Extraction and Quality Evaluation of Biodiesel from Six Familiar Non-Edible Plants Seeds

Biodiesel produced from non-edible plant sources is cost-effective, biodegradable, environment friendly, and compatible with petro-diesel, but new sources and extraction processes still need to be discovered. Here, we explored the fuel properties of seeds from six non-edible plant sources, including Sapindus mukorossi (Soapnut, SP), Vernicia fordii (Tung, TO), Ricinus communis (Castor, CA), Toona sinensis (Juss. TS), Ailanthus altissima (Heaven tree, AA), and Linum usitatissimum L. (Lin seed, LS) from China. The optimum extraction conditions were obtained by optimizing the most important variables (reaction temperature, ratio of alcohol to vegetable oil, catalyst, mixing intensity, and purity of reactants) that influence the transesterification reaction of the biodiesel. All six plants contained high seed oil content (SOC; % w/v) with the highest in the TO-54.4% followed by SP-51%, CA-48%, LS-45%, AA-38%, and TS-35%, respectively, and all expressed satisfactory physico-chemical properties as per international standards of ASTM D6751 and EN14214. Our data provide a scientific basis for growing these plants in unproductive agricultural lands as an alternative energy sources for biodiesel production either standalone or blended with petro-diesel.


Introduction
The impact of emissions of greenhouse gases on climate change and growing energy needs has compelled the world community to focus on alternate energy sources, such as biodiesel production from plant, algae [1], and other waste materials [1,2]. Biofuel production from biomass is an effective strategy to reduce both crude oil consumption and pollution [3]. The demand and production of biofuels have grown rapidly (approximately 23% per annum) during the last two decades [4,5]. Globally, more than 350 oil-bearing plant sources [6] have been explored for biodiesel production using different extraction and optimization strategies. The vegetable plants used as biodiesel feedstock include soybean, rapeseed, sunflower, palm oils, mustard, peanut, sunflower, and cottonseed [7]. The challenge, however, remains to resolve as far as the high viscosity of vegetable oils is concerned. Generally, the high viscosity at room temperature is considered unsuitable in diesel engines. During the combustion process, oxygen in the air quickly reacts with the outer surface of the oil droplet and releases a huge amount of heat, and further initiates the intricate reactions (charring, choking, and polymerization). The oils with higher viscosity tend to form larger drops and further elevate the polymerization reaction, particularly those with a higher degree of unsaturation [8]. The risk gets even higher as more viscous

Sapindus mukorossi (Soapnut, SP)
Sapindus mukorossi tree produces soapnut fruit. The plant inhabits tropical and subtropical regions comprising Asia, America, and Europe. S. mukorossi and S. trifoliatus are commonly found in India, Nepal, Bangladesh, and Pakistan. S. trifoliatus contains an average of 51.8% oil content of total seed weight [24]. Soapnut contains potential non-edible oil for biodiesel production [25,26]. Although soapnut fruit shells have been used for medicinal [27], surfactant [28], as well as laundry purposes [29], after using pericarp portion, seeds are wasted, which makes them feasible for use as biodiesel (Figure 1a,b).

Ricinus communis (Castor, CA)
Ricinus communis, a small wooden tree with 6 m maximum height, belongs to the Euphorbiaceae family and generally known as the castor oil plant (Figure 3a,b). Although it has African origin, it can be widely observed in the tropical and subtropical regions of the world. Its seeds contain higher oil content ranging from 48% to 60%, with 500-1000 L of oil/acre production potential. The salient features of Ricinus communis oil are 90% ricinoleic acid (18 carbon atoms and hydroxyl group at position 12), more oxygen atoms suitable for transesterification, and challenging high viscosity (240.12 mm 2 /s at 313 K) [34].

Preparation of Seeds for Feed Stocking
In this study, after washing the seeds with distilled water, these were allowed to dry under sunlight for 48 h and later on were oven-dried at 60 °C for removal of moisture. The seeds were then ground using a grinder (XIANTAOPAI XTP-10000A, Zhejiang, China). The size of seed particles after grinding ranged from 0.21 to 1.0 mm. The seeds were again processed for oven-drying at 60 °C for 90 min to minimize the moisture content, and the seeds were then processed for oil extraction to maximize the purity and minimize any contamination after the grinding and pulverization.

Oil Extraction
The seed oil content (SOC) from all six seed sources was extracted using soxhlet [47] and mechanical oil extractor (Fangtai Shibayoufang FL-S2017 China and Fangtai Shibayoufang J508, Guangdong, China) (Table S1, Supplementary Materials). The oil extraction occurred at 90 °C for 7 h, and different solvents were also used during this process which comprised petroleum ether, acetone, dichloromethane, and ethyl acetate. Petroleum ether was finally used for all plant sources during Soxhlet extraction. Filter papers (pore size 30-50 µm) were used for removing the impurities, and solvent was removed at 80 °C by employing a rotary evaporator (Tokyo Rikakikai Co. Ltd. N-1210B, Tokyo Japan) under lower pressure (0 to 0.01 MPa). Finally, the oil extracted was stored and allowed to dry over anhydrous sodium sulfate prior to use.

Evaluation of Potential of the Six Plants as Biodiesel Resources
We assessed the potential of each plant as a biodiesel resource by its seed oil content (SOC). The detailed procedure of Soxhlet and mechanical extraction are given in Supplementary Material S1 and S2.

Acid-Catalyzed Esterification Process
The esterification process is most appropriate for the unrefined or waste cooking oils with high free fatty acid (FFAs). Sulfuric acid (H2SO4) is commonly used as a catalyst. The procedure is devoid of using pre-treatment of oil with an alkali for lowering its FFA content. The process has few challenges to overcome, including slow rate, need of methanolto-oil molar ratio; water production causes hindrance in the esterification of triglycerides following reaction of FFA with the alcohol. The overall biodiesel yield is hampered due to the burning of oil followed by treating it with the higher levels of the acid [48].
The methanol (469 g) and H2SO4 (11.4 g) were added into soapnut oil, tung oil, linseed oil, Haven tree, Toona sinensis, and castor oil (for all 2751 g) as a prerequisite of esterification pre-treatment and mixed vigorously at 60 °C for 1 h as a reaction time. Then, esterified soapnut oil was processed for methanolysis for 1 h, the molar ratio of methanol to 6, reaction temperature (60 °C), and KOH catalyst quantity on the basis of oil weight of 1% (w/w). The reactor had a Pyrex glass structure having 17.2 cm inside diameter and 5.5 L volume and properly connected with water jacket for marinating the reaction temperature. The reactor's design was founded on the shape factor criteria of a standard six-blade turbine [49]. The solution was allowed to settle down overnight, and thus separation phase was achieved. Later on, an ester phase was carried out, and saturated sodium chloride solution (three times the volume of the ester phase) was used to wash the outcome to remove any residual methanol, KOH, or glycerol. Ultimately, the water traces in the soapnut oil methyl esters (SPME) were removed by adding up anhydrous magnesium sulfate, subsequently to filtration [50].

Base-Catalyzed Transesterification Process
The seed oil biodiesel from all six sources was prepared by using crude oil (50 g), methanol (10 mL) at a molar ratio of 5:1, and KOH catalyst (2.3 w/v%). The reaction was conducted for 1 h at 65 °C under reflux, and the agitation rate was 600 rpm [51]. The procedure was conducted with an excess of methanol (99.99%), having methanol to oil ratio (5:1), and KOH ratio (2.9 w/w%), [52]. The reflux condenser had a reactor to cool down the methanol after coming out of the reaction mixture. The resultant reaction mixture was shifted into the funnel, and it was kept overnight in order to separate the biodiesel, soap, and glycerol constituents with glycerol at the bottom and biodiesel at the uppermost layer. Once the reaction got completed, we separated the crude glycerin using gravity, and KOH was also separated, followed by treating 3-4 times with hot distilled water. The phenolphthalein indicator was used for assessing the complete removal of the catalyst. The vacuum distillation procedure was applied on the leftover un-reacted methanol and moisture until the achievement of the final product and stable FAMEs weight loss. The crude FAMEs underwent further washing for 3-4 times heated de-ionized water, centrifugation, and drying with a vacuum dryer to ascertain its purity. The phase separation was fast and observable within 10 min. The biodiesel phase had a cloudy appearance; it, however, became pretty clean and clear after at least 20 h of setting duration. We used all analytical reagent-grade chemicals.

Fourier Transform Infrared (FT-IR) Study
The FT-IR bands spectroscopy was evaluated by FT-IR spectrometer (Bruker Vertex 70, Ettlingen, Germany) at a resolution of 1 cm −1 , scanning 15 times, and employing Nujol mull as a dispersive medium in the range of 400 to 4000 cm −1 , to originate the produced biodiesel which has been described through characterization of different functional groups.

Nuclear Magnetic Resonance (NMR) Study
The FAMEs NMR spectrum was carried out by NMR Spectrometer (Bruker Avance Ш 400, Karlsruhe, Germany) at 400 MHz (1H-NMR) or 100 MHz (13C-NMR). Denatured chloroform was used as solvent and tetramethylsilane as the internal standard. The biodiesel 1 H NMR (300 MHz) spectrum was documented with a cycle delay of 1.0 s and several scans of 8 times, with pulse duration of 30°. A carbon 13 C NMR (75 MHz) spectrum was recorded with a pulse duration of 30° and a cycle delay of 1.89 s, and a scan of 160 times [22,32,33].

Gas chromatography-Mass Spectrometry (GC-MS) Study
The FAMEs outcome was evaluated by GC-MS (QP2010SE, Shimadzu, Tokyo, Japan), [22]. GC-MS conditions were followed as per our previous study [22] and are listed in Table 1.

Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES) and Elemental Analysis Study of Biodiesel
The presence of metals in the FAMEs was studied using Inductively Coupled Plasma Spectrometer (Spectro-blue, Kleve, Germany) and Elemental Analyser (Vario EL CUBE, Hanau, Germany) for all six plant sources [22].

Effect of Methanol to Oil Molar Ratio on FAMEs yield
The results of the optimization process for all six plant sources are given in Table 2; the details of the process are given in Supplementary Material (Table S2 AAOB; Table S3 CAOB; Table S4 LSOB; Table S5 SPOB; Table S6 TOOB; Table S7 TSOB). The types of alcohol and methanol to oil molar ratios also have a significant impact on biodiesel production. Short-chain alcohols are preferred for biodiesel production as these are cost-effective and possess greater efficiencies and higher reaction speeds. In the current study, the optimum of methanol to oil molar ratios were 6:1 for TOBD, AABD, LSBD, and CABD; however, these were 5:1 for TSBD and 7:1 for SPBD. Overall, we applied 4:1 to 7:1 methanol to oil molar ratios for all sources in the present study (Details in Supplementary data). Moreover, methanol to oil molar ratios from 4:1 to 5:1 produced comparatively low FAME yield, and 6:1 was best as it gave an optimum yield, and an equilibrium reaction was established. However, beyond 7:1, the soap content was increased, and FAME content was decreased (Figure 9). The difficulty was further aggravated for separating glycerol due to high methanol solubility rate and reversible equilibrium reaction with the methyl esters to form mono-glycerides [53].

Effect of Catalyst Concentration on FAMEs Yield
Biodiesel production largely depends upon the choice of appropriate, cost-effective, and environment-friendly catalysts based on the nature of oil [54], which greatly help in the transesterification of oil. In our study, among all the tested catalysts (Details are given in Supplementary Material; (Table S2 AAOB; Table S3 CAOB; Table S4 LSOB; Table S5 SPOB; Table S6 TOOB; Table S7 TSOB), KOH was evaluated as effective in the context of FAME's yield. The optimum concentration of KOH was 0.32 (g) for TOBD, AABD, LSBD, CABD, and SPBD, respectively, whereas it was 0.42 (g) for TSBD. The percentage yield of biodiesel in our study was highest in the LSBD (98%), followed by the TOBD (97.2%), and CABD and SPBD (96% each), TSBD (95.2%), and AABD (93.9%), respectively ( Figure 10).

Effect of Temperature and Stirring Intensity on Fames Yield
Both temperature and stirring have a remarkable influence on mass transfer during transesterification of biodiesel production [55]. The optimum temperature (65 °C) and stirring intensity (700 rpm) were the same for all the sources in the present study. Details are given in Supplementary Material (Table S2 AAOB; Table S3 CAOB; Table S4 LSOB; Table  S5 SPOB; Table S6 TOOB; Table S7 TSOB). As the boiling point of methanol is 64.7 °C, so 65 °C was normal, as, beyond this temperature, it gets broken and produces a negative impact on biodiesel production. The stirring intensity at 500-600 rpm gave a low FAME yield as it required more shaking. However, at 700 rpm it gave the optimum yield. We found that beyond 700 rpm, it did not produce good results due to the binding of oil and methanol solution and extra shaking, which resulted in a negative impact on FAME yield (Figures 11 and 12).
Generally, biodiesel production generates about 10% (w/w) glycerol as the major byproduct. The excessive production of glycerol will pose challenges to the refined glycerol market [56]. In our study, the percentage yield of glycerol was highest in the AABD (6%), followed by the lowest in TOBD (1.8%). This highlights that glycerol outcome was at optimized levels in our study, and the plant sources reliably fall in a feasible range of biodiesel production (Figures 11 and 12). However, the glycerol produced as a byproduct can be used as animal feed rations and other value-added chemicals [56].
Biodiesel contains soap as one of the impurities which may create issues in engine operation as well fuel storage [57]. We found that the soap percent content was 0 for LSBD and CABD, 0.1 for AABD, and 1 each for TOBD, TSBD, and SPBD, respectively. This highlights that the biodiesel we obtained from six plant sources had minimum soap content which supports the notion of their testing and fuel usability at promising levels ( Figures  11 and 12).

Effect of Reaction Time on FAMEs Yield
The heterogeneous and homogeneous catalysis need about 4 and 1 h, respectively, as the reaction time to achieve the maximum biodiesel yield [58]. In our study, the optimum reaction time was 80 min for TOBD, LSBD, and TSBD; however, it was 60 min for AABD, CABD, and SPBD, respectively. Details are given in Supplementary Materials (Table S2  AAOB; Table S3 CAOB; Table S4 LSOB; Table S5 SPOB; Table S6 TOOB; Table S7 TSOB). In fact, the reaction time ranges from 60 to 80 min based on the sources for clear separation of biodiesel, glycerol, and soap. In addition, if we allowed more setting time to the final biodiesel product, the possibility of reversible reactions occurred, and the resultant FAME's yield was compromised ( Figure 13).

FTIR Analysis of Non-Edible Seed Oil Sources
To identify the functional groups and the bands corresponding to various stretching and bending vibrations in six biodiesel samples, the FT-IR spectroscopy of the mid-infrared region was used, as presented in Figure 14 and the Supplementary Materials Table S8. The two resilient ester representative absorption bands were detected from carbonyl (νC=O) around 1750-1730 cm −1 and C-O at 1300-1000 cm cm −1 [59]. The stretching vibrations and bending vibrations (ρCH2) of CH3, CH2, and CH appeared at 2980-2950, 2950-2850, 3050-3000 cm −1 , and at 1475-1350, 1350-1150, 722 cm −1 , correspondingly [60]. The absorption peaks of the sample were detected in all biodiesel samples to be 3464, 3007, 2927, 2854, 1743, 1641, 1435, 1361, 1170, 1016, and 723 cm −1 , respectively. The peaks presence in all biodiesel FAMEs at 1430 and 1167 cm −1 specifies the conversion of crude oil to biodiesel. The strong absorption peak at 2852-2859 cm −1 and 2920-2927 cm −1 is just because of the alkane group of C-H stretching vibration. The C-H bending vibration appeared at 1430-1466 cm −1 due to strong absorption. All of the single bands symbolize saturated functional groups. The C=O stretching frequency peak at 1741.30 cm −1 is due to strong absorption, which is composed of an unsaturated functional group and is called an ester. Further, due to the C-O stretching vibration of the ester, the strong band appeared at 1017-1093 cm −1 and 1161-1174 cm −1 .

NMR Analysis of Non-Edible Seed Oil Sources FAMEs
The nuclear magnetic resonance (NMR) comprising 1 HNMR (Figure 15; Supplementary Materials Table S9); 13 C NMR spectrum ( Figure 16; Supplementary Materials Table S10) were used for the characterization of the FAMEs of all six plant sources.

1 H NMR Analysis of Non-Edible Seed Oil Sources FAMEs
In the 1 H NMR spectra, the terminal methyl proton (C-CH3) appearance signals were detected between 0.76-0.88 and 0.92-0.98 ppm, and the aliphatic chain (-(CH2) n-) associated signals were observed between 1.19-1.29 and 1.47 ppm. The β-methylene ester (CH2-C-CO2Me) bands' appearance signals were detected around 1.55-1.62 ppm. Correspondingly, the methylene proton peaks near the base (-CH2-C = C-) appeared between 2.00-2.07 ppm attached to the allylic group. The methylene proton peaks signal were attributed between 2.19 and 2.30 ppm existing near the carbonyl group proton (-CH2-COOMe), the existence of a methylene proton (-C = C-CH2-C = C-) between the allylic groups associated between 2.74-2.80 ppm, and the single sharp peak at 3.52-3.66 ppm was the expressed ester bond (CH3COO-CH) linked CH3 group. The proton (-CH = CH-) from the glycerol moiety appeared between 5.23-5.34 ppm.

13 C NMR Analysis of Non-Edible Seed Oil Sources Fames
In the 13 C NMR spectrums, a signal which showed the occurrence of an ester carbonyl carbon (-COO-) was observed at 174.14-174.46 ppm. The band's signals were observed in the spectrum between 126.68-128.00 and 130.14-130.45 ppm indicating the existence of unsaturation in the methyl ester. The occurrence of ester (C-O) methoxy carbon was observed at 51.35 ppm due to the long carbon chain methylene carbon of the fatty acid methyl ester. The band's occurrence detected between 22.20-34.10 ppm, and the terminal carbon of the methyl group peaks was observed at 13.80-14.12 ppm, respectively ( Figure  16).

GC-MS Analysis of Non-Edible Seed Oil Sources FAMEs
The biodiesel obtained from crude oil of six plant sources and modified by methyl ester was evaluated by the gas chromatography and mass spectrometry (GC-MS) ( Figure  17). The peaks were identified by the NIST-14 library matching software. After assessment, every single peak was matched with fatty acid methyl ester [61]. The retention time (min) and position of the determined peaks are presented in Table 3

Conclusions
The non-edible plant seed oil, obtained from Sapindus mukorossi (Soapnut, SP), Vernicia fordii (Tung, TO), Ricinus communis (Castor, CA), Toona sinensis (Juss. TS), Ailanthus altissima (Heaven tree, AA), and Linum usitatissimum L. (Lin seed, LS) from China, provided promising biodiesel yield. The transesterification produced good quality biodiesel production possessing compatible combustion properties with that of diesel. The biodiesel obtained has compatible combustion properties with that of diesel. The seed oil contents obtained were TO-54.4% followed by SP-51%, CA-48%, LS-45%, AA-38%, and TS-35%, respectively. These plants thus can be cultivated on barren lands in order to enhance the feedstock and can be used as an alternate renewable, cost-effective, and environmentally friendly energy source for biodiesel production, either standalone or blended with petro-diesel.