Utilization of Madhuca indica ash as a green catalyst for transesterification of Madhuca indica oil: process optimization and engine analysis

1 Introduction

Dependence on fossil fuels has become a pressing environmental issue as their combustion emissions large quantities of carbon dioxide and other greenhouse gases, triggering climate change and global warming. The use of fossil fuels, which now meets over 80% of the entire world’s energy needs, produces over 70% of the greenhouse gas emissions induced by humans, causing enormous negative environmental impacts, such as land deterioration, loss of biodiversity, and water and air contamination (Ali Ijaz Malik et al., 2024). By 2030, economic growth, industrial expansion, and population growth are expected to lead to a significant increase in the world’s energy consumption. Growing oil consumption will help meet some of the rising demand, reflecting a greater dependence on fossil fuels. Increased pollution and greenhouse gas emissions are two significant environmental consequences of continuing to rely on fossil fuels. To mitigate ecological harm and secure safe energy sources, it is crucial to address these issues (Wan Osman et al., 2024).

Currently, biodiesel is receiving considerable attention as a viable alternative to conventional diesel derived from fossil fuels. Biodiesel consists primarily of fatty acid methyl esters (FAMEs) produced by the transesterification of triglycerides with short-chain alcohols (commonly methanol) (Islam et al., 2013). Recent optimization studies using microwave-assisted and conventional transesterification approaches are further contextualizing process intensification strategies for biodiesel synthesis (Wati et al., 2025). Biodiesel differs from petroleum-based diesel in that it has higher manufacturing costs and sourcing costs. According to the current screenplay, biodiesel is a renewable alternative to petroleum diesel, which is running out and will eventually run out (Madhu et al., 2017). Given the high value of vegetable wastes as food commodities, the feedstock vacuum is the primary source of concern with biodiesel products. Non-edible oils are investigated to avoid food vs. fuel conflicts; M. indica is a non-edible, locally abundant feedstock (Dhanuprabha et al., 2025). Consequently, non-food products like rubber seed, jatropha, and Karanja oil are being investigated as possible sources of fuel for the creation of biodiesel (Chowdhury et al., 2021).

M. indica is a deciduous tree native to South Asia and widely distributed across India. The species produces oil-rich seeds that have been used traditionally as a non-edible feedstock for biodiesel production, making it an attractive local resource for decentralized biofuel supply. In this study, we focus on M. indica seed oil because of its availability, low cost, and suitability for transesterification after appropriate pre-treatment. Reported seed oil compositions contain a mixture of saturated and unsaturated fatty acids that are amenable to conversion to FAMEs. Emphasizing M. indica as a non-food feedstock avoids direct competition with edible oils and supports the valorisation of an underutilized agricultural product. Therefore, assessing low-cost, waste-derived heterogeneous catalysts for M. indica oil transesterification are of practical and environmental interest (Patel et al., 2012; Gupta et al., 2012; Oloyede et al., 2023a).

The seeds of M. indica are generally appertained to as short-lived. Although no details on its storehouse physiology have been proven, the seeds are believed to be recalcitrant. Recalcitrant seeds need high humidity content to retain their viability (Nabora et al., 2019), and trees producing similar seeds are abundant in tropical timbers. Numerous economically important tropical species similar as Shorea robusta, Honoria speciosa, Theobroma cacao, Hevea brasiliensis, Persea americana, and Cocos nucifera, produce recalcitrant seeds, heterogeneous catalysts and biomass ash catalysts (Demirbas et al., 2016). Oil made from M. indica has been used as a non-food feedstock for biodiesel. The principal native habitats of the species are South Asia and the Indian Key (Puhan et al., 2005). The cost of the finished product is significantly influenced by the feedstock choice, and using comparable garbage that is readily available in big quantities could provide marketable and feasible amounts of biodiesel (Baskar et al., 2017).

Catalysts for biodiesel production are commonly classified as alkaline, acidic, or enzymatic, and as homogeneous (soluble) or heterogeneous (solid) catalysts. Homogeneous base catalysts (e.g., NaOH, KOH) typically provide high conversion under mild conditions but complicate downstream separation and wastewater treatment. Heterogeneous solid catalysts (metal oxides, mixed oxides, supported bases, and biomass-derived ashes) offer easier separation, potential reusability, and reduced effluent treatment, which can improve process economics at scale. However, heterogeneous catalysts may exhibit lower intrinsic activity, mass-transfer limitations, and deactivation mechanisms such as surface fouling and leaching of active species. Therefore, careful characterization (composition, basicity, phase identification) and reusability testing are essential. Biomass-derived ash catalysts are of particular interest as low-cost basic catalysts because they can be rich in alkali/alkaline earth oxides (K₂O, CaO, MgO) that provide basic sites for methoxide formation. Nevertheless, potential sodium or potassium leaching and compositional variability must be quantified (Semwal et al., 2011; Leung et al., 2010; Helwani et al., 2009; Demirbas, 2009a). Recent studies on peel and agricultural ash catalysts are providing comparative data on basicity, leaching behavior, and catalytic performance that are directly relevant to the present ash-derived catalyst approach (Rajak et al., 2024).

Homogeneous transesterification is a commercial method for producing biodiesel. The technology can improve yield and conversion, but the product is really expensive, which raises the question of why biodiesel has been around for so long (Babadi et al., 2022). Homogeneous catalysts increase the cost of the end product since it complicates the process of separation and purification and thus make the catalyst unusable (Dhawane et al., 2016). As a result, it became necessary to intensify the process by switching out homogenous catalysts for diverse catalysts. It is said that an acid-catalyzed process converts vegetable-based wastes less effectively into biodiesel than a ground-catalyzed transesterification process. One kind of gemstone that can be utilized as a non-toxic base catalyst wherever in the globe is limestone (Demirbas, 2009b). The best and most prevalent strategies for producing sizable amounts of biodiesel are transesterification (Sharma et al., 2008). In heterogeneous catalysis, the catalyst is in a different phase than the reactants. In common cases, the catalyst source is a solid that contacts gaseous or liquid reactants. Heterogeneous catalysis is very common in industrial processes due to its benefits (Okechukwu et al., 2022). The most notable benefit is simply the ease of separation of the catalyst from the reaction mixture, which makes it easier to recover, reuse, and thereby reduce costs and optimize efficiency. Heterogeneous catalysts also tend to be thermally stable and are often used under relatively harsh conditions, making them more suited to bulk manufacturing processes, and are also generally more robust to deactivation and poisoning when compared to their homogeneous counterparts (Siddiki et al., 2022).

The primary objectives of this study are to (Ali Ijaz Malik et al., 2024) evaluate a calcined and NaOH-treated M. indica shell ash catalyst for transesterification of M. indica oil (Wan Osman et al., 2024), optimize key process parameters (methanol:oil ratio, temperature, catalyst loading) using RSM-CCD (Islam et al., 2013), characterize catalyst composition and stability (XRF/XRD/FTIR and reusability tests), and (Wati et al., 2025) assess engine performance and emissions for B10 and B20 blends and perform a preliminary techno-economic analysis.

2 Materials and methodology

2.1 Materials

The M. indica seeds (Figure 1) were bought from a nearby farm and were deshelled and dried. For oil extraction, n-hexane (Loba chemicals Pvt Ltd., purity of 99%, India) is used. Methanol (Sigma-Aldrich Pvt Ltd., purity of 99.5%, India) is utilized for transesterification. In addition, chemicals such as NaOH (Merck Pvt Ltd., purity of 99.0%, Germany), ethanol (Sigma-Aldrich Pvt Ltd., purity of 99.5%, India) were used.

2.2 Preparation of catalyst

To synthesize a catalyst, the M. indica shell (Figures 2a,b) was calcinated in a furnace at 400 °C at a heating rate of 10 °C/min for 4h to burn most volatile materials and improve rigidity (Prajapati et al., 2023). Then, the biomass ash is treated with alkali solution (2N NaOH) in a ratio of 1:15 (w/v) at controlled conditions of 80 °C for 24 h. NaOH treatment dissolves amorphous silica as sodium silicate, thereby reducing SiO₂ content and enriching basic oxides (K₂O, CaO, MgO) in the residual ash. Then the residual ash is washed with distilled water to remove the trace Na from the catalyst. Then, it is dried in an oven at 60 °C for 24 h to remove the remaining water vapour and to be recognized as a catalyst pyrogenic biomaterial.

2.3 Extraction of oil from M. indica seed

M. indica biomass is dried in a controlled atmosphere to evaporate the moisture in the biomass. The 200g powdered M. indica seeds were packed in the thimble and placed inside the soxhlet apparatus. The extraction was performed using n-hexane as a solvent at 60 °C for 8h. The Soxhlet apparatus is filled with powdered seeds of M. indica and hexane. During this process, vaporised hexane is condensed when it reaches the condenser and drips down on the biomass, thereby extracting oil. The distillation unit was used to separate M. indica oil from an oil-hexane mixture by distillation with a yield of 34.2%.

2.4 Transesterification

M. indica oil was transesterified with methanol at a varying molar ratio. A catalyst derived from MISA was added at a concentration of 1%–4% (w/w), and the reaction was performed in a pressure tube at a temperature ranging from 60 °C–80 °C for 6 h under a constant stirring of 200 rpm. The resulting mixture was centrifuged at 2,100 g for 10 min to separate the catalyst. The biodiesel layer underwent further processing, including heating for excess methanol removal and washing with water to remove impurities (Rahul et al., 2024).

2.5 Optimization

Response Surface Methodology (RSM) was used to thoroughly assess and enhance the transesterification process variables, i.e., reaction temperature (60 °C–80 °C), methanol-to-oil molar ratio (15:1–30:1), and catalyst loading (1-4 wt%). Central Composite Design (CCD) was employed to understand not only the individual but also the combined effects of these parameters on the biodiesel yield, while also reducing the number of experimental runs required in the conventional one-variable-at-a-time method. The experimental design comprises 20 runs in total, out of which there were 8 factorial points, 6 axial points, and 6 centre points. This allowed the estimation of the experimental error and provided confidence in the model. The center points serve as a method to verify the reproducibility of the data and to detect any curvature in the response surface. Table 1 displays the coded and actual levels of the independent variables. A reaction time of 6 h was maintained throughout all experimental runs so that the effects of selected independent variables could be focused. The biodiesel yield (%) corresponded to the response variable (Singh and Singh, 2024).

TABLE 1. RSM-CCD for the transesterification process.

In order to represent the relationship between the response and the independent variables, a second-order polynomial model (quadratic model) was constructed. The model’s significance and adequacy were checked through analysis of variance (ANOVA). The model validation was done using statistical indicators such as the F-value, p-value (p < 0.05), coefficient of determination (R ²), adjusted R ², and lack-of-fit test. Furthermore, the importance of contributing individual linear, quadratic, and interaction terms was evaluated. The process parameters were optimized by employing the desirability function to increase the yield of biodiesel. Statistical analytics and regressional modeling were executed with Minitab (Version 16.2.2, Minitab Inc., USA) statistical software. The experimentally obtained results at predicted optimum conditions corroborated the accuracy and reliableness of the model developed.

2.6 Characterization

The Physicochemical properties of catalysts and biodiesel products were carried out using a variety of analytical tools, and the data of M. indica oil is mentioned in Table 2. X-ray fluorescence (XRF) helped in identifying the elemental compositions of raw and residual ash samples, giving insights about the metal oxides that are catalytically active. In order to recognize the functional groups in the catalyst and biodiesel, Fourier Transform Infrared (FTIR) spectroscopy was utilized. To reinforce the FAME generation through typical methyl ester signals in proton NMR, Nuclear Magnetic Resonance (NMR) spectroscopy was employed for further confirmation. To analyze the biodiesel element composition and purity levels, and also to FAME quantification, Gas Chromatography coupled with Mass Spectrometry (GC-MS) analysis was conducted.

TABLE 2. Physicochemical properties of M. indica oil.

The GC-MS analysis was conducted on a PerkinElmer Clarus 500 system with TurboMass software (version 5.2.0). Helium was used as a carrier gas at a flow of 1 mL/min. As a stationary phase, an Elite-5MS (30 m length, 0.25 mm ID, 0.25 µm film thickness) was used for the separation. The sample injector was set at 250 °C, and the gas flow conditions were 53.6 kPa system pressure, helium carrier gas, and a linear velocity of 36.3 cm/s. The mass spectrometer was operable using electron ionization mode at 70 eV with a scan range of 40, 600 amu and a purge flow of 3.0 mL/min. Finally, the optimized biodiesel samples were checked for their FAME composition and biodiesel yield via GC-MS analysis. The schematic representation of the entire process is provided in the Figure 3.

3 Performance and emission analysis

The performance and emission characteristics were tested using a Kirloskar model 4 S single-cylinder diesel engine at a constant speed of 1,500 rpm, rated power output of 3.75 kW, and a compression ratio of 17.5:1 (Figure 4). For uninterrupted air intake, a large drum with a volume 500 times the engine’s swept volume was used. The fuel supply was delivered through three-hole mechanical injectors, and the injection pressure (IP) and injection timing were set to 200 bar and 23° BTDC, respectively. The “IC engine soft” software was used to analyse the engine’s performance and combustion dynamics. Pressure data were obtained from the IP sensor, and supplementary sensors located on the engine head, with crank angle data accurately recorded using precision encoders. This approach ensured reliable measurement of combustion timing, enhancing the validity of the experimental findings. The experimental setup also featured an Eddy current dynamometer with an arm length of 185 mm, capable of handling up to 7.5 kW at a broad speed range of 1,500–6,000 rpm. This supported robust performance testing under varied operating conditions. Data acquisition was performed over ten cycles, and the “IC engine soft” program synthesised all incoming data to compute various performance and combustion parameters for each cycle, ultimately reporting average results. The AVL DI N five-gas analyzer measures NO and NO₂. NOx is reported as NO + NO₂ and is expressed (g/kWh or ppm). CO and HC are reported (g/kWh or ppm) and smoke opacity as % (Bosch scale). Analyzer calibration and detection limits are provided in Methods (Varalakksmi et al., 2025).

Uncertainty of this experimental setup

= u n c e r t a i n i t y o f l o a d 2 + p r e s s u r e 2 + s p e e d 2 + t e m p e r a t u r e 2 + f u e l 2

= u n c e r t a i n i t y o f 0.2 2 + 0.4 2 + 0.2 2 + 0.25 2 + 0.2 2

= ± 0.4455 %

The precision of the data may not be impacted by the experimental setup’s ±0.4455% uncertainty.

4 Techno-economic analysis

Techno-economic analysis (TEA) is an efficient research technique for evaluating the technical and financial feasibility of proposed process designs. The techno-economic and profitability assessments were organized as distinct processes that involved sensitivity analysis and cost computation (total capital investment, production costs, and profitability and income ratios) (Gholami et al., 2021). AspenPlus (Advanced System for Process Engineering), a well-known process simulation application, was developed by Aspen Technology. In chemical engineering, it is widely used to model and simulate a wide range of chemical processes, such as the production of chemicals, the refinement of petroleum, and other industrial operations (Introducing Aspen Plus, 2016). By simulating unit operations and chemical reactions under a variety of operating conditions, AspenPlus helps to create, improve, and assess the processes. Using Thermodynamic rules and precise mathematical models, the software predicts the behavior of processes. Process simulations and economic analysis tools using Aspen Plus may be used to economically evaluate the feasibility of different process configurations, operating conditions, and design alternatives during techno-economic study (Dougher et al., 2023). The case studies on hydro treating and green diesel production illustrate TEA best practices and assumptions that inform equipment specification and costing choices (Kopli et al., 2024).

5 Result and discussion

5.1 Characterization of catalyst

A multitude of catalyst samples were subjected to infrared examination to figure out type of metal-oxygen bonds present in the catalyst. The FTIR spectra of each type of sample match the stretching frequencies of all major functional groups (different bonding types) present in those samples (Sahani et al., 2018). Figure 5 represents the FTIR spectra of the produced heterogeneous catalyst. FTIR assignments (catalyst and biodiesel): peaks at 3,559–3,456 cm^-1 correspond to O–H stretching (adsorbed water/hydroxyl groups); peaks at 2,922 and 2,853 cm^-1 correspond to aliphatic C–H stretching; the peak at ∼1713 cm^-1 corresponds to ester C=O stretching (biodiesel); peaks at 1,277 and 1,201 cm^-1 correspond to C–O stretching of esters. Assignments follow standard references (e.g., Silverstein et al., 2005). Water particles or ions containing hydroxide might be the root cause of additional levels inside the catalyst composition (Oloyede et al., 2023b).

XRF analysis indicated that the ash sample contained 10.64% Al₂O₃, 17.69% CaO, 28.02% SiO₂, and 24.49% Na₂O. And naturally, biomass ash is silica-dominant, so processing has to be done to enhance its catalytic activity. Alkaline process was used to extract silica from this ash, and after analysis was found to consist of 36.08% K₂O, 18.61% CaO, 8.70% MgO, and a quite smaller concentration of 11.26% of silica (Table 3). Potassium is said to increase catalytic activity in the transesterification reaction and hence the abundance of potassium oxide in the leftover ash would serve as a great candidate as a catalyst in the formation of biodiesel. Therefore, the residual ash has a higher likelihood of serving as a catalyst for the production of biodiesel.

TABLE 3. XRF analysis of ash.

The X-ray diffraction pattern of ash and residual ash is shown in Figure 6. The crystalline structure of the sample appears to be noticeably altered by alkali treatment. The ash sample has some weaker peaks in the ash from (2θ) 10.89°, 24.02°, and 28.17°. The most intense peak corresponds to the amorphous structure and slightly crystalline material at 24.02°. XRD analysis (Figure 6) was indexed against ICDD reference patterns. Major peaks at 22.5°, 26.7°, and 34.0° are consistent with phases such as K₂CO₃/K₂O and CaO. The enrichment of potassium-containing phases is likely providing basic sites that facilitate methoxide formation and transesterification. However, potential leaching of the K species under reaction conditions is being discussed and quantified by leaching tests. The formation of new phases and crystallinity suggests that the residual ash can function as a heterogeneous catalyst, which would increase its efficiency in transesterification processes for the manufacture of biodiesel.

5.2 Optimization (response surface methodology)

Following the effective production from the heterogeneous MISA catalyst along with analysis of its exterior characteristics, the catalytic activity during the transesterification of M. indica seed oil in the biodiesel was evaluated. For the RSM investigation, twenty experimental studies were conducted out. With this context, the impact of numerous variables, including weight percentage of the catalyst, temperature, and methanol to crude oil ratio, was examined. ANOVA, as well as an algebraic polynomial theory, were applied to assess the reliability and applicability of the suggested model. The appropriate parameter is significantly greater when the test statistic P-value is less. Since the estimated F-value and probabilities statistic P-value are 736.92 and 0.000, respectively, the analysis of variance (ANOVA) indicates that the Pearson correlation coefficient was extremely significant (Table 4). The determination coefficient R ² value was (99.67%). The mathematical value of the interaction Adjacent R ² was (99.38%). Figure 7 shows the response surface graph for the interaction effects between the chosen factors, temperature and methanol-to-oil ratio (a and b), methanol to oil molar ratio, and catalyst wt% (c and d), temperature and catalyst wt% (e and f).Where A is the methanol-to-oil molar ratio, B is the temperature, and C is the catalyst concentration.

B i o d i e s e l y i e l d % = 758.9 + 12.57 A + 17.68 B + 53.21 C − 0.19 A 2 C 2 − 0.027 A B − 0.55 A C − 0.411 B C

TABLE 4. Analysis of Variance for biodiesel production.

The highest yield of (92.3%) was attained during a process temperature of 70 °C, reaction time fixed at 6 h during RSM, and not modelled as a factor, a catalytic level of 2.5 weight percent, as well as a methanol: crude oil ratio of 22.5:1, which is obtained from the response surface methodology data (Table 5). The equation represents the regression equation in uncoded units from the RSM. Figure 8 indicates the residual plots for the methyl ester yield of M. indica predicted vs. actual yield.

TABLE 5. Outcomes of experiments using central composite design to produce biodiesel by oil extracted from M. indica.

5.3 Characterization of biodiesel

From Figure 9, the FTIR analysis of biodiesel obtained from the transesterification process of M. indica seed oil revealed an extensive lineup of absorbance from 3,338.5 cm^-1 to 728.27 cm^-1. The chemical structure of transesterification contains the following groups: C-O, C=O, and C-H at different ranges of peaks, indicating the conversion of M. indica oil to biodiesel. The emergence of an ester group C-O is shown by the peaks located at 1,277 cm^-1, and the peaks located at 1713 cm^-1 indicate the presence of the C=O group of an ester, followed by the presence of C-H groups shown by the peak located at 2,922 cm^-1 and 2,853 cm^-1, and all other reported peaks are due to the movement of the fatty acid molecule (Nisar et al., 2017; Devaraj et al., 2020).

A novel approach to biodiesel characterization were applications of NMR spectroscopy, which is used to identify and quantify the molecular moieties of the molecules, especially the unsaturated long-chain alkyl esters, as well as to fully characterize and assign the structural makeup of biodiesel molecule samples. In Figure 10, it has been found that the presence of the Ester carbonyl peak at 173.29 ppm, which indicates the carbonyl carbon (C=O) of the fatty acid ester linkage, and multiple peaks in the range of 13–35 ppm correspond to various carbon atoms within the long aliphatic chains of the fatty acid alkyl groups. This is the key signature of biodiesel. In Figure 11, it has been found that the presence of methyl ester protons at 3.63 ppm and alkenyl hydrogens multiplets at 5.31 ppm from proton NMR examination results, which also indicates the key signature of biodiesel (Kara et al., 2018).

The analysis conducted through GC-MS of the biodiesel sample provided a complex composition of fatty acid methyl esters (FAMEs). Figure 12 shows the GC chromatogram of the produced biodiesel. The sample was found to contain methyl hexadecanoate (hexadecanoic acid, methyl ester), methyl 9-octadecenoate (9-octadecenoic acid, methyl ester), and methyl 9,12-octadecadienoate (9,12-octadecadienoic acid, methyl ester) as principal components. These esters are representative of biodiesel from vegetable oils. All the samples also contained significant amounts of dodecane, 2,6,11-trimethyl-, eicosane, and nonadecane. The presence of these compounds confirms that the conversion of oil to biodiesel is successful. The total area peak percentage indicates a conversion yield of approximately 92.3% biodiesel. There are several studies conducted for biodiesel production from different feedstocks with different catalyst have been listed in Table 6.

TABLE 6. The comparison of biodiesel production from various feedstock using ash-based catalysts is reported in the literature.

5.4 Reusability of the catalyst

Five continuous cycles of biodiesel synthesis were conducted to evaluate the ability of the residual ash to function as an active catalyst (Figure 13). A consistent decline was seen in activity with each additional use due to possible surface fouling or even partial deactivation of active sites. Eventually, the biodiesel production at the fifth cycle was reduced only by 3.37%, as compared to that of the first production, denoting strong catalytic stability. The repeatable activity of the catalyst over several cycles gives an economic attraction to leading industries’ ecologically sustainable biodiesel production. Figure 14 shows the M. indica oil and the produced M. indica biodiesel.

6 Engine analysis

6.1 Performance analysis of biodiesel

Brake thermal efficiency (BTE) and brake specific fuel consumption (BSFC) were tested in the engine performance analysis of two biodiesel blends (B10 and B20) and compared to commercial diesel (Figure 15). It was found that the BTE of both biodiesel blends was generally lower than commercial diesel but increased with increasing engine load; for instance, with full load, both biodiesel blends, B10 and B20, show 49.39% and 38.91% of BTE, which was 11.48% and 30.26% lower compared to the diesel, respectively, but this may result from the lower calorific value of biodiesel fuel blends. Alternatively, the BSFC of both biodiesel blends was 17.65% and 23.53% greater than the diesel, reflecting greater fuel consumption. This concurs with previous studies that indicate that biodiesel blends have higher BSFC; due to the low energy density and high viscosity of biodiesel blends, these fuel properties can affect fuel atomization and combustion efficiency (Ansari et al., 2018). Despite these variations, biodiesel blends indicated more optimal combustion characteristics with lower emissions of CO and HC; as such, it is a clear alternative to diesel with environmental advantages.

6.2 Emission analysis

Figure 16 represents the CO and NOx emissions of biodiesel blend B10 and B20 with respect to those of diesel. Carbon monoxide emissions occur because of deficient oxygen availability and lower combustion temperatures, often suggesting the conditions of partial combustion commonly attributed to a fuel-rich mixture and charge mixing deficiency (Kalargaris et al., 2017). For all load conditions, the BSCO emissions for diesel were more than those of biodiesel blends, and this could be attributed to the high oxygen availability of the latter that enhances the oxidation reactions. At maximum performance, the biodiesel blends B10 and B20 showed a remarkable reduction of 12.78% and 20.3%, respectively, in BSCO emissions compared with diesel (Ansari et al., 2018; Arumugam and Ponnusami, 2019).

Generally, the emission of nitrogen oxides tends to increase with an increase in the load of the engine because the combustion temperatures are higher in those cases. However, a reverse trend was observed in the brake-specific nitrogen oxide (BSNO) emissions: diesel showed lower NOx at full load compared to B10 and B20, which increased by 8.35% & 17.1%, respectively. The oxygen content in biodiesel blends can explain the reasons behind increased NOx emissions; NOx enhances combustion temperatures. Thus, such emissions pose difficulties regarding environmental laws relating to NOx reduction to diminish air pollution (Sathish, 2024). Just like biodiesel from soybean, jatropha, and palm oil, M. indica has been linked with increased NOx emissions due to its oxygen. Control measures of NOx emissions involve SCR, EGR, and LNTs. Changes in the engine parameters, such as injection timing and the air-fuel ratio, along with the use of single or dual-fuel strategies, provide further possibilities for the reduction of NOx emissions (Ashok et al., 2017).

Figure 17 displays the comparison of smoke emission and HC emission of biodiesel with that of commercial diesel. It is the smoke emissions that most frequently come from CI engines due to suboptimal combustion processes, insufficient availability of oxygen, and the existence of fuel-dominant regions in the combustion chamber. The problem occurs at elevated loads where significantly more fuel is needed at constant engine speed than at lower loads (Mahla et al., 2018). Under peak load conditions, diesel exhibited a smoke opacity of 70.3%, which was higher compared to that obtained with B10 and B20 by 5.97% and 12.23%, respectively. The higher oxygen content in the biodiesel blends (B10, B20) not only enhances combustion efficiency but also largely reduces smoke emissions. The emission of hydrocarbons (HC) generated by biodiesel combustion represents a significantly lower percentage than that derived from conventional commercial diesel fuel. Because of the oxygen contained in the molecular structure of biodiesel, complete combustion is facilitated (Arumugam et al., 2024). In addition, biodiesel has a lower aromatic content, which contributes to lower unburned hydrocarbons and hence, cleaner exhaust. The HC emission of the B10 and B20 in the present work shows 21.43% and 27.55% reduction compared to that of commercial diesel (Ashok et al., 2017). The properties of producing biodiesel blends are mentioned in Table 7.

TABLE 7. Physicochemical properties of M. indica biodiesel blends.

6.3 Techno-economic analysis

The process of making biodiesel from M. indica oil was modelled with Aspen Plus. The process flow chart is given in Figure 18. The production process has different stages: feed preparation, transesterification, product separation, and purification. The first step is that methanol and oil are mixed (MIX-1) and then transported to the reactor system (CSTR-1) with the help of a pump-1. The transesterification reaction is performed in a continuous stirred tank reactor (CSTR) operating at 60 °C–70 °C and atmospheric pressure; under these conditions, 95% conversion efficiency is obtained, which is based on experimental optimization. The output from the reactor is sent to a catalyst separator (centrifuge), where the solid catalyst is separated and recycled into the process.

The liquid is then sent to a methanol recovery section made up of a heat exchanger and a flash separator, where the unreacted methanol (∼90%) is recovered and recycled to the reactor. The residual blend after methanol recovery is sent to the decanter that performs phase separation of biodiesel and glycerol. The biodiesel rich phase is further purified in the distillation column to achieve about 98% purity, while the glycerol rich phase is withdrawn as a by product stream. The process features multiple recycle streams, especially for methanol and catalyst, to increase process efficiency and decrease operating costs. All the main operating units, including the mixer, pumps, reactor, centrifuge, flash unit, decanter, and distillation column, were simulated at the respective operating conditions, and mass balance convergences were obtained at the interface of all streams. The simulation included a detailed stream table and unit specifications for the validation of the process design and to provide a basis for the techno-economic analysis.

The cost summary is being provided in Table 8. The techno-economic analysis (TEA) of biodiesel production from M. indica oil has been conducted by performing process simulation in Aspen Plus. Supplementary Tables S1, 2 in the Supplementary Material shows the breakdown of CAPEX and OPEX of the simulated model. The plant is supposed to manufacture approximately 82, 774, 000 kg of biodiesel yearly, using the methodology. The economic evaluation was done by means of market prices, which are 0.91 USD/kg for biodiesel, 0.36 USD/kg for glycerol, 0.54 USD/kg for methanol, and 0.33 USD/kg for feedstock oil. The total capital investment, including equipment and installation, was set at 4,050,180 USD, with the equipment cost being 703,300 USD and the total installed cost 3,824,130 USD. Operating costs, including raw materials and utilities, were estimated at 74, 200, 000 USD/yr, with raw materials being the largest part of the cost (67, 245, 000 USD/yr) and utilities, only 111,565 USD/yr. Product revenue annually, at the given prices, is estimated at 75, 325, 100 USD/yr. This means a payback period of 3.6 years, considering the desired rate of return that is 20% per year.

TABLE 8. Cost summary for producing Biodiesel from M. indica oil.

Even though the profit margin is quite low due to the costly raw materials, the process is large-scale and economically feasible, and further patenting of methanol recovery, catalyst reusability, and feedstock optimization may make the overall process more profitable.

The sensitivity analysis indicated that biodiesel selling price strongly influences process profitability (Figure 19). Increasing biodiesel price improved annual net profit significantly, whereas reductions in selling price resulted in economic losses. This demonstrates that market fluctuations in biodiesel value play a critical role in determining the commercial feasibility of the process.

7 Conclusion

In conclusion, the current research demonstrates that lignocellulose biomass—in this case, the shell of M. indica can be utilized as a heterogeneous catalyst for biodiesel production from M. indica oil. The optimal conditions determined by response surface methodology yielded an ester content of 92.3% (GC-MS area-based) at the optimal temperature of 70 °C, an optimal adjustment for a 6-h process time from the test, an optimal catalytic amount of 2.5 wt%, and an optimal methanol-to-oil ratio of 22.5:1. Engine performance tests for biodiesel blends B10 and B20 demonstrated significant reductions in emissions over commercial diesel; hydrocarbon emissions reductions of 21.43% for B10% and 27.55% for B20, carbon monoxide emissions reductions of 12.78% for B10% and 20.3% for B20, and smoke emissions reductions of 5.97% for B10% and 12.23% for B20. NOx levels were increased by 8.37% and 17.1% for B10 and B20 blends. The techno-economic analysis was performed for the process in Aspen Plus economic analyser and showed that the process is feasible with a payback period of 3.6 years. Overall, results contribute to sustainability in biofuels by providing an alternative solution to environmental issues while utilizing waste biomass as a catalyst.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

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.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/ffuel.2026.1840089/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Malinee Sriariyanun, Kmutnb, Thailand

Reviewed by: Adityas Agung Ramandani, Yuan Ze University, Taiwan

Ibrahim Fetuga, Clarkson University, United States

Nida Arshad, King Mongkut’s University of Technology North Bangkok, Thailand