SYNTHESIS AND CHARACTERIZATION OF FATTY ACID METHYL ESTER MIXTURES DERIVED FROM ACORN KERNEL OIL
Biodiesel fuels are produced via transesterification of a triacylglycerol (TAG, e.g. vegetable oil, waste cooking oil, or animal fats) with a short-chain alkyl alcohol in the presence of a suitable catalyst. Alternative TAG sources, ones not derived from plants used as human food sources, have been of particular recent interest. In this work, the oil extracted from the endosperm of acorns, acorn kernel oil (AKO), was used as an alternative TAG source for the synthesis of biodiesel fuels. Acorns were collected from various species of oak trees (Quercus spp.) in the city of Nacogdoches, Texas. AKO was extracted from the acorn endosperm. The AKO was then subjected to acid-catalyzed and base-catalyzed transesterification with methanol and ethanol to produce acorn kernel oil methyl esters (AKOME) and acorn kernel oil ethyl esters (AKOEE) respectively. Concentrated H2SO4 was used as the acid catalyst and K2CO3 was used as the base catalyst. The effect of using a room temperature ionic liquid on percentage conversion for base-catalyzed transesterification was also investigated. Product mixtures were characterized using 1H-NMR spectroscopy. The NMR data were used to confirm the presence of transesterified products as well as to quantify the percentage conversion for the reaction. Percent conversion results ranged from 96 to 98% for AKOME products and 96 to 97% for AKOEE products.Abstract
Environmental and economic concerns have led to the search for alternatives to fossil fuels. Biodiesel is a possible alternative to petroleum-based diesel fuels. Biodiesel fuels can be synthesized via transesterification of a fat or triacylglycerol (TAG) and an alcohol yielding a mixture of fatty-acid alkyl esters (FAAE), as depicted in Figure 1.
Citation: Texas Journal of Science 71, 1; 10.32011/txjsci_71_1_Article8
In Figure 1, when R4OH is methanol, the result of this transesterification process is a mixture of fatty-acid methyl esters (FAME). However, when R4OH is ethanol, the result is a mixture of fatty-acid ethyl esters (FAEE). The transesterification reaction can be performed using acid catalysis, base catalysis, or enzyme catalysis (Knothe & Van Gerpen 2005). A variety of TAG sources can be used for biodiesel production. Common TAG sources include canola (rapeseed) oil, soybean oil, peanut oil, coconut oil, palm oil, pork lard, beef tallow, and many others (Pinto et al. 2005). The resulting increase in production of biodiesel fuels has brought forth a number of interesting and, potentially contentious, issues. One issue in particular is the “food vs. fuel” dilemma, i.e. a hectare of arable land that is used to raise crops destined for biofuel production is a hectare of land that is not being used to produce crops for food. This is a particularly serious issue in regions in which arable land is limited (Thompson 2012). The debate over whether or not arable land should be used for the production of crops for food or the production of crops for fuel has stimulated the search for alternative TAG sources for biodiesel production. Stephen F. Austin State University (SFASU) is located in the middle of the Piney Woods region of East Texas; hence, the surrounding forests provide a rich source of potential TAG feedstocks for biodiesel synthesis. This report presents the initial results of a proof-of-concept study into the viability of using acorns, the fruit of oak trees (Quercus spp.), as a TAG source for biodiesel production.
Oak trees (Quercus spp.) are a common type of tree found in many parts of the world. The Quercus genus consists of approximately 600 species of trees and shrubs widely distributed over the Northern Hemisphere (Bainbridge 1986a). Acorns are an important food source for a variety of animals including squirrels, swine, deer, bears, and many others (Cantos et al. 2003). Throughout human history, acorns have been used as a food source, for example in ancient Greece, ancient Iberian cultures, Native American tribes in California, and are still used as a food source in Korean culture (Bainbridge 1986b, Meyers et al. 2006). Acorns are not, however, presently a primary human food source in many parts of the world. Moreover, many species of Quercus produce acorns that are unsuitable for use as human food. This is primarily due to such acorns having a high tannin content (Ofcarcik & Burns 1971). Acorns do have a relatively high fat content (approx. 30 g total fat per 100 g of acorn) irrespective of whatever their tannin content might be (USDA 2018). The mixture of TAGs extracted from acorns is commonly referred to as acorn kernel oil (AKO).
Acorns present an intriguing potential TAG source for biodiesel production, as they are plentiful and inexpensive. Some initial reports of biodiesel production using AKO from oak trees native to Turkey have been published (Karabaş 2013a; 2013b; 2014). A number of studies in which the fatty acid composition of acorns from various species of Quercus have been published (Bernardo-Gil et al. 2007; Al-Rousan et al. 2013).
The use of 1H-NMR spectroscopy as a means of quantifying percentage conversion from TAG to FAME and FAEE in biodiesel synthesis has been reported (Gelbard et al. 1995; Knothe 2000; Knothe 2001; Faraguna 2017). FAME mixtures are quantified using integration data from the signal for the methoxy protons of the esterified product, which is a singlet at ~3.6 ppm (Figure 2, peak a). The integration value for the methoxy signal is compared to that for the α-methylene protons of the fatty acid chain, which is seen as a triplet at ~2.3 ppm (Figure 2, peak b). Additionally, the signal for olefinic protons of any unsaturated fatty acid chains that comprise a portion of the FAME products can be seen at ~5.3 to 5.4 ppm (Figure 2, peak c).
Citation: Texas Journal of Science 71, 1; 10.32011/txjsci_71_1_Article8
Usage of the integration of the α-methylene signal is crucial to the NMR analysis. In addition to accounting for FAME products, the α-methylene signal also appears in the spectrum if any triacylglycerol, diacylglycerol, or monoacylglycerol species are present in the product mixture. Thus, comparison of the integration values of the methoxy signal to the α-methylene signal allows for the percent conversion to be determined as noted in equation 1,
where IME is the integration value for the methoxy signal at 3.6 ppm and Iα–CH2 is the integration value for the α-methylene signal. The coefficients of 2 and 3 in equation 1 are normalization factors that take into account the fact that the methoxy group has three protons, whereas the α-methylene group has only two protons.
NMR quantification of FAEE products is different. The ethoxy group of the FAEE product shows two signals: a triplet for the methyl group at 0.9 ppm and a quartet for the methylene group at 4.15 ppm. NMR analysis of FAEEs is complicated due to the fact that it is more difficult to locate a signal in the spectrum that is uniquely attributable to the FAEE product. In spite of the analytical difficulties, approaches to quantification of FAEE mixtures using 1H-NMR spectroscopic data have been proposed in the literature (Garcia 2006; Ghesti et al. 2007; de Jesus et al. 2015). The method proposed by Garcia (2006) uses a portion of the signal for the methylene protons of the ethoxy group as a basis, and is shown in equation 2,
where IC4 is the integration value of the peak in the 4.07 to 4.08 ppm range and IDD+EE is the integration value of all peaks within the 4.05 to 4.35 ppm range. The coefficient of 8 in Eq. 2 is a normalization factor arising from the fact that the integration value from the 4.05 to 4.35 ppm range represents ⅛ of the methylene protons of the ethoxy group of the FAEE product. This methodology has been applied to analysis of FAEE products in several studies (Paiva et al. 2013; Carvalho et al. 2015). An example of the peaks and integration values used in the analysis of 1H-NMR spectral data of FAEE products is provided in Figure 3. These quantification methods were applied to the analysis of AKOME (acorn kernel oil methyl ester) and AKOEE (acorn kernel oil ethyl ester) product mixtures.
Citation: Texas Journal of Science 71, 1; 10.32011/txjsci_71_1_Article8
Materials & Methods
Experimental Information.–Methanol (ACS Reagent Grade) was purchased from VWR Chemicals BDH; absolute ethanol (USP Grade) was purchased from VWR Koptec; K2CO3 (ACS Reagent Grade) was purchased from Fischer Scientific; [BMIM+ [BF4]− (1-butyl-3-methylimidazolium tetrafluoroborate, ≥98 %), CDCl3 (99.8 atom % D with 0.1 % (v/v) TMS), and TMS (ACS Reagent NMR grade) were purchased from Sigma-Aldrich. All reagents were used without any further purification. A Jeol ECS-400 FT-NMR spectrometer was used for all NMR analyses. NMR data were processed using the SpinWorks freeware program (version 4.2.8), published by the Department of Chemistry of the University of Manitoba, Manitoba, Canada (Chemistry NMR Lab). All NMR spectra were recorded using CDCl3 as solvent with TMS as an internal chemical shift standard.
Extraction of AKO from acorns.–East Texas is home to a wide variety of species of oak trees including white oak (Quercus alba), live oak (Quercus virginiana), shumard oak (Quercus shumardii), and many others. These oak trees produce acorns, which begin to fall to the ground in mid-October. Acorns were gathered by hand from various locations around the SFASU campus and around the city of Nacogdoches, Texas. No effort was made to separate acorns by species of Quercus. The collected acorns were then dried at 70°C for at least 24 h, the shells (pericarp) were removed, and the endosperm material was pulverized using a kitchen blender. The TAGs were extracted from the endosperm material using heptane (Hopkins & Chisholm 1953). A sample of endosperm material was placed in a 250-mL round bottom flask and 150 mL of heptane were added, along with a stir bar. The system was heated, with stirring, at reflux for 4–6 h and then allowed to cool to room temperature. The solid endosperm material was removed using simple gravity filtration. Fresh endosperm material was then added to the heptane and heating was repeated for an additional 4–6 h. This process was repeated 4–6 times in order to accumulate a sufficient amount of AKO in the heptane to be used for subsequent transesterification reactions. Once the extraction process was completed, the solvent was removed using rotary evaporation. The AKO was then stored in a capped vial until needed.
Transesterification of AKO.–The process of transesterification is different than that of esterification. In esterification, two species undergo a reaction in which a new ester moiety is created. This can be done using a carboxylic acid and an alcohol, as in the classic Fischer Esterification, or by reacting an acid chloride with an alcohol. Transesterification, on the other hand, begins with a substrate that already contains one or more existing ester moieties. In transesterification, the alkoxy group on the ester is replaced by a different alkoxy group. In this work, a triacylglycerol, which is composed of three fatty acid groups that share a common glycerol backbone, is allowed to react with methanol (or ethanol) and the methoxy (or ethoxy) group takes the place of the glycerol backbone. Thus, three methyl esters (or ethyl esters) are formed in the process. A larger, more structurally-complex ester is converted into three simpler esters.
Acid-Catalyzed AKOME Synthesis.–A 100-mL round bottom flask was charged with AKO (3.00–8.70 g), methanol (40–50 mL), 1 mL of concentrated H2SO4, and a magnetic stir bar. The system was heated at reflux for 16–18 h. After heating, the mixture was allowed to cool to room temperature. The reaction mixture was transferred to a separatory funnel and washed with two 25 mL portions of saturated aqueous NaHCO3, two 25 mL portions of saturated aqueous NaCl, and then two 25 mL portions of deionized H2O. The organic phase was then separated and dried using anhydrous MgSO4. The product mixture was then filtered into a 100-mL round bottom flask. Unreacted methanol was removed by rotary evaporation for 30 min. The purified product was then weighed and stored in a capped vial until it was able to be analyzed using NMR spectroscopy. In a typical reaction, 4.582 g of AKO yielded 4.504 g of AKOME, 98.3% mass conversion.
A similar process was used for the transformation of AKO to AKOEE, resulting in a 97.3% mass conversion.
Base-Catalyzed AKOME Synthesis.–A 100-mL round bottom flask was charged with AKO (3.0–7.8 g) and a stir bar and pre-heated for 5–10 min. After pre-heating, methanol (50 mL) and K2CO3 (20–60 mg, approx. 0.7% (w/w) with respect to AKO) were added to the flask. The system was then heated at reflux for 16–18 h. At the end of the reflux period, the mixture was allowed to cool to room temperature. The product mixture was then transferred to a separatory funnel. The mixture was washed with two separate 25 mL portions of commercially-available white vinegar (to neutralize unreacted K2CO3), two separate 25 mL portions of saturated aqueous NaCl, and two separate 25 mL portions of deionized H2O. The organic phase was then separated and dried using anhydrous MgSO4. The product mixture was then filtered into a 100-mL round bottom flask. Unreacted methanol was removed by rotary evaporation for 30 min. The purified product was then weighed and stored in a capped vial until it was able to be analyzed using NMR spectroscopy. In a typical reaction, 5.207 g of AKO yielded 5.087 g of AKOME, 97.7% mass conversion.
A similar process was used for the transformation of AKO to AKOEE, resulting in a 96.9% mass conversion.
Base-Catalyzed AKOME Synthesis with Ionic Liquid Catalyst.–The procedure for base-catalyzed AKOME synthesis using an ionic liquid catalyst was identical (reaction, purification, and isolation) to that specified in the Base-Catalyzed AKOME Synthesis section, except that [BMIM]+[BF4]− (5 mol % based on the mass of K2CO3 used) was added along with the methanol and K2CO3.
Base-Catalyzed AKOEE Synthesis with Ionic Liquid Catalyst.–The procedure for base-catalyzed AKOEE synthesis using an ionic liquid catalyst was identical (reaction, purification, and isolation) to that specified in Base-Catalyzed AKOME Synthesis section, except that [BMIM]+[BF4]− (5 mol % based on the mass of K2CO3 used) was added along with absolute ethanol and K2CO3.
Results & Discussion
The attempted transesterification of AKO was successful. Methanolysis of AKO yielded AKOME products with excellent conversion. Ethanolysis of AKO yielded AKOEE products, also with excellent conversion. The results are summarized in Table 1.
The use of 1H-NMR spectroscopic data for qualitative and quantitative analysis of AKO biodiesel mixtures worked well. The NMR spectrum of the AKO feedstock contains several distinct signals that do not appear in the spectra after transesterification. Typical 1H-NMR spectra of the neat AKO feedstock, AKOME product mixture, and AKOEE product mixture are shown in Figure 4.
Citation: Texas Journal of Science 71, 1; 10.32011/txjsci_71_1_Article8
Acid-Catalyzed Transesterification.–Acid-catalyzed trancesterification of AKO worked well for synthesis of both AKOME and AKOEE products (Table 1, Entries 1 & 4). The only drawback to the use of acid-catalyzed transesterification is that the reaction itself is considerably slower under these conditions, reaction times of 18–24 h are typically necessary. The relatively slow rate of reaction for acid-catalyzed transesterification is a limitation imposed by the nature of the reaction mechanism associated with the need to activate the acyl group, and this topic has been well-discussed (Otera 1993).
Base-Catalyzed Transesterification.– Synthesis of AKOME from AKO using base-catalyzed transesterification yielded good results (Table 1, Entries 2 & 6). Initial attempts at base-catalyzed transesterification using AKO and 95% ethanol yielded no detectable transesterification product (Table 1, Entry 5). Water is known to inhibit transesterification under base-catalyzed conditions (de Oliveira et al. 2005). The use of the RTIL catalyst for the base-catalyzed transesterification reactions did not appear to influence yield of AKOME or AKOEE percent conversion significantly (Table 1, Entries 3 & 7). The putative benefit of using an RTIL like [BMIM]+[BF4]− in biodiesel synthesis is that the solubility of the alkoxide anion in the low-polarity triacylglycerol reactant phase is increased, thus raising the effective concentration of the nucleophile in the triacylglycerol (Hallett & Welton 2011). In this initial study of the feasibility of using AKO as a biodiesel feedstock, reactions were allowed to proceed over a fairly long period (16–18 h). Attempts are underway to determine if similar conversions can be obtained over shorter periods of time. It is possible that the effect of the RTIL might be more noticeable when the reaction is performed over a 2–3 h period.
As noted above, one advantage offered by the base-catalyzed transesterification using potassium carbonate is the reduced risk of saponification. The problems caused by saponification manifest themselves during the purification process and resulted in the suggestion of a multi-step procedure (Karabaş 2013b). Since this study was an initial attempt to determine if acorns could serve as a potential feedstock for biodiesel synthesis, the reduced time and labor demands of our process were deemed superior.
When potassium carbonate is used as a base, however, there are no complications from saponification. The carbonate anion does not undergo nucleophilic addition with the ester moiety of the triacylglycerol; its function in the reaction system is able to deprotonate methanol or ethanol. This produces a low, steady-state concentration of alkoxide anion, which can then undergo nucleophilic addition with the triacylglycerol substrate. Moreover, no insoluble carboxylate salts are formed to complicate the separation process. Any unreacted carbonate ion present at the end of the reaction can be quenched by adding commercially-purchased vinegar. The AKOME product is then purified by a series of washing steps to remove impurities and unreacted materials. No centrifugation is necessary. The entire purification process can be performed using an ordinary separatory funnel.
Each method of catalyzing the transesterification reaction has its advantages and disadvantages. The acid-catalyzed method requires a longer reaction time, but the product purification and isolation tends to be cleaner and easier. The base-catalyzed method, on the other hand, requires a shorter reaction time, but the product isolation and purification can be more complicated. The results of this work show that either method is viable for synthesizing biodiesel fuels from acorns on a laboratory scale. This allows for the fuels to be tested, characterized, and studied. The reaction and purification methodologies would be considerably different if the process were to be scaled up with larger-scale biodiesel production in mind.
Conclusions
This work represents a proof-of-concept that it is indeed possible to synthesize biodiesel fuels using AKO from acorns collected from oak trees in East Texas. Acorns were gathered, shelled, and the AKO was extracted from the acorn endosperm material. Transesterification of the AKO from the acorns using methanol yielded AKOME in excellent yields under acid-catalyzed (98.3% conversion) and base-catalyzed conditions (97.7% conversion). Similarly, trancesterification of the AKO using ethanol also yielded AKOEE in excellent yields under acid-catalyzed (96.9% conversion) and base-catalyzed (97.7% conversion) conditions. In all cases, the resulting product mixtures were characterized using 1H-NMR spectroscopy. Integration of the NMR data was used to quantify the percentage conversion values for the reactions utilizing established literature methodology.
The use of acorns as a potential biodiesel source presents several interesting possibilities for future work, including: (1) refinement of the shell removal and AKO extraction procedure to increase its efficiency, (2) repetition of the extraction-transesterification-purification-analysis protocol for individual species of Quercus, (3) determination of free fatty acid content of the AKO feedstock for individual species of Quercus, (4) further analysis using GC and/or GC-MS in order to determine the specific composition of the various ester species in the product mixtures, (5) reduction of reaction time to determine if the use of the RTIL provides a beneficial effect on percent conversion, and (6) determination of combustion energies, viscosities, and cloud point for the AKO ester products.
Synthesis of fatty-acid alkyl ester (FAAE) mixtures via catalytic transesterification of triacylglycerols (TAGs) and alcohols (R4OH).
A portion of the 1H-NMR spectrum (2.0 to 6.0 ppm region) of a typical acorn kernel oil methyl ester (AKOME) product mixture from transesterification of acorn kernel oil (AKO) with methanol. The singlet for the methoxy protons of the ester is seen at approximately 3.60 ppm (a), the triplet for the α-methylene protons of the fatty-acid chain is seen at ~ 2.3 ppm (b), and the signal for the olefinic protons of unsaturated fatty acid chains are seen at ~5.2 to 5.3 ppm (c).
A portion of the 1H-NMR spectrum (3.85 to 4.40 ppm region) of a typical acorn kernel oil ethyl ester (AKOEE) mixture from transesterification of acorn kernel oil (AKO) with ethanol. Quantification is done using the integration value for the C4 peak and the integration value for the 4.05 to 4.35 ppm region of the spectrum.
A comparison of the 1H-NMR spectra of neat AKO, an AKOME mixture, and an AKOEE mixture. Confirmation of successful transesterification is based on (1) disappearance of the glyceridic peaks from the TAG and (2) on appearance of the characteristic ester peaks, depending on whether methanol (methoxy singlet at 3.60 ppm) or ethanol (ester methylene quartet at 4.12 ppm) was used.
Contributor Notes