Article pubs.acs.org/IECR
Efficient Production of Alkanolamides from Microalgae Ajith Yapa Mudiyanselage,† Haoyi Yao,† Sridhar Viamajala,‡,§ Sasidhar Varanasi,‡,§ and Kana Yamamoto*,†,§ †
Department of Chemistry and Biochemistry, University of Toledo, 2801 West Bancroft Street, Toledo, Ohio 43606, United States Department of Chemical and Environmental Engineering, University of Toledo, 1650 Westwood Avenue, Toledo, Ohio 43606, United States § School of Green Chemistry and Engineering, University of Toledo, 2801 West Bancroft Street, Toledo, Ohio 43606, United States ‡
S Supporting Information *
ABSTRACT: Fatty acid alkanolamides (FAAA) are lipid derivatives with industrial applications as biosurfactants and biolubricants. Although conventionally produced from vegetable oils, use of alternative renewable sources that do not compete with the food supply chain, such as microalgae, is desirable. We studied the production of FAAA through direct in situ amidation of algal biomass or by amidation of fatty acid methyl esters (FAME) recovered from in situ transesterification of algae. In situ transesterification resulted in spontaneous formation of a distinct FAME phase, which could be easily recovered and converted to FAAA. With this two-step transesterification-followed-by-amidation method, >95% of algal lipids were recovered as FAAA products. In situ amidation did not result in a separate product phase, likely because of the amphiphilic nature of the product. However, extraction with ethyl acetate allowed recovery of nearly 90% of the biomass lipids as FAAA after in situ amidation.
■
INTRODUCTION The development of algae-based fuels and products has been attracting interest in recent years because of the recognition of the limitation of petroleum resources and increasing global carbon emissions. Use of microalgae as biomass feedstocks has several advantages. Microalgae are fast-growing, can be cultivated on marginal lands using low-quality nutrients and water (e.g., seawater or wastewater), and exhibit high lipid productivities.1b The economic viability of algal biorefineries can be improved through higher-value products, such as oleochemicals, in addition to fuels. As a general rule of thumb, a bulk price of most specialty chemicals is ∼$3/kg, while fuel costs are ∼$1/kg.1b The objective of this study was to develop efficient and economically feasible methods for production of a class of highvalue specialty chemicals, fatty acid alkanolamides (FAAA), to improve the overall economics of algal biorefineries. (A provisional patent which details a part of this work has been filed.) FAAA are lipid derivatives which are found naturally in plants and animal tissues.2 Industrially, they are used primarily as biosurfactants or biolubricants.1b Some alkanolamides have important biological roles such as anti-inflammatory activity,3 attenuation of pain sensation,4 and pro-apoptotic and anorexic effects.5 Apart from its biological functions in living tissues, this class of lipid derivatives is used in personal care products, pharmaceuticals, detergents, rust inhibitors,6 ink formulations,7 and many other applications. FAAA are mainly manufactured from vegetable oil with annual global demand estimated at 90 000 t.1b A number of studies have reported conversion of fatty acid triglycerides (FAG) from terrestrial biomass (i.e., vegetable oil) to alkanolamides.7,8 In general, FAAA can be synthesized from alkanolamine with a fatty acyl donor, such as free fatty acids,9 fatty acid chlorides,10 fatty acid alkyl esters,11 and FAG.7,12 Acyl © XXXX American Chemical Society
chlorides have been used to deliver FAAA products with sufficient purity for biological studies4,13 but are likely unsuitable for industrial-scale production because of cost and the corrosive nature of the reagents. Fatty acid conversion to FAAA can be accomplished using sodium methoxide catalyst but requires harsh reaction conditions due to ionic salt formation.14 Milder conversion is conceivable with boronbased catalyst, although it has not been reported in the context of FAAA synthesis.15 Use of lipase is also possible for the same transformation at lower temperature and is reported to result in better product quality and color.6,9,12a,16 However, enzyme cost and longevity remain a concern for the biocatalyst approach. Most commonly, direct conversion of FAG to FAAA has been studied using ethanolamine as a solvent with7,12a,b or without12c sodium methoxide catalyst. A conversion of FAG to FAAA through fatty acid methyl ester (FAME) has also been reported with analogous reaction conditions and with excellent yields for both steps.11 Although FAAA production from vegetable oils (or their derivatives) is well-known (as described above), use of alternative renewable sources that do not compete with the food supply chain, such as microalgae, is desirable. However, traditional methods used to recover FAG from oil seeds such as mechanical “pressing” are not effective with microalgae because of the microscopic size of cells and relatively tough cell walls.17 For microalgae, FAMEs can be more easily produced through in situ transesterification by reacting the lipid-containing Special Issue: Scott Fogler Festschrift Received: October 10, 2014 Revised: December 20, 2014 Accepted: December 21, 2014
A
DOI: 10.1021/ie503980g Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX
Article
Industrial & Engineering Chemistry Research biomass directly with a mixture of methanol and catalyst.17,18 Transesterification of FAG to FAME can be catalyzed either by a base or an acid.19 However, because algal biomass can contain free fatty acids, acid-catalyzed in situ transesterification is preferred to avoid saponification in the presence of a base.17,18b Both homogeneous (HCl, H2SO4)20 and heterogeneous14a acid catalysts are applicable, with preference to the latter because of ease of catalyst separation and recyclability.14a We have recently developed a method in which in situ transesterification of algal biomass in acidified methanol (containing H2SO4) produces a spontaneously separable FAME phase when reactions are carried out at high biomass concentration.20 As a result, solvent extraction of FAME from methanol and subsequent energy-intense solvent recovery are avoided. In this study, we have extended this approach to use heterogeneous acid catalysts (instead of H2SO4) to allow catalyst recovery. The crude FAME, recovered as a separate phase, was then converted to FAAA. In addition, we have also synthesized FAAA through direct in situ amidation by reacting lipid-containing algae with ethanol amine. To our knowledge, there have been no previous reports describing synthesis and isolation of FAAA from algal biomass or its lipids.
solution (containing 5% (v/v) H2SO4) in a sealed crimp-top GC vial at 90 °C for 2 h using an oil bath (containing silicone oil) to convert algal lipids to FAMEs. After the incubation period was complete, the GC vial was removed from the oil bath and cooled to room temperature. The contents of the GC vial were then carefully transferred to an 8 mL screw-cap vial containing 4 mL of hexane to extract the FAMEs from the acidified methanol reaction mixture. The GC vial was rinsed with an additional 1 mL of hexane to remove any residual solids and also transferred to the 8 mL vial. Extraction of FAME into hexane was performed by heating the screw-cap vials at 90 °C for 15 min. After being cooled, the vials were weighed and evaporation losses, if any, were compensated by addition of fresh hexane (usually 1−2 drops). Finally, an aliquot from the hexane layer was analyzed using GC, and FAME concentrations were quantified with calibration curves of reference standards. The lipid content of microalgae biomass used in this study was determined to be 290−330 mg per gram of biomass of lipid. Alkanolamide Synthesis and Recovery by in Situ Amidation. Freeze-dried microalgae biomass (0.1 g) and ethanolamine (0.5 mL) were added to a 1.5 mL crimp-top GC vial. The vial was sealed and heated at 120 °C in an oil bath for 1.5 h with continuous stirring at 300 rpm. After the reaction period, the vial was cooled to room temperature and the contents were carefully transferred to an 8 mL glass centrifuge tube by first diluting with 1 mL of ethyl acetate and subsequently with three additional rinses with 1 mL of ethyl acetate and 2 mL of water (1 mL × 3). After the transfer, the 8 mL centrifuge tube contained 2 mL of water and 2 mL of ethyl acetate in addition to the reaction mixture. Ethyl acetate formed a separate phase from the reaction mixture and allowed selective extraction of the FAAA. A 0.1 g sample of NaCl was also added to the two-phase mixture to prevent emulsion formation and to minimize the partitioning of glycerol into the ethyl acetate phase. The contents of the centrifuge tube were shaken vigorously to facilitate rapid transfer of FAAA to the ethyl acetate phase and allowed to settle back into two phases. The ethyl acetate layer was recovered and transferred to a round-bottom flask. The aqueous phase was contacted two more times with fresh ethyl acetate (2 mL × 2) to extract any FAAA not recovered during the initial extraction. Pooled ethyl acetate fractions from the three extraction stages were dried using a rotary evaporator. The solid crude FAAA remaining in the flask was then dissolved in a mixture of bis(trimethylsilyl)amine and pyridine (1 mL each) along with catalytic amounts of trifluoroacetic acid (20 μL) for silation and quantification (described in more detail below). Alkanolamide Synthesis by Concurrent in Situ Transesterification and Amidation. This method was analogous to the previous method for in situ amidation, but the reaction was carried out with a mixture of methanol and ethanolamine instead of ethanolamine alone. The following two mixture compositions were used in these reactions: (1) 0.1 mL of methanol and 0.4 mL of ethanolamine and (2) 0.25 mL of methanol and 0.25 mL of ethanolamine Alkanolamide Synthesis by in Situ Transesterification Followed by Amidation. Freeze-dried microalgae biomass (0.1 g) and Amberlyst 15 (40% g/g-biomass) were mixed with methanol (0.5 mL) in a 1.5 mL crimp-top GC vial, which was sealed and heated at 90 °C for 6 h in an oil bath with continuous stirring at 300 rpm. After the reaction was complete, the vials were removed, cooled to room temperature, and centrifuged at 4000 rpm for 10 min. Two well-separated
■
MATERIALS AND METHODS Equipment. 1H NMR analyses were performed using Varian VXRS 400 4 (400 MHz) or Bruker Avance 600 (600 MHz) NMR spectrometers at ambient temperatures. Mass spectrometric analyses were performed using Hewlett-Packard Esquire Ion Trap LC-MS (electrospray) spectrometer. GC analyses were performed using HP 5890 series II chromatograph equipped with an auto injector HP 7672A (HewlettPackard, Palo Alto, CA) and an FID detector. A Rtx-Biodiesel TG (fused silica) or an MXT Biodiesel TG (Siltek-treated stainless steel) capillary column (Restek, Bellefonte, PA) was used for the GC. The following method was used for all the GC analyses for this study: oven temperature, 60 °C → 370 °C (10 °C/min; 6 min hold at 370 °C; total time 37 min); injector temp, 250 °C; detector temp, 370 °C; injection volume, 1 μL; split mode, 40:3. Dry N2 was used as a carrier gas with a column flow rate of 15 mL/min. Calibration curves were prepared for C14:0, C15:0, C16:0, and C22:6 fatty acid methyl esters and TMS derivatives of fatty acid ethanolamides for quantification (r2 > 0.99). Chemicals. Most reagents were purchased from commercial suppliers and used without further purification. Thin-layer chromatography (TLC) was carried out on glass backed silica plates, purchased from Sorbent Technology. The plates were visualized under ultraviolet (254 nm) light or by staining with either potassium permanganate or phosphomolybdic acid reagents and gentle heating. Silica gel column chromatography was carried out using 20−60 μm dry silica purchased from Sorbent Technology. Microalgae Biomass. The microalgae strain Schizochitrium limacinum SR21 was used in this study. The microalgal cultures were heterotrophically grown, harvested, and freeze-dried as previously described.21 This microalgal strain is reported to have a high growth rate and lipid content22,23 and is thus wellsuited as a model substrate for our study. Analysis of Lipid Content and Composition. The method is based on the in situ transesterification procedure for quantification of FAME content of microalgal biomass followed by Morita and others.22 A 20−30 mg sample of freezedried biomass was heated with 0.5 mL of acidified methanol B
DOI: 10.1021/ie503980g Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX
Article
Industrial & Engineering Chemistry Research
Figure 1. Isolation strategies of FAAA from algal biomass. Reactants and products are shown in red blocks.
Scale-Up of Alkanolamides Synthesis and Quality Upgrade. For upgrading the quality of FAAA, larger-scale reactions were first performed. A 1.00 g sample of dried biomass and Amberlyst 15 (40 w/w%) were suspended in methanol (5.0 mL/g) in a 8 mL screw cap vial and heated at 90 °C in an oil bath with continuous stirring at 300 rpm. After completion of transesterification, the vial was cooled to room temperature and centrifuged at 4000 rpm for 10 min. The upper layer was separated using a gastight syringe. The crude lower phase was treated with hexane (10.0 mL/g of initial biomass weight) to extract the remaining products and combined with the upper phase. Hexane was removed under reduced pressure, and the crude liquid product was mixed with ethanolamine (1.0 mL/g of initial biomass weight). This mixture was heated for 3 h at 120 °C and cooled to room temperature. The reaction mixture was distilled using a Kügelrohr distillation apparatus under 30 mmHg vacuum at 70 °C. The resulting crude FAAA was dissolved in a minimal amount of methanol and treated with 10 wt % activated charcoal and heated to 50 °C with stirring for 10 min. The mixture was filtered through a cellite pad under vacuum, and the filtrate was concentrated to afford a light brown solid. Derivatization Method of FAAA for GC Analysis. This method was developed based on the procedure by Thabuis2 and Wang.24 A mixture of bis(trimethylsilyl)amine (0.75 mL), pyridine(0.75 mL), and trifluoroacetic acid (20 μL) and a known mass of FAAA (40−80 mg) was heated at 70 °C for 30 min with a heating block with constant stirring at 500 rpm. The mixture was cooled to room temperature and diluted to 2 mL with hexane for GC analyses.
liquid phases were obtained. The upper FAME layer was recovered using a gastight syringe. Thereafter, 1 mL of hexane was added into the vial to extract the remaining products and then recovered and combined with the upper layer. Hexane was evaporated from the pooled liquid samples; the residual liquid was placed back in a 1.5 mL crimp-top GC vial, and ethanolamine (0.1 mL) and NaOMe (5% w/w) were charged. This mixture was incubated at reaction temperature (90 or 120 °C) for 1−15 h. After desired incubation periods, the vials were cooled to room temperature and a portion of the crude mixture (40−80 mg) was recovered and silylated (see the sub-section: Derivatization Method of FAAA for GC Analysis) by the following procedure and subjected to GC analysis for quantification against the reference standards. Synthetic Procedure of Alkanolamide Reference Standards. The substrate (free fatty acid) was placed in a three-necked flask and dissolved in methanol (12 mL/g). Acetyl chloride (1.2 equiv.) catalyst was added and the mixture was heated to reflux under nitrogen atmosphere for several hours. After completion of methyl ester formation, a part of the solvent (methanol) was removed from the reaction mixture under reduced pressure, and the reaction mixture was contacted with the immiscible extraction solvent diethyl ether to extract FAMEs, and the diethyether layer was washed three times with 5% NaHCO3 and dried over anhydrous MgSO4. Dried diethyl ether layer was filtered and crude methyl ester was recovered by evaporating diethyl ether using a rotary evaporator. The methyl ester was carried to the next step without further purification. The crude material was treated with ethanolamine (∼6 equiv) for several hours at ambient temperature and was recrystallized from methanol (except for C22 FAAA synthesis where purification was achieved by column chromatography using 3:7 acetone:hexane mixture as the eluent) to obtain the pure alkanolamides. The identity and purity of the isolated products were confirmed by 1H NMR, 13C NMR, mass spectrometry, and by GC (NMR spectra are shown in Supporting Information).
■
RESULTS AND DISCUSSION In Situ Amidation of Algal Biomass. The previously developed reactive extraction of FAME from biomass was accomplished by treatment of algal biomass with acidified methanol (containing 5% H2SO4) at 90 °C for 90 min. FAME produced at the end of this reaction forms a separate light C
DOI: 10.1021/ie503980g Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX
Article
Industrial & Engineering Chemistry Research phase distinct from the heavier methanol.20 Our understanding of the mechanism suggests that methanol first diffuses into microalgae cells and transesterifies FAG (and other lipids), generating FAME. Once generated, FAMEs form a separate phase after exceeding their solubility limit in methanol. Analogously, initial efforts on isolation of FAAA from the biomass were made to directly amidate FAG by treatment of the biomass with alkanolamine with or without sodium methoxide catalyst (Figure 1(A)). In general, this procedure suffered from emulsion formation that resulted in no visible phase separation at 90 °C (Figure 2a) and solidified upon
amine used in the in situ amidation method, to also reduce usage of ethanolamine (a more expensive chemical relative to methanol) (Figure 1(B)). Under these conditions also, nearly quantitative yield of FAAA was achieved, but the reaction formed a turbid mixture (Figure 2b) at reaction temperature without discernible phase separation and solidified at room temperature. Although FAAA can be recovered with multiple extractions with ethyl acetate using both of the in situ amidation procedures described above, solvent recovery could be energy intensive. Therefore, we sought methods that would eliminate the use of the solvent, while maintaining near quantitative conversion of biomass lipids to FAAA. Two-Step Conversion of Cellular Lipids to FAAA. Because our previously established reactive extraction with methanol resulted in near-complete transesterification of cellular lipids and provided an easily separable FAME phase (Figure 2c), we further pursued a two-step approach. Cellular lipids were recovered as phase-separable FAME in the first step; subsequently, crude FAME was converted to FAAA in the second step (Figure 1(C)). As described below, both steps of this two-step procedure were further optimized giving nearly quantitative recovery of the desired FAAA (Table 1, reaction scheme C). Because catalyst recovery and reuse is more difficult with homogeneous catalysts (such as H2SO4), we first pursued in situ transesterification of algal biomass using solid-supported catalysts. Use of Amberlyst 15 for esterification of algal lipids has been studied, and it has been shown that the catalyst can be recycled and reused without sacrificing reaction yield.14a We first optimized the transesterification step using the reported protocol as a guideline. Reaction conversion was monitored over time for the reaction containing biomass and Amberlyst 15 (40 (w/w)% biomass) in methanol (5 mL/g biomass) at 90 °C (Figure 3). In contrast to previous studies with Amberlyst 15,
Figure 2. (a) Image showing the reaction mixture after treatment of biomass with ethanolamine showing emulsion formation interfered with phase separation of FAAA from the reaction mixture; however, FAAA can be selectively extracted into a solvent immiscible with the reaction mixture. (b) Image showing the reaction mixture after treatment of biomass with mixture of ethanolamine and methanol showing partial phase separation of a mixture of FAMEs and FAAAs (upper phase) from methanol and ethanol amine (lower phase). Isolation of FAMEs and FAAAs still requires the use of an extraction solvent immiscible with the reaction mixture. (c) Image showing the reaction mixture after treatment of biomass with methanol showing clear separation of FAMEs (upper phase) and methanol (lower phase). The separated FAME layer is reacted in a second-step with ethanolamine to form FAAA.
cooling to ambient temperature after reaction completion. While a distinct phase formation of FAAAs was not observed, in situ conversion of FAGs to FAAAs did take place under these reaction conditions at high yield as was evident from the amount of FAAA we were able to selectively extract from the reaction mixture into the immiscible solvent ethyl acetate (Table 1, reaction scheme A). To assess if supplementation of methanol to the reaction mixture would prevent emulsification of amides and facilitate product phase separation, a reactant mixture comprising ethanolamine and methanol was used (Figures 1(B); Table 1, reaction scheme B). Our expectation was that methanol would first react with FAGs to form a separate FAME layer, and ethanolamine would then react with the FAMEs to produce FAAAs, with both steps taking place in the same pot. In these experiments, the total amount of mixed-solvent (methanol and ethanolamine) was kept the same as the amount of ethanol-
Figure 3. Kinetics of in situ transesterification of algal lipids with Amberlyst 15 catalyst. Reaction conditions: biomass (0.1 g), methanol (0.5 mL), and Amberlyst 15 (40% w/w-biomass). Reactions were carried out in a sealed GC vial at 90 °C.
Table 1. Isolation of FAAAs from Algal Biomass Lipids yieldf (as wt % of lipids in algae) rxn. scheme
solvent/reactant
catalyst
rxn. time (min)
rxn. temp (°C)
phase sep.
FAME
FAAA
A B
EAa MeOH/EA (4:1) MeOH/EA (1:1) MeOH;b EAc
none none none Amb. 15b,d NaOMec
90 90 90 360b 90c
120 90 120 90b 120c
none turbid turbid clean
− ∼47e trace amount −
∼95e ∼30e ∼ 94e >99
C a
Ethanolamine. bTransesterification step. cAmidation step after phase separation. The reaction required 3 h without NaOMe catalyst. dAmberlyst 15. after extraction into ethyl acetate. fPerformed with 100 mg of biomass containing 31−33 mg of lipid content.
e
D
DOI: 10.1021/ie503980g Ind. Eng. Chem. Res. XXXX, XXX, XXX−XXX
Article
Industrial & Engineering Chemistry Research
loadings lower than 5 % (w/w), the reaction conversion and recovery of FAAA were significantly low under otherwise identical conditions (Table 2, entry 2 versus 3 and 4). However, with extended reaction time (3 h), we found that the reaction could proceed even without a catalyst (entry 7). We determined that the optimal reaction conditions for this step are: mass ratio of biomass to ethanol amine 1:1; reaction temperature 120 °C; reaction time 1.5 h (with 5% (w/w) sodium methoxide catalyst) or 3 h (without catalyst). Both procedures (i.e., with and without catalyst) were implemented on gram scale biomass input (entries 7, 8). GC analysis of products showed that the major products at the end of the reaction were FAAA with small amounts of other impurities after distillation of residual methanol as well as ethanolamine (Figures S1 and S2, Supporting Information). GC-MS analysis of the crude FAAA suggested that some of the remaining impurities may be monosaccharide derivatives. Other possible impurities could be product isomers from acyl rearrangement and their derivatives.24 The typical color of the isolated alkanolamide was dark brown but was improved after charcoal treatment and filtration (Figure S3, Supporting Information).
we observed that our reactions, performed with high concentrations of biomass, produced FAME as a separate phase. To accurately quantify total FAME produced from the reaction, FAME was recovered from both phases after reaction. The upper phase was collected first, then FAME dissolved in the lower phase was extracted using hexanes. The upper phase and solvent extract were combined, and solvents were removed under vacuum. From Figure 3, it can be seen that the reaction completed after 6 h. A previous study with solid-supported catalyst14a suggested a two-step process for obtaining FAMEs from biomass: in situ pre-esterification with Amberlyst of “free fatty acids” in the lipid, followed by a base-catalyzed transesterification of the “triglyceride fraction” of the lipids. However, our experiments show that nearly all of the cellular lipids were obtained as FAME in a single treatment step. It is possible that transesterification efficiencies are strain-dependent. Although hexane extraction was employed in this study to perform quantitative mass balance, we anticipate that extraction will not be necessary in larger-scale operations. In a continuous process, the methanol phase saturated with FAME would be easy to recover after separating out solid catalyst and biomass residues. After supplementing with small amounts of fresh methanol (to compensate for consumption during FAME synthesis), the methanol phase can be recycled for subsequent reactions. Because of the high FAME concentration in the recycled methanol, the subsequent reactions will lose only little, if any, FAME to dissolution in the methanol phase. Thus, upon recycle of the methanol media, most, if not all, of the FAME produced in subsequent reactions is expected to stay in the upper phase.20 After recovery, the resulting isolated FAME was treated with ethanolamine (1 g/g biomass, ∼10 equiv of the corresponding FAME) in the presence of sodium methoxide catalyst. With use of 5 % (w/w) sodium methoxide at 120 °C, a minimum of 1.5 h was required for reaction completion (Table 2, entries 1 and 2). At 90 °C, the reaction did not go to completion even after 15 h (entry 6). Reducing the amount of ethanolamine by half also led to lower product recovery under the otherwise identical reaction conditions (entry 5). These observations are consistent with previous studies.25 We have also tested the possibility of reducing the amount of the catalyst. At catalyst
■
CONCLUSIONS Methods for FAAA synthesis and isolation from lipidcontaining algal biomass were developed. In situ amidation produced quantitative conversion of biomass lipids to FAAA but required extraction with ethyl acetate. Alternately, biomass lipids were first converted to FAME though in situ transesterification with Amberlyst 15 catalyst and recovered as a spontaneously separable phase. Subsequently, crude FAME was converted to FAAA. The new method circumvents solventbased recovery and purification of FAG or FAME prior to conversion to FAAA, resulting in fewer unit operations. In addition, our procedure uses recyclable solid acid catalyst and operates under mild conditions (95% of the total lipid in the biomass. This study represents the f irst report on production of FAAA from algal biomass.
■
Table 2. Optimization of Reaction Conversion from FAMEs to FAAAsa entry
cat. (w/w%)
temp. (°C)
time (h)
FAME (%)
FAAA (%)
1 2 3 4 5b 6 7c 8d
5 5 1 3 5 5 0 5
120 120 120 120 120 90 120 120
1.0 1.5 1.5 1.5 1.5 15 3.0 1.5
