- Email: [email protected]
Obtaining biodiesel from microalgae oil using ultrasound-assisted in-situ alkaline transesterification
Fuel 202 (2017) 512–519
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
Obtaining biodiesel from microalgae oil using ultrasound-assisted in-situ alkaline transesterification Natalia Martínez a, Nicolás Callejas a, Etiele G. Morais b, Jorge A. Vieira Costa b, Iván Jachmanián a, Ignacio Vieitez a,⇑ a b
Departamento de Ciencia y Tecnología de los Alimentos, Facultad de Química, Universidad de la República, General Flores 2124, Montevideo, Uruguay Laboratório de Ingenieria Bioquímica, Escuela de Química y Alimentos, Universidade Federal de Rio Grande, Rua Engenheiro Alfredo Huch 475, Rio Grande, Brazil
h i g h l i g h t s Efficiency of in-situ alkaline transesterification of dry biomass. Ultrasound-assisted process and addition of different co-solvents to the reaction medium. When process was carried out with ultrasound-assisted (80 W), ester content increased to 97.6%. Ultrasound increased yield of in-situ alkaline transesterification of Spirulina sp. biomass.
a r t i c l e
i n f o
Article history: Received 13 February 2017 Received in revised form 6 April 2017 Accepted 8 April 2017 Available online 26 April 2017 Keywords: Microalgae Spirulina sp. In-situ transesterification Biodiesel Ultrasound Co-solvents
a b s t r a c t Efficiency of in-situ alkaline transesterification of dry biomass from a crop of Spirulina sp., using ultrasound-assisted and/or the addition of different co-solvents to the reaction medium was studied. The starting biomass had a total lipid content of 7.0% (Folch method), comprised mainly by the following fatty acids: 16:0 (44.2%), 18:3 (n–6) (23.3%), and 18: 2 (n–6) (11.1%). When the transesterification reaction was carried out without ultrasound, it only extracted 12.8% of lipids (0.9% of the biomass) after 2 h of reaction, achieving only a 63.6% ester content. When the process was carried out under the same conditions except for the ultrasound-assisted (80 W), the weight yield remained mostly unchanged, but the ester content increased to 97.6%. Additionally, when chloroform was added to methanol as a co-solvent in a 2:1 ratio it was possible to extract 43% of lipids. However, the ester content was reduced to 69.7%. Moreover, with maximum ultrasound power (180 W) a high ester content (96.9%) and a relatively high extraction yield (26%) were obtained, even without the co-solvent. These results show that the ultrasound significant increased the yield of the in-situ alkaline transesterification of Spirulina sp. biomass, promoting a higher percent recovery of lipids as well as ester content from the starting material. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Increasing energy consumption and the adverse effects of the indiscriminate use of fossil fuels are growing concerns. This problem has been a driving force for developing an array of alternative renewable energy sources [1]. Unlike other sources of renewable energy, that require longterm assessment, biodiesel technology is already known and only requires minor adjustments to adapt to local raw materials and production capacities. For this reason, adoption of biodiesel has gradually increased and is regarded as a viable alternative for curb⇑ Corresponding author. E-mail address: [email protected] (I. Vieitez). http://dx.doi.org/10.1016/j.fuel.2017.04.040 0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.
ing fossil-fuel use and complying with international agreements to reduce emissions of greenhouse gases [1]. The cost of raw materials accounts for about 75% of total costs of biodiesel production [2]. Therefore, the choice of suitable raw materials is crucial to minimizing production costs. Transesterification is the key reaction in biodiesel production, and triglycerides are the fundamental reagent. Traditionally, the main source of triglyceride has been vegetable oil, such as soybean, rapeseed, palm, sunflower oils. However, whether these materials should be used for biodiesel production or not is under debate, since they are predominantly used as edible oils. Liquid biofuels can be categorized into different generations according to the type of raw material they are made from [1]. First generation biofuels come from vegetable oils (edible raw
N. Martínez et al. / Fuel 202 (2017) 512–519
materials), while the second generation ones come from nonedible raw materials, such as oils coming from Jatropha curcas L. [3–5], Croton megalocarpus [6], Cerbera odollam [7], castor-oil plant, macauba, mahua [8] and jojoba [9]. However, growing demand of these raw materials for the food industry, coupled with limited availability of land for these crops, render first- and second-generation biofuels unsustainable. Third generation biofuels, on the other hand, are derived from microand macroalgae, which prooves to be advantageous over the two preceding categories [10]. Microalgae are photosynthetic, microscopic organisms that are found in seawater and freshwater [11,12], and are able to convert 1.83 tons of atmospheric CO2 into 1 ton of biomass [13]. Microalgae are considered attractive for biofuel production since they have greater photosynthetic efficiency, increased biomass production and faster growth, compared to traditional crops [14]. Furthermore, growing microalgae can be exceedingly simple. For instance, wastewater and land that is unsuitable for conventional agriculture can be used for this purpose [15,16]. Lipid accumulation inside cells is quite appealing, being around 25–75% on the dry basis for these type of microalgae [17]. Reported yields [15] are at least 10 times higher in microalgae compared to palm trees, the highest-yielding oil crop. Interest in microalgae culture has grown considerably in recent years. Research in this area, particularly in efficient conversion of microalgae oils into biofuels has become a new frontier in the field of renewable energy. Microalgae culture is quite promising, since it can be done in ponds, open lakes or photobioreactors, has a high yield per square mile, and does not compete with traditional food crops [18]. Spirulina sp. is a filamentous, spiral-shaped, photosynthetic, and multicellular plankton. It belongs to the filum Cyanobacteriae, formerly known as blue-green microalgae due to the presence of green (chlorophyll) and blue (phytocyanins) pigments. This microflora thrives in alkaline saline waters, up to a pH of 11.0 and can exist in a variety of habitats, such as soils, freshwater, brackish and sea waters, hot springs, as well as in industrial and household wastewaters. It has a high amount of phytonutrients and pigments with applications in foodstuffs, therapies and diagnostics [19,20]. Although, Spirulina sp. is used mostly for food as supplement because it has small amount of lipids (which are mostly structural lipid) there are several references of it use as feedstock for biodiesel production [18,21]. For example, Nautiyal et al., 2014 [18] studied a single stage extraction–transesterification from Spirulina algae biomass with a total lipid content of 8.6%. In the optimum conditions was achieved a maximum biodiesel yield of 75%. However, there are not too many references to the ester content achieved. Key processes in biodiesel production from microalgae include culturing, harvesting, biomass processing, extraction of lipids, and subsequent transesterification. Lipid extraction from the microalgae is the most complex and expensive of these processes [22]. Extraction of the microalgae oil requires a mechanical, chemical, or biological, cell disruption followed by some oil collection method by means of solvents. The main difficulty in oil extraction comes from the cell walls of some microalgae species, which are strong and thick, considerably hindering the oil extraction process. The water content of the microalgae further reduce the overall yield [22]. Oils from microalgae are usually converted into biodiesel by transesterification using methanol and different types of catalysts (alkaline or acidic catalysis, depending on the oil characteristics). However, recent studies have shown that in situ transesterification (direct method) may be a more efficient process. In this case, oil extraction and conversion into biodiesel occurs simultaneously [22].
513
In situ transesterification happens in a single stage, including both free fatty acid esterification and triglyceride transesterification from the biomass-derived oil. This simplifies the production process and improves the biodiesel yield in compared to the conventional extraction, since the fewer stages the less oil is lost [18]. The in situ transesterification reaction can be carried out both in presence as well as in absence of co-solvents [23,24]. However, the use of a co-solvent may be crucial for a successful in situ transesterification. On the one hand, it acts as an extraction agent, and on the other hand, forms a homogeneous system with the microalgae oil, methanol and the catalyst. Usually, stirring and heating are required to promote the reaction of transesterification. However, co-solvent-assisted in situ transesterification should require less energy in the form of heating or stirring [25]. Releasing the lipids contained inside the cells of the microalgae could be difficult during in situ transesterification, therefore large volumes of solvent are generally required. This takes a lot of energy and substantially increases costs. Another factor to be considered is that oil must be released and extracted without contamination from other cellular components, such as DNA or chlorophyll [26]. Several approaches have been proposed for the selectively disrupting the cell wall, for instance, using ultrasound, microwaves, enzymes and pressurized fluids [27–29]. Although the mechanisms behind each of these techniques are different, most involve rupturing the cells to release the lipids within the cytoplasm. The principle behind the ultrasound method is the generation of sound waves that propagate through the fluid causing alternate cycles of high and low pressures. During the high pressure cycle, small bubbles, which were generated during the low pressure cycle, violently collapse and result in the phenomenon called cavitation [1]. High pressure and liquid velocities create shear forces upon the microalgae during cavitation, mechanically breaking the cellular structures and enhancing the transfer of extracted lipids. This methodology improves the lipid yield (output) between 50 to 500% and produces up to a 10-fold reduction in extraction times [1]. Martinez-Guerra et al., 2014 [30] studied the effect of ultrasound on the in situ transesterification of lipids from Chlorella sp. using ethanol as co-solvent for lipid extraction and as a reagent in the transesterification reaction. Optimal conditions were: a 1:6 to 1:9 ratio of microalgae:ethanol (wtv), 2% w/w of catalyst (sodium hydroxide) and a 6-min treatment time at 490 W of ultrasound. The weight yield was 18.5% and the conversion to ethyl esters was 95.0%. On the other hand, Keris-Sen et al., 2014 [31] tested different ultrasound intensities (0.1–0.5 W/mL), at a 30 kHz frequency for 5 to 60-min cycles. They also studied the effect of ultrasound on the lipid extraction efficiency in presence of co-solvents (hexane or a chloroform/methanol mixture). The results showed that in the sample with the mixture of co-solvents and no ultrasound, 13.6% of lipids could be extracted from the dry biomass. However, the biomass still contained a significant amount of lipids that had not been removed (43.4%). When the extraction process was performed by ultrasound, the use of the co-solvents mixture managed to extract 26.8% of lipids from the dry biomass. Consequently, ultrasound significantly increased the extraction of lipids [31]. Therefore, additional information is required to evaluate this type of biomass used for biodiesel conversion, which needs a process that is much more complex than that traditionally used to obtain this biofuel from vegetable oils or animal fats. In this study, there were shown results for in situ transesterification of dried biomass of Spirulina sp. Several variables related to the reaction system, as well as the components of the reaction mixture were studied. Addition of different co-solvents (chloroform and hexane), and the use of ultrasound in in situ transesterification method were evaluated. The ultrasound power and treatment
514
N. Martínez et al. / Fuel 202 (2017) 512–519
times in absence and presence of different co-solvents were also studied. Since there are not too many references to the ester content achieved with dried biomass, this study focused on the main differences with respect to lipid extraction, the weight yield relative to the starting biomass, as well as the content of fatty acid methyl esters (FAME), in order to define the most suitable conditions and co-solvents for biodiesel production. 2. Materials and methods 2.1. Materials Dry Spirulina sp. LEB 18 powder was provided by the School of Chemistry and Foods of the Universidade Federal do Rio Grande of Brazil. Biomass was obtained by culturing in Zarrouk medium [32] using Raceway ponds at room temperature and daylight as the light source. In all cases, the biomass was separated by filtration from the crop, then pelletized and dried in an oven for 6 h at 60 °C and then stored at room temperature. Before using it in the different methods, any residual moisture was eliminated by freeze-drying. Standards, reagents and solvents in this study were analytical grade provided by Sigma-Aldrich (USA). 2.2. Methods 2.2.1. Biomass preparation Strands of Spirulina sp. LEB 18 were ground with a domestic coffee grinder to obtain a fine powder. The powdered biomass was kept in a capped container at - 18 °C until use. 2.2.2. Determination of lipid content Lipid content was measured by gravimetry after extraction with a solvent mixture (chloroform: methanol 2:1 v/v, using the method by Folch, Less and Stanley, adapted from Christie and Han, 2010 [33]). The total lipid content was measured in triplicates. About 0.2 g of biomass were weighed, to which was added 5.0 mL of the solvent mixture, and stirred in a vortex. Then, the mixture was sonicated for about 15 min. The phases were then separated by centrifugation at 3500 rpm. This procedure was repeated three more times. Centrifugation supernatants were pooled together and filtered through filter paper. An aqueous solution of 0.88% KCl (equivalent to one fourth of the total supernatant volume) was added, mixed, and centrifuged. The organic phase separated and washed with one fourth of the mixture volume using a mixture of methanol:water (2:1). It was then mixed and centrifuged. The organic phase was separated and a small amount of anhydrous Na2SO4 was added. This was then centrifuged and separated. The organic phase was dried to constant weight. The acid value as well as free fatty acids were titrimetrically determined using AOCS method (Cd 3d-63) [34]. 2.2.3. Determination of the fatty acid composition and the percentage of lipids that can be converted into esters (purity of the Folch extract) About 30 mg of sample were placed inside a tube with a screw cap and 2.0 mL of 0.05 M MeONa in MeOH solution were added. The air inside the tube was flushed with a stream of nitrogen and the mixture was refluxed for 10 min (100 °C). After that time, it was allowed to cool down and 2.0 mL of a methanolic HCl solution (prepared by dissolving 8 mL of acetyl chloride (C2H3ClO) in 100 mL of methanol) were added and further refluxed for 10 min. Then, the mixture was allowed to cool down and the internal standard was added (heptadecanoic acid methyl ester). Then 5.0 mL of hexane and 7.0 mL of distilled water were added to extract the FAME formed in the reaction. This was stirred and centrifuged at
4000 rpm for 10 min and the organic phase was separated. Onto the organic phases, a small amount of anhydrous Na2SO4 was added. Then it was stirred, centrifuged and the solution of FAME was separated. A volume of 1.0 lL of the FaME solution was injected into a gas chromatograph (Shimadzu, model GC-14B with FID detector, Kyoto, Japan). A capillary column was used (Supelco SP2330, 30 m 0.25 mm 0.2 mm), using nitrogen as the carrier gas. Temperature program started at 60 °C for 2 min, then the temperature was raised to 200 °C at a rate of 10 °C/minute, and was finally raised at a rate of 5 °C/min to 240 °C and kept at that temperature for 7 min. The composition in fatty acid was measured in triplicate. Using this method, the amount lipids that can be converted into FAME can be calculated using the following equation:
mfat ¼ ½mstd :ðAtotal Astd Þ=Astd %convertible lipids ¼ mfat 100=mbiomass 2.2.4. In situ alkaline transesterification Fifty milliliters of methanol containing 0.5 g of potassium hydroxide (2% of catalyst) were added to 25 g of biomass in a 500 mL batch reactor. If a co-solvent was added to the reaction mixture, the volume was calculated to produce a 2:1 co-solvent: methanol ratio. Chloroform or hexane were tested as co-solvents. The temperature was set to 55 °C and the treatment time to 2 h. The reaction mixture remained under continuous stirring. When the time of reaction was up, the phase liquid was separated from the phase solid. The solid phase was washed several times with co-solvent, which pooled with the previously collected liquid phase. FAME were separated from the liquid phase using several washes with distilled water, in order to eliminate the catalyst and glycerol excess. Glycerol was formed as a reaction byproduct. A small amount of anhydrous sodium sulfate was added to the organic phase, and then the mixture was centrifuged and separated. The organic phase was dried to constant weight and the recovered weight yield was calculated (wt%). 2.2.5. Ultrasound-assisted in situ alkaline transesterification The same procedure used in 2.2.4 was performed in order to evaluate the efficiency of the ultrasound-assisted transesterification process. Ultrasound equipment (Elmasonic model P60 H, Singen, Germany) is capable of setting the ultrasound power to different values up to 180 W and features temperature control (30–80 °C). The temperature was set to 55 °C. Due to the temperature increase derived from ultrasound, a system of cooling (coil) was used to keep the temperature at the set value. We studied the following transesterification times: 15, 30, 60 and 120 min; and different ultrasound intensities, 44% (80 W), 60% (108 W), 80% (144 W) and 100% (180 W) at a 37 kHz frequency. As it is explained in Section 2.2.4, the biomass:methanol ratio was always 50:100 (weight/volume) and, if applicable, the volume of co-solvent (hexane or chloroform) was 200 parts. The reaction mixture was kept inside the equipment and no stirring system was used. Upon completion of the reaction, under all the different conditions, the liquid phase was separated from the solid phase. The solid phase was washed three times with hexane (50–75 mL), which was pooled with the collected liquid phase from the previous step. The FAME were then separated from the mixture and the excess of catalyst and glycerol were eliminated with several washes with water distilled. Once the pH of the wash water was equal to that of the distilled water, a small amount of anhydrous sodium sulfate was added to the organic phase to eliminate residual water, and was filtered through filter paper. The solvent was
N. Martínez et al. / Fuel 202 (2017) 512–519
evaporated to constant weight using a rotary evaporator and the ‘‘gross” weight yield was calculated (quantity of product with respect to the starting biomass). 2.2.6. Purification by solid phase extraction (SPE) The transesterification product obtained was purified using SPE. One gram of active silica was prepared and properly packed into a column using solvent. Ten milliliters of hexane or petroleum ether were added to the column and allowed to elute until 1 to 2 mL of solvent were sitting on top of the column. Between 200 and 300 mg of sample were dissolved in the smallest possible volume of solvent and sewed onto the column. Once sample entered the column, 5 mL of a petroleum ether:diethylether (95:5) mixture were added to the column and eluted, taking care not to let the column dry up. The eluate was collected in a clean and dry tube. This procedure was repeated 3–4 more times. The collected eluate (about 20 mL) was dried under a nitrogen stream to constant weight, in order to measure the ester content later on. 2.2.7. Determination of the ester content The ester content was determined according to the EN 14103 standard. In a 10 mL volumetric flask, 200 mg of the purified sample was dissolved using hexane. In a 4 mL vial, 200 lL of the previously prepared sample solution were mixed with 200 lL of a 4.0 mg/mL solution of internal standard (17:0). This mixture was analyzed in a Shimadzu-14B gas chromatograph using a SP-2330 capillary column. The temperature program was as follows: starting temperature of 160 °C, then heating at 4 °C/min up to 230 °C, and then maintenance at 230 °C for 10 min. The ester content was analyzed in triplicate. 3. Results and discussion 3.1. Lipid content and fatty acid composition The lipid content was found to be 7.0 ± 0.4% on dry basis, using the method by Folch, Less, and Stanley. Oil contained a very low percentage of free fatty acids (0.50 mg KOH/g oil). Table 1 shows the fatty acid composition of the oil extracted from Spirulina sp. LEB 18. The main fatty acids were 16:0 (palmitic acid), followed by 18:3 (n–6) (linolenic acid) and 18:2 (n–6) (linoleic acid). Additionally, the purity in the Folch extract was 30.2% (according to Section 2.2.3). The amount of lipids that can be converted into FAME was calculated multiplying the total lipid content (Folch method) by the purity of the extract. Accordingly only 2.1% of the lipids (7.0%) were convertible into FAME. 3.2. Effect of time and co-solvent type on the efficiency of the ultrasound-assisted in situ alkaline transesterification Treatment time plays a crucial role in the in situ transesterification reaction, not only because it needs to be enough to release
Table 1 Fatty acid composition of Spirulina sp LEB 18. Fatty acid
Percentage (wt%)
14:0 16:0 16:1 n7 18:0 18:1 trans 18:1 n9 18:1 n7 18:2 n6 18:3 n6 Total identified FAs
3.9 44.2 3.7 2.1 1.6 5.4 1.2 11.1 23.3 96.5
515
lipids into the reaction medium, but also to allow transesterification to occur. The main reported advantage of the ultrasound method is that it requires shorter treatment times [31,35]. When the transesterification reaction was carried out under the conditions described in Section 2.2.4 (without ultrasound), the weight yield was 12.8% (0.9% with respect to biomass) after 2 h of reaction, and the ester content was just 63.6%. Weight yields (with respect to dry biomass) and ester contents are shown in Fig. 1. These results correspond to ultrasoundassisted in situ alkaline transesterification at 80 W of ultrasound power, no co-solvent, and different treatment times (1, 2, 4, and 8 h). These results show that, without using co-solvents, the yield tends to decrease after 2 h of reaction. However, no significant differences in yields were observed when comparing 1 h and 8 h. These results suggest that the maximum yield was reached at 2 h of reaction. Addition of a co-solvent (chloroform or hexane) to the reaction medium increased the yield with compared to absence of cosolvent (0.6% and 0.9% with respect to biomass). Hexane yielded 1.0% in weight, with respect to biomass. Chloroform, on the other hand, considerably improved the weight yield, allowing to extract 43% of lipids (3.0% with respect to biomass). Therefore, the effect of the chloroform and methanol mixture extracted lipids more efficiently than hexane and methanol. The results showed that combining the cell ‘‘disruption” effect of ultrasound with co-solvents increases lipid extraction, compared working without co-solvents. Fig. 1 also shows the ester content vs treatment times, under the different conditions at various treatment times (1, 2, 4 and 8 h), with or without using co-solvents (chloroform or hexane). The results show that without co-solvents, ester content above 90% were obtained in all those cases. In particular, the ester content was particularly high at 1 and 2 h of reaction (96.8% and 97.6% respectively), exceeding the requirements of international regulations (96.5%). This shows that ultrasound was clearly favorable, since the ester content was only 63.6% in its absence. Therefore, regulatory requirements were met with short treatment times (under two hours). The use of co-solvents, as previously mentioned, enhances the extraction of lipids from the biomass by improving mass transfer. The type of co-solvent affects which lipids are extracted. Some of these lipids, such as pigments or polar lipids, are not subsequently converted to FAME during the transesterification reaction [30]. Therefore, the choice of co-solvent used in the process may also affect the final ester content. The treatment time chosen for testing the co-solvent (hexane or chloroform) was 2 h, because it had produced the greater ester content in absence of co-solvent. Fig. 1 shows that the ester content was lower when co-solvents were used, compared to similar conditions without them. None of those cases complied with regulations. The ester content using Hexane was 92.8%, and using chloroform was 69.7%. These results can be explained because cosolvents extract different kinds of lipids, which reduce the transesterification efficiency, therefore lowering the ester content. Hexane enables the extraction of non-polar lipids (triglycerides), which are the main substrate of the reaction. Since chloroform is miscible with methanol in certain proportions, together they may extract both polar and non-polar lipids, leading to lower conversion of lipids into esters. This was shown in the ester content, which did not exceed 70% when chloroform was used as co-solvent. To assess the effect of co-solvents in the overall yield (efficiency of the process), we studied calculated the product between the weight yield and the ester content. The result is the amount of FAME per gram of biomass. This parameter reveals the effect of extraction and the efficiency of the reaction as a whole. The results are shown in Fig. 2.
516
N. Martínez et al. / Fuel 202 (2017) 512–519
Fig. 1. Weight yield and ester content under different conditions of ultrasound-assisted in situ alkaline transesterification (80 W). (Versus treatment time and co-solvent, if applicable).
Fig. 2. Amount of FAME per gram of biomass vs treatment time and type of co-solvent.
In absence of a co-solvent, the highest amount of FAME per gram of biomass was obtained at two hours of reaction. For higher treatment times, this value dropped. This happens because the ester content decreases to a greater extent than weight yield. Chloroform produced the highest quantity of FAME per gram of biomass because its weight yield was high. Apparently, two opposing effects come into play when using co-solvents. On the one hand, the co-solvent can improve the weight yield by enhancing lipid extraction. On the other hand, however, these extracted lipids can be very diverse and not be able to be converted into esters. Therefore, the choice of co-solvent represents a compromise between weight yield and ester content. 3.3. Effect of ultrasound power, treatment time and type of co-solvent on the weight yield and ester content Until this point, the results corresponded to 80 W of ultrasound power. In order to observe the effect of ultrasound power on
weight yield and ester content, we tested the following conditions: 108 W, 144 W and 180 W (maximum power of the equipment), using a treatment time of only 30 min. The greater ultrasound intensities were expected to contribute to the disruption of the microalgae’s cell wall, facilitating the release of lipids into the reaction medium. However, previous reports have suggested otherwise. This could be attributed to excessive cavitations forcing bubbles to join each other instead of interacting with the cells, thus preventing their disruption [31]. The effects of ultrasound power on weight yield and ester content are shown in Fig. 3. These results indicate that increasing ultrasound power increases the weight yield. However, this increase was not significant; weight yield increased from 1.1% (15.3% of lipid material) to 1.6% (22.8% of lipid material) when ultrasound power was increased from 80 W to 180 W. It is worth mentioning that those yields were achieved in short treatment times (30 min), a favorable aspect that is consistent with previous literature [29,30,1].
N. Martínez et al. / Fuel 202 (2017) 512–519
517
Fig. 3. Weight yield and ester content under different conditions of in situ alkaline transesterification and 30 min of treatment time versus ultrasound power in absence of cosolvent or using hexane as a co-solvent.
The results in Fig. 3 show that the ester content increases with higher ultrasound intensities. For instance, the ester content after 30 min of reaction at 80 W was 82.6%, which increased to 91.6% and 95.2% using 144 W and 180 W ultrasound power, respectively. Therefore, results show that the increasing the ultrasound power in the in situ alkaline transesterification process (only 30 min of treatment time in absence of a co-solvent) improves the process since a larger amount of lipids are extracted, improving both the weight yield and the ester content. This can be clearly seen in Fig. 4, which shows the process efficiency in terms of the amount of FAME per gram of biomass (calculated as the product of weight yield and ester content). Since co-solvents enhance the extraction of lipids, we also studied the effect on weight yield and the content ester, as shown in Fig. 3. The reaction was carried out using 180 W, under the same conditions as those mentioned above, except for the addition of a co-solvent (2:1 (v/v) with respect to methanol). In this case, hexane was the chosen co-solvent, since our previous results showed it did not decrease the ester content compared to chloroform. The
extraction yield after 30 min of reaction, under the same operating conditions was 31% (2.2% with respect to biomass), which was higher than that obtained with 80 W and the same co-solvent. Therefore, lipid extraction was enhanced by adding a co-solvent and increasing ultrasound power. The increase of ultrasound power (over 80 W) has a greater effect on weight yield at short treatment times, since longer treatment times (1 h or more) showed lower values, as shown in Fig. 1. As already mentioned, the effect of the co-solvent in the ester content primarily depends on the types of lipids that are extracted, as well as the transesterification reaction. The ester content was 85.9% (30 min of reaction), which is lower than in absence of co-solvent. Therefore, treatment times have two different effects. Shorter treatment times (30 min) improve lipid extraction efficiency, thus increasing weight yield, where both ultrasound power and presence of co-solvent make a positive contribution. Nevertheless, the ester content falls below the required value of 96.5%, suggesting that the transesterification reaction requires of more time to reach completion.
Fig. 4. Amount of FAME per gram of biomass versus presence or absence of hexane as co-solvent.
518
N. Martínez et al. / Fuel 202 (2017) 512–519
Based on previous results, the optimum ultrasound power is 180 W (the highest power of the equipment) since it produced the highest ester content and weight yield without any cosolvent. However, these results were carried out using a fixed treatment time of 30 min. Thus, we were interested in finding how treatment times affects extraction under this optimum condition. We studied three additional treatment times, 15, 60, and 120 min, in order to observe how the process evolves. The results of the weight yield and ester content versus treatment time are shown in Fig. 5. Yields obtained in these times did not vary significantly. Nevertheless, the shortest treatment time (15 min) showed the best weight yield (2.0% of biomass) which represents a 28% extraction of lipids). According to the expected behavior, increasing the treatment time increased the ester content. The highest ester content achieved was 96.9% after 120 min of reaction at 180 W (the weight yield was 1.8% of biomass). However, 30, 60 and 120 min of reaction did not show significant differences in terms of ester content. This is also consistent with other reports indicating that short treatment times are required to achieve suitable conversions [26]. In this study, 180 W of ultrasound power and 30 min of treat-
ment time were enough to yield a 95.2% ester content, under the studied conditions. The value of weight yield for 80 W was 0.9% (2 h of reaction) and 1.8% for 180 W (2 h of reaction), whereas the ester contents were 97.6% and 96.9%, respectively. Thus, higher ultrasound power increased the weight yield but not the ester content. To evaluate the combined effect of the lipid extraction and the efficiency of the transesterification reaction, we once again calculated the process efficiency parameter. This equates to the amount of FAME per gram of biomass (Fig. 6). According to Fig. 6 and as mentioned, treatment time was shown to have two different effects: shorter times extract more lipids improving weight yields, while longer times increase the ester content. While there were large differences in the amount of FAME for the different times, the greater amount of FAME per gram of biomass was reached for 15 to 120 min. Consequently, choosing the treatment time is a compromise between maximizing weight yields or ester content. In our case, after 15 min the ester content was hardly over 85%, whereas, after 120 min, it reached 96.5%.
Fig. 5. Weight yield and ester content versus treatment times at maximum ultrasound power (180 W) in the in situ alkaline transesterification reaction.
Fig. 6. Amount of FAME per gram of biomass versus treatment time at maximum power (180 W) in ultrasound-assisted in situ alkaline transesterification.
N. Martínez et al. / Fuel 202 (2017) 512–519
4. Conclusions Results show that ultrasound is a suitable tool to improve the extraction of lipids from biomass because it produces enough cellular disruption. Additionally, it increases the reaction extent of the transesterification reaction, therefore enabling higher ester contents under suitable reaction conditions. In situ alkaline transesterification reaction without a co-solvent yielded an ester content above the value required by regulations (96.5%). The use of co-solvents seems to produce two opposing effects. On the one hand, it can improve the weight yield by enhancing extraction efficiency. However, it also modifies the quality of extracted lipids, increasing the proportion of ‘‘nonconvertible” lipids, and consequently decreasing the ester content. Therefore, priority must be given to one of these two parameters. The maximum power (180 W) was the best condition when treatment times were short (30 min). Under these conditions, the weight yield and ester content were 1.6% and 95.2%, respectively. At the maximum power (180 W) the ester content increased with longer times. The ester content was 96.9% after 2 h of reaction at 180 W. Weight yields did not vary significantly over time, but the maximum value corresponded to the shorter times (15 min at 180 W). The optimal power was 180 W (the highest possible with our equipment) which produced a high ester content (96.9%) and a high weight yield (1.8% with respect to biomass, corresponding to a 25.2% extraction of lipids) without needing a co-solvent. Therefore, if the limiting factor it was the capacity the equipment, major ultrasound powers could be studied to see a possible improve in the efficiency process. Acknowledgment This work was funded by ANCAP and CSIC (Uruguay) with financial support and PEDECIBA-Química programs (Uruguay). References [1] Mubarak M, Shaija A, Suchithra TV. A review on the extraction of lipid from microalgae for biodiesel production. Algal Res 2015;7:117–23. [2] Ahmad AL, Mat Yasin NH, Derek CJC, Lim JK. Microalgae as a sustainable energy source for biodiesel production: A review. Renew Sustainable Energy Rev 2011;15:584–93. [3] Guo F, Fang Z, Tian XF, Long YD, Jiang LQ. One-step production of biodiesel from Jatropha oil with high-acid value in ionic liquids. Bioresour Technol 2011;102:6469–72. [4] Yee KF, Wu JCS, Lee KT. A green catalyst for biodiesel production from jatropha oil: optimization study. Biomass Bioenergy 2011;35:1739–46. [5] Yang CY, Fang Z, Li B, Long Y. Review and prospects of Jatropha biodiesel industry in China. Renew Sustainable Energy Rev 2012;16:2178–90. [6] Kafuku G, Lam MK, Kansedo J, Lee KT, Mbarawa M. Croton megalocarpus oil: a feasible non-edible oil source for biodiesel production. Bioresour Technol 2010;101:7011–5. [7] Kansedo J, Lee KT, Bhatia S. Cerbera odollam (sea mango) oil as a promising non-edible feedstock for biodiesel production. Fuel 2009;88:1148–50. [8] Ghadge SV, Raheman H. Process optimization for biodiesel production from mahua (Madhuca indica) oil using response surface methodology. Bioresour Technol 2006;97:379–84.
519
[9] Canoira L, Alcantara R, Garcia-Martinez J, Carrasco J. Biodiesel from Jojoba oil wax: transesterification with methanol and propeties as a fuel. Biomass Bioenergy 2006;30:76–81. [10] Daroch M, Geng S, Wang G. Recent advances in liquid biofuel production from algal feedstocks. Appl Energy 2013;102:1371–81. [11] Demirbas A. Use of algae as biofuel sources. Energy Convers Manage 2010;51:2738–49. [12] Brennan L, Owende P. Biofuels from microalgae—a review of technologies for production, processing, and extractions of biofuels and co-products. Renew Sustainable Energy Rev 2010;14:557–77. [13] Chisti Y. Biodiesel from microalgae beats bioethanol. Trends Biotechnol 2010;26:126–31. [14] Miao X, Wu Q. Biodiesel production from heterotrophic microalgal oil. Bioresour Technol 2006;97:841–6. [15] Mata TM, Martins AA, Caetano NS. Microalgae for biodiesel production and other applications: a review. Renew Sustainable Energy Rev 2010;14:217–32. [16] Khan SA, Rashmi M, Hussain MZ, Prasad S, Banerjee UC. Prospects of biodiesel production from microalgae in India. Renew Sustainable Energy Rev 2009;13:2361–72. [17] Malcata FX. Microalgae and biofuels: a promising partnership? Trends Biotechnol 2011;29:542–9. [18] Nautiyal P, Subramanian KA, Dastidar MG. Kinetic and thermodynamic studies on biodiesel production from Spirulina platensis algae biomass using single stage extraction–transesterification process. Fuel 2014;135:228–34. [19] Becker EW. Microalgae. Cambridge, New York: Cambridge Univ. Press; 1994. [20] Vonshak A, Tomaselli L. Arthrospira (Spirulina) systematic and ecophysiology. In: Whitton BA, Potts M, editors. The Ecology of Cyanobacteria in the Netherlands. Kluwer Academic Publishers; 2000. [21] El-Shimi HI, Nahed K, Attia ST, El-Sheltawy GI, El-Diwani. Biodiesel production from Spirulina-Platensis microalgae by in-situ transesterification process. J Sustainable Bioenergy Syst 2013;3:224–33. [22] Hidalgo P, Toro C, Ciudad G, Navia R. Advances in direct transesterification of microalgal biomass for biodiesel production. Rev Environ Sci Biotechnol 2013;12:179–99. [23] Baumgartner TRS, Burak JAM, Baumgartner D, Zanin GM, Arroyo PA. Biomass production and ester synthesis by in situ transesterification/esterification using the microalga Spirulina platensis. Int J Chem Eng 2013:1–7. [24] Hidalgo P, Toro C, Ciudad G, Schober S, Mittelbach M, Navia R. Evaluation of different operational strategies for biodiesel production by direct transesterification of Microalgal Biomasa. Energy Fuels 2014;28:3814–20. [25] Xu R, Mi Y. Simplifying the process of microalgal biodiesel production through in situ transesterification technology. J Am Oil Chem Soc 2011;88:91–9. [26] Scott SA, Davey MP, Dennis JS, Horst I, Howe CJ, Lea-Smith DJ, et al. Biodiesel from algae: challenges and prospects. Curr Opin Biotech 2010;21:277–86. [27] Lee JY, Yoo C, Jun SY, Ahn CY, Oh HM. Comparison of several methods for effective lipid extraction from microalgae. Bioresour Technol 2010;101:75–7. [28] Ranjan A, Patil C, Moholkar SV. Mechanistic assessment of microalgal lipid extraction. Ind Eng Chem Res 2010;49:2979–85. [29] Araujo GS, Matos LJBL, Fernandes JO, Cartaxo SJM, Gonçalves LRB, Fernandes FAR, et al. Extraction of lipids from microalgae by ultrasound application: Prospection of the optimal extraction method. Ultrason Sonochem 2013;20:95–8. [30] Martinez-Guerra E, Gude VG, Mondala A, Holmes W, Hernandez R. Microwave and ultrasound enhanced extractive-transesterification of algal lipids. Appl Energy 2014;129:354–63. [31] Keris-Sen UD, Sen U, Soydemir G, Gurol MD. An investigation of ultrasound effect on microalgal cell integrity and lipid extraction efficiency. Bioresour Technol 2014;152:407–13. [32] Zarrouk C. Contribution à l’étude d’une cyanophycée. Influence de divers facteurs physiques et chimiques sur la croissance et photosynthese de Spirulina maxima Geitler Ph.D. Thesis. University of Paris; 1966. [33] W.W. Christie, X. Han, Lipid analysis: Isolation, separation, identification and lipidomic analysis (4th Ed.), Oily Press Lipid Library Series No. 24, 2010. [34] AOCS methods for biodiesel feedstock quality. In: Official Methods and Recommended Practices of the AOCS, (6th Ed.), Second printing, 2012. [35] Gude VG, Patil PD, Martinez-Guerra E, Deng S, Khandan NN. Microwave energy potential for biodiesel production. Sustainable Chem Process 2013;1(5):1–31.
Contents lists available at ScienceDirect
Fuel journal homepage: www.elsevier.com/locate/fuel
Full Length Article
Obtaining biodiesel from microalgae oil using ultrasound-assisted in-situ alkaline transesterification Natalia Martínez a, Nicolás Callejas a, Etiele G. Morais b, Jorge A. Vieira Costa b, Iván Jachmanián a, Ignacio Vieitez a,⇑ a b
Departamento de Ciencia y Tecnología de los Alimentos, Facultad de Química, Universidad de la República, General Flores 2124, Montevideo, Uruguay Laboratório de Ingenieria Bioquímica, Escuela de Química y Alimentos, Universidade Federal de Rio Grande, Rua Engenheiro Alfredo Huch 475, Rio Grande, Brazil
h i g h l i g h t s Efficiency of in-situ alkaline transesterification of dry biomass. Ultrasound-assisted process and addition of different co-solvents to the reaction medium. When process was carried out with ultrasound-assisted (80 W), ester content increased to 97.6%. Ultrasound increased yield of in-situ alkaline transesterification of Spirulina sp. biomass.
a r t i c l e
i n f o
Article history: Received 13 February 2017 Received in revised form 6 April 2017 Accepted 8 April 2017 Available online 26 April 2017 Keywords: Microalgae Spirulina sp. In-situ transesterification Biodiesel Ultrasound Co-solvents
a b s t r a c t Efficiency of in-situ alkaline transesterification of dry biomass from a crop of Spirulina sp., using ultrasound-assisted and/or the addition of different co-solvents to the reaction medium was studied. The starting biomass had a total lipid content of 7.0% (Folch method), comprised mainly by the following fatty acids: 16:0 (44.2%), 18:3 (n–6) (23.3%), and 18: 2 (n–6) (11.1%). When the transesterification reaction was carried out without ultrasound, it only extracted 12.8% of lipids (0.9% of the biomass) after 2 h of reaction, achieving only a 63.6% ester content. When the process was carried out under the same conditions except for the ultrasound-assisted (80 W), the weight yield remained mostly unchanged, but the ester content increased to 97.6%. Additionally, when chloroform was added to methanol as a co-solvent in a 2:1 ratio it was possible to extract 43% of lipids. However, the ester content was reduced to 69.7%. Moreover, with maximum ultrasound power (180 W) a high ester content (96.9%) and a relatively high extraction yield (26%) were obtained, even without the co-solvent. These results show that the ultrasound significant increased the yield of the in-situ alkaline transesterification of Spirulina sp. biomass, promoting a higher percent recovery of lipids as well as ester content from the starting material. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Increasing energy consumption and the adverse effects of the indiscriminate use of fossil fuels are growing concerns. This problem has been a driving force for developing an array of alternative renewable energy sources [1]. Unlike other sources of renewable energy, that require longterm assessment, biodiesel technology is already known and only requires minor adjustments to adapt to local raw materials and production capacities. For this reason, adoption of biodiesel has gradually increased and is regarded as a viable alternative for curb⇑ Corresponding author. E-mail address: [email protected] (I. Vieitez). http://dx.doi.org/10.1016/j.fuel.2017.04.040 0016-2361/Ó 2017 Elsevier Ltd. All rights reserved.
ing fossil-fuel use and complying with international agreements to reduce emissions of greenhouse gases [1]. The cost of raw materials accounts for about 75% of total costs of biodiesel production [2]. Therefore, the choice of suitable raw materials is crucial to minimizing production costs. Transesterification is the key reaction in biodiesel production, and triglycerides are the fundamental reagent. Traditionally, the main source of triglyceride has been vegetable oil, such as soybean, rapeseed, palm, sunflower oils. However, whether these materials should be used for biodiesel production or not is under debate, since they are predominantly used as edible oils. Liquid biofuels can be categorized into different generations according to the type of raw material they are made from [1]. First generation biofuels come from vegetable oils (edible raw
N. Martínez et al. / Fuel 202 (2017) 512–519
materials), while the second generation ones come from nonedible raw materials, such as oils coming from Jatropha curcas L. [3–5], Croton megalocarpus [6], Cerbera odollam [7], castor-oil plant, macauba, mahua [8] and jojoba [9]. However, growing demand of these raw materials for the food industry, coupled with limited availability of land for these crops, render first- and second-generation biofuels unsustainable. Third generation biofuels, on the other hand, are derived from microand macroalgae, which prooves to be advantageous over the two preceding categories [10]. Microalgae are photosynthetic, microscopic organisms that are found in seawater and freshwater [11,12], and are able to convert 1.83 tons of atmospheric CO2 into 1 ton of biomass [13]. Microalgae are considered attractive for biofuel production since they have greater photosynthetic efficiency, increased biomass production and faster growth, compared to traditional crops [14]. Furthermore, growing microalgae can be exceedingly simple. For instance, wastewater and land that is unsuitable for conventional agriculture can be used for this purpose [15,16]. Lipid accumulation inside cells is quite appealing, being around 25–75% on the dry basis for these type of microalgae [17]. Reported yields [15] are at least 10 times higher in microalgae compared to palm trees, the highest-yielding oil crop. Interest in microalgae culture has grown considerably in recent years. Research in this area, particularly in efficient conversion of microalgae oils into biofuels has become a new frontier in the field of renewable energy. Microalgae culture is quite promising, since it can be done in ponds, open lakes or photobioreactors, has a high yield per square mile, and does not compete with traditional food crops [18]. Spirulina sp. is a filamentous, spiral-shaped, photosynthetic, and multicellular plankton. It belongs to the filum Cyanobacteriae, formerly known as blue-green microalgae due to the presence of green (chlorophyll) and blue (phytocyanins) pigments. This microflora thrives in alkaline saline waters, up to a pH of 11.0 and can exist in a variety of habitats, such as soils, freshwater, brackish and sea waters, hot springs, as well as in industrial and household wastewaters. It has a high amount of phytonutrients and pigments with applications in foodstuffs, therapies and diagnostics [19,20]. Although, Spirulina sp. is used mostly for food as supplement because it has small amount of lipids (which are mostly structural lipid) there are several references of it use as feedstock for biodiesel production [18,21]. For example, Nautiyal et al., 2014 [18] studied a single stage extraction–transesterification from Spirulina algae biomass with a total lipid content of 8.6%. In the optimum conditions was achieved a maximum biodiesel yield of 75%. However, there are not too many references to the ester content achieved. Key processes in biodiesel production from microalgae include culturing, harvesting, biomass processing, extraction of lipids, and subsequent transesterification. Lipid extraction from the microalgae is the most complex and expensive of these processes [22]. Extraction of the microalgae oil requires a mechanical, chemical, or biological, cell disruption followed by some oil collection method by means of solvents. The main difficulty in oil extraction comes from the cell walls of some microalgae species, which are strong and thick, considerably hindering the oil extraction process. The water content of the microalgae further reduce the overall yield [22]. Oils from microalgae are usually converted into biodiesel by transesterification using methanol and different types of catalysts (alkaline or acidic catalysis, depending on the oil characteristics). However, recent studies have shown that in situ transesterification (direct method) may be a more efficient process. In this case, oil extraction and conversion into biodiesel occurs simultaneously [22].
513
In situ transesterification happens in a single stage, including both free fatty acid esterification and triglyceride transesterification from the biomass-derived oil. This simplifies the production process and improves the biodiesel yield in compared to the conventional extraction, since the fewer stages the less oil is lost [18]. The in situ transesterification reaction can be carried out both in presence as well as in absence of co-solvents [23,24]. However, the use of a co-solvent may be crucial for a successful in situ transesterification. On the one hand, it acts as an extraction agent, and on the other hand, forms a homogeneous system with the microalgae oil, methanol and the catalyst. Usually, stirring and heating are required to promote the reaction of transesterification. However, co-solvent-assisted in situ transesterification should require less energy in the form of heating or stirring [25]. Releasing the lipids contained inside the cells of the microalgae could be difficult during in situ transesterification, therefore large volumes of solvent are generally required. This takes a lot of energy and substantially increases costs. Another factor to be considered is that oil must be released and extracted without contamination from other cellular components, such as DNA or chlorophyll [26]. Several approaches have been proposed for the selectively disrupting the cell wall, for instance, using ultrasound, microwaves, enzymes and pressurized fluids [27–29]. Although the mechanisms behind each of these techniques are different, most involve rupturing the cells to release the lipids within the cytoplasm. The principle behind the ultrasound method is the generation of sound waves that propagate through the fluid causing alternate cycles of high and low pressures. During the high pressure cycle, small bubbles, which were generated during the low pressure cycle, violently collapse and result in the phenomenon called cavitation [1]. High pressure and liquid velocities create shear forces upon the microalgae during cavitation, mechanically breaking the cellular structures and enhancing the transfer of extracted lipids. This methodology improves the lipid yield (output) between 50 to 500% and produces up to a 10-fold reduction in extraction times [1]. Martinez-Guerra et al., 2014 [30] studied the effect of ultrasound on the in situ transesterification of lipids from Chlorella sp. using ethanol as co-solvent for lipid extraction and as a reagent in the transesterification reaction. Optimal conditions were: a 1:6 to 1:9 ratio of microalgae:ethanol (wtv), 2% w/w of catalyst (sodium hydroxide) and a 6-min treatment time at 490 W of ultrasound. The weight yield was 18.5% and the conversion to ethyl esters was 95.0%. On the other hand, Keris-Sen et al., 2014 [31] tested different ultrasound intensities (0.1–0.5 W/mL), at a 30 kHz frequency for 5 to 60-min cycles. They also studied the effect of ultrasound on the lipid extraction efficiency in presence of co-solvents (hexane or a chloroform/methanol mixture). The results showed that in the sample with the mixture of co-solvents and no ultrasound, 13.6% of lipids could be extracted from the dry biomass. However, the biomass still contained a significant amount of lipids that had not been removed (43.4%). When the extraction process was performed by ultrasound, the use of the co-solvents mixture managed to extract 26.8% of lipids from the dry biomass. Consequently, ultrasound significantly increased the extraction of lipids [31]. Therefore, additional information is required to evaluate this type of biomass used for biodiesel conversion, which needs a process that is much more complex than that traditionally used to obtain this biofuel from vegetable oils or animal fats. In this study, there were shown results for in situ transesterification of dried biomass of Spirulina sp. Several variables related to the reaction system, as well as the components of the reaction mixture were studied. Addition of different co-solvents (chloroform and hexane), and the use of ultrasound in in situ transesterification method were evaluated. The ultrasound power and treatment
514
N. Martínez et al. / Fuel 202 (2017) 512–519
times in absence and presence of different co-solvents were also studied. Since there are not too many references to the ester content achieved with dried biomass, this study focused on the main differences with respect to lipid extraction, the weight yield relative to the starting biomass, as well as the content of fatty acid methyl esters (FAME), in order to define the most suitable conditions and co-solvents for biodiesel production. 2. Materials and methods 2.1. Materials Dry Spirulina sp. LEB 18 powder was provided by the School of Chemistry and Foods of the Universidade Federal do Rio Grande of Brazil. Biomass was obtained by culturing in Zarrouk medium [32] using Raceway ponds at room temperature and daylight as the light source. In all cases, the biomass was separated by filtration from the crop, then pelletized and dried in an oven for 6 h at 60 °C and then stored at room temperature. Before using it in the different methods, any residual moisture was eliminated by freeze-drying. Standards, reagents and solvents in this study were analytical grade provided by Sigma-Aldrich (USA). 2.2. Methods 2.2.1. Biomass preparation Strands of Spirulina sp. LEB 18 were ground with a domestic coffee grinder to obtain a fine powder. The powdered biomass was kept in a capped container at - 18 °C until use. 2.2.2. Determination of lipid content Lipid content was measured by gravimetry after extraction with a solvent mixture (chloroform: methanol 2:1 v/v, using the method by Folch, Less and Stanley, adapted from Christie and Han, 2010 [33]). The total lipid content was measured in triplicates. About 0.2 g of biomass were weighed, to which was added 5.0 mL of the solvent mixture, and stirred in a vortex. Then, the mixture was sonicated for about 15 min. The phases were then separated by centrifugation at 3500 rpm. This procedure was repeated three more times. Centrifugation supernatants were pooled together and filtered through filter paper. An aqueous solution of 0.88% KCl (equivalent to one fourth of the total supernatant volume) was added, mixed, and centrifuged. The organic phase separated and washed with one fourth of the mixture volume using a mixture of methanol:water (2:1). It was then mixed and centrifuged. The organic phase was separated and a small amount of anhydrous Na2SO4 was added. This was then centrifuged and separated. The organic phase was dried to constant weight. The acid value as well as free fatty acids were titrimetrically determined using AOCS method (Cd 3d-63) [34]. 2.2.3. Determination of the fatty acid composition and the percentage of lipids that can be converted into esters (purity of the Folch extract) About 30 mg of sample were placed inside a tube with a screw cap and 2.0 mL of 0.05 M MeONa in MeOH solution were added. The air inside the tube was flushed with a stream of nitrogen and the mixture was refluxed for 10 min (100 °C). After that time, it was allowed to cool down and 2.0 mL of a methanolic HCl solution (prepared by dissolving 8 mL of acetyl chloride (C2H3ClO) in 100 mL of methanol) were added and further refluxed for 10 min. Then, the mixture was allowed to cool down and the internal standard was added (heptadecanoic acid methyl ester). Then 5.0 mL of hexane and 7.0 mL of distilled water were added to extract the FAME formed in the reaction. This was stirred and centrifuged at
4000 rpm for 10 min and the organic phase was separated. Onto the organic phases, a small amount of anhydrous Na2SO4 was added. Then it was stirred, centrifuged and the solution of FAME was separated. A volume of 1.0 lL of the FaME solution was injected into a gas chromatograph (Shimadzu, model GC-14B with FID detector, Kyoto, Japan). A capillary column was used (Supelco SP2330, 30 m 0.25 mm 0.2 mm), using nitrogen as the carrier gas. Temperature program started at 60 °C for 2 min, then the temperature was raised to 200 °C at a rate of 10 °C/minute, and was finally raised at a rate of 5 °C/min to 240 °C and kept at that temperature for 7 min. The composition in fatty acid was measured in triplicate. Using this method, the amount lipids that can be converted into FAME can be calculated using the following equation:
mfat ¼ ½mstd :ðAtotal Astd Þ=Astd %convertible lipids ¼ mfat 100=mbiomass 2.2.4. In situ alkaline transesterification Fifty milliliters of methanol containing 0.5 g of potassium hydroxide (2% of catalyst) were added to 25 g of biomass in a 500 mL batch reactor. If a co-solvent was added to the reaction mixture, the volume was calculated to produce a 2:1 co-solvent: methanol ratio. Chloroform or hexane were tested as co-solvents. The temperature was set to 55 °C and the treatment time to 2 h. The reaction mixture remained under continuous stirring. When the time of reaction was up, the phase liquid was separated from the phase solid. The solid phase was washed several times with co-solvent, which pooled with the previously collected liquid phase. FAME were separated from the liquid phase using several washes with distilled water, in order to eliminate the catalyst and glycerol excess. Glycerol was formed as a reaction byproduct. A small amount of anhydrous sodium sulfate was added to the organic phase, and then the mixture was centrifuged and separated. The organic phase was dried to constant weight and the recovered weight yield was calculated (wt%). 2.2.5. Ultrasound-assisted in situ alkaline transesterification The same procedure used in 2.2.4 was performed in order to evaluate the efficiency of the ultrasound-assisted transesterification process. Ultrasound equipment (Elmasonic model P60 H, Singen, Germany) is capable of setting the ultrasound power to different values up to 180 W and features temperature control (30–80 °C). The temperature was set to 55 °C. Due to the temperature increase derived from ultrasound, a system of cooling (coil) was used to keep the temperature at the set value. We studied the following transesterification times: 15, 30, 60 and 120 min; and different ultrasound intensities, 44% (80 W), 60% (108 W), 80% (144 W) and 100% (180 W) at a 37 kHz frequency. As it is explained in Section 2.2.4, the biomass:methanol ratio was always 50:100 (weight/volume) and, if applicable, the volume of co-solvent (hexane or chloroform) was 200 parts. The reaction mixture was kept inside the equipment and no stirring system was used. Upon completion of the reaction, under all the different conditions, the liquid phase was separated from the solid phase. The solid phase was washed three times with hexane (50–75 mL), which was pooled with the collected liquid phase from the previous step. The FAME were then separated from the mixture and the excess of catalyst and glycerol were eliminated with several washes with water distilled. Once the pH of the wash water was equal to that of the distilled water, a small amount of anhydrous sodium sulfate was added to the organic phase to eliminate residual water, and was filtered through filter paper. The solvent was
N. Martínez et al. / Fuel 202 (2017) 512–519
evaporated to constant weight using a rotary evaporator and the ‘‘gross” weight yield was calculated (quantity of product with respect to the starting biomass). 2.2.6. Purification by solid phase extraction (SPE) The transesterification product obtained was purified using SPE. One gram of active silica was prepared and properly packed into a column using solvent. Ten milliliters of hexane or petroleum ether were added to the column and allowed to elute until 1 to 2 mL of solvent were sitting on top of the column. Between 200 and 300 mg of sample were dissolved in the smallest possible volume of solvent and sewed onto the column. Once sample entered the column, 5 mL of a petroleum ether:diethylether (95:5) mixture were added to the column and eluted, taking care not to let the column dry up. The eluate was collected in a clean and dry tube. This procedure was repeated 3–4 more times. The collected eluate (about 20 mL) was dried under a nitrogen stream to constant weight, in order to measure the ester content later on. 2.2.7. Determination of the ester content The ester content was determined according to the EN 14103 standard. In a 10 mL volumetric flask, 200 mg of the purified sample was dissolved using hexane. In a 4 mL vial, 200 lL of the previously prepared sample solution were mixed with 200 lL of a 4.0 mg/mL solution of internal standard (17:0). This mixture was analyzed in a Shimadzu-14B gas chromatograph using a SP-2330 capillary column. The temperature program was as follows: starting temperature of 160 °C, then heating at 4 °C/min up to 230 °C, and then maintenance at 230 °C for 10 min. The ester content was analyzed in triplicate. 3. Results and discussion 3.1. Lipid content and fatty acid composition The lipid content was found to be 7.0 ± 0.4% on dry basis, using the method by Folch, Less, and Stanley. Oil contained a very low percentage of free fatty acids (0.50 mg KOH/g oil). Table 1 shows the fatty acid composition of the oil extracted from Spirulina sp. LEB 18. The main fatty acids were 16:0 (palmitic acid), followed by 18:3 (n–6) (linolenic acid) and 18:2 (n–6) (linoleic acid). Additionally, the purity in the Folch extract was 30.2% (according to Section 2.2.3). The amount of lipids that can be converted into FAME was calculated multiplying the total lipid content (Folch method) by the purity of the extract. Accordingly only 2.1% of the lipids (7.0%) were convertible into FAME. 3.2. Effect of time and co-solvent type on the efficiency of the ultrasound-assisted in situ alkaline transesterification Treatment time plays a crucial role in the in situ transesterification reaction, not only because it needs to be enough to release
Table 1 Fatty acid composition of Spirulina sp LEB 18. Fatty acid
Percentage (wt%)
14:0 16:0 16:1 n7 18:0 18:1 trans 18:1 n9 18:1 n7 18:2 n6 18:3 n6 Total identified FAs
3.9 44.2 3.7 2.1 1.6 5.4 1.2 11.1 23.3 96.5
515
lipids into the reaction medium, but also to allow transesterification to occur. The main reported advantage of the ultrasound method is that it requires shorter treatment times [31,35]. When the transesterification reaction was carried out under the conditions described in Section 2.2.4 (without ultrasound), the weight yield was 12.8% (0.9% with respect to biomass) after 2 h of reaction, and the ester content was just 63.6%. Weight yields (with respect to dry biomass) and ester contents are shown in Fig. 1. These results correspond to ultrasoundassisted in situ alkaline transesterification at 80 W of ultrasound power, no co-solvent, and different treatment times (1, 2, 4, and 8 h). These results show that, without using co-solvents, the yield tends to decrease after 2 h of reaction. However, no significant differences in yields were observed when comparing 1 h and 8 h. These results suggest that the maximum yield was reached at 2 h of reaction. Addition of a co-solvent (chloroform or hexane) to the reaction medium increased the yield with compared to absence of cosolvent (0.6% and 0.9% with respect to biomass). Hexane yielded 1.0% in weight, with respect to biomass. Chloroform, on the other hand, considerably improved the weight yield, allowing to extract 43% of lipids (3.0% with respect to biomass). Therefore, the effect of the chloroform and methanol mixture extracted lipids more efficiently than hexane and methanol. The results showed that combining the cell ‘‘disruption” effect of ultrasound with co-solvents increases lipid extraction, compared working without co-solvents. Fig. 1 also shows the ester content vs treatment times, under the different conditions at various treatment times (1, 2, 4 and 8 h), with or without using co-solvents (chloroform or hexane). The results show that without co-solvents, ester content above 90% were obtained in all those cases. In particular, the ester content was particularly high at 1 and 2 h of reaction (96.8% and 97.6% respectively), exceeding the requirements of international regulations (96.5%). This shows that ultrasound was clearly favorable, since the ester content was only 63.6% in its absence. Therefore, regulatory requirements were met with short treatment times (under two hours). The use of co-solvents, as previously mentioned, enhances the extraction of lipids from the biomass by improving mass transfer. The type of co-solvent affects which lipids are extracted. Some of these lipids, such as pigments or polar lipids, are not subsequently converted to FAME during the transesterification reaction [30]. Therefore, the choice of co-solvent used in the process may also affect the final ester content. The treatment time chosen for testing the co-solvent (hexane or chloroform) was 2 h, because it had produced the greater ester content in absence of co-solvent. Fig. 1 shows that the ester content was lower when co-solvents were used, compared to similar conditions without them. None of those cases complied with regulations. The ester content using Hexane was 92.8%, and using chloroform was 69.7%. These results can be explained because cosolvents extract different kinds of lipids, which reduce the transesterification efficiency, therefore lowering the ester content. Hexane enables the extraction of non-polar lipids (triglycerides), which are the main substrate of the reaction. Since chloroform is miscible with methanol in certain proportions, together they may extract both polar and non-polar lipids, leading to lower conversion of lipids into esters. This was shown in the ester content, which did not exceed 70% when chloroform was used as co-solvent. To assess the effect of co-solvents in the overall yield (efficiency of the process), we studied calculated the product between the weight yield and the ester content. The result is the amount of FAME per gram of biomass. This parameter reveals the effect of extraction and the efficiency of the reaction as a whole. The results are shown in Fig. 2.
516
N. Martínez et al. / Fuel 202 (2017) 512–519
Fig. 1. Weight yield and ester content under different conditions of ultrasound-assisted in situ alkaline transesterification (80 W). (Versus treatment time and co-solvent, if applicable).
Fig. 2. Amount of FAME per gram of biomass vs treatment time and type of co-solvent.
In absence of a co-solvent, the highest amount of FAME per gram of biomass was obtained at two hours of reaction. For higher treatment times, this value dropped. This happens because the ester content decreases to a greater extent than weight yield. Chloroform produced the highest quantity of FAME per gram of biomass because its weight yield was high. Apparently, two opposing effects come into play when using co-solvents. On the one hand, the co-solvent can improve the weight yield by enhancing lipid extraction. On the other hand, however, these extracted lipids can be very diverse and not be able to be converted into esters. Therefore, the choice of co-solvent represents a compromise between weight yield and ester content. 3.3. Effect of ultrasound power, treatment time and type of co-solvent on the weight yield and ester content Until this point, the results corresponded to 80 W of ultrasound power. In order to observe the effect of ultrasound power on
weight yield and ester content, we tested the following conditions: 108 W, 144 W and 180 W (maximum power of the equipment), using a treatment time of only 30 min. The greater ultrasound intensities were expected to contribute to the disruption of the microalgae’s cell wall, facilitating the release of lipids into the reaction medium. However, previous reports have suggested otherwise. This could be attributed to excessive cavitations forcing bubbles to join each other instead of interacting with the cells, thus preventing their disruption [31]. The effects of ultrasound power on weight yield and ester content are shown in Fig. 3. These results indicate that increasing ultrasound power increases the weight yield. However, this increase was not significant; weight yield increased from 1.1% (15.3% of lipid material) to 1.6% (22.8% of lipid material) when ultrasound power was increased from 80 W to 180 W. It is worth mentioning that those yields were achieved in short treatment times (30 min), a favorable aspect that is consistent with previous literature [29,30,1].
N. Martínez et al. / Fuel 202 (2017) 512–519
517
Fig. 3. Weight yield and ester content under different conditions of in situ alkaline transesterification and 30 min of treatment time versus ultrasound power in absence of cosolvent or using hexane as a co-solvent.
The results in Fig. 3 show that the ester content increases with higher ultrasound intensities. For instance, the ester content after 30 min of reaction at 80 W was 82.6%, which increased to 91.6% and 95.2% using 144 W and 180 W ultrasound power, respectively. Therefore, results show that the increasing the ultrasound power in the in situ alkaline transesterification process (only 30 min of treatment time in absence of a co-solvent) improves the process since a larger amount of lipids are extracted, improving both the weight yield and the ester content. This can be clearly seen in Fig. 4, which shows the process efficiency in terms of the amount of FAME per gram of biomass (calculated as the product of weight yield and ester content). Since co-solvents enhance the extraction of lipids, we also studied the effect on weight yield and the content ester, as shown in Fig. 3. The reaction was carried out using 180 W, under the same conditions as those mentioned above, except for the addition of a co-solvent (2:1 (v/v) with respect to methanol). In this case, hexane was the chosen co-solvent, since our previous results showed it did not decrease the ester content compared to chloroform. The
extraction yield after 30 min of reaction, under the same operating conditions was 31% (2.2% with respect to biomass), which was higher than that obtained with 80 W and the same co-solvent. Therefore, lipid extraction was enhanced by adding a co-solvent and increasing ultrasound power. The increase of ultrasound power (over 80 W) has a greater effect on weight yield at short treatment times, since longer treatment times (1 h or more) showed lower values, as shown in Fig. 1. As already mentioned, the effect of the co-solvent in the ester content primarily depends on the types of lipids that are extracted, as well as the transesterification reaction. The ester content was 85.9% (30 min of reaction), which is lower than in absence of co-solvent. Therefore, treatment times have two different effects. Shorter treatment times (30 min) improve lipid extraction efficiency, thus increasing weight yield, where both ultrasound power and presence of co-solvent make a positive contribution. Nevertheless, the ester content falls below the required value of 96.5%, suggesting that the transesterification reaction requires of more time to reach completion.
Fig. 4. Amount of FAME per gram of biomass versus presence or absence of hexane as co-solvent.
518
N. Martínez et al. / Fuel 202 (2017) 512–519
Based on previous results, the optimum ultrasound power is 180 W (the highest power of the equipment) since it produced the highest ester content and weight yield without any cosolvent. However, these results were carried out using a fixed treatment time of 30 min. Thus, we were interested in finding how treatment times affects extraction under this optimum condition. We studied three additional treatment times, 15, 60, and 120 min, in order to observe how the process evolves. The results of the weight yield and ester content versus treatment time are shown in Fig. 5. Yields obtained in these times did not vary significantly. Nevertheless, the shortest treatment time (15 min) showed the best weight yield (2.0% of biomass) which represents a 28% extraction of lipids). According to the expected behavior, increasing the treatment time increased the ester content. The highest ester content achieved was 96.9% after 120 min of reaction at 180 W (the weight yield was 1.8% of biomass). However, 30, 60 and 120 min of reaction did not show significant differences in terms of ester content. This is also consistent with other reports indicating that short treatment times are required to achieve suitable conversions [26]. In this study, 180 W of ultrasound power and 30 min of treat-
ment time were enough to yield a 95.2% ester content, under the studied conditions. The value of weight yield for 80 W was 0.9% (2 h of reaction) and 1.8% for 180 W (2 h of reaction), whereas the ester contents were 97.6% and 96.9%, respectively. Thus, higher ultrasound power increased the weight yield but not the ester content. To evaluate the combined effect of the lipid extraction and the efficiency of the transesterification reaction, we once again calculated the process efficiency parameter. This equates to the amount of FAME per gram of biomass (Fig. 6). According to Fig. 6 and as mentioned, treatment time was shown to have two different effects: shorter times extract more lipids improving weight yields, while longer times increase the ester content. While there were large differences in the amount of FAME for the different times, the greater amount of FAME per gram of biomass was reached for 15 to 120 min. Consequently, choosing the treatment time is a compromise between maximizing weight yields or ester content. In our case, after 15 min the ester content was hardly over 85%, whereas, after 120 min, it reached 96.5%.
Fig. 5. Weight yield and ester content versus treatment times at maximum ultrasound power (180 W) in the in situ alkaline transesterification reaction.
Fig. 6. Amount of FAME per gram of biomass versus treatment time at maximum power (180 W) in ultrasound-assisted in situ alkaline transesterification.
N. Martínez et al. / Fuel 202 (2017) 512–519
4. Conclusions Results show that ultrasound is a suitable tool to improve the extraction of lipids from biomass because it produces enough cellular disruption. Additionally, it increases the reaction extent of the transesterification reaction, therefore enabling higher ester contents under suitable reaction conditions. In situ alkaline transesterification reaction without a co-solvent yielded an ester content above the value required by regulations (96.5%). The use of co-solvents seems to produce two opposing effects. On the one hand, it can improve the weight yield by enhancing extraction efficiency. However, it also modifies the quality of extracted lipids, increasing the proportion of ‘‘nonconvertible” lipids, and consequently decreasing the ester content. Therefore, priority must be given to one of these two parameters. The maximum power (180 W) was the best condition when treatment times were short (30 min). Under these conditions, the weight yield and ester content were 1.6% and 95.2%, respectively. At the maximum power (180 W) the ester content increased with longer times. The ester content was 96.9% after 2 h of reaction at 180 W. Weight yields did not vary significantly over time, but the maximum value corresponded to the shorter times (15 min at 180 W). The optimal power was 180 W (the highest possible with our equipment) which produced a high ester content (96.9%) and a high weight yield (1.8% with respect to biomass, corresponding to a 25.2% extraction of lipids) without needing a co-solvent. Therefore, if the limiting factor it was the capacity the equipment, major ultrasound powers could be studied to see a possible improve in the efficiency process. Acknowledgment This work was funded by ANCAP and CSIC (Uruguay) with financial support and PEDECIBA-Química programs (Uruguay). References [1] Mubarak M, Shaija A, Suchithra TV. A review on the extraction of lipid from microalgae for biodiesel production. Algal Res 2015;7:117–23. [2] Ahmad AL, Mat Yasin NH, Derek CJC, Lim JK. Microalgae as a sustainable energy source for biodiesel production: A review. Renew Sustainable Energy Rev 2011;15:584–93. [3] Guo F, Fang Z, Tian XF, Long YD, Jiang LQ. One-step production of biodiesel from Jatropha oil with high-acid value in ionic liquids. Bioresour Technol 2011;102:6469–72. [4] Yee KF, Wu JCS, Lee KT. A green catalyst for biodiesel production from jatropha oil: optimization study. Biomass Bioenergy 2011;35:1739–46. [5] Yang CY, Fang Z, Li B, Long Y. Review and prospects of Jatropha biodiesel industry in China. Renew Sustainable Energy Rev 2012;16:2178–90. [6] Kafuku G, Lam MK, Kansedo J, Lee KT, Mbarawa M. Croton megalocarpus oil: a feasible non-edible oil source for biodiesel production. Bioresour Technol 2010;101:7011–5. [7] Kansedo J, Lee KT, Bhatia S. Cerbera odollam (sea mango) oil as a promising non-edible feedstock for biodiesel production. Fuel 2009;88:1148–50. [8] Ghadge SV, Raheman H. Process optimization for biodiesel production from mahua (Madhuca indica) oil using response surface methodology. Bioresour Technol 2006;97:379–84.
519
[9] Canoira L, Alcantara R, Garcia-Martinez J, Carrasco J. Biodiesel from Jojoba oil wax: transesterification with methanol and propeties as a fuel. Biomass Bioenergy 2006;30:76–81. [10] Daroch M, Geng S, Wang G. Recent advances in liquid biofuel production from algal feedstocks. Appl Energy 2013;102:1371–81. [11] Demirbas A. Use of algae as biofuel sources. Energy Convers Manage 2010;51:2738–49. [12] Brennan L, Owende P. Biofuels from microalgae—a review of technologies for production, processing, and extractions of biofuels and co-products. Renew Sustainable Energy Rev 2010;14:557–77. [13] Chisti Y. Biodiesel from microalgae beats bioethanol. Trends Biotechnol 2010;26:126–31. [14] Miao X, Wu Q. Biodiesel production from heterotrophic microalgal oil. Bioresour Technol 2006;97:841–6. [15] Mata TM, Martins AA, Caetano NS. Microalgae for biodiesel production and other applications: a review. Renew Sustainable Energy Rev 2010;14:217–32. [16] Khan SA, Rashmi M, Hussain MZ, Prasad S, Banerjee UC. Prospects of biodiesel production from microalgae in India. Renew Sustainable Energy Rev 2009;13:2361–72. [17] Malcata FX. Microalgae and biofuels: a promising partnership? Trends Biotechnol 2011;29:542–9. [18] Nautiyal P, Subramanian KA, Dastidar MG. Kinetic and thermodynamic studies on biodiesel production from Spirulina platensis algae biomass using single stage extraction–transesterification process. Fuel 2014;135:228–34. [19] Becker EW. Microalgae. Cambridge, New York: Cambridge Univ. Press; 1994. [20] Vonshak A, Tomaselli L. Arthrospira (Spirulina) systematic and ecophysiology. In: Whitton BA, Potts M, editors. The Ecology of Cyanobacteria in the Netherlands. Kluwer Academic Publishers; 2000. [21] El-Shimi HI, Nahed K, Attia ST, El-Sheltawy GI, El-Diwani. Biodiesel production from Spirulina-Platensis microalgae by in-situ transesterification process. J Sustainable Bioenergy Syst 2013;3:224–33. [22] Hidalgo P, Toro C, Ciudad G, Navia R. Advances in direct transesterification of microalgal biomass for biodiesel production. Rev Environ Sci Biotechnol 2013;12:179–99. [23] Baumgartner TRS, Burak JAM, Baumgartner D, Zanin GM, Arroyo PA. Biomass production and ester synthesis by in situ transesterification/esterification using the microalga Spirulina platensis. Int J Chem Eng 2013:1–7. [24] Hidalgo P, Toro C, Ciudad G, Schober S, Mittelbach M, Navia R. Evaluation of different operational strategies for biodiesel production by direct transesterification of Microalgal Biomasa. Energy Fuels 2014;28:3814–20. [25] Xu R, Mi Y. Simplifying the process of microalgal biodiesel production through in situ transesterification technology. J Am Oil Chem Soc 2011;88:91–9. [26] Scott SA, Davey MP, Dennis JS, Horst I, Howe CJ, Lea-Smith DJ, et al. Biodiesel from algae: challenges and prospects. Curr Opin Biotech 2010;21:277–86. [27] Lee JY, Yoo C, Jun SY, Ahn CY, Oh HM. Comparison of several methods for effective lipid extraction from microalgae. Bioresour Technol 2010;101:75–7. [28] Ranjan A, Patil C, Moholkar SV. Mechanistic assessment of microalgal lipid extraction. Ind Eng Chem Res 2010;49:2979–85. [29] Araujo GS, Matos LJBL, Fernandes JO, Cartaxo SJM, Gonçalves LRB, Fernandes FAR, et al. Extraction of lipids from microalgae by ultrasound application: Prospection of the optimal extraction method. Ultrason Sonochem 2013;20:95–8. [30] Martinez-Guerra E, Gude VG, Mondala A, Holmes W, Hernandez R. Microwave and ultrasound enhanced extractive-transesterification of algal lipids. Appl Energy 2014;129:354–63. [31] Keris-Sen UD, Sen U, Soydemir G, Gurol MD. An investigation of ultrasound effect on microalgal cell integrity and lipid extraction efficiency. Bioresour Technol 2014;152:407–13. [32] Zarrouk C. Contribution à l’étude d’une cyanophycée. Influence de divers facteurs physiques et chimiques sur la croissance et photosynthese de Spirulina maxima Geitler Ph.D. Thesis. University of Paris; 1966. [33] W.W. Christie, X. Han, Lipid analysis: Isolation, separation, identification and lipidomic analysis (4th Ed.), Oily Press Lipid Library Series No. 24, 2010. [34] AOCS methods for biodiesel feedstock quality. In: Official Methods and Recommended Practices of the AOCS, (6th Ed.), Second printing, 2012. [35] Gude VG, Patil PD, Martinez-Guerra E, Deng S, Khandan NN. Microwave energy potential for biodiesel production. Sustainable Chem Process 2013;1(5):1–31.