Sustainable valorization of waste cooking oil via low-temperature transesterification using BaO/ZnO nanocatalyst: process optimization and mechanistic studies

Abstract

Biodiesel is widely regarded as a promising renewable energy source with minimal environmental impact. In our study, a novel catalyst, BaO/ZnO (BZO), was synthesized using a wet-impregnation method and used in biodiesel production from waste cooking oil. The synthesized catalysts were physico–chemically characterized using different analytical techniques, including TGA, XRD, FTIR, BET, XPS, HR-SEM, and HR-TEM. NMR analysis was employed to quantify the synthesized biodiesel. The Box–Behnken Design (BBD) model in the Response Surface Methodology (RSM) approach was implemented to optimize various reaction parameters involved in biodiesel production. The maximum biodiesel conversion of 97.3% and yield of 97.1% were obtained under optimized transesterification reaction conditions of 2.6 wt% catalyst loading at 39.9 °C reaction temperature with a MeOH to oil molar ratio of 10.9 : 1 for a reaction time of 29.8 min. In addition, the synthesized BZO catalyst was recyclable up to five times, suggesting higher catalytic efficacy and stability throughout the reaction. The turnover frequency of the proposed catalyst was found to be 15.52 h⁻¹. Kinetic and thermodynamic studies revealed that the obtained values of activation energy (E_a), enthalpy of activation (ΔH^#), and entropy of activation (ΔS^#) were 37.64 kJ mol⁻¹, 35.76 kJ mol⁻¹, and −150.73 J mol⁻¹ K⁻¹, respectively. Moreover, various fuel properties like kinematic viscosity, calorific value, flash point, pour point, and cloud point were consistent with ASTM D-6751 international standards. A green metrics study demonstrates that the overall biodiesel production process is sustainable and environmentally friendly.

---

1 Introduction

Excessive use of fossil fuels globally has prompted significant environmental and financial challenges, particularly in light of depleting resources and intensifying greenhouse gas emissions.¹ As a result, research on biofuels as alternative energy sources has been gaining attention in recent years. Biofuels include biomethanol, bioethanol, biobutanol, and biodiesel.² Among these, biodiesel is considered a potential replacement for petrodiesel as a renewable and sustainable fuel with some advantages, including non-corrosiveness, biodegradability, and suitability for existing diesel engines when blended with petrodiesel.³ Additionally, it is devoid of aromatics and sulfur, which release lower amounts of CO₂, CO, and other greenhouse gases during combustion.⁴ Biodiesel consists of fatty acid methyl esters (FAME), which are usually prepared through a transesterification reaction between methanol and triglyceride sources such as edible oils,⁵ non-edible oils,⁶ microalgal oil,⁷ or animal fats,⁸ in the presence of an efficient catalyst, resulting in biodiesel and by-product glycerol as illustrated in Scheme 1. Glycerol has potential applications in the pharmaceutical and food industries after being valorized into valuable products such as solketal, acrolein, glycerol carbonate, lactic acid, and other derivatives.^9–12

Scheme 1 Transesterification reaction of biodiesel synthesis.

Around 70–80% of the biodiesel production cost relies on feedstock selection.¹³ According to recent research, countries such as Japan, China, Malaysia, United States, and Europe produce approximately 16.5 million tons of waste cooking oil (WCO) annually.¹⁴ The disposal of WCO is worsening the soil and water quality, leading to substantial environmental pollution.¹⁵ Therefore, utilization of WCO as feedstock for biodiesel production has recently gained great attention due to its wide availability and low-cost source generated as waste from the food industry.^16,17 The concept of “waste to wealth” not only lowers the cost of biodiesel production but may also be the most effective approach to waste management and environmental protection. The cost-effectiveness of biodiesel production depends on selecting an effective catalyst for the transesterification process. Biodiesel synthesis employs different types of catalysts, such as homogeneous, heterogeneous, and enzymatic. Homogeneous basic catalysts such as NaOH and KOH in biodiesel synthesis from waste cooking oil (WCO) are found to be less effective as they possess several challenges, including saponification, emulsification, and hydrolysis of the product, which consequently lead to poor FAME yield due to high free fatty acid (FFA) content and greater impurities in WCO.^18–20 Nevertheless, non-recyclability, difficulties in separation, wastewater generation during product purification, and the corrosive nature of homogeneous catalysts render them unsuitable for biodiesel synthesis.^21,22 Therefore, heterogeneous catalysts are employed in the transesterification of WCO as they are highly recyclable, inexpensive, easy to separate from the product mixture, less corrosive, and generate less waste during separation.^23,24 Despite having these advantages, heterogeneous catalysts require comparatively higher temperatures and time to obtain more than 90% biodiesel yield.²⁵ Researchers have focused more on metal oxide-based catalysts due to their high catalytic activity, stability, and heterogeneous nature. Abdul Rahim et al. reported 98.78% biodiesel conversion from WCO using a CaO–TiO₂ nanocatalyst at 65 °C temperature for a reaction time of 150 min.²⁶ Aghel et al. synthesized a MgO/CaO catalyst from industrial waste and employed it in the transesterification reaction of WCO. They achieved FAME conversion of more than 93.32% utilizing 9 wt% catalyst at a temperature of 63 °C and a reaction time of 120 min.²⁷ Both studies took a comparatively longer time to convert WCO into biodiesel. Therefore, scientists have shifted to using BaO-based catalysts in the transesterification reaction, due to their stronger basicity and higher catalytic activity than other alkaline earth metal oxides like CaO, MgO, and SrO.²⁸ However, BaO is highly susceptible to leaching during the reaction period due to its slight solubility in methanol and has a high affinity toward atmospheric CO₂ to form barium carbonate (BaCO₃).²⁹ Consequently, it reduces the available surface area and number of active basic sites and lowers catalytic activity. To address these challenges, incorporating BaO into a metal oxide support like zinc oxide (ZnO) has been proposed, as it can enhance particle dispersion, improve surface area, and thereby enhance the stability and recyclability of the catalyst.

In the current study, we designed a heterogeneous nanocatalyst, barium oxide supported on zinc oxide (BaO/ZnO), and employed it in biodiesel production from WCO. The reported catalyst demonstrated outstanding catalytic efficiency for low-temperature biodiesel synthesis. The synthesized catalyst was characterized using several instrumental methods, including thermogravimetric analysis (TGA), Fourier Transform Infrared Spectroscopy (FTIR), X-ray Diffraction (XRD), Temperature Programmed Desorption (CO₂−TPD), X-ray Photoelectron Spectroscopy (XPS), Brunauer–Emmett–Teller (BET) surface area analysis, high-resolution scanning electron microscopy (HR-SEM), and high-resolution transmission electron microscopy (HR-TEM). The optimization of transesterification process parameters using the one variable at a time (OVAT) method is time-consuming and requires lots of reagents to conduct a large number of experiments. Therefore, the RSM–BBD approach was employed to optimize the influencing parameters related to the transesterification reaction. Moreover, kinetic and thermodynamic studies were performed to evaluate the energy of activation, enthalpy of activation, entropy of activation, and Gibbs free energy of the methanolysis reaction. The studies of recyclability and cost analysis were also performed to validate the production cost of biodiesel. In addition, several green parameters, including E-factor and PMI, were evaluated to assess the sustainability of the overall biodiesel production process.

2 Materials and methods

2.1 Chemicals and raw materials

Waste cooking oil (WCO) was acquired from the mess of Prof. S. N. Bose hostel, IIT(BHU), Varanasi. The collected WCO was subjected to pretreatment processes, such as filtration and drying, to remove impurities. Several essential physicochemical characteristics of WCO were examined, as demonstrated in Table S1. Chemicals such as zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O, 99.9% pure), sodium carbonate (Na₂CO_3, EMPLURA, 99% pure), barium nitrate (Ba(NO₃)₂, AR grade, 99.5% pure), hexane (C₆H_14, EMPLURA, >95%, Merck Life Science Pvt. Limited) and methanol (CH₃OH, EMPLURA, 99% pure) were purchased from Sigma-Aldrich and used for the experiments without further purification. Deionized (DI) water was used throughout the experiment.

2.2 Catalyst synthesis techniques

2.2.1 Synthesis of the ZnO support. The co-precipitation method was followed during the preparation of the ZnO support. At first, 7.43 g of precursor zinc nitrate hexahydrate salt was dissolved in 50 mL of distilled water under stirring conditions at 600 rpm for 1 h. Subsequently, the freshly prepared 0.5 M Na₂CO₃ solution was added dropwise into the solution until the complete precipitation of zinc carbonate (ZnCO₃) had occurred. During the precipitation, the overall pH of the solution was maintained at 9–10, and the resulting mixture was agitated for another 5–6 h. Then, the precipitate was separated through centrifugation at a constant speed of 12 000 rpm and washed with DI water multiple times to remove excess Na₂CO₃. The resulting white mass was then dried in a hot air oven at 80 °C overnight, followed by calcination at 550 °C for 4 h in a muffle furnace to obtain ZnO nanoparticles, denoted as ZO.

2.2.2 Synthesis of the BaO/ZnO nanocatalyst. The catalyst barium oxide supported on zinc oxide (BaO/ZnO) was prepared using a simple and cost-effective wet impregnation method. In this method, 2 g of barium nitrate was dissolved in 100 mL of DI water in a 250 mL beaker, and subsequently, 1.34 g of zinc oxide was added to it. Then the resultant solution mixture was aged with continuous stirring on a magnetic stirrer for 3 h at 600 rpm. After the formation of a homogeneous mixture, the temperature of the solution was increased to 120 °C, and the mixture was heated till complete dryness of the solution. The mixture was then kept overnight in a hot air oven at 85 °C. The next day, the resultant white mass was crushed finely in a mortar and pestle and subjected to calcination at 800 °C in a muffle furnace for 4 h to obtain the BaO/ZnO nanocatalyst, which is abbreviated as BZO. The synthesis procedure of the BZO nanocatalyst is shown in Fig. 1. Three different BZO catalysts were prepared using varying Ba to Zn atomic ratios. The synthesized BZO catalysts with varied Ba to Zn atomic ratios, labeled 23 BZO (Ba : Zn 2 : 3), 11 BZO (Ba : Zn 1 : 1), and 32 BZO (Ba : Zn 3 : 2), were characterized using numerous analytical techniques, as described below.

Fig. 1 Schematic illustration of the synthesis procedure of the BaO/ZnO nanocatalyst.

2.3 Characterization techniques

Various characterization techniques were employed to examine the physicochemical characteristics of the synthesized catalysts and biodiesel product. A detailed, comprehensive description of the characterization techniques is provided in the SI file (Text S1).

2.4 Implementation of the transesterification reaction and analysis of the product

The catalytic efficiency of BaO/ZnO as a heterogeneous basic catalyst in biodiesel production was studied using a lab-scale transesterification reaction. The transesterification reaction of WCO for biodiesel production was accomplished in a 100 mL three-necked round-bottom (RB) flask by assembling with a reflux condenser, a hot plate with a magnetic stirrer, and a thermometer. The following setup was put into an oil bath. A thermometer was immersed through one neck of the RB to examine the reaction temperature. Before performing every experimental run, the catalyst was activated in a hot-air oven at 85 °C for 15 minutes to remove the moisture adsorbed on the catalyst surface. Then, the required amount of BZO catalyst and methanol, followed by 15 g of oil, were put into the reaction vessel. The temperature of the hot plate was elevated to a particular temperature, and the stirring speed was maintained constant at 600 rpm. The entire reaction was executed under refluxing conditions. To evaluate optimal conditions for producing biodiesel through the transesterification process, several transesterification runs were carried out with varying reaction conditions of catalyst loading (1.0–3.0 wt%), reaction temperature (30–50 °C), MeOH to oil molar ratio (10 : 1–14 : 1), and reaction time (15–45 min). After completion of each run, the catalyst was separated from the reaction mixture through filtration using Whatman 41 filter paper, followed by washing multiple times with hexane and methanol to remove adsorbed impurities on the catalyst surface. Then, the filtrate product mixture was placed in a separating funnel and left for one day to allow the product mixture to settle completely. Three layers had developed the next day in the separating funnel: glycerol (side product) at the bottom, biodiesel (major product) in the middle, and excess methanol at the top. The crude glycerol was separated, and excess methanol was recovered through vacuum distillation using a rotary evaporator. The synthesized biodiesel was quantified by estimating the conversion of WCO into waste cooking oil methyl ester (WCOME) using proton nuclear magnetic resonance (¹H-NMR) analysis. The biodiesel conversion was calculated using eqn (1).³⁰where I_OCH3 represents the integration value of methyl ester protons appearing at a chemical shift value of 3.68 ppm and I_α−CH2 indicates the integration value of protons of the α-methylene unit, appearing at a chemical shift value of 2.32 ppm, respectively. Furthermore, the WCOME yield was evaluated using eqn (2).³¹

2.5 Design of the experiment and statistical analysis using RSM

Response surface methodology (RSM) is a combination of statistical and mathematical tools used in designing experiments and optimizing different independent variables or responses of the desired experiments. Numerous methods can be used for optimization through RSM, such as Box–Behnken Design (BBD), Definitive Screening Design (DSD), Central Composite Design (CCD), and D-Optimal Design (DOD).³² In our case, optimization of the reaction parameters involved in the biodiesel synthesis from WCO was performed through the RSM in the BBD method using Design Expert 13.0.5.0 software. The experimental range of every reaction parameter was defined according to the coded levels: −1 (minimum value), 0 (middle value), and +1 (maximum value), as illustrated in Table 1. A second-order regression polynomial equation in terms of different independent variables, such as MeOH to oil molar ratio (P), reaction temperature (Q), catalyst loading (R), and reaction time (S), was applied to evaluate biodiesel conversion (%), as shown in eqn (3).³³where Y represents biodiesel conversion, a₀ denotes a constant called intercept, and a_i, a_ii, and a_jk are expressed as linear coefficients, quadratic coefficients, and interaction factors, respectively. x_i, x_j, and x_k are independent variables. i, j, and k are the positions of the variables, and n is the total sum of the investigated variables. The experimental error is expressed by b.³⁴ Analysis of variance (ANOVA) was executed during the optimization study using RSM to check the significance of various model coefficients and evaluate statistical parameters such as p-value, f-value, coefficient of determination (R²), and adjusted statistical coefficient (adj R²).³⁵

Table 1 Experimental range and levels in BBD

Variables	Symbol	Levels
		−1	0	+1
MEOH to oil molar ratio	P	10 : 1	12 : 1	14 : 1
Temperature (°C)	Q	30	40	50
Catalyst loading (wt%)	R	1	2	3
Time (min)	S	15	30	45

---

2.6 Kinetic and thermodynamic studies

The kinetic study establishes a relationship between reaction rate and temperature in a particular process. A thermodynamics study predicts whether the process is energetically favorable or not. For the current study, various kinetic parameters such as activation energy (E_a), rate constant (k), pre-exponential factor (A), and thermodynamic parameters such as enthalpy of activation (ΔH^#), entropy of activation (ΔS^#), and Gibbs free energy (ΔG^#) were experimentally evaluated for the transesterification process of WCO using the BZO nanocatalyst. The transesterification process involves the reaction between one mole of waste cooking oil (WCO) and three moles of methanol in the presence of a catalyst, resulting in the production of three moles of waste cooking oil methyl ester (WCOME) and one mole of byproduct glycerol (Gly). The overall stoichiometric balanced equation of the transesterification process is shown in eqn (4) below.

WCO + 3CH₃OH ↔ 3WCOME + Gly (4)

The rate equation for the above transesterification reaction is expressed in eqn (5) as follows.

Rate = k₁ [WCO] [CH₃OH]³ (5)

where k₁ denotes the rate constant of the transesterification reaction. [WCO] and [CH₃OH] denote the concentration of WCO and methanol, respectively. Eqn (4) demonstrates that the transesterification reaction is highly reversible. To promote the reaction in the forward direction, an excess amount of methanol relative to oil was used during the reaction process. Consequently, it was found that the concentration of methanol remained constant under the experimental conditions. Therefore, the rate of the transesterification process relies on the concentration of WCO only. Thus, it is worth mentioning that the above transesterification process follows pseudo-first-order kinetics.³⁶ Consequently, the rate equation has been rewritten below in eqn (6).

Rate = k [WCO], or d[WCO]/dt = k [WCO] (6)

where k = k₁ [CH₃OH]³, the effective rate constant associated with the transesterification process. Upon integrating and simplifying eqn (6), the ultimate expression for determining the rate constant is obtained [eqn (7)].

−ln[1 − X_WCOME] = k·t (7)

where k denotes the reaction rate constant at a particular temperature, and X_WCOME represents the extent of methyl ester conversion at an interval of time t. Now, the slope of the plot of ln[1 − X_WCOME] against ‘t’ (min) gives the value of ‘k’ (min⁻¹) at a particular temperature. Furthermore, the activation energy and pre-exponential factor were calculated using the Arrhenius equation, as illustrated in eqn (8).³⁷

After simplifying eqn (8), we get

Here, A is the pre-exponential factor (in min⁻¹), and E_a (in kJ mol⁻¹) is the energy of activation for the transesterification reaction. R and T represent the universal gas constant (8.314 J mol⁻¹ K⁻¹) and absolute reaction temperature (in Kelvin), respectively. The slope and intercept of the plot of ln(k) against 1/T provide the values of E_a and A, respectively [eqn (9)]. Moreover, thermodynamic parameters such as the enthalpy of activation (ΔH^#) and entropy of activation (ΔS^#) can be evaluated using the Eyring–Polanyi equation, as represented below in eqn (10).³⁸Where k_b denotes the Boltzmann constant (1.38 × 10⁻²³ J K⁻¹) and h represents Planck's constant (6.626 × 10⁻³⁴ J s), respectively. Furthermore, the resultant values of ΔH^# and ΔS^# obtained from eqn (10) are employed in the Gibbs–Helmholtz equation to evaluate the Gibbs free energy (ΔG^#) at the respective temperature [eqn (11)].

ΔG^# = ΔH^# − TΔS^# (11)

3 Results and discussion

3.1 Characterization of the synthesized catalyst

3.1.1 TGA of uncalcined catalysts. The thermal stability of uncalcined ZO and BZO catalysts was examined using thermogravimetric analysis (TGA) under an inert atmosphere (N₂). The weight loss against temperature plot depicted in Fig. 2(a) was studied to explain the different weight loss phases. In the case of the ZO catalyst, a sharp weight loss was observed between 180 °C and 500 °C, attributed to the decomposition of zinc carbonate (ZnCO₃) into stable zinc oxide (ZnO).³⁹ Thus, 550 °C was considered the optimal calcination temperature for the ZO catalyst. From the TGA graph, the total weight loss of the ZO catalyst was estimated to be 26.35%. Similarly, the TGA plot of uncalcined BZO ascertained three distinct weight-loss phases at different temperatures. The beginning phase of weight loss was observed at around 105 °C, probably due to eliminating physically bound water molecules on the catalyst surface.⁴⁰ The second weight loss associated with the temperature range of 300–550 °C corresponds to the removal of chemically bound water molecules from the lattice structures. The final weight loss happened in the temperature range of 600 °C to 720 °C, mainly due to the thermal decomposition of barium nitrate [Ba(NO₃)₂] into barium oxide (BaO) as a stable catalyst along with the formation of nitrogen dioxide (NO₂) and oxygen (O₂) as shown in eqn (12).⁴¹ The overall weight loss for the uncalcined BZO catalyst was calculated to be 31.69%. Beyond 720 °C, no further weight loss of the sample was observed, signifying that the catalyst became thermodynamically stable at 720 °C. Therefore, the optimum calcination temperature for the BZO catalyst was chosen to be 800 °C, at which maximum biodiesel conversion was obtained.

Fig. 2 (a) TGA plot of uncalcined ZO and BZO catalysts, (b) XRD patterns and (c) FTIR spectra of ZO, 23 BZO, 11 BZO, and 32 BZO catalysts, (d) CO₂-TPD profile of the 11 BZO catalyst, and BET adsorption–desorption isotherm (inset: BJH pore size distribution) of (e) ZO and (f) 11 BZO catalysts.

3.1.2 X-ray powder diffraction analysis. The crystalline structure, purity of phase, and crystallite size of ZO and BZO catalysts were examined using X-ray powder diffraction analysis (XRD). Fig. 2(b) reveals the XRD patterns of ZO, 23 BZO, 11 BZO, and 23 BZO catalysts. The major characteristic diffraction peaks were observed at 2θ values of 31.7°, 34.4°, 36.2°, 47.5°, 56.5°, 62.8°, 66.3°, 67.9°, 69.1°, 72.5°, and 76.9°, associated with Miller indices (hkl) of (100), (002), (101), (102), (110), (103), (200), (112), (201), (004), and (202), respectively, confirming the successful formation of the hexagonal wurtzite phase of ZnO (ZO), matching with JCPDS card no. 36-1451.⁴² No extra peaks were observed, implying that ZnO is present in the pure phase. Other notable peaks with Miller indices of (110), (101), (200), (201), (211), (220), (102), (310), and (212) at 2θ of 13.4°, 16.1°, 18.8°, 23.0°, 24.9°, 26.7°, 27.8°, 29.9°, and 33.7°, respectively, correspond to the tetragonal phase of BaO, consistent with the JCPDS card no. 26-0178.^43,44 Moreover, distinct peaks of both ZnO and BaO were observed in the case of composite BZO catalysts, suggesting the successful impregnation of BaO on the ZnO catalyst support. Furthermore, no peaks due to any impurities were found in the XRD patterns of 23 BZO and 11 BZO catalysts, indicating the high purity of the samples. But, in the case of 32 BZO some extra peaks due to the formation of BaCO₃ as an impurity were observed at 23.9°, 27.6°, 29.5°, 39.5°, 42.8°, 44.1°, 46.6°, and 48.9°, corresponding to hkl values of (111), (002), (012), (220), (041), (202), (113), and (222), respectively, matched with the JCPDS card no. 05-0378.⁴⁵ This may happen due to an excess stoichiometric quantity of Ba in 32 BZO, lowering the overall basicity of the catalyst. It is worth mentioning that the peak intensities of the BaO phase were higher in 11 BZO compared to the other two catalysts, indicating the existence of highly active basic sites, resulting in a very high biodiesel conversion at low temperature.

The average crystallite size of the prepared ZO and BZO catalysts was evaluated using the Debye–Scherrer equation as illustrated in eqn (13).⁴⁶

where D represents the average crystallite size, K is the crystallite size coefficient (K = 0.9), λ denotes the wavelength of the X-ray radiation source (Cu Kα), β refers to the full width at half maximum (FWHM) in the XRD pattern, and θ symbolizes Bragg's diffraction angle. The average crystallite size of the synthesized ZO and BZO catalysts, as calculated using the Debye–Scherrer equation, is illustrated in Table 2. The ZO catalyst possessed an average crystallite size of 23.65 nm. This value shows that the prepared supporting oxide was in the nanoscale and exhibited a high surface area, making it a superior support of BaO for low-temperature biodiesel production. Furthermore, an increase in crystallite size was observed for BZO catalysts compared to ZO, which may be due to the higher calcination temperature (800 °C) applied during the synthesis of the BZO catalyst. The average crystallite size of the BZO catalyst decreased from 32.82 nm for 23 BZO to 28.57 nm for 11 BZO and further increased to 35.13 nm for 32 BZO. Thus, the lower crystallite size has a higher degree of disorder in the crystal structure of 11 BZO, making it highly catalytically active in biodiesel production due to its numerous active basic sites on the catalyst surface.⁴⁷

Table 2 Average crystallite size, surface properties, and total basicity of the catalysts

Catalyst	Average crystallite size (nm)^a	Specific surface area (m^2 g^−1)^b	Total pore volume (cm^3 g^−1)^c	Average pore diameter (nm)^c	Total basicity (mmol g^−1)^d
a Calculated using the Debye–Scherrer equation.b Evaluated using the BET method.c Determined by the BJH method.d Calculated by the CO2-TPD method.
ZO	23.65	10.2	0.033	30.6	0
23 BZO	32.82	1.51	0.059	55.5	4.28
11 BZO	28.57	6.0	0.046	36.5	5.76
32 BZO	35.13	1.38	0.096	94.8	3.12

---

3.1.3 FTIR analysis. The existence of various functional groups on the catalyst samples was assessed by FTIR analysis. Fig. 2(c) demonstrates the FTIR spectra of ZO, 23 BZO, 11 BZO, and 32 BZO catalyst samples. The characteristic peak observed around 485 cm⁻¹ was attributed to the stretching vibration of Zn–O bonds in both ZO and BZO catalysts.⁴⁸ A sharp peak originating at 690 cm⁻¹ corresponds to the Ba–O bond vibration found in all three BZO catalysts.^49,50 The broad peak between 3350 and 3600 cm⁻¹ was ascribed to the O–H stretching band, probably due to absorbed atmospheric moisture on the catalyst surface.⁵¹ Most interestingly, the peak intensity of the Ba–O bond was highest in 11 BZO, demonstrating the highest catalytic activity in the transesterification reaction. Thus, the Ba–O bond plays a major catalytic role in biodiesel production, which correlates with similar results obtained from the XRD patterns of BZO catalysts [Fig. 2(b)]. The FTIR spectrum of 32 BZO contains some extra peaks at 850 cm⁻¹, 1058 cm⁻¹, 1426 cm⁻¹, and 1745 cm⁻¹, indicating the presence of O–C–O in-plane bending, symmetric C–O stretching, asymmetric stretching of C–O, and C=O bond vibration, respectively, which is due to the formation of impure BaCO₃ phases with the oxide phases.⁵² Probably due to the presence of excess Ba in 32 BZO, it readily absorbs atmospheric carbon dioxide and forms carbonate species, reflected in both XRD patterns [Fig. 2(b)] and FTIR spectra [Fig. 2(c)].

3.1.4 Basicity analysis using CO₂-TPD. The overall basic strength of the synthesized catalysts was evaluated using the CO₂-temperature programmed desorption method, which relies on the number of CO₂ molecules adsorbed on the available basic sites of the catalyst.⁵³ Generally, the CO₂-TPD profile exhibits three distinct temperature regions. The TCD signal observed in temperature regions below 250 °C indicates the presence of weak basic sites. Moreover, the CO₂ desorption signal found in the temperature range of 250 °C to 450 °C corresponds to the moderate basic sites, and temperature regions above 450 °C are associated with the stronger basic desorption sites.⁵⁴ Fig. 2(d) illustrates the CO₂ desorption profile of the 11 BZO catalyst, which is associated with six major desorption peaks at 504 °C, 570 °C, 623 °C, 681 °C, 712 °C, and 787 °C, respectively. This indicates that all the peaks correspond to strong basic sites in the 11 BZO catalyst. Furthermore, no peaks due to weak and moderate basic sites were observed below 500 °C in the TPD profile, demonstrating the strong basicity of 11 BZO, which results in higher WCOME conversion. The area under each desorption peak was integrated, and their sum was used to determine the overall basicity of the synthesized catalysts.⁵⁵ Consequently, the total basicity of 23 BZO, 11 BZO, and 32 BZO catalysts was evaluated to be 4.28, 5.76, and 3.12 mmol g⁻¹, respectively (Table 2). The strong interaction with BaO in the ZnO support creates a higher number of basic sites in 11 BZO, which is also supported by both the XRD and FTIR results. Thus, the 11 BZO catalyst was selected further to optimize the reaction parameters involved during biodiesel synthesis from WCO.

3.1.5 BET surface area analysis. The specific surface area of the catalyst significantly affects the transesterification reaction rate during biodiesel production. The Brunauer–Emmett–Teller (BET) nitrogen physisorption isotherm was used to estimate the surface area of the ZO and 11 BZO catalysts, with the resulting plots presented in Fig. 2(e) and (f), respectively. Additionally, other surface characteristics, such as pore volume and pore diameter of the catalysts, were quantified using the Barrett–Joyner–Halenda (BJH) method. The obtained values of surface area, pore diameter, and pore volume of all the catalysts are listed in Table 2. The obtained BET plots of both ZO and 11 BZO catalysts show a Type-IV isotherm with an H1 hysteresis loop, and the BJH pore size distribution confirms that the pore diameters of both catalysts were in the range of 2–50 nm.⁵⁶ These prove that both the catalysts are mesoporous. Moreover, the obtained specific surface area value was decreased from 10.2 m² g⁻¹ for ZO to 6.0 m² g⁻¹ for 11 BZO, which may be due to the mesoporous filling of BZO particles on the ZO support. Additionally, the pore diameter and pore volume values in the ZO catalyst were 30.6 nm and 0.033 cm³ g⁻¹, while those in 11 BZO were 36.5 nm and 0.046 cm³ g⁻¹, respectively. It was also observed that 11 BZO possessed a higher surface area than the other two BZO catalysts. The higher surface area and mesoporous surface topology combined constitute the 11 BZO as an effective catalyst in transesterification, with higher biodiesel conversion.⁵⁷

3.1.6 HR-SEM and EDX analysis. The morphological characteristics of catalyst surfaces were viewed through a high-resolution scanning electron microscopy (HR-SEM) instrument. Fig. 3(a) illustrates the morphology of ZO, demonstrating a homogeneous mixture of variable particle shapes, like circular, cubic, hexagonal, and irregular.⁵⁸ The homogeneity of particles in the catalyst surface decreases in BZO as compared to ZO, and this may be due to the calcination of the uncalcined catalyst at a very high temperature (800 °C).⁵⁹ In Fig. 3(b), the SEM image of the 23 BZO catalyst shows that the particles are irregular in shape throughout the surface, suggesting successful impregnation of BaO on the ZnO support. An increase in heterogeneity of the catalyst surface with uneven particle shape, size, and orientation, as shown in Fig. 3(c), makes the 11 BZO catalyst superior in the transesterification reaction due to possessing a higher number of active sites for adsorbing WCO on the catalyst surface and converting it into the biodiesel product.⁶⁰ The SEM image of 32 BZO, as illustrated in Fig. 3(d), demonstrates that the particles are larger in shape and agglomerated in nature. When the stoichiometric ratio of Ba to Zn increased from 1 : 1 to 3 : 2, there was a formation of white barium carbonate (BaCO₃) as an impure phase, which was confirmed by the SEM image of 32 BZO, where the needle-like particles accumulated throughout the ZO support.⁶¹ Energy-dispersive X-ray (EDX) analysis was employed to verify the atomic ratio of constituent elements in ZO and BZO catalysts. Moreover, the EDX spectrum of ZO, as shown in Fig. 3(e), reveals that Zn and O are present in the catalyst sample according to their atomic and weight percentages, whereas the EDX histogram of the 11 BZO catalyst, as indicated in Fig. 3(f), demonstrates the presence of Ba, Zn, and O as constituent elements in the sample. No other elemental peak was observed, confirming that BaO and ZnO were present in the pure phase. Fig. 3(g–i) illustrates the elemental mapping of Ba, Zn, and O, which validates the successful incorporation of BaO on the ZnO support in the 11 BZO catalyst.

Fig. 3 HR-SEM images of (a) ZO, (b) 23 BZO, (c) 11 BZO, and (d) 32 BZO, EDX of (e) ZO and (f) 11 BZO, and elemental mapping of (g) Ba, (h) Zn, and (i) O in the 11 BZO catalyst utilized for biodiesel production.

3.1.7 HR-TEM analysis. Microstructural characteristics like nanostructures, particle diameter, and crystallinity of catalyst samples were studied using high-resolution transmission electron microscopy (HR-TEM) analysis, as illustrated in Fig. 4. The TEM image of ZO, as shown in Fig. 4(a), indicates that the particles are of semi-circular to irregular shape. Moreover, the particles are uniformly distributed and less agglomerated throughout the surface. Furthermore, in Fig. 4(b), the SAED pattern showing the bright dotted ring-like electron diffraction pattern suggests that ZnO has a polycrystalline structure, which is in good agreement with XRD results. Fig. 4(c) demonstrates the lattice fringes with an interplanar spacing of 0.281 nm corresponding to the (100) plane of the ZnO lattice.⁶² From the particle size distribution plot, as depicted in Fig. 4(d), the average particle diameter of the ZO support was calculated to be 22.61 ± 0.76 nm, suggesting the ZO support is nano-dimensional. Similarly, in Fig. 4(e), the TEM micrograph of the 11 BZO catalyst shows that the particles are almost spherical-shaped and evenly dispersed with no agglomeration. The brilliant dots in the SAED pattern depict the BZO lattice as highly crystalline [Fig. 4(f)]. Furthermore, an interplanar spacing of 0.319 nm between the lattice planes corresponding to the (hkl) value of (102) was observed, confirming the presence of BaO in the 11 BZO catalyst [Fig. 4(g)]. Similarly, in Fig. 4(h), the particle size distribution plot shows that the average particle size in the 11 BZO catalyst was 28.28 ± 1.06 nm, revealing that the 11 BZO catalyst also has nanoscale dimensions.

Fig. 4 High-resolution TEM image, SAED pattern, crystal plane indexing, and particle size distribution of (a–d) ZO and (e–h) BZO catalysts.

3.1.8 XPS analysis. X-ray photoelectron spectroscopic analysis was executed to evaluate the elemental composition and corresponding oxidation state of the constituent elements present in the 11 BZO catalyst. The XPS spectra of the 11 BZO catalyst are shown in Fig. 5, which clarifies that Ba, Zn, and O are present on the catalyst surface. Two characteristic peaks of Ba 3d spectra originated at the corresponding binding energies of 779.28 eV and 794.58 eV, signaling the spin–orbit doublet of Ba 3d_5/2 and Ba 3d_3/2, respectively, as shown in Fig. 5(a). The difference of 15.3 eV binding energy between Ba 3d_5/2 and Ba 3d_3/2 peaks demonstrates that Ba is in a + 2 (II) oxidation state.⁶³ Similarly, the high-resolution spectrum of Zn 2p, as shown in Fig. 5(b), contains two characteristic peaks at 1021.5 eV and 1044.5 eV, represented by the Zn 2p_3/2 and Zn 2p_1/2 doublet, respectively. The binding energy difference of 23.0 eV between these two peaks indicates the presence of Zn²⁺ on the BZO catalyst surface.⁶⁴ Fig. 5(c) shows the deconvoluted spectrum of O 1s, corresponding to three distinct peaks due to three different oxygen species. The first peak at 529.8 eV was observed due to lattice oxygen (O_lat) in the metal–oxygen bond. Moreover, the second peak was found at 531.3 eV, ascribed to the –OH functionality due to moisture adsorbed on the catalyst surface, and the peak at 535.9 eV was observed due to chemically adsorbed oxygen (O_ads) in the crystal lattice, respectively.^65,66 The peak area of O_lat was higher than that of O_ads, which resulted in higher catalytic activity in the transesterification of WCO using the BZO catalyst.⁶⁷ Fig. 5(d) represents the survey spectra of the 11 BZO catalyst in XPS analysis.

Fig. 5 XPS spectra of (a) Ba 3d, (b) Zn 2p, and (c) O 1s, and (d) survey scan of the 11 BZO catalyst.

3.2 Optimization of various parameters affecting biodiesel production

Optimizing biodiesel production parameters is essential for maximizing yield, improving fuel properties, ensuring economic viability, and promoting sustainable and cost-effective production. Consequently, the optimization of both catalyst synthesis parameters and transesterification reaction parameters was studied, and the results obtained are discussed below.

3.2.1 Influence of Ba to Zn atomic ratio and catalyst calcination temperature. The Ba to Zn stoichiometric ratio and catalyst calcination temperature must be optimized to obtain the maximum catalytic activity of the BaO/ZnO nanocatalyst in the transesterification reaction. Three different catalysts (23 BZO, 11 BZO, and 32 BZO) were prepared by varying the stoichiometric ratio of Ba and Zn. Fig. 6(a) shows that the WCOME conversion increases from 82.5% using 23 BZO (Ba : Zn 2 : 3) to 97.6% using the 11 BZO (Ba : Zn 1 : 1) catalyst under the reaction conditions of 2.5 wt% catalyst loading using a 12 : 1 MeOH to oil molar ratio at a temperature of 40 °C for 30 min. The maximum WCOME conversion using 11 BZO demonstrates the highest catalytic efficiency among all BZO catalysts, possibly due to the synergistic effect of excellent basicity and high surface area. Furthermore, the reduction in WCOME conversion was found utilizing the 32 BZO catalyst (Ba : Zn 3 : 2). This was because of its low surface area and poor basicity due to some amount of BaCO₃ formed as an inactive phase. Therefore, the catalyst having a Ba to Zn stoichiometric ratio of 1 : 1 was regarded to be the most effective one compared to others in biodiesel production.

Fig. 6 Influence of reaction parameters on transesterification: (a) Ba to Zn atomic ratio (MeOH to oil molar ratio = 12 : 1, catalyst loading = 2.5 wt%, temperature = 40 °C, time = 30 min), (b) catalyst calcination temperature (MeOH to oil molar ratio = 12 : 1, catalyst loading = 2.5 wt%, temperature = 40 °C, time = 30 min), (c) reaction temperature (MeOH to oil molar ratio = 12 : 1, catalyst loading = 2.5 wt%, time = 30 min), and (d) reaction time (MeOH to oil molar ratio = 12 : 1, catalyst loading = 2.5 wt%, temperature = 40 °C).

On the other hand, activating the uncalcined BZO catalyst at an appropriate temperature is crucial for creating the active phase of BaO, which drives proton abstraction from MeOH and consequently leads to biodiesel formation. The calcination temperature for catalyst preparation was varied from 600 °C to 900 °C. A set of batch experiment was performed under the aforementioned reaction conditions, followed by measuring the WCOME conversion. It is worth mentioning that biodiesel conversion increases from 600 °C to 800 °C and further decreases at 900 °C, as shown in Fig. 6(b). The reduced catalytic efficacy at very high calcination temperatures is likely due to the sintering effect and blockage of the active sites.⁶⁸ The maximum WCOME conversion of 97.6% was obtained using the 11 BZO catalyst activated at 800 °C. Thus, 800 °C was considered the optimal calcination temperature for the catalyst preparation.

3.2.2 Influence of reaction temperature and time. Temperature has a significant influence on the transesterification reaction rate. For conveying the catalytic efficacy of the synthesized 11 BZO catalyst in biodiesel production, the transesterification reaction was executed by varying the temperature from 25 °C to 45 °C. The results illustrated in Fig. 6(c) reveal that the WCOME conversion was comparatively lower (65.7%) at 25 °C and increased to 40 °C with a maximum conversion of 97.6%. Beyond 40 °C, no further appreciable increase in biodiesel conversion was observed, suggesting that achieving optimum kinetic energy for effective collisions among reactant molecules leads to product formation.⁶⁹

Reaction time is also one of the crucial parameters that must be optimized to obtain maximum biodiesel conversion at a minimum time. Therefore, the reaction time was increased from 15 to 35 min, keeping all other parameters at optimized conditions (2.5 wt% catalyst loading, 12 : 1 MeOH to oil molar ratio, 40 °C temperature). Allowing a long time results in improved mass transfer and an effective contact period among all the reactant molecules. Fig. 6(d) demonstrates that up to 30 min, a gradual increase in WCOME was observed, and the highest conversion of 97.6% was achieved at 30 min. Beyond that, the biodiesel conversion remains almost unchanged, indicating that the reaction has achieved equilibrium conditions.⁷⁰

3.3 RSM modeling for the optimization of biodiesel conversion

3.3.1 Box–Behnken Design (BBD). The optimization of transesterification reaction parameters was executed using the response surface methodology (RSM) approach in the Box–Behnken design (BBD) model. According to the predicted statistical BBD model, 30 batch experiments were conducted. The results obtained are shown in Table S2, which corresponds to the WCOME conversion ranging from 62.24% to 98.14%. The resulting data was put in the Design Expert software, and the statistical model predicted whether it followed linear, 2FI, cubic, or quadratic polynomials.⁷¹ According to the fit summary, the quadratic model was recommended as the best-suited model for optimizing the WCOME conversion (Table S3). The quadratic polynomial equation for the predicted WCOME conversion with the coded factors is illustrated in eqn (14).

WCOME conversion (%) = +97.37 + 0.3383P + 0.5875Q + 12.90R + 5.02S + 0.79PQ − 0.7725PR + 1.21PS − 1.38QR + 0.58QS + 3.15RS − 4.11P² − 1.43Q² − 15.83R² − 4.73S² (14)

where P, Q, R and S are the model variables for the MeOH to oil molar ratio, temperature, catalyst loading, and time, respectively.

Analysis of variance (ANOVA) was employed to assess the impact of various regression model variables for obtaining maximum biodiesel conversion. The outcomes of ANOVA studies related to the WCOME conversion are illustrated in Table S4. The significance of the proposed quadratic model and each constituent variable impacting model responses is analyzed using the regression model coefficients, such as F-value and p-value.³⁵ As mentioned in Table S5, the F-value for the selected model is 908.12, along with the p-value of <0.0001, which is lower than 0.05. These results elucidate that the proposed quadratic model is statistically significant, and the lack of fit value is insignificant. The model terms Q, R, S, PQ, PR, PS, QR, RS, P², Q², R², and S² having high F-values and p-values <0.05 are found to be significant and terms P and QS have lower F-values of 4.20 and 4.12, in addition to the higher p-values of 0.0582 (>0.05) and 0.0605 (>0.05), respectively.⁷² This clarifies that these two factors hold diminished importance in the biodiesel production from WCO. The fit statistics data in Table S5 demonstrate that the R² value is 0.9988, which justifies the model fitting. The predicted R² of 0.9938 is in reasonable agreement with the Adjusted R² of 0.9977, as the difference is less than 0.2. The obtained low value of standard deviation (0.5716) ensures a good fitting of the proposed model and demonstrates satisfactory concordance between the experimental and model-predicted data. The adequate precision value for the proposed model is 88.655 (a ratio of more than 4 is desirable), indicating an acceptable signal-to-noise ratio and offering a good degree of precision. Furthermore, a low coefficient of variation (CV) of 0.6575% (<10%) indicates a significant correlation between the obtained and predicted outcomes.⁷³

The actual and predicted responses are close to a straight line with a regression coefficient value, R² = 0.9988, establishing an excellent agreement between the experimentally observed and predicted WCOME conversion, as shown in Fig. S1(a).⁷⁴ It is important to determine whether the residuals adhere to a normal distribution to assess the suitability of the regression model. Fig. S1(b) shows that all the data points align along a regression line in the residuals plot of normal % probability, indicating the uniform distribution of the residuals.⁷⁵ The plot of residuals versus predicted WCOME conversion illustrated in Fig. S1(c) demonstrates that the data responses are arbitrarily scattered across a reference line, suggesting the goodness of the model's prediction for biodiesel synthesis from WCO. In Fig. S1(d), the plot of externally studentized residuals against run number, it is clear that all the WCOME conversion responses predicted by the proposed model fall within the ±4% error range compared to the experimentally obtained values.⁷⁶ This indicates that the regression model has high accuracy and a strong foundation for predicting biodiesel conversion precisely under arbitrary experimental conditions. The correlation between all the process parameters affecting WCOME conversion can be illustrated by the perturbation plot (Fig. S2), where all tested variables are set at a middle value (P = 12 : 1, Q = 40 °C, R = 2 wt%, S = 30 min). The impact of each parameter on biodiesel production was investigated while keeping the other three at constant values. The results are represented by different line curves corresponding to each variable. According to Fig. S2 and the ANOVA findings provided in Table S4, factor R (catalyst quantity) is the most significant parameter for the WCOME conversion. Additionally, factor S (time) also has a considerable influence. In contrast, factor P (MeOH to oil molar ratio) demonstrates the least effect on biodiesel conversion from WCO.

3.3.2 Assessment of the interaction effect of various reaction parameters using the 3D response surface plot. Three-dimensional response surface plots have been used to investigate the complementary effect of two independent reaction parameters on WCOME conversion, keeping other parameters constant at their median value, as shown in Fig. 7. The interaction effect of both the MeOH to oil molar ratio and temperature on WCOME conversion was examined, as demonstrated in Fig. 7(a). In this analysis, the catalyst loading and time were held constant at the center values of 2 wt% and 30 min, respectively. It was observed that WCOME conversion increases to a lower extent with a simultaneous increase in the molar ratio and time. It reached 98.14% conversion at the center value of 2 wt% catalyst loading and 30 min reaction time. Moreover, the flat nature of the plot suggests that the combined effect of the MeOH to oil molar ratio and temperature did not significantly influence biodiesel conversion.⁷⁷ This is evident by the fact that factor P (MeOH to oil molar ratio) has the least impact on WCOME conversion, indicated by its insignificant p-value of 0.0582 (>0.0001) and a very low F-value of 4.20, as mentioned in the ANOVA result (Table S5). Fig. 7(b) illustrates the combined impact of catalyst loading and MeOH to oil molar ratio on WCOME conversion under fixed conditions of 40 °C reaction temperature and 30 min reaction time. The WCOME conversion had increased drastically with increasing both catalyst loading and MeOH to oil molar ratio, showing the synergistic influence of both parameters towards biodiesel production. A higher amount of catalyst provides more readily accessible basic sites for proton abstraction from MeOH, consequently enhancing biodiesel conversion. Furthermore, with the increment of the MeOH to oil molar ratio, the concentrations of reactant molecules increase, thus shifting the overall equilibrium toward the product side according to Le Chatelier's principle and resulting in a higher WCOME conversion. The maximum 98.14% WCOME conversion was achieved under the foremost conditions of 2 wt% catalyst loading in a 12 : 1 MeOH to oil molar ratio. However, increasing the catalyst loading above 2 wt% creates higher viscosity due to agglomeration of the catalyst, improper mixing of reactant molecules, and availability of lower active basic sites for converting WCO into biodiesel.⁷⁸ Furthermore, excess methanol results in emulsion formation between unreacted oil, MeOH, and freshly converted biodiesel.⁷⁹ These contribute towards a slight decrease in WCOME formation. The cumulative effect of MeOH to oil molar ratio and time on WCOME was assessed keeping the catalyst loading and temperature constant at 2 wt% and 40 °C, respectively [Fig. 7(c)]. The augmentation of reaction time substantially affected WCOME conversion; however, raising the MeOH to oil molar ratio did not improve biodiesel conversion because greater WCOME conversion was attained at a comparably lower MeOH to oil molar ratio. This is also supported by ANOVA results, which show that the MeOH to oil molar ratio has the highest p-value of 0.0582 (>0.05) and lowest F-value (4.20) among all experimental parameters, making it the least significant factor in biodiesel production. Fig. 7(d) portrays the interaction effect of catalyst loading and temperature. With an increase in both parameters at a fixed value of 12 : 1 MeOH to oil molar ratio and 30 min time, WCOME conversion increases appreciably. However, the simultaneous increase in catalyst loading and reaction temperature results in no significant increase in the biodiesel conversion after 2 wt% catalyst loading and 40 °C temperature. This implies that these two parameters significantly influence WCOME conversion at lower catalyst loading and temperature values than at higher values. In Fig. 7(e), the influence of reaction time and temperature on WCOME conversion was studied while keeping a constant MeOH to oil molar ratio of 12 : 1 and catalyst loading of 2 wt%. The biodiesel conversion was boosted by the elevation of both time and temperature. However, after a certain point, when the equilibrium conditions were achieved, WCOME conversion did not increase substantially with an increase in both parameters. Fig. 7(f) shows the complementary effect of time and catalyst loading on WCOME conversion at a temperature of 40 °C and MeOH to oil molar ratio of 12 : 1. The WCOME conversion was greatly enhanced by increasing time, and the maximum conversion of 98.14% was obtained at 30 min; after that, conversion slightly declined maybe due to the reversibility of the transesterification reaction.⁸⁰ With increasing catalyst concentration from 1 to 2 wt%, biodiesel conversion significantly increased due to the higher basicity of the catalyst, and conversion decreased beyond 2 wt%. Thus, time and catalyst loading synergistically affected biodiesel production from WCO.

Fig. 7 3D surface plots on the BBD model with the interaction effect of (a) temperature and MeOH to oil molar ratio, (b) catalyst loading and MeOH to oil molar ratio, (c) time and MeOH to oil molar ratio, (d) catalyst loading and temperature, (e) time and temperature, and (f) time and catalyst loading.

3.3.3 Numerical optimization of biodiesel conversion. The numerical optimization method was employed in the RSM–BBD model to determine the ideal parametric conditions under which the maximum biodiesel conversion was obtained. The following regression model predicted a WCOME conversion of 98.15% under the optimum conditions of MeOH to oil molar ratio of 10.9 : 1, temperature of 39.9 °C, catalyst loading of 2.6 wt%, and reaction time of 29.8 min with a desirability factor of 1.000 as depicted in Fig. S3. The experimentally obtained WCOME conversion using the proposed conditions was 97.3%. The 0.85% difference between the predicted and actual WCOME conversion indicates that the proposed quadratic statistical model was highly accurate and effectively justified in predicting the optimal conditions for achieving maximum biodiesel conversion.

3.4 Kinetic and thermodynamic studies of the transesterification process

A series of transesterification experiments with temperatures ranging from 25 to 40 °C were conducted to assess the kinetic and thermodynamic parameters involved in WCOME conversion from triglycerides utilizing the 11 BZO catalyst. During this study, each set of reactions at a fixed temperature was performed under varying reaction times from 10 to 30 min at the constant catalyst loading of 2.0 wt% and MeOH to oil molar ratio of 12 : 1. The extent of WCOME conversion, X_WCOME, was calculated from the obtained biodiesel conversion value at 5-minute intervals and put in eqn (7). The plot of −ln(1 − X_WCOME) against time (t) at four different temperatures (25 °C, 30 °C, 35 °C and 40 °C) gives straight lines and their slopes were used to calculate the rate constant (k) values at their respective temperatures as shown in Fig. 8(a). The rate constant value was obtained at 0.0446 min⁻¹ at 25 °C and increased to 0.0958 min⁻¹ at 40 °C, as illustrated in Table 3. From the above findings, it is anticipated that the transesterification reaction rate mainly depends on oil concentration. Furthermore, the regression coefficient (R²) values of the corresponding straight lines in the kinetic plot were close to unity (0.98–0.99), demonstrating the pseudo-first-order kinetics for the methanolysis reaction of WCO. The activation energy (E_a) and pre-exponential factor (A) values for the transesterification reaction were determined from the slope and intercept values of the Arrhenius plot of ln(k) versus 1/T, as shown in Fig. 8(b). The activation energy and pre-exponential factor were determined to be 37.64 kJ mol⁻¹ and 1.74 × 10⁵ min⁻¹, respectively. The obtained energy of activation value is consistent with the previously reported values ranging from 31.2 to 85.8 kJ mol⁻¹.^81,82

Fig. 8 The kinetic study using (a) the plot of −ln(1 − X_WCOME) vs. time (min), (b) Arrhenius plot of ln(k) vs. 1/T (K⁻¹), the thermodynamic study using (c) the Eyring–Polanyi plot of ln(k/T) vs. 1/T (K⁻¹), and (d) heterogeneity test of the 11 BZO catalyst for the production of WCOME.

Table 3 Kinetic and thermodynamic parameters for the synthesis of biodiesel from WCO using the 11 BZO catalyst

Temperature (K)	k (min^−1)	ΔG^# (kJ mol^−1)	ΔH^# (kJ mol^−1)	ΔS^# (J mol^−1 K^−1)
298	0.0446	80.67	35.76	−150.73
303	0.0605	81.43
308	0.0702	82.18
313	0.0958	82.93

---

Thermodynamic parameters such as the enthalpy of activation (ΔH^#) and the entropy of activation (ΔS^#) were calculated using the Eyring–Polanyi plot of ln(k/T) versus 1/T shown in Fig. 8(c). From the slope of the ln(k/T) versus 1/T plot, the enthalpy of activation (ΔH^#) value was calculated to be 35.76 kJ mol⁻¹. The positive sign ΔH^# value indicates that the biodiesel production process is endothermic. That means heat is essential for carrying out the transesterification reaction of WCO. Similarly, the entropy of activation (ΔS^#) was evaluated using the intercept value of the ln(k/T) versus 1/T plot, and it was found to be −150.73 J mol⁻¹ K⁻¹. The negative ΔS^# value signifies that the process is entropically unfavorable and follows an associative mechanism in which triglycerides and methanol molecules come close to each other to form a tetrahedral intermediate in the transition state associated with the transformation of WCO into biodiesel (Scheme 2).⁸³ Furthermore, the Gibbs free energy (ΔG^#) was evaluated by substituting the value of ΔH^# and ΔS^# in the Gibbs–Helmholtz equation [eqn (11)]. The values of ΔG^# at 25 °C, 30 °C, 35 °C, and 40 °C were observed to be 80.67, 81.43, 82.18, and 82.93 kJ mol⁻¹, respectively (Table 3). The positive value of the Gibbs free energy indicates that the transesterification process of biodiesel synthesis is non-spontaneous.

Scheme 2 Sequential mechanistic steps for the transesterification reaction of WCO using the 11 BZO nanocatalyst.

3.5 Heterogeneity study of the synthesized catalyst

The heterogeneity study of the synthesized 11 BZO catalyst was conducted by performing a leaching experiment using the hot Sheldon filtration method.⁸⁴ In this process, a methanolysis reaction of WCO was conducted under the optimal reaction conditions. After 10 minutes of reaction, the catalyst was filtered out from the hot reaction mixture, and the WCOME conversion was measured, which was found to be 38.6% during that time. The reaction was then continued for an additional 50 minutes without the catalyst. It was observed that there was only a 1.47% increase in the biodiesel conversion [Fig. 8(d)]. This suggests that no significant quantity of active components of the catalyst was leached into the product mixture during the reaction period. Consequently, the synthesized 11 BZO catalyst exhibited its heterogeneous characteristics in the generation of biodiesel from WCO.

3.6 ¹H and ¹³C NMR analysis of waste cooking oil and synthesized biodiesel

The structural identification of WCO and the synthesized biodiesel was carried out through NMR analysis. To confirm the formation of biodiesel, the ¹H and ¹³C NMR signals of both the reactant and product were compared and verified as shown in Fig. S4 and S5. The characteristic peaks obtained in the ¹H NMR spectra at the chemical shift (δ) values of 0.85 to 1.29 (m) ppm correspond to –CH₂–CH₃ protons. Signals at δ values of 1.62 (m), 2.04 (m), 2.31 (t), and 2.76 (t) ppm are attributed to =CH–CH₂, –CH₂–CH=CH, α-CH₂ (methylene protons), and =CH–CH₂–CH= protons, respectively. Other characteristic peaks around 4.1 to 4.3 (d–d) were observed due to the –CH₂–O and –CH–O protons of triglyceride backbones as depicted in Fig. S4(a). Additionally, peaks at δ values of 5.25 to 5.38 (m) originated due to –CH=CH– (olefinic protons).^85,86 The absence of a characteristic ¹H NMR signal between 4.1 and 4.3 ppm, along with a new singlet peak of high intensity at 3.68 ppm due to –OCH₃ (methoxy) protons in the synthesized product, confirms the successful synthesis of waste cooking oil methyl ester, i.e., biodiesel [Fig. S4(b)]. Similarly, in the ¹³C NMR spectrum of WCO illustrated in Fig. S5(a), the major characteristic ¹³C signals at 62.2 and 68.8 ppm are attributed to glyceride carbons (–CH₂–O and –CH–O), which were absent in the ¹³C NMR spectrum of the product [Fig. S5(b)]. Furthermore, a new peak appears at 51.5 ppm due to the methoxy (–OCH₃) carbon, confirming the effective synthesis of biodiesel from WCO.⁸⁷ Moreover, the signal at 173.24 ppm is attributed to the ester carbonyl carbon in both samples.

3.7 Mechanistic investigation of the transesterification reaction of WCO using the 11 BZO nanocatalyst

The mechanistic pathway for the methanolysis reaction of waste cooking oil utilizing the 11 BZO catalyst is provided based on previously published literature. The mechanism of the following transesterification reaction has been portrayed using the Langmuir–Hinshelwood (L–H) model, which goes through several sequential steps as illustrated in Scheme 2. This mechanism involves the diffusion of both triglyceride and methanol molecules on the catalyst surface and adsorption into the different active sites of the catalyst. In the first step, adsorbed methanol undergoes deprotonation to form a reactive methoxide ion. Notably, the Ba–O bond in the 11 BZO catalyst plays a significant catalytic role in the deprotonation of MeOH. The carbonyl group of triglyceride molecules interacts with the catalytic sites, and as a consequence, the electrophilicity of the carbonyl carbon is increased. Subsequently, in the second step, methoxide (⁻OCH₃) attacks the carbonyl carbon center and forms a tetrahedral intermediate, which breaks down into biodiesel and diglycerides.⁸⁸ This process repeats in the third step and leads to the formation of monoglycerides and biodiesel. Finally, the monoglycerides are also attacked by ⁻OCH₃ ions in the same fashion to generate biodiesel and by-product glycerol.

3.8 Recyclability study of the catalyst

Catalyst recyclability plays a vital role in the transesterification reaction of biodiesel synthesis in terms of economic viability, environmental sustainability, and industrial scalability.⁸⁹ In this process, several catalytic cycles of biodiesel production were repeated using the recycled 11 BZO catalyst in a batch reaction mode under the optimized reaction conditions of 2.6 wt% catalyst loading using a 10.9 : 1 MeOH to oil molar ratio at 39.9 °C for 29.8 min. After each cycle, the catalyst was separated from the product mixture, thoroughly washed 5–6 times with methanol and n-hexane, and finally kept in a hot air oven overnight for drying. Furthermore, the recycled catalyst was activated by heating in a muffle furnace at 800 °C for 2 h to decompose any adsorbed organic moieties at the catalytic sites and unwanted carbonate species. According to Fig. 9(a), it is observed that the catalytic activity of 11 BZO was almost completely retained up to the third cycle, with a WCOME conversion of 90.8%. Further substantial reduction in biodiesel conversion was observed from the third to fifth cycles, and 82.1% WCOME conversion was obtained after the fifth catalytic cycle. This can be explained by the evidence of XRD spectra of the 11 BZO catalyst after five cycles, demonstrating similar XRD patterns to that of the fresh 11 BZO catalyst and having all the peaks with relatively lower intensities compared to the fresh catalyst, which may be due to the leaching of active components during each catalytic cycle [Fig. 9(b)]. To quantify the concentration of leached metal ions, an ICP analysis of biodiesel samples was performed, which shows that after the first cycle, leaching of Ba²⁺ and Zn²⁺ ions into biodiesel was not severe. However, in the third cycle onwards, the extent of leaching was drastically enhanced, and after the fifth cycle, the total concentration of Ba²⁺ and Zn²⁺ leached into the biodiesel sample was found to be 120.4 mg L⁻¹ (Table S10). Besides this, the existence of carbonate species as observed in both the XRD patterns and FTIR spectra of the recycled catalyst could be another reason for lowering the catalyst's basicity, and consequently decreasing WCOME conversion in each cycle. Moreover, the active sites of the reused catalyst were blocked by triglyceride molecules, as confirmed by the FTIR spectra of the reused catalyst (Fig. S6). The HR-SEM image of the reused catalyst also shows a similar surface topology to that of the fresh catalyst, and the EDX spectrum demonstrates the existence of carbonate species in the recycled catalyst (Fig. S7). Furthermore, in Fig. S8, the BET isotherm of the recycled 11 BZO catalyst reveals that the specific surface area was decreased from 6.0 m² g⁻¹ in the fresh catalyst to 4.2 m² g⁻¹ in the reused catalyst, and also the pore diameter was reduced from 36.5 nm to 24.6 nm, as illustrated in Table S11. This reduction may be attributed to the agglomeration of catalyst particles or blockage of pores by impurities like unreacted oil and by-product glycerol. Moreover, to investigate the reason behind deteriorating stability in the reused catalyst, XPS analysis of the recycled 11 BZO catalyst was executed as demonstrated in Fig. S9. Both deconvoluted spectra of O 1s and C 1s indicate the presence of inactive BaCO₃, and the shifts in binding energy values of peak positions suggest the attachment of impurities at the catalyst surface and thereby a reduction in catalytic stability after five reuse cycles.

Fig. 9 (a) Recyclability of the 11 BZO catalyst in the transesterification of WCO and (b) XRD pattern of the 11 BZO catalyst after five cycles.

3.9 Economic cost analysis of biodiesel production

According to various reports, the total 60–70% cost of biodiesel production depends on the selection of feedstocks. Besides this, the cost of catalyst preparation also contributes to the overall cost of biodiesel production. The factors that affect the price of biodiesel production include the amount of chemicals, electricity consumption, and manpower.⁹⁰ Here, we utilized waste cooking oil as a feedstock, which was collected from the hostel mess at IIT(BHU) free of cost. Therefore, the feedstock's cost is considered to be zero. The amount and price of all the used chemicals as well as the energy consumption for producing 1 L of biodiesel are illustrated in Table S6. To synthesize 1 g of 11 BZO catalyst, 1.14 g of barium nitrate (0.0028 USD), 0.408 g of sodium carbonate (0.0005 USD), and 1.14 g of zinc nitrate hexahydrate (0.0028 USD) were utilized. In the transesterification reaction of WCO, 2.6 wt% catalyst was utilized. Thus, 25 g of 11 BZO was required, and the total cost of catalyst preparation was found to be 0.152 USD. Moreover, the methanol-to-oil molar ratio of 10.9 : 1 suggests that 0.46 L of methanol was employed for 1 L of WCOME generation, which costs around 0.25 USD. Nevertheless, 50 mL of n-hexane and 40 mL of MeOH were used for washing in the catalyst's regeneration process. This came to a cost of 0.1 USD. Moreover, the total cost of electricity consumption, including the heating process in the muffle furnace and stirring in the magnetic stirrer, was calculated to be 0.28 USD. Consequently, the total cost of 1 L biodiesel production from WCO utilizing 11 BZO came to only 0.782 USD. It is worth mentioning that the excess methanol can be recovered and reused. The proposed catalyst is also reusable for up to five cycles. Considering these two facts, the overall biodiesel production cost could be reduced to below 0.782 USD.

3.10 Comparative study of green metrics and turnover frequency of catalytic cycles

Studying green parameters in chemical reactions is essential to assessing the sustainability, effectiveness, and environmental impact of various chemical processes. Moreover, these green metrics align with green chemistry principles, which aim to minimize waste generation, diminish the use of hazardous chemicals, and design a safer and cleaner process.⁹¹ The current study deals with two crucial green parameters, including the environmental factor (E-factor) and process mass intensity (PMI), which are examined to assess the ‘greenness’ compatibility in the biodiesel production process using the 11 BZO catalyst. E-factor is defined as the amount of waste generated relative to the amount of product, as illustrated in eqn (15).⁹²

The E-factor value for the following transesterification reaction was 0.14, which falls within the range of 0–1, indicating a lower environmental impact and minimal waste generation during the process. Likewise, the PMI refers to the total amount of materials utilized in a reaction relative to the amount of desired product, as mentioned in eqn (16).⁹³

The calculated value of PMI for this methanolysis reaction was 1.43, indicating the sustainable production of biodiesel from WCO using the 11 BZO catalyst. Interestingly, the E-factor value increases in every catalytic cycle, which could be due to the increase in excess methanol and unreacted oil and the decrease in biodiesel production. However, excess methanol can be recovered, and by-product glycerol can be valorized into valuable products such as solketal, glycerol carbonate, and other derivatives. Similarly, the PMI value gradually increases from 1.57 in the 1st cycle to 3.45 in the 5th cycle, possibly due to decreasing biodiesel yield in each cycle (Table S7). A comparative literature study, as illustrated in Table S9, demonstrates that the obtained values of E-factor and PMI are lower compared to those of other reported processes, which indicates that the following biodiesel production process utilizing the 11 BZO catalyst is clean and environmentally friendly. Turnover frequency (TOF) is another crucial parameter that helps to determine how fast a catalyst facilitates a reaction. In the transesterification reaction, TOF can be defined as the number of triglycerides converted into biodiesel per basic site of the catalyst per unit time, as shown in eqn (17).⁹⁴

The TOF value of the 11 BZO catalyst in WCOME production was calculated to be 15.52 h⁻¹, comparable to that of reported homogeneous catalysts used in the transesterification reaction. High TOF makes 11 BZO an efficient catalyst for biodiesel production from WCO. However, a gradual decrease in TOF value during each catalytic cycle was observed, indicating a loss of catalytic efficiency for the 11 BZO catalyst after each run.

3.11 Comparative study on the performance of the 11 BZO catalyst with the reported catalysts

In the current study, the catalytic efficacy of the BaO/ZnO (BZO) catalyst in the transesterification of biodiesel production has been compared with that of other previously reported barium-based heterogeneous basic catalysts, as shown in Table 4. Previously reported catalysts, including BaCeO₃,⁹⁵ Cs modified BaZrO₃,⁹⁶ and barium lanthanum oxide,⁹⁷ utilize a high methanol-to-oil molar ratio to produce biodiesel through the transesterification reaction. Likewise, the utilization of other catalysts, such as BaO-rGO,⁹⁸ ZSM-5 zeolite-supported BaO,⁹⁹ and Ba₂SiO₄,¹⁰⁰ leads to the completion of the transesterification reaction in a longer time. Al-Abbasi et al. generated biodiesel from sunflower oil and achieved a comparatively lower yield of 78.38% utilizing a 4.7 wt% BaO catalyst and a higher MeOH to oil molar ratio of 20 : 1 at a higher temperature of 70 °C under a very long duration of 240 min.¹⁰¹ Similarly, Balakrishnan et al. synthesized biodiesel from waste cooking oil using a Ba/CaO catalyst. They obtained 88% yield under optimal experimental conditions of 3 wt% catalyst loading, 9 : 1 MeOH to oil molar ratio, 65 °C temperature, and 180 min reaction time.¹⁰² In our study, the 11 BZO catalyst exhibited excellent catalytic activity in the methanolysis reaction of WCO and showed 97.3% biodiesel conversion only at 29.8 min. In addition, using a minimal catalyst loading (2.6 wt%) and a MeOH to oil molar ratio (10.9 : 1) renders the process economically viable. Moreover, the following reaction proceeded at a lower temperature of 39.9 °C, making the process economically viable and energy efficient. Thus, the proposed 11 BZO catalyst is highly efficient and catalytically active compared to other reported catalysts for low-temperature biodiesel production from waste cooking oil.

Table 4 Comparative performance analysis of the 11 BZO catalyst with other reported barium-based heterogeneous catalysts in biodiesel synthesis

Catalyst	Preparation method	Feedstock	Transesterification reaction conditions				Yield (Y) or conversion (C) (%)	Cycle no.	Ref.
			MeOH to oil molar ratio	Catalyst loading (wt%)	Temperature (°C)	Time (min)
BaCeO3	Sol–gel	Karanja oil	19 1	1.2	65	100	98.4 (C)	6	95
Cs modified BaZrO3	Solid-state reaction	MillettiaPinnata oil	22.1 : 1	1.72	65	138	97.27 (C)	7	96
Ba/CaO	Co-precipitation	Waste cooking oil	9 : 1	3	65	180	88 (Y)	3	102
BaO-rGO	One-pot hydrothermal	Palm oil	12 : 1	5	65	180	88.43 (Y)	4	98
BaO	Sol–gel	Sunflower oil	20 : 1	4.7	70	240	78.38 (Y)	5	101
ZSM-5 zeolite-supported BaO	Hydrolysis	Yellow grease	15 : 1	2	70	180	95.9 ± 0.94 (Y)	6	99
BaO–MoO2	Co-precipitation	Balanites aegyptiaca seed oil	12 : 1	4.5	65	120	97.8 (Y)	6	103
Barium tin oxide@rGO	Wet-impregnation	Waste cooking oil	14.93 : 1	3.21	68.83	27.95	97.03 (C)	6	104
Ba2SiO4	Solid state mixing	Anabaena oil	12 : 1	4	65	240	98.4 (C)	6	100
Barium lanthanum oxide	Sol–gel auto-combustion	Madhuca indica oil	18 : 1	1.5	65	120	97 ± 0.5 (C)	5	97
BaO/ZnO (11 BZO)	Wet-impregnation	Waste cooking oil	10.9 : 1	2.6	39.9	29.8	97.3 (C)	5	Present work
							97.1 (Y)

---

3.12 Study of the physicochemical properties of the synthesized biodiesel

The suitability of WCOME in CI engines was examined by evaluating the synthesized biodiesel's physicochemical properties. The obtained values of various fuel characteristics of WCO-derived biodiesel are verified by comparing them with the American Society for Testing and Materials (ASTM) D-6751 standard limits. Biodiesel fuel properties, such as acid value, kinematic viscosity at 40 °C, density, carbon residue, flash point, pour point, cloud point, and calorific value, are assessed and listed in Table S8. The acid value of WCOME is 0.28, which is within the ASTM standard limits. Two major physical characteristics, density and kinematic viscosity at 40 °C, have values of 0.864 g cm⁻³ and 5.2 mm² s⁻¹, respectively. The good agreement of density and kinematic viscosity values with ASTM D-6751 suggests the compatibility of the biodiesel fuel in the proper functioning of various engine parts during the combustion process. Furthermore, the flash point of the produced biodiesel is 143 °C (>130 °C), which is relatively higher than that of conventional diesel. The high flash point value indicates the higher thermal stability of the fuel during its storage and transportation. Moreover, the cold-flow properties, such as pour and cloud points, demonstrate the suitability of using biodiesel in engines in cold climates. Pour and cloud points of WCOME are found to be −3 °C and 2 °C, respectively, which are within the acceptable limits of the ASTM standard. The carbon residue value is only 0.01% of the total mass. In addition, the calorific value of the produced biodiesel is 37.61 MJ kg⁻¹, which is quite close to that of conventional petrodiesel. Since the physicochemical properties of WCO-biodiesel adhere to ASTM standard limits, it is a viable replacement for petroleum diesel in internal combustion engines.¹⁰⁵

4 Conclusions

The current study demonstrates a sustainable and efficient route for biodiesel production from inexpensive and readily available waste cooking oil through low-temperature transesterification using the BaO/ZnO (BZO) nanocatalyst, optimized through RSM in the BBD method. Three BZO catalysts with Ba to Zn stoichiometric ratios of 2 : 3, 1 : 1, and 3 : 2 were synthesized. Diverse analytical studies such as XRD, FTIR, XPS, CO₂-TPD, and BET revealed that 11 BZO (Ba : Zn 1 : 1) possessed maximum catalytic activity in biodiesel production due to exhibiting stronger basicity, numerous catalytic sites, and improved surface area. Process optimization using RSM helped to evaluate optimized parametric conditions with an enhancement of conversion efficiency while minimizing resource input. A high biodiesel conversion of 97.3% was obtained at a low temperature of 39.9 °C in a very short period of 29.8 min. Kinetic and thermodynamic studies elucidate that the transesterification process of biodiesel production using the BZO catalyst follows an endothermic non-spontaneous pathway. Likewise, the proposed catalystexhibits excellent reusability with achieving 82.1% biodiesel conversion across five catalytic cycles, suggesting that ZnO provides superior support to BaO, making it highly stable and catalytically active in biodiesel generation from WCO. In conclusion, biodiesel production from waste-derived feedstocks using BaO/ZnO has emerged as an energy-efficient, economical, and greener approach, offering a dual advantage of waste mitigation and renewable energy generation. Future studies on developing magnetic catalysts to simplify the separation process and enhance long-term stability during recyclability studies, integrated with technological evaluations, will offer deeper insights into the commercialization of this sustainable approach.

Consent for publication

All authors consented to the publication of this research article.

Author contributions

Surajit Pradhan: conceptualization; methodology; investigation; resources; data acquisition; formal analysis; writing the original draft. Hiralal Pramanik: conceptualization; formal analysis; review and editing. Yogesh Chandra Sharma: conceptualization; methodology; formal analysis; review and editing; supervision. All the authors have read and approved the final manuscript.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d5se00875a.

Acknowledgements

The authors would like to acknowledge the central instrumental facilities at the Indian Institute of Technology (BHU), Varanasi, for providing catalyst characterization and product analysis, and the Rajiv Gandhi Institute of Petroleum Technology (RGIPT), Amethi, for providing facilities for CO₂-TPD analysis. The authors also want to thank the Central Discovery Centre (BHU) for providing the facilities for XPS analysis. S.P. also sincerely thanks the Ministry of Human Resource Development (MHRD), Govt. of India, for providing financial support.

References

1. D. Bousba, C. Sobhi, E. Zouaoui, K. Rouibah, A. Boublia, H. Ferkous, A. Haddad, A. Gouasmia, I. Avramova, Z. Mohammed, V. I. Parvulescu, K. K. Yadav, M. Hasan, M. M. S. Cabral-Pinto, N. Elboughdiri and Y. Benguerba, Energy Convers. Manag., 2024, 300, 118021.
2. S. A. Padder, R. Khan and R. A. Rather, Biomass Bioenergy, 2024, 185, 107220.
3. A. H. Abu-Ghazala, H. H. Abdelhady, A. A. Mazhar and M. S. El-Deab, Fuel, 2023, 351, 128897.
4. G. S. Medeiros Leão, M. D. Silva Ribeiro, R. L. d F. Filho, L. B. Saraiva, R. R. Peña-Garcia, A. P. d C. Teixeira, R. M. Lago, F. A. Freitas, S. de Sá Barros, S. D. Junior, Y. L. Ruiz and F. X. Nobre, ACS Appl. Mater. Interfaces, 2024, 44, 60103–60121.
5. I. B. Laskar, K. Rajkumari, R. Gupta, S. Chatterjee, B. Paul and L. Rokhum, RSC Adv., 2018, 8, 20131–20142.
6. S. Ao, S. P. Gouda, M. Selvaraj, R. Boddula, N. Al-Qahtani, S. Mohan and S. L. Rokhum, Energy Convers. Manag., 2024, 300, 117956.
7. V. Das, A. M. Tripathi, M. J. Borah, N. T. Dunford and D. Deka, Renewable Energy, 2020, 161, 1110–1119.
8. G. T. Jeong, H. S. Yang and D. H. Park, Bioresour. Technol., 2009, 100, 25–30.
9. J. Kaur, A. K. Sarma, M. K. Jha and P. Gera, Biotechnol. Rep., 2020, 27, e00487.
10. G. Pradhan and Y. C. Sharma, Fuel, 2021, 284, 118966.
11. S. Maurya and Y. C. Sharma, Energy Fuels, 2024, 38, 14375–14386.
12. A. A. F. da Costa, A. d N. de Oliveira, R. Esposito, C. Len, R. Luque, R. C. R. Noronha, G. N. da Rocha Filho and L. A. S. Do Nascimento, Catalysts, 2022, 12, 570.
13. M. Athar and S. Zaidi, J. Environ. Chem. Eng., 2020, 8, 104523.
14. H. Hosseinzadeh-Bandbafha, C. Li, X. Chen, W. Peng, M. Aghbashlo, S. S. Lam and M. Tabatabaei, J. Hazard. Mater., 2022, 424, 127636.
15. N. Ghosh, M. Patra and G. Halder, Sustainable Energy Fuels, 2024, 8, 1105–1152.
16. A. Caporusso, M. Radice, A. Biundo, R. Gorgoglione, G. Agrimi and I. Pisano, J. Biotechnol., 2025, 400, 48–65.
17. Z. Yaakob, M. Mohammad, M. Alherbawi, Z. Alam and K. Sopian, Renew. Sustain. Energy Rev., 2013, 18, 184–193.
18. I. K. Hong, H. Jeon, H. Kim and S. B. Lee, J. Ind. Eng. Chem., 2016, 42, 107–112.
19. A. Larimi, A. P. Harvey, A. N. Phan, M. Beshtar, K. Wilson and A. F. Lee, Catalysts, 2024, 14, 701.
20. V. Cordero-Ravelo and J. Schallenberg-Rodriguez, J. Environ. Manage., 2018, 228, 117–129.
21. S. Y. Chua, L. A. Periasamy, C. M. H. Goh, Y. H. Tan, N. M. Mubarak, J. Kansedo, M. Khalid, R. Walvekar and E. C. Abdullah, J. Ind. Eng. Chem., 2020, 81, 41–60.
22. Y. Wang, D. Wang, M. Tan, B. Jiang, J. Zheng, N. Tsubaki and M. Wu, ACS Appl. Mater. Interfaces, 2015, 7, 26767–26775.
23. S. B. Monika and V. V. Pathak, Energ. Nexus, 2023, 10, 100209.
24. M. R. Khodadadi, I. Malpartida, C. W. Tsang, C. S. K. Lin and C. Len, Mol. Catal., 2020, 494, 111128.
25. A. H. Abu-Ghazala, H. H. Abdelhady, A. A. Mazhar and M. S. El-Deab, Chem. Eng. J., 2023, 474, 145784.
26. N. Abdul Rahim, M. Rosdzimin Abdul Rahman, K. Amali Bin Ahmad, A. Rahim Mat Sarip, S. Hasbullah and A. Suhel, Mater. Today: Proc., 2023 DOI:10.1016/j.matpr.2023.08.197.
27. B. Aghel, A. Gouran and P. Shahsavari, Chem. Eng. Technol., 2022, 45, 348–354.
28. S. Yan, H. Lu and B. Liang, Energy Fuels, 2008, 22, 646–651.
29. L. Du, Z. Li, S. Ding, C. Chen, S. Qu, W. Yi, J. Lu and J. Ding, Fuel, 2019, 258, 116122.
30. O. A. Mawlid, H. H. Abdelhady and M. S. El-Deab, Energy Convers. Manag., 2022, 273, 116435.
31. T. Maneerung, S. Kawi, Y. Dai and C. H. Wang, Energy Convers. Manag., 2016, 123, 487–497.
32. I. Veza, M. Spraggon, I. M. R. Fattah and M. Idris, Results Eng., 2023, 18, 101213.
33. A. N. Amenaghawon, K. Obahiagbon, V. Isesele and F. Usman, Clean Eng. Technol., 2022, 8, 100501.
34. V. Singh, L. Belova, B. Singh and Y. C. Sharma, Energy Convers. Manag., 2018, 174, 198–207.
35. S. K. Behera, H. Meena, S. Chakraborty and B. C. Meikap, Int. J. Min. Sci. Technol., 2018, 28, 621–629.
36. P. Mwenge, B. Djemima, S. Zwane, S. Muthubi, H. Rutto and T. Seodigeng, J. Environ. Sci. Health, 2024, 59, 499–511.
37. J. Mercy Nisha Pauline, R. Sivaramakrishnan, A. Pugazhendhi, T. Anbarasan and A. Achary, Fuel, 2021, 285, 119108.
38. P. Nautiyal, K. A. Subramanian and M. G. Dastidar, Fuel, 2014, 135, 228–234.
39. Y. Liu, J. Zhao, H. Zhang, Y. Zhu and Z. Wang, Thermochim. Acta, 2004, 414, 121–123.
40. A. Clarà Saracho, S. K. Haigh, T. Hata, K. Soga, S. Farsang, S. A. T. Redfern and E. Marek, Sci. Rep., 2020, 10, 10168.
41. C. J. Bardwell, R. I. Bickley, S. Poulston and M. V. Twigg, Thermochim. Acta, 2015, 613, 94–99.
42. P. Jayabal, S. Gayathri, V. Sasirekha, J. Mayandi and V. Ramakrishnan, J. Nanopart. Res., 2014, 16, 2640.
43. M. A. Kamran, S. Siddique, S. Ullah, T. Alharbi, M. Raza, M. Usama and B. Zou, J. Energy Storage, 2023, 67, 107598.
44. S. S. Abdul-Jabbar and K. H. Harbbi, AIP Conf. Proc., 2023, 3018, 020029.
45. M. Raessi, H. Q. Alijani, F. F. Nematollahi, R. S. Baty, G. El-Saber Batiha, A. U. Khan, N. Hashemi, S. Iravani, I. Sharifi, M. Aflatoonian and M. Khatami, Ceram. Int., 2021, 47, 21045–21050.
46. M. Haghighi, L. B. Zare and M. Ghiasi, Chem. Eng. J., 2022, 430, 132777.
47. Y. S. Erchamo, T. T. Mamo, G. A. Workneh and Y. S. Mekonnen, Sci. Rep., 2021, 11, 6708.
48. R. S. Brishti, M. Ahsan Habib, M. H. Ara, K. M. Rezaul Karim, M. Khairul Islam, J. Naime, M. M. Hasan Rumon and M. A. Rayhan Khan, Results Chem., 2024, 7, 101441.
49. M. A. Ansari and N. Jahan, Materials Highlights, 2021, 2, 23.
50. R. Singh, A. Kumar and Y. Chandra Sharma, Energy Fuels, 2019, 33, 1175–1184.
51. F. Chen, P. Zhang, L. Xiao, J. Liang, B. Zhang, H. Zhao, R. Kosol, Q. Ma, J. Chen, X. Peng, G. Yang and N. Tsubaki, ACS Appl. Mater. Interfaces, 2021, 13, 8191–8205.
52. B. Sreedhar, C. Satya Vani, D. Keerthi Devi, M. V. Basaveswara Rao and C. Rambabu, Am. J. Mater. Sci., 2012, 2, 5–13.
53. T. Li, Q. Yu, F. Barzagli, C. Li, M. Che, Z. Zhang and R. Zhang, Carbon Capture Sci. Technol., 2023, 6, 100099.
54. H. Fu, S. Sun and H. Lian, J. CO2 Util., 2023, 69, 102415.
55. T. Kamsuwan, C. Krutpijit, S. Praserthdam, S. Phatanasri, B. Jongsomjit and P. Praserthdam, Heliyon, 2021, 7, e07682.
56. S. Maurya and Y. Chandra Sharma, RSC Adv., 2024, 14, 39511–39522.
57. X. Li, S. Zhang, X. Jia, W. Li and J. Song, Catalysts, 2025, 15, 239.
58. Y. Sun, L. Chen, Y. Bao, Y. Zhang, J. Wang, M. Fu, J. Wu and D. Ye, Catalysts, 2016, 6, 188.
59. H. Yang, Y. Zhang, J. Zhou, Z. Wang, J. Liu and K. Cen, Int. J. Hydrogen Energy, 2016, 41, 3339–3348.
60. B. Changmai, C. Vanlalveni, A. P. Ingle, R. Bhagat and L. Rokhum, RSC Adv., 2020, 10, 41625–41679.
61. M. G. Ma, Y. J. Zhu, G. F. Cheng and Y. H. Huang, Mater. Lett., 2008, 62, 3110–3113.
62. Y. Pu, Y. Niu, Y. Wang, S. Liu and B. Zhang, Particuology, 2022, 61, 11–17.
63. T. Roy, S. Sahani, D. Madhu and Y. Chandra Sharma, J. Clean. Prod., 2020, 265, 121440.
64. M. Claros, M. Setka, Y. P. Jimenez and S. Vallejos, Nanomaterials, 2020, 10, 471.
65. Z. Xu, Z. Zhang, F. Gao, Y. Zhu and H. Zhao, Energy Mater., 2024, 4, 400076.
66. R. R. Leite, R. Colombo, A. J. Moreira, J. M. Farias, F. Eduardo Bimbi Júnior, J. B. Lima, M. Roberto de Vasconcelos Lanza, H. da Silva Barud and M. I. Basso Bernardi, Chemosphere, 2024, 369, 143888.
67. Y. Fan, P. Guo and J. Liu, Catalysts, 2025, 15, 299.
68. Z. Ren, H. Zhang, G. Wang, Y. Pan, Z. Yu and H. Long, ACS Omega, 2020, 5, 33357–33371.
69. R. Chauhan, H. Thakkar, M. Patel, S. Thakore and A. K. Prajapati, ACS Appl. Nano Mater., 2025, 8, 8076–8087.
70. R. Kasirajan, S. Afr. J. Chem. Eng., 2021, 37, 161–166.
71. P. E. Ohale, C. F. Uzoh and O. D. Onukwuli, S. Afr. J. Chem. Eng., 2017, 24, 43–54.
72. S. Yi, Y. Su, B. Qi, Z. Su and Y. Wan, Sep. Purif. Technol., 2010, 71, 252–262.
73. H. V. Lee, R. Yunus, J. C. Juan and Y. H. Taufiq-Yap, Fuel Process. Technol., 2011, 92, 2420–2428.
74. B. Salamatinia, H. Mootabadi, S. Bhatia and A. Z. Abdullah, Fuel Process. Technol., 2010, 91, 441–448.
75. H. Hamze, M. Akia and F. Yazdani, Process Saf. Environ. Prot., 2015, 94, 1–10.
76. S. Dharma, H. H. Masjuki, H. C. Ong, A. H. Sebayang, A. S. Silitonga, F. Kusumo and T. M. I. Mahlia, Energy Convers. Manag., 2016, 115, 178–190.
77. V. Birur Jayanna, V.M. Kulkarni, K. N. Krishnamurthy, G. C. Manjunath Patel, H. S. Nagendra Prasad, O. D. Samuel and E. Linul, Results Eng., 2025, 25, 104359.
78. A. Rezaeifar, M. Mansouri and B. Maleki, Sci. Rep., 2025, 15, 7067.
79. P. Maheshwari, M. B. Haider, M. Yusuf, J. J. Klemeš, A. Bokhari, M. Beg, A. Al-Othman, R. Kumar and A. K. Jaiswal, J. Clean. Prod., 2022, 355, 131588.
80. M. O. Daramola, K. Mtshali, L. Senokoane and O. M. Fayemiwo, J. Taibah Univ. Sci., 2016, 10, 675–684.
81. A. F. Lee, J. A. Bennett, J. C. Manayil and K. Wilson, Chem. Soc. Rev., 2014, 43, 7887–7916.
82. M. D. Putra, C. Irawan, U. Udiantoro, Y. Ristianingsih and I. F. Nata, J. Clean. Prod., 2018, 195, 1249–1258.
83. O. A. Mawlid, H. H. Abdelhady, M. G. Abd El-Moghny, A. Hamada, F. Abdelnaby, M. Kased, S. Al-Bajouri, R. A. Elbohy and M. S. El-Deab, J. Clean. Prod., 2024, 442, 140947.
84. R. Devasan, J. V. L. Ruatpuia, S. P. Gouda, P. Kodgire, S. Basumatary, G. Halder and S. L. Rokhum, Sci. Rep., 2023, 13, 2570.
85. G. Kupriyanova, M. Smirnov, I. Mershiev, A. Maraşlı, C. Okay, G. Mozzhukhin and B. Rameev, J. Food Compos. Anal., 2024, 126, 105877.
86. A. Barison, C. W. P. Da Silva, F. R. Campos, F. Simonelli, C. A. Lenz and A. G. Ferreira, Magn. Reson. Chem., 2010, 48, 642–650.
87. K. I. Doudin, Fuel, 2021, 284, 119114.
88. D. W. Lee, Y. M. Park and K. Y. Lee, Catal. Surv. Asia, 2009, 13, 63–77.
89. A. D. Burmana, R. Tambun, B. Haryanto, M. Sarah and V. Alexander, Case Stud. Chem. Environ. Eng., 2023, 8, 100510.
90. Y. Liu, Z. Zhu, R. Zhang and X. Zhao, Biomass Convers. Biorefin., 2024, 14, 28635–28660.
91. G. Pradhan and Y. Chandra Sharma, J. Clean. Prod., 2020, 264, 121258.
92. E. Martinez-Guerra and V. G. Gude, Appl. Sci., 2017, 7, 869.
93. M. Yadav and Y. C. Sharma, Energy Convers. Manag., 2019, 198, 111795.
94. K. Liu, G. Wei, Y. Zhu, L. Zhang and Z. He, J. Environ. Chem. Eng., 2023, 11, 111015.
95. S. Sahani, T. Roy and Y. Chandra Sharma, J. Clean. Prod., 2019, 237, 117699.
96. D. Kumar and B. Singh, Renewable Energy, 2019, 133, 411–421.
97. S. Sahani and Y. C. Sharma, Energy Convers. Manag., 2018, 171, 969–983.
98. C. Yan Loh, A. Rahman Mohamed and S. Yuan Lai, Fuel, 2025, 386, 134210.
99. A. S. Yusuff and M. O. Onibonoje, Mater. Renew. Sustain. Energy, 2023, 12, 199–208.
100. G. Pradhan, S. Pradhan, P. Chaudhary, D. Madhu and R. Singh, Mol. Catal., 2024, 556, 113876.
101. A. Al-Abbasi, F. Almahdi, M. Almaky, R. Izriq, A. Milad, S. Salim and A. Najar, Inorg. Chem. Commun., 2023, 158, 111620.
102. K. Balakrishnan, M. A. Olutoye and B. H. Hameed, Bioresour. Technol., 2013, 128, 788–791.
103. F. Jamil, M. Al-Riyami, L. Al-Haj, A. H. Al-Muhtaseb, M. T. Z. Myint, M. Baawain and M. Al-Abri, Int. J. Energy Res., 2021, 45, 17189–17202.
104. M. Safaripour, M. Saidi and H. R. Nodeh, Renewable Energy, 2023, 217, 119199.
105. T. Roy, S. Sahani and Y. Chandra Sharma, J. Clean. Prod., 2020, 247, 119166.

---