Fermentation, membrane separation, PEF and cascade biorefineries are redefining the recovery of side streams. The industrial challenge is to obtain ingredients with predictable characteristics, consistent production and sustainable balances of energy, water and costs throughout the entire production line.
By Roberto Tognella
In recent years, food upcycling has gradually shifted its focus. Identifying proteins, fibres, pigments or bioactive compounds in industrial side streams is no longer, in itself, the main technological challenge. The challenge lies in transforming streams that are often moist, perishable, seasonal and variable in composition. To become ingredients, these streams must acquire reproducible characteristics that are compatible with safety, functionality and sensory quality requirements.
It is at this stage that the gap between a laboratory trial and a process that can actually be scaled up to production is now measured. Fungal fermentation, single-cell proteins, membrane-integrated enzymatic bioprocesses, pulsed electric fields and cascade biorefineries tackle the problem with different strategies, but converge on the same objective: to control the variability of residual material until products with controllable specifications are obtained. It is in this transition from waste to specification that upcycling begins to be applied on an industrial scale.
Transforming the matrix
Fungal fermentation is one instance where the side stream becomes the substrate for a transformation, rather than merely providing compounds to be recovered. In solid-state and submerged configurations, the microorganism’s metabolic activity alters the composition, structure and functional properties of the matrix, paving the way for the production of micoproteins, enzymes and higher-value ingredients and, in some cases, facilitating subsequent separation operations [1].
BionovFOOD takes this approach to the demonstration scale [2]. In the study by Hoxha et al., grape marc and wine lees are used to cultivate Neurospora intermedia via submerged fermentation in a 1,300-litre bubble column reactor, with an operating volume of 1,000 litres; the biomass obtained is then used in the formulation of meat substitutes. This case thus links a real industrial matrix, a bioprocess on a significant scale and a potential end-use food application [3].
Another application emerges in the form of brewers’ grains: Rusbjerg-Weberskov et al. demonstrate that solid-state fermentation with Rhizopus oryzae increases the protein content on a dry matter basis from 16.8 to 23.3 per cent and that subsequent extraction improves the solubility and functional properties of the resulting fraction [4].
Single-cell proteins: the frontier remains open
A second area of biological research concerns the direct production of high-protein microbial biomass. Single-cell proteins have a well-established technological history; the current frontier lies in using complex agri-food streams as substrates, whilst maintaining productivity, microbiological control and output characteristics that are sufficiently stable to allow for scale-up. The focus therefore shifts to the pre-treatment and standardisation of the feedstock, the continuity of fermentation, and the downstream processing required to achieve repeatable food-grade production.
Two recent studies demonstrate just how diverse the pathways for converting complex side streams into protein biomass can be. Moscariello et al., by combining hemp residues and whey in a sequential anaerobic–aerobic fermentation process, obtained biomass with a protein content of up to 51 per cent and developed a preliminary technical and economic assessment [5].
Amini et al., on the other hand, converted deproteinised whey first into volatile fatty acids and then into biomass from purple phototrophic bacteria, with protein contents of up to 64 per cent; however, the application remains predominantly geared towards animal feed, confirming the gap that still separates some of these solutions from human food use [6]. PROSCALE, a European demonstration project launched in September 2026, is working on this transition. It aims to utilise real agri-food side streams in continuous fermentation systems and scale them up by 2030 [7]. These are therefore project targets, not yet established industrial operating data.
Designing the ingredient: enzymes and membranes
With enzymes and membranes, the problem changes: the complete transformation of the matrix is replaced by a more targeted selection of the fractions of interest. Enzymatic hydrolysis and membrane separations are well-established processes; the innovation lies in designing them together, ensuring that the reaction and fractionation work in tandem to define not only yield and purity— —but also the properties of the final ingredient.
In the study by Jenssen et al., side streams from sea bream and sea bass are subjected to enzymatic hydrolysis followed by 3 kDa membrane fractionation, yielding permeate and retentate with different chemical, biological and techno-functional properties; the retentate, in particular, exhibits good emulsifying, foaming and oil-binding capabilities [8].
Aspevik et al. shift the focus to sensory quality: microfiltration and nanofiltration of poultry protein hydrolysates alter the composition of the fractions, reducing particulate matter, lipids and oxidation compounds, but they also show that the separation process can affect bitterness [9]. The downstream process, therefore, is not merely an ancillary purification stage: it helps determine which ingredient is obtained and what properties it will possess when incorporated into a formulation.
PEF: intensifying the process
Not all the innovations that are redefining upcycling require new transformations of the matrix or new fractionation schemes. In some cases, the breakthrough lies in positioning an already mature technology at the point in the production line where it can make subsequent operations more efficient. This is the case with pulsed electric fields (PEF), used to increase cell permeability and facilitate mass transfer, pressing and extraction.
Pataro et al. demonstrate this using tomato skins from industrial processing: a PEF pre-treatment at 5 kV/cm and 5 kJ/kg increased the rate of lycopene extraction by 27–37 per cent and the yield by 12–18 per cent compared with untreated samples [10]. The industrial significance, however, goes beyond the numerical increase: PEF acts as a pre-treatment and alters the conditions under which the subsequent operation takes place, making the value contained in the side stream more accessible. Performance must therefore be assessed across the entire production line.
In a context other than direct upcycling, Landi et al., when comparing PEF and HTST pasteurisation of orange juice, demonstrate how energy consumption and environmental impact depend heavily on integration with heat recovery and on the overall process configuration [11]. The study concerns a primary feedstock rather than the valorisation of a by-product, but it provides a useful reference for evaluating PEF as a component of a plant system rather than as an isolated unit operation.
Cascade biorefineries: from individual technologies to system architecture
In cascade recovery, the focus of the design is no longer on the individual operation, but on the entire process sequence. When a single side stream contains multiple recoverable fractions, the aim is to organise successive stages so that each recovers value without compromising what remains available for the next stage. Fermentation, extraction, membrane separation, PEF and other unit operations can thus be integrated into the same architecture.
In the case study by Bianchi et al., coffee silverskin is valorised through a cascade of ‘green’ extractions, comprising supercritical CO₂ and subcrit , to recover compounds such as caffeine, phytosterols and polyphenols; the extraction residues are then further valorised as fillers for biocomposites [12]. The final stage therefore falls outside the strictly food-related scope, but highlights the multi-product logic of the cascade: progressively channelling the various components of the biomass towards different applications, rather than focusing the entire process on a single target compound. However, recovering more fractions does not automatically improve the environmental balance.
The LCA analysis by Carpentieri et al. highlights how electricity consumption, freeze-drying and chemical treatments can become significant hotspots and how, in the scenario relating to cherry press-cake, the greater complexity of the cascade resulted in the highest footprint among the pathways compared [13]. The industrial viability of a biorefinery therefore depends on the added value generated in relation to the energy, water, separation processes and complexity it introduces. The technologies considered so far become, at this level, components of a single integrated system.
When upcycling truly enters the factory
The transition to industrial scale hinges on the robustness of the entire system when the raw material changes. A high yield or a successful pilot trial is not enough if variations in the availability and quality of the feedstock compromise process stability, safety or ingredient specifications. Downstream, consumption and economic sustainability are factored into the same equation: what can be optimised separately in the laboratory must function together on the actual production line.
The technologies analysed thus converge on a common criterion: transforming a variable waste stream into an ingredient that the end-user can formulate and purchase with predictable characteristics and a consistent supply. The industrialisation of upcycling will likely depend less on the emergence of new unit operations than on the ability to integrate existing ones, control variability and balance the technical, energy, water and economic aspects of the process. When these conditions align, upcycling can become a genuine process option for the food industry, and not merely a strategy for recovering value from a waste stream.
References
[1] Khatami K., Qazanfarzadeh Z., Jiménez-Quero A. Fungal fermentation: The blueprint for transforming industrial side streams and residues. Bioresource Technology, 440 (2026), 133426. DOI: 10.1016/j.biortech.2025.133426.
[2] European Commission – CORDIS. BionovFOOD – Green processing and valorisation of organic by-products from distilleries, as novel approaches for climate-smart protein applications in alternative foods and the circular bio-economy. Marie Skłodowska-Curie Postdoctoral Fellowship, Grant Agreement 101105437.
[3] Hoxha L., Sucic I., Taherzadeh M.J., Marangon M. Development of new meat analogues from filamentous fungi cultivated on oenological by-products: A quality perspective. Innovative Food Science & Emerging Technologies, 108 (2026), 104409. DOI: 10.1016/j.ifset.2025.104409.
[4] Rusbjerg-Weberskov C.E., Foley J.D., Yang L., Terp M., Gregersen Echers S., Orlien V., Lübeck M. Combined Rhizopus oryzae fermentation and protein extraction from brewer’s spent grain improves protein functionality. Food and Bioprocess Technology, 18 (2025), 9574–9593. DOI: 10.1007/s11947-025-03989-1.
[5] Moscariello C., Matassa S., Oliva A., Lens P.N.L., Esposito G., Pirozzi F., Papirio S. One-stage single-cell protein production from hemp (Cannabis sativa L.) biomass residues and cheese whey through sequential anaerobic-aerobic fermentation. Waste Management, 193 (2025), 462–471. DOI: 10.1016/j.wasman.2024.12.029.
[6] Amini A., Remelli A., Amirian Mojarad V., El Achkar M., Dell’Anno M., Rossi L., Mirra G., Di Giancamillo A., Grassi Scalvini F., Tedeschi G., Parati K., Turolla A. From cheese whey to single-cell protein production: By-product valorisation through acidogenic fermentation and purple phototrophic bacteria. Bioresource Technology, 435 (2025), 132894. DOI: 10.1016/j.biortech.2025.132894.
[7] Circular Bio-based Europe Joint Undertaking – CBE JU. PROSCALE. Innovation Action – Demonstration, project 2026–2030, launched on 1 September 2026.
[8] Jenssen M., Sone I., Grasso F., Turrini F., Tardanico F., De Negri Atanasio G., Méndez Paz D., Vázquez Sobrado R., Alonso Martínez M., Boggia R., Grasselli E., Lian K. ‘Valorisation of aquaculture by-products from sea bream and sea bass by enzymatic hydrolysis and fractionation: chemical and biological insights’. Frontiers in Nutrition, 12 (2025), 1663294. DOI: 10.3389/fnut.2025.1663294.
[9] Aspevik T., Dankel E.K., Leivers S., Afseth N.K., Oterhals Å., Haugen J.-E., Skov M., Gaarder M.Ø. Upcycling of poultry protein hydrolysates using membrane filtration technology – Effects on sensory properties and chemical composition. Food Chemistry: X, 38 (2026), 104181. DOI: 10.1016/j.fochx.2026.104181.
[10] Pataro G., Carullo D., Falcone M., Ferrari G. Recovery of lycopene from industrially derived tomato processing by-products by pulsed electric field-assisted extraction. Innovative Food Science & Emerging Technologies, 63 (2020), 102369. DOI: 10.1016/j.ifset.2020.102369.
[11] Landi G., Benedetti M., Sforzini M., Eslami E., Pataro G. Comparative analysis of cost, energy efficiency and environmental impact of pulsed electric fields and conventional thermal treatment with integrated heat recovery for fruit juice pasteurisation. Foods, 14 (2025), 2239. DOI: 10.3390/foods14132239.
[12] Bianchi S., Vannini M., Sisti L., Marchese P., Mallegni N., Rodríguez Ó., Kohnen S., Tchoumtchoua J., Cinelli P., Celli A. Valorisation of coffee silverskin by cascade extraction of valuable biomolecules: preparation of eco-friendly composites as the final step. Biofuels, Bioproducts and Biorefining, 18 (2024), 524–542. DOI: 10.1002/bbb.2587.
[13] Carpentieri S., Ghanem A., Khwaldia K., Silva A.S., Ferrari G. Life cycle assessment of agro-industrial residue valorisation processes to obtain phenolic-rich extracts. Frontiers in Sustainable Food Systems, 9 (2025), 1693181. DOI: 10.3389/fsufs.2025.1693181.
