Green VALORisation cascade approach of fISH waste and by-products through fermentation towards a zero-waste future

Projektinformationen

VALORISH

ID Finanzhilfevereinbarung: 101135078

DOI

10.3030/101135078

EK-Unterschriftsdatum 21 Mai 2024

Startdatum 1 Mai 2024

Enddatum 31 Oktober 2027

Finanziert unter

• Food, Bioeconomy Natural Resources, Agriculture and Environment

Gesamtkosten

€ 3 935 845,00

EU-Beitrag

€ 3 935 845,00

Koordiniert durch

IDENER RESEARCH & DEVELOPMENT AIE
Spain

Periodic Reporting for period 1 - VALORISH (Green VALORisation cascade approach of fISH waste and by-products through fermentation towards a zero-waste future)

Berichtszeitraum: 2024-05-01 bis 2026-04-30

Zusammenfassung vom Kontext und den Gesamtzielen des Projekts

Every year, more than 90 million tonnes of fish and shellfish are caught worldwide. Once harvested, fish typically undergo processing, which generates significant amounts of by-products such as heads, bones, guts, and shells. These by-products can represent between 30% and 70% of the whole fish. Although they are rich in valuable macro- and micronutrients, they are often discarded or underutilised, leading to both environmental harm and economic losses. Addressing this issue is essential for building a truly circular and sustainable bioeconomy. The VALORISH project seeks to turn this challenge into an opportunity by developing and scaling up an innovative, computationally assisted biorefinery that transforms fish processing by-products into high-value products for food applications. The biorefinery process begins with the extraction of oils and the breakdown of proteins using selected bacteria. These microorganisms are identified through advanced bioinformatic modelling to ensure optimal efficiency. The resulting protein-rich material can be used directly as a dietary supplement or further processed to produce beneficial compounds such as vitamin B12, astaxanthin (a natural antioxidant), and bacteriocins (natural food preservatives). In line with a zero-waste approach, any remaining materials such as fish bones and residual biomass, are also recovered and reused. Minerals are extracted from bones, while the remaining organic material is processed into biogas through anaerobic digestion. All resulting products undergo thorough testing to ensure they meet EU safety, nutritional, and regulatory standards. In parallel, the entire biorefinery system is digitally modelled to support efficient design, simulate optimize, support the scale-up and help the assessment of the environmental and socio-economic impacts. Furthermore, societal acceptance of the resulting products will be analysed. Spanning 42 months and supported by a €3.9 million budget, VALORISH aims to deliver a fully operational pilot at Technology Readiness Level 5. By demonstrating how fish by-products can be transformed into valuable market-ready resources, the project support EU climate and circular economy goals, and offers a replicable blueprint for sustainable innovation across Europe and beyond.

Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse

During the first 24 months, VALORISH established the main scientific and technological building blocks of the cascade biorefinery. Ten fish-processing side-streams were characterised in terms of composition, availability and suitability for valorisation. Representative lipid-rich and protein-rich feedstocks were selected, preserved in compliance with food hygiene requirements and directed to the most appropriate processing routes.
Three green oil extraction technologies (supercritical CO2 (scCO2), subcritical water extraction (SWE) and natural deep eutectic solvents (NADES)) were optimised and benchmarked against conventional methods. Oil quality was assessed to support technology selection, and scCO2 was chosen for scale-up. For the subsequent proteolytic fermentation, strains from culture collections and computational screening were tested using hake, shrimp and surimi side-streams. Bacillus subtilis, Bacillus mojavensis and Priestia megaterium exceeded the 20% hydrolysis target under selected conditions. B. mojavensis and P. megaterium were prioritised for bioreactor trials, and the resulting FPHs were stabilised for further characterisation and use in the cascade. Residual fermentation biomass was also characterised in preparation for anaerobic digestion tests. An analytical framework was established for the chemical, nutritional, safety and sensory assessment of oils and FPHs. Bitterness, astringency and fish-related flavours were identified as the main sensory limitations of FPHs, while nanofiltration reduced flavour intensity and informed the design of purification and debittering strategies.
FPHs were evaluated as alternative substrates for microbial bioproduction. Astaxanthin was produced and quantified in yeast fermentations using FPH-based media, with one shrimp-derived hydrolysate slightly outperforming the commercial reference medium. Cultivation and analytical methods were also established for vitamin B12 and bacteriocins, including sakacin-producing strains.
Fishbone valorisation advanced through the characterisation of salmon and cod bones, collagen extraction and calcination studies. Enzymatic treatment increased collagen solubilisation from below 5% to above 40%. Thermal analysis supported the definition of calcination conditions, and a laboratory-scale calcination system was designed for the recovery of hydroxyapatite or calcium-rich products.
Computational work has supported three areas. A combined proteolytic and metabolic workflow screened 81 food-relevant bacterial strains, prioritised 17 candidates and supported the selection of four for experimental testing. Genome-scale metabolic models were used to assess candidates for astaxanthin, vitamin B12 and bacteriocin production from FPH-based media. Mathematical models were also developed for the core biorefinery modules to support integration, optimisation and scale-up.
The LCA framework was defined and applied to the first available process data. Preliminary results identified scCO2 as the most favourable oil extraction route and showed promising performance for collagen and hydroxyapatite, while highlighting yield, drying and energy demand as improvement areas for FPH and astaxanthin production. The project also initiated social acceptance and stakeholder engagement activities. A first stakeholder survey assessed awareness, expectations and perceptions, and preparations for qualitative interviews began.

Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)

The first 24 months produced measurable results supporting a flexible, computationally assisted cascade biorefinery:
• A nutritional dataset was generated for ten fish-processing by-products, supporting feedstock-specific valorisation routes.
• Three green oil extraction processes were optimised and documented: scCO2 achieved up to 98% oil recovery, SWE reached 85% at 4 L scale, and NADES-based extraction achieved 61–64%. These routes provide alternatives to conventional thermal or solvent-based processing.
• B. subtilis, B. mojavensis and P. megaterium exceeded 20% hydrolysis. The latter two were first identified computationally and then validated experimentally.
• Astaxanthin was produced in FPH-based media, while strains and cultivation conditions were selected for vitamin B12 and sakacin production. This provides initial validation of a sequential fermentation route in which fish proteins are first converted into FPHs and then used for higher-value bioproduction.
• Two bioinformatic protocols were established. The first screened 81 proteolytic candidates, prioritising 17. The second combined genome mining, GEM reconstruction, pathway integration and FBA/dFBA to assess 183 astaxanthin, 167 cobalamin and four nisin A candidates.
• Fish oil extraction models calibrated with project data achieved deviations below 15%; models for the remaining modules were defined for later calibration and integration.
• Collagen solubilisation increased from below 5% to above 40%, and an innovative laboratory-scale calcination reactor was designed for the remaining mineral fraction.
Further uptake depends on pilot-scale validation, sufficient product quantities for safety and end-user testing, model calibration with pilot data, completion of environmental and techno-economic assessments, and clarification of regulatory and standardisation requirements.