Harnessing pongamia shell hydrolysate for triacylglycerol agglomeration by novel oleaginous yeast Rhodotorula pacifica INDKK

Background To meet the present transportation demands and solve food versus fuel issue, microbial lipid-derived biofuels are gaining attention worldwide. This study is focussed on high-throughput screening of oleaginous yeast by microwave-aided Nile red spectrofluorimetry and exploring pongamia shell hydrolysate (PSH) as a feedstock for lipid production using novel oleaginous yeast Rhodotorula pacifica INDKK. Results A new oleaginous yeast R. pacifica INDKK was identified and selected for microbial lipid production. R. pacifica INDKK produced maximum 12.8 ± 0.66 g/L of dry cell weight and 6.78 ± 0.4 g/L of lipid titre after 120 h of growth, showed high tolerance to pre-treatment-derived inhibitors such as 5-hydroxymethyl furfural (5-HMF), (2 g/L), furfural (0.5 g/L) and acetic acid (0.5 g/L), and ability to assimilate C3, C5 and C6 sugars. Interestingly, R. pacifica INDKK showed higher lipid accumulation when grown in alkali-treated saccharified PSH (AS-PSH) (0.058 ± 0.006 g/L/h) as compared to acid-treated detoxified PSH (AD-PSH) (0.037 ± 0.006 g/L/h) and YNB medium (0.055 ± 0.003 g/L/h). The major fatty acid constituents are oleic, palmitic, linoleic and linolenic acids with an estimated cetane number (CN) of about 56.7, indicating the good quality of fuel. Conclusion These results suggested that PSH and R. pacifica INDKK could be considered as potential feedstock for sustainable biodiesel production.


Background
Global population is increasing exponentially and likely to reach ~ 8.6 billion by 2030, raising big concerns about energy [1]. Demand for plant oils has also been shooting up in parallel since they have many industrial applications like epoxy biopolymers [2], drug delivery systems [3], bio-lubricants [4], pharmaceuticals [4] and biodiesel [5]. However, plant oil-derived biodiesel production has raised many questions related to the sustainable use of food crops for cleaner energy production [6]. Therefore, oleaginous microbes having fatty acid profile similar to vegetable oils are considered as suitable alternative for biodiesel [7]. Among oleaginous microbes, micro-algae are being widely used for lipid production, but it requires vast region of land for large-scale cultivation, longer incubation time and specific light exposure [8]. Currently, oleaginous yeasts are of special interest as they can produce high lipid titres in short duration and require limited space [9]. Additionally, yeast has the ability to utilize various renewable carbon sources along with the potential to grow at low pH, which prevents bacterial contamination [10]. Together, these characteristics facilitate the

Open Access
Biotechnology for Biofuels *Correspondence: naseem@icgeb.res.in; nasgaur@hotmail.com 1 International Centre for Genetic Engineering and Biotechnology (ICGEB), New Delhi 110067, India Full list of author information is available at the end of the article development of oleaginous yeast-based technologies for economically attractive industrial process.
Presently Rhodotorula, Rhodosporidium, Lipomyces and Trichosporon are considered as potential yeast for lipid-based biodiesel production. Yeast strains belonging to these genera can accumulate intracellular lipids more than 60% of their dry cell weight (DCW), displayed tolerance to pre-treatment-derived inhibitors along with the ability to assimilate wide range of carbon sources [10]. Consequently, screening and identification of oleaginous yeast isolates from unexplored natural habitats are still relevant [10]. However, the cost of microbial oil production is exorbitant due to the high substrate cost [11]. Therefore, to establish a sustainable microbial lipid production, extra endeavours are requisite such that yeast efficiently utilize renewable and low-cost carbon sources [12]. In recent years, inexpensive lignocellulosic carbon sources like rice straw hydrolysate [13], elephant grass hydrolysate [14], sugarcane bagasse hydrolysate [15], groundnut shell hydrolysate [16], wheat straw [17] and waste office paper hydrolysates [18] have been used for microbial lipid production. But, the conversion of these renewable feedstock like lignocellulosic materials into lipids in a cost-effective manner is a key challenge [19]. The common steps involved in converting lignocellulosic materials to microbial lipids for biodiesel production include: hydrolysis of lignocellulosic materials into fermentable sugars; utilization of released sugars by microbes for lipid production; and biodiesel production from lipids. However, during thermo-chemical pre-treatment such as acid/alkali and steam explosion process, hydrolysates are laden with weak acids, furans and phenolic compounds like 5-hydroxymethylfurfural (5-HMF), furfural and acetic acid which inhibit yeast growth and lipid accumulation. Therefore, for removal of pre-treatment-generated inhibitors, detoxification methods like treatment with activated charcoal and over liming [20] are employed to facilitate efficient fermentation of hydrolysate sugars into lipids. Various studies focus on the utilization of the detoxified hydrolysate for the yeast lipid production. Patel et al. [21] explored lignocellulosic wastes such as cassia fistula for biodiesel production using Rhodosporidium kratochvilovae. Huang et al. [22] explored the rice straw hydrolysate which was detoxified by activated carbon for lipid production by Trichosporon fermentans with lipid titer of 12.1 g/L after 10 days fermentation. Recently, Liu et al. [17] reported lipid production on wheat hydrolysates using different oleaginous yeasts. Moreover, lignocellulosic hydrolysates contain wide range of C5 and C6 sugars including glucose, xylose and arabinose along with the inhibitors. Therefore, yeast isolates utilizing C5 and C6 sugars derived from lowcost feedstock along with the potential to tolerate high concentration of pre-treatment derived inhibitors are much essential for lipid production.
In this regard, Pongamia pinnata was explored as source of substrate for microbial lipid production. It is a non-edible oilseed tree, which belongs to Leguminosae family and grows in the semiarid regions such as Asia, Australia and Florida. It was estimated that one hectare of land can produce approximately 6.8 tons of seeds in shell which can generate 1.12 tons of oil, 1.9 tons of meal, 2.67 tons of pod shells and others [23]. The air-dried pongamia shells (PS) consist of 46.02% carbon, 0.23% nitrogen, 42.46% oxygen, 5.58% hydrogen and 5.7% ash [24]. However, the PS are generally discarded or burned after oil extraction from seeds. Recently, pongamia shells have been utilized as fuel briquettes [24], but their utilization as feedstock for biodiesel production remain unexplored. The compositional analysis of widely available waste PS unveils its potential application for lipid production by microbes.
Utilizing low-cost materials required development of more efficient process like employing trans-esterification process to produce a high-quality FAME biodiesel. FAME production process involves some advantages due to low energy utilization, flexibility in feedstock consumption, reduced capital cost and faster reaction by employing accelerated trans-esterification at lower temperature. However, renewable or green diesel produced via hydro-processing of vegetable oils involves costly additional steps of isomerization and cracking at higher temperature and pressure [25]. Interestingly, any alteration in the fatty acid profile influences the biodiesel properties during trans-esterification process. The relative fatty acids composition in oleaginous yeasts was found to be C18:1 > C16:0 > C18:2 = C18:0 that can be altered depending on the feedstock provided and their growth conditions [26]. The biodiesel quality is also affected by the refining process, production process and postproduction parameters. Hence, international standards namely European (EN 14214), American Society for Testing and Materials (ASTM D6751) and Indian standards (IS15607-05) have been set up to monitor the parameters and quality of biodiesel. The important parameters for potential biodiesel are cetane number (CN), high heating value (HHV), cold filter plugging point (CFPP, °C), oxidative stability (OS, h), viscosity (mm 2 /s), iodine value (IV, mgI 2 /100 g), density (kg/m 3 ) and saponification value (SV, mg KOH/g oil). The oil composition and biodiesel properties were evaluated in this study to ensure the biodiesel quality and compared to the ASTM, IS and EN biodiesel standards specifications.
The aim of this study was to explore an inexpensive and renewable raw material pongamia shell hydrolysate (PSH) for yeast lipid production. First, the high lipid-accumulating oleaginous yeast was screened using microwave-aided Nile red spectrofluorimetry method. Alkaline pre-treatment, acid pre-treatment, enzymatic hydrolysis, detoxification and lipid production by yeast fermentation in PSH were then conducted and optimized. To the best of our knowledge, this is the first report on lipid production from waste PSH using yeast fermentation. This study provides valuable information for researchers on microbial lipid production using PSH as sole carbon source.

Screening and molecular identification of the selected yeast isolate
In this study, we collected a pool of potential oleaginous yeast isolates (57) including strains procured from collection centres in India (NCIM and MTCC) as well as by screening samples from various sites of biomass degradation (Additional file 1: Table S1). Next, molecular identification of new yeast isolate was carried out by PCR amplification of the ITS region (using genomic DNA template) followed by phylogenetic relationship analysis.  [27]. The evolutionary background of the taxa was determined by the maximum likelihood method [28]. The bootstrap consensus of tree was obtained from 1000 replicates and units of branch length are represented by number of nucleotide substitutions per site. As shown in Fig. 1

R. pacifica INDKK assimilated wide range of sugars and displayed inhibitor-tolerant phenotypes
Hydrolysates of lignocellulosic biomass contain mixture of C5 and C6 sugars and toxic inhibitors generated during pre-treatment such as furfural, acetic acid and 5-HMF [34]. These inhibitors reduce cell growth as well as lipid yield and productivity [35]. Therefore, yeast isolates capable of assimilating wide range of sugars (C5 and C6) along with enhanced tolerance to pre-treatment inhibitors are very important for economical microbial lipid production [36]. Interestingly, R. pacifica INDKK was able to grow on all the tested C6, C5 and C3 sugars except cellobiose and rhamnose (Fig. 4a). Moreover, the presence of lignocellulosic hydrolysate inhibitors such as 5-HMF (2 g/L), furfural (0.5 g/L) and acetic acid (0.5 g/L) in culture media did not show any significant effect on R. pacifica INDKK cell growth ( Fig. 4b-d).

Microbial lipid production using PS
Pongamia tree bears non-edible fruits whose shells after oil extraction from the seeds are generally discarded or burned [37]. Compositional analysis showed that PS contains 56.8% w/w holocellulose, 12% w/w cellulose and 8%  (Table 1). However, PS has not been considered as source of carbon and nitrogen for microbial cell growth thus far. In this regard, we explored hydrolysate of Pongamia pinnata shells for growth and lipid production by our newly isolated yeast isolate R. pacifica INDKK.
PS were subjected to acid as well as alkali pre-treatments (Additional file 5: Table S2). Liquid fraction of acid treatment showed higher sugar concentration (37.38 g/L) while alkali-treated liquid fraction obtained negligible amount of fermentable sugars (0.7 g/L total sugars). Therefore, liquid fraction of acid treatment was detoxified by activated charcoal, which reduced acetic acid concentration from 5.61 ± 0.035 g/L to 0.11 ± 0.005 g/L, completely removed furfural and 5-HMF with slight reduction in sugar concentration (14.63%). The acidtreated and detoxified liquid fraction of PSH (AD-PSH) contains 31.91 ± 0.042 g/L of total sugars (0.45 g/L glucose, 29.01 g/L xylose and 2.45 g/L arabinose). Time course study revealed that R. pacifica INDKK consumed all the sugars after 120 h of growth and produced 10.63 ± 0.004 g/L of DCW with 4.48 ± 0.02 g/L of lipid titre and 0.037 ± 0.001 g/L/h of lipid productivity (Fig. 5. Panel-1 5a).
The solid fraction of both acid-treated and alkalitreated PS were subjected to enzymatic hydrolysis (as   There was a drastic reduction (17-fold) in calcium ions which might be due to increase in calcium influx into the cell to cope up with the physiological stress conditions [38,39]. Time course analysis of R. pacifica INDKK in AS-PSH showed that all the sugars (C6 and C5) were consumed after 120 h of growth and maximum 12.6 ± 0.5 g/L of DCW, 7.02 ± 0.7 g/L of lipid titre, 0.104 ± 0.004 g/L/h of biomass productivity and 0.058 ± 0.006 g/L/h of lipid productivity were achieved (Fig. 5, Panel-1 5b). Remarkably, the DCW and lipid productivities were 1.02 and 1.12-fold higher in AS-PSH as compared to YNB (0.101 ± 0.005 g/L/h of biomass productivity and 0.052 ± 0.003 g/L/h of lipid productivity) (Fig. 5, Panel-1 5c). Moreover, after 120 h 13% utilization of trace elements corresponding to Ca, Na, Mg, P, K and Mn were also observed. The lipid accumulation in PSH batch-cultivated cells was also confirmed by confocal microscopy. The average cell size (7.43 ± 0.99 µm) and average LD size (5.47 ± 0.68 µm) was 1.67 and 1.27-fold higher in AS-PSH cultivated cells as compared to YNB medium, respectively (Fig. 5, Panel-2). However, cell size and LD size was 1.18 and 1.56-fold higher in AS-PSH grown cells as compared to AD-PSH (6.26 ± 0.78 and 3.49 ± 0.56 µm), respectively. Together, batch cultivation in AS-PSH significantly showed more biomass and lipid productivity as compared to AD-PSH (P value < 0.05).

Discussion
Biodiesel-derived from lignocellulosic materials is often challenging and costly because of additional material processing steps such as biomass pre-treatment to release sugars and removal of pre-treatment-generated inhibitors in the hydrolysates that hinder fermentation. Therefore, use of potential oleaginous yeast that could simultaneously utilize mixed carbon sources and show tolerance to inhibitors will reduce the major obstacles of biodiesel production. Over the past few years many inhibitor-tolerant oleaginous yeast have been found, but their lipid production performances are still substandard. Hence, the quest for robust oleaginous yeast is still relevant. Yeast isolates isolated from sites of biomass degradation have shown great potential for TAG accumulation [41]. To untap the potential of oleaginous yeasts isolated from natural habitats related to lignocellulosic biodiesel production, 57 yeast isolates were screened.
High-throughput microwave-aided Nile red staining was found to be quick, effective and easy method for screening high TAG accumulating oleaginous yeasts as compared to other traditional methods for lipid estimation. This method clearly differentiates the RFU values between oleaginous and non-oleaginous yeast. Among 57 yeasts, 6 strains belonging to Rhodotorula and Rhodosporidium species showed higher RFU values. Further, gravimetric analysis data showed that four strains (R. pacifica INDKK, R. toruloides, R. kratochvilovae, R. rubra) displayed > 20% lipid content. Interestingly, when kinetic analysis was performed, our isolate R. pacifica INDKK displayed high lipid productivity with effective sugar utilization rate, which could be by stimulation of genes related to growth and lipid production [42]. Inhibitors present in lignocellulosic hydrolysates (acetic acid, 5-HMF, furfural, etc.) inhibit yeast growth. Acetic acid inhibits growth by repressing the expression of genes involved in nutrient transporters such as glucose transporters (HXT1 and HXT3) [43,44]. 5-HMF inhibits dehydrogenases and glycolysis, whereas furfural reduces growth by inhibiting the key enzymes of carbon metabolism, increased production of radical oxygen species which damage DNA, protein and membranous structures [45].
Interestingly, increased growth was observed in 0.2 g/L concentration and displayed reduction of growth at 0.6 g/L of acetic acid. The increase in growth at 0.2 g/L could be due to the utilization of acetic acid as carbon source at this concentration. The yeast cells were also tolerant to 5-HMF (2 g/L) and furfural (0.5 g/L), beyond this concentration significant decrease in growth was observed. According to previously reported studies, most Rhodosporidium species could not tolerate furfural at 0.5 g/L concentration while Rhodotorula glutinis showed growth inhibition by 5-HMF > 0.5 g/L [36]. Therefore, in this study enhanced tolerance to inhibitors was observed by R. pacifica INDKK. It also showed ability to utilize all the tested C6, C5 and C3 sugars effectively. However, no significant growth was observed on rhamnose and cellobiose in comparison to glucose (P < 0.05). R. pacifica INDKK showed similar carbon source utilization profile as reported earlier in Rhodosporidium mucilaginosa [36]. The data elucidate that R. pacifica INDKK was tolerant to pre-treatment-generated inhibitors with potential to utilize different carbon sources present in lignocellulosic hydrolysates. The results were consistent with previously reported literature, wherein yeast such as Trichosporon fermentans and Cryptococcus curvatus were reported with similar potential but with low lipid productivity [6,33].
Lignocellulosic lipid production is a multistep process where feedstock collection and valorization itself accounts for 70-80% biodiesel production cost. Therefore, we exploited abundantly available lignocellulosic waste PS as feedstock to reduce the major obstacle of biodiesel production using oleaginous yeast. Notably, when isolate R. pacifica INDKK was tested with AD-PSH and AS-PSH as carbon sources, we achieved lipid titre of 4.48 and 7.02 g/L at 30 °C in 120 h. The higher biomass and lipid was observed with AS-PSH, which could be due to preferable utilization of glucose as carbon source as compared to xylose. Also, AS-PSH contains more nutrients which supports the growth of R. pacifica INDKK. The gravimetric data were in co-relation to confocal microscopy study (P < 0.05). The lipid productivity of R. pacifica INDKK in AS-PSH (0.058 g/L/h) was higher than the previously reported oleaginous yeasts on different lignocellulosic hydrolysates such as 0.041 g/L/h on waste office paper enzymatic hydrolysate [18], 0.02 g/L/h on saccharified sweet sorghum juice [46,47], and 0.029 g/L/h on corn stover enzymatic hydrolysate [48] as shown in Table 4. To the best of our knowledge, no yeast isolate reported has produced equivalent lipid titre to isolate R.  Hoekman et al. [49] reported that differences in degree of unsaturation and carbon chain length influence the properties and performance of biodiesel. For better OS, biodiesel should have high amount of SFA and MUFA, but low amount of multi-unsaturated FAME. While biodiesel should have low amount of long-chain SFA for good low-temperature performance [49], the fatty acid composition of R. pacifica INDKK was rich in C16:0 and C18:1 depicting improved biodiesel properties such as CN and OS [50,51]. The biodiesel obtained must meet the fuel standards (EN 14214, ASTM D6751, and IS) specifications before using it as a pure fuel. The CN parameter of diesel engine determines the auto-ignition quality of the fuel [49]. The high MUFA content elucidates balance between CFPP and OS for better quality of biodiesel [52]. The biodiesel properties obtained were in range of the biodiesel standard specifications illustrating the vehicular quality. The KV and density were similar to rape seed oil and jatropha oil. Results indicated that high CN of biodiesel from R. pacifica INDKK cultivated on AS-PSH significantly affects engine performance and start of injection with improved ignition characteristic. Longer OS and low CFPP properties of biodiesel obtained lead to longer shelf life and good engine performance, respectively. Hence, biodiesel production from R. pacifica INDKK cultivated on PSH was environmentally friendly and cost-efficient. Therefore, this process could be considered as an important step for the development of a cost-effective biodiesel production process.

Methods and materials
Media and other chemicals PS were collected from International Centre for Genetic Engineering and Biotechnology (ICGEB), New Delhi campus. All analytical reagents and solvents (chloroform, methanol, n-hexane, diethyl ether, glacial acetic acid, H 2 SO 4 ) were of high-performance liquid chromatography (HPLC) grade. Nile red (9-diethylamino-5-benzo[α] phenoxazinone), Bodipy 493-503 nm, heptadecanoic acid (internal standard) and FAME external standard (Supelco 37 component FAME mix) for GC-MS analysis were procured from Sigma (USA). Standard for

Microwave-aided Nile red staining and screening of oleaginous yeast
Yeast isolates were screened for lipid accumulation by using microwave-assisted Nile red staining protocol [54] with modifications. Briefly, single colony culture of each strain was grown overnight in YPD medium at 30 °C and 200 rpm (pre-culture). The pre-culture was centrifuged and washed twice with Milli-Q (MQ) water, re-suspended in 100 mL YNB medium with glucose (3%) and (NH 4 ) 2 SO 4 (0.5%) to optical density (OD) of 0.2 at 600 nm and incubated at 30 °C for 3 days at 200 rpm. Cells corresponding to OD 1 of the above grown cultures were centrifuged (5000 × g, 4 min) and re-suspended in 50 µL of dimethyl sulphoxide (DMSO) followed by microwave treatment (1250 watts power for 60 s). Cells were mixed with Nile red solution (10 µg/mL) and again subjected to microwave treatment (1250 watts power, 60 s). Four replicates of each treatment were prepared and relative fluorescence intensity (RFU) was measured at exciting and emission wavelengths of 475 and ~ 580 nm, respectively. Relative neutral lipid content was represented as RFU of LD [55].

ITS sequencing and phylogenetic analysis
The 18S rDNA sequence was PCR amplified from the genomic DNA by using ITS1-F (TCC GTA GGT GAA CCT GCG ) and ITS4-R (TCC TCC GCT TAT TGA TAT  GC)

Cell growth in presence of different carbon sources and pre-treatment inhibitors
To know the carbon source utilization efficiency by R. pacifica INDKS, the experiment was conducted on individual carbon sources. For this, cells were grown in 100 mL YNB medium supplemented with 2% carbon source individually (sucrose, cellobiose, glucose, mannose, galactose, rhamnose, arabinose, xylose and glycerol) and incubated at 30 °C, 200 rpm for 3 days. The inhibitor tolerance profile was tested by growing cells in YNB with 2% glucose supplemented with varying concentrations of 5-HMF (0.5 to 3 g/L), furfural (0.5 to 3 g/L) and acetic acid (0.2 to 0.7 g/L). YNB without inhibitors was used as control medium for cell growth.

Biomass and lipid production on PSH
PS were washed thoroughly with water, dried in oven (60 °C for 48 h) and crushed in grinder. ~ 20 g of dried powder was subjected to dilute acid (2% v/v) as well as alkaline (2% v/v) treatments in autoclave at 121 °C for 90 min. The liquid fraction of acid treatment was detoxified by activated charcoal (15% w/v) at 30 °C for 3 h. Solid fraction of both acid and alkali treatments were neutralized and subjected to enzymatic saccharification using 20 FPU of cellulases/g (Sigma, USA) of biomass at 50 °C, 150 rpm for 72 h [57]. The hydrolysate was filtered and used as cultivation medium after supplementation of micro-nutrients such as NH 4 SO 4 (0.5 g/L), MgSO 4 (1.5 g/L), KH 2 PO 4 (1.5 g/L). In parallel, YNB supplemented with 46.5 g/L of sugars (glucose 28 g/L, xylose 18.18 g/L, and arabinose 0.3 g/L) was used as control for cell growth and lipid accumulation. Lipid production was carried out in Erlenmeyer flasks (250 mL) containing 100 mL medium by adding overnight grown cells of OD 0.2 and incubating at 30 °C, 200 rpm for 196 h with initial pH of 6.8. Cells were harvested by centrifugation, washed with MQ water, lyophilized (Labconco, USA) and DCW (g/L) was determined. Lipids were extracted from lyophilized cells by using modified Bligh and Dyer method [58], lipid titre (g/L) was measured gravimetrically. Biomass productivity (g/L/h), lipid productivity (g/L/h) and lipid contents (%) were calculated as described previously [52].

Lipid analysis
Lipid analysis was performed by confocal microscopy (Nikon, India) after staining the cells with Bodipy dye (0.5 µg/mL DMSO) [21]. Cell sizes and LD sizes were measured by using Nikon software. TAG analysis was performed on TLC plates (Merck, India) with triolein as standard in hexane: diethyl ether: acetic acid (85:15:1, v/v/v) solvent system. TLC plates were immersed in methanolic-MnCl 2 solution, dried and heated at 120 °C (20 min) [59]. The TAG estimations were performed by using Image-J software.

Analytical methods
Sugar (glucose, xylose, arabinose) and inhibitor (5-HMF, furfural and acetic acid) concentrations in PSH were determined by using HPLC (Agilent 1260 Series) equipped with Aminex HPX-87H column (Bio-Rad, USA) and refractive index (RI) detector. The mobile phase H 2 SO 4 (4 mM) at a flow rate of 0.3 mL/min at column temperature of 40 °C and the sugar as well as inhibitor were quantified by dividing the peak area of the sample with the peak area of standard (1.0 g/L) at specific retention time [28]. Trace elements were determined by inductively coupled plasma-induced ion chromatography-mass spectroscopy (ICP-MS) analysis (Agilent 7900) using argon as carrier gas and sample flow rate was 2.0 mL/min with approximately 2.5 min total analysis time per sample. The samples were acidified with nitric acid to pH below 2.0 and filtered through a 0.45-μm pore diameter membrane filter. The calibration curves were prepared by diluting ICP multi-element standard solution, including the blank [52]. The protein concentration was estimated by Bradford method [60]. Holocellulose, cellulose and xylan content were determined by using standard Association of Official Analytical Chemists (AOAC) methods of analysis [61]. The dried PS powder was also subjected to energy dispersive X-ray (EDX) elemental analysis.

Transesterification and GC-MS analysis
Transesterification was performed by previously described method [62] with some modifications. Briefly, lyophilized yeast cells and 6% methanolic-H 2 SO 4 in 1:20 ratio were taken in teflon-sealed tube and heated at 80 °C for 1 h. FAMEs were extracted into hexane phase and analysed by GC-MS (7890A series) equipped with Omegawax (30 m × 0.25 mm ID, 0.25 µm thickness) and Agilent 7000 QQQ MS [63]. Identification and quantification of FAMEs were performed by NIST (National Institute of Standards and Technology) mass spectral database, AMDIS (Automated Mass Spectral Deconvolution and Identification System) and mass hunter software. Physical properties of biodiesel were computed by using previously reported experimental equations [49] and collated with rape seed oil methyl ester, jatropha oil methyl ester and to EN 14214, ASTM D6751 and Indian standards IS156907 [64].

Statistical analysis
All experiments were performed in minimum three replicates. Average values with standard deviation were mentioned. One-way ANOVA test with post hoc analysis by Tukey's test was performed using Microsoft Office Excel 2013 (Microsoft, USA) to analyse statistical significance of the results. Statistical differences at P ≤ 0.05 were considered as significant.