Rice Bran Biorefinery: A Zero-Waste Approach to Bioactive Oil and Biopolymer Production

Abstract

Rice is a staple food for global nutrition, and its processing generates large volumes of waste with a consequent environmental impact. The industry needs to improve its capacity to manage and treat this waste with more sustainable options than traditional management methods, thereby mitigating the environmental impact of the rice industry. Among the waste streams generated, rice bran represents a significant fraction that is largely underutilized. This study proposes a comprehensive approach to rice bran recovery, aiming to transform 100% of the waste into bio-based products through a three-stage biorefinery approach that combines chemical and biological operations. The process began with the ethanolic extraction of rice bran, which yielded 20.58% (w·w⁻¹) rice bran oil. This oil, evaluated through both in vitro and in vivo trials, has demonstrated effectiveness when combined with commercial edible coatings, reducing post-harvest damage in grapes and lemons by 15–20%. Following extraction, the remaining defatted rice bran, accounting for 79.42% (w·w⁻¹) of the initial material, was used as a carbon-rich substrate for microbial fermentation by Haloferax mediterranei. This step converts 28.75% (w·w⁻¹) of rice bran into microbial biomass and 12.75% (w·w⁻¹) into polyhydroxybutyrate-valerate. The undigested residual biomass, comprising 37.95% (w·w⁻¹) of the starting material, was further valorized through the purification of high-value products such as cellulose (13.08% (w·w⁻¹)), hemicellulose (14.58% (w·w⁻¹)), and lignin (10.29% (w·w⁻¹)). Overall, the biorefinery model recovers 100% of the initial waste and demonstrates, under laboratory conditions, the model’s ability to transform rice bran into six products of industrial interest, offering an option with the potential to effectively manage rice bran waste and help circularize the production model of an industry that traditionally operates under a linear production model.

1. Introduction

The urgency to incorporate circular economy models into agricultural practices is becoming increasingly evident, particularly in the context of efficient resource use.

Biorefineries are being promoted as preferred management models for agricultural waste by international frameworks such as Agenda 2030 and organizations such as the United Nations and its specialized agency, the Food and Agriculture Organization of the United Nations, which urge in their guidelines the circularization of traditional agricultural production models to move towards zero-waste production processes [1,2].

From a technical perspective, the deployment of biorefineries remains limited, despite the increasing prevalence of circular-economy roadmaps. The commercial implementation of biomass-to-products is hindered by insufficient cost competitiveness, as bio-based outputs frequently fail to match the prices of their fossil-derived counterparts. High capital and operational expenditures, feedstock logistics, and scale-up risks have consistently been identified as significant barriers [3,4].

Regarding the rice industry, the second most important crop in the world in terms of production volume, approximately 30% of its yield consists of by-products such as rice husks, straw and bran [5], which are mostly discarded directly or used for low-value purposes, such as animal feed or energy recovery [6].

In this context, rice bran stands out as a high-value by-product because of its rich content of proteins, carbohydrates, and lipids, including bioactive compounds. Rice bran oil (RBO), a valuable oil extracted from rice bran (RB), can be obtained using cold pressing and solvent extraction methods without degradation [7,8,9]. RBO contains several compounds (γ-oryzanol, tocopherol, etc.) that are recognized for their antimicrobial, anticancer, and antioxidant properties [10,11]. Numerous studies have been conducted on the application of RBO in food formulations, cosmetics and nutritional supplements [7,12]. However, other industries, such as horticulture, are introducing greener strategies to improve postharvest protection. For example, raw cinnamon extracts have been used in vaporized form with some success [13], but there is very little evidence of the efficacy of lipophilic raw liquid extracts in post-harvest protection applications, although their in vitro properties against food pathogens have been reported frequently [14,15].

On the other hand, the core product of biorefineries continues to be bio-based polymers obtained through fermentation. Polymers such as polyhydroxybutyrate-valerate (PHBV) are environmentally friendly alternatives to traditional polymers (high-density polyethylene or polypropylene), which typically take more than 50 years to decompose and are used in the manufacture of single-use products or packaging [16], and petrochemical-based plastics, which have a very high environmental footprint [17,18]. However, the most significant barrier to the widespread adoption of bio-based plastics is their high production cost, which limits their economic feasibility [19,20]. Since the early 2000s, research in this field has increased steadily, focusing on cost-effective solutions for bioplastics [21]. The most successful strategies include reducing fermentation and culture media costs by using agricultural byproducts and minimizing purification expenses. For example, several studies have successfully used residues and byproducts in fermentation [20,22,23]. Huang et al. (2006) [24] demonstrated that culture media formulated from agricultural waste (corn and rice bran) reduce fermentation costs without excessively affecting PHBV accumulation and, when combined with extreme halophilic production strains, traditional toxicity and post-processing costs for cell lysis can be further reduced, as intracellular PHBV can be released by osmotic shock with tap water.

Finally, the common element that biorefineries must address is the recalcitrance of plant waste associated with lignocellulosic structures. Energy recovery processes have been successfully applied to manage lignocellulosic waste [25]; however, there are also well-established industrial processes for transforming lignin, hemicellulose and cellulose from various plant biomasses. Adapting these proven extraction methods to the final stages of biorefinery models offers a way to create economically viable and sustainable systems [26,27].

Summarizing, although there are many studies on the topic, there is a lack of scientific evidence reporting on rice biorefinery models that validate comprehensive zero-waste proposals, test the results of applying sequential valorization stages, identify the technologies and yields per product that can be obtained, and provide scientific evidence on (1) the economic performance of transforming waste into valuable and competitive products with respect to traditional counterparts and (2) the actual capacity of cascade strategies for the efficient management of waste effluents associated with globally prioritized crops that generate high volumes of waste during production.

This study aimed to develop a biorefinery model capable of transforming nearly 100% of the RB into high-value products. The assessed model integrates three main processes: extracting RBO to be used as an active ingredient in novelty applications such as fruit coatings to reduce postharvest damage; utilizing defatted rice bran to decrease the production costs of PHBV; and recovering cellulose, hemicellulose, and lignin from non-assimilated residues of defatted rice bran (Figure 1).

2. Materials and Methods

2.1. Rice Bran Material and Oil Extraction

The RB used as the starting material was obtained from a local supplier in Valencia, Spain, and stored in 1 kg vacuum-packed batches until its use. RBO was extracted according to the protocol described by Martillanes et al. (2018) [9], where the optimal extraction conditions were established as follows: 100% ethanol (EtOH) (PanReac AppliChem (Castellar del Vallès, Barcelona, Spain)) at 60 °C for 97 min.

2.2. Properties of the Rice Bran Oil

As the raw material used was the same as that used by Martillanes et al. (2018) [9] and did not have any extra value in determining the specific profile of bioactive compounds, this study validated the antioxidant activity of RBO following the method described by Turoli et al. (2004) [28] as a tool to verify the effect of storage conditions (20 °C, dark) over the preservation of RB throughout the years.

2.3. In Vitro Evaluation of the Rice Bran Oil

The pathogenic fungi Geotrichium candidum and Rhizopus stolonifer were obtained from the private fungal collection of the Biotechnology and Sustainability Department of CICYTEX and isolated from residual fruits and vegetables. Both microorganisms were cultured on Potato Dextrose Agar (PDA) plates (Merck KGaA, Darmstadt, Germany) for 5 days at 25 °C. The hyphal solution of G. candidum and spore solution of R. stolonifer required for the assays were obtained by resuspending the contents of the previously cultured plates in 8 mL of Potato Dextrose Broth (PDB) (Conda Laboratories, Madrid, Spain) supplemented with the antibiotic chloramphenicol (PanReac AppliChem, Castellar del Vallès, Spain) using a swab. After filtration, the concentration was adjusted to 10⁴ colony-forming units (CFU)·mL⁻¹ using a Neubauer chamber (Brand GmbH + Co KG, Wertheim, Germany).

The antifungal activity of RBO was determined using two methods. The first was the CFU count, in which 10 µL of the solution of G. candidum hyphae or R. stolonifer spores together with different volumes of RBO (from 10 to 90 µL to G. candidum and from 10 to 90 to R. stolonifer), 1% DMSO (PanReac AppliChem, Castellar del Vallès, Spain), and PDB with chloramphenicol were surface-seeded on Rose Bengal plates (Oxoid, Basingstoke, UK) and incubated for 3 days at 25 °C. The controls comprised: (i) formulations in which RBO was replaced with sterile distilled water, serving as the reference for uninhibited fungal growth; and (ii) a vehicle control lacking RBO but containing 1% DMSO to confirm that this DMSO level did not suppress fungal growth. The antifungal activity was determined based on the inhibitory capacity of RBO on fungal growth (summarized in Table S1, Supplementary Materials).

The second method involved a linear growth assay. Culture plates were prepared with PDA and a solution containing 1 g RBO, 0.5 mL DMSO, 4.5 mL glycerol (density 1.26 g·cm⁻³), and 9.5 mL distilled water. A similar mixture was used as a control, but water was substituted for RBO in the case of the white control and ketoconazole (0.1 mg·mL⁻¹) (Sigma-Aldrich, St. Louis, MO, USA) was used as the positive control group. Once the culture plates were prepared, a circular incision of 9 mm in diameter was made in the agar, and a section obtained from the plates containing G. candidum and R. stolonifer was placed on the agar. After three days of incubation at 25 °C, the inhibition ratio was determined using Equation (1), where D_o is the diameter of the fungal section, D_c is the diameter of the fungal colony in the blank control, and D_s is the diameter of the fungal colony on the plate with RBO or ketoconazole as a positive control.

Measures of in vitro fungal growth on image plates were performed using the software integrated into the automatic colony counter SCAN 500 (Interscience, Saint-Nom-la-Bretêche, France).

$$\mathrm{I}\mathrm{n}\mathrm{h}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{r}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\left(\mathrm{\%}\right)={\displaystyle \frac{\left({\mathrm{D}}_{\mathrm{c}}-{\mathrm{D}}_{\mathrm{o}}\right)-\left({\mathrm{D}}_{\mathrm{s}}-{\mathrm{D}}_{\mathrm{o}}\right)}{({\mathrm{D}}_{\mathrm{c}}-{\mathrm{D}}_{\mathrm{o}})}}\times 100$$

(1)

2.4. Assessment of Rice Bran Oil for Effective Post-Harvest Disease Control in Grapes and Lemons

The grapes (Vitis vinifera L.) and lemons (Citrus lemon L.) used were commercially ripe. The fruits were disinfected by immersion in 2% sodium hypochlorite (NaClO) (Sigma-Aldrich, Darmstadt, Germany) for 5 min and allowed to dry under a laminar flow hood before use.

After disinfection, the stalks were removed from the grapes, and the lemons were cut into three parts because of their size. The fruit was immersed in a solution of G. candidum or R. stolonifer spores at a concentration of 5 × 10³–10⁴ CFU·mL⁻¹. After drying for 10 min, the samples were immersed in a coating solution containing 1%, 3%, or 15% RBO; 4 g soy lecithin (Laguilhoat, Fuenlabrada, Spain) as a compatibilizer; and 100 mL of a solution of a commercial lipophilic carrier without inherent bioactive properties (NATURCOVER M:water = 1:6, v·v⁻¹) (Decco Ibérica Post Cosecha, S.A., Paterna, Spain) as a vehicle for RBO inclusion. A reference control was included consisting of samples in which water replaced RBO to assess the intrinsic (per se) effect of the commercial coating. The fruits were incubated for 5 days at 25 °C with 100% relative humidity, and the growth of fungi in the inoculated area was classified as (rotting or no rotting).

We calculated inhibition as the reduction in rotting compared to the control (water without RBO) using Equation (2):

$$\mathrm{I}\mathrm{n}\mathrm{h}\mathrm{i}\mathrm{b}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\left(\mathrm{\%}\right)=\left({\displaystyle \frac{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\left(\mathrm{\%}\mathrm{r}\mathrm{o}\mathrm{t}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g}\right)-\mathrm{T}\mathrm{r}\mathrm{e}\mathrm{a}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}\mathrm{t}\left(\mathrm{\%}\mathrm{r}\mathrm{o}\mathrm{t}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g}\right)}{\mathrm{C}\mathrm{o}\mathrm{n}\mathrm{t}\mathrm{r}\mathrm{o}\mathrm{l}\left(\mathrm{\%}\mathrm{r}\mathrm{o}\mathrm{t}\mathrm{t}\mathrm{i}\mathrm{n}\mathrm{g}\right)}}\right)\times 100$$

(2)

Owing to the small number of sampling points in each group analyzed for each data block (each fungus and fruit analyzed), linear interpolation was adjusted to allow the determination of the IC₅₀ instead of sigmoidal interpolation.

These experiments are exploratory due to the lack of technical replicates; nevertheless, the heterogeneity across independent units (stemming from samples collected from different lots and at varying ripening stages) is informative and therefore reported despite limited statistical support.

2.5. Microbial Production of PHBV by Haloferax mediterranei

Haloferax mediterranei was selected based on a study by Huang et al. (2006) [24] on the biological production of PHBV through the fermentation of RB and extruded corn starch residues. Initially, H. mediterranei was grown in 200 mL of standardized commercial medium to generate the pre-inoculum and incubated for seven days at 39 °C and 75 rpm in an orbital incubator (Optic Ivymen System, Selecta Group, Barcelona, Spain). The medium composition included glucose (1 g·L⁻¹) (Scharlab S.L., Barcelona, Spain), proteose/peptone (5 g·L⁻¹) (Scharlab S.L., Barcelona, Spain), yeast extract (10 g·L⁻¹) (Scharlab S.L., Barcelona, Spain), and a Subov salt solution (833 mL) composed of NaCl (234 g·L⁻¹), MgCl₂·6H₂O, MgSO₄·7H₂O, CaCl₂·2H₂O, KCl (6 g·L⁻¹), NaHCO₃ (0.2 g·L⁻¹), NaBr (0.7 g·L⁻¹), and 9 drops of FeCl₃·6H₂O (5% aqueous solution), all reagents from Thermo Fisher Scientific (Waltham, MA, USA). After seven days, the culture was initiated by inoculating a 5% v·v⁻¹ pre-inoculum (OD600 = 0.27) into a medium in which the conventional carbon source was replaced with d-RB, with the nitrogen-to-carbon ratio set to 0.73 to favor intracellular PHBV accumulation; batch fermentation was then run for 14 days in a glass bioreactor (Applikon Bio, Delft, The Netherlands), 7.5 L working with a volume of broth of 4 L, at 41 °C with continuous aeration via a sparger (18.6 L·h⁻¹; 12 kPa), automated monitoring, and pH control at 6.8 using 1 M NaOH and 1 M HCl. The detailed medium composition included: yeast extract 28.95 g·L⁻¹, d-RB 21.05 g·L⁻¹, and a generic salinized trace element solution as a base to meet the requirements of halophilic microorganisms, composed of NaCl 234 g·L⁻¹, MgCl₂·6H₂O 19.5 g·L⁻¹, MgSO₄·7H₂O 30 g·L⁻¹, CaCl₂ 1 g·L⁻¹, KCl 5 g·L⁻¹, NaHCO₃ 0.2 g·L⁻¹, and NaBr 0.5 g·L⁻¹.

2.6. PHBV Extraction and Purification

The PHBV extraction procedure was based on the method described by Rawte and Mav (2002) [29]. Cell lysis was performed using 2% (w·v⁻¹) NaClO, and the suspension was incubated in an orbital shaker at 37 °C and 75 rpm. The mixture was then centrifuged at 8000 rpm for 20 min at room temperature (RT). After removing the supernatant, the pellet was washed thrice with chloroform (CHCl₃) (Sigma-Aldrich, Darmstadt, Germany), and the organic chloroform fraction was retained. Finally, the chloroform was removed under vacuum using a rotary evaporator. The extraction yields were determined gravimetrically using Equation (3).

$$\mathrm{E}\mathrm{x}\mathrm{t}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \mathrm{y}\mathrm{i}\mathrm{e}\mathrm{l}\mathrm{d}\left(\mathrm{\%}\right)=\left({\displaystyle \frac{{\mathrm{W}}_{\mathrm{P}\mathrm{H}\mathrm{B}\mathrm{V}}}{{\mathrm{W}}_{\mathrm{D}\mathrm{c}}}}\right)\times 100$$

(3)

where W_Dc is the dry weight of the cells, and W_PHBV is the weight of the extracted PHBV.

2.7. Lignin, Hemicellulose, and Cellulose from Fermented Rice Bran

The extraction method used to obtain lignin, hemicellulose, and cellulose was described by Xu et al. (2006) [30]. The lignin fraction obtained was compared with that obtained using the alkali destructive method of lignin extraction described by Hernández-Coronado et al. (1997) [31].

2.8. Statistical Analysis

Statistical analyses were performed as follows. For the in vitro CFU counting method, model parameter evaluations included normality testing, polynomial fitting, and generation of the corresponding graphical representations, which displayed prediction intervals rather than standard deviations (SD) to simplify interpretation and provide practical estimates for predicting RBO concentrations for subsequent in vivo studies. Each data point denotes the mean of the experimental unit measured in triplicates. For RBO in vitro linear growth, data are reported as mean ± SD; after confirming normality, comparisons were performed using Student’s t-test (p < 0.05). For RBO in vivo assays, after confirming normality, the data were scaled to the 0–1 range and expressed as percentages. The IC₅₀ was calculated using linear interpolation instead of sigmoidal interpolation because of the small set of sampling points. Four and five independent experimental units were analyzed for lemon and grape samples, respectively. PHBV production and the extraction yields of cellulose, lignin, and hemicellulose are reported as mean ± SD of three independent experiments. All statistical analyses were performed using Origin 2023 software (OriginLab Corporation, Northampton, MA, USA).

The yields obtained per process are expressed in their respective results sections based on the initial biomass of the process and the product or products obtained at the end of the stage, that is, the conversion yield at that stage. In the graphical abstract and Section 4.4, the yields shown consider all stages of the biorefinery and its success in the percentage transformation of rice bran into each product, which is why they differ from the values expressed per stage.

3. Results

3.1. Properties of Bran Oil

The extraction yield was 20.58%, which is similar to that reported by Martillanes et al. (2018) [9]. Furthermore, the results showed a high total antioxidant activity of 13.32 ± 0.84 mmol Trolox·mL⁻¹. Antioxidant activity validation showed that the preservation of the raw material for 6 years in the dark, vacuum packing, and storage at room temperature preserved the bioactivity of RB (

Table 1. Overview of RBO extraction methods, antioxidant activity (ABTS), and source details in various studies.

Origin and Year of Harvest	Variety	Year of RBO Extraction	Extraction Process	Antioxidant Activity by ABTS Assay	SD	Research
Valencia (Spain), 2018	Seria (Senia-Bahia)	2018	Ethanolic extraction	1.47 (mg TEAC·g−1 DW of RB)	-	[9]
Valencia (Spain), 2018	Seria (Senia-Bahia)	2024	Ethanolic extraction	13.32 (mmol Trolox·mL−1 of RBO)	0.84	Our research
Chumphon (Thailand), 2014	Dok-kham	2014	Ethanolic extraction	34.94 (mg TEAC·g−1 DW of RB)	1.26	[32]
Phang-nga (Thailand), 2014	Dok-kha			8.36 (mg TEAC·g−1 DW of RB)	1.04
Satum (Thailand), 2014	Khem-ngen			7.23 (mg TEAC·g−1 DW of RB)	0.27
Chumphon (Thailand), 2014	Nang-dam			3.73 (mg TEAC·g−1 DW of RB)	0.19
China, 2019	Oryza sativa L. Varieties: Nanjing, Wuyoudao, Yanfeng, Suijing, Longjing	2021	N’Hexane extraction assisted by ultrasonic	1592.38–2106.47 (µmol TEAC·100 g−1 of RB)	NA	[33]

).

3.2. In Vitro Evaluation of the Antifungal Activity of Bran Oil

The results of the antifungal activity of RBO against R. stolonifer and G. candidum based on the CFU count and linear growth are shown in Figure 2 and Figure 3, respectively. Additional details are provided in Figures S1–S6 in the Supplementary Materials.

First, the effect of RBO on the growth of G. candidum and R. stolonifer was assessed by evaluating the CFU across a concentration gradient. The data shown in Figure 2 and Figure 3 were normally distributed. Additionally, no significant differences were observed between the water and DMSO controls, indicating that 1% DMSO did not inhibit fungal growth.

For G. candidum (Figure 2), the CFU count showed a slight but consistent decrease with increasing RBO concentrations. The fitted quadratic regression model, y = 539.01 − 0.061x − 0.023x² yielded an adjusted R² of 0.73107, suggesting a moderate correlation between RBO concentration and fungal growth inhibition. By setting the first derivative to zero to locate the maximum of the curve, we calculated the peak value at approximately 14.05 µL of RBO. This indicates a threshold beyond which further increases in RBO concentration result in a decrease in G. candidum CFU mL⁻¹. These findings suggest that RBO concentrations above this threshold may exert an inhibitory effect on G. candidum, potentially reducing its growth and viability. However, given the R² value, 10 and 20 µL were validated instead of 14.05 µL. The observed growth inhibition did not align with the model’s prediction, resulting in G. candidum growth at all studied values (Supplementary Materials, Figure S4C,D).

Similarly, for R. stolonifer (Figure 3), the CFU counts initially increased slightly before declining at higher RBO concentrations. The quadratic regression model y = 373.80 + 2. 297x − 0.034x² provided a higher adjusted R² of 0.84104, indicating a stronger fit than that of the first-order model. The calculated peak CFU, obtained by setting the first derivative to zero, was approximately 34.05 µL of RBO. This concentration appears to represent a threshold for R. stolonifer growth, beyond which RBO begins to inhibit its growth. Similarly to the G. candidum validation, we decided to validate 30 and 40 µL instead of the 34.05 µL suggested by the model because of R². In this case, the results were satisfactory, with a count of zero in both samples. This adjustment aligned the validation with the model (Supplementary Materials, Figure S4A,B).

Regarding the evaluation of linear growth (Figure 4), the antifungal efficacy of RBO at a final concentration of 3% was evaluated against R. stolonifer and G. candidum, and the inhibition percentages were compared with those obtained by substituting RBO with an antifungal agent (the positive control).

For R. stolonifer, the inhibition achieved by RBO was approximately 20%, whereas the positive control displayed an inhibition of nearly 100%, indicating complete suppression of fungal growth. Similarly, for G. candidum, RBO showed an inhibition of approximately 5%, whereas the positive control showed nearly 100% inhibition. Statistical analysis confirmed significant differences in the inhibition rates between RBO and the antifungal control for both fungal species.

3.3. In Vivo Evaluation of the Antifungal Activity of Bran Oil

The in vivo activity of RBO against R. stolonifer and G. candidum was evaluated using grapes and lemon samples. As shown in Figure 5, in the case of grapes, treatment with 1% RBO did not reduce the percentage of rotten grapes compared to the control, but higher concentrations of the oil reduced the growth of the fungus. With the 3% RBO treatment, 60% of the grapes remained non-rotted. The calculated IC₅₀ values indicated that approximately 2.67% RBO is required to inhibit 50% of R. stolonifer growth on grapes. In the case of lemons infected with R. stolonifer, none of the RBO treatments sufficiently reduced the number of lemons in a rotting condition, and the IC₅₀ was estimated to be above the tested concentration range (>15% RBO).

In fruits infected with G. candidum, RBO considerably decreased fungal growth. The percentage of fruit in a rotting condition ranged from 60% in the control treatment to 20% using 1% and 3% RBO, with an estimated IC₅₀ of less than 1% RBO for grapes. For lemons, the percentage of G. candidum rot decreased as the RBO concentration increased: 100% for the water control, 75% for the 1% RBO treatment, and 50% for the 3% RBO treatment, with an IC₅₀ of approximately 3% RBO. Additional details are provided in Figures S7–S10 in the Supplementary Materials.

3.4. PHB Production with Haloferax mediterranei

In the second phase of cascade valorization, the leftover solid waste from RBO extraction was used as a carbon source to produce PHBV via fermentation. After 14 days, d-RB was extensively digested with H. mediterranei deposited on the undigested RB at the bottom of the reactor. The pH adjustment data showed that to maintain the pH at SetPoint 6.8, 197.2 mL of 1 M NaOH was added over 14 days. Additional details are provided in Figures S11 and S12A in the Supplementary Materials.

3.5. PHB Extraction and Purification

Gravimetric quantification of the crude culture broth yielded 4.18 ± 0.678 g of wet biomass and 0.77 ± 0.085 g of dry biomass from 63.64 g of culture broth. As for PHBV production, 0.2366 g of PHBV was obtained, which represented 0.37% of the fermented raw broth, 5.66% in terms of wet biomass, and 30.73% in terms of dry biomass. The removal of chloroform generated a very thin and brittle PHB film (Figure S12B, Supplementary Materials).

3.6. Lignin, Hemicellulose and Cellulose Extraction Yield from Digested Rice Bran

The serial extraction method allowed the recovery of 1.6465 ± 0.015 g cellulose (16.46%), 1.8310 ± 0.1827 g hemicellulose (18.31%), and 1.297 ± 0.073 g lignin (12.97%) from 10 g of defatted and unfermented RB. In contrast, the alkali method (destructive) yielded 0.4993 g of lignin (9.80%) from 5.002 g of the sample.

4. Discussion

4.1. Extraction, In Vitro Validation, and Application of the Crude Rice Bran Extract in Edible Coatings

As shown in Table 1, the results indicate that proper storage of RB under vacuum at room temperature maintains its oil bioactivity over long periods, in contrast to previous studies that suggested treatments such as infrared radiation, dry heat, microwave, or low-temperature storage to prevent rapid oxidation [34,35,36]. This stability emphasizes the influence of factors such as variety, origin, harvest season, and extraction methods on RBO’s bioactive properties of RBO [37].

The antifungal activity of RBO exhibited different levels of inhibition depending on the fungal species and conditions, confirming its potential role as a natural alternative to protect fruits from postharvest damage. The in vitro assays highlighted effective concentrations for inhibiting fungal growth: for G. candidum, our model indicated an inhibitory effect at approximately 14.05 µL of RBO, where concentrations above this point significantly reduced CFU. In the case of R. stolonifer, a stronger correlation was observed, with a peak effect at 34.05 µL of RBO, indicating a threshold for inhibition. It should be noted that the doses used are intentionally higher than the endogenous levels of antimicrobial phytochemicals typically present in fresh fruit tissues and, as is customary in plant antimicrobial studies, exceed naturally occurring (and often sensorially acceptable) concentrations in foods to overcome matrix effects [38,39]. Furthermore, our bioassay with lemons and grapes used open/wounded fruit, which removes the main natural resistance barriers (cuticle and cell wall/waxes) and increases susceptibility to pathogen infections [40,41]. Therefore, the activity observed under these conditions is likely to be a conservative estimate and may underestimate the efficacy in intact, unwounded fruit.

Although the differences between the two fungi are significant, the antifungal activity of RBO against both R. stolonifer and G. candidum appears to result from direct contact with the functional compounds of the oil, possibly attributable to the high levels of oryzanol group [42]. Family of oryzanols may disrupt the electron transport chain in fungal cells, leading to a loss of proton motive force, reduced ATP synthase activity, and decreased cell viability [43,44]. Additionally, RBO polyphenols can interact with microbial cell membranes and cause structural alterations [45]. Our findings are consistent with previous theoretical explanations and the demonstrated antifungal activity of RBO against other fungi, such as Rhizoctonia solani, Pyricularia oryzae, Colletotrichum gloeosporioides, and Fusarium graminearum [42].

Figure S6 (Supplementary Materials) illustrates how an increase in concentration affects fungal morphology, with direct contact with RBO disrupting growth, particularly in G. candidum. These macroscopic changes align with the previously described actions of plant antimicrobials (membrane disruption, oxidative stress, and impaired β-glucan/chitin remodeling), leading to hyphal collapse and surface roughening [46]. This is also consistent with recent SEM studies reporting membrane and cell wall damage and deformed, shrunken hyphae after exposure to botanical extracts or essential oils [47], mirroring our macroscopic observations. In contrast, it was challenging to make similar observations for R. stolonifer because its aerial structures hide the agar surface. Furthermore, as shown in the images in Figure S5 (Supplementary Materials), challenges were encountered in achieving a homogeneous dispersion of RBO in the solid PDA medium used for fungal cultivation. The lipophilic nature of RBO complicates its distribution within a hydrophilic agar structure. Although the addition of DMSO and glycerol improved dispersion, it remained uneven, indicating that standard in vitro methods may need to be refined to achieve more accurate CFU counts when evaluating lipophilic compounds.

Overall, the results obtained using the proposed methodology allowed us to adjust the data using polynomial regression, which can guide the selection of the optimal concentration for in vivo trials. Our model suggests that a range of 3–5% RBO (oil·coating⁻¹) is likely to achieve significant reductions in fungal growth.

Our in vivo tests showed partial antifungal efficacy of a commercial cover with 3% RBO, inhibiting both R. stolonifer and G. candidum on grapes and lemons. The rough texture of the fruits likely facilitated the immobilization of the RBO coating, enhancing its antifungal properties and minimizing the repulsive forces between the hydrophilic and lipophilic compounds. As shown in Figure 5, in grapes, a coating with 3% RBO preserved 60% of the fruit in a non-rotting state, with IC₅₀ values of 2.67% for R. stolonifer and less than 1% for G. candidum. In contrast, lemons required higher concentrations of RBO, with an IC₅₀ of 3% for G. candidum and over 15% for R. stolonifer, indicating that the efficacy of RBO varies depending on the fruit type and fungal species (additional information is available in Figures S7–S10 of the Supplementary Materials).

Traditionally, fruit coatings are designed to effectively reduce water loss, gas exchange, and respiration, thereby extending fruit shelf life [48,49]. Our in vivo results suggest that the lipophilic matrix with RBO not only creates a uniform protective layer on the fruit but also adds an active function to traditional formulations, protecting against postharvest damage. Thus, our active coating showed results consistent with research on essential oils, such as those using alginate with grapefruit seed extract, which has demonstrated efficacy against postharvest fungi, such as Penicillium digitatum [15,48].

4.2. Biological Production of PHB(V) from Defatted Rice Bran

In the second valorization step, owing to the high production costs of bio-based plastics such as PHBV [19,20], our research incorporated their production into biorefinery models. This approach aims to offset losses with the benefits gained in the earlier stages while reducing input costs by using d-RB as a replacement for commercial carbon sources. Building on previous research [24,50], our study achieved a 47% yield of PHBV compared to the results of Huang et al. (2006) [24], which were based on a fed-batch process using an RB and cornstarch mixture (1:8 ratio). Differences in yield may be attributed to various factors, such as the use of different carbon sources [23] and the absence of specific stress induction in our process, such as N starvation [22]. Additionally, variations in the results might be explained by the potential degradation occurring between 120 and 336 h of culture, as described in previous studies on Cupriavidus necator [51].

For extraction and purification, our study also indicated that, consistent with Koller et al. (2015) [50], storing the culture under cold conditions and analyzing it after 12 days yielded results similar to those obtained immediately after the end of the culture period. This suggests that the process is effective for extracting and purifying PHBV owing to the low rate of degradation and simplicity of the extraction process, in which H. mediterranei is inactivated after fermentation [52]. The extraction process is simple and environmentally friendly, taking advantage of the extreme salinity conditions of fermentation. Dilution with tap water causes cell rupture by osmotic shock, facilitating the extraction of the desired compounds and eliminating the need for harsh chemicals or complex procedures, as previously reported [53]. However, to align the process with green chemistry trends, there is an urgent need to eliminate the use of halogenated solvents, such as chloroform, in favor of green solvents or extraction techniques.

Nevertheless, it remains unclear whether using d-RB as a carbon source leads to higher PHBV yields than using non-defatted sources does.

4.3. Lignin, Cellulose, and Hemicellulose from Defatted and Fermented Rice Bran

In the last step of our biorefinery model, by extracting lignin, cellulose, and hemicellulose, the proposed biorefinery model provides a solution for handling almost all initial RB.

Through serial extraction, 30% less cellulose was obtained than that in the characterization studies of Arun et al. (2020) [54] and Y. Liu et al. (2021) [37]. However, the result is positive, since the differences can be explained by losses across biological degradation during cultivation with H. mediterranei, a desirable effect indicating that our substrate-conditioning route improves bioaccessibility for the inoculated microorganism, consistent with reports of cellulose/hemicellulose depletion during microbial fermentation [55]. The same is true for holocellulose, which is based on sugars and is very likely to have reduced its volume after fermentation [55]. However, holocellulose should be interpreted more broadly, as lignin content, structure, and lignin–carbohydrate complexes strongly govern carbohydrate accessibility and hydrolysis, complicating cross-study comparisons [56]. As indicated by Casas et al. (2019) [57], depending on the hydrolytic process, an important part of both fractions can adhere to each other via covalent bonds; therefore, their quantification and purification are very complex, explaining the 32% difference between the evaluated lignin extraction processes in this study.

4.4. Biorefineries

Biorefineries and biofoundries are essential climate tools that are needed now more than ever before. The fight against climate change demands faster and smarter solutions, and biorefineries and biofoundries are stepping up to the challenge. These advanced facilities transform waste, such as agricultural byproducts, into sustainable alternatives. Consequently, investment in biorefineries and/or biofoundries is growing worldwide as their potential becomes undeniable [58,59]. These facilities are no longer niche research projects but are rapidly becoming essential tools in the global effort to achieve sustainability targets. The ultimate goal is to harness cutting-edge science and automation to develop real and scalable solutions that reduce greenhouse gas emissions, while creating economic value. Our research (summarized in Figure 6), like that of others [20], reimagines agroindustrial waste (RB) into valuable materials and products (RBO, PHBV, Biomass, cellulose, hemicellulose, and lignin). By integrating chemical processes with microbial fermentation, we can efficiently and sustainably address climate change at its root and generate economic profits.

To provide a rough estimate, if we scale our obtained yields to 1 ton of rice bran and assess each fraction based on current market values (the calculation considered 1 US$ ≈ €0.95), the potential gross revenue per product could be: rice bran oil (20.6% w·w⁻¹) €1100; PHB (12.8% w·w⁻¹) €490–730 [60]; microbial biomass (28.8% w·w⁻¹) €190–270 for feed-grade or €1000–1440 for premium SCP [61]; cellulose (13.1% w·w⁻¹) €75 [62]; hemicelluloses (14.6% w·w⁻¹) €70–80 if sold as glucose syrup or €2770–5540 as xylooligosaccharides (XOS) [62,63]; and lignin (10.29% w·w⁻¹) €30–70 [64]. Depending on the product combination, the total revenue per ton can vary significantly: in a conservative scenario (feed-grade biomass + hemicelluloses as syrup), it is €1950, whereas in an optimistic scenario (premium SCP + XOS), it can reach €8950 per ton. These figures are merely indicative as they exclude inputs such as steam, water, electricity, waste stabilization/handling, and logistics. Furthermore, as mentioned, the economic return on the products obtained is highly sensitive to the purity, quality and end use of the product. Nevertheless, they provide a useful framework for assessing the economic viability of the proposed biorefinery.

5. Conclusions

At the laboratory scale, we validated the central hypothesis that three-stage biorefinery coupling chemical and biological operations can valorize rice bran with near-zero waste and reintroduce the whole stream into the value chain as six products. Quantitatively, the cascade achieved an overall mass closure of ≈100% with step-resolved yields of 20.58% rice bran oil, 28.75% microbial biomass, 12.75% PHBV, 13.08% cellulose, 14.58% hemicellulose, and 10.29% lignin. This convergent accounting provides an end-to-end demonstration in a single framework, clarifying how the initial hypothesis is met and defining the specific advance of this work: a measured, unit-by-unit route from residue to multiple products of industrial relevance.

Conceptually, the research ambition is to achieve effective transfer to industry and promote economically viable circularization of the rice sector. Therefore, in the mid-short term, the real interest lies in analyzing the practical impact of the biorefinery model on an industrial scale. The technological perspective/limitation is covered, as valorization already includes stages that use mature technologies that have been tested in the industry, and their application and transferability do not entail any complications. The environmental and regulatory perspective on waste promoted by Europe has already established regulatory requirements for waste management in the industrial sector, specifically requiring the implementation of biorefineries as a priority solution for waste management, rather than traditional methods (incineration, landfilling, etc.). Therefore, the main limitation identified is economic viability, where the authors highlight two critical issues that must be resolved: (i) replicating cascade valorization in a pilot-scale infrastructure to confirm the operational robustness and performance of the products obtained at representative industrial scales and (ii) maximizing revenue potential by developing market-adapted applications for each line of compounds obtained. In the quest to maximize economic benefits, the exploratory results obtained on the applicability of RBO as a post-harvest treatment point to a real market option for RBO aimed at meeting a specific need demanded by society, rather than returning a product with no specific application focus. This increases the added value, competitiveness, and potential of this type of bio-based product to reach the market, improving the overall viability of biorefinery implementation.

The proposed biorefinery is a good candidate for implementation, as it combines environmental benefits with solid technical preparation and can be applied to a wide variety of waste products. However, their adoption will ultimately depend on capital costs and the value that can be obtained from these products. Nevertheless, the high initial cost of the necessary infrastructure means that the overall economy and acceptance by the industry will be the decisive factors.