Energy Reports 5 (2019) 37–40

Contents lists available at ScienceDirect

Energy Reports
journal homepage: www.elsevier.com/locate/egyr

Research paper

Simultaneous extraction–reaction process for biodiesel production

from microalgae
∗
Reem Shomal, Hiyam Hisham, Amal Mlhem, Rawan Hassan, Sulaiman Al-Zuhair
Chemical Engineering Department, UAE University, 15551 Al-Ain, United Arab Emirates

highlights graphical abstract

• Supercritical CO2 (SC-CO2 ) was used

as an extraction solvent and reaction
medium.
• Immobilized lipase was used as the
catalyst.
• Effects of temperature, reaction time,
and methanol:oil molar ratio were
studied.
• A biodiesel production yield of 19.3%
was obtained.

article info a b s t r a c t

Article history: Biodiesel is a promising sustainable alternative to non-renewable petrodiesel. In this work, oils extracted
Received 20 August 2018 from microalgae are used for biodiesel production. To simplify the process, the extraction of the oils and
Received in revised form 2 November 2018 their reaction are made to take place simultaneously in one step. Immobilized lipase was used as the
Accepted 15 November 2018
catalyst, and supercritical CO2 (SC-CO2 ) was used as an extraction solvent and reaction medium. The use of
Available online 28 November 2018
SC-CO2 allows easy separation of the products and leaves the leftover biomass uncontaminated, allowing
Keywords: it to be utilized in food and pharmaceutical applications. The effects of temperature (35–50 ◦ C), reaction
Biodiesel time (2-6 h), and methanol:oil (M:O) molar ratio (8:1–16:1) on biodiesel yield were investigated. Within
Microalgae 6 h, the maximum biodiesel production yield was found to be 19.3% at a temperature of 35 ◦ C and at an
Supercritical CO2 M:O molar ratio of 8:1. The results hold promise in simplifying the microalgae-to-biodiesel production
Extraction
process.
Transesterification
© 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction is renewable, non-toxic, biodegradable, and does not contribute to

the net accumulation of greenhouse gases (Fjerbaek et al., 2009;
Finding alternatives to conventional fossil fuels is becoming Al-Zuhair et al., 2007; Mekhilef et al., 2011). Microalgae have
increasingly important, because of their unsuitability and harmful received significant attention as a potential oil source, owing to
environmental impact. Among the promising sustainable alterna- their high oil content and growth rates (Lu et al., 2009). Cultivating
tives is biodiesel, the physical and chemical properties of which are microalgae cells does not require land development on freshwater,
similar to those of petrodiesel, allowing it to be used in conven- as several strains of algae have been found to grow in seawater and
tional engines without requiring any modifications. In addition, it wastewater.
The conventional technique used for oil extraction from mi-
∗ Corresponding author. croalgae is the organic solvent extraction method (Mercer and
E-mail address: s.alzuhair@uaeu.ac.ae (S. Al-Zuhair). Armenta, 2011). However, when an organic solvent is used, an

https://doi.org/10.1016/j.egyr.2018.11.003
2352-4847/© 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-
nd/4.0/).
38 R. Shomal, H. Hisham, A. Mlhem et al. / Energy Reports 5 (2019) 37–40

additional downstream unit is required to separate the solvent To reduce the overall cost associated with the microalgae-to-
from the products, resulting in an additional production cost. biodiesel conversion process and to avoid the complexity associ-
Moreover, most of these solvents are toxic and volatile, posing ated with the separate extraction of lipids, followed by the trans-
environmental problems and limiting further utilization of the left- esterification of the extracts, a simultaneous extraction–reaction
over biomass. Once the oil is extracted, it is converted to biodiesel process (SERP) has been proposed (Qian et al., 2008; Skorupskaite
through a catalytic reaction with methanol. The use of lipase et al., 2016; Ehimen et al., 2010). The SERP was tested on Chlorella
enzymes has been receiving increased attention, owing to their sp. microalgae using sulfuric acid as the catalyst (Wahlen et al.,
capacity to produce biodiesel from different sources of oils, with 2011), and an optimum conversion rate of 91% was achieved after
easy product separation and without any pretreatment or soap 8 h of reaction at 60 ◦ C. Sulfuric acid was also used as the cata-
formation. However, the enzyme should be used in an immobilized lyst when the SERP was tested on other microalgae strains such
form to overcome the high cost of the enzyme (Samukawa et al., as Synechocystis sp. PCC 6803, Synechococcus elongatus, Chlorella
2000; Adamczak and Krishna, 2004). In this case, i.e., when an sorokiniana, Tetraselmis suecica, and Chaetoceros gracilis (de Luna
immobilized lipase is used, an organic solvent needs to be added to et al., 2017). However, the use of an acidic catalyst can be highly
increase the solubility of the reaction substrates. This reduces the corrosive and is not recommended for fuel production. A base
inhibition effect and lowers the viscosity and transport limitations catalyst, i.e., LiOH-pumice, has been recently used in the SERP
of the reaction mixture (Yang et al., 2010; Rathore and Madras, of Chlorella sp. microalgae (Shirazi et al., 2017). The highest pro-
2007). Among the many organic solvents tested, n-hexane is by duction yield was 47%, obtained after a reaction time of 3 h at
far the most commonly used. The transesterification yield of tallow 80 ◦ C. Near supercritical methanol has recently been tested for
fats was found to be 94.8% using Mucor miehei lipase, obtained after the SERP of Spirulina microalgae (Eggers and Schwudke, 2016).
5 h with n-hexane as the solvent, whereas in a solvent-free system, The optimum conversion rate was achieved at an extremely high
the yield did not exceed 20% (Nelson et al., 1996). As mentioned temperature of 300 ◦ C and a pressure of 120 bar. The use of such
earlier, most organic solvents are toxic and require an additional an energy intensive process may not be economical. Moreover,
separation unit. when the temperature was 200 ◦ C, a significant drop in the yield
Supercritical carbon dioxide (SC-CO2 ) has been suggested to was observed, with a conversion rate of only 1.78%. Owing to the
replace conventional organic solvents for both oil extraction and advantages of enzymes over chemical catalysts, it was interesting
reaction processes. CO2 is a non-toxic and cheap fluid with mod- to study the SERP for the enzymatic biodiesel production from
erate critical parameters (Celia et al., 2005; del Valle et al., 2004). microalgae using SC-CO2 , which is the focus of this work.
Moreover, the products can be easily separated via a simple de-
pressurization process. SC-CO2 has been used for enhanced oil 2. Materials and methods
extraction from several sources such as oil crops (del Valle et al.,
2004; Reverchon and Marrone, 2001) and microalgae cells (An-
2.1. Strain, Chemicals, and Enzyme
drich et al., 2005; Halim et al., 2011). The lipid extraction yield
from Scenedesmus sp. microalgae increased by 20% using SC-CO2
A dried biomass of Scenedesmus sp. microalgae was provided
at a temperature of 50 ◦ C and a pressure of 350 bar, as compared
by AlgaOil Limited, Philippines. The biomass was cultivated in
to that obtained using n-hexane (Taher et al., 2014a). This is mainly
an organic fertilizer (NPK, grade 14-14-14) and then sun dried.
because of the higher diffusivity and lower viscosity of SC-CO2 ,
An analytical grade methanol with a purity ≥ 99% was obtained
which allows it to penetrate easier within the biomass matrix.
from Fisher chemicals, USA. Novozym⃝ R
435, with an activity of
Higher extraction yields and rates have been reported when SC-
11,900 PLU g−1 , was provided by Novozymes, Denmark. n-Hexane
CO2 was used to extract lipids from Nannochloropsis sp. Andrich
(purity of 96%) was obtained from DAEJUNG Co., Korea. A standard
et al. (2005) and Chlorococum sp. Halim et al. (2011).
solution of high purity fatty acid methyl esters (FAME) containing
Because of the low critical temperature of CO2 (31 ◦ C), which is
4% myristic acid (C14:0), 10% palmitic acid (C16:0), 6% stearic acid
below the denaturation temperature of most proteins, lipases can
maintain their activity in SC-CO2 . The moderate critical pressure (C18:0), 35% oleic acid (C18:1), 36% linoleic acid (C18:2), 2% arachi-
of CO2 (73 bar) has minimal effect on enzyme activity (Celia et al., donic acid (C20:0), and behenic acid (C22:0) was obtained from
2005; Novak et al., 2003). Moreover, the diffusion of the reaction Sigma–Aldrich, USA. Ultra-pure air was supplied by Abu-Dhabi
substrates and products to and from the enzyme active sites is Oxygen Company, UAE, and high-purity helium was supplied by
enhanced with the use of SC-CO2 , which overcomes the main Air Product Company, UAE.
drawback of immobilized enzymes. In the transesterification of
oil extracted from palm kernel in the presence of Novozym⃝ R
435, 2.2. Experimental method
an optimum conversion rate of 63.2% was obtained using SC-CO2 .
However, even with double the amount of enzyme, the maxi- The harvested microalgae cells were lyophilized in a freeze drier
mum conversion rate was only 58.3% using n-hexane (Oliveira and operated at −80 ◦ C and 0.01 mbar for 6 h. The lyophilized cells
Oliveira, 2001). were then grinded for a short period of 15 s. Particles with sizes
Although employing a high-pressure SC-CO2 process can be jus- between 150 and 355 µm, determined using sieve analysis, were
tified for oil extraction, this may not be the case for fuel production, used. The total oil content was found to be 5.8 ± 0.16% using
despite the enhanced reaction yield and ease of product separa- the Folch method (Eggers and Schwudke, 2016) with a chloro-
tion. Therefore, an integrated process was proposed, in which the form:methanol solvent mixture of 2:1 (v:v).
extracted oils already dissolved in SC-CO2 are directly fed into a The experimental setup comprises a CO2 cylinder, a CO2 high-
bioreactor packed with immobilized lipase, to produce biodiesel pressure syringe pump with a maximum capacity of 500 bar
without the need for further pumping. Thus, the advantages of (Model 260D, ISCO, USA), a pump controller (ISCO, SFX 200, USA),
performing the reaction in an SC-CO2 medium are maintained a high-pressure stainless steel reaction cell with a volume of
while avoiding the additional cost of pumping. Such an integrated 10 ml, and a temperature-controlled incubator with a maximum
system was tested for the extraction-reaction of fats from lamb temperature of 150 ◦ C (ISCO, SFX 220). The pressure within the
meat (Al-Zuhair et al., 2012) and from microalgae lipids (Taher chamber was measured and controlled by the system, whereas
et al., 2014b). In these studies, the extraction and reaction steps the temperature was measured and controlled in the incubator.
were carried out in separate units. The precision of the temperature measurements of the extraction
 R. Shomal, H. Hisham, A. Mlhem et al. / Energy Reports 5 (2019) 37–40 39

Fig. 1. Schematic of the supercritical CO2 extraction–reaction system.

system was ±0.1 ◦ C. Fig. 1 shows the schematic of the supercritical

fluid extraction (SFE) apparatus used.
A sample of lyophilized cells (1 g) was placed along with 2.7%
w/w enzyme loading and a pre-specified amount of methanol in
the reaction cell, which was covered by two 5/8 filters. Glass wools
were placed at the top and bottom of the sample to prevent particle
carryover. The cell was tightly sealed and heated up to the desired
temperature. SC-CO2 was then passed from the CO2 cylinder into
the high-pressure syringe pump and was pressurized to the desired
pressure. Once the desired pressure was reached, the reaction cell
was filled with SC-CO2 . At this point, the reaction starts. After a
specified reaction time, the products dissolved in SC-CO2 were
eluted by depressurizing the cell. The reaction products were di-
luted in 10 ml of n-hexane and taken for analysis to determine
the FAME content using gas chromatography (GC). The experiment
was carried out at a pressure of 400 bar, at different temperatures
(35, 40, and 50 ◦ C), and at different methanol:oil (M:O) ratios (8:1,
12:1, and 16:1). As the rate-determining step in this work was Fig. 2. Effects of reaction time on the FAME yield at 400 bar, 35 ◦ C, 2.7% w/w enzyme
loading, and different methanol:oil (M:O) molar ratios.
found to be the lipid extraction process, a pressure of 400 bar
was used. This is because it was shown that the extraction yield
from the same strain of microalgae used in another study increased
with the increase in the pressure of SC-CO2 (Taher et al., 2014a). using the same enzyme, oil extracted from the same microalgae
This is mainly because of the increase in the density, which has a biomass, at the same temperature but at a lower pressure of 200
positive effect on the solubility. At the same time, the pressure was bar, with a similar enzyme loading, and for the same M:O molar
not excessively high, thus avoiding any possible negative effect on ratio, a much better yield of 62% was obtained after 4 h (Taher
the enzyme. The quantity of FAME produced was represented as et al., 2014b). This is expected, as the oil in this case was readily
a percentage of the total oil content, determined using the Folch available for the reaction, whereas in the current study, the two
method as expressed in Eq. (1). processes take place simultaneously. This also suggests that the
lipid extraction is the rate-determining step.
mFAME
FAME yield = × 100% (1) The experiment was then repeated using a higher M:O molar
moil content ratio of 12:1 and tested at reaction times of 4 and 6 h and at the
where mFAME and moil content are the weights of the FAME produced same temperature and pressure. As shown in Fig. 2, the yield de-
and the oil in the biomass used, respectively. The experiments were creases with the increase in the M:O molar ratio at both the tested
carried out in duplicate, and the presented results are the average times, mainly because of the inhibition of methanol. This result is
values (with the standard deviation shown in the form of error bars in good agreement with that of the previous study conducted on
in the figures). the extracted oil, wherein the optimum yield was obtained at an
M:O molar ratio of 9:1 (Taher et al., 2014b). It noteworthy that the
3. Results and discussions error bars increase at 6 h. Nevertheless, the standard deviations are
relatively low. The most important takeaway is the successful use
The FAME production yield was determined after 2, 4, and 6 of the SC-CO2 system for a simultaneous extraction–reaction.
h of reaction at 400 bar, for a 2.7% w/w enzyme loading, at a To test the effect of temperature, the experiment was repeated
temperature of 35 ◦ C, and at an M:O molar ratio of 8:1. As shown at different temperatures but at the same pressure, i.e., 400 bar,
in Fig. 2, the highest production yield is 19.3% obtained after 6 h. and for an enzyme loading of 2.7%. The results, shown in Fig. 3,
Comparing the results of this work with previous results obtained indicate that the production yield decreases with the increase
40 R. Shomal, H. Hisham, A. Mlhem et al. / Energy Reports 5 (2019) 37–40

References

Adamczak, M., Krishna, S.H., 2004. Strategies for improving enzymes for efficient
biocatalysis. Food Technol. Biotechnol. 42 (4), 251–264.
Al-Zuhair, S., Hussein, A., Al-Marzouqi, A.H., Hashim, I., 2012. Continuous produc-
tion of biodiesel from fat extracted from lamb meat in supercritical CO2 media.
Biochem. Eng. J. 60, 106–110.
Al-Zuhair, S., Ling, F.W., Jun, L.S., 2007. Proposed kinetic mechanism of the produc-
tion of biodiesel from palm oil using lipase. Proc. Biochem. 42 (6), 951–960.
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 (6), 381–386.
Celia, E., Cernia, E., Palocci, C., Soro, S., Turchet, T., 2005. Tuning Pseudomonas
cepacea lipase (PCL) activity in supercritical fluids. J. Supercrit. Fluids 33 (2),
193–199.
Eggers, L.F., Schwudke, D., 2016. Liquid extraction: Folch. In: Wenk, M. (Ed.), Ency-
clopedia of Lipidomics. Springer, Dordrecht.
Ehimen, E.A., Sun, Z.F., Carrington, C.G., 2010. Variables affecting the in situ trans-
esterification of microalgae lipids. Fuel 89 (3), 677–684.
Fjerbaek, L., Christensen, K.V., Norddahl, B., 2009. A review of the current state of
Fig. 3. Effect of temperature on FAME yield after 4 h in SC-CO2 at 400 bar for 2.7% biodiesel production using enzymatic transesterification. Biotechnol. Bioeng.
w/w enzyme loading and different methanol: oil (M:O) molar ratios. 102 (5), 1298–1315.
Halim, R., Gladman, B., Danquah, M.K., Webley, P.A., 2011. Oil extraction from
microalgae for biodiesel production. Bioresour. Technol. 102 (1), 178–185.
Lu, J., Chen, Y., Wang, F., Tan, T., 2009. Effect of water on methanolysis of glycerol
in the temperature. However, when the extracted oil was used trioleate catalyzed by immobilized lipase Candida sp. 99–125 in organic solvent
instead of the biomass, the optimum temperature is 50 ◦ C (Taher system. J. Mol. Catal. B 56 (2–3), 122–125.
de Luna, M.D.G., Doliente, L.M.T., Ido, A.L., Chung, T.W., 2017. In situ transesterifi-
et al., 2014b). This difference is due to the pressure used in the
cation of Chlorella sp. microalgae using LiOH-pumice catalyst. J. Environ. Chem.
current simultaneous extraction–reaction system, i.e., 400 bar, Eng. 5 (3), 2830–2835.
whereas that used in the previous separate units was 200 bar. In Mekhilef, S., Siga, S., Saidur, R., 2011. A review on palm oil biodiesel as a source of
addition, the temperature in the current study affected not only renewable fuel. Renewable Sustainable Energy Rev. 15 (4), 1937–1949.
the reaction, as in the previous study, but also the solubility in SC- Mercer, P., Armenta, R.E., 2011. Developments in oil extraction from microalgae.
Eur. J. Lipid Sci. Technol. 113, 539–547.
CO2 . As mentioned earlier, the rate-determining step in the SERP
Nelson, L.A., Foglia, T.A., Marmer, W.N., 1996. Lipase-catalyzed production of
is the lipid extraction process. As the temperature increases, the biodiesel. J. Am. Oil Chem. Soc. 73 (9), 1191–1195.
solubility of the oils in SC-CO2 decreases because of the reduction Novak, Z., Habulin, M., Krmelj, V., Knez, Å.E., 2003. Silica aerogels as supports for
in the SC-CO2 density (Al-Zuhair et al., 2012). This in turn reduces lipase catalyzed esterifications at sub- and supercritical conditions. J. Supercrit.
the extraction rate. Fluids 27 (2), 169–178.
Oliveira, D., Oliveira, J.V., 2001. Enzymatic alcoholysis of palm kernel oil in n-hexane
The main challenge in this work was the low extraction rate, and SCCO2 . J. Supercrit. Fluids 19 (2), 141–148.
which affected the overall biodiesel production. This was mainly Qian, J., Wang, F., Liu, S., Yun, Z., 2008. In situ alkaline transesterification of cotton-
because of the low lipid content in the biomass used. Therefore, it seed oil for production of biodiesel and nontoxic cottonseed meal. Bioresour.
is recommended to extend this work by testing other microalgae Technol. 99, 9009–9012.
Rathore, V., Madras, G., 2007. Synthesis of biodiesel from edible and non-edible
biomasses having a higher lipid content for a more efficient extrac-
oils in supercritical alcohols and enzymatic synthesis in supercritical carbon
tion. dioxide. Fuel 86 (17–18), 2650–2659.
Reverchon, E., Marrone, C., 2001. Modeling and simulation of the supercritical
4. Conclusions CO2 extraction of vegetable oils. J. Supercrit. Fluids 19 (2), 161–175.
Samukawa, T., Kaieda, M., Matsumoto, T., Ban, K., Shimada, Y., Noda, H., et al.,
2000. Pretreatment of immobilized Candida antarcticalipase for biodiesel fuel
Biodiesel was directly produced from microalgae via a simulta- production from plant oil. J. Biosci. Bioeng. 90 (2), 180–183.
neous extraction–reaction process (SERP) using SC-CO2 . The effects Shirazi, H.M., Karimi-Sabet, J., Ghotbi, C., 2017. Biodiesel production from Spir-
of temperature, reaction time, and methanol:oil (M:O) molar ratio ulina microalgae feedstock using direct transesterification near supercritical
on the biodiesel yield were investigated. A biodiesel production methanol condition. Bioresour. Technol. 239, 378–386.
Skorupskaite, V., Makareviciene, V., Gumbyte, M., 2016. Opportunities for simul-
yield of 19.3% was obtained after a reaction time of 6 h, at a tem-
taneous oil extraction and transesterification during biodiesel fuel production
perature of 35 ◦ C, and at an M:O molar ratio of 8:1. The production from microalgae: A review. Fuel Proc. Technol. 150, 78–87.
yield was found to decrease with the increase in the temperature Taher, H., Al-Zuhair, S., Al-Marzouqi, A.H., Haik, Y., Farid, M., 2014a. Supercritical
and M:O molar ratio. Despite having a lower yield compared to carbon dioxide extraction of microalgae lipid: Process optimization and labora-
that obtained via separate extraction and reaction processes, the tory scale-up. J. Supercrit. Fluids 86, 57–66.
Taher, H., Al-Zuhair, S., Al-Marzouqi, A.H., Haik, Y., Farid, M., 2014b. Enzymatic
successful production of biodiesel via the simultaneous process is biodiesel production of microalgae lipids under supercritical carbon dioxide:
expected to simplify the overall biodiesel production process and process optimization and integration. Biochem. Eng. J. 90, 103–113.
make it more economical. del Valle, J.M., Rivera, O., Mattea, M., Ruetsch, L., Daghero, J., Flores, A.S., 2004.
Supercritical CO2 processing of pretreated rosehip seeds: effect of process scale
on oil extraction kinetics. J. Supercrit. Fluids 31 (2), 159–174.
Acknowledgment
Wahlen, B.D., Willis, R.M., Seefeldt, L.C., 2011. Biodiesel production by simultaneous
extraction and conversion of total lipids from microalgae, cyanobacteria, and
The work was funded by SURE Plus Fund, UAE University (Grant wild mixedcultures. Bioresour. Technol. 102 (3), 2724–2730.
No. : G00002455). Yang, Z., Zhang, K.P., Huang, Y., Wang, Z., 2010. Both hydrolytic and transesterifi-
cation activities of Penicillium expansum lipase are significantly enhanced in
ionic liquid [BMIm][PF6 ]. J. Mol. Catal. B 63 (1–2), 23–30.