseed, sunflower is the third major cultivated oilseed crop in the world.^92,93 Sunflower meal is the main byproduct of the sunflower oil industry, and each year, approximately 20 Mt of sunflower meal is produced globally, accounting for 36% of the processed seeds’ volume.^101,104,105 The exact composition of this byproduct depends on different factors, such as crop variety, growing conditions, and processing techniques. On average, however, it contains a high amount of proteins (20–30%), fibers (20–40%), residual oil (15%), moisture (8%), vitamins, minerals, etc.^101,106,107 Sunflower meal is very widely used as livestock feed as a result of its good nutritional values, but human consumption is limited to highly processed meal derivatives ascribed to the high content of phenolic and antinutrient compounds (protease inhibitor, arginase inhibitor).^106,107 More recently, emerging techniques have been developed to obtain higher-value products from sunflower meals, such as the extraction of proteins, bioactive compounds, and fibers.^{101,108−112} Although its potential nutritional value is high, this byproduct remains underutilized, and it is often disposed in landfills or used as solid fuel.

2.2.3. Cereal Industry

Corn, with a global production above 1100 Mt per year, is the world’s most cultivated cereal crop.¹¹³ Several varieties of maize exist, and although their composition does not vary significantly, the main difference between them is their color, spanning from white, yellow, red, and black. The main uses of corn are maize flour and meal; in the past decades, however, the use of maize for the production of bioethanol increased exponentially, reaching up to 40% of maize production in the U.S., the major global corn producer.¹¹⁴ Ethanol is produced from corn through wet and dry milling, and it constitutes 70% of global production.¹¹³ The wet-milling process produces different byproducts, including corn gluten meal and feed, corn syrup, corn oil, and germ meal, while dry-milling results in the production of the main byproduct known as dried grains with solubles.^115−117 Dried grains are currently only practically employed as animal feed, primarily for ruminants due to the high fiber and protein content.^115−117 However, this coproduct is highly rich in sulfur, which can alter the homeostasis of the ruminal microbial population, resulting in thiamine deficiency (causing polioencephalomalacia) and enzymatic inhibition.^115,116 Corn gluten feed contains digestible fibers, starch (20%), and up to 25% of proteins, which makes it a valuable source of nutrients for cattle and ruminants and means it can be fed in large amounts.^115,116 Corn gluten meal, however, produced during wet milling after germ, fibers, and starch have been removed, has higher protein content (up to 70%) that is mostly ruminantly undegradable. Because of its high content of sulfur and phosphorus, and to its bitter taste, it is consumed moderately by animals and used mainly as a food additive,^115,116 although it can be fermented to improve its nutritional properties by producing soluble peptides. Corn syrup, produced from dry-grinding plants, has a high moisture level up to 70% and is rich in protein (up to 40% dry mass) and oil (up to 20%). However, because of the abundant presence of Na, K, and P, long-term effects of the use of this byproduct in animal diets have been questioned.^115−117 During corn milling processes for nutritional applications, the outer layers of the grains are removed in the form of maize bran, which is largely used as livestock feed due to the presence of dietary fibers and micronutrients, such as vitamins and minerals, and is also used to extract edible oil and dietary fibers.^115,116

Corn, on average, contains approximately 72% starch, 10% proteins, and 4% fats.¹¹³ Zein, the predominant alcohol-soluble protein¹¹⁸ in corn, is a major protein of numerous byproducts, such as corn gluten meal and distilled grains produced during the milling process in starch and maize oil production and by the bioethanol industry.^118,119 The corn kernel contains 9–12% proteins; however, only half of this amount is an industrially valuable zein protein. Although a balance of yield, quality, and purity is desired in zein extraction, most commercial extraction techniques result in high purity zein, and the yields are still low.⁹³ The imbalanced amino acid profile and poor water solubility make zein far from ideal for human consumption. Zein was previously considered to be a waste protein and was incorporated into animal feed until recently, when novel applications are taking advantage of its low water solubility and biocompatibility properties, expanding its application into many products, such as coating, textiles, and adhesives,^118−123 and in biomedical applications.^124,125

Gluten, defined as a proteinaceous blend of glutenin and gliadin, results from the isolation of starch from wheat or corn flour, and is one of the most abundant plant proteins, being a major byproduct of starch and the bioethanol industry. As gluten is widely available and relatively inexpensive, extensive research has focused on valorization of gluten-rich materials for nonfood applications, especially as a result of its good thermoplastic behavior and processability. These studies include the development of gluten-based materials, such as gels, scaffolds, micro- and nanoparticles for drug delivery, tissue engineering, and medical applications.^{116,126−128} Another gluten-rich byproduct that is increasingly produced, rice gluten meal, derives from the wet milling of rice and the production of starch from rice. Compared to corn and wheat gluten meals, it contains lower protein content (40–47% on a dry basis), and there are currently no well-documented uses for rice gluten meal outside of livestock nutrition. Furthermore, its benefits for animal nutrition have not yet been well established.^129,130

Rice is one of the most important food sources around the world, with production above 600 Mt.^131,132 Rice husk and rice bran are the main byproducts of the rice milling process, representing the external layers of the grain that correspond to approximately 25% and 12% of the total weight of the kernel, respectively, with more than 70 Mt being produced globally.^96,133 In common with other cereal species, rice kernel outer layers are rich in nutrients and bioactive compounds, such as proteins (14–16%), lipids (12–23%), dietary fibers (8–10%), carbohydrates (34–52%), vitamins, antioxidants, and essential unsaturated fatty acids.^131,132,134 Because of their nutritive value, rice husk and bran are mostly utilized as an animal feed ingredient, fertilizer, or fuel. More recently, it is also used for the extraction of vegetable oil and bioactive ingredients.^132−135 The main limitations for valorization of these byproducts for human nutrition or higher-value applications are the presence of antinutrient compounds and their fast oxidation and spoilage, which would require a rapid and efficient stabilization treatment.¹³⁴

3. General Processing Routes for Food Protein Waste

Through the process from raw materials to on-shelf products, the food-processing industry generates large amounts of industrial food waste, which has always been a prevalent and recurring issue.¹³⁶ This food waste can come in the form of discarded solids from the raw material or process effluents, which become part of industrial wastewater. Containing untapped sources of macromolecules, such as fats, proteins, and carbohydrates, this waste demonstrates the potential to be reprocessed into value-added products.¹³⁷

To optimally reutilize these industrial wastes, many processes have been proposed, studied, and performed over the years in order to extract and recover proteins for further applications. In Figure 3 and Table 1, we summarize a total of 128 entries comprising articles and patents regarding the extraction and recovery of animal- and plant-based protein waste, showing a diverse range of extraction techniques used to recover proteins from waste.

Figure 3.

Sankey diagram summarizing the main extraction and recovery processes for proteins from animal- and plant-based industrial waste.

Table 1. Reviewed Literature on Protein Waste and Its Extraction Process into Recovered Proteins.

extraction process	protein	advantages	disadvantages	ref
acid	collagen	high solubility of proteins, easy operation	harsh conditions: possible hydrolysis and functionality alteration	(138−149)
	gelatin			(150−153)
	cereals			(154,155)
hydrothermal	gelatin	easy operation, faster reaction from pressure aid	high equipment cost and energy demand	(156−160)
	cereals			(161−167)
	soy/okara			(168)
ionic liquid	collagen	greener than organic solvents, nonvolatile, can be regenerated	high initial cost, potential toxicity	(169)
	keratin			(170−176)
enzymatic	collagen			(177−184)
	gelatin	efficient, mild extraction conditions, lower environmental impact	costly for large-scale application	(185−187)
	keratin			(188)
	cereals			(161−167)
	soy/okara			(189−194)
	oilseeds			(99,193,195−200)
alkaline	gelatin	high solubility of proteins, easy operation	harsh conditions: possible hydrolysis and functionality alteration	(201−203)
	keratin			(204−208)
	cereals			(119,123,209−214)
	soy/okara			(215−222)
	oilseeds			(99,111,223−232)
alcohol	cereals	extract alcohol-soluble proteins	high solvent usage, high environmental impact	(119,123,233,234)
salt	soy/okara	mild extraction condition, minimal protein alteration	low extraction yield	(235)
	oilseeds			(99,236−243)
membrane filtration	whey	suitable for large scale application	membrane fouling	(244−248)

3.1. Dairy Industry

In dairy wastewater treatment, numerous technologies have been developed to recover whey proteins and reutilize them for various applications. Because of the dilute content of whey proteins in industrial effluents, the liquid stream is usually first concentrated by ultrafiltration,³⁹ followed by diafiltration to remove lactose.^249−251 This produces a whey protein concentrate (WPC), which can either be further purified to produce WPI or fractionated into different protein types. WPC and WPI are both highly nutritious food products containing essential amino acids, bioactive peptides, and antioxidants,²⁵² which have consequently found wide usage in the food industry while also being consumed as protein supplements and meal replacements.

3.2. Meat Industry

To reutilize leftover parts from meat production, rendering is employed to separate fat from solid animal waste using high heat, producing edible fat, such as lard and tallow.^67,253 The residual insoluble meal contains an enriched content of protein, from which collagen and gelatin can be extracted and isolated using dilute acid, enzymes, and hydrothermal treatments involving high temperature and pressure;^53,254,255 whereas, other tissue proteins, such as myofibrillar proteins, are extracted with alkaline, acid, salt, and enzymes.^53,255 Apart from solid waste, process effluents also contain substantial levels of muscle and blood proteins, which can be recovered by isoelectric precipitation after solubilization with alkali.²⁵⁶

The extraction of keratin proves to be difficult due to its resistance to dissolution by common solvents.²⁵⁷ Harsh extraction conditions using strong acids and alkalis, sodium sulfide, and reducing agents have proven to be effective in disrupting the compact structure of keratin and aiding dissolution,²⁵⁸ but the toxic nature of such chemicals limits their usage.^204,259,260 Less toxic extraction methods have also been investigated, such as aqueous extraction using chaotropes and reducing agents,²⁶⁰ microwave-assisted extraction,²⁶¹ ionic liquids,^171,262 steam flash explosion,²⁶³ and enzymes.²⁶⁴ The commercial extraction of keratin produces keratin hydrolysates through either high-temperature hydrolysis with acid and alkali or microbial treatment,²⁶⁵ which are then used in hair care products and leather tanning.²⁶⁶

3.3. Seafood Industry

Because of the difference in protein solubility, acid, alkaline, salt, and enzymes are frequently employed to extract proteins from seafood waste.^267,268 Salt and alkaline extraction are the most common techniques for protein extraction, while acid extraction is employed to dissolve acid-soluble collagen.^269−271 Enzymatic extraction relies on proteases to break down cell walls and hydrolyze proteins into shorter chain polypeptides, producing hydrolysates which are recovered by membrane technology.^272,273 Under harsh extraction conditions, e.g., heat or extreme pH,¹⁶⁰ hydrolysis of collagen occurs, which results in the formation of gelatin.²⁷⁴ This imparts a higher water solubility compared to collagen, while also possessing thermoreversible gelling properties, and thus finds broad usage in both the food and biomedical industry.²⁷⁵

3.4. Plant Industry

Seed proteins can be classified according to the Osbourne classification: water-soluble albumins, salt-soluble globulins, alcohol-soluble prolamins, and alkaline-soluble glutelins.²⁷⁶ The extraction and isolation of proteins from oilseeds are commonly performed using acids, alkaline solutions, salt, and enzymes. Proteins from acid and alkaline solutions are frequently separated by isoelectric precipitation between pH 4–5, and a protein isolate is obtained.^193,277,278

Extraction of plant proteins from meals is often performed with alkaline solutions. The major challenge of extracting protein isolates from oilseed meals is the coextraction of polyphenolic antinutritional compounds, such as glucosinolates and phytic acid in rapeseed,²⁷⁷ chlorogenic acid in sunflower,²⁷⁹ trypsin inhibitors and phytic acid in soybean,²⁸⁰ and tannins and agglutinin in peanut,²⁸¹ which either are toxic for consumption or impart unpalatable tastes.^279,282,283 To produce light-colored protein isolates while retaining their nutritional benefits, two main methods have been explored to remove and separate the phenolic compounds from proteins: removal of phenolic compounds using organic solvents prior to protein extraction;^284−288 and separation of phenolic compounds and protein through adsorption on resins.²⁸⁹ This improves digestibility and palatability, while allowing the safe consumption of protein isolates. Apart from phenolic compounds, carbohydrates are also removed by solvents to increase the protein content of the isolate.²⁹⁰

In addition to alkaline extraction, plant proteins can also be extracted using neutral salt solutions, which could be considered to be a milder method through the dissolution of proteins with ionic strength. After extraction, protein solutions are diluted with, or dialyzed against, water to lower the ionic strength of the solution.^238,239 As the main proteins extracted are water-soluble albumins and salt-soluble globulins, the lowering of ionic strength would result in precipitation of globulins.

Another class of proteins is cereal and grain proteins, such as wheat, rice, oat, barley, and corn. Compared to oilseed proteins, cereal proteins contain a higher percentage of prolamins and glutelins, while having a lower globulin content, with the exception of rice and oat.²⁹¹ Alcohol–water solvents, such as 70% ethanol/water, are used to extract prolamins, such as gliadin, zein, hordein from wheat, corn, and barley, respectively, after which proteins can be precipitated simply by the addition of water.^291−295 Upon removing the prolamins, the remaining albumins, globulins, and glutelins can be extracted by normal alkaline and salt extraction.

In the production of large-scale protein isolates, proteins are generally extracted with alkaline or salt solutions.²⁷⁷ Extracted protein solutions typically go through an ultrafiltration process followed by diafiltration to remove unwanted coextracted compounds, after which they are precipitated at their respective isoelectric points or diluted with water to yield protein isolates.^239,296,297 Different extraction conditions, such as meal-to-solvent ratio, temperature, pH, and additives, can be adjusted to optimize protein extraction yield.

4. Revalued Food

Food proteins re-entering the food-value chain as revalued food ingredients are generally characterized by a decrease in the sensorial, nutritional, physical, and chemical properties of the food profile associated with the foodstuff within which they are integrated. Moreover, the extracted proteins from food waste must meet FDA requirements regarding hygiene and safe consumption, which usually requires that the process of producing edible food from food waste be stringent. Two important factors of food consumption from a nutritional perspective are amino acid intake and protein digestibility,²⁹⁸ with an emphasis on the content of essential amino acids, as they can only be obtained from food consumption. Lysine is the most sought-after amino acid in food, while methionine plays a major role in cell metabolism and antioxidative protection,²⁹⁹ making their content in food an important consideration in nutritional intake.³⁰⁰

In the animal food industry, whey, collagen, and keratin are the main sources of industrial waste proteins. Liquid whey has been reprocessed into food for human consumption, such as whey protein isolates, as a result of their high protein content and complete amino acid profiles.³⁰¹ Whey protein has also been commonly incorporated into foods, such as meat replacements, bakery and confectionery products, yogurt, and ice cream.³⁰² Collagen is mainly extracted, purified, and processed into hydrolyzed type I and II collagen, and commonly used as a food nutraceutical supplement to promote collagen production to overcome aging and drying of the skin.³⁰³ The hydrolyzed/degraded form of collagen, gelatin, has already been extensively used in food applications, such as consumables, food additives, and gelling agents.²⁵⁵ In addition to collagen and gelatin, efforts have been made to produce protein hydrolysates from other types of proteins, such as muscle proteins, as they also provide a complete amino acid profile for nutrition.³⁰⁴ In the fishing and seafood industry, much interest exists in collagen-rich waste from fish skin, bones, and scales.³⁰⁵ The major advantages of consuming collagen and gelatin from fish waste over animal waste are the lower risk of diseases, toxins, and contaminants, lower immunogenic and inflammatory responses, and religious prescriptions. Similarly, studies have involved recovering collagen as protein hydrolysates for different food applications,^75,306 some of which have been commercialized for consumption and use in beverages.²⁷² One notable example of reusing leftover fish parts is fermentation into fish sauce, primarily found in the culinary sector in Asia and Europe. Additionally, nutritious fish oil containing high contents of polyunsaturated fatty acids is extracted for human consumption, leaving residual fishmeal rich in protein with balanced amino acid profiles and good digestibility.²⁶⁸

Extracted keratin proteins in the form of hydrolyzed peptides are mainly utilized in hair formulation products and hair growth-promoting food supplements due to the high content of cysteine. Research involving keratin as food hydrolysates demonstrated the potential for keratin to be used as an alternative to dairy protein,^307,308 although, as mentioned above, the main drawback of keratin is the unbalanced amino acid profile which limits its usage as a dietary supplement.

In recent years, oilseed proteins and cereal proteins have been explored for their protein extraction³⁰⁹ and application as food.³¹⁰ For example, to extract and isolate soy proteins from soybean meal, processing methods from Figure 3 are used to produce soy protein isolates (SPI) commonly used in human nutrition as an alternative to animal-based products. Being a staple in Asian cuisine, soy protein provides an amino acid profile similar to that from animal sources, as well as high digestibility of more than 90%.^311,312 Furthermore, soy protein isolates exhibit good emulsifying, water adsorption, and textural properties, which find wide use in food additives, confectionary, pastry-making, beverages, and alternative meat.³¹² Soy protein can also be further hydrolyzed to produce hydrolysates to additionally improve the physicochemical properties.³¹³

Another form of waste from soybeans which has the potential for reintroduction into the food chain is okara.³¹⁴ While most commonly employed as livestock feed due to its protein content between 20–30%,³¹⁴ okara is consumed in China and Japan as a side dish and human dietary supplement to control diabetes, hyperlipidemia, and obesity, as it is rich in fibers, isoflavones, and minerals.⁸⁷ Additionally, fermented okara can be used for the production of bioactive compounds and other nutritional ingredients.³¹⁵ Okara has also shown improvements in nutrition through fermentation with different bacteria strains, increasing accessibility to macronutrients and improving digestibility.^314,316,317 Alternatively, proteins can be extracted from okara to produce protein isolates, which have similar amino acid profiles to soy protein.^317,318

In addition to soy protein, an increasing trend exists of searching for soy alternatives due to the allergenicity of soy to certain groups of people.³¹⁹ Oilseed proteins provide an alternative avenue for nonallergenic plant protein sources. The main hurdle, however, lies in the removal of antinutritional factors, which are toxic and decrease palatability. The shift toward plant-based diets has made more progress with studies demonstrating the viability of incorporation of oilseed proteins into the food industry as a result of their nutrition, physicochemical properties, and technofunctionalities,³²⁰ as shown by the increasing market production of commercial nutritious and functional protein isolates, such as those from rapeseed,^226,321 sunflower,^322,323 and peanut.³²⁴ Examples of some recent commercial protein powders produced from oilseed meal include (rapeseed) CanolaPRO from DSM,³²⁵ (rapeseed) Supertein and Puratein by Burcon,³²⁶ and (sunflower) Sunprotein by Biotechnologies.³²⁷

Zein is the main protein of corn and can be extracted from corn gluten meal with acid, alkaline, and alcoholic solvents.¹¹⁹ Although native zein is not recommended for consumption due to the unbalanced amino acid profile, composed mostly of hydrophobic residues and almost no essential amino acids,¹¹⁹ enzymatic and chemical modifications can be made to enhance zein’s functional properties to be reintroduced into food,¹²² where zein can serve as a network-forming additive to, for example, produce strong meat analogues.³²⁸ Wheat gluten is a byproduct of wheat starch processing from wheat, and apart from gluten-free diets, has found many food applications, such as food texturizers, food fortification, and mock meat (seitan).³²⁹ Solubilized wheat protein isolate has also been shown to improve flavor and texture and was well received by consumers in a study.³³⁰ One key modification of wheat gluten is deamidation, either chemically or enzymatically, which has been shown to not only improve the emulsifying and foaming properties of gluten, but also to reduce allergenicity.^331,332

Containing 10–16% protein and also bioactive compounds,³³³ rice bran is regarded as a high-value byproduct, and has been incorporated into baking and pastry goods to enhance their nutritional value.^334,335 Rice bran also provides the added benefit of being gluten-free, finding use as meat binders, cereal, and nutritional bars.³³⁶ Rice protein concentrates and isolates have also been shown to possess promising emulsifying capabilities, which make them suitable for beverages and coffee whiteners,³³⁴ while a balanced amino acid profile serves well as a nondairy alternative supplement.^163,337,338 Rice dreg, a byproduct from the making of rice syrup, has been shown to be a viable alternative to soy isolates.³³⁹

Proteins from animal sources, such as meat and milk, are considered to be complete proteins, possessing the dietary amounts of all of the 20 essential and nonessential amino acids.³⁴⁰ Collagen and gelatin, as mentioned above, are low in nutritional value, and are known to be deficient in tryptophan and other essential amino acids.³⁴¹ Harsh extraction conditions of collagen and gelatin could result in degradation of cysteine and methionine, lowering their availability.³⁴² On the other hand, the amino acid profiles of plant proteins are not as balanced as compared to animal-derived proteins, with the most significant difference being the lower lysine content and digestibility.³⁴³ Among the different vegetable families, legumes are richer in lysine and deficient in sulfur-containing amino acids; whereas, cereal proteins show the opposite composition, being richer in sulfur-containing amino acids and poorer in lysine.^{301,311,344,345} This imbalance of amino acids among the plant proteins consequently requires the consumption of a mix of proteins and not relying on a single source of protein for nutrition (Figure 4).^301,346

Figure 4.

General comparison of the sources of several essential amino acids among proteins recovered from industrial food waste.

The development of new protein sources for human consumption brings new alternatives for the food industry, potentially benefiting those who are allergic to dairy and soy products, vegetarian, or have religion-based diet restrictions. Although the reutilization of industrial food processing waste is warranted by its large volume generation and variety, several considerations, such as production, safety for consumption, and nutritional quality, would need to be assessed when processing further into food.

5. Upgrading Food Protein Waste beyond the Food-Value Chain

We next consider the application of food protein waste beyond food-value chains and discuss the three pillars of bioplastics, water purification, and renewable energy in detail.

5.1. Bioplastics

In the past decades, plastics have become one of the most abundant man-made materials, and have brought enormous benefits to society due to the countless applications of these low-cost materials characterized by lightweight, high performance, and durability. At the same time, however, the uncontrolled production of synthetic plastic materials, combined with their poor disposal management, has resulted in a catastrophic accumulation of plastic waste and debris (e.g., microplastics) in the environment.^{14,15,347,348} Currently, plastic pollution is a major global issue, with a direct negative impact on climate change, and therefore on human and environmental health.^14,15,348 Many of the environmental burdens derive from the fact that nearly 80% of global plastic production relies on petrochemical feedstocks, implying a negative environmental impact during extraction, as well as disposal (e.g., the release of greenhouse gases during incineration).^348−350 Additionally, a major problem is that approximately 80% of plastic waste is discharged in landfills and accumulate in the environment, where they remain for hundreds or even thousands of years, causing devastating effects on land and sea life due to their toxicity.^{14,15,348,351−353} These issues are also drastically accelerated by the intensive use of single-use plastics, such as packaging materials, that are characterized by a very short lifetime between production and disposal. The use of single-use plastics does also bring certain benefits as, for instance, they can increase the shelf life of products. Consequently, it is very frequently the case that their use cannot be avoided, and thus solutions for limiting their negative impact are urgently needed. Recycling of plastic materials, for example, is a potential approach to mitigate this issue, but unfortunately, this practice is limited to less than 9% of the total plastics produced. This is because it remains a technically and an economically challenging process that can be applied only to certain plastic materials and results in plastics that are often characterized by lower performance and properties.^354−357

More recently, as a result of increasingly strong regulations derived from growing environmental awareness, the plastic market is shifting toward more sustainable materials, with a particular focus on bioplastics. Bioplastics can be produced by fossil-based sources or by biobased materials with the additional advantage of also being renewable.^358−363 By definition, bioplastics are obtained from polymeric materials that are either biodegradable, generated from bioresources, or a combination of both.^364,365 In fact, some biodegradable polymers can be produced from fossil fuel resources, such as polybutylene adipate terephthalate (PBAT) or polycaprolactones (PCL). Additionally, biobased polymers, such as biopolyethylene terephthalate (PET), biopolyethylene (PE), or biopolyamides (PA), are considered bioplastics by definition, because, although not biodegradable, they are obtained from renewable resources like sugar cane or wheat grain. Those polymers that are both obtained from renewable resources and are biodegradable can therefore be considered as the most eco-friendly and sustainable bioplastics. The environmental degradation of biopolymers results from multiple mechanisms that can be nonbiotic (i.e., photolysis, radiolysis, or oxidation) or biotic (i.e., enzymatic activity), and their biodegradation rate depends on numerous different parameters, including the environmental conditions (e.g., aerobic vs anaerobic) but also the physicochemical properties of the biopolymers.^364−366 More recently, it has been demonstrated that when the chemical structure of biopolymers is modified during the bioplastic production process, the biodegradation rate and mechanism can vary compared to those of the unmodified biopolymer: for example, the biodegradation of cellulose acetate involves different microorganisms and enzymes compared to unmodified cellulose.³⁶⁵

Generally, biodegradable polymers can be grouped into two main categories: biopolyesters and agrobiopolymers.³⁶⁷ Biopolyesters can be extracted by micro-organisms, such as polyhydroxy alkenoates (PHA) and polyhydroxybutyrate (PHB), by petrochemical products, such as polycaprolactones (PCL) and aromatic and aliphatic copolyesters, or by synthesis from bio monomers, such as polylactide. Agrobiopolymers instead derive from biomass products, and are based, for example, on polysaccharides, such as starches, pectin, chitin, or lignocellulose, or on proteins, such as zein, soy, gluten, whey, or gelatin.^367,368

During the past few years, intensive research has shown that proteins are one of the most promising classes of biopolymers for developing bioplastics, as they are renewable and abundantly available in nature, resulting in environmentally friendly bioplastics that are sustainable and can be biodegradable, compostable, or even edible. In addition, because of the reactivity of the numerous functional groups characterizing proteins, they offer unique opportunities to develop composite and functional bioplastics. In fact, a very large body of literature has reported that both plant and animal protein isolates can be successfully used to produce bioplastics; unfortunately, however, their practical application remains limited due to the high cost of raw materials (purified proteins/protein isolates), especially when compared to the selling price of petroleum-based plastics.

More recently, different authors proposed the development of biobased plastics through food-waste valorization.^369,370 Most of these works, however, largely focus on sugar-rich biowastes, aiming at the production of biopolymers, such as PHA and PLA, via bacterial fermentation,^{126,371−377} often resulting in low-yield and high-cost processes.

Typically, protein-rich food byproducts used to produce bioplastics are processed into two main categories, wet and dry processing, depending on their respective moisture content and the targeted application. Wet processing involves techniques, such as solvent casting followed by evaporation, while dry processing involves thermoforming techniques, such as injection molding, 3D printing, and extrusion (Figure 5). In what follows, we will discuss only those few works that have focused on the production of protein-based bioplastics starting from food waste with particular attention given to avoiding expensive protein extraction and purification steps.

Figure 5.

Moisture content of food waste, common processing techniques, and targeted applications.

5.1.1. Diary

Although many investigations have been performed on the potential valorization of dairy byproducts into bioplastics, most of them focused on the production of biopolyesters via microbial fermentation of unprocessed liquid whey.^{372,375,376,378,379} Moreover, a few researchers suggested that whey protein isolate can also be transformed into bioplastic films that are transparent, colorless, and edible.^380−383 These studies show that whey proteins possess good film-forming capabilities due to the molecular aggregation of proteins facilitated by hydrophobic interactions, electrostatic interactions, disulfide, and hydrogen bonds, resulting in materials characterized by good water stability and excellent oxygen barrier properties, but also face challenges such as brittleness, stiffness, and high water vapor permeability (WVP) as a result of the intrinsic hydrophilicity of proteins, limiting their practical applications in their pure form. To avoid these constraints, whey protein can be combined with plasticizers (typically water, glycerol, or sorbitol) or blended with other synthetic or natural polymers. For example, blending whey proteins with other natural polymers, such as polysaccharides, improves the thermoplastic properties of whey proteins, allowing the production of films and other shaped products at the industrial scale.^381,384 Additionally, mixing whey proteins with one or more polymers can assist in improving the thermal and water stability of the resultant materials while also reducing the high costs connected to animal proteins, which remains a major limitation to the widespread use of these materials when starting from protein isolates.^385−387 More recently, a few studies reported that dairy waste and byproducts, such as sweet whey,³⁸⁸ acid whey,³⁸⁷ and whey protein concentrate,³⁸⁹ could be used directly as main building blocks to produce films for packaging and edible coatings^390−392 without requiring challenging, time-consuming, and expensive purification steps. Valorization of these byproducts in their original form or after mild pretreatment offers the major advantage of substantially decreasing the costs of raw materials. Certain disadvantages also exist, such as that the exact composition of whey depends on several different parameters, and therefore producing films with reproducible properties may become challenging. Furthermore, whey is highly rich in lactose, which might crystallize during film formation, resulting in inhomogeneous and sticky films. Finally, sweet whey and acidic whey are extremely diluted wastewater that is expensive to transport and store without previous upconcentration treatments.

5.1.2. Meat Processing

Fish^393−398 and animal food waste^399,400 that are rich in collagen have been successfully valorized as main components or as additives to other compounds into transparent and flexible films with good mechanical properties suitable for packaging, edible coatings, and medical applications, with minimal pretreatment.^{73,397,401−404} In addition, partially hydrolyzed collagen, in the form of gelatin, is suitable for these applications with the additional advantages of improved thermoplastic behavior, reversible gelation, and facilitated cross-linking.^{398,405−408} As mentioned above, fish-derived collagen is typically preferred because there are fewer ethical and sanitary concerns compared to those obtained from land animals. Bloodmeal^409−416 and plasma proteins^{411,417−420} derived from slaughterhouses, due to their thermoplastic properties, have also been valorized as ingredients for producing bioplastics by extrusion, injection molding, 3D printing, and solvent casting.

Finally, animal-based byproducts that are rich in keratin, such as chicken feathers^421−429 and sheep wool,^{423,430−432} have been intensively studied and successfully valorized by the production of bioplastics, due to the high stability and low water solubility of keratinaceous materials that are rich in the amino acid cysteine, allowing numerous disulfide bonds and extensive cross-linking that are resistant to water, weak acids, and organic solvents. Yet, as mentioned previously, the extraction of keratin from biowaste requires harsh treatments that might compromise its functionality and performance, while dramatically decreasing its average molecular weight, thus setting a limit for various applications. Films based purely on keratin are usually brittle; consequently, this protein is frequently blended with other polymers to form composite materials to improve their mechanical performance and functionality.

Although many examples exist for animal protein-based bioplastics and edible films, plant proteins are more appealing for these applications, as there are fewer limitations connected to ethical, regulatory, economic, and environmental aspects compared to animal-based products.

5.1.3. Soy

At the beginning of the last century, scientists realized the unique versatility of soybeans and began proposing and patenting the first soy-based plastics.⁴³³ A few years later, scientists at Henry Ford’s factories started to develop and produce textiles and automobile parts, such as exterior panels and frames, knobs and buttons and pedals, using soybean meal and soy proteins (see Figure 6). During World War II, however, the plastic industry focused mostly on petroleum-based polymers, delivering numerous novel synthetic materials with high performance and the ability to meet all of the customers’ requirements. For this reason, research and development of plastic materials based on soy made no significant progress until the 1980s, when environmental awareness catalyzed the development of alternative biodegradable plastics.

Figure 6.

(a) Ford Motor Co. promotional material asserting that agriculture and industry must be considered as one entity (1944). (b) Henry Ford’s experimental “soybean car”. (c) Henry Ford wearing a suit made of soy fibers.⁴³⁴

More recently, because of the enormous quantity of soybeans processed annually, soy protein isolates have become cheap materials that are abundantly available in the market. Therefore, they have been studied intensively for their use in the development of edible films^435−439 and biocomposite materials.^440−445 These works demonstrated that soy proteins could be successfully transformed into bioplastics with satisfactory mechanical properties by solvent casting, as well as by thermoplastic techniques, such as compression and injection molding. The mechanical properties can be further improved by the addition of other ingredients, such as polysaccharides and lipids.^369,446 These soy-based materials, however, are characterized by low water resistance and solubility, which can be further enhanced by acetylation or decreased with cross-linking and post treatments.^369,447,448

Additionally, some studies focused on using byproducts of the soy industry, such as soy whey and okara, and valorized them with minimal pretreatments by producing edible films, functional coating, or films for packaging.^{445,449−456} The use of unprocessed side streams of the soy industry to produce bioplastics offers multiple advantages: the cost of the raw materials decreases substantially, and more importantly, this waste contains different compounds, such as cellulose and lignin, that can assist in improving the physical properties of the resultant materials.⁴⁵⁷ Moreover, soybean waste has a higher glass transition temperature than other bioplastics that, however, can be tuned by changing the composition of the blend.^448,458

5.1.4. Oilseed

Examples also exist for plant protein-based bioplastic^459−463 and edible films^464,465 that have been produced by oilseed protein isolates.^466,467 The bioplastics obtained from these protein isolates are characterized by good gas barrier properties and low moisture content but often also by poor mechanical performance. However, their mechanical properties can be improved by different optimization processes, such as by blending with additives⁴⁶¹ or chemical modification.⁴⁶⁸

Some investigators also suggested that canola/rapeseed cake or meals can be valorized with minimal pretreatment into various biodegradable composite materials, edible films, and films for packaging applications.^{466,467,469−474} Similarly, sunflower protein isolates have been extensively studied as a potential building block for developing edible films and biodegradable composite materials.^{459,462,475,476} To decrease the cost of the raw materials, other studies focused on developing packaging materials and bioplastics using nonprocessed or mildly processed sunflower cake or meals.^{463,476−479} These investigations show that rapeseed cake and sunflower meal can be processed into bioplastics by either solvent casting techniques or by thermoplastic processes and, because of the high concentration of proteins, fiber, polysaccharides, and lipids, they present good potential as reinforcing fillers and raw materials for film production.^474,480 The resulting bioplastics are characterized by lower mechanical performance and lower water stability compared to synthetic materials, but these drawbacks can be partially overcome by tuning the processing parameters (e.g., pressure and temperature) and the blend formulation, as well as by irradiation, heat-induced or enzymatic cross-linking, and acylation.⁴⁸⁰

5.1.5. Cereals

In the past decades, the prolamine protein zein attracted tremendous attention for the development of biobased materials because it is abundantly available, biocompatible, biodegradable, and water-insoluble. Zein, in fact, has been extensively studied for its film-forming capabilities and thermoplastic properties, which are useful for numerous different applications in biomedicine and pharmaceuticals,^124,125,481 packaging,^482−485 and even edible films.^486−489 In particular, zein films can be formed by solvent-casting of ethanol solutions, or alternatively, extruded as the main compound or blended with other polymers. Although the films obtained through solvent evaporation are insoluble in water, grease-resistant, and possess good hydrophobicity, they tend to remain brittle.^482,490 On the other hand, when mixtures of zeins and other biopolymers, such as polysaccharides, are extruded or processed by injection molding, the resulting plastics have higher mechanical properties but lower water stability, which, however, can be improved by cross-linking treatments.^482,491 Most of these studies are based on highly purified zein, implying the relatively high cost of the commonly used alcohol-based extractions, which is limiting widespread applications of this protein isolate in the bioplastics industry. To overcome this major constraint, a few authors proposed the production of bioplastics from raw or mildly purified zein-rich byproducts, such as corn gluten meal.^492−496 This material presents a thermoplastic behavior with a rather high glass transition temperature (approximately 190 °C)⁴⁹⁷ that can be tuned with the addition of plasticizers by compounding or mixing.^497,498 The mechanical properties of the resultant materials and their water stability depend on the blend composition, and especially the ratio between fiber and proteins.⁴⁹⁹ Similarly, distillers’ dry grains can be extruded or processed by injection molding for the production of bioplastic materials.^500−502 In distillers’ dry grains, the protein content is approximately half compared to that of corn gluten meal, while the fiber concentration is 10 times higher, resulting in materials with lower performance compared to those produced with corn gluten meal.⁴⁹⁹

Gluten, from corn and wheat, has also been intensively studied for film-forming properties by solvent casting or thermoplastic processing^503−507 for many different bioplastics in the form of edible films and coatings,^508−512 packaging,^{493,505,513−516} and fibers.^517,518 The properties of the resulting biomaterials vary depending on several parameters, such as the processing technique and the blend composition. For example, the films obtained by glutenin have higher tensile strength and lower elongation at break and WVP compared to gliadin-based films that are also characterized by higher solubility.⁵¹⁹ Gluten bioplastics have good thermal processability that can be enhanced by the addition of plasticizers and by processing in alkaline conditions⁵²⁰ favoring protein bonding and cross-linking. Because of the high price and the nutritional value of pure gluten, some researchers proposed the use of corn and wheat gluten meals without intensive purifications steps as the main component and, more often, as an additive for the production of bioplastic composite materials.^{493,521−525} In fact, gluten meals can be coprocessed with different polymers and biomaterials, such as PBAT,⁵²⁶ PCL,⁵²⁷ or even wood⁵²⁸ to improve the biodegradability of the composites and to achieve targeted properties depending on the applications.

Finally, rice proteins have also been transformed into biodegradable films. Because of the low solubility of these proteins, however, they first require complicated protein extraction and purification processes,^529,530 and also the use of potentially harmful solvents and additives for the production of biomaterials. To overcome these limitations, a few authors focused on valorizing protein-rich rice byproducts, such as rice bran^531−536 and rice wine meal,⁵³⁴ avoiding expensive pretreatments and extraction processes, and used them as building blocks or as additives to develop edible films^534,537,538 and composite bioplastics.^{531−533,535,536,539,540} For instance, the addition of rice bran microparticles into starch films allows improvement of their mechanical properties and water stability.⁵⁴¹ In addition, rice bran can be processed by injection molding, and the properties of the resulting bioplastics can be tuned by the processing conditions and the addition of plasticizers,⁵⁴² as well as by the blend components and the pretreatments of the byproduct. In fact, defatted rice bran results in bioplastics with higher viscoelastic moduli and higher performance compared to untreated bran, whereas, in contrast, lipids provide a plasticizing effect.⁵⁴³

5.1.6. Bioplastics from Denatured Proteins

Although preventing protein denaturation is highly important for food applications, the development of protein-based materials often requires a certain degree of denaturation, which can assist to improve their processability and the properties of the resultant materials, such as their strength, aggregation, and water stability.⁵⁴⁴ Generally, to allow the production of proteinaceous materials with good performance, it is necessary to obtain a high cohesive strength between the protein units, typically promoted by a high degree of entanglement, adhesion, and bonds. These are usually achieved due to exposed domains, such as hydrophobic, polar, hydroxyl, and sulphydryl groups⁵⁴⁴ that become more available in partially hydrolyzed, disordered, and denatured proteins.

However, in most cases reported above, the proteins were utilized in their native state, frequently resulting in materials with poor mechanical properties, high oxygen permeability, and high water solubility. A few studies that focused on purified proteins, however, showed that denaturation can improve the processability and the properties of the resulting films.^545,546 For example, Jerez et al. showed that wheat gluten, albumin, and rice proteins show a thermosetting behavior at temperatures above the denaturation temperature, which assists in improving the properties of the resulting materials by further thermomechanical treatments.⁵⁴⁷ Studies on whey protein films revealed that heat-induced denaturation induces covalent cross-linking between the proteins, resulting in improved tensile properties and lower water solubility,⁵⁴⁶ and also that protein hydrolysis allows reduction of the amount of plasticizer compared to films made with native proteins.⁵⁴⁸ Even in the case of oilseed byproduct proteins,⁵⁴⁹ some studies showed that protein unfolding and hydrolysis resulted in films and bioplastic materials with improved mechanical properties and processability.^549−552

In the very few reports in which the proteins are fully denatured and converted into amyloid fibrils to produce bioplastics,^553−556 the authors demonstrated remarkable improvements in the film mechanical and optical performance, antioxidant activity, water stability, and WVP. However, in these works, high-purity grade proteins were used, resulting in costs that are prohibitive for their widespread application. It is reasonably asserted that the possibility of inducing amyloid self-assembly in protein-rich byproducts, and avoiding expensive purification steps, could have the potential of developing cheaper protein-based bioplastics with enhanced performance.

5.2. Water Purification

Although water purification is the most ecologically friendly approach for addressing the global water issue, developing an efficient, robust, and sustainable technology remains challenging.⁵⁵⁷ Typical water purification technologies are energy-intensive and involve unsustainable synthetic materials.^558,559 As a natural-based solution, proteins can be used for water reclamation due to their functionality and adsorption affinity to the broad range of aqueous contaminants.⁵⁶⁰ Proteins can be employed for fabricating functional superabsorbents, such as films and hydrogels with a hydrophilic network structure and excellent water holding capacity, which can reach thousands of times their dry weight.^561−568 Notably, they possess a strong chelation affinity toward heavy metals due to supramolecular chemical coordination between amino acids available on the protein surface and metal ions.^569−575 Moreover, local electrostatic,⁵⁷⁶ hydrophobic interactions,^577,578 and hydrogen bonding⁵⁷⁹ can also play a role in binding events with organic molecules (Figure 7).⁵⁶⁰ The protein’s primary amino acid residues that contribute to binding with pollutants are cysteine (Cys) and methionine (Met) as sulfhydryl and thioether-containing amino acids, histidine (His) as an imidazole-containing amino acid, aspartic acid (Asp), and glutamic acid (Glu) as carboxyl-containing amino acids, serine (Ser), threonine (Thr), and tyrosine (Tyr) as hydroxyl-containing amino acids.⁵⁸⁰

Figure 7.

A schematic of different binding mechanisms between protein and water pollutants.

Figure 8 presents an overview of various protein adsorbents for water purification. Zein^581−585 and soy^586−588 proteins in their pure native forms have been used more than any other proteins to remove heavy metals and dyes from water. In one work, hollow zein nanoparticles were synthesized using sodium carbonate cores as templates, and then efficiently removed reactive dyes from simulated postdyeing wastewater.⁵⁸² Superabsorbent hydrogels based on zein proteins were synthesized by graft copolymerization of acrylic acid monomers on hydrolyzed protein backbones in the presence of initiators and a cross-linker. The resultant hydrogels showed an excellent water absorbency of 240 g/g and a good copper ion chelation capacity of 208 mg/g.⁵⁸⁹ The generality of using zein as an adsorbent for water purification was recently highlighted by using a zein micro/nanofibrous membrane to remove a wide range of pollutants, including oils, organic dyes, and heavy metal ions, from water. Using an electrospinning system, three types of membranes with different zein fiber morphologies of rod fibers, ribbon fibers, and groove ribbon fibers were designed. The results revealed the superiority of groove ribbon fiber membranes in both mechanical properties and water purification performance. In the optimum condition, adsorption capacities up to 94 g/g for motor oil, 168 mg/g for Congo red, and 189 mg/g for Pb²⁺ ion were obtained.⁵⁹⁰ Moreover, zein in the form of electrospun nanoribbons removed lead from water with an adsorption capacity of 89 mg/g.⁵⁸³

Figure 8.

Various protein adsorbents with different purity grades and processing history used for water purification.

Similarly, amphoteric soy protein-rich fibrous membranes were prepared by coelectrospinning of soy protein and PVA and used for rapid and selective adsorption of cationic and anionic dyes from water.⁵⁹¹ Furthermore, the soy isolate hollow microsphere exhibited an acceptable removal efficiency for Zn, Cr, Cd, Cu, Pb, and Ni.⁵⁹² In other related works, soy protein isolate was incorporated and chemically immobilized in the deacetylated konjac glucomannan matrix to form adsorbents for purifying water from methylene blue⁵⁹³ and copper.⁵⁹⁴ Zhuang et al. introduced a facile synthesis route to fabricate biocomposite aerogels of soy protein and graphene and showed the porous aerogels’ capability to remove two antibiotics from water: ciprofloxacin⁵⁹⁵ and tetracycline.⁵⁸⁷ Additionally, soy protein isolate can be used for biomineralization of inorganic adsorbents for water purification. For instance, cellulose grafted with soy protein isolate was used for biomineralization of hydroxyapatite rod-like nanocrystals. The resulting sustainable adsorbent showed an excellent adsorption capacity of 454 mg/g for methylene blue.⁵⁹⁶

Gluten is another protein with remarkable potential for water purification application, both as a natural adsorbent and gel-forming agent for fabricating composite materials.⁵⁹⁷ Yet, the disulfide bonds inhibit forming porous structures with the untreated gluten, resulting in poor adsorption performance for pollutants. To improve gluten’s properties and adsorption capacity, Zhang et al. proposed biofermentation and acid bath coagulation methods. The result showed the superiority of the acid bath coagulation method for producing more loose and porous adsorbent, leading to a congo red adsorption capacity of 211 mg/g.⁵⁹⁸ In another work, hybrid gluten/PVA nanofibers were prepared by electrospinning their corresponding mixtures with different ratios. Nontoxic and biodegradable nanofiber mats were used for removing nano pollutants from water. The results revealed increased adsorption capacities by increasing the gluten content of fibers, which can be ascribed to the gluten functional groups and hydrophilic nature. In the optimum ratio and condition, the hybrid nanofiber mats presented high removal efficiency of 99% toward Au and Ag nanoparticles (NPs), with a maximum capacity of 37 mg/g for Au NPs and 32 mg/g for Ag NPs.⁵⁹⁹ Various other reports exist on the hybridization of glutens with graphene⁶⁰⁰ and iron⁶⁰¹ to remove pollutants from water. Pirsa et al. prepared different gluten/pectin/Fe₃O₄ composite hydrogel, and applied them to remediate a wide range of contaminations in Lake Urmia, Iran. In the optimum condition, the water purification performance was as follows: 80% reduction in biochemical oxygen demand (BOD), 60% reduction in chemical oxygen demand (COD), 50% reduction in total organic carbon (TOC), and 71% reduction in total heavy metals (THMs).⁶⁰²

Although these reports demonstrated proteins’ capability to remove pollutants from water with acceptable efficiencies, they have been used in their pure forms, which are edible and expensive precursors. Furthermore, in most applications, there is a need to hybridize the protein with other materials to form adsorbents, hindering their direct wide application.

Yet, there are a few reports on applying food waste and byproducts containing proteins directly for water purification.^603−606 For instance, functionalized soy waste biomass after oil extraction was employed for removing Pb, Cu, and Ni from water. Compared to other adsorbents, however, the obtained adsorption capacities were lower.⁶⁰³ In its pure form or in combination with other materials, okara has been widely utilized for water purification from dyes,^607−609 heavy metals,^610,611 and phosphorus.^612,613 Chemical modification of okara can improve dye removal performance. For example, while raw okara’s methylene blue adsorption capacity is 238 mg/g, this value for sodium dodecyl sulfate (SDS) activated okara is 335 mg/g.⁶⁰⁸ Additionally, acid-treated okara can adsorb Acid Red 14 and Reactive Red 15 with adsorption capacities of 217 and 244 mg/g, respectively.⁶⁰⁷ Okara can also adsorb a broad range of heavy metals in water, such as Pb, Zn, and Cd, attributable to hydroxyl-, carboxyl-, and sulfur-based binding sites on its rough surface.^610,611 Soybean residue from bean curd and soymilk production has also been hybridized with poly(acrylic acid) to form a hydrogel for heavy metal removal. Although the adsorption capacities in this work were acceptable for Cd and Pb, hybridizing with synthetic polymers decreased the sustainability profile of the technology.⁶⁰⁴ Finally, cationization of okara using metal salts was used to activate its phosphorus capture ability. The PO₄^3– anions adsorption capacity of zirconium, iron/zirconium, and iron-loaded okara were 48, 41, and 16 mg/g, respectively.⁶¹²

Rapeseed cake was also applied directly and without any purification to remove Cu and Cd from water; however, the resulting adsorption capacity was not promising compared to other adsorbents.⁶⁰⁵ It was also pyrolyzed at 700 °C to form a biochar adsorbent. Although the adsorption capacities for Cu and Zn increased slightly, the carbonization process at high temperatures cannot be considered to be sustainable due to the high carbon footprint.⁶⁰⁶

In another work, human hair, dog hair, chicken feathers, and degreased wool as four typical waste keratin fibers were used directly as adsorbents for the simultaneous removal of eight metal ions (Cr, Mn, Co, Ni, Cu, Zn, Cd, and Pb) from water.⁶¹⁴ The removal performance efficiency was in the order of degreased wool > chicken feathers > human hair > dog hair. Besides some environmental factors, such as sunlight, chlorinated water, and frequent shampooing, that could cause partial oxidation of the keratin biomaterial surfaces, differences in the keratin chemical structure in various sources should be considered. For instance, wool keratin has more sites for disulfide bonds and higher molecular weights than feather keratin, making it a superior adsorbent for heavy metals removal from water.⁶¹⁴ Moreover, keratinous-composed fibers can be chemically modified to improve removal performance.⁶¹⁵

In all of these reports, the waste was used directly for water purification, and as it can contain other compounds, such as polysaccharides, and the prominent role of proteins cannot be unambiguously acknowledged.⁶¹⁶ Furthermore, the significant difference between the purification performance of pure proteins and waste may suggest the inaccessibility of protein-binding sites in the waste to water pollutants.

Finally, a third approach is to extract proteins from waste and then apply them for water purification.^617,618 In this context, keratin^619−621 and collagen⁶¹⁸ were successfully extracted from waste, and then used for removing pollutants from water. As an abundant byproduct of the poultry industry, chicken feathers contain approximately 91% keratin. Zahara et al. extracted keratin from feathers and successfully removed a wide range of metals from water by preparing different pure and modified adsorbents.⁶¹⁷ However, fabricating keratin films and membrane adsorbents is challenging as a result of their weak mechanical properties. In a more recent approach, Song et al. addressed this issue by a molecular network reconstruction strategy, in which β-crystalline structure of silk fibroin template transforms keratin’s free unfolded molecular chains to β-sheet conformation, resulting in a controllable manipulating of the keratin films’ mechanical properties.⁶²² The reinforced composite keratin film exhibited high adsorption efficiency and capacity as high as 191 mg/g and 99%, respectively, for reactive brilliant blue, as well as ideal regeneration performance.⁶²²

In another study, collagen was extracted from leather shaving scrap, which is a byproduct of the leather industry.⁶¹⁸ The extracted collagen was then mixed with alginate and alumina to form adsorbent beads for heavy metal removal. However, the adsorption capacities for Cu and Pb were only 1.7 and 0.4 mg/g, respectively. Gelatin has also been used in the water purification sector as a gel-forming agent to fabricate hydrogel nanocomposite adsorbents.⁶²³

Recently, self-assembly of proteins, including protein waste and byproducts, into amyloid nanofibrils supports the application of proteins as efficient adsorbents and membranes for water purification by endowing them with excellent properties, such as higher surface area, charge density, and available functional groups. Additionally, protein nanofibrils allow the development of advanced membranes, hydrogels, and aerogels compared to monomers. Our previous work comprehensively reviewed and discussed the application of protein nanofibrils for water purification, their advantages, disadvantages, and their sustainability footprints.⁵⁶⁰ The general state of the art in protein nanofibrils for water purification and desalination is summarized in Figure 9. Since their first introduction in the water purification context in 2016,⁶²⁴ research and development of amyloid protein nanofibrils has rapidly increased. Currently, they can be used in most water purification technologies, such as filtration,^624,625 adsorption,^626,627 coagulation–flocculation,⁶²⁸ and distillation,⁶²⁹ significantly improving sustainability footprints over other water purification technologies.⁵⁶⁰ Furthermore, as observed in Table 2 and Figure 9, protein nanofibrils remove a wide range of water contaminants from different categories, such as metals,^{624,625,627,629} radioactive compounds,⁶²⁴ bacteria and viruses,⁶³⁰ pesticides,⁶²⁶ pharmaceuticals,⁶²⁶ dyes,^626,627,629 per- and polyfluoroalkyls (PFASs),⁶³¹ microplastics,⁶²⁸ and natural organic matter (NOM),⁶²⁸ exhibiting a universal role in water purification, which was only previously found in benchmark membrane-based technologies, such as reverse osmosis and nanofiltration. However, protein nanofibrils are characterized by significantly greater energy consumption.⁵⁵⁸

Figure 9.

A summary of the general state of the art in protein nanofibrils for water purification and desalination.

Table 2. List of Pollutants That Can Be Removed from Water by Protein Amyloid Fibrils Technology.

pollutants	form	technology	ref
mono- and multivalent metal ions	film	filtration	(625,637,638)
	aerogel	adsorption	(627)
		solar distillation	(629)
dyes	aerogel	adsorption	(626,627)
		solar distillation	(629)
pesticides	aerogel	adsorption	(626)
pharmaceuticals	aerogel	adsorption	(626)
natural organic matter (NOM)	colloids	coagulation/flocculation	(628)
radioactives	film	filtration	(637,639)
per- and polyfluoroalkyls	film	filtration	(631)
bacteria and viruses	film	filtration	(630)
	aerogel	solar distillation	(629)
microplastics	colloids	coagulation/flocculation	(628)

Protein nanofibrils for water purification have been produced from both highly purified proteins, such as β-lactoglobulin^624,626 oat,⁶³² lysozyme,^628,633 and BSA,^634,635 as well as refined protein byproducts, such as whey.^625,629 The possibility of producing protein nanofibrils for water purification directly from food protein waste is only very recently starting to be recognized.⁶³⁶ In this respect, future research should be directed toward valorizing proteins from food waste and applying them for water purification, by using affordable approaches and minimal processing.

5.3. Renewable Energy

The use of protein-based materials for renewable energy has already been reported and has relied so far on biohybrid materials as diverse as silk,⁶⁴⁰ photoactive proteins,⁶⁴¹ light harvesting chlorophyll proteins,⁶⁴² pigment–proteins,⁶⁴³ and other exotic energy-harvesting proteins.⁶⁴⁴ However, the use of food proteins, and in particular food proteins from waste, for this task remains largely unexplored. This is because, for the great majority of these proteins, no active role of charge carriers, charge transfer, or photocurrent generation is known. A few examples are known, however, which rely on proteins from animal sources, such as insulin and whey, converted into amyloid fibrils to template fully organic or hybrid solar cells and light emitting devices (LEDs) with highly promising physical and optoelectronic properties.^645−649 Because amyloids can be produced from food protein waste, these examples highlight the strong potential of protein-enabled technologies for templating renewable energy devices. The multiple benefits of using amyloid fibrils to template the active layer in optoelectronic devices, converting either electricity into light (LEDs) or solar energy into electricity (photovoltaic solar cells, PSCs), have been demonstrated by Inganas’ group on fully organic devices^645−648 and by our group on hybrid devices.⁶⁴⁹ The role of amyloids has been shown to be many-fold: increase in external quantum efficiency (EQE) by more than 1 order of magnitude, outstanding power conversion efficiency (PCE), high charge mobility, as reflected by the fill factor (FF), and high charge generation, as indicated by high short-circuit currents J_SC.

In a first report, Tanaka et al.⁶⁴⁵ combined bovine insulin amyloids with a polyfluorene (PPF) to generate fully organic polymer light emitting diodes (PLEDs) (Figure 10). The researchers used a classical multilayered device, in which the active layer was made of either pure PPF or a blend of amyloids and PPF. Because the amyloids and PPF combination introduces roughness of the active layer, upon evaporation sputtering of the LiF cathode, the active layer cross-sectional area is increased, thus enhancing electron injection. This lowering of the cathodic barrier height enables efficient electron injection through tunneling at an applied voltage below 5 V, leading to a 10-fold increase in the EQE of the amyloids and PPF combination compared to the pure PPF case.

Figure 10.

(a) Polymer light emitting devices (PLEDs) based on amyloid and polyfluorene combined active layers. Adapted with permission from ref (645). Copyright 2008 American Chemical Society. (b) Illustration of white-light PLED by a combination of dyed-insulin amyloid fibrils and a blue-emitting polyfluorene. Adapted with permission from ref (646). Copyright 2010 American Chemical Society.

In a follow-up publication, the concept was advanced to generate white light PLED.⁶⁴⁶ Insulin amyloid fibrils were first dyed with red and yellow phosphorescent Ir complexes, and then mixed with a blue-emitting polyfluorene to obtain white emitted light. It was found that the performance was increased by the presence of the amyloids compared to the bare Ir complexes. Specifically, by incorporating the Ir complexes in the amyloid fibrils, the authors were able to suppress the undesirable Dexter back transfer process from the phosphorescent emitters to the polyfluorene matrix, consequently increasing efficiency and lifetime of the photons.

A particularly appealing possibility demonstrated by the Inganas group is that of using amyloid fibrils to engineer the active layer of fully organic photovoltaic cells. Barrau et al.⁶⁴⁷ used the usual insulin amyloid fibrils in combination with a polyfluorene to establish an electron donor pseudocomponent and then mixed this with a PCBM electron acceptor component in a thin film acting as an active layer of a sandwiched solar cell (Figure 11). Several characteristics of the resultant solar cells were carefully studied. The short-circuit currents J_SC, which is a parameter that is sensitive to the morphology of the active layer, was found to have a maximum value for amyloid concentration of 57 g mL^–1. At this concentration, the morphology was found to obtain a high charge generation. The PCE, which is strongly correlated to the active layer morphology and the FF, and associated with the mobility of the charge carriers, was also determined to have a maximum value for amyloid concentration of 57 g mL^–1. All of the data together indicated high charge generation and high charge mobility in the presence of the amyloids, which led to an increase in the roughness of the active layers and an improvement of photovoltaic performance.

Figure 11.

(a) Fully organic photovoltaic cells based on amyloid fibrils: polyfluorene electron donors and PCBM electron acceptors. Adapted with permission from ref (647). Copyright 2008 AIP Publishing. (b) Hybrid photovoltaic cells with amyloid fibrils are utilized to template both the electron donors (based on the amyloid and polythiophene combination) and electron acceptors (based on the amyloid and TiO₂ combination). Adapted with permission from ref (649). Copyright 2012 Wiley.

An alternative approach to use amyloids to assist the formation of the active layer in hybrid photovoltaic cells was proposed by Bolisetty et al.⁶⁴⁹ In this approach, the amyloid fibrils were made of β-lactoglobulin and played a crucial role in the design of the active layer: they were first blended with TiBALDH precursor to template TiO₂ nanoparticles directly on the surface of the amyloids, as a first electron acceptor pseudocomponent, then β-lactoglobulin was mixed with polytiophene to form the electron donor pseudocomponent, the two were then assembled into an active layer of a sandwiched photovoltaic hybrid cell. Importantly, the resulting device was found to have a fill factor (FF) as high as 0.53 and a power conversion efficiency approaching 1% (0.72%). This clearly illustrates how the performance of amyloid-templated solar cells can be increased over fully organic devices to acceptable performances when hybrid devices are correctly engineered.

Most importantly, some general trends emerge from the use of amyloid fibrils in renewable energy applications: their high surface to volume ratio generally increases the exchange surface of the electron donor–acceptor heterojunctions and thus their performance, their amino acids availability can be used to template electron donor and acceptor species and, in the case of hybrids devices, even assist their chemical reduction, finally, their extreme expect ratio promote unidirectional hole/electron transport in most optoelectronic applications. These are some of the key structural features that highlight the superiority of amyloid fibrils over native proteins in templating renewable energy devices.

6. Sustainability and Life Cycle Assessment (LCA) of the Technologies

Environmental protection is often viewed as being more or less synonymous with sustainability. The reason for this is the concern for intergenerational equity in welfare between this and future generations because Nature’s resources and services are regarded as the foundation for society.⁶⁵⁰ As a powerful tool to quantify the environmental impact of products and technologies, LCA facilitates choosing the most eco-efficient route to achieve a specific functionality or service.^650,651

Although valorization of proteins from sidestreams, in essence, can be considered a sustainable practice and complies with circular economy principles,⁶⁵² the entire upcycling process, including extraction, materials fabrication, and application, also needs to be sustainable.⁶⁵³ In some cases, the upcycling process involves toxic chemicals⁶⁵⁴ or requires high energy inputs,⁶⁵⁵ which jeopardizes the green aspects of the entire valorization process.

Here, we evaluate the environmental footprints of valorization of two model animal and plant proteins derived from waste, i.e., whey and sunflower protein, for water purification and bioplastic fabrication, using four separate LCAs. We also compare the environmental impacts with another biobased product for each technology, taken as a benchmark. The LCAs were attributional and prospective for emerging products and performed based on the ISO14040/44 standard.⁶⁵⁶ The source of life cycle inventory (LCI) was the Ecoinvent 3 database, and SimaPro v. 8.3 was used to generate life cycle models. The cradle-to-use life cycle impact assessment (LCIA) was obtained using the ReCiPe midpoint and end point for a broad range of environmental impact categories and cumulative energy demand (CED) for energy consumption. The assessed 18 midpoint impacts are climate change (CC), ozone depletion (OD), terrestrial acidification (TA), freshwater eutrophication (FE), marine eutrophication (ME), human toxicity (HTOX), photochemical oxidant formation (POF), particulate matter formation (PMF), terrestrial ecotoxicity (TTOX), freshwater ecotoxicity (FTOX), marine ecotoxicity (MTOX), ionizing radiation (IR), agricultural land occupation (ALO), urban land occupation (ULO), natural land transformation (NLT), water depletion (WAT), metal depletion (MET), and fossil depletion (FOS). The impacts on human health, ecosystems, and resources are also considered for endpoint impacts.⁶⁵⁷ Finally, direct and indirect energy use throughout the valorization life cycle and the sources are determined by the CED method.

6.1. LCA for Protein Valorization for Bioplastic Fabrication

In this section, via different LCAs, we assess cradle-to-use life cycle impacts of producing 10 000 m² (as a functional unit) of hybrid bioplastic packaging films made from whey and sunflower protein (SFP). As a benchmark, we further compare the LCAs results with the environmental impacts of producing packaging films of polylactic acid (PLA), which is a well-known bioplastic. As observed in Supporting Information, Figures S1 and S2, the hybrid protein bioplastics were made largely of protein powder, along with a biopolymer (starch) and a plasticizer (glycerol). In addition, water was used as a solvent in the fabrication process of both bioplastics.⁵⁵³ Because inventory data for sunflower protein were not available in the Ecoinvent 3 database, the LCA for its valorization with acceptable purity grade from sunflower meal was considered based on the process⁶⁵⁸ shown in Supporting Information, Figure S2, in the whole LCA for sunflower bioplastics. Our laboratory experiments provided the process data for both hybrid protein bioplastics, including the mass and energy balances. For PLA, the LCA was built on inventory data for PLA pellets production via Ingeo technology from corn,⁶⁵⁹ followed by an extrusion step to produce packaging films. The LCI for bioplastics made of whey, sunflower protein, and PLA are listed in Supporting Information, Tables S1–3, respectively. For packaging film thicknesses of 70 μm, the yield is 9.2, 10.3, and 10.8 m²/kg⁶⁶⁰ for sunflower protein, whey, and PLA bioplastics, respectively. Therefore, for producing 10 000 m² of each packaging film, 1087.9, 974.6, and 925.9 kg of sunflower protein, whey, and PLA bioplastics, respectively, are needed. Consequently, these values were used as inputs for building the comparative LCA.

The exact and normalized environmental impacts of producing bioplastics from whey, sunflower protein, and PLA based on the ReCiPe Midpoint method are presented in Table 3 and Figure 12a, respectively. As observed, producing bioplastics by valorizing whey or sunflower protein in almost all categories results in lower impacts than PLA. Among all 18 impacts, MTOX, FTOX, HTOX, and FE have the highest relative adverse effects on the environment. Higher amounts of marine, freshwater, and human ecotoxicities for PLA production are attributable to a more significant release of toxic elements, such as heavy metals generated by the electricity production process. The LCA result also reveals phosphate as the primary contributor to FE, which consequently can cause anoxic events, accelerated growth of algae blooms, alteration of biomass, and species composition.⁶⁶¹ While producing 10 000 m² packaging film of PLA approximately results in 9.1 ton CO₂ emission, production of hybrid bioplastics of whey and sunflower protein only emits 2.1 and 3.6 ton CO₂, respectively (Figure 12b). This highlights the superiority of protein valorization scenarios for Bioplastic applications in reducing greenhouse gas emissions and minimizing global warming. As observed in Figure 12c, PLA bioplastic production depletes the water resources by approximately 4.5 and 9 times more than the hybrid bioplastic of whey and sunflower protein, respectively, primarily ascribed to the lactic acid fermentation step⁶⁶² and dependency on corn as a water-intensive crop,⁶⁶³ as well as higher synthesis energy demands. The CED results in Figure 12d further reveal the higher energy demands for PLA bioplastic production than the two other bioplastics, in which the energy consumption for producing bioplastics from whey, sunflower, and PLA was 34, 60, and 158 GJ, respectively. All 18 midpoint impacts can cause three end-point damages to human health, ecosystems, and resources. The exact and normalized end-point environmental damages are presented in Table 4 and Figure 12e, respectively. As observed, compared to the two other protein-based bioplastics, PLA production imparts the most severe damage to human health, mainly due to the high emission of carbon dioxide, manganese, nitrogen oxides, particulates, and sulfur dioxide. Production of sunflower protein-based and PLA bioplastics produces greater damage to the ecosystem than whey-based bioplastics. In this context, the main contributors to ecosystem damage are land occupation and forest conversion for agricultural activities and energy-related carbon dioxide emission to the air. Furthermore, PLA production, as the most energy-intensive process examined in the current study, results in the most serious damage to resources and primarily to fossil energy sources. In aggregate, these results support that bioplastics production by valorizing the protein from sidestreams is a sustainable solution for the pressing plastics pollution challenges, which also fits well in the circular economy model.

Table 3. LCA Impact Assessment for Bioplastics Produced from Whey, Sunflower Protein, and PLA Based on the ReCiPe Midpoint Method.

impact category	unit	whey	sunflower protein	PLA
climate change	kg CO2 eq	2094.448	3671.44	9077.985
ozone depletion	kg CFC-11 eq	0.000101	0.000409	0.000476
terrestrial acidification	kg SO2 eq	12.94995	19.16364	47.73888
freshwater eutrophication	kg P eq	0.873418	1.13642	4.493611
marine eutrophication	kg N eq	3.211778	7.049526	6.64812
human toxicity	kg 1,4-DB eq	633.6074	804.5115	2903.606
photochemical oxidant formation	kg NMVOC	5.597357	10.43946	27.18646
particulate matter formation	kg PM10 eq	5.969955	6.539966	28.65374
terrestrial ecotoxicity	kg 1,4-DB eq	1.234396	6.255091	7.581558
freshwater ecotoxicity	kg 1,4-DB eq	15.57127	24.7231	80.89547
marine ecotoxicity	kg 1,4-DB eq	14.83008	18.25125	72.17246
ionizing radiation	kBq U235 eq	231.6501	66.98405	1204.864
agricultural land occupation	m2a	477.8613	4691.397	1619.349
urban land occupation	m2a	11.92866	9.476093	101.6272
natural land transformation	m2	0.509642	0.455588	1.180786
water depletion	m3	35.93095	70.60064	320.0767
metal depletion	kg Fe eq	38.45285	46.75312	124.1194
fossil depletion	kg oil eq	527.7857	1115.715	2293.289

Figure 12.

LCA of bioplastics produced from whey, SFP, and PLA. (a) Normalized environmental impact profile obtained via the ReciPe Midpoint method. (b) LCA results comparison based on climate change impact. (c) LCA results comparison based on water depletion impact. (d) Energy demand comparison based on CED impact. (e) Normalized environmental impact profile obtained via the ReciPe Endpoint method.

Table 4. LCA Damage Assessment for Bioplastics Produced from Whey, Sunflower Protein, and PLA Based on the ReCiPe Endpoint Method.

damage category	unit	whey	sunflower protein	PLA
human health	DALY	0.004932	0.007406	0.022213
ecosystems	species.yr	0.0000269	0.000124	0.000105
resources	$	90.00143	187.7567	387.8626

6.2. LCA for Protein Valorization for Water Purification

To shed light on the environmental impacts of the valorization of protein sidestreams for water purification, we perform different LCAs by considering two model protein adsorbents based on amyloid fibrils from β-lactoglobulin (β-LgF), the main protein in whey and high-grade sunflower protein (SFPF). The results are then compared with the impacts of using another biobased fibrillar adsorbent: cellulose nanofibrils.

Supporting Information, Figure S3, shows the processes for the valorization of β-Lg from sweet whey,²⁴⁹ followed by the fibrilization step for producing amyloid fibrils.⁶³⁸ The procedures for obtaining high-purity sunflower protein from sunflower meals, and its amyloid fibrils preparation are depicted in Supporting Information, Figure S4.⁶⁶⁴ The mass and energy balances for these processes for both proteins were obtained by our laboratory experiments. Moreover, for CNF, the inventory data were acquired from the LCA performed by Li et al.⁶⁶⁵ to prepare TEMPO-oxidized CNFs via the homogenization route, which has the lowest environmental impact compared to the other preparation processes. The LCI for β-LgF, SFPF, and CNF for water purification are listed in Supporting Information, Tables S4–6, respectively.

Without loss of generality, the LCA functional unit was set for treating 10 m³ of water polluted by toxic gold salts to mimic electroplating-polluted water with a concentration of 100 ppb (removal of 1 kg gold) via an adsorption process. The gold ions adsorption capacities of β-LgF, SFPF, and CNF are 168.2, 149.0, and 15.4 mg/g,⁶⁶⁶ respectively. Therefore, to achieve the LCA goal of treating 10 m³ of 100 ppb gold-polluted water, 5952.4, 6711.4, and 64745.9 kg of β-LgF, SFPF, and CNF, respectively, are needed and employed as inputs for the comparative LCA.

The midpoint environmental impacts of purifying water from gold salts by using amyloid fibrils from β-LgF and SFPF, compared to CNF, are listed in Table 5. The relevant normalized values for each impact are shown in Figure 13a. As observed, compared to β-LgF and SFPF, almost in all categories, purifying water using CNF has more deleterious effects on the environment, particularly by eutrophication and human, terrestrial, marine, and freshwater ecotoxicities. The high energy and ethanol used in the CNF preparation process contribute most to these environmental impacts, particularly by releasing heavy metals, metalloids, and pesticides. As can be seen in Figure 13b, purifying water using CNF causes 74 and 17 times more CO₂ and greenhouse gas emissions compared to β-LgF and SFPF, respectively. LCA results indicate the higher CO₂ emissions in the CNF case are caused mainly by higher energy demands and consequently the use of more fossil fuels, as well as deforestation and land transformation. It should be noted that compared to typical membrane filtration technologies, such as nanofiltration and reverse osmosis, protein nanofibrils technology has a lower operating energy demand and thus causes a reduced amount of CO₂ release in the atmosphere. By considering that the key building block of the technology, i.e., proteins, are valorized from waste and sidestreams, these results show that protein nanofibrils constitute a clean and zero-emission technology for water purification. In contrast, producing CNF is a water-intensive process and has a devastating effect on water resources, largely attributable to the use of ethanol and the release of chemical precursors in water, as well as higher energy demands (Figure 13c).⁵⁶⁰ To elucidate the energy demands for using each adsorbent for water purification, we compare the requisite energy consumption for producing the needed amounts of each adsorbent via the CED method. The energy demand for using CNF in this application is 182 TJ, which is 60 and 10 times higher than for β-LgF (3 TJ) and SFPF (18 TJ), respectively (Figure 13d). This result confirms energy consumption as the “Achilles’ heel” of CNF production.⁶⁶⁷ The exact and normalized endpoint damage to human health, ecosystems, and resources is presented in Table 6 and Figure 13e. As can be seen, purifying water with CNF, compared to β-LgF and SFPS, produces the most serious damage in all three categories. The most significant contributors to the damage from CNF to human health are CO₂ emissions from fossil fuels, land-use change, and release of manganese, particulates and nitrogen oxides, and sulfur dioxide. In the SFPF case, using hexane for defatting and enhancing the protein extraction yield can be regarded as a potential threat to human health; however, this is not comparable with the severe damage of using CNF. Eventually, the process of using hexane for defatting can either be recovered and reused, or replaced by green solvents with lower environmental impacts.⁶⁶⁸ In addition to CO₂ emissions, land occupation and intensive forest transformation impart the most severe ecosystem damage when using CNF, which can mainly be ascribed to cellulose pulp and ethanol production. As discussed previously, compared to CNF, using β-LgF and SFPS is a more energy-efficient process, resulting in lower consumption of fossil fuels and resources. Taken together, the LCA again demonstrates the high potential of valorizing proteins by converting waste protein streams into amyloid fibrils to be used as adsorbents for water purification and thus mitigating two major global environmental issues at the same time: waste management and water scarcity.

Table 5. LCA Impact Assessment for Water Purification by Amyloid Fibrils from β-LgF, SFPF, and CNF, and Based on the ReCiPe Midpoint Method.

impact category	unit	β-LgF	SFPF	CNF
climate change	kg CO2 eq	106688.7	470727.7	7854172
ozone depletion	kg CFC-11 eq	0.018476	0.196777	0.488478
terrestrial acidification	kg SO2 eq	1147.886	2868.183	43394.05
freshwater eutrophication	kg P eq	26.58099	140.469	3380.303
marine eutrophication	kg N eq	247.7508	247.3965	9514.275
human toxicity	kg 1,4-DB eq	23399.57	147311.1	2606326
photochemical oxidant formation	kg NMVOC	258.5838	3510.245	35426.55
particulate matter formation	kg PM10 eq	263.2823	1026.103	25493.61
terrestrial ecotoxicity	kg 1,4-DB eq	157.671	234.4036	51601.31
freshwater ecotoxicity	kg 1,4-DB eq	909.6991	4149.839	66650.65
marine ecotoxicity	kg 1,4-DB eq	541.6348	3475.326	57056.02
ionizing radiation	kBq U235 eq	2100.426	67634.13	829372.9
agricultural land occupation	m2a	17409.7	162640.6	5374454
urban land occupation	m2a	261.9081	3755.068	231588.5
natural land transformation	m2	82.90611	375.6684	31385.01
water depletion	m3	2208.802	8201.955	461814.2
metal depletion	kg Fe eq	1044.194	20966.76	149138.5
fossil depletion	kg oil eq	59017.23	396008.4	1679906

Figure 13.

LCA of water purification by nanofibrils obtained from β-Lg, SFP, and cellulose. (a) Normalized environmental impact profile obtained via the ReciPe Midpoint method. (b) LCA results comparison based on climate change impact. (c) LCA results comparison based on water depletion impact. (d) Energy demand comparison based on CED impact. (e) Normalized environmental impact profile obtained via the ReciPe Endpoint method.

Table 6. LCA Damage Assessment for Water Purification by Amyloid Fibrils from β-LgF, SFPF, and CNF, and Based on the ReCiPe Endpoint Method.

damage category	unit	β-LgF	SFPF	CNF
human health	DALY	0.234267	1.030643	19.46426
ecosystems	species.yr	0.001374	0.007618	0.215977
resources	$	9828.054	66979.04	288297.5

7. Outlook and Conclusions

As a further perspective on the sustainability aspects examined in this review, we conclude by evaluating the environmental impact of the proposed food waste valorization strategies with respect to carbon dioxide emissions based on the planetary boundaries suggested by Rockström et al.¹ Using the data available in Figure 2, we propose two scenarios for decreasing the environmental impact of food waste generated in the food industry. In particular, we consider that half of the industrial byproducts volume is used to extract protein for developing protein-based bioplastics, while the remaining 50% is recovered and reintroduced into the food supply chain at a rate between 25% and 100%.

As summarized in Figure 2a, the total production of protein-rich industrial waste amounts to approximately 1 Gton/year. By considering that half of it can be valorized for the production of bioplastics and that the total average protein content of these byproducts is 32% (Figure 2b), 160 Mtons of protein are potentially available for the production of bioplastics. This amount of protein has the potential to substitute 43.7% of global plastic production, which itself accounts for 4.5% of global CO₂ emissions (2.25 Gton/year),⁶⁶⁹ enabling a CO₂ emission reduction of 1Gton/year.

Annual global CO₂ emissions are reported to be 50 Gton/year,⁶⁷⁰ of which food production comprises 37% (corresponding to 18.7 Gton/year).⁶⁷¹ Global food production is estimated to be 4 Gton/year, indicating that for each kg of food, 4.7 kg of CO₂ is generated and released into the environment. As mentioned in the introduction, approximately one-third of the food produced globally becomes wasted or lost, and it has recently been reported that this food loss accounts for 8–10% of total GHG emissions, equating to ∼4–5 Gton/year of CO₂ emissions.³ By taking the protein-rich byproducts discussed in this paper into account, in which the average protein content is approximately 32%, we find that for each kg of protein wasted, 14.6 kg of CO₂ is emitted in the environment. This review, however, focuses on protein-rich byproducts; we expect the average protein content in food waste to be lower than 32%, further increasing the predicted emissions of CO₂ for each kg of wasted protein to be well beyond 15 kg. In fact, as shown in Figure 14 a, CO₂ emissions derived from the production of proteins can be much higher than this value: the carbon footprint can be as high as 640 kg of CO₂ for each kg of beef proteins produced.⁶⁷²

Figure 14.

(a) Carbon footprints for producing one kg of protein from different sources. (b) Projected CO₂ reductions by a combined food waste management approach relying on reintroduction of food waste into the food-value chain together with implementation of food protein waste into sustainable technologies (the numbers in the figure are derived for a case study considering reintroducing 25–100% of half of global industrial food waste back into the food-value chain, and converting the protein content of the other half into biodegradable plastics. (a) is derived from data presented in ref (672). Copyright 2021 Elsevier. (b) left, is adapted with permission from ref (1). Copyright 2021 Nature.

By reducing the amount of food waste through increasing production efficiencies and reintroducing part of the food waste back into the food supply chain, it would be possible to satisfy global food demands while reducing net food production and subsequently its carbon footprint. For example, by reintroducing between 25% and 100% of half of the entire industrial protein-rich sidestreams (Figure 2) back into the nutrition chain, it would be possible to save 0.6–2.4 Gton/year of CO₂, respectively.

To summarize, the synergistic effect of reintroducing part of industrial protein-rich sidestreams into the nutrition supply chain, as well as for bioplastic production, could potentially decrease CO₂ emissions by 1.6–3.4 Gton/year.

According to the work of Rockström et al., the proposed safe boundary of CO₂ concentration in the atmosphere should not exceed 350 ppm.¹ Unfortunately, atmospheric CO₂ concentration reached 387 ppm in 2009 and has still been rising steadily, reaching an alarming 420 ppm in 2021.^673,674 This corresponds to an increase of 33 ppm of CO₂ over the past 12 years, equating to a rise of 2.75 ppm/year. On the basis of the conversion rate of 1 ppm of atmospheric CO₂ to ∼7.8 Gton CO₂,⁶⁷⁵ the potential reduction of 1.6–3.4 Gton/year of CO₂ by the strategies discussed above would result in a reduction of 0.2–0.4 ppm/year CO₂, or a reduction of 7–16% of excess CO₂ accumulation each year. Although it is clear that the reduction of emissions associated with food waste cannot solely bring the CO₂ atmospheric concentration back to the safety boundary level, these results appear remarkable when considering that food waste contributes to 8–10% of total GHG emissions. The proposed strategies would therefore allow mitigation of the accumulation of CO₂ in the environment in a significant way, as shown in Figure 14. Moreover, if similar measures are also implemented in other fields contributing to global GHG emissions (e.g., fossil energy consumption, emissions associated with water purification, etc.), the climate change issue could be approached in a much more comprehensive and effective manner.

In aggregate, in this review, we have discussed how it is possible to change the carbon footprint of food waste from positive to negative, focusing on the protein part of food waste as its most valuable and versatile component. The all-encompassing and overarching approach to achieve this objective is to perform minimal processing of proteins found in food waste, allowing either their reintroduction within the food-value chain (revalued food) or their use as templates for sustainable technologies, such as those showcased here: biodegradable plastics, water purification membranes, and renewable energy devices. For example, for both energy and water purification, it is found advantageous to reprocess the native or extracted proteins into amyloid fibrils to increase the surface-to-volume ratio and, therefore their efficiency. For bioplastics, the denaturation step is beneficial to unfold the protein and expose the hydrophobic residues and functional groups, thus decreasing the hydrophilicity of ensued bioplastics. By doing so, it should become feasible not only to simultaneously reach several of the 17 sustainable development goals on the UN 2030 agenda but also to provide novel possibilities to the portfolio of protein-enabled sustainable technologies. As a consequence, progress will be catalyzed and accelerated toward an improved equilibrium between social, technological, and economical aspects in a continuously growing world population.

Biographies

Dr. Mohammad Peydayesh received his Ph.D. degree in Chemical Engineering from Iran University of Science and Technology in 2018. He was awarded as a first-ranked graduate of the department in all B.Sc, M.Sc, and Ph.D. programs. Then he joined ETH Zurich as a postdoctoral fellow under the supervision of Prof. Dr. Mezzenga. Since September 2021, he has been working as a senior assistant at the Food and Soft Material Laboratory, ETH Zürich. His research interests are soft matter, self-assembly phenomena, amyloid fibrils, advanced sustainable materials, waste valorization, water purification, and environmental engineering.

Dr. Massimo Bagnani obtained a M.Sc. in Biomedical Engineering at Politecnico di Milano in 2015. He obtained his Ph.D. in 2020 from the Department of Health Sciences and Technology of ETH Zürich under the supervision of Prof. Raffaele Mezzenga, with a thesis focused on amyloid fibrils self-assembly and their liquid crystalline behavior. Now he is working as a postdoctoral researcher in the Food and Soft Materials Laboratory, focusing on sustainable technologies, including food waste valorization and bioplastics development.

Wei Long Soon received his Bachelor’s degree in Materials Science and Engineering from Nanyang Technological University in 2019. He was awarded the Nanyang President Graduate Scholarship for his Ph.D. study. He is currently a visiting graduate student at the Food and Soft Materials group at ETH Zürich. His research interests are materials science, waste valorization, sustainable materials, soft matter, and bioinspired materials.

Raffaele Mezzenga received his Ph.D. from EPFL Lausanne in 2001 and spent 2001–2002 as a postdoctoral scientist at University of California, Santa Barbara, working on the self-assembly of polymer colloids. In 2003, he moved to the Nestlé Research Center in Lausanne as research scientist, working on the self-assembly of surfactants, natural amphiphiles, and lyotropic liquid crystals. In 2005, he was hired as Associate Professor in the Physics Department of the University of Fribourg, and he then joined ETH Zurich on 2009 as Full Professor. His research focuses on the fundamental understanding of self-assembly processes in polymers, lyotropic liquid crystals, food, and biological colloidal systems. He has pioneered the use of protein-based materials in the establishment of new technologies for environmental remediation, health, and advanced materials design. Prof. Mezzenga has been recipient of several national and international distinctions such as the 2011 AOCS Young Scientist Research Award, the 2013 Dillon Medal, and the 2017 Fellowship of the American Physical Society, the Biomacromolecules/Macromolecules 2013 Young Investigator Award of the American Chemical Society, the 2004 Swiss Science National Foundation Professorship Award, and the 2018 Spark Award.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.chemrev.2c00236.

• Process flowcharts and life cycle inventory data for the production of bioplastics and adsorbents for water purification (PDF)

Author Contributions

^∥ M.P., M.B., and W.L.S. contributed equally to this work.

The authors declare the following competing financial interest(s): M.P., M.B., and R.M. are inventors of a patent on protein-based bioplastics filed on behalf of ETH Zurich.