Continuous production of biodiesel from microalgae by extraction coupling with transesterification under supercritical conditions

Continuous production of biodiesel from microalgae by extraction coupling with transesterification under supercritical conditions

Accepted Manuscript Continuous production of biodiesel from microalgae by extraction coupling with transesterification under supercritical conditions ...

1MB Sizes 0 Downloads 70 Views

Accepted Manuscript Continuous production of biodiesel from microalgae by extraction coupling with transesterification under supercritical conditions Dan Zhou, Baoquan Qiao, Gen Li, Song Xue, Jianzhong Yin PII: DOI: Reference:

S0960-8524(17)30604-1 http://dx.doi.org/10.1016/j.biortech.2017.04.097 BITE 17992

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

8 February 2017 22 April 2017 24 April 2017

Please cite this article as: Zhou, D., Qiao, B., Li, G., Xue, S., Yin, J., Continuous production of biodiesel from microalgae by extraction coupling with transesterification under supercritical conditions, Bioresource Technology (2017), doi: http://dx.doi.org/10.1016/j.biortech.2017.04.097

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Continuous production of biodiesel from microalgae by extraction coupling with transesterification under supercritical conditions Dan Zhou a, Baoquan Qiao a, Gen Li a, Song Xue b, Jianzhong Yin a, a

State Key Laboratory of Fine Chemicals, School of Chemical Machinery, Dalian University of Technology, Dalian 116024, P. R. China b

Marine Bioproducts Engineering Group, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 16023, China

Abstract Raw material for biodiesel has been expanded from edible oil to non-edible oil. In this study, biodiesel continuous production for two kinds of microalgae Chrysophyta and Chlorella sp was conducted. Coupling with the supercritical carbon dioxide extraction, the oil of microalgae was extracted firstly, and then sent to the downstream production of biodiesel. The residue after decompression can be reused as the material for pharmaceuticals and nutraceuticals. Results showed that the particle size of microalgae, temperature, pressure, molar ration of methanol to oil, flow of CO2 and nhexane all have effects on the yield of biodiesel. With the optimal operation conditions: 40 mesh algae, extraction temperature 60 ⁰C, flow of n-hexane 0.4ml/min, reaction temperature: 340 ⁰C, pressure: 18-20MPa, CO2 flow of 0.5L/min, molar ration of methanol to oil 84:1, a yield of 56.31% was obtained for Chrysophyta, and 63.78% for Chlorella sp due to the higher lipid content.



Corresponding author. E-mail address: [email protected] (J. Yin) 1

Key words: biodiesel;microalgae;supercritical fluid; extraction coupling with reaction; continuous;

1. Introduction To avoid the crisis for the competition between human’s food and feed stock of biodiesel, traditional biodiesel has a great rely on the edible oil, some researchers have attempted to make nonedible lipids as the feed for biodiesel, which include non-edible seeds from plants, waste oil and microalgae. That is Jatropha, chinaberry seeds, yellow fruit, Sterculia foetida and rubber tree, Mahua oil, castor oil, tobacco, etc (Karmakar et al., 2010). It realizes the process of waste recovery and reuse that turn waste oil to biodiesel, and reduce the pollution for environment. However, most of them have high acid value and contain too many impurities which are kinds of polymers as well as degradation products, and the supply has a great fluctuation, which is hard to use as the stable raw oil for large-scale industrialization. Microalgae cultivation has low requirement for space and high efficiency of photosynthesis. They have been considered as a bright prospect for biodiesel, as can be produced using saline and waste water with a remarkable growing rate and product yield. Compared with other material, microalgae has obvious advantages which are easy to cultivate, using non drinking water, transition solar to chemicals (Galadima and Muraza, 2014), and short growth period. Moreover, sulfur free feed also reduce the sulfide in the tail gas. Algae have much application value in Pharmaceuticals, Nutraceuticals, Cosmetics and Aquaculture purpose. Besides biodiesel, methane, ethanol and hydrogen can also be generated from microalgae. Due 2

to the high content of polyunsaturated fatty acids such as linolenic acid, EPA and DHA, abundance of vitamins, minerals, and trace elements, they can be made as the feed of health products (Chauton et al., 2015). Recent research works on microalgae have identified this new bio-material as a promising technology for bioenergy production, wastewater treatment, the development of high value added products and CO2 capture. Although the oil content of microalgae is similar to other feeds, the global annual output is larger. One of the first issues is extracting oil from microalgae in order to transfer the microalgae oil to biodiesel. The most common extraction processes are solvent extraction, ionic liquid extraction, and subcritical water extraction and so on, in which solvent extraction is the most mature technology and have general use in industrial. However, the toxicity of organic solvent and the high energy consumption for recovery make it harmful for the environment. In particular, the residue can be used as the material for pharmaceuticals and nutraceuticals. ScCO2 as a clean and green solvent, easy to separate with extracts and no residue in product, all these characters made it a suitable solvent for the extraction of microalgae (Halim et al., 2012). Andrich using scCO2 to extract oil from microalgae and obtained a extraction curve(Andrich et al., 2005). As the extraction time prolong, the extraction yield decreased. Within 5000s, the extraction amount was over 80%, while exceeding 10000s, the extraction amount was not increasing a lot, and the trend of increase become slower. That’s because the driving force of extraction is the oil concentration between bulk scCO2 and the inner cell. The initial concentration difference is large, so the extraction 3

speed was fast, extend the extraction time, the concentration difference become smaller and the driving force reduced, so the tendency become flat. Taher studied the extraction conditions for microalgae (Taher et al., 2014), and found that the temperature and pressure has strong impact on the extraction yield, but less effect on the extraction yield. The optimal extraction conditions are temperature at 53ºC, 50MPa pressure and the flow of CO2 was 1.9g/min, and the extraction yield reached 7.41wt% (dry algae). Solana utilized the ethanol as the entrainer to improve the extraction yield (Solana et al., 2014). The results showed that the extraction yield reached the highest at the temperature 60ºC, 30MPa pressure, and CO2 flow 0.4kg/h and 5% ethanol. Furthermore, there is a crossover pressure 25MPa for extraction algae oil. That is when the pressure above the crossover pressure, the key factor for extraction is the vapor pressure of solute rather than the density of CO2. Cheng et al compare the extraction with organic solvent and scCO2, ethyl acetate and methanol mixed solvent has an extraction yield of 98.7%, much more than the extraction for scCO2, which is 61.6% (Cheng et al., 2011). But ball-milling pretreatment can enhance the scCO2 extraction to 98.7%. Millao et al investigated the extraction of oil and cartenoids from nannochloropsis gaditana using supercritical carbon dioxide (Millao and Uquiche, 2016), by response surface methodology, the maximum oil yield 152.2g/kg dry substrate was gained at 64⁰C and 59.3MPa. They found that temperature had a greater effect than CO2 density. The extracted oil can be used to prepare biodiesel by supercritical methanol transesterification method, and its yield was affected by the operation conditions, such 4

as temperature, reaction time, ratio of alcohol to oil (ratio of methanol to algae) and pressure. Reddy conducted the transesterification of algae (Nannochloropsis Salina) in the condition temperature 265ºC, 20min, dry algae to ethanol 1:9 (wt/v), and a maximum yield 67% was obtained (Reddy et al., 2014). Patil also used this one-step process for direct liquefaction and conversion of wet biomass containing about 90% of water to biodiesel under supercritical methanol conditions (Patil et al., 2011), in the optimal condition: wet algae to methanol (wt./vol.) ratio of around 1:9, reaction temperature and time of about 255 °C, and 25 min lead to yield above 80%. They used the Nannochloropsis Salina, too. Due to the deviation of species and culture environment, the ingredients of microalgae are different, but the free fat acid and water content have a great impact on the production process. When algae changed to Nannochloropsis gaditana as feed (Jazzar et al., 2015), only 48% yield was received in the optimized condition at 255–265 °C, 50 min reaction time, and using a methanol to dry algae ratio of 10:1 (vol./wt.) Batch reactions were employed in the above work. Continuous flow process (He et al., 2007; ZHOU et al., 2010) can be brought so as to provide much bigger manufacturing scale, and cut the cost. Nan et al studied the continued production of biodiesel through non-catalytic transesterification of microalgae oil with methanol and ethanol (Nan et al., 2015), and optimization of continuous process by RSM showed that the best condition were 320 °C, 15.2 MPa, 19:1 molar ratio, 31 min, 7.5 wt% of water content for methanol transesterification, for ethanol, that is 340 °C, 17MPa, 33:1 molar ratio, 35 min, and also 7.5 wt% of water content. The corresponding yields of fatty acid 5

methyl ester (FAME) and fatty acid ethyl ester (FAEE) were 90.8% and 87.8%, respectively. To our knowledge, little has been reported in regard to coupling supercritical CO2 (scCO2) extraction and continuous transesterification for the microalgae. Most works are focusing the in-situ transesterification, although the in-situ process can utilize the wet microalgae directly, the residue after the reaction can’t be used again as the raw material of health products. That is a waste for the other ingredients of algae. Therefore, in this work, the continuous supercritical methanol transesterification coupling with scCO2 extraction without catalyst for the production of biodiesel was investigated, and two kinds of microalgae Chrysophyta and Chlorella sp were adopted as the feed stock. This work aims to expand the feed of biodiesel, and utilize the green solvent CO2 so as to recover and reuse the residue. Through the investigation of the influence parameters of supercritical extraction and continuous transesterification for microalgae, the optimal operation conditions were obtained. We expected to expand and comprehensively utilize the raw material of biodiesel, and realize the continuous production in industry. 2. Materials and Methods 2.1 Materials and instruments Chrysophyta and Chlorella sp were provided by Dalian institutions of Chemical Physics, with fatty acid contents of 10.5 wt% and 19.8 wt% respectively (analyzed by in-situ transesterification method due to the reason that the lipids extracted by organic solvents for microalgal case are very complicated, including the components without the fatty acid chains such as lipid-soluble pigments, sterol and so on, which cannot be 6

converted to FAME, and the lipid content obtained by extraction method is usually higher than the actual value of the lipid components that could be converted to FAME) (Liu et al., 2015). Their GC profiles and ingredients content are listed in supplementary. Methanol and n-hexane with the purification of analytical grade were obtained from Beijing chemical works. N2 and CO2 were supplied by Dalian Guangming Special Type Gas Co., Ltd. The quality of biodiesel is analyzed by Agilent 6890 gas chromatography. 2.2. Analysis The fatty acid methyl ester (FAME) content of the samples was determined by an Agilent 6890 N gas chromatograph, equipped with a HP-5 capillary column (30m × ϕ0.32mm × 0.25 μm). The column temperature program was set at a starting temperature of 160 °C followed by an initial rate of 15 °C/min to 215 °C, held at 215 °C for 2 min, and then it was ramped at 0.5 °C/min to 220 °C, held at this temperature for 2 min, continue to ramp at 15°C/min to 270°C, held for 2min. Nitrogen was used as carrier gas. The inlet pressure was 43.4 kPa. The split ratio was 40:1. The injector and detector temperatures were set at 270 °C and 290 °C, respectively. The FAME content of the products was determined by external standard method and the standard sample was prepared using sulfuric acid catalysis method. 2.3. Experimental procedure As is depicted in Fig.1, in a typical experiment, 20g algae powder (40 mesh) mixed with certain amount glass bead were loaded in the extractor 14, separated into 4 layers by wire mesh and sealed. CO2 was pumped into extractor until a certain pressure was reached. Keep the pressure for 12 h so that CO2 and the algae can contact enough. Some 7

methanol was firstly pumped by high-pressure constant pump 18 to preheater 21 and reactor 24 and preheated to the required temperature. Then CO2, n-hexane and methanol was loaded, and the pressure and flow of CO2 showed by wet type gas flow meter were tuned by valve 26 and 28 so that the pressure in buffer tank 27 was around half the total. Started the timer when the system was stable and first drop of product was accumulated. A fixed bed reactor 24 using Dixon rings (Qiao et al., 2017) as packing was employed, which can be seen in our previously report. The products were vacuum distillated to remove excess methanol and n-hexane, and washed three times with some n-hexane and deionized water, the upper layer were obtained after static stratification, and centrifuged for 10min in 7000rpm. The upper clear liquid was reduced pressured distillated, the obtained mother liquor diluted by n-hexane was analyzed by gas chromatography. 3. Results and discussion 3.1. The optimal operation parameters of scCO2 extraction The extraction mechanism was provided by studying the extraction curve against the time with the condition: 20g chrysophyta in 40 mesh size, 60⁰C, 30MPa and 300min. As is shown in Fig.2, the extraction yield was increasing along with time, and the increasing tendency became gently after 240min. Microalgae possess a cell wall, which is a thick and rigid layer composed of complex carbohydrates and glycoproteins with high mechanical strength and chemical resistance (Kim et al., 2013). Due to the benign diffusivity, CO2 can get into and out cell wall easily. The extraction process was presumed as follows (Halim et al., 2012): CO2 penetrate cell wall across membrane into the cytoplasm, solubilize fatty acid and wrap them up; then the combined substance 8

pass through the membrane and cell wall into the bulk scCO2 phase due to the concentration gradient, and stay in the bulk scCO2. Mass of fatty acid= mass of crude oil× fatty acid content analyzed by GC (here, the crude oil mass means the oil extracted in an hour at that time point) Extraction yield =total mass of fatty acid extracted at that time point / (mass of algae×total content of fatty acid) Total content of fatty acid was measured by in-situ transesterification, which for Chrysophyta is 10.5%, and Chlorella sp is 19.5%. Since the extraction yields were no longer growing after 240min, the extraction time was set as 240min. The scCO2 extraction for chrysophyta was examined firstly to obtain the optimal operation parameters for the coupling continuous process. Chrysophyta was smashed, and three kinds of grain sizes 10, 20 and 40 mesh raw material were chosen to investigate the effect of feedstock’s granularity to the extraction. Results showed that on the condition of 20g algae, 60 °C, 20 MPa, flow of CO2 2L/min, the extraction yield was 47.7%, 56.1% and 63.5% respectively according to 10, 20 and 40 mesh size of algae. This phenomenon verified that the mass transfer can be reinforced through smashing the feedstock containing fatty acid, and the cell wall maybe crashed in the crushing, which is beneficial for CO2 entering and exiting. The smaller the feedstock is, the more easily the extraction agent passed through the cell, but the more tightly the feedstock is stacked, attendant to the increasing resistance, the harder that CO2 can permit into the feed , and the less contact the CO2 and feed is. Keep increasing the the mesh size, the residue after extraction was mottled, which means the 9

poor contact between extraction agent and algae lead to the incomplete extraction. So 40 mesh was selected as the optimal size after comprehensive investigation. The impacts of temperature, pressure as well as flow of the co-solvent and CO2 on the extraction are investigated respectively. Table 1 presents the influence of temperature on the fatty acid weight extracted per hour. Other parameters were set as 20 MPa、flow of CO2 2 L/min、20g algae with 40 mesh size. From 40⁰C to 60⁰C, the mass of fatty acid per hour varied directly with the temperature, but further increasing, as at 70⁰C, the oil extracted was lower than that of 50 and 60⁰C. Vapor pressure of the solute increases with the increase of temperature, enhancing the diffusion in scCO2, beneficial for extraction. Whereas the density of CO2 decreases along with the increasing temperature, negative for its dissolving capacity. Higher than 60⁰C, the decrease of the CO2 density dominates over the increase of the solute vapor pressure. As is shown in Table 2, the pressure has a great influence on the extraction yield. Enlarged the pressure from 20MPa to 25MPa and extracted for 120 minutes, the mass of extracted fatty acid were magnified about 0.2g, from 0.93g to 1.13g, but further increased, the amount was only magnified 0.13g. As time went by, the difference became smaller and smaller. High pressure leads to a big density of scCO2, which is good for the effect of extraction, but increases the energy consumption. Moreover, due to the limit of the transesterification reactor, the pressure in the after coupling process was set as 20MPa. Adding n-hexane as co-solvent is beneficial for oil extracting (Da Silva et al., 2014), since it is a non-polar solvent and had been widely used in the extraction of oil, 10

and previous studies reported that n-hexane can improve the yield of biodiesel(Rojas et al., 2014; Xu et al., 2015) . The larger the flow of n-hexane is mixed in, the larger amount of oil can be obtained. N-hexane improved the extraction of scCO2 because of the solvation association between it and the solute. In the first hour, listed in Table 3, the mass of extracted oil was 0.5, 0.59, 0.68 and 0.77g with the hexane flow of 0, 0.1, 0.2 and 0.4 ml/min, at the end of the process, the total mass of extracted fatty acid in the 4 hours were 1.33, 1.52, 1.65 and 1.83g in turns. For the flow of 0.4 ml/min, the final extraction yield was 87.15%, which is larger than others, so 0.4ml/min was selected. It is important to choose an optional flow of CO2. In the first hour, which was demonstrated in Table 4, the extraction oil was soaring from 0.43g to 0.68g, with the flow of CO2 from 0.5L/min to 1L/min, finally amplified to 0.77g in the flow of 2L/min. At the end of the extraction(extracted 4 hours), the total extraction oil mass were 1.46g, 1.70g and 1.83g in turns with CO2 flow of 0.5L/min, 1L/min and 2L/min. Therefore, in the module of extraction, high flow of CO2 leads to a good extraction yield. 3.2 Transesterification coupling with supercritical extraction 3.2.1The influence of time It can be concluded from the algae extraction experiments that the extraction rate increased firstly then decreased, accordingly, the FAME yields were constrained as time extended. Fig.3 illustrated that the FAME yield and its amount per unit time changed over time. In the first 2h, the yield and the mass of FAME grew along with time, and reached a top spot at 2h. After that, the FAME amount decreased as the extracted algae oil reduced, and the change of yields became modest. That’s because the 11

transesterification was subjected to the feed oil’s extraction step, the oil content in the microalgae was decreased as the extraction went on. So the feed oil amount extracted from the soybean was decrease, dragged the reaction rate, and then the extraction progress became the rate determining step. In the industrial scale, this problem can be solved through an appropriate process make the optimal extraction period and transesterification matched appropriately. Specifically, three parallel extractors are set to switch in order to implement the quasi-continuous production. 3.2.2 The influence of CO2 flow Flow of CO2 influenced the extraction, as well as the residence time of the reactants. For the fixed reactor, the residence time was 2.28 min、1.14 min and 0.57 min corresponding to the flow of 0.5 L/min, 1 L/min and 2 L/min. The residence time must be longer than 4 min at least, otherwise the yield of FAME is too low (Saka and Kusdiana, 2001). As was illustrated in Fig.4, the extraction and reaction temperatures are 60 and 300ºC, n-hexane 0.4 mL/min, 18-20 MPa, molar ratio of methanol to oil 84:1, the FAME yield decreased as CO2 flow grew, but the amount of FAME was irregular (Table 5). Before 120min, the FAME amount of 1L/min CO2 was the highest, but after that point of 120min, the FAME amount of 0.5L/min CO2 became the highest, 1L/min CO2 went to the second place. Since the FAME is the target product and its yield is the key factor that should be considered, the flow of CO2 was set as 0.5L/min in the following coupling process. 3.2.3 The influence of temperature Four reaction temperatures 250, 300, 340 and 360⁰C were examined for the 12

optional temperature’s selection. The other condition were: extraction temperature 60⁰C,n-hexane flow 0.4 mL/min, pressure 18-20MPa, the molar ration of methanol to oil 84:1 and CO2 flow 0.5 L/min. The results showed that the yield at 250 ⁰C was 19.31% in 240 min, and then increased to 36.56% at 300 ⁰C, and 54.56% at 340 ⁰C. As was illuminated in Fig. 5, further increased temperature to 360 ⁰C, the yield decreased to 44.14%. That’s because the FAME was not thermal stable and hydrolysis when the temperature higher than 350⁰C (Imahara et al., 2008),That phenomenon was consistent to the results of batch(Yin et al., 2008) and continuous production(Zhou et al., 2017) for soybean oil. Higher temperature raised energy cost, elevated the risk of methyl ester pyrolysis and the coke may form in that situation, which will block the tube. 3.2.4 The influence of methanol to oil ratio. The transesterification reaction is reversible,elevating the concentration of the reactants can prompt the reaction to the forward direction. Usually, raising the molar ratio of methanol to oil was employed. The results were obtained and shown in Table 6 in the condition: Extraction temperature 60ºC, reaction 340ºC, CO2 0.5 L/min, n-hexane 0.4 mL/min, 18-20 MPa. The yields of FAME increased firstly then reduced as the molar ration grew. When the molar ratio is 21:1, the yield was lower than 10%, and it was added to 56.31% at 120min till the ratio of 84:1, finally, the yield reduced a little at the ratio of 126:1. Although the higher molar ratio of methanol to oil is beneficial for the transesterification, the excess methanol would dilute the reactant too much when it exceeded to a certain extent. Our studies about soybean oil showed that the best ratio of 13

methanol to oil was 42:1(Yin et al., 2012), and this paper was 84:1, twice as the former reports. Convert to dry algae to methanol (wt./vol.) ratio which is a parameter usually used for algae, this value is 1:4. That’s because the algae’s fatty acid content was only 10.5%, much lower than soybean, and the extraction amount per unit time was too little, along with the massive CO2 full of the reactor, led to a poor contact between the oil of methanol, so the methanol must be elevated to improve the touch. 3.2.5 The influence of n-hexane N-hexane can increase the mutual solubility of the oil and alcohol at lower reaction temperatures, as well as accelerated the reaction rate at supercritical conditions (Friedrich, 1982). But it has been proved that this intensification only occurred in certain thermodynamic conditions (temperature above 300°C), and the effect is limited. In this work, it was found that at the temperature of 340⁰C, coupling pressure of 18-20 MPa, CO2 flow rate of 0.5 L/min, without adding n-hexane yielded 21.62% of FAME after 120 min. As was depicted in Fig. 6, increased the flow rate of n-hexane from 0 to 0.1, 0.2, 0.4 ml/min, with the yield increased to 30.81%, 45.84%, and 56.31%. Further increase the flow rate to 0.6 ml/min, the yield had a little decrease. It confirmed that a certain amount of n-hexane had a positive effect on the transesterification reaction, but an excess amount performed a negative effect on the production of FAME. For one thing, adding n-hexane as entrainer can increase extraction speed of feed oil, which promoted the reaction towards the direction of transesterification, but after a certain time, the oil extracted on high flow rate shorten the residence time of the reaction medium, resulted in an insufficient contact; for another, n-hexane’s promotion for the 14

transesterification existed an optimum amount, once exceeded, the effect was not significant. Combined with the extraction, the flow rate of n-hexane was set as 0.4 ml/min. 3.3 Chlorella sp After the systematic research about the biodiesel production with Chrysophyta through supercritical methanol method coupling with scCO2 extraction, to further develop the scope of microalgae for biodiesel, another kind of algae was selected as the raw material to prepare biodiesel in the same process with the optimal operation conditions above, for extraction which are: 20g microalgae in 40 mesh, temperature 60⁰C, 18-20MPa, flow of n-hexane 0.4ml/min, flow of CO2 0.5L/min, and for reaction: 340⁰C, pressure: 18-20MPa, molar ratio of methanol to oil 84 : 1. In these conditions, a biodiesel yield of 63.78% at 120min was obtained, when flow of CO2 was changed from 0.5 L/min to 1 L/min and 2 L/min, the yield decreased to 46.74% and 38.08%. That 63.78% was larger than the highest yield for Chrysophyta, which is 56.31%. That’s because the oil content of Chlorella sp is 19.8%, much more than that of Chrysophyta’s oil content 10.5%. It is thus clear that it is possible to use the oil containing microalgae to produce biodiesel with the continue process coupling with scCO2 extraction, and higher oil content species should be chosen as far as possible. From the perspective of feedstock integrated utilization, residue after extracting can be reused to produce highvalue added products such as medicine and health products, because other gradients such as complex carbohydrates, vitamins, minerals, and trace elements which are different from lipids, they are not damaged in a modest extraction temperature, and can 15

be separated completely and easily from the solvent CO2 through decompression, So the residue after supercritical CO2 extraction is safe for the other utilization in Pharmaceuticals and Nutraceuticals. 4. Conclusion A continuous coupling process of supercritical extraction and non-catalytic supercritical methanol transesterification method for preparing biodiesel was put forward. Using scCO2 as the extraction solvent avoiding the damage of high-value compositions in high temperature, and the residue can be recycled. The optimum reaction condition was that 40 mesh algae, the extraction and reaction temperature 60⁰C and 340⁰C, 18-20 MPa, molar ration of methanol to oil 84:1 and the flow rate of CO2 and n-hexane were 0.5 L/min and 0.4 mL/min, respectively. In these circumstances, a yield of 56.31% was obtained for Chrysophyta, while 63.78% for Chlorella sp. Acknowledgements The Author would like to thank the financial support from the National Natural Science Foundation of China (21376045, 21506027), Petrochemicals Joint Fund of National Natural Science Foundation of China and China National Petroleum Corporation (U1662130), Chinese Postdoctoral Science Foundation (2015M571307) and the Open Project Program of State Key Laboratory of Catalysis (Dalian Institute of Chemical Physics, N-15-01). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version.

References 16

1. Andrich, G., Nesti, U., Venturi, F., Zinnai, A., Fiorentini, R., 2005. Supercritical fluid extraction of bioactive lipids from the microalga Nannochloropsis sp. Eur. J. Lipid Sci. Technol.107. 381--386. 2. Chauton, M.S., Reitan, K.I., Norsker, N.H., Tveterås, R., Kleivdal, H.T., 2015. A techno-economic analysis of industrial production of marine microalgae as a source of EPA and DHA-rich raw material for aquafeed: Research challenges and possibilities. Aquaculture, 436, 95-103. 3. Cheng, C., Du, T., Pi, H., Jang, S., Lin, Y., Lee, H., 2011. Comparative study of lipid extraction from microalgae by organic solvent and supercritical CO2. Bioresource Technol., 102, 10151-10153. 4. Da Silva, F.M., Pinho, D.M.M., Houg, G.P., Reis, I.B.A., Kawamura, M., Quemel, M.S.R., Montes, P.R., Suarez, P.A.Z., 2014. Continuous biodiesel production using a fixed-bed Lewis-based catalytic system. Chem. Eng. Res. Des., 92, 1463-1469. 5. Friedrich, J.P.A.L., 1982. Petroleum-free extraction of oil from soybeans with supercritical CO2. J. Am. Oil Chem. Soc., 59, 288--292. 6. Galadima, A., Muraza, O., 2014. Biodiesel production from algae by using heterogeneous catalysts: A critical review. Energy, 78, 72-83. 7. Halim, R., Danquah, M.K., Webley, P.A., 2012. Extraction of oil from microalgae for biodiesel production: A review. Biotechnol. Adv., 30, 709-732. 8. He, H., Wang, T., Zhu, S., 2007. Continuous production of biodiesel fuel from vegetable oil using supercritical methanol process. Fuel, 86, 442-447. 9. Imahara, H., Minami, E., Hari, S., Saka, S., 2008. Thermal stability of biodiesel in supercritical methanol. Fuel, 87, 1-6. 10. Jazzar, S., Olivares-Carrillo, P., Pérez De Los Ríos, A., Marzouki, M.N., Acién-Fernández, F.G., Fernández-Sevilla, J.M., Molina-Grima, E., Smaali, I., Quesada-Medina, J., 2015. Direct supercritical methanolysis of wet and dry unwashed marine microalgae (Nannochloropsis gaditana) to biodiesel. Appl. Energ., 148, 210-219. 11. Karmakar, A., Karmakar, S., Mukherjee, S., 2010. Properties of various plants and animals feedstocks for biodiesel production. Bioresource Technol., 101, 7201-7210. 12. Kim, J., Yoo, G., Lee, H., Lim, J., Kim, K., Kim, C.W., Park, M.S., Yang, J., 2013. Methods of downstream processing for the production of biodiesel from microalgae. Biotechnol. Adv., 31, 862-876. 13. Liu, J., Liu, Y., Wang, H., Xue, S., 2015. Direct transesterification of fresh microalgal cells. Bioresource Technol., 176, 284-287. 14. Millao, S., Uquiche, E., 2016. Extraction of oil and carotenoids from pelletized microalgae using supercritical carbon dioxide. J. Supercrit. Fluids, 116, 223-231. 15. Nan, Y., Liu, J., Lin, R., Tavlarides, L.L., 2015. Production of biodiesel from microalgae oil (Chlorella protothecoides) by non-catalytic transesterification in supercritical methanol and ethanol: Process optimization. The Journal of Supercritical Fluids, 97, 174-182. 16. Patil, P.D., Gude, V.G., Mannarswamy, A., Deng, S., Cooke, P., Munson-McGee, S., Rhodes, I., Lammers, P., Nirmalakhandan, N., 2011. Optimization of direct conversion of wet algae to biodiesel under supercritical methanol conditions. Bioresource Technol., 102, 118-122. 17. Qiao, B., Zhou, D., Li, G., Yin, J., Xue, S., Liu, J., 2017. Process enhancement of supercritical methanol biodiesel production by packing beds. Bioresource Technol., 228, 298-304. 18. Reddy, H.K., Muppaneni, T., Patil, P.D., Ponnusamy, S., Cooke, P., Schaub, T., Deng, S., 2014. Direct conversion of wet algae to crude biodiesel under supercritical ethanol conditions. Fuel, 115, 720726. 19. Rojas, O.J., Gómez, L.M., López, J.G., Barrera-Zapata, R., Giraldo, S.A., 2014. Producción de 17

biodiesel a partir de aceite de palma en un reactor de lecho fijo con metóxido de calcio como catalizador. Ingeniería y competitividad, 16, 109-118. 20. Saka, S., Kusdiana, D., 2001. Biodiesel fuel from rapeseed oil as prepared in supercritical methanol. Fuel, 80, 225-231. 21. Solana, M., Rizza, C.S., Bertucco, A., 2014. Exploiting microalgae as a source of essential fatty acids by supercritical fluid extraction of lipids: Comparison between Scenedesmus obliquus, Chlorella protothecoides and Nannochloropsis salina. J. Supercrit. Fluids, 92, 311-318. 22. Taher, H., Al-Zuhair, S., Al-Marzouqi, A.H., Haik, Y., Farid, M., Tariq, S., 2014. Supercritical carbon dioxide extraction of microalgae lipid: Process optimization and laboratory scale-up. J. Supercrit. Fluids, 86, 57-66. 23. Xu, W., Gao, L., Xiao, G., 2015. Biodiesel production optimization using monolithic catalyst in a fixed-bed membrane reactor. Fuel, 159, 484-490. 24. Yin, J., Ma, Z., Shang, Z., Hu, D., Xiu, Z., 2012. Biodiesel production from soybean oil transesterification in subcritical methanol with K3PO4 as a catalyst. Fuel, 93, 284-287. 25. Yin, J., Xiao, M., Wang, A., Xiu, Z., 2008. Synthesis of biodiesel from soybean oil by coupling catalysis with subcritical methanol. Energ. Convers. Manage., 49, 3512-3516. 26. ZHOU, C., WANG, C., WANG, W., WU, Y., YU, F., CHI, R., ZHANG, J., 2010. Continuous Production of Biodiesel from Soybean Oil Using Supercritical Methanol in a Vertical Tubular Reactor: I. Phase Holdup and Distribution of Intermediate Product along the Axial Direction. Chinese J. Chem. Eng., 18, 626-629. 27. Zhou, D., Qi, L., Qiao, B., Xu, Q., Yin, J., 2017. Continuous production of biodiesel from soybean flakes by extraction coupling with transesterification under supercritical conditions: Original research article. J. Supercrit. Fluids, 120, 395-402.

18

List of figure captions

1) CO2 cylinder;2) reducing valve;3) filter;4) cold trap;5), 8), 10), 16), 19), 26), 28) needle valve; 6) plunger pump;7) check valve;9) 22) mixer;11) 18) HPLC pump;12) n-hexane container;13) CO2 preheater coiler;14) extractor;15) water bath;17) methanol container;20), 25) temperature control unit;21) preheater coiler;23) furnace;24) reactor;27) buffer tank;29) collector;30) wet type flow meter; T: thermocouple; P:pressure gauge Fig. 1 Experiment apparatus for supercritical extraction coupling with supercritical transesterification Fig.2 Microalgae oil extraction curve by scCO2 Fig. 3 The influence of operation time on the yield and the amount per unit time of FAME (Extraction 60 ºC , reaction 340 ºC, CO2 0.5 L/min, n-hexane 0.4 mL/min, 18-20 MPa, 84:1 ) Fig. 4 The influence of CO2 flow rate on the yield of FAME in the fixed bed reactor; Fig. 5 The influence of reaction temperature on the yield of FAME in the fixed bed reactor Fig. 6 The influence of the flow rate of n-hexane on the yield of FAME in the fixed bed reactor (The extraction and reaction temperatures are 60 and 340 ºC, 84:1, CO2 0.5 L/min, 18-20 MPa)

19

Tables

Table 1 Mass of fatty acid extracted in an hour at different temperatures Mass of fatty acid (g) time(min) 40⁰C 50⁰C 60⁰C 60 0.3310 0.4040 0.5040 120 0.3251 0.3520 0.4221 180 0.2340 0.2460 0.2770 240 0.1330 0.1819 0.1302

70⁰C 0.3331 0.3570 0.2411 0.1892

Table 2 Mass of fatty acid extracted in an hour in different pressures Mass of fatty acid (g) time 20MPa 25MPa 30MPa 60 0.5040 0.6130 0.6901 120 0.4221 0.5141 0.5731 180 0.2770 0.3030 0.2760 240 0.1302 0.0920 0.1180

Table 3 Mass of fatty acid extracted in an hour on different n-hexane flows Mass of fatty acid (g) time(min) 0 mL/min 0.1 mL/min 0.2 mL/min 0.4 mL/min 60 0.5040 0.5920 0.6789 0.7661 120 0.4221 0.4240 0.5290 0.7491 180 0.2770 0.2990 0.2589 0.2480 240 0.1302 0.2090 0.1789 0.0670

Table 4 Mass of fatty acid extracted in an hour on different CO2 flows Mass of fatty acid (g) time(min) 0.5L/min 1 L/min 2 L/min 60 0.4330 0.6794 0.7661 120 0.4830 0.5708 0.7489 180 0.3990 0.3005 0.2480 240 0.1451 0.1455 0.0672

20

Table 5 Mass of FAME in an hour on different CO2 flows Mass of fatty acid (g) time(min) 0.5L/min 1 L/min 2 L/min 60 0.1342 0.1333 0.0834 120 0.1483 0.1648 0.1108 180 0.1266 0.0996 0.0518 240 0.0702 0.0401 0.0225

Table 6 Yield of FAME at different molar ratio of methanol to oil. (40 mesh, the extraction and reaction temperatures are 60 and 340 ˚C, the flow rate of n-hexane and CO2 are 0.4 mL/min and 0.5 L/min, the pressure is 18-20 MPa)

Time(min) 60 120 180 240

21 6.35 8.69 7.31 8.29

Yield (%) 42 84 19.5 42.79 29.05 56.31 30.21 53.87 26.96 54.56

21

126 48.98 52.69 46.74 48.47

1) CO2 cylinder;2) reducing valve;3) filter;4) cold trap;5), 8), 10), 16), 19), 26), 28) needle valve; 6) plunger pump;7) check valve;9) 22) mixer;11) 18) HPLC pump;12) n-hexane container;13) CO2 preheater coiler;14) extractor;15) water bath;17) methanol container;20), 25) temperature control unit;21) preheater coiler;23) furnace;24) reactor;27) buffer tank;29) collector;30) wet type flow meter; T: thermocouple; P:pressure gauge Fig. 1 Experiment apparatus for supercritical extraction coupling with supercritical transesterification

22

Extraction Yield(%)

90

60

30

0

0

60

120 180 Time(min)

240

300

Fig.2 Microalgae oil extraction curve by scCO2

23

Yield (%)

0.3

40

0.2

20

0.1 Yield FAME amount

0

0

60

120 Time (min)

 FAME Amount (g/h)

60

0.0 240

180

Fig.3 The influence of operation time on the yield and the amount per unit time of FAME (Extraction 60 ºC, reaction 340 ºC, CO2 0.5 L/min, n-hexane 0.4 mL/min, 18-20 MPa, 84:1 )

24

40

(a)

Yield (%)

30 20 0.5L/min 1 L/min 2 L/min

10 0

0

60

120

180

240

Time(min)

Fig. 4 The influence of CO2 flow rate on the yield of FAME in the fixed bed reactor;

25

60

Yield (%)

45 30 15 0

250C 340C 0

60

120 Time (min)

300C 360C 180

240

Fig. 5 The influence of reaction temperature on the yield of FAME in the fixed bed reactor

26

60

Yield (%)

45 30 15 0

0 mL/min 0.2 mL/min 0.6 mL/min

0

60

120 Time (min)

0.1 mL/min 0.4 mL/min

180

240

Fig.6 The influence of the flow rate of n-hexane on the yield of FAME in the fixed bed reactor (The extraction and reaction temperatures are 60 and 340 ºC, 84:1, CO2 0.5 L/min, 18-20 MPa)

27

Extraction

Reaction

Chrysophyta

Chlorella sp

scCO2

28

Highlights

1. Two kinds of microalgae Chrysophyta and Chlorella sp were selected as feedstock. 2. A continuous process of transesterification coupling scCO2 extraction was conducted. 3. No catalyst loaded and residue after extraction can be reused for health products.

29