Synergistic Action of Rhizopus oryzae Lipase and Fusarium oxysporum Cutinase in the Enzymatic Transesterification of a High‐Free Fatty Acid Oil

Abstract

High‐free fatty acid oil cannot be converted to biodiesel through conventional chemical processes and accumulates as a waste stream in biodiesel industries. In this work, a bienzymatic system comprising Rhizopus oryzae lipase (ROL) and Fusarium oxysporum cutinase (FOC) was developed for its transesterification. Preliminary experiments, followed by response surface methodology, identified the optimal conditions of ROL/FOC ratio of 3.9:1 (mg/mg), a methanol‐to‐oil molar ratio of 2.6:1 applied through three stepwise additions, and a temperature of 31.8 °C, yielding a maximum conversion of 69.6% after 48 h. Under these conditions, FOC alone led only to ~4% conversion, while the combined ROL/FOC system outperformed ROL alone by approximately 18% after 72 h, demonstrating a potential synergistic effect between the two enzymes. To enhance process sustainability, FOC was immobilized on four carriers, with macroporous divinylbenzene and amino C6 methacrylate to perform the best, enabling the reuse over three consecutive cycles with minimal activity loss and achieving conversions up to 91% after 72 h in the second use. High efficiency was maintained upon scale‐up to 100 mL, demonstrating process robustness. The synergistic action of ROL and FOC provides an effective biocatalytic route for the valorization of low‐value, free fatty acid‐rich lipid feedstocks into biodiesel.

A bienzymatic system of Rhizopus oryzae lipase and Fusarium oxysporum cutinase was optimized for the transesterification of high‐free fatty acid waste oil into biodiesel. Synergistic enzyme action, immobilization‐enabled reusability, and successful 100 mL scale‐up highlight the industrial potential of the suggested biocatalytic approach.

1. Introduction

Biodiesel is a renewable and environmentally friendly alternative to conventional diesel, offering lower greenhouse gas emissions, biodegradability, and reduced toxicity. It is classified according to its origin: first‐generation biodiesel comes from edible oils, raising concerns on food availability and costs, while second‐generation is produced from nonedible feedstocks, such as waste oils [1, 2]. One such feedstock is high‐free fatty acid (FFA) oil, a byproduct of the biodiesel industry that challenges chemical transesterification [3]. In these processes, conventional acid or base catalysis is widely used, but suffers from soap formation, difficult catalyst recovery, and harsh conditions in the reaction [4]. Enzymatic transesterification, typically catalyzed from hydrolases, offers a sustainable alternative avoiding these drawbacks, operating under mild, solvent‐free conditions and converting efficiently FFAs and triglycerides (TGs) into fatty acid methyl esters (FAMEs) [5, 6].

Lipases (EC 3.1.1.3) and cutinases (EC 3.1.1.74) catalyze ester bond hydrolysis or synthesis, with broad industrial use in biodiesel, food, detergent, and pharmaceuticals, due to their activity under mild conditions [7]. Lipases, due to their affinity for hydrophobic surfaces and their substrate specificity, are effective in lipid transformations [8]. Lipases, due to the geometry of their active site, are typically 1,3‐selective, while there are also some nonselective lipases. The selectivity, however limits the overall yield in transesterification reactions. Cutinases, on the other side, have accessible active sites, enabling the conversion of polymers, such as cutin, and TGs.

Among lipases, a 1,3‐specific lipase, Rhizopus oryzae lipase (ROL), has drawn significant attention for biodiesel production. ROL has been used both in immobilized [9, 10] and free form [11] for FAMEs production. From cutinases, Fusarium oxysporum cutinase (FOC) has been widely studied, showing robust catalytic activity, broad specificity, and potential for immobilization. It has been applied in aqueous and organic media for transesterifications using model esters and alcohols [12, 13, 14], however, never so far in second‐generation biodiesel. Other cutinases also show interesting activity with TGs: Fusarium solani pisi cutinase achieved up to 90% conversion with 1‐butanol, and 75% with methanol [15], with engineered variant improving methanol tolerance and, thus productivity. Methanol inactivation of the enzymes is a common drawback in production of FAMEs [5, 16, 17], so important that Aspergillus nidulans cutinase could only produce 2% FAMEs under optimal conditions [18]. Apart from the alcohol‐induced enzyme inactivation, the enzymatic biodiesel production still faces challenges, including high enzyme cost and slow kinetics [19]. Advances in immobilization, protein engineering, and process optimization aim to improve enzyme stability, reusability, and cost‐efficiency [20].

Combination of lipases and cutinases in biodiesel production could enhance overall conversion, offering an innovative strategy for the valorization of high‐FFA feedstocks. The rationale for combining ROL with FOC lies in their complementary catalytic features. ROL exhibits pronounced 1,3‐regioselectivity toward TGs, which can limit its efficiency toward partially hydrolyzed or structurally hindered intermediates. In contrast, FOC displays broader substrate accessibility, enabling effective turnover of a wider range of esterified and FFA species under the reaction conditions. This complementarity suggests a synergistic catalytic system capable of improving overall conversion in complex, high‐FFA feedstocks. Synergistic action between these enzymes has been reported in other biocatalytic processes, such as the enhanced hydrolytic performance of a chimeric Thermomyces lanuginosus lipase, Thielavia terrestris cutinase enzyme [21], or the improved PET depolymerization by Candida antarctica lipase B and Humicola insolens cutinase [22]. The combined action of a lipase and a cutinase in biodiesel synthesis is underexplored. This study investigates the synergistic behavior of R. oryzae lipase (ROL) and F. oxysporum cutinase (FOC) in the solvent‐free transesterification of high‐FFA oil. By evaluating their cooperative effects, we optimize enzymatic biodiesel production and advance the development of efficient and sustainable processes.

Toward this direction, enzyme immobilization enhances stability, reusability, and activity of biocatalysts in industrial biocatalysis. Various supports, such as diatomaceous earth [23, 24, 25], methacrylic resins with octadecyl and amino functional groups, and macroporous divinylbenzene resins, have been employed for lipases and cutinases [26, 27]. Hydrophobic supports mimic oil–water interface and favors the active conformation of lipases [28]. However, enzyme immobilization can also present challenges, such as enzyme leaching, potential diffusional restrictions, and structural rigidity that may reduce catalytic efficiency [29]. Thus, optimizing immobilization techniques to enhance hydrophobic interactions while minimizing enzyme deactivation is essential for maximizing enzymatic efficiency in biodiesel production.

Herein, we describe the optimization of a bienzymatic process with ROL and FOC in solvent‐free transesterification of high‐FFA oil via response surface methodology (RSM). The process was evaluated at 100 mL scale, to assess its performance under larger‐scale conditions. Toward a more sustainable design, the immobilization of FOC on various carriers was also investigated.

2. Results and Discussion

2.1. Preliminary Experiments for Definition of Single‐Parameter Range

Preliminary experiments showed that the ROL/FOC combination outperformed ROL alone in solvent‐free transesterification of acid oil (Figure S1). Reaction conditions were systematically studied through single‐parameter screening to narrow the factor ranges for subsequent RSM. As this prescreening was not performed in full replicates, it served solely to guide the RSM design; the RSM results confirmed the observed trends.

As shown in Figure 1a, 30 °C was identified as the best‐performing temperature under pH‐controlled conditions (pH 7.0), with 3 equivalents of methanol‐to‐oil molar ratio, 32% (w/w) water, and ROL/FOC ratio of 5:1 (mg/mg). For all preliminary experiments, methanol was added in three stepwise additions, following the same addition strategy throughout the study. Notably, enzymatic activity ceased after the first 24 h under these conditions. This finding is consistent with the literature, as fungal cutinases, such as the ones from Fusarium species, typically exhibit optimal activity at 30–40 °C [30], and enzymatic transesterification of FFA‐rich substrates generally performs best at 30–50 °C [31].

FIGURE 1.

Effect of individual variables on FAME conversion in solvent‐free transesterification reactions at different incubation times (24, 48, and 72 h). (a) Temperature, (b) enzyme ratio ROL/FOC (mg/mg), (c) pH adjustment (7.0) using NaOH or not, and (d) methanol‐to‐oil molar ratio. If not otherwise stated, the reaction is performed at 30 °C, with pH adjustments, with 32% water (w/w) relative to the oil, using 3:1 methanol‐to‐oil ratio and ROL/FOC ratio 5:1.

Ratios of ROL to FOC ranging from 5:1 to 1:3 (mg/mg) were tested. The most favorable ratio was 0.5 mg ROL to 0.1 mg FOC (pure protein), corresponding to approximately 35 U of ROL (70.0 ± 1.1 U/mg) and 37.7 U of FOC (377.7 ± 28.8 U/mg), or 0.052 and 0.056 U/mg of oil, respectively (Figure 1b). Further increases in FOC loading did not improve, and in some cases reduced conversion, suggesting that ROL is the primary catalyst and that the two enzymes interact synergistically in a ratio‐dependent manner. A comparable conclusion was reported by Moschona and coworkers [32], who found a 5:1 ROL/Biolipasa‐IN ratio optimal for acid oil transesterification.

Given the acidic aqueous phase resulting from the high‐FFA content of the oil, pH adjustment prior to reaction was considered. Unexpectedly, reactions without pH adjustment achieved higher conversions over 72 h and remained active beyond 24 h, whereas NaOH‐adjusted reactions stalled after the first 24 h (Figure 1c). This is likely attributed to soap formation: alkali addition causes FFAs to react to form sodium soap, that binds the enzyme and inhibits downstream biodiesel processing. pH adjustment was therefore excluded from further optimization. Despite saponification of FFAs, which is happening during pH adjustment, it may significantly alter the reaction medium, the rapid loss of enzymatic activity observed under these conditions after 24 h is likely not solely due to soap formation. Additional effects, including changes in phase behavior and reduced interfacial area, as well as possible enzyme destabilization under locally alkaline conditions, may also contribute to the observed deactivation.

Methanol addition was optimized without pH adjustment for the aforementioned reasons. A 3:1 molar ratio achieved the higher conversion (70%) at 72 h, outperforming the 4:1 ratio (64%), which also caused the reaction to stall after 48 h (Figure 1d). Importantly, the reduced performance observed at higher methanol loading is attributed to methanol‐induced enzyme inhibition, which becomes more pronounced above the optimal operating range. The 3:1 ratio provides sufficient methanol to drive transesterification while avoiding enzyme inhibition, a well‐documented balance in the literature for both lipases and cutinases. A study on the enzymatic transesterification of Jatropha oil using immobilized Burkholderia cepacia lipase reported that the highest biodiesel yield was achieved at a methanol‐to‐oil molar ratio of approximately 3.3:1. Increasing the molar ratio beyond this point led to a decrease in production rate, due to methanol‐induced enzyme inhibition [33]. In addition, enzyme inhibition and deactivation effects due to methanol were significant factors affecting the reaction kinetics of a cutinase‐catalyzed transesterification in a reversed micellar system [15].

2.2. Optimized Transesterification Parameters by Response Surface Methodology

To assess the interactive effects of the three independent variables, contour surface plots were generated (Figure 2). Figure 2a illustrates the interaction between enzyme ratio and methanol‐to‐oil molar ratio; Figure 2b depicts the combined effects of temperature and methanol‐to‐oil molar ratio; and Figure 2c shows the interplay between enzyme ratio and temperature.

FIGURE 2.

Contour plots of FAMEs content (%) as a function of: (a) enzymes ratio (mg/mg) ROL/FOC and methanol‐to‐oil molar ratio, (b) temperature and methanol‐to‐oil molar ratio, and (c) enzymes ratio (mg/mg) ROL/FOC and temperature.

The prescreening successfully defined the parameter ranges for the RSM design, with optimal values consistently falling near the center of each contour plot. Figure 2a indicates an optimal enzyme ratio of approximately 3.9:1 (ROL/FOC, mg/mg) and a methanol‐to‐oil molar ratio of about 3:1. Excessive enzyme loading reduces biodiesel yield. This observation is consistent with reports of aggregation‐induced loss of catalytic efficiency at high enzyme concentrations [34], while methanol beyond the optimum concentration inhibited enzyme activity, such as in the study of Maceiras and coworkers on Jatropha oil transesterification [35]. Figure 2b identifies an optimal temperature of 30–32 °C at a methanol‐to‐oil ratio of about 3:1. These values agree with the known activity profile of fungal cutinases (30–40 °C) [30] and with the thermal instability reported for related enzymes at elevated temperatures. For instance, Sulaiman and coworkers [36] reported that LC‐cutinase lost activity rapidly at high temperatures, with half‐lives of 80 min at 60 °C and only 7 min at 80 °C. Figure 2c confirms the optimal conditions at 30–32 °C and an enzyme ratio of about 3.9:1 (ROL/FOC, mg/mg), consistent with the preceding plots.

RSM analysis yielded the following optimal conditions: methanol‐to‐oil molar ratio 2.6:1, ROL/FOC ratio 3.9:1 (mg/mg, pure protein), and temperature 31.8 °C, with a predicted maximum conversion of 69.6% after 48 h. These correspond to approximately 27.3 U of ROL and 37.7 U of FOC, or 0.04 and 0.056 U/mg oil, respectively. A quadratic model was identified as the best fit (p < 0.05; Table S1), expressed in coded values as:

$$\begin{array}{cc}\mathrm{Conversion}& =69.05-2.51\cdot \mathrm{A}+0.6049\cdot \mathrm{B}-0.2377\cdot \mathrm{C}\\ & -10.96\cdot \mathrm{AB}+1.06\cdot \mathrm{AC}+1.29\cdot \mathrm{BC}\\ & -9.32\cdot {\mathrm{A}}^2-6.00\cdot {\mathrm{B}}^2-8.80\cdot {\mathrm{C}}^2\end{array}$$

ANOVA confirmed the model’s statistical significance (F = 16.57, p = 2.91 × 10⁻⁸; Table S2). Among the main effects, only the methanol‐to‐oil ratio (A) was individually significant (p = 0.034), while temperature (B, p = 0.594) and enzyme ratio (C, p = 0.834) were not, though this does not preclude their influence on the process. The AB interaction (methanol‐to‐oil ratio × temperature) was highly significant (p = 9.86 × 10⁻⁸), indicating a strong combined effect. The AC and BC interactions were nonsignificant (p = 0.475 and 0.388, respectively). All three quadratic terms (A², B², C²) were significant (p < 0.05), confirming nonlinear responses and the existence of optima for each factor.

The model exhibited good overall fit: R ² = 0.861, adjusted R ² = 0.809, and predicted R ² = 0.704, with a coefficient of variation (CV) of 11.78% and adequate precision of 12.32. These metrics suggest a reliable and consistent model with a good signal‐to‐noise ratio. The significant lack‐of‐fit (p = 0.004) indicates that the model does not fully account for all variability in the system, suggesting the possible interference of additional unmodeled factors. Nevertheless, the model adequately captures the main trends within the studied design space and remains useful for process optimization under the investigated conditions. The pure error term (353.02) is smaller than the model sum of squares, indicating that most of the variation is explained by the model.

2.3. Model Validation and Interpretation

Validation at the RSM‐optimized conditions yielded an average conversion of 66.0 ± 1.1%, consistent with the predicted value of 69.6% and within the 95% confidence interval. To assess the individual enzyme contributions, reactions were conducted with each enzyme separately under the same conditions. ROL alone achieved 49% conversion at 48 h and 56% at 72 h, whereas FOC alone yielded only 4% conversion. Based on the sum of the individual enzyme contributions, an additive conversion of approximately 60% would be expected after 72 h. However, the combined ROL/FOC system reached 74% conversion, exceeding the expected additive outcome by 14 percentage points and supporting a synergistic interaction between the two enzymes.

The poor performance of FOC alone suggested substrate inhibition by either FFAs or methanol. To investigate this potential effect, reactions were conducted in n‐hexane with increasing acid oil concentrations (100–1000 mM FFA), while maintaining a constant methanol‐to‐oil ratio, so that methanol concentration increased proportionally. As shown in Figure 3, conversion declined progressively with increasing FFA concentration: 27% at 100 mM FFA, 18% at 250 mM, 9% at 500 mM, 6% at 750 mM, and 4% at 1000 mM, while FAME concentration reaching a plateau at ~45 mM, indicative of inhibition. Since ROL maintained high conversion under identical FFA concentrations, FAMEs are not acting as inhibitor. Furthermore, FOC also showed very low conversion with sunflower oil, a substrate that consists only of TGs, suggesting that methanol is the primary inhibitor rather than the FFAs.

FIGURE 3.

Effect of varying FFAs concentrations in the acid oil on FAMEs.

These findings support the following mechanistic interpretation, that ROL initiates transesterification and consumes a portion of the methanol, thereby alleviating FOC inhibition. FOC in turn esterifies FFAs to FAMEs, reducing system acidity and enabling continued reaction progress. This is consistent with the lower performance of ROL alone in acid oil relative to sunflower oil (Figure S1). Additionally, FOC can catalyze transesterification at the sn‐2 position, thus contributing to higher overall conversions, an effect also observed in sunflower oil transesterification, where FOC addition improved conversion by 22%. The interplay between the two enzymatic activities thus yields conversions, exceeding the sum of individual contributions.

To assess process scalability, the reaction volume was increased 50‐fold to 100 mL in a flask under vigorous agitation, using the RSM‐optimized conditions. Conversion reached 75% at 48 h and 81.3% at 72 h, approximately 10% higher than at small scale. This improvement may be associated with factors such as enhanced mixing, heat dissipation, and mass transfer at larger volume; however, these parameters were not directly evaluated in the present study. This result indicates that the process can be successfully translated from laboratory to larger scales without loss of performance, which is encouraging for its industrial applicability.

2.4. Immobilization of FOC

Enzyme immobilization was explored as a strategy for cost reduction through biocatalyst reuse. Immobilization of ROL was not investigated in the present work, as has been described in our previous work [10]. Two immobilization methods were employed for FOC: physical adsorption, in which the enzyme attaches to the carrier surface through a combination of hydrophobic interactions, electrostatic forces, and van der Waals interactions [37]; and covalent binding, which forms stronger chemical bonds between enzyme and carrier [38, 39]. Five carriers were evaluated: diatomaceous earth (a low‐cost, sustainable natural support), octadecyl methacrylate, macroporous divinylbenzene, and amino C2 and C6 methacrylate. Immobilization progress was monitored by Bradford assay (enzyme depletion from supernatant, Figure 4a) and pNPB photometric assay (relative activity of the free enzyme during immobilization, Figure 4b); the latter also was used to detect any enzyme deactivation during the process.

FIGURE 4.

Immobilization of FOC on different carriers and evaluation of enzyme stability during the immobilization process. (a) Immobilization yield (%) after 1, 3, 5, and 24 h of incubation with each carrier. (b) Relative activity (%) and (c) residual activity of the free enzyme during incubation in the immobilization medium. Values represent the mean of three independent experiments (n = 3).

FOC achieved complete or near‐complete immobilization (95%–‍100%) after 24 h incubation with all carriers, though the initial binding rates varied considerably (Figure 4a). The highest initial binding efficiency was demonstrated with diatomaceous Earth (approximately 85% within 1 h), suggesting strong affinity between the enzyme and its siliceous surface. Amino C6 methacrylate followed with ~65% immobilization at 1 h, likely facilitated by the greater flexibility conferred by the longer C6 spacer arm relative to its C2 counterpart. Notably, the covalent carriers did not bind faster than the physically adsorptive ones; diatomaceous earth outperformed all commercial carriers, including those employing covalent attachment. Overall, the results suggest that support chemistry and surface accessibility strongly influence immobilization kinetics, while spacer length may affect enzyme orientation and accessibility to the active site. These factors, together with carrier hydrophobicity and pore structure, appear to contribute to the differences in catalytic performance observed among the tested supports.

As expected, the relative activity in the supernatant decreased over time, inversely correlating with immobilization yield (Figure ‍4b). The residual activity of the free enzyme (Figure 4c) increased during the process, peaking between 1 and 3 h, which is consistent with literature reports. This may reflect conformational changes or favorable microenvironment effects arising from enzyme–support interactions, which temporarily can enhance catalytic activity prior to full immobilization.

Immobilization of FOC on the specific carriers studied here is not documented. Su and coworkers [40] reported efficient immobilization of cutinases on Lewatit VP OC 1600, with improved solvent tolerance and thermal stability. Nikolaivits et al. [13] prepared cross‐linked FOC aggregates, enhancing thermostability and ester synthesis efficiency.

In terms of specific activity, FOC performed best on macroporous divinylbenzene (0.89 ± 0.01 U/mg), amino C6 methacrylate (0.87  ± 0.00 U/mg), and diatomaceous earth (0.81 ±  0.00 U/mg), with lower values on amino C2 methacrylate (0.78 ± 0.01 U/mg) and octadecyl methacrylate (0.58 ± 0.01 U/mg). These differences reflect the influence of support surface chemistry, properties and porosity on enzyme‐carrier interactions, though no straightforward correlation with immobilization protocol was apparent.

Reusability was assessed over eight consecutive reaction cycles (Figure 5). Diatomaceous earth exhibited a steep activity decline, retaining only 25% of its initial activity by the 8^th cycle, likely due to enzyme leaching from weak physical interactions. Octadecyl methacrylate and macroporous divinylbenzene performed considerably better, maintaining 72% and ~70% residual activity, respectively, the latter’s porous structure likely providing a more protective microenvironment to FOC. Amino C2 methacrylate retained about 50% activity, with the short spacer arm possibly restricting enzyme flexibility and inducing conformational strain over repeated cycles. Amino C6 methacrylate outperformed all carriers, retaining over 94% residual activity after eight cycles, attributed to the longer spacer arm enabling favorable enzyme orientation and reduced steric hindrance. Based on their specific activity and reusability profiles, macroporous divinylbenzene and amino C6 methacrylate were selected for further transesterification studies. It should be noted that reusability trends observed in the photometric assay may not directly translate to transesterification performance, as the latter involves longer reaction times and exposure to methanol, oil, and reaction intermediates. Consequently, carrier performance is influenced not only by enzyme retention but also by stability under process‐relevant reaction conditions.

FIGURE 5.

Residual activity of FOC immobilized on different carriers over eight consecutive reaction cycles. Specific activity after each cycle was expressed as a percentage relative to the activity measured in the first cycle (set as 100%). Values represent the mean of three independent experiments (n = 3).

The reusability of these two carriers was then evaluated over three consecutive transesterification cycles (each lasting 72 h) under RSM‐optimized conditions, using immobilized FOC and free ROL. FOC immobilized on macroporous divinylbenzene showed only 4% activity loss after three cycles, reaching a maximum conversion of 81% (Figure 6a), suggesting effective enzyme retention within its porous structure. FOC on amino C6 methacrylate achieved a higher maximum conversion of 91% on cycle 2, but exhibited an 11% activity loss by cycle 3 (Figure ‍6b), possibly reflecting cumulative conformational changes under repeated reaction conditions, an effect not apparent in the shorter photometric assay. A minor contribution from residual bound intermediates or FAMEs after each cycle cannot be fully excluded, although extensive washing steps were applied between cycles to minimize any possible carryover effects. Both carriers demonstrated strong reusability, with macroporous divinylbenzene offering greater stability and amino C6 methacrylate delivering higher peak conversions.

FIGURE 6.

Reusability of immobilized FOC on (a) macroporous divinylbenzene and (b) amino C6 methacrylate in the transesterification of acid oil, performed with immobilized FOC and free ROL under RSM‐optimized conditions. FAMEs conversion (%) was measured after 24, 48, and 72 h over three consecutive reaction cycles.

3. Conclusion

This study demonstrates the synergistic potential of a bienzymatic ROL/FOC system for the conversion of high‐FFA waste oil to FAMEs under solvent‐free conditions. Systematic optimization via preliminary experiments to identify the range and RSM identified optimal conditions (a methanol‐to‐oil ratio of 2.6:1, ROL/FOC ratio 3.9:1, 31.8 °C), under which the combined system achieved approximately 25% higher conversion than ROL alone, confirming a genuine synergistic rather than additive effect. Mechanistic investigation revealed that methanol inhibition of FOC is a key factor governing the interaction between the two enzymes, with ROL playing a critical role in depleting methanol and enabling FOC activity. Immobilization of FOC on macroporous divinylbenzene and amino C6 methacrylate enabled effective biocatalyst reuse over multiple cycles, with amino C6 methacrylate showing a maximum conversion of 91% after the second cycle, demonstrating strong catalytic performance under the studied conditions. Successful scale‐up to 100 mL without loss of efficiency further supports the robustness and industrial relevance of the process. Overall, this work establishes the ROL/FOC combination as a promising biocatalytic platform for valorizing low‐cost, FFA‐rich lipid feedstocks into biodiesel, contributing to more sustainable and economically viable biorefinery strategies.

4. Experimental Section

4.1. Chemicals and Materials

Biolipasa‐R (ROL) was purchased from Biocon (Spain). FOC was produced in E. coli BL21 (DE3) as previously described [12]. p‐Nitrophenol (pNP, ‍99%) and p‐nitrophenyl butyrate (pNPB, ‍99%) were acquired from Alfa Aesar and Fluorochem, respectively. Diatomaceous earth, glutaraldehyde, n‐hexane (99.9%), and tetradecane (99+%) were obtained from Sigma‐Aldrich. Dimethyl sulfoxide (DMSO, 99.9%), potassium dihydrogen phosphate, and potassium hydrogen phosphate were supplied by Fisher Scientific. Lifetech ECR1090M (macroporous divinylbenzene), ECR8309M (amino C2 methacrylate), ECR8409M (amino C6 methacrylate), and ECR8806M (octadecyl methacrylate) were all products of Purolite. Methanol (99.8%) was purchased from Honeywell. The high‐FFA oil, a waste stream generated from biodiesel chemical synthesis and a significant challenge for waste management, was kindly provided by NEWENERGY S.A. (Greece); its composition is reported in Table S3. All other reagents and solvents were of analytical grade or higher.

4.2. Noncovalent Immobilization Assay

Prior to immobilization, the carriers (diatomaceous earth, octadecyl methacrylate, or macroporous divinylbenzene) were washed 2–4 times with potassium phosphate buffer (KPi; 50 mM, pH 7.5) and filtered to remove excess buffer. An appropriate amount of enzyme dissolved in KPi buffer was then mixed with the carrier to achieve a carrier‐to‐enzyme ratio of 20:1 (w/w) for the commercial carriers and 60:1 (w/v) for diatomaceous earth. The mixture was incubated for 24 h at 25 °C and 150 rpm. Samples of the supernatant were collected at defined time intervals for determination of protein concentration [41] and enzymatic activity (see respective paragraph). After incubation, the immobilized enzyme was recovered by centrifugation (6000 ×  g, ‍10 min, ‍4 °C) and washed four times with KPi buffer. The immobilized preparations were dried either in a vacuum desiccator at 4 °C or using a speed‐vac.

4.3. Covalent Immobilization Assay

The amino C2 and C6 methacrylate carriers were equilibrated by repeated washing (2–4 times) with KPi buffer (50 mM, pH 7.5). Activation was carried out by adding a freshly prepared glutaraldehyde solution (1%, v/v) at a resin‐to‐glutaraldehyde ratio of 1:4 (w/v) and incubating for 1 h at 25 °C and 150 rpm. After incubation, the supernatant was removed, and the carrier beads were washed four times with KPi buffer. The remaining steps followed the procedure described for noncovalent immobilization, using a carrier‐to‐enzyme ratio of 20:1 (w/w), with the addition of a final washing step with KPi buffer containing 0.5 M NaCl, followed by a wash with KPi buffer. Immobilization yield was determined using the Bradford assay [41].

4.4. Photometric Assay

Lipase activity was assessed by monitoring the hydrolysis of pNPB through the release of pNP, measured photometrically at 405 nm and 30 °C (Multiskan Sky Microplate photometer, Thermo Scientific). In a typical reaction, 20 μL of the 10 mM pNPB stock solution in DMSO was mixed with 160 μL of KPi buffer. After thorough mixing, 20 μL of enzyme solution in KPi buffer was added to initiate the reaction, and absorbance was recorded for 5 min. The specific activity of the immobilized biocatalysts was determined as follows, 0.1 mg of immobilized preparation was added to 1 mL of a 1 mM pNPB solution in KPi buffer containing 10% (v/v) DMSO to initiate the reaction. The reaction was conducted at 30 °C and 1,000 rpm for 9 min. Supernatant samples were collected at defined time intervals, and the released pNP was quantified photometrically at 405 nm. To assess the recyclability of the immobilized FOC, the same protocol was applied with the following modification: after each cycle, the immobilized enzyme was separated from the reaction mixture, washed four times with KPi buffer, and reintroduced into a fresh reaction mixture to initiate the subsequent cycle. One unit (U) of activity was defined as the amount of enzyme required to release 1 μmol of pNP per minute under the conditions specified above. All reactions were performed in triplicate.

4.5. Transesterification of Acid Oil

The study examined the individual and interactive effects of three selected variables on biodiesel conversion (FAME content, %): (i) enzyme ratio (mg/mg) ROL/FOC (5:1, 5:2, 5:3, 3:1, 3:2, 3:3, 1:1, 1:2, 1:3), (ii) temperature (25–40 °C) in 5 °C increments, and (iii) methanol‐to‐oil molar ratio (1:1, 2:1, 3:1, 4:1). RSM was applied using a three‐level, three‐factor face‐centered central composite design implemented in Design Expert 13.0 software (Stat‐Ease Inc., Minneapolis, MN, USA), requiring a total of 34 experimental runs (Table S4). The range of each factor was selected based on preliminary experiments conducted over a broader parameter space, which also evaluated the effect of pH adjustment to 7.0 via NaOH addition. The reaction setup at optimized conditions was as follows: 32% (w/w) water relative to oil weight was added to the high FFA oil, in which 0.1 mg of FOC was dissolved. A total of 0.670 g of oil (equivalent to 1.79 mmol of FFAs, FAMEs, TGs, and diglycerides) was combined with tetradecane (1.13 mmol) as internal standard and methanol (4.68 mmol). Methanol was introduced stepwise at 0, 6, and 24 h. The reaction was initiated by the addition of FOC (0.1 mg free enzyme or 1.0 mg immobilized on 20 mg carrier) and 0.39 mg ROL, corresponding to approximately 27.3 U of ROL, 37.7 U of free FOC, and 377 U/mg of immobilized FOC. These represent enzyme dosages of 0.04 U/mg of oil for ROL, 0.056 U/mg of oil for free FOC and 0.56 U/mg of immobilized FOC. Reactions were carried out at 31.8 °C for 72 h at 1000 rpm for both free and immobilized enzyme preparations. The same procedure was applied for scale‐up reactions in a 100 mL volume. All reactions were performed in triplicate.

For reactions conducted in n‐hexane, the high‐FFA oil was tested at concentrations ranging from 100 to 1000 mM, with methanol added at three equivalents relative to the available FFAs. Tetradecane (50 mM) was used as an internal standard, and predried n‐hexane served as the solvent. Reactions were performed with 0.1 mg FOC per mL of reaction mixture at 31.8 °C for 72 h under continuous stirring at 1000 rpm. All reactions were performed in triplicate.

At defined time points, 2 μL samples were withdrawn from the oil phase after centrifugation, diluted with 198 μL of n‐hexane, and analyzed. Samples were injected at a split ratio of 1:20 on a GC Nexis 2030 gas chromatograph equipped with a flame ionization detector, using a MEGA‐BIODIESEL 103 column (0.25 mm × 30 m, 0.25 μm) and helium as a carrier gas at 24 mL/‍min. The temperature program was: 120 °C for 3 min, ramp at 10 °C/min to 220 °C, and hold for 10 min. FAMEs quantification was performed using a methyl oleate calibration curve with tetradecane as internal standard, according to the equation y = 0.1403x, with blank samples included in all measurements. The corresponding blank values were systematically subtracted from the final results to correct for any background signal. FAME conversion (%) was calculated as the percentage of the theoretical maximum amount of FAMEs obtainable from the fatty acid equivalents present in the feedstock, based on its experimentally determined composition (Table S3). Therefore, 100% conversion refers to complete conversion of the available fatty acid equivalents into FAMEs and not to complete conversion of TGs alone.

Funding

This work was supported by the RESEARCH – CREATE – INNOVATE (Project number: T2EDK‐00573) and Special Account of Research Funds of the University of Crete (Project number: 11512).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supplementary Material

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.