See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/251581597

Comparative studies of thermochemical liquefaction

characteristics of microalgae using different organic solvents
ARTICLE in FUEL AND ENERGY ABSTRACTS NOVEMBER 2011
DOI: 10.1016/j.energy.2011.09.031

CITATIONS

READS

29

61

8 AUTHORS, INCLUDING:
Xing-Zhong Yuan

Jingyu Wang

Hunan University

Huazhong University of Science and Technology

134 PUBLICATIONS 1,535 CITATIONS

27 PUBLICATIONS 547 CITATIONS

SEE PROFILE

SEE PROFILE

Huajun Huang

Zhifeng Liu

Jiangxi Agriculture University

Hunan University

42 PUBLICATIONS 533 CITATIONS

35 PUBLICATIONS 493 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate,

letting you access and read them immediately.

SEE PROFILE

Available from: Huajun Huang

Retrieved on: 11 December 2015

Energy 36 (2011) 6406e6412

Contents lists available at SciVerse ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Comparative studies of thermochemical liquefaction characteristics of microalgae

using different organic solvents
Xingzhong Yuan a, b, *, Jingyu Wang a, b, Guangming Zeng a, b, Huajun Huang a, b, Xiaokai Pei a, b, Hui Li a, b,
Zhifeng Liu a, b, Minghui Cong c
a
b
c

College of Environmental Science and Engineering, Hunan University, Changsha 410082, PR China
Key Laboratory of Environment Biology and Pollution Control (Hunan University), Ministry of Education, Chang 410082, PR China
College of Chemical Engineering, Shandong University of Science and Technology, Qingdao 266510, PR China

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 15 March 2011
Received in revised form
9 September 2011
Accepted 16 September 2011
Available online 15 October 2011

The effect of different organic solvents, such as methanol, ethanol and 1,4-dioxane, on thermochemical
liquefaction characteristics of Spirulina (a kind of high-protein microalgae) was systematically studied.
The liquefaction experiments were conducted in a 1000 mL autoclave at different temperatures from 573
to 653 K with a xed solid/liquid ratio. Liquefaction of Spirulina processed in methanol and ethanol
favored the conversion rate and bio-oil yield compared with that in 1,4-dioxane solvent. The bio-oil
generated in methanol contained higher C and H concentrations but a lower O content, resulting in
a higher caloric value (39.83 MJ/kg). The results of FT-IR (Fourier Transform Infrared Spectroscopy) and
GCeMS (Gas ChromatographyeMass Spectroscopy) analyses indicated that the compositions of bio-oil
products were greatly affected by the type of solvent used for the liquefaction process. The major
component of bio-oil produced with methanol was hexadecanoic acid methyl ester (C17H34O2, 35.53%).
However, ethanol favored the formation of hexadecanoic acid ethyl ester (C18H36O2, 26.27%). When
Spirulina were operated with 1,4-dioxane, the bio-oil was dominated by hexadecanenitrile (C16H31N,
22.7%). The presence of methanol and ethanol might promote the formation of esters. Low-boiling-points
compounds with phenol ring structure or heterocyclics can be generated when 1,4-dioxane was
employed as solvent.
2011 Elsevier Ltd. All rights reserved.

Keywords:
Organic solvent
Microalgae
Sub- and supercritical liquefaction
Bio-oil

1. Introduction
Nowadays, alternative fuels from biomass have been explored
worldwide due to the shortage of traditional fossil fuels and associated environmental problems. The utilization of biofuels can contribute to the reduction of greenhouse gases emissions because of its
carbon-neutral life-cycle [1]. The rst generation biofuel could be
produced from food crops (e.g., sugar beet, sugarcane, wheat, corn,
etc.) through anaerobic fermentation [2]. It is not feasible to produce
bioenergy considering the competition for food supply. The next
generation biofuels are produced by lignocellulosic agriculture and
forest residues such as rice straw [3] and woody biomass [4].
However, the possibility to develop the second generation biofuel
production is restricted to land availability. In recent years, much

* Corresponding author. College of Environmental Science and Engineering,

Hunan University, Changsha 410082, PR China. Tel.: 86 731 88821413; fax: 86
731 88823701.
E-mail address: [email protected] (X. Yuan).
0360-5442/$ e see front matter 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.energy.2011.09.031

thrust has been put on the production biofuel from microalgae. The
fast growing microalgae have a high capacity for carbon dioxide
xation and can be grown in intensive culture on limited areas [5,6].
Therefore, microalgae have been recognized as attractive feedstock
for the third generation biofuel.
A signicant amount of work has been reported on the potential
use of microalgae to produce biofuel. In these articles, the technologies are mainly focused on pyrolysis and liquefaction [7e10].
The pyrolysis process requires a relatively dry biomass but the
latter is tolerant to high moisture content and so is ideally suited to
biomass from an aquatic origin such as microalgae [10]. Furthermore, the liquefaction process has many advantages such as (1) the
presence of solvent could dilute the concentration of the products
and prevent the cross-linked reactions between hydrocarbon and
aromatics compounds generating tar compounds and (2) relative
low reaction temperature (less energy consumption) in comparison
with pyrolysis and gasication [11].
The products of microalgae liquefaction depend on numerous
parameters, including solvent type, substrate type, heating condition, reactor conguration and catalyst. In particular, the type of

X. Yuan et al. / Energy 36 (2011) 6406e6412

solvent has remarkable effect on the liquefaction reaction. Some

researchers have reported the liquefaction characteristics of microalgae in sub- and supercritical water [9,10,12e15]. However, the
critical point of water (647.3 K, 22.1 MPa) means that the liquefaction
process was operated under challenging conditions. Moreover, the
bio-oil produced in water contained higher oxygen concentration
thus cutting down its quality and storage stability. To enhance the
yield of bio-oil with lower oxygen content, a few researchers investigated the liquefaction characteristics of microalgae in sub- and
supercritical organic solvents, such as 1-methylnaphthalene and
toluene [15], ethylene glycol [16], ethanol [17], and methanol [18].
The presence of organic solvents, whose critical points are much
lower than that of water, could succeed in converting microalgae into
bio-oils with lower oxygen content.
As mentioned above, solvolysis liquefaction of microalgae has
been carried out with various organic solvents [15e18]. However,
these investigations were not comparable due to the difference in
the separation of products and denition of the liquid products. To
the best of the authors knowledge, there are few studies carried
out at identical conditions to understand the effect of organic
solvent type on the thermochemical liquefaction characteristics of
microalgae. In the present paper, Spirulina were chosen as the renewable energy source and three organic reagents (methanol, ethanol
and 1,4-dioxane) were adopted as the liquefaction solvent. The
inuences of solvent type on thermochemical liquefaction characteristics of Spirulina were investigated in the same autoclave at
different temperatures from 573 to 653 K with a xed solid/liquid
ratio (solid: Spirulina, liquid: solvent) 0.075 g/mL for 20 min. In
addition, the physical and chemical properties of bio-oils obtained
with different liquefaction solvents were also compared through
various analyses, such as elemental analysis, viscosity, acid value,
density, FT-IR and GCeMS.

6407

dried microalgae and 200 mL organic solvent (methanol, ethanol,

1,4-dioxane, respectively) were loaded into the reactor. The reactants were agitated vertically using a stirrer (75 rpm). The autoclave
was heated to the desired temperature at 5 K/min by an external
electric furnace. Then the temperature was maintained for 20 min.
After each run, the reactor was cooled down to room temperature
by cool water.
The procedure for separation of liquefaction products is illustrated in Fig. 1. The liquid and solid products were removed from
the autoclave into a beaker. Then the wall of the autoclave was
washed three times with 150 mL acetone (3  50 mL). The products
in the mixture were ltered by vacuum and acetone was added to
wash the fraction remained. The acetone soluble fraction was
designated as bio-oil. The acetone insoluble fraction was dened as
residue after dried at 378 K until constant weight. The extractant
acetone was evaporated under reduced pressure at 343 K using
a rotary evaporator. The conversion rate and bio-oil yield were
calculated by the following formulas.


Conversion rate wt:%


1  mass of residue
 100%
mass of microalgae


mass of bio-oil
 100%
mass of microalgae


Bio-oil yield wt:%

Three duplicate runs were performed under nominally identical

conditions for all the measurements to ensure the repeatability of
the results. The standard deviations were also calculated and reported as experimental uncertainties. The maximum standard
deviations for conversion rate and bio-oil yield were both less
than 5%.
2.3. Analysis

2. Experimental
The density, viscosity and acid value of bio-oils were determined
according to ASTM standard methods [6,19]. Elemental compositions of bio-oil were analyzed by CHNOS Elemental Analyzer Vario

2.1. Materials
The dried microalgae cells of Spirulina were purchased from
Xigema Biological Technology Co., Ltd. (Fujian, China). Samples
were ground and screened into portions of particle diameter
between 0.2 and 0.9 mm, and then stored in a desiccator. The
protein, lipid and carbohydrate of microalgae strains were determined by J. Waterborge method [9], Soxhlet extract method and
phenol sulfuric acid method [12], respectively. The elemental
compositions of liquefaction feedstock were measured by an
elemental analyzer. The analytical results of microalgae are listed in
Table 1. Three different organic solvents including methanol,
ethanol and 1,4-dioxane (analytical grade), were supplied by
Tianjin Chemicals Company (Tianjin, China).

Mixing
Microalgae

Solvent

Liquefaction

Gas products

Liquid and solid products

2.2. Experimental procedure and separation

Extraction

All liquefaction experiments were performed in a 1000 mL

GSHA-1 type autoclave with a magnetic stirrer. In a typical run, 15 g
Table 1
Analytic results of Spirulina.
Chemical
composition (wt.%)

Organic matter
composition (wt.%)

Elemental
compositions (wt.%)

Moisture
Ash
Volatiles

Protein
Lipid
Carbohydrate

C
H
Oa
N
S

4.3
10.9
89.1

Determined by difference.

69.4
8.2
22.4

42.96
8.49
39.19
8.87
0.49

Acetone soluble fraction

Evaporation
Bio-oil

Acetone insoluble fraction

Drying
Residue

Fig. 1. Procedure for separation of liquefaction products.

6408

X. Yuan et al. / Energy 36 (2011) 6406e6412

EL III (Elementar Analysensysteme GmbH, Germany). The caloric

value was determined using Dulong equation (3), where C, H and O
are the contents of carbon, hydrogen and oxygen in bio-oil,
respectively.

Caloric valueMJ=kg 0:3383C 1:442H  O=8

(3)

The FT-IR spectra for raw material and bio-oil fractions were
conducted on a WQF-410 type FT-IR spectrometer to examine the
major organic components based on the absorbance peaks of the
functional groups. The GCeMS analysis was carried out using
a Trace GC, Palaris Q GCeMS spectrometer (Thermo-Finnigan, USA)
with a carbon capillary column, DB-1 (30 m  0.25 mm  0.25 mm).
The ow rate of carrier gas He was 1 mL/min. The column
temperature of GC was programmed from 463 to 543 K with an
increasing rate of 8 K/min. The temperatures of transfer line and
injection chamber were 543 and 553 K, respectively. Mass range
was 40e500 m/z.
3. Results and discussion
3.1. Effect of solvent type on conversion rate

3.2. Effect of solvent type on bio-oil yield

Fig. 3 presents the effect of solvent type on the yields of bio-oil.
The trends of bio-oil yield lines varied between different reagents.
According to this gure, when 1,4-dioxane was adopted as liquefaction solvent, the bio-oil yield increased gradually when the
temperature increased from 573 to 653 K. But in the case of
methanol and ethanol, the yields of bio-oil leveled off at temperatures higher than 633 K. When the temperature was relative low,
the bio-oil yields obtained in methanol and ethanol were higher

84

58

82

56

80

54
Bio-oil yield (wt%)

Conversion rate (wt%)

The effect of solvent type on conversion rate of Spirulina is

depicted in Fig. 2. As clearly shown from the gure, the conversion
rate increased continuously with raising temperatures. Meanwhile,
it was noted that the type of solvent had an obvious inuence on
the conversion rate, which was in good agreement with previous
literatures conducted on the effect of solvent type on the liquefaction of lignocellulosic biomass [11,20e22]. Among the three
tested solvents, methanol had the best promotion inuence on the
conversion of Spirulina followed by ethanol and 1,4-dioxane,
respectively. The maximum conversion rates were all achieved at
653 K, 82 wt.% for methanol, 80 wt.% for ethanol, and 75 wt.% for
1,4-dioxane.
It was found that decomposition and repolymerization reactions
were both involved during the liquefaction process [23]. The high
molecules were decomposed and depolymerized into smaller
molecules by cleavage, dehydration, decarboxylation and deamination forming intermediates. Many of these fragments are
unstable and can be recombined into larger ones [24]. The increase
of reaction temperature could enhance the decomposition of Spirulina, leading to a signicant increment of conversion rate.

During the liquefaction process, depolymerization reactions

could be achieved by solvolysis of main macromolecules. Behrent
et al. proposed that the depolymerization of biomass strongly
depended on the used solvent. And the interactions between
solvent and substrate would be increased when the tested solvent
tended to be a product derivable from the biomass itself (e.g.,
alcohols, acids, phenol and its derivatives) [22]. Compared with
adopting 1,4-dioxane as solvent, higher conversion rates of Spirulina using methanol and ethanol were obtained, which further
illustrated Behrents view. Alcoholic solvents could react with the
degradation products of high molecules, which can be conrmed
by the analysis of chemical compositions in bio-oil products elaborated in Section 3.3.3.
Moreover, this decomposition process proceeded via a range of
reactions, such as cracking, isomerization, defragmentation, and
rearrangement. These reactions are almost related to the polarity of
liquefaction solvent [21]. Methanol and ethanol are polar protic
solvents, but 1,4-dioxane is dipolar aprotic solvent. The polarities of
the three tested solvents are sequenced as: methanol (76.2) > ethanol
(65.4) > 1,4-dioxane (16.4). The polarity might play an important role
in determining the conversion process of biomass liquefaction [20].
According to the comparisons of conversion rates using different
solvents, polar protic solvents could promote the conversion of
Spirulina better than dipolar aprotic solvent. Among the three tested
solvents, methanol has the highest polarity. Therefore, the conversion
rate of Spirulina liqueed in methanol was higher than that of
microalgae liqueed in the other two solvents. There were similar
trends in the conversion rate curves in the case of methanol and
ethanol for their similar polarity values. However, 1,4-dioxane with
a lower polarity gave lower conversion rate.

78
76
74
72

52
50
48
46

70

44

68

42
66

40
570

580

590

600

610

620

630

640

650

660

Temperature (K)

Fig. 2. Effect of solvents on the conversion rate as a function of the temperature with
Spirulina 15 g, organic solvent 200 mL, holding time 20 min. C: methanol, -: ethanol,
:: 1,4-dioxane.

570

580

590

600

610

620

630

640

650

Temperature (K)

Fig. 3. Effect of solvents on the bio-oil yield as a function of the temperature with
Spirulina 15 g, organic solvent 200 mL, holding time 20 min. C: methanol, -: ethanol,
:: 1,4-dioxane.

X. Yuan et al. / Energy 36 (2011) 6406e6412

than that obtained in 1,4-dioxane. Meanwhile, higher yields of biooil could be achieved with ethanol than that with methanol at
higher reaction temperatures.
Possible reasons to account for the improvement of bio-oil
yields at higher temperatures would be: (1) formation of oil fractions through the repolymerization of fragments (derived from the
decomposition of Spirulina) and (2) depolymerization of solids and
aggregation of gases to bio-oils. The higher the reaction temperature was, the easier the fragments were decomposed into a liquid
oil-rich phase [22]. When the temperature was increased from 573
to 653 K, the density of solvent would be increased. This means that
the dissolving ability of solvents became stronger. More products
would be dissolved into liquid phase as the density increased.
Consequently, the formation of the bio-oil was enhanced with the
increment of temperatures.
As described above, the different densities of solvents were most
likely one cause of the different liquefaction behaviors. Furthermore,
the polar protic solvent (methanol and ethanol) contains hydrogen
bonds. Hence they could act as hydrogen-donor solvents [25]. When
methanol or ethanol was employed, the intermolecular hydrogen
bonds might be decomposed into hydrogen free radical (H) which is
reactive for the hydrocracking of long-chain polymers (such as
protein, lipid and carbohydrate) into low molecular fragments, and
can be an effective stabilization agent, preventing the repolymerization reactions of small biomass-derived fragments/intermediates
to form solid residues [1]. In addition, hydrogen transfer reactions
between liquefaction solvent and free radicals or intermediates were
promoted by the sub- and supercritical methanol and ethanol.
Huang et al. also found that the free radicals and intermediates from
the decomposing of Spirulina would react with supercritical ethanol
to produce bio-oil products. These reactions prevented the formation of residue and improved the bio-oil yield [17]. As regard to 1,4dioxane having no capability to donate hydrogen free radical, those
intermediates generated during the liquefaction process might
repolymerize to larger compounds, thus the reactions of residue
formation were promoted.
Furthermore, the solvent would cause the swelling of Spirulina
that resulted in an increase in internal surface area and separation
of structural linkages between high weight molecules. As a result,
the intermolecular interactions of polymeric chains were weakened which promoted cleavage of macromolecules or repolymerization of fragments into bio-oils. However, various solvents had
different effects on swelling Spirulina.
Attention should also be paid to the fact that the bio-oil yield
changes curve of Spirulina obtained in 1,4-dioxane was found to be
quite different from that obtained in methanol and ethanol. The
reason may be that the density of 1,4-dioxane was higher than that
of methanol and ethanol, which could accelerate the solvolysis of
Spirulina. Apart from that, the molecular structure of 1,4-dioxane
has one oxygen atom more than that of methanol and ethanol.
However, the real reason for this phenomenon remains to be
discovered.

6409

Table 2
Comparison results of physicochemical properties between fossil oil and microalgae
bio-oils.
Properties

Unit

Bio-oilsa
Methanol

Acid value
mgKOH/g
Densityd
g/cm3
pa$s
Viscosityd
Elemental analysis
C
wt.%
H
wt.%
e
wt.%
O
N
wt.%
S
wt.%
Caloric value
MJ/kg
a
b
c
d
e

Fossil oilb
Ethanol

1,4-dioxane

0.9
1.25
0.42

0.7
0.98
0.94

1.8
1.02
0.65

ec
0.75e1.0
2e1000

69.62
12.53
9.91
7.46
0.48
39.83

68.72
11.83
11.04
7.92
0.49
38.32

66.01
11.56
12.43
9.54
0.46
36.76

83.0e87.0
10.0e14.0
0.05e1.5
0.01e0.7
0.05e5.0
42

Obtained using methanol, ethanol and 1,4-dioxane, respectively.

The data were taken from Miao et al. [8].
Not specied.
Density and viscosity were measured at 313 K.
Determined by difference.

For the application of microalgae bio-oil, the acid value should

be as low as possible and not exceed 0.5 mgKOH/g [19]. As can be
seen in Table 2, the bio-oils produced from microalgae had higher
acid values of 0.7e1.8 mgKOH/g. These values were much lower
than that of bio-oil from fast pyrolysis of cellulose materials [26],
though there is no information available on the value of fossil oil.
Compared with fossil oil, the microalgae bio-oils had slightly higher
density between 0.98 and 1.25 g/cm3, relative lower viscosity of
0.4e0.9 pa$s and lower caloric value of 38.3 MJ/kg on average.
Table 2 also reports the elemental compositions of microalgal
bio-oils produced using different reagents. Compared with the

3.3. Effect of solvent type on characterization of bio-oil

The physical and chemical properties of bio-oil products are of
particular interest as they are important for the usage efciency
and application of microalgae bio-oil.
3.3.1. Physicochemical analysis
The elemental analysis and caloric value along with other
properties of bio-oils obtained at 633 K using different organic
solvents were analyzed. The results are summarized in Table 2. For
a comparative purpose, relevant properties of fossil oil are also
listed in Table 2 [8].

Fig. 4. FT-IR analysis of raw material and bio-oils obtained at 633 K with Spirulina 15 g,
organic solvent 200 mL, holding time 20 min using different organic solvents. (a) raw
material, (b) methanol, (c) ethanol, (d) 1,4-dioxane.

6410

X. Yuan et al. / Energy 36 (2011) 6406e6412

composition of raw material, it was a satisfying upgrading that the

oxygen content in bio-oils was reduced from 39.19 wt.% to a range
of 9.91e12.43 wt.% and the caloric value was greatly increased.
Among the three tested solvents, the bio-oil obtained in methanol
had higher carbon and hydrogen concentrations but lower oxygen
content compared with those of bio-oils produced in ethanol and
1,4-dioxane, resulting in a higher caloric value of 39.83 MJ/kg.
The reduction of oxygen fractions in bio-oils was probably
ascribed to de-oxygenation reactions during the liquefaction of
biomass including (1) reduction of oxygen in the form of CO and
CO2, and (2) hydrogen-donor reduction of oxygen combined with
hydrogen to produce H2O [17,25]. The polar protic solvent, such as
methanol and ethanol, could act as hydrogen-donor solvent and
promote the formation of hydrogen free radical and dehydration
reactions. Therefore, the bio-oils produced by methanol or ethanol
could result in a lower content of oxygen compared with bio-oil
obtained in 1,4-dioxane.
However, a fact should not be ignored that the content of oxygen
and nitrogen in microalgal bio-oil were still higher than that of
fossil oil. In addition, the high content of nitrogen was due to the
high-protein fraction in Spirulina. NOx pollutions compounds
would form after the burning of bio-oil products, which was

undesirable for environmental reasons. The further removal of

oxygen and nitrogen in bio-oils is necessary for their application.
Rening processes of bio-oil should be investigated and studies
focusing on stability of bio-oils should also be carried out.
3.3.2. FT-IR analysis
The FT-IR spectra of raw Spirulina and bio-oils obtained using
different organic solvents (methanol, ethanol and 1,4-dioxane) at
633 K are shown in Fig. 4. Overall, there are quite similar features
on the chemical structures of bio-oils detected by FT-IR spectroscopy. The greater intensity and higher resolution of bands at
3000e2800 cm1 suggested bio-oils to be more highly aliphatic in
character. The band at 1360 cm1 attributed to CeN stretching
vibration in the spectra of raw Spirulina was disappeared in the
spectra of bio-oils. This might show that the liquefaction of Spirulina proceeded mainly by the scission of peptide linkages [15].
The C]O stretching vibration centered at 1701 cm1 and CeO
vibration bonds between 1210 and 1160 cm1 indicated the existence of esters in bio-oil fractions. Absorption at 1600 cm1 and
1400 cm1 were contributed to aromatic groups (1600 cm1 for
the aromatic ring backbone vibration) and represented more
benzene substitutions. Furthermore, the broad band at 3290 cm1

Fig. 5. Total ion chromatograms of bio-oils obtained at 633 K with Spirulina 15 g, organic solvent 200 mL, holding time 20 min using different organic solvents. (a) methanol,
(b) ethanol, (c) 1,4-dioxane.

X. Yuan et al. / Energy 36 (2011) 6406e6412

was ascribed to combination and overlap of absorbance peaks of

aromatic and aliphatic OeH, CeH on the aromatic ring and NeH
stretching vibrations [11].
However, some distinct differences could also be observed
when comparing the spectra of bio-oils produced with different
solvents. Among the FT-IR spectra, the bio-oil obtained using
methanol as solvent supplied the simplest spectra with the fewest
functional groups, which implied that the presence of methanol
during the decomposition process could produce more selective
and controllable products. No absorption bands belonging to
phenols were detected in the bio-oil produced by methanol. In the
treatments with methanol and ethanol, more intensive C]O
stretching vibration at 1701 cm1 and CeH absorbance peak at
3000e2800 cm1 were found to be large portion of aliphatic esters
in the bio-oil products. The relatively strong bands between 3300
and 3000 cm1 for the bio-oils obtained using ethanol and 1,4dioxnae were assigned to OeH stretching vibration for phenols.
The strong bands between 1383 and 1275 cm1 in the bio-oils
produced using ethanol and 1,4-dioxane might be attributed to
heterocyclic compounds.
3.3.3. GCeMS analysis
The bio-oils obtained at 633 K using different solvents (methanol, ethanol and 1,4-dioxane) were also analyzed by GCeMS.
Hundreds of peaks were detected in the total ion chromatograms of
bio-oils (Fig. 5), implying that the bio-oils were quite complex
mixtures of organic compounds. The tentative identication of
GCeMS peaks was performed using the MS search le (NIST
library). Table 3 only shows the compounds with more than 0.8% of
the total area (dened by the percentage of the compounds chromatographic area out of the total area). The total area for all the
listed compounds is also presented in the table.
Spirulina has a high content of protein as shown in Table 1. Thus
it is important to understand the degradation mechanism of

6411

protein as in the case of liquefaction of microalgae Spirulina. The

liquefaction of protein is a result of the rupture of CeN peptide
linkages between the carboxyl and amine groups of amino acids,
which are the building blocks of protein. And the main reactions are
deamination and decarboxylation reactions [10,24]. In this study,
sub- and supercritical conditions will allow these degradation
products recombine to long-chain hydrocarbons, phenols and
nitrogen heterocycles such as indole and pyrrole derivatives.
As shown in Table 3, the major compounds of bio-oils were
phenols, alkanes and alkenes, esters (long-chain or aromatic
esters), organic acids, nitriles and nitrogen heterocycles (indoles
and pyridines). This was consistent with results reported by Ross
et al. who investigated hydrothermal liquefaction of microalgae
including Chlorella vulgaris and Spirulina using alkali and organic
acids as catalyst. They also found that heterocycles and phenolic
derivatives were identied in the bio-oil products and the higher
protein containing Spirulina resulted in higher amounts of indoles
and pyridines [10].
Fig. 5 illustrates that compounds with medium-boiling-points
(retention time less than 29 min) were greatly more than compounds
with low-boiling-points (retention time less than 18 min) irrespective of which type of solvent was used. The bio-oil produced with
methanol as solvent had fewer major peaks than those obtained with
ethanol and 1,4-dioxane as solvent. This indicated that the distribution of components in bio-oil produced with methanol comparatively narrowed. Liquefaction process using methanol produced
a large numbers of esters (hexadecanoic acid methyl ester, 35.53%;
octadecanoic acid methyl ester, 11.18%; linoleic acid ethyl ester,
10.4%). The most abundant compounds of bio-oil produced in
ethanol were also esters (hexadecanoic acid ethyl ester, 26.27%;
linoleic acid ethyl ester, 4.82%; heptadecanoic acid 15-methyl-ethyl
ester, 4.73%). On the contrary, the bio-oil obtained with 1,4-dioxane
was dominated by nitriles (hexadecanenitrile, 22.7%; octadecanenitrile, 5.02%). Organic solvents could act as substrates reacting with

Table 3
Major compounds of bio-oils.
No.

RT (min)

Name of compound

Solvents (area %)
Methanol

Ethanol

1,4-dioxane

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26

10.62
11.38
12.02
12.38
13.53
13.93
13.99
14.03
14.86
15.43
15.90
16.46
17.25
18.58
20.16
20.87
20.88
21.13
21.53
21.86
23.15
23.32
23.84
24.01
24.04
24.56

Phenol, 4-methyl-(C7H8O)
Pyridine, 5-ethyl-2-methyl-(C8H11N2)
Pyrrole, 2-ethyl-3,5-dimethyl-1H-(C8H13N)
Phenol, 2,4-dimethyl-(C8H10O)
Phenol, 3-ethyl-5-methyl-(C9H12O)
Pentanedioic acid, diethyl ester (C9H16O4)
Benzenepropanoic, methyl ester (C10H12O2)
Benzenepropanoic, ethyl ester (C11H14O2)
4-(2,5-Diohyro-3-methoxyphenyl) butylamine (C11H19NO)
1H-Indole, 3-methyl-(C9H9N)
1H-Indole, 5,7-dimethyl-(C10H11N)
Benzonitrile, 2,4,6-trimethyl-(C10H11N)
1H-Indole, 5,6,7-trimethyl-(C11H13N)
Butanoic acid, 3-methyl-, 3,7-dimethyl-2, 6-octadienyl ester (C15H26O2)
1-Naphthalenol, 4a, 8-dimethyl-2-(2-propenyl)-(C15H24O)
9-Hexadecanoic acid (C16H32O2)
Hexadecanenitrile (C16H31N)
Hexadecanoic acid, methyl ester (C17H34O2)
Nonanedioic acid, dibutyl ester (C17H32O4)
Hexadecanoic acid, ethyl ester (C18H36O2)
Octadecanenitrile (C18H35N)
Octadecanoic acid, methyl ester (C19H38O2)
Linoleic acid, ethyl ester (C20H36O2)
7, 10, 13-Eicosatrienoic acid, methyl ester (C21H36O2)
Heptadecanoic acid, 15-methyl-, ethyl ester (C20H40O2)
8-Octadecanoic acid, methyl ester (C19H36O2)

ea
e
e
e
e
e
0.91
e
e
0.84
0.92
e
e
0.97
1.39
2.75
e
35.53
2.63
e
e
11.18
10.4
0.86
e
e

e
2.51
0.95
e
2.34
1.2
e
1.89
2.36
1.24
e
0.82
e
e
e
e
1.42
e
e
26.27
e
e
4.82
e
4.73
e

1.53
e
1.08
1.23
4.77
e
e
e
1.43
2.04
1.53
1.1
0.92
2.63
e
6.64
22.7
e
e
0.93
5.02
e
e
e
e
1.77

Total area

68.38

50.55

55.32

Not detected or peak area less than 0.8% of the total area.

6412

X. Yuan et al. / Energy 36 (2011) 6406e6412

the free radicals and intermediates [11]. Amino acids undergo

deamination reaction to produce ammonia and organic acids. When
methanol or ethanol was employed, organic acids might react with
solvent and form methyl- or ethyl-ester compounds. For instance,
hexadecanoic acid methyl ester was produced via esterication
reactions between hexadecanoic acid and methanol. Furthermore,
amino acids also preceded decarboxylation to produce carbonic acid
and amines. Hexadecanenitrile might be formed between hexadecanoic acid and 1,4-dioxane solvent.
In the case of methanol treatment, the low-boiling-points
compounds were not observed except one ester (benzenepropanoic acid methyl ester, 0.91%) and two indoles (3-methyl-1H-indole,
0.84%; 5,7-dimethyl-1H-indole, 0.92%). However, ethanol and 1,4dioxane favored the formation of low-boiling-points compounds
with phenol ring structure or heterocyclics. High amounts of pyridines (5-ethyl-2-methyl-pyridine, 2.51%), phenols (3-ethyl-5methyl-phenol, 2.34%), amines (4-(2,5-Diohyro-3-methoxyphenyl)
butylamine, 2.36%) were detected in the bio-oil produced by ethanol.
Microalgae processed with 1,4-dioxane resulted in high amounts of
pyrroles (2-ethyl-3,5-dimethyl-1H-pyrrole, 1.08%), phenols (4methyl-phenol, 1.53%; 2,4-dimethyl-phenol, 1.23%; 3-ethyl-5methy-phenol, 4.77%), indoles (3-methyl-1H-indole, 2.04%; 5,7dimethyl-1H-indole, 1.53%). These compounds were almost
nitrogen-contenting heterocyclic rings, which were conrmed by
the results of elemental analysis in Table 2.
In summary, the polar protic solvent (methanol, ethanol) and
dipolar aprotic solvent (1,4-dioxane) are able to convert microalgae
into bio-oils and useful chemical materials. And the effects of
organic solvents on characteristic of bio-oil were signicantly
different. The formation of esters was promoted by methanol and
ethanol, while the bio-oil obtained with 1,4-dioxane was dominated by nitriles. The acid and ester derivatives were decomposed
into smaller radicals to produce low-boiling-points compounds
with phenols and heterocyclics in the presence of 1,4-dioxane.
Based on the results presented, the conversion rate and bio-oil
yield increased continuously with raising temperatures. Compared
with 1,4-dioxane, methanol and ethanol are more effective to
convert Spirulina into bio-oils. The results of elemental analysis
showed that bio-oils obtained using methanol and ethanol had
higher carbon and hydrogen contents as well as higher caloric
values. The FT-IR and GCeMS analyses indicated that the components of bio-oils could be strongly affected by the employed
solvents. Liquefaction of Spirulina using methanol and ethanol
produced larger amounts of esters. In addition, the formation of
low-boiling-points compounds with phenol ring structure or
heterocyclics was promoted by 1,4-dioxane.
It is interesting to mention that a plenty of esters were formed
when methanol and ethanol were employed. However, ethanol
appears to be a more promising solvent for Spirulina liquefaction,
compared with other n-alcohols (such as methanol, propanol,
butanol, and pentanol). This is because not only ethanol is effective
for liquefaction, but it is a renewable source derivable from bioconversion of cereals, sugars and lignocellulosic materials. Consequently, ethanol may be the optimum solvent for the degradation
of Spirulina to obtain bio-oils in the three tested solvents.
4. Conclusions
The effect of solvent type (polar protic solvent and dipolar aprotic
solvent) on the liquefaction characteristics of microalgae was
investigated. This study has shown that the different solvent types
could change the conversion rate, bio-oil yield and properties of biooils. Spirulina processed with methanol and ethanol (polar protic
solvent) resulted in higher conversion rate than that obtained using
1,4-dioxane (dipolar aprotic solvent) and simultaneously a large

number of esters were produced. While 1,4-dioxane liquefaction

process favored the formation of nitriles. From the viewpoint of
efciency and reproducible ability, ethanol may be one more
promising solvent for microalgae liquefaction.
Acknowledgments
The research was supported by National Natural Science Foundation of China (No. 50978087) and the Key Project for Science and
Technology Research, Ministry of Education of China (No. 108100).
References
[1] Zhang LH, Champagne P, Xu CB. Bio-crude production from secondary pulp/
paper-mill sludge and waste newspaper via co-liquefaction in hotcompressed water. Energy 2011;36(4):2142e50.
[2] Nigam PS, Singh A. Production of liquid biofuels from renewable resources.
Prog Energy Combust Sci 2011;37(1):52e68.
[3] Li H, Yuan XZ, Zeng GM, Tong JY, Yan Y, Cao HT, et al. Liquefaction of rice straw
in sub- and supercritical 1,4-dioxane-water mixture. Fuel Process Technol
2009;90(5):657e63.
[4] Qian YJ, Zuo CJ, Tian J, He JH. Structural analysis of bio-oils from sub- and
supercritical water liquefaction of woody biomass. Energy 2007;32(3):196e202.
[5] Singh J, Gu S. Commercialization potential of microalgae for biofuels
production. Renew Sust Energy Rev 2010;14(9):2596e610.
[6] Haik Y, Selim M, Abdulrehman T. Combustion of algae oil methyl ester in an
indirect injection diesel engine. Energy 2011;36(3):1823e35.
[7] Zou SP, Wu YL, Yang MD, Li C, Tong JM. Pyrolysis characteristics and kinetics of
the marine microalgae Dunaliella tertiolecta using thermogravimetric
analyzer. Bioresour Technol 2010;101(1):359e65.
[8] Miao XL, Wu QY. High yield bio-oil production from fast pyrolysis by metabolic controlling of Chlorella protothecoides. J Biotechnol 2004;110(1):85e93.
[9] Biller P, Ross AB. Potential yields and properties of oil from the hydrothermal
liquefaction of microalgae with different biochemical content. Bioresour
Technol 2011;102(1):215e25.
[10] Ross AB, Biller P, Kubacki ML, Li H, Lea-Langton A, Jones JM. Hydrothermal
processing of microalgae using alkali and organic acids. Fuel 2010;89(9):
2234e43.
[11] Liu ZG, Zhang FS. Effects of various solvents on the liquefaction of biomass to
produce fuels and chemical feedstocks. Energy Convers Manage 2008;49(12):
3498e504.
[12] Zou SP, Wu YL, Yang MD, Imdad K, Li C, Tong JM. Production and characterization of bio-oil from hydrothermal liquefaction of microalgae Dunaliella
tertiolecta cake. Energy 2010;35(12):5406e11.
[13] Yang YF, Feng CP, Inamori Y, Maekawa T. Analysis of energy conversion characteristics in liquefaction of algae. Resour Conserv Recycl 2004;43(1):21e33.
[14] Sawayama S, Minowa T, Yokoyama SY. Possibility of renewable energy
production and CO2 mitigation by thermochemical liquefaction of microalgae.
Biomass Bioenergy 1999;17(1):33e9.
[15] Matsui T, Nishihara A, Ueda C, Ohtsuki M, Ikenaga N, Suzuki T. Liquefaction of
micro-algae with iron catalyst. Fuel 1997;76(11):1043e8.
[16] Zou YF, Wu YL, Yang MD, Li C, Junmao T. Thermochemical catalytic liquefaction of the marine microalgae Dunaliella tertiolecta and characterization of
bio-oils. Energy Fuels 2009;23:3753e8.
[17] Huang HJ, Yuan XZ, Zeng GM, Wang JY, Li H, Zhou CF, et al. Thermochemical
liquefaction characteristics of microalgae in sub- and supercritical ethanol.
Fuel Process Technol 2011;92(1):147e53.
[18] Patil PD, Gude VG, Mannarswamy A, Deng SG, Cooke P, Munson-McGee S,
et al. Optimization of direct conversion of wet algae to biodiesel under
supercritical methanol conditions. Bioresour Technol 2011;102(1):118e22.
[19] Chen YH, Chen JH, Luo YM, Shang NC, Chang CH, Chang CY, et al. Property
modication of jatropha oil biodiesel by blending with other biodiesels or
adding antioxidants. Energy 2011;36(7):4415e21.
[20] Mazaheri H, Lee KT, Bhatia S, Mohamed AR. Sub/supercritical liquefaction of
oil palm fruit press ber for the production of bio-oil: effect of solvents.
Bioresour Technol 2010;101(19):7641e7.
[21] Yip J, Chen MJ, Szeto YS, Yan SC. Comparative study of liquefaction process
and liqueed products from bamboo using different organic solvents. Bioresour Technol 2009;100(24):6674e8.
[22] Behrent F, Neubauer Y, Oevermann M, Wilmes B, Zobel N. Direct liquefaction
of biomass. Chem Eng Technol 2008;31(5):667e77.
[23] Yuan XZ, Li H, Zeng GM, Tong JY, Xie W. Sub- and supercritical liquefaction of
rice straw in the presence of ethanol-water and 2-propanol-water mixture.
Energy 2007;32(11):2081e8.
[24] Toor SS, Rosendahl L, Rudolf A. Hydrothermal liquefaction of biomass:
a review of subcritical water technologies. Energy 2011;36(5):2328e42.
[25] Xu CB, Etcheverry T. Hydro-liquefaction of woody biomass in sub- and supercritical ethanol with iron-based catalysts. Fuel 2008;87(3):335e45.
[26] Rutkowski P, Kubacki A. Inuence of polystyrene addition to cellulose on
chemical structure and properties of bio-oil obtained during pyrolysis. Energy
Convers Manage 2006;47(6):716e31.