Process Optimization of Biodiesel Production Using Waste Snail Shell as a Highly Active Nanocatalyst

,


Introduction
The substantial use of fuel-based energy for transportation that results from global industrialization promotes the depletion of fossil fuels and contributes to global warming.As a result, renewable fuels are viewed as a substitute for conventional energy sources in the fight against environmental pollution and fuel depletion [1].According to the literature studies, fossil fuel reserves for oil, gas, and coal may persist for an additional 40, 70, and 200 years, respectively [2].Fatty acid methyl ester (FAME), a biodiesel, offers a promising alternative to conventional petroleum diesel, as it can be seamlessly integrated into existing fuel infrastruc-ture.Compared to fossil fuels, it has significant environmental advantages, for example, having low sulphur content (up to 57.7%), low toxicity, and low CO (up to 58.9%) and low CO 2 (up to 8.6%) emissions [3], in addition to being biodegradable and renewable [4,5].Furthermore, it has a very good fuel property like having a high cetane number, less emission, good lubricity, and high combustion efficiency.For the manufacturing of biodiesel, many processes such as microemulsification, pyrolysis, direct and indirect usage, transesterification, and dilution method are employed [6,7].Transesterification is the easiest and most affordable method for producing biodiesel which includes converting triglycerides to methyl esters (with methanol) or ethyl esters (with ethanol) using the right catalyst.The process of transesterification used to make biodiesel is shown in Figure 1.The significance of biodiesel as a clean energy solution has urged intensive research to optimize its production methods.One pivotal area of investigation lies in the development of efficient and sustainable catalysts, as they play a central role in the transesterification process [8].Traditional catalysts, such as homogeneous acids or bases, have shown promise but often come with drawbacks, including high cost, toxicity, and the generation of significant waste streams.Also, homogeneous base catalysts (NaOH and KOH) are incapable of converting used cooking oils and inedible oils to biodiesel due to their high FFA content, which ultimately causes soap formation in the product.Furthermore, a huge amount of water is needed to purify biodiesel which inevitably results in the generation of a huge quantity of waste water.In addition, it is difficult to recover and reuse the catalyst [9].
To overcome these boundaries, a heterogeneous catalyst was developed due to its various benefits for instance easy catalyst recovery and reusability, thus making the overall process economically efficient [10].Several works have been reported on biodiesel production using heterogeneous catalysts.There are numerous types of heterogeneous catalysts like acids, bases, and enzymes.Since enzyme and acid catalyst requires very high operating conditions (for example high methanol-to-oil ratio, high temperature, and a longer reaction period), it is typically preferred to use a basic catalyst while making biodiesel.Shu et al. [11] conducted a study on the synthesis of biodiesel from waste vegetable oil utilizing a carbon-based solid acid catalyst.Their approach involved the use of high operating conditions, including a temperature of 220 °C and a substantial methanol-to-oil ratio (MTOR) of 16.8 M. Similarly, Narkhede et al. focused on producing biodiesel from waste cooking oil with a solid acid catalyst.Their process required an extended reaction duration of 8 h and achieved a conversion rate of only 86%, which was relatively low [12].In another investigation, Alhassan et al. explored the production of biodiesel from mixed waste vegetable oil employing ferric hydrogen sulfate as a catalyst.Their method demanded a high reaction temperature of 205 °C and a significant molar ratio of 15 : 1 [13].
For enzymatic transesterification of jatropha oil, Kumari et al. reported a prolonged reaction duration of 48 h, with a conversion rate of just 68%, indicating relatively low efficiency [14].Similarly, Tamalampudi et al. conducted enzy-matic biodiesel production from jatropha oil, which also necessitated an extended reaction time of 60 h to obtain the desired product [15].
In contrast, in the case of biodiesel production using a magnetic solid base catalyst, Guo et al. [16] employed much milder operating conditions.Their process operated at a lower temperature of 60 °C, had a shorter reaction time of 100 min, and utilized a more moderate molar ratio of 7 : 1.Additionally, Liu et al. [17] synthesized biodiesel from soybean oil utilizing calcium methoxide as a solid base catalyst.Their approach was notably efficient, completing the reaction in just 2 h at a temperature of 65 °C and a methanolto-oil ratio of 1 : 1.Thus, to date, significant growths in terms of improved catalytic activity, selectivity, and stabilization of active sites have been observed in supported catalysts: K 2 O/CaO-ZnO [18], Zn@CaO [19], CaO@La 2 O 3 [20], CaO@SnO 2 [21], hydroxyapatite-encapsulated g-Fe 2 O 3 nanoparticles (HAp-g-Fe 2 O 3 ) [22], Fe 3 O 4 @HKUST-1 [23], Fe 3 O 4 @silica [24], WO 3 @ZrO 2 [25], lipase/GO-Fe 3 O 4 [26], MgO/Fe 2 O 3 -SiO 2 [27], and Cu/ZnO [28].For instance, in the case of K 2 O/CaO-ZnO, the incorporation of K 2 O on the CaO-ZnO catalyst enhanced the catalytic transesterification activity by elevating both its basicity and surface area, whereas in the case of CaO@SnO 2 , tin oxide can act as a stabilizing matrix for calcium oxide (CaO) nanoparticles by preventing their aggregation or deactivation during the reaction.Also, the use of nanomagnetic catalyst such as MgO/Fe 2 O 3 -SiO 2 in the production of biodiesel offers notable advantages such as higher specific surface area and good catalytic activity.Nanocatalysts with high specific surface area and large porosity are useful for the catalyst to bond with the sublayer which enhances the duration, reproducibility, and the yield of a transesterification reaction.However, the synthesis of the above catalysts contains multiple steps thereby making the overall processes lengthy, tedious, and complex.The total cost of the product is further increased due to the high chemical costs of the precursors.
Hence, utilizing waste materials as a heterogeneous base catalyst will solve the environmental issues while also improving the economic sustainability of biodiesel production.The choice of raw materials for biodiesel production is a critical aspect of its overall sustainability and economic viability.Common feedstock for biodiesel production includes vegetable oils and animal fats.While traditional edible oils remain a common feedstock, efforts to utilize

2
International Journal of Energy Research alternative feedstock and address issues related to oil acidity are essential for making biodiesel production more sustainable, economically viable, and less dependent on valuable food resources.Effective catalyst management and pretreatment of acidic oils are key strategies to ensure efficient transesterification, even when using calcium oxide as a catalyst [29].To produce biodiesel from various vegetable oils, a variety of waste products have been used as sources of CaO catalyst, including egg, oyster, crab, snail, and chicken manure shells [30].Maneerung et al. [31] developed CaO from bottom ash waste produced during the gasification of woody biomass and turned it into an active CaO catalyst by simply calcining it at 800 °C for the making of biodiesel through transesterification of palm oil.Boonyuen et al. [32] extracted CaO from Turbo jourdani shells (Turbinidae), a marine shell, as a raw material and converted it into an active catalyst calcined at 900 °C for 5 h.The optimal parameters were found to be catalyst concentration of 10 wt.%, methanol: oil ratio 3 : 1, and a reaction period of 7 h to give a >99% oil to FAME conversion level.Roschat et al. [33] used hydrated lime-derived CaO as a catalyst for the transesterification of palm oil.A yield of 97% was attained under optimal conditions, for example, catalyst loading 6 wt.%, time of 2 h, methanol : oil ratio of 15 : 1, temperature of 65 °C, and calcination temperature of 800 °C.Niju et al. [6] produced CaO from conch shells, a suitable heterogeneous base catalyst used for the transesterification of Moringa oleifera oil (MOO).Here, a two-step transesterification procedure was used to synthesize the biodiesel.The optimal process parameters were catalyst concentration 8.02 wt.%, methanol : oil molar ratio 8.66 : 1, and 2.17 h reaction duration.Under these conditions, a methyl ester conversion of 97.06% was produced.Although shell-derived catalysts are reported for biodiesel production, they usually have low surface area and modest reactivity.Hence, to expand the surface area and boost the overall reactivity of CaO, several researchers performed further hydration followed by dehydration of the produced shell-derived CaO [34][35][36].Additionally, our research team has reported Mangifera indica peel [37], orange peel ash [38], and Musa acuminata peel ash [39] as sustainable catalysts for the synthesis of biodiesel, which is in line with our area of expertise in renewable energy.Snails are a member of the phylum Mollusca and belong to the class Gastropoda.With more than 100,000 documented species, snails, which are members of the phylum Mollusca, are considered the second-highest diversity of living species among all arthropods on earth [40][41][42].The annual market value for edible snails is EUR 1 billion.Europe consumes around 100,000 tonnes of eatable snails each year.Snail shells are left over as garbage after being digested [43].They have a high calcium carbonate (CaCO 3 ) content, which can be easily converted into calcium oxide (CaO) and can be used in the transesterification reaction to produce biodiesel [44,45].
The use of nanocatalysts has recently been the focus of numerous researchers related to biodiesel synthesis.Nanocatalyst represents materials with particle size between 1 and 100 nm [46].Nowadays, nano biodiesel production has emerged as an exciting field for enhancing the efficiency and sustainability in biodiesel production [47].A way to improve the catalytic activity of CaO is to increase its specific surface area by preparing it in nanoform.The goal of this work is to evaluate and improve the catalytic performance in the transesterification of soybean oil with methanol using a nano-CaO-based catalyst derived from wastelong snail shells.
Using response-surface methodology (RSM), a statistical optimization procedure was used to investigate the maximal biodiesel output from soybean oil.The effect of processinput variables like reaction duration, catalyst load, temperature, and methanol to oil molar ratio (MTOR), along with their interactions, on the yield of biodiesel was studied using central composite design (CCD).

Synthesis of CaO Nanoparticle from Long Waste Snail
Shell.CaO nanoparticles were synthesized from waste-long snail shells using the hydration-dehydration method.At first, these shells were washed thoroughly to get rid of any flesh or unwanted materials adhered onto the surface of the shells, and then, it was dried for 12 h at 100 °C.In order to make sure complete removal of impurities from the shell, the process has been repeated for several times.The snail shells were then sieved with mesh sizes ranging from 125 to 250 mm, pounded to a very fine particles using mortar and pestle with an outside diameter of 4 inches and a height of 3 inches, and then calcining it in a muffle furnace at 900 °C for 4 h.The calcium carbonate present in the snail shell will break down into CaO and CO 2 at a temperature above 800 °C, as shown in Equation (1).The calcined snail shell powder was further refluxed with distilled water at 80 °C for 6 h and then oven dried at 100 °C for 10 h.Further, the resultant powder was dehydrated by calcination at 900 °C for about 4 h to obtain ultrasmall size, highly porous CaO nanoparticles.The produced CaO nanoparticles from the snail shell are then kept in an airtight desiccator.
The basic strength (H_) of the catalyst was determined by the Hammett method using the following indicators such as bromothymol blue (pKa = 7 2), phenolphthalein (pKa = 9 8), 2-4 dinitroaniline (pKa = 15 0), and 4-International Journal of Energy Research nitroaniline (pKa = 18 4).Approximately, about 25 mg of the catalyst was placed in a conical flask containing a 5 mL methanol solution of Hammett indicators, and then, the mixture was agitated.After two hours when the reaction reached equilibrium, a colour change in the solution was observed.The catalyst basic strength was found to be greater than the indicator used when the solution exhibited a colour change.However, when there is no colour change, the catalyst exhibited a lower basic strength than that of the indicator used.A colour change was observed from colourless to pink when phenolphthalein (pKa = 9 8) was used as an indicator and from yellow to mauve when 2-4 dinitroaniline (pKa = 15) was used, but in the case of 4-nitroaniline (pKa = 18 4), it failed to show any change in colour.Hence, the catalyst basic strength was found to be 15 < H <18 4 and is considered as a strong base for transesterification reaction [48].
2.3.Characterization of the CaO Nanocatalyst.The calcined snail shell structure and morphology were studied using XPS, FT-IR, BET, TGA, XRD, SEM-EDX, and TEM analyses.XPS spectra of the elemental analysis were gained from the Thermo Fisher Scientific ESCALAB Xi + instrument.The source type is aluminium K-α, the power of the XPS instrument is 1486.6 eV, and the calibration standard is set at C-C bond binding energy of 284.8 eV.FT-IR study was performed, and IR spectra were obtained in the range of 400-4000 cm -1 using a 3000 Hyperion FT-IR spectrometer (Bruker, Germany) to identify the presence of functional groups in the material.BET analysis was obtained with a Quanta chrome ASIQwin surface area and porosity analyzer to measure N 2 adsorption-desorption isotherms.TGA analysis of the catalyst was done using Perkin Elmer Instrument (model: STA 800, USA) under a heating rate of 10 °C min -1 under a nitrogen atmosphere, and the weight loss of the curve has been recorded.XRD analysis was performed using a Panalytical X' pert 3 powder machine, which emits a wavelength of 0.154 nm, with Cu-Kα and monochromatic radiation.Intensity range was from 10 °to 90 °at a scanning scale of 1 °min -1 .The structural morphology of the catalyst was examined with the help of Sigma Field Emission microscopy (Model: Zeiss, Germany).This system is equipped with a LaB6 field emitting electron gun having three different types of detectors, viz, InLens, SE2, and ESB detectors.This instrument can operate with acceleration of 0.2 to 30 kV.Transmission electron microscopy (TEM) images were procured on a JEOL, JEM-2100 Plus Electron Microscope.

Transesterification
Reaction for FAME Production.Soybean oil was transesterified by combining 0.874 g of SO, 10 mmol of methanol, and 52.4 mg (6 wt.%) of the prepared catalyst in a pressure tube kept on a magnetic stirrer (pressure tube reduces the pressure inside the vessel, thereby increasing the boiling point of the solvent, and hence, methanol evaporation is avoided).Reaction conditions such as methanol to oil molar ratio of 4 : 1-9 : 1, catalyst concentration of 3-8 wt.% and time of 1-8 h were tested during experimentation.The reaction was performed at a temperature range of 30-80 °C.The reaction progress was monitored using the thin layer chromatography (TLC) technique.After completion of the reaction, catalyst was separated from biodiesel using centrifugation process for 10 min at 4000 rpm.The synthesized biodiesel product and composition were confirmed by NMR and GC-MS techniques.The recovered catalyst was then thoroughly cleaned with distilled water and dried in an oven at 100 °C before being regenerated by calcination at 900 °C.

Experimental Method and Response Surface Methodology (RSM).
A five-level, four-factorial central composite design (CCD) was employed in this study.The input parameters were MTOR (A), catalyst loading (B), temperature (C), and reaction time (D).The input factors were coded from -1 to +1, and their levels for analysis were developed from primary laboratory trials [49].Experiments were carried out in agreement with central composite design to determine the finest combination and to assess the influence of process factors on biodiesel yield, and the outcomes are depicted in Table 1.Regression analysis was used to investigate the experimental data from the CCD, and a second-order polynomial model was fitted to determine the possible interactions of chosen parameters with the response function: The Sheldon test was performed to determine the heterogeneous nature of the CaO nanocatalyst [51,52].The synthesized catalyst was filtered off from the reaction medium after 1.5 h of the reaction.The reaction was then continued for another 4 h under catalyst-free reaction condition, and it was then filtered off.TLC and GC-MS techniques were used to observe the progress of the reaction.Catalyst reusability was investigated by performing consecutive batch of transesterification reaction.After every run, the catalyst was separated from the reaction mixture by centrifugation; then, it was washed thoroughly with dis-tilled water and then allowed to dry overnight in an oven at 100 °C.Also, catalyst was reactivated by calcining it at 900 °C for about 4 h.Again with the calcined catalyst, a new set of reaction was conducted with fresh reactants.This procedure was repeated up to six times.SEM-EDX analysis was done to examine the morphology of the recycled catalyst.
2.8.Reaction Kinetics.The transesterification reaction is predicted to obey pseudo-first-order kinetics, as excessive amount of methanol was present in the reaction mixture, permitting the reverse reaction to be ignored [53].Thus, Equation ( 5) can be used to calculate the response rate (r): where k denotes the rate constant, O represents the concentration of SO, and t represents time.First-order rate constants were calculated by using Equation ( 6).The  7), was employed to calculate the activation energy (E a ) for the transesterification processes [54].

Results and Discussions
3.1.Characterization of the Catalyst 3.1.1.XPS.XPS analysis revealed important information about the elements constituting the catalyst surface.Accord-ing to the findings, the catalyst surface is composed of calcium, oxygen, and carbon.A survey scan for binding energy was performed in the range of 100-900 eV (Figure 2(a)).The existence of carbon, calcium, and oxygen in the sample was discovered by three peaks in the survey scan of the catalyst at 283 eV, 346.78 eV, and 531.14 eV.Narrow scans were then performed in the C1s region, Ca 2p region, and O1s region [55].The existence of carbon (C1s, Figure 2(c)) as support from the catalyst results in a sharp peak at 284.81 eV and a weak peak at 289.61 eV, which is attributed due to the presence of carbonate [56].The O1s peak at 531.5 eV (Figure 2(d)) resembles the metal oxide lattice oxygen [57].It can be observed that Ca2p displayed two main distinct peaks at 346.91 eV and 349.82 eV accredited to Ca 2p 1/2 and Ca 2p 3/2 (Figure 2(b)), respectively [49,58].

FT-IR.
The FT-IR spectroscopy was used to know the formation of CaO.The peak observed in Figure 3 at 716 cm -1 could be assigned as a CaO bond, matched with Hussein et al. [59].Moreover, the presence of CaO after the 6 International Journal of Energy Research calcination of snail shells can be seen at 1440 cm -1 and 876 cm -1 , which is in agreement with Guan et al. [60].
3.1.3.BET.The pore size distribution and nitrogen adsorption-desorption isotherm of the calcined long snail shell are shown in Figure 4.According to IUPAC's classification, N 2 isotherms revealed a Type IV hysteresis loop which is a graphical representation of the mesoporous material's specific behaviour [61].The presence of bigger mesopores in the sample was indicated by a shift in the hysteresis loop to higher P/P o values (>0.8) [62].The surface area and pore volume of the CaO nanocatalyst were found to be 19.451m 2 g -1 and 0.034 cc g -1 , respectively.The high surface area and pore volume may be due to the formation of porosity in the calcined snail shells.The evolution of steamy carbonization products (CO 2 in our case) and the creation of CaO were both responsible for the pore formation in the produced catalyst [63].The pore-size distribution study (inset, Figure 4) reveals a pore radius of 13.84 Å, and the pore diameter lies within the range of 2-50 nm, indicating mesoporous nature [64].The catalytic performance of the transesterification reaction is controlled by pore structure and surface area.As our catalyst is prepared in the nanoform, it possesses a high surface area, leading to excellent  7 International Journal of Energy Research catalytic activity.The mesoporous material's pores allow the reactants to diffuse into the pores thereby speeding up the reaction rate.The reaction rate in mesoporous materials is substantially higher, but in microporous materials, the rate is lower because the reaction is used to hold at the entrance of the pores [65].Thus, the calcined mesoporous CaO could be used to increase the production of biodiesel from SO, a clean fuel additive.
3.1.4.TGA.The thermal stability of the snail shell has been investigated by performing the thermogravimetric analysis (TGA). Figure 5 shows the TGA curve where the weight loss of snail shell has been analyzed under the influence of temperature (room temperature to 1000 °C).The catalyst weight loss has been noticed in two stages as depicted in Figure 5. Firstly, a weight loss of about 4% occurred between the temperature range of 55-616 °C which may be due to the evaporation of moisture and decay of organic substances.The major weight loss was observed during the second stage at 41% between the temperature range of 616-789 °C which was caused by the decomposition of CaCO 3 to CaO by releasing CO 2 .No weight loss was observed after the temperature of 800 °C [66,67].
3.1.5.XRD.The X-ray powder diffraction technique is mainly used to determine the crystallographic nature of the material.Figure 6 shows the XRD profile of uncalcined and calcined snail shells (CSS).It may be seen from Figure 6(b) that CaCO 3 is the main constituent of the natural snail shell and exists in two different crystalline phases, calcite, and aragonite, examined from the stated data (JCPDS file no.41-1475 and 29-0306) [68].After calcination at 900 °C for 4 h, nearly all of the CaCO 3 got converted to CaO, but we can still see some CaCO 3 peaks in Figure 6(a) which may be accredited due to the presence of atmospheric CO 2 [48].This is also supported by the TGA result as depicted in Figure 5.In Figure 6(a), the XRD pattern of the CaO nanoparticle shows perceptible peaks at 29.02 °, 32.37 °, 37.54 °, 53.99 °, 64.27 °, and 67.57°which corresponds to (011), ( 111), (002), ( 022), (113), and (222) cubic plane systems.The plane values of the XRD pattern correspond perfectly with the reported data of CaO measurements (JCPDS card No. 00-004-0777) [68,69].The catalyst's crystalline structure was revealed by the appearance of strong and sharp peaks of the calcined snail shells.The average crystalline size of CaO NPs is calculated using the Debye-Scherrer equation.
where D is average crystalline size, K is the Scherrer constant, λ is the X-ray wavelength (0.154 nm), θ stands for Bragg's angle, and β is line broadening at FWHM in radians.
From Equation ( 8), the mean crystalline size was found to be 45.05 nm.
3.1.6.SEM-EDX.Scanning electron microscopy (SEM) technique has been used to study the surface morphology of the natural and snail shell calcined at 900 °C.Natural snail shells displayed asymmetrical sizes of rod-like particles and also exhibited a typical layered architecture (Figure 7(a)) [30].After calcination, it was observed that the particles had a semispherical form with a significant agglomeration of the catalyst due to the large surface area (Figure 7(b)).Most importantly, the CaO catalyst exhibited a porous structure that could provide the catalyst with an active site [48].The porous surface resulted from calcination might be caused by the release of water and carbon dioxide in the course of the decomposition of CaCO 3 to CaO [70].Energy dispersive X-ray spectroscopy (EDX) provides information about the chemical composition of the catalyst.In the EDX spectrum of calcined CaO, calcium element displayed a higher peak than oxygen which is also supported by the elemental mapping of calcium (Figure 7(c)), oxygen (Figure 7(d)), and carbon (Figure 7(e)).The higher the concentration of elements in a specimen, the higher will be the intensity of the peak in the spectrum.
3.1.7.TEM.The transmission electron microscopy technique is used to analyze the structure and composition of the materials at the nanoscale level.Figures 8(a)-8(d) show that the particles are porous, well-ordered, and spherical.The polycrystalline property of the material was confirmed by SAED imaging (Figure 8(e)).The average particle size distribution of CaO nanoparticle was found to be 42.44 nm (Figure 8(f)), which has a good resemblance with the particle size obtained from XRD results using the Debye-Scherrer equation.

Modelling Outcomes and Data Analysis Using Response Surface Method (RSM).
Experimental data obtained from central composite design (CCD) were used to determine the ideal combination of various operating parameters (such as temperature, methanol/oil molar ratio, time, and catalyst concentration) to maximize the generation of biodiesel.Table 1 summarizes the actual and predicted responses for 30 experiments.Advanced multiple regression analysis was subjected to the experimental data to obtain a secondorder polynomial equation with regression coefficients, which were then assessed for statistical significance.
The biodiesel yield from soybean oil ranged from 59.1 to 96.1% from the experimentations conducted in the laboratory.The second-order polynomial equation predicted by the model for maximum biodiesel is shown in where A is the MTOR (methanol to oil molar ratio), B is the catalyst loading (wt.%), C is the temperature ( °C), and D is the reaction period (h).
The linear terms (A, B, C, and D) have a positive impact on the yield of biodiesel as can be seen from Equation (9).The terms A 2 , B 2 , C 2 , and D 2 represent the squared terms of the independent variables/factors within the secondorder polynomial model.These terms represent the squared values of the corresponding factors and play a crucial role in capturing the curvature and nonlinear effects in the response surface.The linear terms in the RSM model capture the straight lines along each axis, but real-world responses are rarely perfectly linear.This is where the square terms, A 2 , B 2 , C 2 , and D 2 , come into play.For each value of independent variables, there would be a corresponding y-value (i.e., the response/biodiesel yield).The RSM model consisting of four independent variables satisfies the quadratic model where y is the function of A, B, C, and D (y = f A, B, C, D ).This quadratic effect suggests that the most effective values of each parameter are located within the intermediate range of the experiment, rather than at its extremes.Thus, based on such dependency, the coefficient comes as to be positive and negative which further explain their role in the model to be significant or not significant.Each quadratic term A 2 , B 2 , C 2 , and D 2 thus explains the interaction with itself and its significance with respect to the biodiesel yield [71]. ANOVA experimental findings are performed and presented in tabular form in Table 2.They include statistical tests like Fischer's test (F-value), the p value that describes the likelihood of having an F-value of any magnitude, and the sum of squares that governs the relevance of parameters concerning model performance [72,73].The model F-value of 66.46 suggests that the model is significant.There is only a 0.01% chance that a high F-value could occur owing to noise.The model is termed significant when p value is less than 0.05 [74].Here, A, B, C, AB, AC, BC, BD, CD, A 2 , and D 2 are significant model terms.Except for the methanol-to-oil ratio (MTOR), reaction time interactions, all interactions, and quadratic terms were found to be significant.The statistical results of the entire process have been summarized in Table 2.The predicted R 2 of 0.9248 is in reasonable agreement with the adjusted R 2 of 0.9693 indicating a difference of less than 0.2.The adjusted R 2 value is a crucial statistical measure used to assess the goodness of fit of a regression model.It represents an advancement over the conventional R 2 (coefficient of determination) by incorporating the complexity of the model.In simple terms, the exclusion of insignificant terms (i.e., p value >0.05) from the models is termed as the adjusted R 2 value [75].Adjusted R 2 can provide a more precise view of the 9 International Journal of Energy Research obtained model correlation by also taking into account how many independent variables are added to a particular model against which the experimental model is measured.This is done because such additions of independent variables usually increase the reliability of that model.That is a reason that the difference between R 2 and adjusted R 2 should be least.Generally, a higher value than the former one ascertains goodness of fit.A high R 2 value of 0.9841 indicated that there is a sta-tistical goodness of fit between the model and experimental data.The R 2 value of 0.9841 indicated that the model explained 98.41% of the variability in the experimental data for biodiesel production.However, when the model predictor variables increases, the R 2 value increases; hence, adjusted R 2 is employed to avoid this undesirable effect.Thus, the reported RSM-CCD approach having adjacent R 2 = 0 9693 showed that 4-parameters/variables fit significantly with the 11 International Journal of Energy Research quadratic model.The adjusted R-square value in RSM has a similar practical significance to that of the R-square.In our reported RSM-CCD approach, the R-square value is observed to be 98% which signifies how effectively the predictor factors in the model can account for variances in the dependent factor and the ease of response very well.As a result, the model can be said to have good practical significance [76][77][78].Adequate precision (AP) determines the signal-to-noise ratio, and a ratio > 4 is considered to be desirable [79].Here, the ratio of 33.186 indicates an adequate signal.The % CV for the model was found to be 1.891where a value of <10% is preferred, signifying a decent connection between actual and anticipated yield values.
From the diagnostic plots depicted in Figure 9, the quality of the regression model has been tested.The normal probability plot has been illustrated in Figure 9(a).The data points were distributed in a linear outline, showing that the normal distribution of the studentized residuals supported the regression results.If this curve adopts an S form, that model cannot be employed since it is considered to be defective for the model and may arise owing to a confidence interval and p value inaccuracy.Figure 9(b) displays the plot of predicted yield vs. actual yield.Externally studentized residuals vs. predicted yield of biodiesel plot are illustrated in Figure 10(a).This figure is predicted to have a random distribution, indicating that the change in the primary observations is independent of the response value.The suggested model provides a decent depiction of the method as supported by the random distribution of the data.For each response, the actual value and predicted value were relatively close.Therefore, this model is appropriate for the empirical data and may be used to estimate the maximum biodiesel output.Figure 10(b) shows the outlier plots for all runs.The proximity of data points to the fit regression line implies a good estimation of the response for changes in the independent variables A-D.Independent residuals revealed no patterns or trends.Because of the patterns in the points, it is possible that nearby residuals are related and therefore not independent.Here, the points were spread randomly around the line within the limits, showing that the model is accurate and correct without any data errors.
Figure 11 depicts the perturbation plots which help to classify the process variables on biodiesel yield while keeping the other factors constant to their center values [80].The steepness of a slope determines how much of an impact a variable has on the yield.Thus, the perturbation plot clearly shows that A, which has the steepest slope from the lower level (-1) to the middle value (0), is the major component followed by C, D, and B, respectively.On the contrary, from the mid value (0) to the higher level (1), D has a dominant effect.Thus, it shows that as we increase the methanol to oil molar ratio from 4 to 6, there was a noticeable change in the yield but with further increment in MTOR, and not much significant impact on process A was observed but rather on the process variable D. Hence, based on the overall steepness of the slope and the results of ANOVA study, A has significant influence over all the variables.1.The red lines denote the range of residuals that are not considered outliers.
13 International Journal of Energy Research in the model graphs can be seen.Figure 12(a) shows the combined influence of surface plots of the catalyst loading and MTOR on the yield of biodiesel, with all other variables held at their center values.It was observed that biodiesel yield increases with a decrease in catalyst loading from 9 to 6 wt.%, and increasing the MTOR from 1 : 4 to 1 : 8. Additionally, it is observed that the existence of a higher catalyst concentration in the reaction medium at a similar MTOR decreases the yield of biodiesel.The catalyst loading has a significant impact on the biodiesel yield since it alters the catalytic activity.In addition to increasing the yield of biodiesel, the right catalyst concentration also prevents unwanted side effects like saponification and hydrolysis [81].In Figure 12(b), at a fixed catalyst loading of 6 wt.%, increasing the temperature from 30 to 70 °C, results in an increase in biodiesel yield.The 3D plots of the combined effect of time and MTOR on biodiesel yields are depicted in Figure 12(c), respectively.Keeping the reaction duration fixed and increasing the MTOR increase the yield of biodiesel.The yield is hardly affected by adding additional methanol to the reaction medium, and the yield of biodiesel is negatively impacted by increasing methanol ratios.Additionally, extending the reaction time significantly reduces the biodiesel yield while keeping the molar ratio constant.As we increased the molar ratio from 4 to 7, the increment is maximum but as we further go from 7 to 8, the increment is not so much which is in agreement with the regression coefficient of the interaction term (-0.15) and molar ratio parameters (D × A) on the regression model (from Equation ( 9)).High oil : methanol ratios are not economically and environmentally viable.Further-more, high methanol concentrations may decrease catalyst concentration in the reactant mixture, thereby slowing the transesterification process.Figure 12(d) depicts the interaction between time and temperature.As we keep the time constant and increase the temperature from 30 to 70 °C, the biodiesel yield also increases.The biodiesel yield is maximum when the temperature reaches 70 °C, and further increase in temperature leads to the evaporation of methanol leading to decrease in biodiesel yield [82].The interaction between temperature and MTOR is depicted in Figure 12(e).It shows that keeping the temperature constant at 70 °C biodiesel yield decreases sharply as we decrease the molar ratio of MTOR. Figure 12(f) illustrates the interaction of time and catalyst loading.It shows that as we keep on increasing the catalyst loading from 6 to 9 wt.% keeping time fixed at 3 h, the biodiesel yield also decreases.This pattern can be attributable to a variety of factors.Firstly, the higher concentrations over a prolonged reaction time result in an emulsion taking place between glycerol, biodiesel, and methanol particles, thereby reducing separation and purification and eventually resulting in a decrease in biodiesel yield [83].The second reason might be due to the reversible nature of the transesterification reaction.An increase in time causes the reaction to move in an unfavorable direction.Last but not least, it could be the result of a saponification reaction brought on by an excess of catalyst concentration, which causes an increase in viscosity and foam formation [84].

Biodiesel Characterization.
The transesterification product, biodiesel, was characterized using 1 H and 13 C-NMR spectroscopy.A Bruker Advance III-500 MHz and Joel 400 MHz FT-NMR were used for analyzing the sample.GC-MS was also used to identify different types of methyl esters produced from triglyceride (soybean oil).Analysis of each sample was carried out on Shimadzu (GC-2030) series GC-MS equipped with Headspace (HS-20) & QQQ Mass spectrometer GC-TQ8040NX and column SH-Rxi-5 SILMS (0.25 X 30 X 0.25).The temperature of the oven was controlled between 60 and 280 °C, while the injector and detector temperature were kept at 200 °C and 300 °C, respectively.The main peaks of the biodiesel produced from soybean oil were identified using Fourier transform infrared (FT-IR) spectroscopy.

FT-IR Analysis.
Figure S4 shows the FT-IR spectrum of synthesized biodiesel.IR spectra were observed in the region 1425-1450 cm -1 which corresponds to the assymetric bending of CH 3 group, and the peaks observed in the range 1181-1450 cm -1 correspond to the bending and oscillating vibrations of OCH 3 group [85].The two peaks observed at 2862 and 2927 cm -1 corresponds to the stretching vibration of C-H bond of alkane.The existence of C=O stretching vibration of carbonyl groups present in triglycerides and esters causes the prominent peak to exist at 1738 cm -1 [86].
3.4.2.NMR Analysis. 1 H-NMR spectroscopy was used to determine the percentage conversion of biodiesel from SO    Figure 12: A 3D graphic depiction (a-f) showing the interaction between A-D and its impact on the production of biodiesel from soybean oil.International Journal of Energy Research nanocatalyst. 1 H-NMR spectra show the presence of a methoxy proton peak at 3.65 ppm as a singlet suggesting the presence of a methoxy group (-OCH 3 ) in the molecule since methoxy protons typically appear as singlets in 1 H-NMR spectra and the existence of α-CH 2 proton peak as a triplet at 2.29 ppm indicate the presence of methylene group in the molecule as depicted in Figure 13(b).The presence of these two distinct characteristic peaks in the 1 H-NMR spectra (Figure 13(a)) verified that methyl ester was formed after the transesterification of soybean oil by the snail shellderived heterogeneous catalyst.The existence of olefinic hydrogen was indicated by the presence of a multiplet at 5.33 ppm.The conversion of soybean oil to biodiesel was calculated using the ratio of the integrated areas below the peaks at 3.65 ppm (methoxy protons) and 2.29 ppm (α-CH 2 protons), and 98.5% conversion was obtained using Equation (3) [87].The findings matched well with the information provided by Gohain et al. [88].

Physicochemical Characteristics of Soybean Oil
Biodiesel.The ASTM-D6751 standards were used to derive the biodiesel fuel properties, including calorific value, flash point, density, acid value, and kinematic viscosity (at 40 °C).Table S1 (Supplementary Information, SI) shows the measured physicochemical biodiesel parameters.Since all of the property values fall within the parameters of ASTM biodiesel standard, they can all be utilized as an alternative fuel in transportation engines.
3.4.4.GC-MS Analysis.GC-MS studies help to identify the chemical composition of the biodiesel as depicted in Figure 14.Table 3 summarizes the fatty acid methyl ester content in SO biodiesel using the GC-MS technique.According to the GC-MS analysis report, major FAMEs identified were 9-octadecenoic acid, methyl ester, (E), hexadecanoic acid, methyl ester, 9, 12-octadecadienoic acid (Z, Z)-, methyl ester, and methyl stearate having the significant peaks.

Kinetic Study of Transesterification.
The evaluation of reaction kinetics between soybean oil to biodiesel is depicted in Figure 15(a) which is a linear plot of -ln 1 − x against time (min) at temperatures ranging from 40 to 70 °C.From Figure 15(a), it can be seen that there is an exponential increment in the transformation of soybean oil to biodiesel thus confirming our prediction that the transesterification  18 International Journal of Energy Research reaction followed pseudo-first-order kinetics [89].The activation energy (E a ) was derived by fitting rate constants to the Arrhenius equation (Equation ( 7)).The rate constant k was calculated from these graphs.The slope (−E a /R) and intercept of the ln k vs. T −1 plot confirmed pseudo−first −order kinetics, and the activation energy E a for the reaction was calculated.From Figure 15(b), E a was found to be 30.45kJ mol -1 by using Equation (7).The E a values for    CaO-catalyzed transesterification of vegetable oil lie within the range of 26-136 kJ mol -1 [48].Thus, the calculated E a value is well within the prescribed range.
3.6.Catalytic Mechanism.Several studies have been made to investigate the mechanism for CaO-catalyzed transesterification reaction [68,90,91].A plausible mechanism of transesterification of SO to biodiesel by using CaO as a basic catalyst is depicted in SI, Figure S2.In the first step, there is a formation of the methoxide anion, arising from the reaction between catalyst CaO and methanol (step 1).Then, the methoxide ion attacks one of the carbonyl carbon of triglyceride leading to the formation of an alkoxy carbonyl tetrahedral intermediate (step 2).Then, that intermediate is rearranged by forming a diglyceride anion and a mole of methyl ester (step 3).The anionic species is then stabilized by proton transfer from the catalyst surface to form diglyceride.Once all three centers of the carbonyl carbon of triglyceride are attacked by the methoxide ions, three moles of methyl esters (biodiesel) and one mole of glycerol are produced [92].

Test for Heterogeneity and Catalyst Reusability.
To study the heterogeneous nature of the CaO nanocatalyst, Sheldon's hot filtration method was performed.For this, after 1.5 h of the reaction, the prepared catalyst was removed by filtration under hot conditions at optimized parameters (8 : 1 MeOH: SO molar ratio, 6 wt.% catalyst loading, and conventional heating to 70 °C), and the yield was observed to be 48 7 ± 0 4%.The reaction was then further extended for 4 h under catalyst-free conditions.Noticeably, a slight increase in yield was observed from 48 7 ± 0 4% to 49 8 ± 0 3%, representing a 1.1% increase (shown in SI, Figure S3).The study showed that there was no significant amount of catalytically active soluble species present in the filtrate.Therefore, it is confirmed that the catalyst is heterogeneous.The main advantages of using such a catalyst for the synthesis of biodiesel are the ease of removal of the catalyst, which lowers the downstream processing costs, and its potential for reuse, which further lowers cost reductions.
Since cost is the key concern for the production of biodiesel, so catalyst reusability test is crucial because it demonstrates the benefit of employing a catalyst to lower the manufacturing cost, making it economically viable and industrially applicable.Under optimized reaction conditions, the reusability of the catalyst was examined, and the outcomes are shown in Figure 16.After every run, the catalyst was centrifuged out of the reaction mixture and then rinsed with hexane and dried overnight at 100 °C and then recalcined it again at 900 °C to activate the catalyst.This process was continued for six reaction cycles to examine the catalyst stability.The experimental findings, which are displayed in Figure 16, demonstrated that an acceptable biodiesel output could be attained in up to six cycles (82%).Beyond that, a substantial decline in yield was seen, which may have been caused by the triglycerides and glycerol poisoning the pores and catalyst surfaces, thereby reducing the number of active sites and lowering the biodiesel yield.After each reuse, the catalyst load was adjusted to get the best reaction conditions.Nonetheless, leaching during the washing stage resulted in some loss of the catalyst throughout each cycle.To determine the structural stability of the reused catalyst, the surface morphology of the spent catalyst was examined, and it was observed that there is a small decrease in Ca amount from 52.3% (fresh catalyst) to 38.0% (reused catalyst) (SI, Figure S5) after the reuse indicating that the catalyst was stable after six reaction cycle.

Comparing CaO Nanocatalyst with Previously Reported
Catalysts.According to the literature survey, a wide range of catalysts have been reportedly used to synthesize biodiesel.Comparing our results with previous studies, it becomes evident that waste snail shell nanocatalysts demonstrate superior catalytic activity.This comparison highlights the innovative potential and frames it as a significant development in biodiesel production technology.The relevant information (i.e., the kind of feedstock, catalyst, working conditions, and biodiesel output) is compiled in Table 3 for comparison with other catalysts developed here.Numerous studies have stated moderate to high biodiesel yields under optimal conditions.Various reported works used a high methanol-to-oil ratio (entries 1, 2, 3, 4, 8, 9, and 10) and high catalyst load (entries 3 and 9) which was a disadvantage.The reaction mixture becomes more viscous with the increased catalyst loading as it inhibits mass transfer in the liquid-liquid-solid system, thereby lowering biodiesel yield.Also in some literature (entries 6 and 7), the time required to complete the reaction is very high which ultimately makes the process very tedious.
Several reports (entries 11 and 12) have highlighted the potential for enhancing snail shells with metal-organic frameworks (MOFs) [93] and activating them with basic catalysts like KOH to enhance their catalytic activity [67].However, it is important to note that the methods employed to achieve these enhancements can be notably lengthy and time-consuming.Furthermore, in entry 5, the reaction parameters are relatively modest in amount, which is advantageous in terms of resource consumption, but the observed yields are relatively low.A comparative assessment of the    yields obtained using the calcined snail shell catalyst in our study, when compared with those from other studies in the literature, which employed similar catalysts but with different feedstocks and nearly identical catalytic treatments, reveals that our prepared catalyst emerges as a sustainable and viable alternative to previously reported studies.
This suggests that our approach offers a promising balance between catalyst effectiveness and sustainability, making it a noteworthy contribution to the field.In the investigation of catalytic activity, the TOF of the reported catalyst was 0.0063 mol g -1 h -1 which was greater than several listed catalysts and likely superior to a number of catalysts listed in Table 4.
3.9.Life Cycle Cost Analysis.Life cycle cost analysis (LCCA) is an economic method used to access all costs incurred throughout the entire life span of a project or product starting from the acquisition of raw materials (feedstock oil, catalyst, and methanol), and other cost (for example, operational cost, transportation cost, and maintenance cost) [101].To assess the economic viability of employing the catalyst and to address the increasing need of energy demands, it is essential to thoroughly examine the costs related to both catalyst synthesis and biodiesel manufacturing [102].The ability to reuse a spent catalyst also enhances the commercial viability of the biodiesel production process.By effectively recycling the catalyst, the total biodiesel production cost is reduced thereby making the process more economic and sustainable.The judicious use of waste precursor material as a catalyst preparation can be a great advantage.The estimated cost for catalyst preparation in the current study helps in analyzing whether the process is viable for practical use by considering all the criterias.Table 5 presents the observations computed in dollar ($) for the preparation of 0.0623 kg of catalyst in order to produce 1 kg of biodiesel.Table 6 summarizes the stepwise biodiesel preparation cost.

Conclusion
A low-cost, economical, biodegradable, and ecologically friendly nano-CaO catalyst generated from waste snail shells was used in the synthesis of biodiesel from soybean oil.The calcined snail shell exhibited a high surface area of 19.451 m 2 g -1 and a pore volume of 0.034 cc g -1 .The prepared catalyst underwent thorough characterization using XPS, FT-IR, BET, TGA, XRD, SEM-EDX, and TEM.Additionally, 1 H and 13 C NMR, GC-MS were used to characterize the produced biodiesel.An excellent 98.5% conversion of SO to biodiesel and a biodiesel yield of 96.11% were obtained under optimal reaction parameters as recorded by 1 H-NMR and RSM numerical optimization methods, respectively.A yield of 82% was observed after the sixth cycle indicating that the catalyst exhibits good stability and can be recycled without a significant loss in its activity.In addition to these advantages, CaO is inexpensively and conveniently available from a variety of waste sources which therefore makes it economically viable.Henceforth, the synthesis of biodiesel from soybean oil under optimized conditions can be regarded as a potential feedstock that might replace petro-diesel fuel in current engines and fulfill the rising demand for fuel oil.

Figure 4 :
Figure 4: N 2 adsorption and desorption curve of a calcined snail shell and pore radius distribution (inset).

Figure 5 :
Figure 5: TGA of the catalyst snail shells.

Figure 9 :Figure 10 :
Figure 9: Diagnostic plots: normal probability plot (a) and plot of predicted versus actual biodiesel yield (b) for experiments 1-30 in Table 1.The trend in colour from blue to red represents the transition from poor to good biodiesel yield.The orange dot in the plot (a) represents the midpoint of biodiesel yield.

Figure 14 :
Figure 14: The GC-MS spectrum of soybean oil biodiesel.
Feedstock costCost of soybean oil for production of 1 kg of biodiesel = amount × soybean oil cost per kg = 1 041 kg × $biodiesel production The cost of methanol for the production of 1 kg of biodiesel after 6 th recycle = amount × methanol cost per kg /6 = 0 304 kg × $0 23 ÷ 6 = $0 011 0.011 Cost of biodiesel production (transesterification process time h × units × per unit cost) Time h × units consumed × cost per unit = 3 h × 1 17 × $0 051 = $0 179 0.179 Net cost (cost of soybean oil +cost of catalyst + cost of methanol + cost of biodiesel production) $0 633 + $0 027 + $0 011 + $0 179 0.85 Overhead cost (10% of net cost) 10% of $0.85 0.085 Total cost of 1 kg biodiesel net cost + over head cost = 0 85 + 0 085 0.935 21 International Journal of Energy Research With the help of 1 H and13C NMR spectroscopy, the esterification product was examined.The Bruker Advance III-500 MHz and Joel 400 MHz FT-NMR spectrometer were used for NMR investigation using TMS (tetramethylsilane) as an internal standard.At a temperature of 28 °C, the data were collected.Using the Knothe and Kenar equation, the methoxy and methylene group (A Me and A CH 2 ) integrals were compared as shown in [50]re β o represents the intercept term; β 1 , β 2 , β 3 , β 4, β 12 , β 13 , β 14 , β 23 , β 24 , and β 34 are their respective coefficients, and A, B, C, and D represent the linear terms.AD, AC, CD, BD, BC, and AB are the interaction terms, and A 2 , B 2 , C 2 , and D 2 are quadratic terms, and their respective coefficients are β 11 , β 22 , β 33 , and β 44 .2.6.Analysis of Product/Biodiesel.whereAMe and A CH 2 represent the integration value of methyl esters and alpha methylene protons.The number of protons in the methylene is represented by factor 2 in the numerator, and the number of protons in the methyl ester is denoted by 3 in the denominator.The biodiesel 4International Journal of Energy Research yield was determined using the Leung and Guo equation[50].

Table 1 :
Design matrix containing experimental variables (A-D) with actual and predicted yield for the transesterification reaction.

Table 2 :
Statistical results for the regression model of transesterification of soybean oil.
3.3.Effects of Process Input Variables on BiodieselProduction.The effect of independent variables A-D (MTOR, catalyst loading, temperature, and reaction time) on the experimental biodiesel output was analyzed using 3D surface plots.While keeping other parameters constant in their center values, the interaction between two variables

Table 3 :
Chemical composition of soybean oil biodiesel.

Table 4 :
Comparison of diverse heterogeneous catalysts for the synthesis of biodiesel.