Biomass and Bioenergy 112 (2018) 138–143

Contents lists available at ScienceDirect

Biomass and Bioenergy

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

Research paper

Catalytic cracking of wax esters extracted from Euglena gracilis for T

hydrocarbon fuel production
Iori Shimadaa,∗, Yoshitaka Nakamurab, Shin Katoa, Ryohei Morib, Haruhisa Ohtac, Kengo Suzukic,
Toru Takatsukad
a
Faculty of Textile Science and Technology, Shinshu University, 3-15-1 Tokida, Ueda, Nagano 386-8567, Japan
b
Graduate School of Science and Technology, Shinshu University, 3-15-1 Tokida, Ueda, Nagano 386-8567, Japan
c
Research & Development Department, euglena Co., Ltd., 75-1 Ono, Tsurumi-ku, Yokohama, Kanagawa 230-0046, Japan
d
Technology Development Unit, Chiyoda Corporation, 4-6-2 Minatomirai, Nishi-ku, Yokohama, Kanagawa 220-8765, Japan

A R T I C LE I N FO A B S T R A C T

Keywords: Wax esters (WEs) synthesized by Euglena gracilis are potential sources for alternative fuels because of their high
Catalytic cracking productivity, recent success in mass cultivation, and low energy consumption in extraction. In this study,
Wax ester deoxygenation of Euglena WE and conversion to hydrocarbons in a catalytic cracking process under a hydrogen-
Euglena gracilis free atmosphere was investigated using a residue fluid catalytic cracking equilibrium catalyst with enhanced
Deoxygenation
hydrogen-transfer activity. The deoxygenation of Euglena WE proceeded more rapidly with higher H2O se-
Hydrogen-transfer
lectivity than that of saturated triglycerides. This is because initial β-elimination of WEs produces saturated fatty
Renewable fuel
acids and higher olefins; the higher olefins rapidly release hydrogen species during cracking, cyclization and
aromatization, and the hydrogen species accelerate hydrodeoxygenation of the saturated fatty acids.
Furthermore, the cracking of Euglena WE produced large amounts of paraffins and olefins instead of aromatics.
Therefore, Euglena WE was confirmed to be a preferable feedstock for the catalytic cracking process for hy-
drocarbon fuel production.

1. Introduction
Microalgae have attracted wide attention because they exhibit high
growth rates on sites that are unsuitable for agriculture, thus avoiding
undesirable competition with food crops. They produce various nu-
trients and can be used in various applications such as supplements,
drugs, and cosmetics. In addition, they are known to produce and store
lipids with much higher productivity than conventional plant crops [1].
Therefore, microalgae-derived renewable fuels are expected as poten-
tial sources to replace fossil fuels and development of their efficient
conversion technology is necessary.
Euglena gracilis is one of such promising microalgae and mass cul-
tivation of it in an open pond has been achieved recently [2]. Euglena
gracilis synthesizes wax esters (WEs) from paramylon, a Euglena specific
polysaccharides, when it is brought from aerobic to anaerobic condi-
tions [3], and the lipid yields can reach 50% of cellular dry weight
[3,4]. WEs are esters composed of saturated fatty acids and fatty al-
cohols. The accumulation of WEs is unique to Euglena because most
other microalgae store lipids as triglycerides (TGs), which are esters
composed of glycerol and saturated or unsaturated fatty acids. Euglena
is also unique in that it has a simple cell structure without cell walls,
and therefore accumulated WEs can be extracted easily without re-
quiring energy-consuming disruption of tough cell walls [5,6].
To use WEs as transportation fuels, they can be converted to bio-
diesel (fatty acid methyl esters; FAMEs) by a transesterification process
[7,8]. However, the biodiesel is an oxygen-containing fuel, whose heat
value is low and thermal and oxidation stabilities are poor. To improve
the heat value and stabilities, deoxygenation of WEs and their con-
version to hydrocarbons are important. The deoxygenation of WEs is
usually performed using a hydrotreating process [9], but using pres-
surized hydrogen atmosphere incurs high operation costs. These are
common problems with the utilization of TGs as transportation fuels.
In contrast, a fluid catalytic cracking (FCC) process can convert TGs
to hydrocarbons under a hydrogen-free atmosphere [10–14]. FCC
process is one of the major oil refining technologies that processes
feedstocks such as heavy gas oil, vacuum gas oil and atmospheric re-
sidue. Typical FCC catalysts consist of zeolite (ultra-stable Y (USY)
partly exchanged with rare earth (RE) ions), filler (kaolin clay) and
matrix (meso- and macroporous silica or alumina) [15]. The heavy oil
feedstocks are converted to high-octane gasoline blend, light olefins

∗
Corresponding author.
E-mail address: iori@shinshu-u.ac.jp (I. Shimada).

https://doi.org/10.1016/j.biombioe.2018.03.004
Received 10 October 2017; Received in revised form 2 March 2018; Accepted 7 March 2018
Available online 13 March 2018
0961-9534/ © 2018 Elsevier Ltd. All rights reserved.
I. Shimada et al. Biomass and Bioenergy 112 (2018) 138–143

(propylene and butene), heavy liquid hydrocarbons (light cycle oil; %). The detailed fatty acid compositions of the two plant oils were
LCO), and coke during cracking, isomerization, cyclization and hy- reported previously [16].
drogen transfer reactions on the acid sites of the USY zeolite. The coke
deposited on the catalysts are burned off in the regenerator and the
2.2. Catalytic activity tests
regenerated catalysts are re-used continuously. In addition to these
reactions, deoxygenation reaction proceeds in TG cracking and pro-
Catalytic activity tests were conducted in a fixed-bed microactivity
duces H2O, CO2 and CO. In particular, we recently found that enhan-
test reactor, details of which were given in a previous publication [22].
cing the hydrogen-transfer activity in the FCC process is effective for
In each trial, E-cat (1.9–5.3 g) was placed in the reactor, which was
removing oxygens in TGs as H2O [16]. The formation of H2O rather
maintained at 470 °C. The feedstock (Euglena WE, coconut oil and
than CO2 and CO is an efficient reaction because all the carbon re-
sunflower oil) was fed into the reactor using a microfeeder while being
sources contained in TGs can be converted to hydrocarbons without
heated electrically in a preheating line. The feed injection time was 75 s
carbon loss. Therefore, the FCC with enhanced hydrogen-transfer ac-
and the weight hourly space velocity (WHSV) was varied between 12
tivity is considered to be a promising process for alternative fuel pro-
and 33 h−1. For these conditions, the weight ratio of the catalyst to the
duction. However, the catalytic cracking of WEs has, insofar as we are
oil (cat/oil) was in the range of 1.5–4.0. N2 gas was added during feed
aware, not been reported. In this study, we therefore investigated the
injection at 19 mL min−1. After each test, the catalyst was stripped by
deoxygenation of WEs in a catalytic cracking process and compared
purging with N2 gas at 5 mL min−1 for 15 min. During the reaction and
their reaction with that of TGs.
stripping stages, the liquid products were collected in a cold trap with
two receiving vessels connected in series and maintained at 0 and
-15 °C, respectively. Simultaneously, the gaseous products were col-
2. Materials and methods
lected in a gas burette by displacement of a saturated NaCl solution. For
all experiments, a mass balance across the collected products was found
2.1. Materials
to be in the range of 92–104 wt% of the process inputs.
The amounts of H2, N2, CO2 and CO in the gaseous products were
An equilibrium catalyst (E-cat), which was obtained from a com-
determined using a GC system (GC-8A, Shimadzu Corp., Kyoto, Japan)
mercial residue FCC (RFCC) unit, was used for all the catalytic cracking
equipped with a packed column (SHINCARBON-ST 50/80, Shinwa
experiments. The RFCC E-cat is known to exhibit strong hydrogen
Chemical Industries, Ltd., Kyoto, Japan) and a thermal conductivity
transfer activity because its hydrothermal stability in the reactor-re-
detector (TCD) with Ar as the carrier gas. Hydrocarbons and oxygenates
generator cycle was enhanced by the high RE loading on the USY
were analyzed with another GC system (GC-2014, Shimadzu Corp.)
zeolite. The effect of RE loading on the catalyst stability and hydrogen
fitted with a capillary column (BP1, SGE Analytical Science Pty. Ltd.,
transfer activity has been discussed in the literature [17–20].
Victoria, Australia) and FID with He as the carrier gas. Biphenyl
The Euglena WE, which was extracted from Euglena gracilis, was
(99.5%, Sigma-Aldrich Co., St. Louis, MO, USA) was used as the internal
purified and supplied by euglena Co., Ltd. The composition of fatty
standard in the GC-FID assay. Quantification of the hydrocarbons and
acids and fatty alcohols of the Euglena WE was analyzed using the
oxygenates was based on the effective carbon number theory [23]. The
method reported by Inui et al. [4] and Ichihara and Fukubayashi [21].
amount of coke deposited on the catalyst was determined from the
WEs were converted to FAMEs and fatty alcohols, which were analyzed
difference between the mass of the reactor before and after testing.
using a gas chromatography (GC) system (GC-390B, GL Sciences Inc.,
Three hydrocarbon fractions were defined based on carbon number or
Tokyo, Japan) equipped with a flame ionization detector (FID) using He
boiling point thresholds: gaseous hydrocarbons (C1 − C4); gasoline (C5
as a carrier gas. The packed columns used for fractionating FAMEs and
– 216 °C) and LCO (216 °C –). In addition, jet fuel fraction was defined
fatty alcohols were a 3 m-length Unisole 3000 coated on Uniport C 80/
as carbon number range of C10 − C15.
100 (GL Sciences Inc.) and a 1 m-length PEG-HT coated on Uniport HP
60/80 (GL Sciences Inc.), respectively. As shown in Table 1, the Euglena
WE mainly consisted of myristyl myristate. In addition, coconut oil 3. Results
(Kaneda Shoji Co., Ltd., Tokyo, Japan) and sunflower oil (Wako Pure
Chemical Industries, Ltd., Osaka, Japan) were also used as reactants. 3.1. Reaction products of WE cracking
Both coconut and sunflower oil consist of TGs, but their fatty acid
composition is different. Coconut oil consists mainly of saturated C12 The reaction product yields from the catalytic cracking of Euglena
and C14 fatty acids (46.9 mol% and 15.7 mol%, respectively), whereas WE are shown in Fig. 1 and compared with those from coconut oil
sunflower oil consists mainly of C18 unsaturated fatty acids (88.5 mol (saturated TG) and sunflower oil (unsaturated TG), which were

Table 1
Fatty acid and fatty alcohol compositions of Euglena WE used in this study.

Fatty acid [carbon number:double bond] Composition [mol%] Fatty alcohol [carbon number:double bond] Composition [mol%]

Capric acid [10:0] Trace Capric alcohol [10:0] Trace

Undecanoic acid [11:0] Trace Undecyl alcohol [11:0] Trace
Lauric acid [12:0] 8.1 Lauryl alcohol [12:0] 5.6
Tridecanoic acid [13:0] 9.3 Tridecyl alcohol [13:0] 10.6
Myristic acid [14:0] 57.8 Myristyl alcohol [14:0] 50.9
Pentadecanoic acid [15:0] 2.5 Pentadecyl alcohol [15:0] 7.3
Palmitic acid [16:0] 9.3 Palmityl alcohol [16:0] 20.9
Palmitoleic acid [16:1] 1.6 Heptadecyl alcohol [17:0] 0.7
Heptadecanoic acid [17:0] Trace Stearyl alcohol [18:0] 1.5
Stearic acid [18:0] 1.9 Oleyl alcohol [18:1] 2.5
Oleic acid [18:1] 4.6
Linoleic acid [18:2] 1.5
Eicosadienoic acid [20:2] 1.2
Eicosatrienoic acid [20:3] 1.1
Eicosatetraenoic acid [20:4] 1.0

139
I. Shimada et al.
100
80 Unknown (include water)
Oxygenates
CO
60 CO2
Yield [wt%]
H2
Gaseous hydrocarbons
40
Gasoline
LCO
20 Coke
0
Euglena Coconut Sunflower
Fig. 1. Reaction product yields from catalytic cracking of Euglena WE, coconut oil and
sunflower oil (470 °C, WHSV = 16 h−1).
investigated in our previous study [16]. Hydrocarbons, which are re-
presented in the figure as gaseous hydrocarbons, gasoline and LCO,
were the main reaction products in all the experiments, whereas CO2,
CO and some oxygenates such as fatty acids, ketones, aldehydes, and
esters were also detected. Water droplets were visible in the liquid
products from all feedstocks, but they were not quantified. The “un-
known” yield in Fig. 1 therefore contains both water and large-molecule
products that were not detected by GC-FID because of their high boiling
points.
The hydrocarbon yield from Euglena WE was higher than that from
coconut oil but lower than that from sunflower oil. The high hydro-
carbon yield from sunflower oil is due to the high reactivity of un-
saturated bonds in the catalytic cracking reaction; they are rapidly
converted to carbenium ions on the Brønsted acid sites [24] and the
carbenium ions are key intermediates for both decomposition and
deoxygenation [16]. Among the saturated feedstocks, Euglena WE
showed higher hydrocarbon yields and lower oxygenate yields than
coconut oil did, which indicates that deoxygenation of Euglena WE
proceeded more rapidly than that of coconut oil. This can be attributed
to the difference in the molecular structure.
Table 2 shows the yields of oxygen-containing species. The catalytic
cracking of Euglena WE produced lower CO2 and CO yields (oxygen
base) than that of coconut oil despite the progress of deoxygenation.
This result suggests that the deoxygenation of Euglena WE was accom-
panied by H2O formation rather than CO2 and CO. The deoxygenation
pathway producing H2O instead of CO2 or CO is efficient because it can
suppress the carbon loss in the conversion of biomass resources to hy-
drocarbons.
Fig. 2 shows the hydrocarbon yields and their composition in gas-
eous, gasoline, jet fuel and LCO fraction from the catalytic cracking of
three feedstocks. The saturated feedstocks (Euglena WE and coconut oil)
produced much gaseous olefins such as propylene and butene, which
can be used as chemical feedstocks. In contrast, the catalytic cracking of
sunflower oil produced more gaseous paraffins (propane and butane)
than gaseous olefins. The butane/butene ratio in the catalytic cracking
of each feedstock was 0.8, 1.1 and 2.4 for Euglena WE, coconut oil and
sunflower oil, respectively. Because the butane/butene ratio is an index
of hydrogen-transfer reaction activity [25], the result indicates the high
activity of the hydrogen-transfer reaction in the catalytic cracking of
sunflower oil. In this case, the hydrogen donors may be olefins and
Table 2
Oxygen-based yields [O%] of oxygen-containing products from catalytic cracking of
various feedstocks (470 °C, WHSV = 16 h−1).
Euglena WE Coconut oil
Oxygenates < 0.4 4.8
CO2 12.6 19.5
CO 11.9 12.4
Biomass and Bioenergy 112 (2018) 138–143
naphthenes that were produced from the cracking of unsaturated
carbon chains in sunflower oil. These hydrogen donors can release
hydrogen species by cyclization and aromatization reaction. Then, the
released hydrogen species are received by light olefins (hydrogen ac-
ceptors) to produce light paraffins. In contrast, the hydrogen-transfer
reaction activity of saturated feedstocks was confirmed not to be so
strong, which is because production of hydrogen donor is not so active.
The gasoline fraction produced from the saturated feedstocks
(Euglena WE and coconut oil) contained a large proportion of olefins,
whereas that from sunflower oil contained a large proportion of aro-
matics. The low olefin yield from sunflower oil cracking can be ex-
plained by the strong activity of the hydrogen-transfer reaction; the
olefins in the gasoline fraction were converted to paraffins and aro-
matics, which results in the lower octane rate of the produced gasoline.
In contrast, olefins in the gasoline fraction produced from WE cracking
were not converted because of the relatively low activity of the hy-
drogen-transfer reaction and therefore the octane rate of the gasoline
produced from the WE cracking becomes high. The light olefins in
gasoline fraction can also be used as chemical feedstocks.
A similar trend was observed in the composition of the jet fuel and
LCO fractions. Although the jet fuel and LCO fraction yields were the
highest in the cracking of sunflower oil, their composition was domi-
nated by polycyclic aromatic hydrocarbons, which have poor combus-
tion properties and are characterized as soot and smoke producers [26].
Indeed, the aromatic content in aviation fuel is limited to less than 25%
(ASTM D-1655). In contrast, the catalytic cracking of saturated feed-
stocks produced jet fuels and LCO containing a large proportion of
paraffins and olefins. Particularly, jet fuel fraction produced from Eu-
glena WE contained a large proportion of iso-paraffins, which are de-
sirable for jet fuel because of their good combustion and cold tem-
perature properties [27–29]. The olefins and aromatics in jet fuel and
LCO fractions can be converted to aviation and diesel fuels by hydro-
genation processes although the hydrogenation of polycyclic aromatic
hydrocarbons consumes large amount of hydrogen. In contrast, the
hydrogenation of olefins requires less hydrogen consumption.
The reason for the high yield of paraffins in the jet fuel and LCO
fractions from Euglena WE can be inferred by the detailed composition
of the produced hydrocarbons, which is shown in Fig. 3. The yields of
the paraffins and olefins with carbon numbers 13 and 14 were re-
markably higher than those of the neighboring carbon numbers. As
shown in Table 1, Euglena WE consists mainly of C14 fatty acid and C14
fatty alcohol. The composition of the reaction products from Euglena
WE therefore probably reflected the fatty acid and fatty alcohol that the
reactant consisted of. This is the reason for the high yield of jet-fuel and
LCO fraction paraffins from Euglena WE cracking, indicating that Eu-
glena WE is suitable as a feedstock for aviation and diesel fuels. The
formation of hydrocarbons reflecting the fatty acid and fatty alcohol
composition of the reactant is a result of the decomposition of the ester
bonds between the fatty acids and the fatty alcohols. This suggests that
in the catalytic cracking of Euglena WE, ester bond decomposition is
faster than the cracking of saturated carbon chains of the fatty acids and
fatty alcohols, which is coincident with the observation of the catalytic
cracking of saturated TGs [16]. In contrast, unsaturated carbon chains
in sunflower oil were cracked rapidly and converted to gasoline-frac-
tion olefins and naphthenes, thus the paraffins and olefins in the jet fuel
and LCO fractions were hardly produced. The excess cyclization, ar-
omatization and condensation of the gasoline-fraction olefins and
naphthenes produced from sunflower oil cracking resulted in the for-
mation of polyaromatics.
Consequently, the saturated feedstocks produced hydrocarbons with
the composition suitable for fuel application (gasoline, jet and diesel
Sunflower oil fuel) in contrast to the unsaturated feedstocks, which did not. Among
the saturated feedstocks, deoxygenation of Euglena WE proceeded more
< 0.7
rapidly and efficiently (less carbon loss) than that of coconut oil. Thus,
13.4
10.5 it was confirmed that Euglena WE is more preferable for the feedstock of
the FCC process for producing hydrocarbon fuels than TGs.
140
I. Shimada et al. Biomass and Bioenergy 112 (2018) 138–143

(a) 30 (b) 70 (c) 40 (d) 40

Butene Monoaromatics Polyaromatics Polyaromatics
Butane 60 Olefins Monoaromatics Monoaromatics

Gaseous hydrocarbon yield [wt%]

Propylene Cycloparaffins Olefins Olefins
Propane i-Paraffins 30 i-Paraffins 30 i-Paraffins
50

Gasoline yield [wt%]

20

Jet fuel yield [wt%]

C -C n-Paraffins n-Paraffins n-Paraffins

LCO yield [wt%]

40
20 20
30
10
20
10 10
10

0 0 0 0
Euglena Coconut Sunflower Euglena Coconut Sunflower Euglena Coconut Sunflower Euglena Coconut Sunflower

Fig. 2. Yields and compositions of the (a) gaseous, (b) gasoline, (c) jet fuel and (d) LCO hydrocarbon fractions from catalytic cracking of Euglena WE, coconut oil and sunflower oil (470 °C,
WHSV = 16 h−1).

12 residence time. As mentioned in the previous section, the C13 and C14
hydrocarbons are the initial products of the deoxygenation reaction.
10 The C13 yield was higher than that of C14 for n-paraffins, whereas the
C14 yield was higher than that of C13 for i-paraffins and olefins (espe-
8 cially in short residence times). The yields of n-paraffins and olefins
Polyaromatics
Yield [wt%]

Monoaromatics decreased as extending residence time, which was the most drastic in
6 the C14 olefin yield. In contrast, the yield of i-paraffins initially in-
Olefins
Cycloparaffins creased and then slightly decreased. These results indicate that n-par-
4 i-Paraffins affins and olefins are initially produced from the deoxygenation reac-
n-Paraffins
tion, whereas i-paraffins are secondary products. With a sufficient
2
residence time, i-paraffins are also cracked into smaller hydrocarbons.

0
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
C11
C12
C13
C14
C15
C16
C17
C18

4. Discussion
Carbon number
The decomposition of esters has been widely investigated in thermal
Fig. 3. Carbon number distributions of hydrocarbons produced from catalytic cracking of and catalytic cracking. Initially, esters undergo thermal transformation
Euglena WE (470 °C, WHSV = 16 h−1).
via a β-elimination mechanism [30,31]. In the case of WE cracking
shown in Fig. 6, the β-elimination reaction produces saturated fatty
3.2. Effect of residence time acids and higher olefins (R1), which were observed in Figs. 4b and 5c,
respectively. The higher olefins are very reactive in the catalytic
To clarify the reaction mechanism of the WE cracking, the reaction cracking process and their reactions can be well explained with a me-
products with different residence times were investigated and are chanism involving carbenium ion intermediates [24]. The higher ole-
shown in Fig. 4. The hydrocarbons, CO2, CO and coke yields increased fins are rapidly converted to carbenium ions on Brønsted acid sites of
as residence time is extended. This relation reflects involvement of the the zeolite catalysts (R2), thus the yield of C14 olefins rapidly decreases
deoxygenation reaction. The oxygen-containing products such as fatty in Fig. 5c. The carbenium ions are converted to i-paraffins, observed in
acids, ketones, aldehydes and esters, which were detected with short Fig. 5b, through the isomerization and hydride abstraction from par-
residence times, became undetectable after a sufficient residence time. affins or naphtnenes (R3). Otherwise, the carbenium ions are converted
Among these oxygenates, the aldehydes yield was the highest. This to gasoline-fraction hydrocarbons (paraffins, olefins and naphthenes)
result was different from the observation in the catalytic cracking of through β-scission reaction (R4). The gasoline-fraction olefins and
saturated TGs, in which the yields of the fatty acids and ketones were naphthenes can be cracked to gaseous hydrocarbons (R5) or can work
higher than that of the aldehydes [16]. Thus, the aldehyde formation as hydrogen donors by releasing hydrogen species through a cyclization
pathway is more active in Euglena WE cracking than in TG cracking. and aromatization reaction (R7 and R8). On the other hand, the fatty
Fig. 5 shows the yield of C13 and C14 hydrocarbons with different acids have three deoxygenation pathways in the catalytic cracking

(a)100 6 (b) 4
Hydrocarbons Fatty acids
CO2 Ketones
80 CO Aldehydes
3
Coke Esters
4
60
Yield [wt%]
Yield [wt%]

Yield [wt%]

2
40
2
1
20

0 0 0
0
0.00 0.02 0.04 0.06 0.08 0.10 0
0.00 0.02 0.04 0.06 0.08 0.10
WHSV-1 [h] WHSV-1 [h]

Fig. 4. Reaction product yields from catalytic cracking of Euglena WE (470 °C): (a) hydrocarbons, CO2, CO and coke, and (b) oxygenates.

141
I. Shimada et al. Biomass and Bioenergy 112 (2018) 138–143

(a) 4 (b) 4 (c) 12

C n-paraffin C i-paraffin C olefin
C n-paraffin C i-paraffin 10 C olefin
3 3
8

Yield [wt%]
Yield [wt%]

Yield [wt%]
2 2 6

4
1 1
2

0 0 0
0.00
0 0.02 0.04 0.06 0.08 0.10 0
0.00 0.02 0.04 0.06 0.08 0.10 0
0.00 0.02 0.04 0.06 0.08 0.10
WHSV [h] WHSV [h] WHSV [h]

Fig. 5. Yields of C13 and C14 hydrocarbons from catalytic cracking of Euglena WE (470 °C): (a) n-paraffins, (b) i-paraffins and (c) olefins.

process; direct decarboxylation (R10), ketonic decarboxylation (R11 – Table 3

R13) and hydrodeoxygenation (R14 – R16) [16]. Among the three Weight-based and mole-based yields of inorganic products from catalytic cracking of
various feedstocks (470 °C, WHSV = 16 h−1).
pathways, the hydrodeoxygenation via aldehyde formation is ac-
celerated by the enhancement of hydrogen-transfer activity of the cat- Euglena WE Coconut oil Sunflower oil
alyst [16] or the coexistence of hydrogen donors [32,33]. Thus, in the
Euglena WE cracking on the RFCC E-cat, which exhibits strong hy- Weight-based yield [wt%]
H2 0.009 0.011 0.018
drogen-transfer activity, the hydrodeoxygenation pathway is ac-
CO2 1.41 3.84 1.77
celerated by receiving hydrogen species released from the gasoline- CO 1.69 3.10 1.77
fraction olefins and naphthenes derived from the higher olefins. This H2O a
6.85 10.23 8.18
can be confirmed by the high aldehyde yield and low CO2 yield as Mole-based yield [mol/molFeed]
shown in Fig. 4b and Table 2, respectively. The deoxygenation of fatty H2 0.018 0.034 0.091
CO2 0.13 0.56 0.40
acids produces jet-fuel/LCO-range paraffins and olefins, and the con-
CO 0.24 0.71 0.63
secutive β-scission of them produces gasoline-range hydrocarbons H2O a
1.50 3.66 4.53
(R17).
a
At this time, the hydrogen species transferred from the hydrogen Calculated value from oxygen balance.
donor to the oxygenates have not been identified. However, the hy-
drogen species should not be molecular hydrogens (H2) because of the oxygenates may be protons and hydride ions, which should be clarified
following two reasons. 1) Under N2 flow, the H2 yield shown in Table 3 in the future work.
corresponds to that the partial pressure of H2 in the reactor is lower Finally we compared the reaction mechanism of WE cracking with
than 7 kPa although typical hydrodeoxygenation processes are operated that of TG cracking. In the case of the saturated TG cracking, the initial
under several MPa of H2. Despite the extremely low H2 partial pressure, β-elimination reaction produces two fatty acids, aldketene, and acrolein
the deoxygenation selectivity to H2O in WE cracking was higher than [36,37]. Due to its instability, the aldketene rapidly decomposes [38].
that to CO2 or CO; the mole-based yield of H2O was even higher than Meanwhile, acrolein decomposes to olefins such as propylene, ar-
that of H2 by almost two orders of magnitude. 2) Hydrodeoxygenation omatizes via a Diels-Alder reaction, or polymerizes to form coke [39],
process under H2 atmosphere requires transition metal or metal sulfide but it does not exhibit as strong a hydrogen-donating ability as the
catalysts that can activate molecular H2 [34,35], but the RFCC catalysts higher olefins produced from Euglena WE. Therefore, the hydro-
do not contain such species. On the other hand, the hydrogen transfer deoxygenation of fatty acids is less active, resulting in the production of
reaction on the zeolite catalysts in petroleum cracking process (naph- a large amount of CO2 or CO (Table 2). In contrast, unsaturated TGs
thene + olefin → aromatics + paraffin) has been investigated by Po- have a strong hydrogen-donating ability because the unsaturated
tapenko et al. [25]. They concluded that the hydrogen species involved carbon chains are very reactive and rapidly produce a large amount of
in the hydrogen transfer reaction are protons from Brønsted acid sites of hydrogen donors (gasoline-fraction olefins and naphthenes) through a
the zeolite and hydride ions from the hydrogen donor. Considering the cracking reaction. Thus, the hydrodeoxygenation of ester bonds is ac-
above, the hydrogen species transferred from the hydrogen donor to the tivated and results in low CO2 and CO yields.

Fig. 6. Schematic diagram of the deoxygenation reaction

mechanism during catalytic cracking of WEs. Red arrows
represent the hydrogen-transfer reaction pathway.
Hydrogen species released from R8 are consumed pre-
ferentially in the hydrodeoxygenation reaction (red solid
arrows) rather than the olefin hydrogenation (red dashed
arrows). (For interpretation of the references to colour in
this figure legend, the reader is referred to the Web version
of this article.)

142
I. Shimada et al.
However, the hydrogen-donating ability of unsaturated TGs is much
higher than that required for the deoxygenation of ester bonds [33].
The excess hydrogen species released from hydrogen donors are con-
sumed by the hydrogen-transfer reaction with olefins in gaseous and
gasoline fractions to produce paraffins. The hydrogenation of olefins
(high octane rate) to paraffins (low octane rate) decreases the octane
rate of the produced gasoline. The progress of the hydrogen-transfer
reaction is confirmed by the high butane/butene ratio in Fig. 2a. In
addition, the hydrogen donors (gasoline-fraction olefins and naph-
thenes) are converted to aromatics by cyclization and aromatization,
which results in the high aromatics yield. Furthermore, a large amount
of carbenium ions produced from unsaturated carbon chains interacts
with paraffins via hydrogen-transfer reactions, which activates paraf-
fins (converts them to carbenium ions by chain reaction mechanism)
and results in a cyclization and aromatization reaction that produce
aromatics. Then, the high concentration of aromatics results in a con-
densation reaction which produces polyaromatics. Therefore, the ga-
soline, jet fuel and LCO fractions from unsaturated TG cracking are
dominated by aromatic hydrocarbons (Fig. 2), which are unsuitable for
fuel applications. In contrast, in the case of the Euglena WE cracking, the
higher olefins produced from the initial β-elimination reaction are
highly reactive and converted to hydrogen donors via a cracking re-
action, but the saturated fatty acids are relatively stable and they work
as hydrogen acceptors in the hydrogen-transfer hydrodeoxygenation
reaction. Because the oxygenated compounds such as fatty acids and
aldehydes preferentially receive hydrogen species (R14 and R15) than
gaseous and gasoline-fraction olefins do (R6 and R9) [33], the hydro-
genation of olefins does not proceed so much in the WE cracking. Thus
the gasoline, jet fuel and LCO fractions contain large amounts of par-
affins and olefins instead of aromatics. Considering the above, Euglena
WE was confirmed to be more suitable than TGs as a feedstock of hy-
drocarbon fuel production in a catalytic cracking process.
5. Conclusions
The deoxygenation and hydrocarbon formation in the catalytic
cracking of Euglena WE was investigated. WEs are initially converted to
saturated fatty acids and higher olefins by β-elimination reaction. The
higher olefins are highly reactive and decomposed to gasoline-fraction
olefins and naphthenes, which can release hydrogen species through
cyclization and aromatization. The hydrogen species are consumed in
the deoxygenation of the saturated fatty acids to produce H2O and jet-
fuel/LCO-range paraffins and olefins. Therefore, the deoxygenation of
WEs in catalytic cracking proceeds with a faster reaction rate, higher
H2O selectivity, and suppression of polyaromatics formation compared
with that of TGs.
Acknowledgements
This work was financially supported by JSPS KAKENHI (Grant
Number JP15K21038) and the JGC-S Scholarship Foundation.
References
[1] Y. Chisti, Biodiesel from microalgae, Biotechnol. Adv. 25 (3) (2007) 294–306.
[2] K. Suzuki, Large-scale cultivation of euglena, in: S.D. Schwartzbach, S. Shigeoka
(Eds.), Euglena: Biochemistry, Cell and Molecular Biology, Springer International
Publishing, Cham, 2017, pp. 285–293.
[3] H. Inui, K. Miyatake, Y. Nakano, S. Kitaoka, Wax ester fermentation in Euglena
gracilis, FEBS Lett. 150 (1) (1982) 89–93.
[4] H. Inui, K. Miyatake, Y. Nakano, S. Kitaoka, Production and composition of wax
esters by fermentation of Euglena gracilis, Agric. Biol. Chem. 47 (11) (1983)
2669–2671.
[5] A.K. Lee, D.M. Lewis, P.J. Ashman, Disruption of microalgal cells for the extraction
of lipids for biofuels: processes and specific energy requirements, Biomass
Bioenergy 46 (2012) 89–101.
[6] K. Sakuragi, P. Li, N. Aoki, M. Otaka, H. Makino, Oil recovery from wet Euglena
gracilis by shaking with liquefied dimethyl ether, Fuel Process. Technol. 148 (2016)
143
Biomass and Bioenergy 112 (2018) 138–143
184–187.
[7] V. Hariram, R. Bharathwaaj, Extraction and optimization of biodiesel yield from
wax esters of Apis melifera (honey bee), Int. J. ChemTech Res. 8 (9) (2015)
433–437.
[8] L. Canoira, R. Alcántara, M. Jesús García-Martínez, J. Carrasco, Biodiesel from
Jojoba oil-wax: transesterification with methanol and properties as a fuel, Biomass
Bioenergy 30 (1) (2006) 76–81.
[9] N. Aoki, H. Matsuda, H. Kato, A. Yoneda, R. Arashida, Y. Marukawa, Method for
Producing Fuel Oil Base, (January 8, 2015) US 2015/0011784 A1.
[10] G.W. Huber, A. Corma, Synergies between bio- and oil refineries for the production
of fuels from biomass, Angew. Chem. Int. Ed. 46 (38) (2007) 7184–7201.
[11] G.W. Huber, S. Iborra, A. Corma, Synthesis of transportation fuels from biomass:
Chemistry, catalysts, and engineering, Chem. Rev. 106 (9) (2006) 4044–4098.
[12] Y.K. Ong, S. Bhatia, The current status and perspectives of biofuel production via
catalytic cracking of edible and non-edible oils, Energy 35 (1) (2010) 111–119.
[13] B. Smith, H.C. Greenwell, A. Whiting, Catalytic upgrading of tri-glycerides and fatty
acids to transport biofuels, Energy Environ. Sci. 2 (3) (2009) 262–271.
[14] N. Taufiqurrahmi, S. Bhatia, Catalytic cracking of edible and non-edible oils for the
production of biofuels, Energy Environ. Sci. 4 (4) (2011) 1087–1112.
[15] E.T.C. Vogt, B.M. Weckhuysen, Fluid catalytic cracking: recent developments on the
grand old lady of zeolite catalysis, Chem. Soc. Rev. 44 (20) (2015) 7342–7370.
[16] I. Shimada, S. Kato, N. Hirazawa, Y. Nakamura, H. Ohta, K. Suzuki, T. Takatsuka,
Deoxygenation of triglycerides by catalytic cracking with enhanced hydrogen
transfer activity, Ind. Eng. Chem. Res. 56 (1) (2017) 75–86.
[17] G. de la Puente, U. Sedran, Conversion of methylcyclopentane on rare earth ex-
changed Y zeolite FCC catalysts, Appl. Catal., A 144 (1–2) (1996) 147–158.
[18] M.C. Galiano, U.A. Sedran, Light alkene selectivity on Y zeolite FCC catalysts, Ind.
Eng. Chem. Res. 36 (10) (1997) 4207–4211.
[19] M.A. Sanchez-Castillo, R.J. Madon, J.A. Dumesic, Role of rare earth cations in Y
zeolite for hydrocarbon cracking, J. Phys. Chem. B 109 (6) (2005) 2164–2175.
[20] D. Wallenstein, K. Schäfer, R.H. Harding, Impact of rare earth concentration and
matrix modification in FCC catalysts on their catalytic performance in a wide array
of operational parameters, Appl. Catal., A 502 (2015) 27–41.
[21] K. Ichihara, Y. Fukubayashi, Preparation of fatty acid methyl esters for gas-liquid
chromatography, J. Lipid Res. 51 (3) (2010) 635–640.
[22] I. Shimada, R. Imai, Y. Hayasaki, H. Fukunaga, N. Takahashi, T. Takatsuka,
Increasing octane value in catalytic cracking of n-hexadecane with addition of *BEA
type zeolite, Catalysts 5 (2) (2015) 703–717.
[23] J.C. Sternberg, W.S. Gallaway, D.T.L. Jones, The mechanism of response of flame
ionization detectors, in: N. Brenner, J.E. Callen, M.D. Weiss (Eds.), Gas chromato-
graphy, Academic Press, New York and London, 1962, pp. 269–285.
[24] B.C. Gates, J.R. Katzer, G.C.A. Schuit, Chemistry of Catalytic Processes, McGraw-
Hill, New York, 1979.
[25] O.V. Potapenko, V.P. Doronin, T.P. Sorokina, O.V. Krol, V.A. Likholobov, A study of
intermolecular hydrogen transfer from naphthenes to 1-hexene over zeolite cata-
lysts, Appl. Catal., A 516 (2016) 153–159.
[26] M. Bernabei, R. Reda, R. Galiero, G. Bocchinfuso, Determination of total and
polycyclic aromatic hydrocarbons in aviation jet fuel, J. Chromatogr. A 985 (1)
(2003) 197–203.
[27] A. Anwar, A. Garforth, Challenges and opportunities of enhancing cold flow prop-
erties of biodiesel via heterogeneous catalysis, Fuel 173 (2016) 189–208.
[28] D.A. Saldana, B. Creton, P. Mougin, N. Jeuland, B. Rousseau, L. Starck, Rational
formulation of alternative fuels using QSPR methods: application to jet fuels, Oil
Gas Sci. Technol. 68 (4) (2013) 651–662.
[29] T.G. Smagala, E. Christensen, K.M. Christison, R.E. Mohler, E. Gjersing,
R.L. McCormick, Hydrocarbon renewable and synthetic diesel fuel blendstocks:
composition and properties, Energy Fuels 27 (1) (2013) 237–246.
[30] C.D. Hurd, F.H. Blunck, The pyrolysis of esters, J. Am. Chem. Soc. 60 (10) (1938)
2419–2425.
[31] R. Taylor, The transition state in ester pyrolysis. Part 9. On the 'surface-catalysed'
mechanism for the elimination, J. Chem. Soc. Perkin Trans. 2 (12) (1979)
1730–1737.
[32] R. Černý, M. Kubů, D. Kubička, The effect of oxygenates structure on their deox-
ygenation over USY zeolite, Catal. Today 204 (2013) 46–53.
[33] I. Shimada, Y. Nakamura, H. Ohta, K. Suzuki, T. Takatsuka, Co-processing of sa-
turated and unsaturated triglycerides in a catalytic cracking process for hydro-
carbon fuel production, J. Chem. Eng. Jpn. (2018), http://dx.doi.org/10.1252/jcej.
17we187.
[34] R.W. Gosselink, S.A.W. Hollak, S.W. Chang, J. van Haveren, K.P. de Jong,
J.H. Bitter, D.S. van Es, Reaction pathways for the deoxygenation of vegetable oils
and related model compounds, ChemSusChem 6 (9) (2013) 1576–1594.
[35] I. Kubičková, D. Kubička, Utilization of triglycerides and related feedstocks for
production of clean hydrocarbon fuels and petrochemicals: a review, Waste Biomass
Valorization 1 (3) (2010) 293–308.
[36] C.-C. Chang, S.-W. Wan, China's motor fuels from tung oil, Ind. Eng. Chem. 39 (12)
(1947) 1543–1548.
[37] S.C. Moldoveanu, Pyrolysis of various derivatives of carboxylic acids, in:
S.C. Moldoveanu (Ed.), Pyrolysis of Organic Molecules with Applications to Health
and Environmental Issues, Elsevier, Amsterdam, 2010, pp. 579–627.
[38] K. Kitamura, Studies of the pyrolysis of triglycerides, Bull. Chem. Soc. Jpn. 44 (6)
(1971) 1606–1609.
[39] A. Corma, G.W. Huber, L. Sauvanauda, P. O'Connor, Biomass to chemicals: catalytic
conversion of glycerol/water mixtures into acrolein, reaction network, J. Catal. 257
(1) (2008) 163–171.