Nutraceutical recovery from dry-grind corn ethanol coproduct streams: a three-lane biorefinery framework

1 Introduction

First generation dry-grind corn ethanol plants generate by-product streams rich in nutrients that are disproportionately routed to minor or low-value uses, like distillers’ dried grains with solubles (DDGS). Many by-product streams contain bioactive compounds that could prove more useful in higher-value applications such as food, feed, and health. Coproduct upgrading is an important lever for plant economics, and market pressures historically favor the development of new coproduct pathways. Within that context, the recovery of defined nutraceutical fractions is an attractive proposition as it can convert portions of existing streams into higher-value food, feed, or health-oriented ingredients while utilizing existing unit operations at scale and supporting a circular-economy concept.

During dry-grind milling, the whole corn kernel is liquefied, saccharified, and then fermented without preceding fractionation (Bothast and Schilcher, 2005). Separation and evaporation post fermentation yields whole stillage, thin stillage, condensed distillers solubles (CDS), distillers corn oil (DCO), and DDGS. The dry-grind process acts to concentrate the non-starch components of the kernel (also including yeast biomass) downstream, which positions dry-grind plants as a source ripe for nutraceutical recovery if adequate and appropriate separation techniques are used (Rausch and Belyea, 2006; Kim et al., 2008; Maitra and Singh, 2021). For the purposes of this review, these five coproducts–whole stillage, thin stillage, condensed distillers solubles (CDS), distillers corn oil (DCO), and wet grains or DDGS–are referred to as the key process streams because they contain most non-starch constituents. These streams also represent the primary practical points of entry for nutraceutical recovery in dry-grind plants.

The key streams are replete with nutraceuticals both lipophilic and hydrophilic in behavior (Kim et al., 2008; Di Lena et al., 2020; Liu and Han, 2011). DCO has been reported to contain higher levels of phytosterols and conjugates thereof, tocopherols/tocotrienols, and carotenoids as compared to standard corn oil, with variability between production facilities shaped by a host of processing conditions (Moreau et al., 2010; Winkler-Moser et al., 2023). Compositionally, syrup and thin stillage are comprised of amino acids, peptides, organic acids, Maillard-derived antioxidants, residual oligosaccharides, and various minerals, which can be selected and isolated with adsorption, membrane, and chromatographic unit operations (Kim et al., 2008; Di Lena et al., 2020). DDGS contains ferulic and p-coumaric acids and usually demonstrates a greater capacity for antioxidants on a mass basis than the source grain, indicating that the phenolic fraction can withstand the unit operations used to produce DDGS (and related solids coproducts) (Luthria et al., 2012; Luthria et al., 2014). Furthermore, distillers-derived arabinoxylan oligosaccharides (AXOS/XOS) demonstrate potential as prebiotics, pointing to additional pathways for value addition (Monteagudo-Mera et al., 2018).

When considering valorization of these nutraceuticals, three benefits are evident.

• Valorization promotes a circular economy by enhancing the existing coproducts into ingredients with higher value while utilizing spent capital and existing unit operations at scale (Rausch and Belyea, 2006).
• Selectively recovering nutraceutical compounds can improve both the consistency and quality of the process coproducts. For example, by lowering the amount of free fatty acids or phosphorus present, the range for end-use markets widens (Liu and Han, 2011; Moreau et al., 2010).
• Technologies with a focus on thin stillage (i.e., secondary fermentation, membrane separations, selective extractions) have developed to a point where economic and environmental co-benefits are feasible (Reis et al., 2017; Fortney et al., 2021).

To date, most studies focus on a single stream (like DCO, DDGS, etc.) or nutraceutical class (like phytosterols, phenolic acids, etc.) (Moreau et al., 2010; Luthria et al., 2012; Sharma et al., 2023; Monteagudo-Mera et al., 2018). This single-stream or single-class focus can bottleneck cross-stream comparisons and analyses of stability, recoverability, and distributions. Studies focusing on enzymatic strategies and process intensifications/additions, like a fiber-degrading enzyme or phytase additions which shift oil partitioning, demonstrate how changes in upstream processes impact the concentration and composition of nutraceuticals downstream, but impacts of these strategies and those like them are sparsely explored across all the key dry-grind streams (Luangthongkam et al., 2015).

This review addresses critical gaps in existing literature by providing an original comprehensive assessment that integrates nutraceutical composition, processing influences, and recovery strategies across dry-grind corn ethanol coproduct streams. Specifically, the objectives are to:

• Map and Compare Nutraceutical Classes: Systematically identify and compare the distribution of major nutraceutical classes across the principal process streams, establishing a baseline for targeted recovery and valorization.
• Evaluate Recovery and Purification Methods: Analyze and assess the technical feasibility, scalability, and regulatory compliance of recovery and purification strategies for nutraceuticals within each stream.
• Develop a Nutraceutical-Integrated Biorefinery Framework: Propose an integrated framework for dry-grind corn ethanol plants that prioritizes target compounds based on value, concentration, and ease of recovery, while aligning with sustainability and operational metrics (Reis et al., 2017; Di Lena et al., 2020; Winkler-Moser et al., 2023).
• Establish a Multi-Stream Baseline and Commercial Potential: Connect compositional profiles and recovery strategies to functional applications and market opportunities, highlighting the commercial viability and potential health benefits of coproduct-derived nutraceuticals.

Collectively, these objectives establish a foundation for addressing the cross-stream knowledge gaps that limit nutraceutical valorization in dry-grind systems. The subsequent sections examine each key process stream in detail, evaluate viable recovery strategies, and integrate these insights into an overarching biorefinery model. An overview of the three-lane biorefinery framework and its associated nutraceutical recovery pathways is presented in the graphical abstract and in Figure 2.

2 Comparative mapping of nutraceutical classes across process streams

Understanding the distribution of nutraceuticals within dry-grind corn ethanol coproducts requires a detailed comparison of their occurrence across the major process streams. By systematically examining the prevalence of carotenoids, phenolic acids, phytosterols and their conjugates, tocopherols/tocotrienols, bioactive peptides, dietary fiber/oligosaccharides, residual sugars/organic acids, and minerals in whole stillage, thin stillage, syrup, DCO, and DDGS, it becomes possible to establish a baseline for nutraceutical potential. This comparative approach highlights key opportunities for targeted recovery and valorization and informs subsequent strategies for biorefinery integration.

2.1 Nutraceutical classes and analytical methods

An effective cross-stream recovery design relies on measurements that are comparable between three matrices: oil, liquid, and solid. This section summarizes typical analytical method families (for example, chromatography-based profiling, spectrophotometric antioxidant assays, and targeted mineral analyses) used to quantify each nutraceutical class. It also aims to uncover why each matrix poses unique challenges and which analytical data are commonly reported to supported further downstream recovery decisions. Table 1 shows a consolidation of these targets and data readouts to support distribution and feasibility claims.

TABLE 1. Nutraceutical classes evaluated across dry-grind corn ethanol coproduct streams, which summarizes primary matrices, common analytical data (readouts), method families, and matrix constraints to support compatibility across streams and downstream recovery options.

Abbreviations: GC, gas chromatography; HPLC, high-performance liquid chromatography; LC/MS, liquid chromatography/mass spectrometry; ICP, inductively coupled plasma; HPAEC-PAD, high-performance anion-exchange chromatography with pulsed amperometric detection; MW, molecular weight; DP, degree of polymerization; OSI, oxidative stability index; MF/UF, microfiltration/ultrafiltration; NF, nanofiltration; RO, reverse osmosis; ED/IX, electrodialysis/ion exchange; LLE, liquid-liquid extraction; QC, quality control; QA, quality assurance; ACE, angiotensin-converting enzyme; DPP-IV, dipeptidyl peptidase IV.

To simplify comparison, the concentration per matrix (for example, mg kg^-1 oil for lipophiles, mg kg^-1 dry basis for solids, and mg kg^-1 dissolved solids for liquids) and stream-level mass loads (mg per ton of fermented mash processed) is reported. This normalization aids the analyses of distribution and feasibility that follow, also aiding in prioritizing targets with high inter-facility variability (for example, tocols and carotenoids in DCO) for downstream recovery screening (Winkler-Moser et al., 2023).

2.2 Process streams and sampling strategy

Process streams vary greatly in matrix properties which drive nutraceutical partitioning and feasibility for recovery. The descriptions below emphasize (i) major matrix qualities of each stream, (ii) nutraceutical classes often reported in that stream, and (iii) what those matrix qualities suggest for downstream recovery choices.

Considering handling constraints across wet streams (whole stillage, thin stillage, CDS), these streams contain a higher water activity and can change rapidly because of microbial activity and oxidation. So, comparability will improve when studies minimize hold times, rapidly cool, and separate solid fractions wherever feasible. Given these wet stream handling constraints (high water activity and composition changes), whole stillage and thin stillage are particularly time-sensitive given their position as the initial split between solids and solubles (Lu and Rosentrater, 2015). Each of the five streams’ sampling and preparation methods are described as:

• Whole stillage: sample at the primary centrifuge feed and stabilize quickly (cooling or thermal stabilization as allowed) (Lu and Rosentrater, 2015).
• Thin stillage: sample at the decanter concentrate line, prioritize rapid cooling and early clarification, treat emulsified oil and ash as key constraints for downstream separations (Kim et al., 2008; Liu and Han, 2011; Di Lena et al., 2020).
• CDS: sample at evaporator outlet before blend-back, document thermal history because it influences peptide and sugar behavior (Rausch and Belyea, 2006; Sharma et al., 2023).
• DCO: sample at oil centrifuge outlet, protect from light and oxidation because lipophiles are stability-sensitive, acknowledge inter-facility variability (Moreau et al., 2010; Winkler-Moser et al., 2023).
• Wet grains/DDGS: document CDS blend-back ratio and drying severity because they drive composition variability relevant to functional positioning (Luthria et al., 2012; Belyea et al., 2004; Kingsly et al., 2010; Liu and Han, 2011).

Whole stillage, the post-distillation slurry entering the primary centrifuge, provides the source matrix where the splitting of liquid and solid streams occurs. Quickly stabilizing the whole stillage stream is important because elevated water content and high temperatures can encourage microbial turnover and oxidation, both of which can degrade volatile compounds. Microbial turnover and oxidation risks be prevented through the immediate cooling or pasteurization paired with early clarification processes (Lu and Rosentrater, 2015).

Thin stillage, which is recovered as concentrate from the decanter unit operation, often has acidic qualities, contains high water content, and is high in soluble nitrogen content (yeast peptides, amino acids), organic acids (mostly lactic acids), glycerol, minerals (Na, K, P), residual oligosaccharides, and emulsified oil which carry a considerable portion of the lipophilic bioactive compounds. Practically, the thin stillage matrix pushes for conditioning in the liquid-lane (clarification and deashing) and membrane-based fractionation, while emulsified oil presents an opportunity for lipid capture, however contributing to fouling and selectivity constraints (Kim et al., 2008; Liu and Han, 2011; Di Lena et al., 2020).

When evaporating thin stillage, CDS is produced, which is a notable and concentrated source of protein, ash, glycerol, oil, and Maillard-relevant sugars, therefore a key substrate for recovering peptides while being sensitive to thermal changes during evaporation. So, composition and functionality are sensitive to high heat, and Maillard-related changes can affect screening results downstream and extractability. In recovery contexts, CDS is posed as a rich source for protein and peptide fractions, requiring conditioning and thermal history documentation that is sensitive to changes in ash to support comparability (Rausch and Belyea, 2006; Liu and Han, 2011; Sharma et al., 2023).

On the backend of the process, centrifugation of the liquid stream provides DCO, a coproduct stream which is rich in phytosterols (as well as conjugates of ferulate), tocopherols/tocotrienols, and xanthophyll carotenoids when compared to standard corn oil. It is important to note that much of the compositional variability between facilities is attributed to variation in processing configurations. This stream aligns with an edible-oil style recovery lane where concentration of unsaponifiables and oxidation control are the major drivers in feasibility (Moreau et al., 2010; Winkler-Moser et al., 2023).

The solids coproduct stream, defined as wet grains whether with or without adding back CDS and the resulting DDGS, is concentrated with compounds including protein, fiber, and bound phenolic acids (i.e., ferulic, p-coumaric, etc.), with a higher antioxidant capacity as compared to source grain The composition, including protein, fat, fiber, glycerol, minerals, and color, is largely driven by processing choices, like CDS addback ratios and drying conditions, than variability in source grain kernel. The dependence of wet grains/DDGS composition on CDS blend-back (also called solubles add-back) ratio, which is defined as the controlled recombination of CDS with wet grains prior to drying, and drying severity affects functional ingredient positioning and contaminant/specification planning. It is notable that additions in enzymes in the upstream processes, like xylanase, protease, and phytase, can change the partitioning, which shifts more oil and solubilized lipophiles to the liquid stream. Because upstream enzyme additions can shift more oil and solubilized lipophiles into the liquid stream, solids-lane strategies emphasize controlled release and fractionation that is compatible with coproduct handling and drying constraints (Luthria et al., 2012; Luthria et al., 2014; Belyea et al., 2004; Kingsly et al., 2010; Liu and Han, 2011; Luangthongkam et al., 2015).

Sampling must account for facility and temporal variability, especially for streams which are sensitive to backend oil recovery configurations, CDS handling, and operating conditions. Analysis on compositions is best represented as a daily average for recovery screening because integrating data across multiple processing facilities and time frames enables a more robust cross-stream synthesis than a single-facility dataset (Belyea et al., 2006; Spiehs et al., 2002; Di Lena et al., 2020; Winkler-Moser et al., 2023; Luangthongkam et al., 2015). With the stream contexts and sources of variability highlighted, the next section maps nutraceutical distributions across streams using matrix-normalized reporting and begins to evaluate recovery feasibility per lane.

2.3 Nutraceutical distribution across streams

Nutraceutical distribution is driven by (i) removal of starch during fermentation which enriches non-starch compounds, (ii) liquid-solid separation of during centrifugation, routing solubles and emulsified oil to thin stillage and proteins and fiber to wet grains, (iii) and evaporation with blend-back concentrating thin stillage into CDS and afterwards recombing with wet grains to produce DDGS. These mechanisms demonstrate why a small number of streams act as practical recovery hotspots (Belyea et al., 2004; Belyea et al., 2006; Kim et al., 2008; Liu and Han, 2011).

A heatmap was constructed and shown in Figure 1 to support a three-lane recovery concept: lipophiles concentrate in DCO in the oil-lane, fiber and bound phenolics concentrate in wet grains/DDGS in the solids-lane, and solubles (including minerals and peptides) accumulate in thin stillage and CDS in the liquid-lane (Belyea et al., 2006; Kingsly et al., 2010; Reis et al., 2017; Sharma et al., 2022; Sharma et al., 2023). Figure 1 is critical because constraints specific to each lane drive which conditioning steps are enabling or optional.

2.3.1 Lipophilic antioxidants and sterol conjugates

The most concentrated source of lipophilic nutraceuticals is the post-fermentation oil which has been extracted from the liquid stream (distillers corn oil, DCO). DCO from U.S.-based facilities consistently contains about 1.5–2.5 wt% total phytosterols, including significant amounts of ferulate phytosteryl esters (FPEs), tocopherols/tocotrienols, and xanthophyll carotenoids (lutein, zeaxanthin, and β-cryptoxanthin) (Moreau et al., 2010; Moreau et al., 2011; Winkler-Moser et al., 2023). Composition of fatty-acids is similar to that in corn and tightly clustered, but the composition of antioxidants is largely variable and inconsistent between plants. Given the high inter-facility variability in antioxidant composition, total tocols and carotenoids display the highest variation and vary together with oxidative stability, which demonstrates that retention of certain compounds coincides with the process (Winkler-Moser et al., 2023). In addition, fermentation and following treatment yield a characteristic “α-T*” tocomonoenol-like peak, indicating processing-derived chromanols that may contribute to oil stability (Moreau et al., 2011). Oils which are linked to solid streams, like wet grains and DDGS, are composed of sterols and some tocols/carotenoids, although they come with a lowered accessibility for a direct recovery in the oil phase. To both enhance the composition of lipophiles in the recoverable liquid fraction and decrease coinciding delivery to DDGS, the process can be configured to send more oil to thin stillage (e.g., additions of xylanase, protease, or phytase) (Luangthongkam et al., 2015; Moreau et al., 2010).

2.3.2 Phenolic acids and antioxidant potential

In the dry-grind corn ethanol process, the solids stream contains a high amount of bound phenolic acids, total phenolic acids averaged around 5.99 mg/g in DDGS versus 1.75 mg/g in corn (about 3.4x), dominated by ferulic and p-coumaric acids accounting for roughly ∼80% of total phenolic acids (Luthria et al., 2012). Additionally, DDGS has around 2.6x more ferric reducing antioxidant power (FRAP) antioxidant capacity by mass as compared to corn. Together, the increased phenolic content and higher FRAP capacity reported for DDGS indicate that dry grinding retains, but also further concentrates, phenolic acids and therefore their functionalities in the final solid streams (Luthria et al., 2012). Wet grains (e.g., pre-solubles and blend-back) offer a more optimal environment for releasing bound phenolics via enzymatic or mild-alkaline methods. Depending on how much CDS blend-back is incorporated as well as the degree of drying, the color, amount/behavior of reactive lysine, and extractability of DDGS is changed (Kingsly et al., 2010). In the liquid phase, phenolics appear at much lower concentrations (thin stillage and CDS) as compared to the solid streams. Despite lower concentrations in thin stillage and CDS, phenolic acids can still be detected analytically and therefore captured alongside peptides after the defatting and deashing steps whenever necessary, as discussed earlier (Kim et al., 2008; Sharma et al., 2023).

2.3.3 Proteins and bioactive peptides

After fermentation, protein builds up in the solids, but nitrogen from yeast propagates out into the liquid lane; amino acid levels go up by 2 to 3.5-fold as compared to corn kernels. The increase in amino acid levels means that wet grains and DDGS are good sources for peptides for proteolysis, while thin stillage and CDS carry soluble nitrogenous species that are rich in yeast signatures (Han and Liu, 2010; Liu and Han, 2011; Kim et al., 2008). Evaporating thin stillage to CDS increases protein density, but it also starts Maillard chemistry and lipid–protein interactions that make extraction increasingly difficult. Still, CDS has become a useful liquid-phase feedstock for protein concentrates and peptide fractions with antioxidant, ACE-inhibitory, and DPP-IV-inhibitory activities (Sharma et al., 2022; Sharma et al., 2023). Compared to CDS-rich DDGS, CDS-lean solids (or wet grains) have more intact protein and less sugar/glycerol. As a result, CDS-lean solids (or wet grains) produce more peptides and make downstream fractionation cleaner (Kingsly et al., 2010).

2.3.4 Dietary fiber and prebiotic oligosaccharides

Insoluble fiber (cell walls rich in arabinoxylan) accumulates in wet grains and DDGS. Enzymatic depolymerization (endo-xylanase ± feruloyl esterase) transforms arabinoxylan-rich insoluble fiber fraction in wet grains/DDGS into arabinoxylan-oligosaccharides (AXOS/XOS), exhibiting bifidogenic activity and significant short-chain fatty acid production in human fecal batch fermentations, particularly when DP ≥ 3 fractions are enriched to decelerate proximal fermentation (Monteagudo-Mera et al., 2018). This puts prebiotic fibers directly in the solids stream, along with peptide recovery from CDS and sterol/xanthophyll capture from DCO.

2.3.5 Residual solubles and organic acids

Thin stillage and CDS are the primary locations in which fermentation-derived solubles, including lactic and acetic acids, glycerol, and residual sugars/oligosaccharides, as well as emulsified oil, are stored (Kim et al., 2008; Di Lena et al., 2020). The acidic pH (≈4.0–4.7) and high water content (≈90–93% in thin stillage) stabilize certain phenolics and facilitate certain solvent partitions, but analogous matrices require deashing or buffering to mitigate their ionic-strength effects on extraction and membranes (Di Lena et al., 2020).

2.3.6 Mineral and phosphorus speciation

Minerals build up in the liquid stream and peak in syrup after evaporation. Sodium, sulfur, potassium, and phosphorus levels are significantly greater than corn baselines because of both concentration effects and chemical additions in plants. Fermentation simultaneously converts phosphorus from phytate to inorganic forms, thereby disrupting phytate-mediated chelation downstream (Liu and Han, 2011; Belyea et al., 2006). These mineral and phosphorus enrichment patterns impact how easy it is to access target nutraceuticals (like phenolic–metal interactions), how efficiently membranes work, and how much polishing is required before utilizing ingredients. Elevated ash and ionic strength in thin stillage/CDS reduce membrane selectivity and flux, increasing fouling and cleaning frequency, so clarification and deashing are positioned as enabling steps before UF/NF fractionation or polishing with solvents. Importantly, the scale of phosphorus accumulation in these streams is not marginal from a resource perspective. Ruffatto et al. (2022) employed plant-wide process models to estimate that the 196 U.S. corn ethanol biorefineries collectively hold a renewable phosphorus (rP) recovery potential of approximately 229,000 tons P per year, which accounted for over 12% of national P fertilizer consumption, with individual dry grind plants able to yield 162 to 4,501 tons P per year depending on grind rate. Corn biorefineries are heavily co-located with agricultural P consumption in the Corn Belt, so recovering phosphorus upstream could simultaneously improve the selectivity of downstream separation in the liquid lane and supply a circular-economy fertilizer coproduct (Ruffatto et al., 2022).

2.3.7 Process variability and blend-back as distribution levers

DDGS composition corresponds to CDS blend-back much more than corn composition does. Higher CDS levels render the color darker, the fat, ash (especially P/Na/S), sugars, and glycerol levels increase, and the protein/fiber levels decrease. Decreased CDS levels have the opposite effect and improve the profiles of limiting amino acids and minerals (Belyea et al., 2004; Kingsly et al., 2010). For lipophiles, the composition of DCO is generally stable (with stable fatty acids and sterols making up about 2% of the oil), but the distributions of antioxidant homologs and carotenoid loads vary, which is due to differences in how the oil is separated and handled at the back end (Winkler-Moser et al., 2023). Enzymatic process aids that break down cell walls and protein barriers can move both oil and lipophiles that are already mixed with the solids to the liquid stream, which renders it easier to extract out of thin stillage (Luangthongkam et al., 2015). CDS blend-back rate, oil recovery configuration/handling, and enzyme process aids help to understand why nutraceutical “hotspots” can be reproduced in the same order (lipophiles in DCO > DG/DDGS oil; phenolics in DG/DDGS ≫ liquids; peptides in CDS/thin stillage ≥ DG) but have different sizes that need to be verified between plants.

2.4 Implications for a biorefinery map

In the cross-stream picture, opportunities for recovery present themselves in three practical lanes that align with existing plant infrastructure: oil refining for lipophilic actives, liquid conditioning and membrane-based fractionation for aqueous solubles, and solids processing for bound phenolics, fiber-derived oligosaccharides, and protein fractions (Belyea et al., 2006; Kingsly et al., 2010; Winkler-Moser et al., 2023). Figure 2 shows recovery/purification options for each lane while emphasizing feasibility, scalability, and compliance.

3 Recovery and purification methods: feasibility, scalability, and compliance

Efficient recovery and purification of nutraceuticals from dry-grind corn ethanol process streams requires a balance of technical feasibility, operational practicality, and regulatory compliance. This section examines the available strategies for isolating key bioactive compounds from liquid, oil, and solid coproduct lanes, highlighting the design principles, process-specific challenges, and quality assurance measures that underpin scalable and food-grade implementation in modern biorefineries. When determining an adequate recovery strategy for a dry-grind biorefinery, it is important that the strategy is stream specific, food-grade when possible, and modular when possible so it can be implemented into preexisting unit operations with ease. A practical implementation strategy is a three-lane recovery framework (oil, liquid, solids) which designed to retrofit downstream of fermentation without penalizing ethanol throughput. The three pathways are liquid as thin stillage to CDS, oil as distillers corn oil and DCO, and solids as wet grains to DDGS. The three-lane recovery framework is strengthened by utilizing conditioning processes in tandem like clarification, deashing, and selection of solvent and quality assurance guidelines that aim to maintain tight yields even with inter-facility variations (Belyea et al., 2006; Kingsly et al., 2010; Winkler-Moser et al., 2023). Figure 3 summarizes the proposed three-lane recovery framework and the main enabling conditioning steps that make each lane operable at plant scale. It is intended as a modular concept map, linking major matrix constraints (for example, ash in liquids, oxidation sensitivity in oils, and release-limited solids) to practical unit operations already used in dry-grind facilities. It is worth noting that an emerging complementary paradigm involves the in situ biosynthesis of high-value compounds during the fermentation stage itself, utilizing genetically engineered yeast strains. Kurambhatti et al. (2021) demonstrated that an engineered Saccharomyces cerevisiae strain was able to produce an oligosaccharide found in human milk, 2′-fucosyllactose (2′-FL), during corn dry grind fermentation. This enriched DDGS with up to 19.8 g 2′-FL/kg dry DDGS without reducing ethanol yield, and process modeling demonstrated favorable economics at moderate DDGS selling prices. Such approaches in the fermentation stage can complement or extend the three-lane framework via introducing an upstream dimension in which enrichment of bioactives occurs prior to the coproduct recovery trains described.

3.1 Design principles and decision criteria

Three criteria guide the train design:

• Value density (compound concentration per unit mass of stream)
• Recoverability (selectivity under feasible pH/ionic strength and viscosity)
• Fit within current infrastructure (ability to place modules upstream of energy-intensive evaporation/drying and to use edible-oil unit operations already present in industry) (Rausch and Belyea, 2006; Belyea et al., 2006).

Given that minerals and phosphorus build up in the liquid stream and change how extraction and membranes work (Liu and Han, 2011), front-end deashing is seen as an enabling technology instead of a luxury. Similarly, QA proxies like ash/P/Na/K for liquids, color/reactive lysine for DDGS, and OSI plus lutein/δ-tocopherol for oils are used to organize lots (Kingsly et al., 2010; Winkler-Moser et al., 2023).

3.2 Lane logic (liquid, oil, and solids)

To translate the three-lane recovery framework, Table 2 summarizes lane-specific recovery targets across the liquid (thin stillage and CDS), oil (DCO), and solids (wet grains or DDGS) matrices, alongside conditioning steps that make each lane operable at scale. As recovery constraints are matrix-specific (for example, mineral and phosphorus buildup in the liquid stream can reduce membrane selectivity and increase fouling, oils require oxidation control during edible-oil style processing, and solids targets are often release-limited and sensitive to blend-back and drying history), each row pairs a candidate product with the minimum enabling conditioning logic needed to address the bottleneck while maintaining fit within existing plant infrastructure and minimizing added duty. Table 2 is intended as a screening-level design aid: it distinguishes enabling versus optional steps, highlights practical feasibility notes (for example, where upstream conditioning can reduce evaporator or dryer duty), and supports cross-lane comparison before detailed discussion, specification planning, and techno-economic evaluation.

TABLE 2. Lane specific recovery targets, conditioning strategies, and feasibility considerations in dry grind ethanol coproduct streams.

3.3 Cross-cutting feasibility: energy, variability, and regulatory posture

Even when a recovery concept is technically feasible, application depends on whether the proposition (i) improves or does not impact plant energy and water intensity, (ii) remains operable under inter-facility variabilities in stream composition, and (iii) yields fractions which meet regulatory requirements and specification for the intended market. Prior work provides guidance on energy leverage points (especially upstream conditioning that reduces evaporator and dryer duty), while specification windows, hazards trends, and documentation requirements for market entry are less consistently reported.

3.3.1 Energy and water

Membranes that capture water and salts before they evaporate reduce the amount of steam and protect peptides and phenolics from damage caused by heat (Rausch and Belyea, 2006; Reis et al., 2017). Aqueous centrifugation already removes some of the “gums” from oils, making edible oil trains more straightforward to use (Moreau et al., 2010).

3.3.2 Variability management

Composition is based on procedures: minerals and ash reach their highest levels in syrup; DDGS properties follow CDS blend-back; and antioxidant levels vary greatly between DCO lots (Belyea et al., 2006; Kingsly et al., 2010; Winkler-Moser et al., 2023). Adding simple gates (ash/P/Na/K, color L/a, OSI + carotenoids, reactive Lys) prevents drift and keeps SOPs for each plant in place.

3.3.3 Regulatory and safety

Using food-grade solvents (like ethanol and water), edible-oil processes, and well-characterized enzymes keeps outputs in line with food/feed pathways (Rausch and Belyea, 2006; Sharma et al., 2023). Not only are minerals and phosphorus monitored for technology choices, but they are also monitored for compliance with downstream specifications (Liu and Han, 2011; Belyea et al., 2006).

3.4 Scale-up and cost

At a plant scale, viable recovery modules include either the reuse of existing vessels/utilities or the reduction of duty of the high energy-intensive steps. Existing techno-economic analysis (TEA) of integrated first- and second-generation ethanol plants and thin stillage valorization demonstrate that implementing fungal upgrading and membrane cascades raises capital investment but may lower overall consumption of energy by around 2%–3% while improving net present value when coproduct revenues are factored (Rajendran et al., 2016; Reis et al., 2017). Studies such as these link together clarifier plus MF/UF chains, ED/IX deashing, and peptide or AXOS lines as supplementary modules sized to feed rates of the evaporator and tonnage of DDGS, where membrane area, enzyme dosage, and any additional pumping dominate costs. Because membrane area, enzyme dosage, and additional pumping typically dominate costs, this review prioritizes recovery options that reduce steam per gallon of ethanol, lower back-end variability, or upgrade syrup blend-back into higher-value fractions, since these levers largely affect feasibility in existing techno-economic scenarios.

3.5 Regulatory pathways

When focusing on the ingredient level, nutraceuticals derived from coproducts will often enter applicable markets as either feed ingredients, self-determined or notified food ingredients under GRAS, or as dietary supplement ingredients requiring New Dietary Ingredient (NDI) notification in the United States, with similar functional food and novel food routes in the EU and other regions (Pawar and Grundel, 2017; Bailey, 2020; EFSA NDA Panel, 2010; EFSA NDA Panel, 2012). Looking at actives in the lipid phase, distillers corn oil currently has a GRAS notice as an edible corn oil, which provides a regulatory precedent that sterol and tocopherol concentrates derived from DCO can be viewed as a higher purity variation of an approved oil as compared to as an entirely new substance (FDA, 2020, GRAS Notice GRN 890). Other, broader regulatory reviews highlight that regulators expect a clear documentation of the identity, manufacturing process, ranges in composition, and safety/quality data, as well as that inconsistent characterization of nutraceuticals is a common pitfall for setbacks (Santini and Novellino, 2018; Komala et al., 2023). Within this regulatory framework (GRAS, NDI, and related pathways), the recovery trains emphasized purposefully rely on food-grade solvents, edible-oil processing steps, and enzymes with an existing history of safe use so that the primary questions regarding regulation shift from novel chemistries to exposure levels, specifications of final products, and proving claims.

3.6 Comparative method fit by stream and class

Taking together feasibility, scale, and regulatory approach defines several different tiers of opportunity. DCO lipophiles that are concentrated through winterization, neutralization plus distillation, and saponification with additional recovery of unsaponifiables can take up a “near-term” tier because they can adapt established edible-oil refining and then connect directly to sterol and tocopherol health precedents (Moreau et al., 2010; Winkler-Moser et al., 2023; EFSA NDA Panel, 2010; EFSA NDA Panel, 2012). The enzymatic depolymerization of DDGS and wet grains to produce AXOS and to release bound phenolics fit into a second tier. The enzyme tools and prebiotic evidence are strong, but a scale-up must then address solids handling, DDGS heterogeneity, and the need for stable linking between structure and function (Luthria et al., 2012; Monteagudo-Mera et al., 2018; Müller et al., 2020). Liquid-lane membranes and peptide lines create a third tier with a high throughput, which is sought-after when paired to evaporator-duty savings, chemical oxygen demand (COD) reduction, and the generation of more defined protein or peptide ingredients from thin stillage and CDS, as opposed to stand-alone nutraceutical trains (Reis et al., 2017; Sharma et al., 2022; Sharma et al., 2023).

4 Framework for a nutraceutical-integrated dry-grind biorefinery

This section compiles the mapping and lane options into an implementation and prioritization framework for retrofitting nutraceutical recovery without disrupting the ethanol throughput. A dry-grind corn ethanol plant can become a multiproduct biorefinery by implementing nutraceutical recovery. Instead of adding front-end fractionation, which would slow down the flow of ethanol, this multiproduct transition is accomplished by exploiting existing downstream concentration and separation procedures to create defined, higher-value ingredients from coproduct streams (Rausch and Belyea, 2006; Reis et al., 2017).

Also, conditioning methods like deashing and clarifying as essential given that mineral accumulation significantly effects downstream partitioning, fouling, and selectivity in thin stillage and CDS (Belyea et al., 2006; Liu and Han, 2011; Reis et al., 2017). By managing known sources of variation like CDS blend-back and oil quality standards, the goal benefits would include: positioning coproducts as a higher-value, tighter-specified ingredient, improving consistency and expanding use options for existing coproducts, and allowing for energy and environmental benefits when selective separation methods are moved upstream from evaporation and drying (specifically in thin stillage valorization scenarios where duty reduction and coproduct revenue can be evaluated together) (Kingsly et al., 2010; Rajendran et al., 2016; Reis et al., 2017). Finally, by concentrating on unit operations and constituent forms that, whenever feasible, already have a clear regulatory posture, the framework is intended to remain consistent with food and supplement pathways. It also identifies the areas at which the primary obstacles to commercialization are characterization, specifications, and claim substantiation (EFSA NDA Panel, 2010; EFSA NDA Panel, 2012; Pawar and Grundel, 2017; Bailey, 2020; Santini and Novellino, 2018; Komala et al., 2023).

4.1 Prioritization

Near-term deployment favors “low-lift, high-fit” modules:

• Liquid lane: Condition thin stillage/CDS to control ash and fouling, then fractionate solubles into peptide and carbohydrate/organic-acid parts (Reis et al., 2017; Sharma et al., 2022; 2023).
• Oil lane: Use concentration compatible with edible-oil to produce sterol/tocopherol (optionally carotenoid) concentrates from DCO (Moreau et al., 2010; Di Lena et al., 2020; Winkler-Moser et al., 2023).
• Solids lane: Target bound phenolics and oligosaccharides derived from fiber with controlled release and polishing (contaminant aware) (Luthria et al., 2012; Monteagudo-Mera et al., 2018; Kingsly et al., 2010).

Enzymatic aids can shift partitioning behavior (such as moving more oil-phase material into liquid streams), so facilities can treat enzyme schemes as a variable effecting lane-routing, reflected in QA targets and module sizing (Luangthongkam et al., 2015).

4.2 Metrics for success (KPIs)

To evaluate if the proposed framework works at a given facility, a KPI set should be used to track performance in areas such as recovery, operability, specification readiness, and plant alignment (sustainability and economics). These steps are aligned with the separation and valorization logic discussed earlier for thin stillage conditioning, edible-oil processing of DCO, and solids valorization. They also match metrics that are often used in thin stillage valorization and integrated techno-economic and environmental assessment work (Rajendran et al., 2016; Reis et al., 2017; Gerrior et al., 2022; Pérez-Almada et al., 2023).

4.2.1 Recovery

KPIs deemed most defensible are overall recovery plus normalized yield (mass basis, tied to original matrix), enrichment factor relative to feed, and purity with lane co-extraction markers that are lane-relevant (Moreau et al., 2010; Luthria et al., 2012; Reis et al., 2017). Co-extraction stand-ins can be selected based on challenges in the streams, like ash and phosphorus carryover in liquid-lane separations and sugar, glycerol, or ash carryover that complicate peptide and phenolic fractionation (Liu and Han, 2011; Kingsly et al., 2010; Sharma et al., 2023; Monteagudo-Mera et al., 2018; Sharma et al., 2022).Operationally, KPIs to factor include evaporator duty or steam consumption differences due to upstream conditioning, membrane operations determined by flux and pressure stability between membranes and the related cleaning burden and downtime attributed to fouling (Rajendran et al., 2016; Reis et al., 2017; Liu and Han, 2011).

4.2.2 Specification readiness

KPIs include marker panel pass rate sorted by matrix (for example, minerals panels for CDS; reactive lysine and color for DDGS; oxidative stability and notable oil-phase actives for DCO), variability of actives and important release specifications between batches (water activity, ash, compositional ranges, etc.), and stability markers applicable to thermal history and type of ingredient (Kingsly et al., 2010; Di Lena et al., 2020; Winkler-Moser et al., 2023; Komala et al., 2023; Sharma et al., 2023).

4.2.3 Plant alignment

Notable and trackable KPIs include, water and energy usage due to unit operation addition, cost of goods per kilogram of ingredient, and a plant-level economic metric (consider Net Present Value (NPV) or payback period) for staged development scenarios (Rajendran et al., 2016; Reis et al., 2017; Gerrior et al., 2022; Pérez-Almada et al., 2023).

4.3 Implementation pathways

The most practical way to install a nutraceutical-integrated dry-grind biorefinery is as a staged retrofit that keeps ethanol throughput high while increasing the value of coproducts. Prioritization should be driven by KPIs, using recovery performance, operability, specification readiness, and plant alignment metrics (energy and water, cost of goods, and plant-level financial indicator) to decide which modules move from screening to pilot to integration. Consistent with biorefinery integration and thin stillage valorization studies, modular add-ons work best when they use existing tanks, centrifuges, evaporators, and utilities, and focus on separations that lower the load on evaporation and drying downstream (Rausch and Belyea, 2006; Reis et al., 2017; Rajendran et al., 2016). Plants can implement effective methodologies in stages, depending on their current setup (oil recovery system, CDS blend-back, evaporator capacity) and the markets they want to reach. Table 3 translates the staged-retrofit concept into a phase roadmap, linking each phase to suitable modules and “gate-to-advance” criteria based on KPIs.

TABLE 3. A suggested staged retrofit divided into practical phases.

Abbreviations: HCSE, hydroxycinnamic acid steryl esters; LDL, low-density lipoprotein; SCFA, short-chain fatty acids; AREDS, Age-Related Eye Disease Study; AREDS2, Age-Related Eye Disease Study 2; AMD, age-related macular degeneration.

Together, these phases highlight that early successes come from modules with low integration burden and clear specification pathways, while later phases require tighter control of variability and separation stability.

4.4 Alignment with sustainability and economic goals

Most sustainability gains originate from lowering thermal load (decreased evaporation and drying duty) and improving mass routing (upstream conditioning and targeted separations); therefore, performance can be tracked via low-impact, plant-relevant indicators. These include energy intensity, water intensity, wastewater and evaporator loading, and (if applicable) life-cycle indicators (Reis et al., 2017; Rajendran et al., 2016; Liu and Han, 2011; Gerrior et al., 2022; Pérez-Almada et al., 2023).

• Energy intensity: steam use and electricity per liter of ethanol, and extra energy per kilogram of nutraceutical ingredient recovered (Reis et al., 2017; Rajendran et al., 2016).
• Water intensity: the amount of water used for every liter of ethanol and every kilogram of ingredient, especially during diafiltration and cleaning cycles (Reis et al., 2017).
• Changes in the COD load and inorganic load (ash, phosphorus) to wastewater and evaporators when conditioning is done upstream (Liu and Han, 2011; Reis et al., 2017).
• Life-cycle indicators (when relevant): greenhouse gas intensity, eutrophication potential, and fossil energy demand utilizing LCA conventions, preferably combined with TEA in a comprehensive evaluation (Gerrior et al., 2022; Pérez-Almada et al., 2023).

Integrated TEA and LCA methodologies are increasingly advocated in biorefinery decision-making as they make clear the trade-offs among energy conservation, additional processing requirements, and the environmental consequences of coproduct distribution and displacement effects (Pérez-Almada et al., 2023; Gerrior et al., 2022).

Comparing the performance of individual modules to plant-wide profitability indicators helps to gauge economic alignment of the recovery-module synthesis with plant-wide profitability. For staged deployment, the most important metrics for making decisions are:

• Incremental capital and operating expenses per unit of throughput for each module, such as per cubic meter per hour of clarified stillage or per ton of DCO processed.
• Cost of goods per kilogram of ingredient (including membranes, enzyme dosing, utilities, solvent recovery, and labor), and how sensitive it is to yield loss caused by changes in the process (Reis et al., 2017; Kingsly et al., 2010).
• Net present value, internal rate of return, and payback period for retrofits that combine energy impacts with coproduct revenues at the plant level (Rajendran et al., 2016; Gerrior et al., 2022).
• Revenue diversification: the percentage of gross margin that comes from nutraceutical and improved coproduct streams, as well as the ability to keep ethanol capacity and reliability (Rausch and Belyea, 2006; Reis et al., 2017).

These indicators help facilities determine if adding nutraceuticals makes their operations more efficient and their margins more stable, all while still being able to work with their current operations and regulatory stance. KPIs should be evaluated as measurements of differences relative to a plant baseline (no recovery modules), reported per functional unit (for example, per L ethanol and per kg ingredient), and paired with assumptions for yield, uptime, membrane area, and enzyme dose. With this, compare trade-offs consistently across candidate retrofits, including duty reduction benefits, added utilities and cleaning burden, and coproduct revenue.

5 Multi-Stream Baseline: linking composition, recovery, and market potential

To realize the full potential of nutraceutical recovery, it is essential to connect compositional profiles and process strategies with functional applications and market opportunities. This section synthesizes the functional and commercial value of coproduct-derived nutraceuticals, reviews regulatory and productization pathways, and assesses risks and evidence gaps to inform go-to-market strategies and future research directions.

5.1 Functional and commercial potential

The same processing methods that concentrate starch into ethanol also concentrate non starch bioactives into side streams with tractable unit operations; therefore, dry grind corn ethanol coproducts can reasonably be positioned as sources of nutraceuticals. Health functions that are most closely linked to the corn matrix include: lipid phase cardiometabolic actives (phytosterols and tocols) derived from distillers corn oil (DCO), carotenoids (importantly, lutein and zeaxanthin) enriched in oil fractions, cereal phenolics (such as ferulic acid) and cereal-fiber derived prebiotic oligosaccharides (AXOS) that are present in thin stillage/syrup as well as DDGS, and protein-derived bioactive peptides generated or freed during fermentation processing or the targeted hydrolysis of protein-rich fractions. The functional evidence base for each nutraceutical class, along with the most plausible product formats, value propositions, and market-entry constraints (food, feed, or supplement), is summarized in Table 4 to move recovery lanes into practical commercialization pathways.

TABLE 4. Summary of nutraceutical classes that are associated with dry-grind corn ethanol coproducts. Displays key streams, the primary recovery lanes, representative uses or value propositions, and supporting references. Stream assignments show trends in enrichment across the dry-grind process and are intended for prioritization uses rather than concentration claims specific to facilities.

5.1.1 Lipid-phase actives (phytosterols and tocols)

In comparison to conventional corn oil, DCO consistently exhibits high levels of unsaponifiables enriched in phytosterols and γ-tocopherol. Despite inter-facility variability, the consistent enrichment of phytosterols and γ-tocopherol in DCO unsaponifiables maintains a favorable value profile for nutraceutical concentration steps, such as winterization, saponification, and molecular distillation (Winkler-Moser et al., 2023). The European Food Safety Authority determined that 1.5–3.0 g/day of plant sterols/stanols lowers LDL-cholesterol by approximately 7%–12%, supporting approved EU health claims when dosage and labeling requirements are fulfilled (EFSA NDA Panel, 2010; EFSA NDA Panel, 2012). Plant sterols/stanols has a well-established clinical function. In addition to sterols, γ-tocopherol contributes to cardiometabolic and anti-inflammatory platforms since it has different antioxidant and anti-inflammatory properties from α-tocopherol in vitro, animal models, and limited human evidence (Jiang, 2014; Jiang et al., 2022). In addition to previous findings that corn process oils and even distillers dried grains (without solubles) (DDG) extracts retain sterols, ferulate-sterol esters, and tocols (Winkler et al., 2007; Winkler-Moser and Breyer, 2011; Moreau et al., 2011), these mechanistic strengths are consistent with the compositional signature of DCO reported across ethanol plants (Winkler-Moser et al., 2023).

5.1.2 Carotenoids (lutein/zeaxanthin)

A previously conducted Age-Related Eye Disease Study 2 (The Age-Related Eye Disease Study 2 Research Group, 2013) randomized trial is the foundation of the lutein/zeaxanthin eye-health narrative. It found that while incorporating lutein/zeaxanthin into (The Age-Related Eye Disease Study 2 Research Group, 2013) formulation did not provide overall additional protection against progression to advanced age-related macular degeneration (AMD), it was a suitable substitute for β-carotene (given β-carotene’s smoker risk signal) and benefited low-dietary intake subgroups (The Age-Related Eye Disease Study 2 Research Group, 2013). These AREDS2 findings support lutein and zeaxanthin’s commercialization as targeted-benefit actives and risk-management alternatives rather than all-purpose cures. Co-recovery with sterols and tocopherols in a combined “functional lipid concentrate” is rationalized process-wise by their co-enrichment in maize process oils (Moreau et al., 2011; Winkler-Moser and Breyer, 2011).

5.1.3 Phenolic acids and cereal-fiber derived prebiotics

Bound ferulic acid (FA), which is abundant in corn and grain streams, exhibits anti-inflammatory and antioxidant properties in both preclinical and mechanistic models as well as in new human data (Zhai et al., 2023; Asbaghi et al., 2020; Jacobo-Velázquez, 2025). In comparison to grain, dry-grind processing can raise the apparent content of phenolic acids in DDGS, maintaining the antioxidant potential for use in food and feed (Luthria et al., 2014). AXOS, a family of prebiotics backed by the International Scientific Association for Probiotics and Prebiotics (ISAPP) standard definition of “selectively utilized by host microorganisms conferring a health benefit,” is produced by the enzymatic or microbial depolymerization of arabinoxylan (Gibson et al., 2017). AXOS-driven bifidogenesis and stool/SCFA advantages are reported in preclinical and human investigations, along with some signs of metabolic regulation (Müller et al., 2020; Neyrinck et al., 2018; Grootaert et al., 2009). The in-vitro prebiotic efficacy and favorable structure-function features of AXOS isolated from wheat DDGS were noteworthy for ethanol coproducts, highlighting the translational potential from distillery streams (Monteagudo-Mera et al., 2018).

5.1.4 Bioactive peptides

Proteolysis during fermentation and the targeted hydrolysis of DDGS or thin-stillage proteins can produce antihypertensive (ACE-inhibitory), DPP-IV–inhibitory, and immunomodulatory peptides. Recent reviews consolidate evidence that food-derived peptides regulate inflammation and gut microbiota, while also highlighting the disparity between in vitro efficacy and clinically significant outcomes. This gap between in vitro bioactivity and clinically meaningful outcomes underscores the necessity for standardized production, bioavailability assessments, and human trials prior to asserting high-value health claims (Chen et al., 2025; comprehensive review of food-derived peptides; also refer to broader peptide–microbiome discussions in Journal of Agricultural and Food Chemistry (JAFC) 2025). These themes correspond with literature review in peptide studies derived from corn coproducts (e.g., Sharma et al., 2022/2023), which indicate encouraging enzyme-inhibition activities but necessitate further translational evidence.

5.2 Regulatory considerations shaping product claims and formats

In the EU, plant sterol/stanol LDL-lowering claims are allowed only under certain conditions of use. This makes it clear how DCO-derived sterol concentrates can be used in functional foods (EFSA NDA Panel, 2010; EFSA NDA Panel, 2012). Most nutraceuticals in the US will be treated as dietary supplements under Dietary Supplement Health and Education Act (DSHEA) (1994). Pre-market safety checks will mostly be done through New Dietary Ingredient (NDI) notifications for ingredients that were not sold as dietary ingredients before 15 October 1994. Authoritative reviews summarize these requirements and common mistakes (Pawar and Grundel, 2017; Bailey, 2020; Coates et al., 2024). Because many of the targets here are found in corn (sterols, tocols, lutein/zeaxanthin, AXOS, peptides), sponsors still must determine if the form (like concentrated steryl esters or enzymatically generated AXOS) and use level trigger NDI or GRAS pathways. They also need to ensure that any structure/function claims have been backed and properly labeled. Evidence may support substitution or subgroup benefit without a broad disease-risk reduction claim, as seen with lutein/zeaxanthin (The Age-Related Eye Disease Study 2 Research Group, 2013). This distinction affects marketing strategy and requires substantiation.

5.3 Commercial feasibility and go-to-market pathways

From a process perspective, the most efficient outcomes arise from lipid-fraction enrichment (sterol/tocopherol ± carotenoids) through supplementary steps to existing DCO lines, and fiber/phenolic valorization via membrane separations or enzymatic pretreatments of thin stillage/syrup, which align with evaporator duty reduction. Multi-plant compositional data for DCO mitigate raw material variability and facilitate the standardization of nutraceutical specifications (Winkler-Moser et al., 2023). Studies and reviews of process integration show that introducing membranes or fermentation before evaporation saves energy and creates new sources of income, even when the capital costs are slightly higher. If AXOS meets ISAPP standards and shows selective use and SCFA responses in human studies, it can be marketed as a prebiotic ingredient. This is an argument for making AXOS or phenolic-rich concentrates along with regular syrup (Reis et al., 2017; Rajendran et al., 2016). Corn-fiber/xylanase-phytase enzyme programs already in use to improve ethanol and oil partitioning may be employed to change carbohydrate profiles toward prebiotic fractions while also increasing DCO supply for sterol/tocopherol recovery (Luangthongkam et al., 2015).

For sterols and tocopherols, the most effective option is a standardized “corn functional lipid concentrate” that meets European Food Safety Authority’s (EFSA) requirements for EU LDL-lowering foods and U.S. dietary supplements that make structure/function claims about LDL-cholesterol support and antioxidant function. Using wheat-DDGS examples (Monteagudo-Mera et al., 2018) in conjunction with a bridging human study on corn-derived AXOS would speed up adoption. For phenolic-rich syrups or DDGS extracts, initial roles as antioxidant flavor-modulators or color-protectants (food-tech roles) may come before high-level health positioning while clinical proof develops. For peptides, initial markets may concentrate on food performance (e.g., bitterness/savory modulation, emulsification) and sports nutrition structure/function claims, whereas extended clinical programs aim at blood pressure and glycemic endpoints.

5.4 Risk and evidence gaps

Three constraints determine whether nutraceutical concepts produced through dry-grinding translate into practical and defensible products: (i) variability and specification control, (ii) evidence maturity, and (iii) regulatory and safety gating.

Variability and specification control present a scale-up risk, because dry-grind plants vary in corn quality, enzymatic processes, fermentation management, and the severity of downstream separation and drying, all of which impact the abundance of target analytes and co-extractives. Phytochemical profiles of DCO display high variation between plants, specifications for sterols, tocopherols, and carotenoids cannot be assumed to be interchangeable between plants without plant-level baselining and routine QA trending (Winkler-Moser et al., 2023; Müller et al., 2020). For the oil lane, elevated free fatty acids and variations in oxidative stability relative to standard corn germ oil can change requirements for refining and shelf-life behavior (Winkler-Moser and Breyer, 2011). For the liquid and solids lanes, drying and evaporation history may alter functional quality attributes (for example, reactive amino acids and Maillard progression). Separations are often constrained by ash and dissolved solids, reinforcing the need for operating windows at the plant level. Variability also includes hazards: contaminants in incoming corn can concentrate into coproducts, and mycotoxins have been documented in DDGS, which elevates the importance of regular screening for any DDGS-derived ingredient positioned for human use (Zhang et al., 2009; Liu, 2011).

The translational readiness and the amount of evidence required for product claims and defensible commercialization varies between compound classes. Many nutraceutical candidates are supported by mechanistic feasibility and in vitro bioactivity; however, these findings do not consistently forecast efficacy in humans, as bioaccessibility, gastrointestinal stability, transport, and food-matrix interactions can significantly diminish the active dose at the target site (Amigo and Hernández-Ledesma, 2020). This translation limitation is especially true for peptide-based ideas from CDS or thin stillage, with abundant in vitro activity literature but scarce human data, uncertain bioavailability, and possible bitterness or sensory barriers that become crucial while developing the product. Some classes, like plant sterols, have more mature human evidence and clearer claim conventions. On the other hand, new ideas, like specific “signatures” of DDGS phenolics or peptide fractions aimed at glycemic endpoints, often need more direct bridging evidence, such as standardized compositional markers, dose justification, and human-relevant endpoints.

Regulatory and safety concerns can take precedence over feasibility. The necessary pathway depends on the market’s intent. For example, food use typically hinges on GRAS positioning, while supplements may require considering NDI based on the history and composition of the ingredients. Feed markets may allow for more variability, but still need safety controls and compositional documentation. Rather than treating regulatory posture as an umbrella downstream step, the framework can be strengthened by explicitly linking each candidate “product definition” to the minimum documentation set (identity, purity, contaminants, stability, intended use level, and manufacturing controls). This emphasis on establishing product definition first aligns with this review’s point regarding lot-gating and QA, since regulatory defensibility depends on having consistent identity and hazard control across lots.

To ensure these risks function within the proposed framework, it is helpful to think of development as stage gated. Each candidate should define a minimum viable specification (identity markers plus allowable ranges), a hazards and contaminants panel that suits the intended market, and a short list of “translation-critical” experiments (for example, stability under realistic storage, simulated digestion or bioaccessibility where appropriate, and at least one human-relevant validation step when claims depend on systemic exposure). The comparative risk matrix below is intended to render these requirements transparent stream by stream, so that prioritization considers the number of items there are and ease of recovery, and how mature the evidence is and difficulties caused with regulations. Table 5 consolidates these risks into a lane-specific screening tool for prioritization.

TABLE 5. Qualitative decision matrix that summarizes, through potential product and lane, (i) scale-up and specification risks determined by variability in stream and matrix effects, (ii) evidence maturity, and (iii) gaps in safety and regulation that limit commercialization. Table 5 consolidates the listed risks into a lane-specific screening tool for prioritization.

5.5 Bottom line for a nutraceutical-integrated biorefinery

A sterol/tocopherol driven concentration from DCO has the strongest immediate functional to commercial match. This is because plant sterols and stanols have a mature human evidence base with defined effective intake ranges and established claim and labeling conventions for cholesterol-related positioning. In addition, a DCO polishing product can be treated as a well-defined ‘unsaponifiables-rich’ concentrate that is straightforward to qualify using routine compositional markers (sterols plus tocols) (EFSA NDA Panel, 2010; EFSA NDA Panel, 2012; Winkler-Moser et al., 2023; Winkler et al., 2007; Winkler-Moser and Breyer, 2011; Moreau et al., 2011). Following behind that is a prebiotic AXOS stream from thin stillage or syrup that is membrane and enzyme enabled, provided that ISAPP-documented selective utilization and human endpoints are reported. Peptide fractions and phenol-rich extracts have more potential, but they need to be validated gradually. Crucially, many of these actions improve fundamental plant economics, such as enhancing DDGS, decreasing evaporator load, or increasing oil partitioning, thus commercial viability can be supported based on both process and health value (Luangthongkam et al., 2015; Reis et al., 2017).

6 Future directions

Future work should emphasize actions that reduce inter-facility variability and shift candidate ingredients from compositional hotspots to specification-ready product definition. Five near-term directions that are most actionable are listed as follows:

• Form multi-plant datasets which link compositional markers to respective operating conditions and the KPI set in this framework. This allows variability to be separated into actionable versus assumed components (Kim et al., 2008; Kingsly et al., 2010; Winkler-Moser et al., 2023).
• Demonstrate that conditioning and separations can be integrated into the process without sacrificing performance through using staged scenarios and progressing only after stability targets are achieved (flux stability, cleaning frequency, specification variation, etc.). Additionally, explicitly track evaporator duty impacts and wastewater routing (Rajendran et al., 2016; Reis et al., 2017).
• Implement mineral and phosphorus profiles as its own enabling module. They drive selectivity and fouling in separations in the liquid-lane, so evaluate whether they are manageable as a coproduct opportunity (depending on plant configuration) (Belyea et al., 2006; Liu and Han, 2011; Reis et al., 2017). Analyze feasibility of phosphorus removal as a revenue-generating coproduct pathway rather than just a process-enabling step, as it is estimated that U.S. corn ethanol biorefineries can collectively generate nearly twice the renewable phosphorus of all wastewater recovery facilities, with Corn Belt states showing strong co-location between phosphorus supply and agricultural demand (Ruffatto et al., 2022). Alongside this, determine how enzyme profiles may move oil and lipophilic routing between lanes, as this impacts where recovery modules should be located and which lots are most suitable per lane (Luangthongkam et al., 2015).
• Match development to the market objective with stage-gated evidence requirements by class: AXOS concepts should align with established prebiotic criteria while incorporating endpoints relevant to human criteria (Gibson et al., 2017; Müller et al., 2020), while peptide and CDS-derived concepts should connect in vitro activities to translational viability through standardized compositions, dose justification, stability and sensory feasibility, and mechanisms relevant to digestion (Sharma et al., 2022; Sharma et al., 2023; Chen et al., 2025). Because regulatory posture differs by ingredient and jurisdiction, planning should be aligned with the appropriate food and supplement pathways and documentation (Bailey, 2020; Pawar and Grundel, 2017; EFSA NDA Panel, 2012; Coates et al., 2024). Finally, compare different retrofit scenarios on an equal standpoint via utilizing TEA and LCA to quantify tradeoffs with added unit operations, energy conservation, effects of coproduct displacement, and expected economic value (Ubando et al., 2020; Gerrior et al., 2022; Pérez-Almada et al., 2023; Reis et al., 2017; Rajendran et al., 2016).
• Evaluate fermentation-stage strategies for in situ biosynthesis of target nutraceuticals using engineered yeast strains. Previous work has shown that genetically modified S. cerevisiae can generate bioactive oligosaccharides such as 2′-fucosyllactose during standard dry grind fermentation without comprising ethanol output (Kurambhatti et al., 2021). Assessing how these approaches interact with downstream recovery trains, coproduct composition, and regulatory positioning would clarify whether fermentation-stage engineering can be integrated with the physiochemical recovery framework proposed here.

7 Conclusion

Dry-grind corn ethanol plants concentrate non-starch components into coproduct streams that are already processed at scale. The concentration lends itself to practical entry points for recovering and enhancing nutraceutical ingredients without front-end kernel fractionation (Rausch and Belyea, 2006; Kim et al., 2008). A consistent distribution is evident from the reported studies: DCO is the primary hotspot for lipophilic actives (phytosterols and conjugates, tocols, and carotenoids), solids retain large fractions of bound phenolic acids and fiber-derived substrates, and thin stillage to CDS carries soluble organics and peptide precursors while concentrating minerals and phosphorus that complicate selective fractionation (Moreau et al., 2010; Moreau et al., 2011; Luthria et al., 2012; Di Lena et al., 2020; Reis et al., 2017; Winkler-Moser et al., 2023).

An important finding is that commercialization can often be limited less by the presence of bioactives rather than specification control and operability across plants. In the oil lane, DCO phytochemicals can be enriched but vary considerably between facilities, pointing to the need for marker panels, stability controls, and defined specification ranges and not assumptions based on conventional corn oil (Moreau et al., 2010; Winkler-Moser et al., 2023). In the liquid lane, ash, ionic strength, and phosphorus profiles greatly influence selectivity, fouling, cleaning cycles, and integration effects relative to evaporation and wastewater handling, strengthening the need for clarification and deashing (Liu and Han, 2011; Reis et al., 2017). In the solids lane, blend-back and thermal history shifts composition, suggesting that defensible products contain clear, standardizable definitions such as explicit DP distributions for oligosaccharide concentrates or marker sets for extracts rich in phenolics (Kingsly et al., 2010; Luthria et al., 2012; Monteagudo-Mera et al., 2018). With these constraints, the most feasible path to implementation is staged, low-fit retrofits that piece into existing unit operations are evaluated using recovery and operability KPIs with plant-level economics. This staged retrofit approach can be supported, where possible, by integrated TEA-LCA comparisons (Rajendran et al., 2016; Reis et al., 2017; Gerrior et al., 2022; Pérez-Almada et al., 2023). Finally, both evidence maturity and regulatory defensibility remain dependent on class: sterols have mature human evidence and clear claim conventions in some jurisdictions, while peptide and complex soluble fractions require a stronger bridge from in vitro to human context, and all novel concentrates require clear identification, contaminant control, and documentation aligned to food and supplement pathways (EFSA NDA Panel, 2010; EFSA NDA Panel, 2012; Amigo and Hernández-Ledesma, 2020; Pawar and Grundel, 2017; Bailey, 2020; Komala et al., 2023).

Several technological trajectories are expected to reshape nutraceutical recovery from dry-grind facilities over the coming decade. Engineered fermentation strains capable of in situ bioactive synthesis during ethanol production (Kurambhatti et al., 2021), upstream phosphorus recovery modules doubling as enabling steps for liquid-lane separations (Ruffatto et al., 2022), multi-plant compositional databases that minimize inter-facility variability (Kim et al., 2008; Winkler-Moser et al., 2023), and maturing regulatory precedent for corn-derived nutraceutical concentrates (Komala et al., 2023) are collectively expected to rapidly progress the transition of dry-grind plants from solely fuel facilities toward multi-stream biorefineries producing specification-grade ingredients alongside ethanol, feed, and recovered nutrients.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. Claude Opus 4.6 was used during the preparation of this manuscript for formatting assistance and literature organization. The authors reviewed and edited all AI-assisted output and take full responsibility for the content of this publication.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Glossary

ACE angiotensin-converting enzyme AXOS arabinoxylan-oligosaccharides CDS condensed distillers solubles COD chemical oxygen demand DCO distillers corn oil DDGS distillers dried grains with solubles DP degree of polymerization DPP-IV dipeptidyl peptidase IV ED/IX electrodialysis/ion exchange EFSA European Food Safety Authority FAE feruloyl esterase FPE ferulate phytosteryl esters FRAP ferric reducing antioxidant power GC gas chromatography GRAS generally recognized as safe HCSE hydroxycinnamic acid steryl esters HPAEC-PAD high-performance anion-exchange chromatography with pulsed amperometric detection HPLC high-performance liquid chromatography ICP inductively coupled plasma ISAPP International Scientific Association for Probiotics and Prebiotics KPI key performance indicator LC/MS liquid chromatography/mass spectrometry LCA life cycle assessment LLE liquid-liquid extraction MF microfiltration MW molecular weight NDI New Dietary Ingredient NF nanofiltration NPV net present value OSI oxidative stability index RO reverse osmosis SCFA short-chain fatty acids TEA techno-economic analysis UF ultrafiltration XOS xylo-oligosaccharides

Footnotes

Edited by: Daehwan Kim, Hood College, United States

Reviewed by: Hu Li, Guizhou University, China

Stephan Barbe, Technical University of Cologne, Germany