Article pubs.acs.org/EF
Novel Nanoparticle-Assisted Room-Temperature Synthesis of Methyl Esters from Aloe vera Seed Oil Puran Singh Rathore,† Poonam Mangalorkar,‡ Padamanabhi S. Nagar,‡ M. Daniel,‡ and Sonal Thakore*,† †
Department of Chemistry and ‡Department of Botany, Faculty of Science, The Maharaja Sayajirao University of Baroda, Vadodara 390002, Gujarat, India S Supporting Information *
ABSTRACT: Aloe vera has been used as a cosmetic and medical remedy since ancient times and has gained increasing popularity in recent years. Despite its widespread use, reports on biodiesel from Aloe vera seeds are lacking. The present investigation reports the fatty acid composition of Aloe vera seed oil (AVSO) and addresses the feasibility of using AVSO as a source of biodiesel. A novel ecofriendly catalyst was developed using triacetin as a model. Interestingly, the room-temperature conversion of AVSO and other nonedible oils to methyl esters could be achieved using this novel catalytic system consisting of ethylene diamine in the presence of nickel nanoparticles (NiNPs). The metal core, capping agent, and amine concomitantly contribute to make the system an effective catalyst.
1. INTRODUCTION Biodiesel prepared by the transesterification of vegetable oil with methanol is an alternative fuel that can be used directly in any existing unmodified diesel engine. Because its properties are similar to those of diesel fuel, biodiesel can be blended at any ratio with diesel fuel. Among the various vegetable oil sources, nonedible oils are suitable for biodiesel production, as edible oils are already in demand for food and are much more expensive than diesel fuel. Among nonedible oil sources, Jatropha curcas1 and Derris indica are some of the species identified as potential biodiesel sources suitable for tropical and subtropical regions of the world.2 However, a species already having a high potential market can provide an extra edge to its selection. Aloe vera is one such species that has been exploited for medicinal, nutraceutical, and cosmetic purposes. Today, mostly the aloe gel from the center of the leaves is processed. It primarily consists of polysaccharides to which many medical properties have also been attributed. However, the potential of Aloe vera seed oil (AVSO) from a biodiesel prospective has not yet been established. We have investigated the fatty acid composition of a number of oil species in the past.3a,b In the present study, we investigated the potential of AVSO for the production of biodiesel. The use of a strong base such as KOH and mineral acids leads to the wastage of water and the large-scale generation of effluents.4 Hence, extensive research has been carried out for the development of suitable catalysts for biodiesel production. Solid catalysts such as ZnAl hydrotalcite,5 sulfated zirconia,6 KF/ZnO,7 hydrous zirconia-supported 12-tungstophosphoric acid,8 zinc dodecatungstophosphate (Zn 1.2 H 0.6 PW 12 O 40 ; ZnPW) nanotubes,9 and acid catalyst10 have some shortcomings including high costs and easy deactivation. Solid organic bases have also been used as catalysts for the production of biodiesel.11 However, the recovery of the catalyst was tedious and expensive. Instead, it is advantageous to use low-boiling amines, which have also exhibited good catalytic activity and have a simpler recovery process.12,13 However, the © 2013 American Chemical Society
vigorous conditions reported previously for such catalysts are not commercially and economically feasible. Because our group has been actively engaged in the synthesis and applications of metal nanoparticles,14a−d we decided to develop an easy method for the nanoparticle-assisted organic-amine-catalyzed synthesis of methyl esters. Currently, metal nanoparticles (NPs) are used widely in many reactions,15 with advantages such as higher specific surface, lower mass-transfer resistance, easy separation, and less fouling than for other catalysts. The high efficiency of a nanoparticle system relies mainly on the approach to the metal core and the structure of the surface. Although a number of organic reactions have been catalyzed, the use of NPs in transesterification reactions has not been extensively reported. In an effort to identify catalyst characteristics that would be ideal for biodiesel synthesis, this study compared the catalytic activities of some organic amines in presence of nickel nanoparticles (NiNPs) with that of the conventional catalyst KOH. Initially, triacetin was used as the model system to simplify the analysis and to accelerate the screening for suitable amine−NP catalytic systems. Consequently, the optimized conditions were used for the synthesis of biodiesel from some recognized oils16a−e,17 and finally applied to AVSO. The products of the new catalyst system were also compared against those obtained from conventional KOH-catalyzed reactions. To our knowledge, this is the first report on the synthesis of biodiesel from AVSO using a novel catalytic system that works at room temperature.
2. EXPERIMENTAL SECTION 2.1. Materials. Nickel acetate [Ni(CH3COO)2·4H2O], soluble starch, sodium borohydride (NaBH4), liquid ammonia, triacetin, KOH, H2SO4, triethylamine (TEA), ethylenediamine (EDA), diethylReceived: November 29, 2012 Revised: April 8, 2013 Published: April 8, 2013 2776
dx.doi.org/10.1021/ef301950j | Energy Fuels 2013, 27, 2776−2782
Energy & Fuels
Article
amine (DEA), methanol, petroleum ether, and chloroform were purchased from Merck, Mumbai, India. Tulsion T-45 BD ion-exchange resin was purchased from Thermax Limited, Pune, India. All solutions were prepared using doubly distilled and demineralized water. 2.2. Synthesis of Nickel Nanoparticle (NiNPs). NiNPs were synthesized through a wet chemical reduction process using 1 M solution of Ni(CH3COO)2·4H2O and soluble starch (1%, w/v) as the metal salt precursor and stabilizing agent, respectively. NaBH4 (10%, w/v) was used as the reducing agent, and liquid ammonia was used as the complexing agent. In a typical process, 10 mL of 1% starch solution and 0.4 mL of 1 M Ni(CH3COO)2·4H2O were mixed together and stirred with a magnetic stirrer at room temperature. The pH of the solution was adjusted to 10 by adding liquid ammonia, and the initial green color of the solution changed to Prussian blue. Finally, 0.6 mL of 10% NaBH4 solution was added dropwise with continuous stirring until a black colloidal suspension was formed. The NPs were isolated by centrifugation carried out at 10000 rpm for 10 min, washed with acetone, and dried at 50 °C under a vacuum. 2.3. Aloe vera Seed Collection and Oil Extraction. Aloe vera seeds were collected from various parts of Gujarat (Saurashtra and Kutch regions) during the favorable season between February and March. All seeds collected were already matured and dried. The seeds were separated from the fruit. Each plant has 15−20 fruits on the spike, and each fruit contains an average of four to eight seeds, which are extremely lightweight. The oil from the seeds was extracted using petroleum ether as the solvent in a Soxhlet apparatus. 2.4. Transesterification Reaction. The molar ratio of methanol to triacetin was taken as 20:1. In a typical reaction, methanol (4 mL), amines (6 wt % of substrate), and NiNPs (150 mg) were sonicated at room temperature (25−30 °C) for 30 min in a round-bottom flask, and then 1 g of the substrate triacetin was added. All reactions were carried out at room temperature initially. In cases where the reaction did not proceed to completion, the mixture was gradually heated to reflux (at 70 °C) with the aim of achieving maximum conversion. Reaction monitoring was done by thin-layer chromatography (TLC). After the completion of the reaction, the NPs were separated by centrifugation. The yield and purity of the final products of the triacetin reaction were analyzed by high-performance liquid chromatography (HPLC) The transesterifications of AVSO and other oils were carried out in a similar manner. After the maximum conversion of the reaction, the NPs were separated from the reaction mixture by centrifuging along with the glycerol phase. The organic amine and unreacted methanol were removed by distillation to get the final product. About 2−3 mL of methanol could be recovered by distillation at 75−80 °C. The isolated NiNPs were washed with water to remove glycerol and then with petroleum ether to remove wax. Finally, after an acetone wash, the NiNPs were dried at 60 °C under a vacuum and used for the next reaction. Transesterification with KOH (3 wt %) was carried out by the conventional method, and the product was obtained by following the usual workup process. After the transesterification, the glycerol byproduct was separated from the resulting biodiesel in a separatory funnel. The excess methanol was removed using a rotary evaporator under reduced pressure. The methyl ester was acidified with dilute H2SO4 to neutralize the excess residual KOH and extracted with chloroform. This was followed by water washings with warm distilled water (50 °C) in a separatory funnel until the pH of the aqueous phase was around 7. The chloroform was removed by evaporation, and the product was dried over anhydrous Na2SO4. Further, purification of the methyl esters was done by passing the mixture through an ionexchange resin (Tulsion T-45 BD) column (1.5 × 20 cm2) that retained the impurities such as residual water, ions, and glycerol. The thus-purified product was compared with standards using gas chromatography (GC). Calibration of the gas chromatograph was done using methyl linoleate, methyl palmitate, methyl stearate, methyl oleate, 1-oleoyl-rac-glycerol, dioleoylglycerol, and triolein purchased from Sigma. n-Pentadecane was used as the internal standard. The areas of GC peaks for individual fatty acid methyl esters obtained from the KOH process were calculated, and their cumulative content was
assumed to be the maximum yield (100%) corresponding to the maximum conversion of AVSO. The yield of methyl esters in NPcatalyzed experiments was expressed in terms of the percentage of methyl esters produced (as analyzed by GC) according to the calibration method reported by Zieba et al.18 for the methanolysis of castor oil. Mixtures were prepared with known amounts of AVSO methyl esters (obtained from KOH process), AVSO, and methanol in proportions corresponding to various conversions of triglycerides from 5% to 100%. From the GC peak areas obtained corresponding to various conversions, a calibration plot was constructed. This plot was used for the calculation of the yield of methyl esters obtained using the EDA−NiNP system as the catalyst and expressed in terms of the percentage of methyl esters produced. 2.5. Characterization Methods. The starch-capped NiNPs were characterized by X-ray diffraction (XRD) using a PANalytical X’PERTPRO instrument with Cu Kα radiation (λ = 0.15406 nm) at a scanning rate of 2°/min for 2θ ranging from 10° to 70°. The size and shape of the NiNPs in solution were determined by transmission electron microscopy (TEM) on a Philips Technai 20 microscope operating at 200 kV. Samples for TEM were prepared by putting one drop of the suspension onto standard carbon-coated copper grids and then drying under an IR lamp for 30 min. Dynamic light scattering (DLS) was carried out on a 90 Plus DLS unit from Brookhaven (Holtsville, NY). Thermogravimetric analysis (TGA) was performed on a TG-DTA 6300 INCARP EXSTAR 6000 instrument in a nitrogen atmosphere in the temperature range of 30−450 °C at a heating rate of 10 °C/min. FTIR spectra of the NiNPs were recorded as KBr pellets on a PerkinElmer RX1 spectrometer in the range of 4000−400 cm−1. The AVSO composition was determined by gas chromatography− liquid chromatography (GC−LC). Two samples were taken for the analysis of the fatty acid composition. The methyl esters of the extracted oil were prepared in accordance with the Bureau of Indian Standards (BIS:548, part III).19 A NUCON-GLC chromatograph with a flame-ionization detector was employed for the analysis using nitrogen as the carrier gas. The column used was a 30 m × 0.53 mm i.d. 5.0-μm DB-1 type MXT-1 capillary column. The column oven temperature was programmed from 80 to 280 °C (at a rate of 10 °C min−1) with injector and detector temperatures of 250 and 280 °C, respectively. The total run time was 40 min. Each component was identified by comparing its retention time with that of a Sigma-Aldrich standard fatty acids mixture. The free fatty acid (FFA) content was determined by a standard titrimetry method.20 Transesterification reactions were monitored by TLC (silica gel 60 F254, Merck, Mumbai, India). The solvent system consisted of hexane/ ethyl acetate (1:1 v/v) for triacetin, hexane/ethyl acetate/acetic acid (95:5:1 v/v/v) for AVSO reaction and (90:10:1 v/v/v) for other oils.17 The spots were detected in iodine chamber. The products of the triacetin reaction (methyl acetate and intermediates) were analyzed by HPLC (Agilent 2000) with a C18 XDB column and a diode array detector. The mobile phase was methanol/water (50:50) and was tested at 205 nm with a flow rate of 1.0 mL/min and a column temperature of 35 °C. Other oils such as Jatropha (Jatropha curcas), Jyotishmati (Celastrus paniculatus), Bacuachi (Psoralea coryfolia), Saragava (Moringa oleifera), Mahuva (Madhuca indica), and Salvadora (Salvadora oleifera) were obtained from a local store. The methyl esters obtained from the transesterification of oils were confirmed by gas chromatography on a Perkin-Elmer Clarus 500 GC instrument with a flame-ionization detector. The capillary column used (70% phenyl polysilphenylenesiloxane) had a length of 30 m with an internal diameter of 0.25 mm. Nitrogen was used as the carrier gas at a constant flow rate. The column oven temperature was programmed from 150 to 250 °C (at a rate of 10 °C min−1) with injector and detector temperatures of 240 and 250 °C, respectively. Finally, the product methyl esters were also analyzed by FTIR (Perkin-Elmer RX1 model) and 1H NMR (Bruker 400 MHz) spectroscopies. Kinematic viscosity measurements were carried out using a Scott Gerate AVS 350b capillary viscometer.21 The viscosity obtained in centipoise was converted to centistokes. The density and specific gravity were measured by means of a pycnometer. 2777
dx.doi.org/10.1021/ef301950j | Energy Fuels 2013, 27, 2776−2782
Energy & Fuels
Article
Figure 1. (a) XRD pattern, (b) DLS analysis, and (c) TEM image of NiNPs.
Figure 2. (a) FTIR spectra and (b) TGA degradation curve of NiNPs.
1620 cm−1 because of the tightly bound water present in the starch. The peak at 1620 cm−1 shifted to 1640 cm−1 in the NiNPs. The shifts observed in the spectra can be attributed to the interaction of the NiNPs with starch. The band at 2910− 2930 cm−1 is characteristic of C−H stretching. A broad band due to hydrogen-bonded hydroxyl groups (O−H) appeared at 3400−3420 cm−1 and is attributed to the complex vibrational stretching, associated with free and inter- and intramolecularly bound hydroxyl groups. The degradation pattern observed by TGA (Figure 2b) also supports the presence of organic matter on the surface of the NiNPs.14a 3.2. Properties and Composition of AVSO. The percentage of seed oil in Aloe vera was found to vary between 20 and 22 wt %. The fatty acid composition is described in Table 1. The unsaturated fatty acids were approximately 85.59% (linoleic acid, 69.63%; oleic acid, 15.96%). Saturated fatty acids accounted for approximately 14% of the total
3. RESULTS AND DISCUSSION 3.1. Characterization of Nickel Nanoparticles (NiNPs). The XRD pattern of NiNPs in Figure 1a clearly shows NiNPs embedded in an amorphous starch matrix. All of the reflection peaks can be indexed as face-centered-cubic NiNPs (Supporting Information). From the DLS analysis in Figure 1b, it was observed that the particles were in the range of 70−95 nm in diameter with a narrow size distribution. The average particle size was confirmed as 75 ± 5 nm by TEM (Figure 1c). The NPs had a somewhat spherical-shaped morphology and were monodispere in nature. The FTIR spectra (Figure 2a) of the NiNPs and pure starch both display the typical profile of polysaccharides in the range of 920−1100 cm−1 (characteristic peaks attributed to C−C/C− O bond stretching). The peaks at 1020−1100 cm−1 are characteristic of the anhydroglucose ring. The peaks at 1404− 1420 cm−1 are due to C−H bending. Other peaks appeared at 2778
dx.doi.org/10.1021/ef301950j | Energy Fuels 2013, 27, 2776−2782
Energy & Fuels
Article
refluxing for 8 h. The order of catalytic activity, namely, EDA > DEA > TEA, is due to the decrease in basicity from primary to tertiary amine and the increase in steric hindrance. Similar results were reported by Wang et al.,12 for the transesterification of crude rapeseed oil in supercritical methanol (>300 °C). To enhance the rate of transesterification at room temperature, NiNPs were used along with the amines. Significant enhancement in reaction was observed for the DEA- and EDAcatalyzed reactions, which were complete within 6 and 3 h, respectively (Table 3, entries 7 and 10). On the other hand, the TEA- and KOH-catalyzed reactions were less affected by the NiNPs. To establish optimum conditions, a preliminary study was carried out using EDA as the catalyst. After several experiments with different concentrations of EDA and methanol and various quantities of NiNPs, the optimum conditions for transesterification were established. The precise mechanism of this transesterification process is not very clear to us. The reactions in the absence of the NPs were observed to be quite slow. To understand the role of the metal, the reaction under the optimized conditions was also carried out with starch-capped copper nanoparticles (CuNPs)14d instead of NiNPs to yield similar results (Table 3, entry 12). The higher efficiency of metal NPs under mild conditions seems to be due to their higher dispersion in the solvent, so that the reactant reaches the catalytic site by diffusion. Possibly, the role of metal NPs is to provide surface binding sites for the reactant (methanol) and milder catalyst (amine) during sonication. This facilitates the generation of nucleophile, namely, methoxide ion, with the help of a weak base,24 such as EDA, which then attacks triacetin as shown in Scheme 1. Further, it is possible that hydrogen bonding between hydroxyl groups of the capping agent starch and methanol also assists in the generation of methoxide ion on the surface of the NiNPs.25 The FTIR spectrum (Figure 3a,b) of the recovered NiNPs provides evidence for the surface functionalization of NiNPs with EDA (Supporting Information). Thus, the metal core, capping agent, and amine do not act independently but rather act concomitantly to make the amine−NiNP system an effective catalyst. EDA gave the highest catalytic activity because of its greater basicity, followed by DEA. TEA was probably too weak to generate the nucleophile, whereas KOH was strong enough for the generation of nucleophile in the absence of NiNPs. At the reflux temperature of the methanol, good to high conversions were obtained in the cases of DEA and EDA, and the reaction was complete within 4 and 1 h, respectively (Table 3, entries 8 and 11). This shows that, as the reaction temperature increases, the reaction time decreased. The sequence of addition of reactants was also observed to be important. If substrate triacetin was added before amine, the reaction slowed, probably because of the preferential adsorption of triacetin which blocks the active site of NiNPs. Hence, it is advantageous that amine and methanol come into contact with NiNPs before triacetin. A similar observation was reported by Liu et al.26 for a hydrotalcite-based catalyst. These results clearly indicate that organic amines can effectively catalyze the transesterification of triacetin with methanol. Among the three amines, EDA showed the best catalytic activity at room temperature, which was further enhanced in the presence NiNPs. Encouraged by these results, we decided to explore the potential of this catalytic system
Table 1. Fatty Acid Composition of AVSO fatty acid
content (wt %)
palmitic acid (16:0) stearic acid (18:0) oleic acid (18:1) eicosadienoic acid (20:2) linoleic acid (18:2)
10.6 2.44 15.96 1.37 69.63
content, with the dominant saturated acid being palmitic acid (10.6%). The oil properties are quite similar to those of sunflower oil.22 The oil properties of AVSO are listed in Table 2. The presence of higher oil with linoleic acid (73.73%) and Table 2. Properties of AVSO saponification value iodine value unsaponifiable matter (USM) content acid value kinematic viscosity (mm2/s) free fatty acid (FFA) content
187.2 (mg/g) 118.4 (g/100 g) 2.65% 39.92 mg of KOH /g 27.8 cSt (40 °C) 21.50%
oleic acid accounting for 90% of the unsaturated fatty acid content gives AVSO the unique property of a low melting point and low viscosity. Thus, the methyl esters formed by the oil also have low melting points between −19.8 and −35.0 °C.23 3.3. Transesterification of Triacetin. The transesterification of triacetin was carried out first with KOH at room temperature. TLC monitoring of the reaction showed stepwise transesterification of each ester group of triacetin, indicating the formation of diacetin followed by monoacetin. At the end of the reaction (after 3 h), all the spots monitored by TLC disappeared because of the complete conversion of triacetin into the product methyl acetate and byproduct glycerol, which were not detected by TLC. The complete conversion of triacetin to methyl acetate was confirmed by HPLC against standard samples (data not shown). To assess the efficiency of amines as base catalysts instead of KOH, the reaction was carried out using each of the three amines without any support. Similar results were observed for all three amines, but the reaction times were much longer (Table 3). In the cases of EDA and DEA, the reaction took 8 and 10 h, respectively, at room temperature, whereas in the case of TEA, complete conversion could be obtained only after Table 3. Transesterification of Triacetin entry
amine
NPs
reaction temperature
reaction time (h)
yield (%, ±2%)
1 2 3 4 5 6 7 8 9 10 11 12
KOH − − TEA TEA DEA DEA DEA EDA EDA EDA EDA
− − Ni − Ni − Ni Ni − Ni Ni Cub
RTa reflux reflux reflux reflux RT RT reflux RT RT reflux RT
3 48 48 8 7.30 10 6 4 8 3 1 3
97 0 0 93 93 94 95 95 97 97 97 97
a
Room temperature. bReference 14d. 2779
dx.doi.org/10.1021/ef301950j | Energy Fuels 2013, 27, 2776−2782
Energy & Fuels
Article
Scheme 1. Proposed Reaction Mechanism for NiNP-Supported Organic-Amine-Catalyzed Transesterification Systems
Table 4. Transesterification of AVSO sample 1 2 3 4 5 6 7 a
amine KOH KOH TEA/EDA/ DEA TEA DEA EDA EDA
NPs
reaction temperature
reaction time (h)
yield (%, ±1%)
− − −
reflux RTa reflux
1.0 3.0 8.0
95.0 95.0 3.3
Ni Ni Ni Ni
reflux reflux reflux RT
8.0 8.0 1.25 4.0
4.6 15.5 94.0 95.0
Room temperature.
the role of the NPs. The importance of the addition sequence of raw materials was similar to that for the triacetin reaction. The products obtained from the two processes were identical, as can be seen from the chromatographic and spectroscopic evidence: GC, TLC (Supporting Information), IR spectroscopy (Figure 4), and 1H NMR spectroscopy (Figure 5). Use of organic amines does not produce soap or an emulsion and thus avoids the need for washing procedures. However, in the absence of NPs, the reaction time, temperature, pressure, and methanol-to-oil molar ratio were relatively high in the organic-amine-catalyzed systems.27,12,13 Because the reactions herein were carried out under ambient conditions, the organic amines can also be easily recovered by distillation,28,13 and the NPs can also be recycled. The catalytic activity of the present system was found to be comparable to that of KOH (Table 3, entry 1). 3.5. Recycling of NiNPs. One of the advantages of using NPs as a catalyst is that they can be easily recovered by centrifugation and then recycled. We found that the catalyst could be reused directly without further purification for five consecutive runs. However, the reaction time increased in each successive recycling experiment (Table 5) reaching 3−4 h in the case of triacetin and 4−5 h for AVSO. This might be because the catalytic activity of the NiNPs decreased with the number of runs, which might be due to various reasons. The
Figure 3. FTIR spectra of (a) NiNPs recovered after the first run, showing traces of bound EDA, and (b) EDA alone.
(EDA−NiNPs) for the transesterification of some recognized seed oils to obtain biodiesel. 3.4. Transesterification of Seed Oils. Our results indicate (Supporting Information) that the transesterification proceeds to completion in 3−4 h, except for that of Saragava (Moringa oleifera) oil, for which the KOH process itself took 12 h to reach a maximum even under reflux conditions. The products obtained from the two processes were identical, and the yields were comparable, as can be seen from the chromatographic and spectroscopic evidence (Supporting Information). Next, the catalytic system was applied to the transesterification of AVSO (Table 4). It is evident that the EDA−NiNP system works efficiently for this reaction at room temperature as well as under reflux conditions, and the results were comparable to those obtained with the KOH process (Table 4). Interestingly, the DEA- and TEA-catalyzed reactions were initiated, but did not proceed to completion even after 8 h of reflux in the presence of NiNPs. These results suggest the importance of the basicity of the amine catalyst and also signify 2780
dx.doi.org/10.1021/ef301950j | Energy Fuels 2013, 27, 2776−2782
Energy & Fuels
Article
Figure 6. XRD pattern of NiNPs recovered after the fifth run.
Figure 4. FTIR spectra of methyl esters of AVSO catalyzed by (a) KOH and (b) EDA−NiNPs.
3.6. Characterization of Methyl Esters of AVSO. The methyl esters formed from AVSO by the two processes were identical as confirmed by GC, TLC (Supporting Information), FTIR spectroscopy (Figure 4) and 1H NMR spectroscopy (Figure 5). The FTIR spectrum reveals a strong band at 1377 cm−1 corresponding to the methyl group. The presence of the ester group is confirmed by the band at 1742 cm−1. The band at 2925−2926 cm−1 is characteristic of C−H stretching. The band at 726 cm−1 is due to the presence of more than three consecutive −CH2 groups, as in case of methyl esters. The percentage of free fatty acids (FFAs) is another critical variable that affects the acid value in the transesterification process.33 For biodiesel preparation, the oil should have acid value of
