Fuel Processing Technology 148 (2016) 184–187

Contents lists available at ScienceDirect

Fuel Processing Technology

journal homepage: www.elsevier.com/locate/fuproc

Short communication

Oil recovery from wet Euglena gracilis by shaking with liquefied

dimethyl ether
Kiyoshi Sakuragi a, Peng Li a,⁎, Nobuo Aoki b, Maromu Otaka a, Hisao Makino a
a
Energy Engineering Research Laboratory, Central Research Institute of Electric Power Industry (CRIEPI), 2-6-1 Nagasaka, Yokosuka, Kanagawa 240-0196, Japan
b
Research & Development Division, JX Nippon Oil & Energy Corporation, 6-3 Otemachi 2-chome, Chiyoda-ku, Tokyo 100-8162, Japan

a r t i c l e i n f o a b s t r a c t

Article history: An efficient, safe, and low-cost method for recovering oil from microalgae is needed for the production of algae
Received 15 January 2016 biofuels. Here, we report a one-step method for the recovery of oil from algae. Wet Euglena gracilis cells were
Received in revised form 2 March 2016 placed in a vessel, mixed with liquefied dimethyl ether, and shaken at room temperature (20 °C). After 5 min
Accepted 5 March 2016
of shaking, approximately 96.7% of the total oil was recovered when the ratio of solvent to sample (wet basis)
Available online 11 March 2016
was 8:1. The molecular weight distribution of oil recovered using this method was comparable to that of oil re-
Keywords:
covered using the conventional n-hexane Soxhlet extraction method. These results showed that it is possible to
Bio-fuel produce oil from wet microalgae by using a simple, economical, and environmentally friendly method.
Algae oil recovery © 2016 Elsevier B.V. All rights reserved.
Euglena gracilis
Dimethyl ether

1. Introduction environmentally friendly because of the properties of dimethyl ether

(DME) [14–16]. As a synthetic fuel, DME also has properties similar to
Global concerns associated with the release of greenhouse gases and those of LP gas [17]. Bench-scale studies are being performed to evaluate
the security of future liquid petroleum supplies have prompted re- the feasibility of future industrial application. However, compared with
searchers to develop methods to produce biofuels from algae as a new other types of biomass, wet microalgae normally form a ropy slurry,
biofuel resource; achievements in this field have attracted much atten- making the scaling up of algae oil extraction complicated and unsatis-
tion worldwide [1–4]. Euglena gracilis has been studied extensively factory. In contrast, in previous flow-type studies, we found that lique-
because of its ability to produce diverse valuable organic components fied DME appears to have excellent ability to penetrate algae cells,
[5,6]. E. gracilis has a simple cell structure. For example, eukaryotes although this phenomenon is still inexplicable experimentally. Based
such as Chlorella sp. consist of a cell wall and a cell membrane, whereas on these findings, we have developed an approach for the recovery of
E. gracilis has a pellicle made up of a protein layer supported by a oil from wet microalgae that only involves shaking E. gracilis with lique-
substructure of microtubules. In a recent patent, a method for the pro- fied DME, with the goal of industrial-scale algae fuel production. The op-
duction of jet fuel from E. gracilis was described that involved a general- timum ratio of solvent to sample and shaking time were investigated.
ised procedure of aerobic and nitrogen-deficient cultivation, harvesting,
drying, cell disruption, n-hexane extraction, and refinement of the re- 2. Materials and methods
covered oil to jet fuel [7]. However, algae oil recovery using convention-
al solvent extraction is difficult because the drying, cell disruption, and 2.1. Microalgae sample
subsequent extraction steps consume large amounts of energy, mainly
because of the high moisture content and small cell size of microalgae A wet E. gracilis sample was obtained from Euglena Co., Ltd. (Tokyo,
[8,9]. Therefore, direct wet extraction methods have been expanded re- Japan). The E. gracilis cells were cultured under conditions designed to
cently [10], and several attempts have been made to use hydrophilic or- increase the content of wax esters, which are the precursors for jet
ganic solvents or supercritical fluid technology in the process. However, fuel synthesis [7]. The water content of E. gracilis was 95.0% in the
to date, methods that are both practical and economical have not yet current study.
been identified. We previously developed a plug-flow-type extractor
for the dewatering and extraction of organic components from high- 2.2. Lab-scale experimental design
moisture biomass [11–13]. This technique is energy efficient and
The concept behind the lab-scale apparatus is shown in Fig. 1. The
⁎ Corresponding author. experimental design is based on our previous finding that liquefied
E-mail address: lipeng_commoner@yahoo.co.jp (P. Li). DME can penetrate the cells of some organisms and remove DME-

http://dx.doi.org/10.1016/j.fuproc.2016.03.005
0378-3820/© 2016 Elsevier B.V. All rights reserved.
 K. Sakuragi et al. / Fuel Processing Technology 148 (2016) 184–187 185

Fig. 1. Concept underlying the dimethyl ether (DME) shaking method. (A) Wet Euglena gracilis is placed in a vessel (open green and closed black circles represent algae cells and oil,
respectively); (B) liquefied DME is added to the vessel, and some of the oil diffuses into the DME layer; (C) the vessel is shaken, and oil is removed from the cells; and (D) oil (liquid)
and algae (solid) are separated.

soluble organic components [11]. Briefly, wet E. gracilis cells were put in time was increased to 5 min, the oil recovery ratio reached its maxi-
an extraction vessel (diameter, 27 mm; length, 235 mm; HPG-10-1, mum; thereafter, the oil recovery rate was almost constant, even up
Taiatsu Techno Corp., Tokyo, Japan) and mixed with liquefied DME. to 120 min. In contrast, when wet microalgae were mixed with DME
The vessel was shaken at 50 rpm (RW 20 digital mixer, IKA, Japan) at and held for 15 h without shaking, only 1.3% of the oil diffused into
a temperature of 20 °C and pressure of 0.51 MPa. The following extrac- the DME layer. These results indicate that shaking is the key step in
tion parameters were tested: (i) ratio of DME to wet weight of E. gracilis this procedure; very little oil remained inside the algae cells under op-
and (ii) shaking time (from 0–120 min, where the 0-min sample was timal experimental conditions.
added to DME and the mixture was held for 15 h without shaking). Fig. 4 shows the GPC curves of the oil recovered using DME shaking
After the shaking step, DME evaporated from the vessel naturally, and and n-hexane Soxhlet extraction; the curves were in good coincidence
the recovered oil floated on the algae sample. To measure the amount with each other. This result suggested that components in the tested
of oil recovered by DME, the vessel was centrifuged (Himac CT6D, E. gracilis are extractable by both DME and n-hexane, although their po-
Hitachi, Ltd., Tokyo, Japan) at 2800 ×g for 1 h. The oil in the upper liquid larities are different. The average molecular weights and number aver-
layer was poured into a separating funnel, extracted with n-hexane, and age molecular weights of the oil were 574 and 741 g mol−1 by using
then analysed gravimetrically. All experiments were performed three DME shaking and were 576 and 720 g mol−1 by using n-hexane Soxhlet
times to verify the reproducibility of the data. extraction, respectively. In a previous study, the oil extracted with n-
hexane from E. gracilis (the same sample used in this study) was mainly
2.3. Total oil content and gel permeation chromatography (GPC) analysis composed of wax esters [7]. The results from the GPC analysis indicated

The total oil content of E. gracilis was determined by gravimetric

analysis using classical n-hexane Soxhlet extraction. The E. gracilis cells
were disrupted using a homogeniser equipped with a saw-tooth gener-
ator probe (Dremel 300 Series; Robert Bosch Tool Corp., IL, USA; 10 mm
diameter) for 5 min at 10,000 rpm and then dried for 5 h at 50 °C in a
vacuum. The time and temperature for n-hexane Soxhlet extraction
were 16 h and 70 °C, respectively. GPC analysis was performed accord-
ing to a recently published method designed for the quantitative analy-
sis of wax esters produced by E. gracilis [7].

3. Results and discussion

The total oil content of E. gracilis was 19.7% (dry basis) based on
n-hexane Soxhlet extraction. Hereafter, the oil recovery ratio obtain-
ed using DME shaking is expressed relative to this total oil content.
Because liquefied DME is also partially miscible with water, the
amount of DME used changed when the algae water content varied.
Therefore, the amount of DME used in this study is expressed as the
ratio of the weight of DME to the weight of wet microalgae. As shown
in Fig. 2, the oil recovery ratio obtained using DME shaking increased
with an increasing solvent-to-sample ratio. At a DME to microalgae
ratio of 8, the oil recovery rate reached a maximum of 96.7%. Therefore,
Fig. 2. Effect of the solvent-to-sample ratio on the recovery of oil from wet E. gracilis. Algae
a solvent-to-sample ratio of 8:1 was used to investigate the effect of oil recovery is expressed relative to the total oil content determined using n-hexane
shaking time on algae oil recovery. As shown in Fig. 3, with a shaking Soxhlet extraction. Each data point represents the mean ± standard deviation (SD) of
time of 1 min, the oil recovery ratio was 84.6%. When the shaking three replicates. Where the error bars are not visible, they fit within the symbols.
186 K. Sakuragi et al. / Fuel Processing Technology 148 (2016) 184–187

that even though DME and n-hexane differ in polarity, the chemical
composition extracted by n-hexane can also be recovered by liquefied
DME. Therefore, considering the extraction yield and extractable chem-
ical compositions, liquefied DME is as effective as n-hexane as an extrac-
tion solvent for algae oil recovery, at least for E. gracilis. However, the
unique properties of DME allowed for the recovery of oil from wet
E. gracilis by shaking using a low amount of DME compared that needed
for other conventional extraction solvents. Moreover, the shaking time
was only 1 min. No experimental evidence showing that liquefied
DME can penetrate cells exists yet, but the results obtained in the cur-
rent study support this hypothesis. Most of the energy consumed in
this process was for DME recycling (evaporation and condensation); re-
search into DME recycling is currently underway, and because of its
near-ambient boiling point and lower vapour pressure than that of
other conventional extraction solvents, the development of a DME
recycling method with lower energy consumption can be expected.
The energy balance study on this method is being carried out and will
be further reported.
Algae provide many potential routes for conversion into biofuels, in-
Fig. 3. Effect of shaking time on the recovery of oil from wet E. gracilis. The triangle cluding hydrothermal liquefaction (HTL). HTL is one of the most prom-
represents the mixture of algae and DME held for 15 h without shaking. Each data point ising technologies for this purpose owing to its advantages of rapid
represents the mean ± SD of three replicates. Where the error bars are not visible, they reaction and the ability to use wet feedstocks with no lipid-content
fit within the symbols.
restriction. However, HTL of microalgae involves the thermochemi-
cal conversion of biomass into liquid fuels by processing in a hot
(523–647 K), pressurized water environment (4–22 MPa) for a
sufficient time to break down the solid biopolymeric structure into
mainly liquid components [18]. Therefore, the aqueous products
generated through HTL contain oxidative and toxic compounds
(e.g., phenols, pyridines), which may inhibit algae re-growth while
being recycled as nutrients [19]. Moreover, biocrude oil produced
via HTL has high contents of O and N elements, and needs further
deoxygenation and denitrogenation before it can be applied as trans-
port fuel [19]. By contrast, liquefied DME has been examined as a
prospective solvent in food processing because of its low toxicity
[16,20]. In addition, the molecular weight distributions of both the
nonpolar and polar lipids obtained by DME extraction are almost the
same as those obtained by hexane Soxhlet (as mentioned above).
Therefore, DME extraction is suitable for value-added foods, chemicals,
fertilizers, and biofuels (e.g., jet fuel, biodiesel). In the previous study,
lipids extracted using plug-flow-type extractor of DME from wet
E. gracilis were analysed by proximate analysis, and according to their
C, H, N, O, and S compositions [21]. In contrast to conventional hexane
Soxhlet extraction, minor element concentrations in the lipids obtained
by liquefied DME were higher, because of the weak hydrogen bonds of
liquefied DME, which indicates that another refining process is required
for further purification.

4. Conclusion

Algae oil can be recovered from wet E. gracilis by means of direct

shaking with liquefied DME. In the current lab-scale study, the optimum
solvent-to-sample ratio and shaking time were determined to be 8:1
and 5 min, respectively. This method does not require complex equip-
ment, high temperatures, toxic solvents, or long extraction times. The
unique phenomenon of DME penetrating some wet organisms should
be explored further. To achieve industrial-scale algae fuel production,
this method will be applied to other biofuel-producing algae species
and an efficient DME recycling method that leads to lower energy con-
sumption will be investigated. In addition, application of DME in a safe
manner should be considered in future commercial-scale process.

Fig. 4. Molecular weight distribution curves of the recovered oil. (A) Oil recovered from
Acknowledgements
wet E. gracilis by using the DME shaking method; (B) oil from E. gracilis pre-treated with
drying and cell disruption, and extracted using the n-hexane Soxhlet method. None.
 K. Sakuragi et al. / Fuel Processing Technology 148 (2016) 184–187 187

References [12] P. Li, H. Makino, Liquefied dimethyl ether: an energy-saving, green extraction sol-
vent, in: F. Chemat, V.M. Abert (Eds.), Alternative Solvents for Natural Products Ex-
[1] M. Aresta, A. Dibenedetto, G. Barberio, Utilization of macro-algae for enhanced CO2 traction, Springer, Berlin Heidelberg 2014, pp. 91–106.
fixation and biofuels production: development of a computing software for an LCA [13] H. Kanda, P. Li, Simple extraction method of green crude from natural blue-green
study, Fuel Process. Technol. 86 (2005) 1679–1693. microalgae by dimethyl ether, Fuel 90 (2011) 1264–1266.
[2] P. Nautiyal, K.A. Subramanian, M.G. Dastidar, Production and characterization of [14] H. Holldorff, H. Knapp, Binary vapor-liquid-liquid equilibrium of dimethyl ether —
biodiesel from algae, Fuel Process. Technol. 120 (2014) 79–88. water and mutual solubilities of methyl chloride and water: experimental results
[3] Z. Wang, S. Adhikari, P. Valdez, R. Shakya, C. Laird, Upgrading of hydrothermal lique- and data reduction, Fluid Phase Equilib. 44 (1988) 195–209.
faction biocrude from algae grown in municipal wastewater, Fuel Process. Technol. [15] J. Wu, Y. Zhou, E.W. Lemmon, An equation of state for the thermodynamic proper-
142 (2016) 147–156. ties of dimethyl ether, J. Phys. Chem. Ref. Data 40 (2011) 023104.
[4] C. Rodriguez, A. Alaswad, J. Mooney, T. Prescott, A.G. Olabi, Pre-treatment techniques [16] European Food Safety Authority, Scientific opinion of the panel on food contact
used for anaerobic digestion of algae, Fuel Process. Technol. 138 (2015) 765–779. materials, enzymes, flavourings and processing aids (CEF) on dimethyl ether as an
[5] H. Takeyama, A. Kanamaru, Y. Yoshino, H. Kakuta, Y. Kawamura, T. Matsunaga, Pro- extraction solvent, EFSA J. 984 (2009) 1–13.
duction of antioxidant vitamins, β-carotene, vitamin C, and vitamin E, by two-step [17] M.C. Lee, S.B. Seo, J.H. Chung, Y.J. Joo, D.H. Ahn, Industrial gas turbine combustion
culture of Euglena gracilis Z, Biotechnol. Bioeng. 53 (1997) 185–190. performance test of DME to use as an alternative fuel for power generation, Fuel
[6] J.S. Rodríguez-Zavala, M.A. Ortiz-Cruz, G. Mendoza-Hernández, R. Moreno-Sánchez, 88 (2009) 657–662.
Increased synthesis of α-tocopherol, paramylon and tyrosine by Euglena gracilis [18] D.C. Elliott, P. Biller, A.B. Ross, A.J. Schmidt, S.B. Jones, Hydrothermal liquefaction of
under conditions of high biomass production, J. Appl. Microbiol. 109 (2010) 2160. biomass: developments from batch to continuous process, Bioresour. Technol. 178
[7] N. Aoki, H. Matsuda, H. Kato, A. Yoneda, R. Arashida, Y. Marukawa, Method for pro- (2015) 147–156.
ducing fuel oil base, JP Patent (2013) WO2013115137 A1 [19] C. Tian, B. Li, Z. Liu, Y. Zhang, H. Lu, Hydrothermal liquefaction for algal biorefinery: a
[8] A. Kumar, S. Ergas, X. Yuan, A. Sahu, Q. Zhang, J. Dewulf, F.X. Malcata, H.V. critical review, Renew. Sust. Energ. Rev. 38 (2014) 933–950.
Langenhove, Enhanced CO2 fixation and biofuel production via microalgae: recent [20] V. Varlet, F. Smith, M. Augsburger, New trends in the kitchen: propellants assess-
developments and future directions, Trends Biotechnol. 28 (2010) 371–380. ment of edible food aerosol sprays used on food, Food Chem. 142 (2014) 311–317.
[9] M.K. Lam, T.K. Lee, Microalgae biofuels: a critical review of issues, problems and the [21] H. Kanda, P. Li, M. Goto, H. Makino, Energy-saving lipid extraction from wet
way forward, Biotechnol. Adv. 30 (2012) 673–690. Euglena gracilis by the low-boiling-point solvent dimethyl ether, Energies 8 (2015)
[10] L. Lardon, A. Hélias, B. Sialve, J.P. Steyer, O. Bernard, Life-cycle assessment of biodie- 610–620.
sel production from microalgae, Environ. Sci. Technol. 43 (2009) 6475–6481.
[11] P. Li, H. Kanda, H. Makino, Simultaneous production of bio-solid fuel and bio-crude
from vegetal biomass using liquefied dimethyl ether, Fuel 116 (2014) 370–376.