Bioresource Technology 165 (2014) 31–37

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Growth and lipid accumulation properties of microalgal

Phaeodactylum tricornutum under different gas liquid ratios
Mingming Song a, Haiyan Pei a,b,⇑, Wenrong Hu a,b, Fei Han a, Yan Ji a, Guixia Ma a, Lin Han a
a
School of Environmental Science and Engineering, Shandong University, 27 Shanda Nan Road, Jinan 250100, China
b
Shandong Provincial Engineering Centre on Environmental Science and Technology, 17923 Jingshi Road, Jinan 250061, China

h i g h l i g h t s

 The optimum gas liquid ratio (GLR) for cell growth and yielding oil was determined.
 The linear relationships among growth, lipid parameters and GLR were built.
 GLR had important effect on the fatty acid profiles contents.

 HCO3 was a promoter for algae growth at low CO2 concentrations (0.03–0.06%).
2
 CO3 /HCO3 mainly influenced the fatty acids synthesis of C16:1.

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

Article history: In this study, the effects of gas liquid ratio (GLR) on growth, lipid and fatty acid production of Phaeodacty-
Received 9 January 2014 lum tricornutum were investigated. The positive linear relationships among specific growth rate, lipid
Received in revised form 12 March 2014 content, fatty acid methyl ester content and GLR were built. GLR of 1.5 vvm was considered as the opti-
Accepted 15 March 2014
mum GLR for P. tricornutum growth and yielding oil, with highest biomass productivity (227.09 mg L1 -
Available online 26 March 2014
d1), highest lipid productivity (48.48 mg L1 d1) and considerable amount fatty acids of saturated
(50.16%) and monounsaturated (48.79%), which gave the finest compromise between oxidative stability
Keywords:
and cold flow properties. pH variation was good controlled by the air bubbling velocity and algal photo-
Microalgae
Gas liquid ratio
synthesis. CO2 
3 /HCO3 ratio influenced the fatty acids synthesis, especially the component of C16:1. The

Biomass productivity average value of the most favorable composition C16:1 was 48.36%, which is comparable with other
Lipid productivity reports.
Fatty acid profiles Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction was far less than the theoretical maximum of 120–150 g DW m2 -
d1 (Tredici, 2010; Zhu et al., 2008; Boyer, 1982). The low biomass
Recently, microalgae have attracted significantly increased productivity of oil-rich microalgae was one of the main bottlenecks
interest (Song et al., 2013; Hoekman et al., 2012; Mata et al., for the widespread utilization of microalgae-derived biofuels. So to
2010). Compared with other potential feedstock for biofuels, they achieve high oil microalgae biomass is fundamentally important to
have proven to be the dominant substitute for petroleum-based develop the algae feedstock as biodiesel production (Liu et al.,
fuels with good qualities such as high production value, less com- 2013; Stephens et al., 2010; Wijffels and Barbosa, 2010). And care-
petition with food production and friendly effect on the environ- fully control of the algal growth conditions to provide optimum
ment (Liu et al., 2013; Mata et al., 2010). Intensive efforts have nurturing environment seem to be a good way to overcome the
been made to achieve the high productivity potential (Lam and bottleneck.
Lee, 2012; Larkum et al., 2012), although they got the higher bio- Recently, light, nutrients, temperature and CO2 levels have to be
mass productivities of 40 g dry weight (DW) m2 d1 in prevailing explored to enhance the microalgae biomass in the photobioreac-
culture systems (Brennan and Owende, 2010; Mata et al., 2010), it tors (Han et al., 2013; Liu et al., 2013; Ota et al., 2009a). However,
another important factor, gas liquid flow, was reported few. Verti-
cal bubble-columns photobioreactors are often used in experiment
⇑ Corresponding author at: School of Environmental Science and Engineering,
Shandong University, 27 Shanda Nan Road, Jinan 250100, China. Tel./fax: +86 531
with good light–dark cycling, low surface/volume and a much
88392983. more chaotic gas liquid flow (Han et al., 2013; Brennan and
E-mail address: haiyanhup@126.com (H. Pei). Owende, 2010; Eriksen, 2008; Mata et al., 2010). A suitable degree

http://dx.doi.org/10.1016/j.biortech.2014.03.070
0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
32 M. Song et al. / Bioresource Technology 165 (2014) 31–37

of gas liquid flow can lead to a better mixing of the microalga cul- mination of 54 lmol-photons m2 s1 provided by daylight fluo-
ture, which prevents sedimentation, increase the mass transfer, rescent tubes. All the experiments were conducted in triplicate.
contributes to a good interaction between cells and nutrients,
and force the cells to move from dark to light zone (Mata et al., 2.3. Experimental design
2010). However, excessive degree may produce cell damage if
the microalgae are susceptible to hydrodynamic and mechanical The algae strain was precultured in a 500 ml Erlenmeyer to
shear forces (Mata et al., 2010; Eriksen, 2008). The optimum level ensure cells for each experiment in steady-state. In the experiment,
of gas liquid flow (above which cell death occurs) is strain depen- a certain volume of algal inoculums was centrifuged and then re-
dent and should be studied in order to prevent decline in produc- suspended in 10 mL fresh sterilized media. After 2 L of culture
tivity (Eriksen, 2008; Barbosa, 2003). Alias et al. (2004) have medium was sterilized by autoclave (120 °C, 15 min) and cooled
observed that the higher the air flow rate the higher the yield of to room temperature, 10 ml of preculture was added to a photobi-
the reactor. But few report studies the lipid content and fatty acid oreactor. Initial cell concentration was 0.09 ± 0.1 g L1 for each
compositions of the microalgae under different air flow rates. experiment.
Someone have studied the effect of concentration of individual For the experiments on the effects of different gas liquid ratios,
species of inorganic carbon on algae growth rate and proposed a seven air flow rates were set as follows, 0, 0.2, 1, 2, 3, 4 and
growth model that high value of HCO 3 may be good for microalgae 5 L min1, corresponding to 0, 0.1, 0.5, 1, 1.5, 2, 2.5 vvm, where
growth while high CO2 possibly inhibits cell growth (CO2 concen- units of vvm are defined as bubbling gas volumetrically divided
tration of 1–50% in the bubbling gas) (Ota et al., 2009b). However, by culture medium per minute at 298 K and 100 kPa. The air was
they only focused on growth inhibition at the high CO2 concentra- filtered in order to avoid contaminating the microalgae by the
tions and without diffusion limiting conditions due to high gas bacterium or other species of microalgae in the air. The pH of the
flow rates used, other studies that further validation of the model culture medium was measured every 24 h by a pH electrode
at low CO2 concentrations have not been determined. In this study, (PHS – 3C, Rex).
we quantitatively investigated the effect of concentration of indi-
vidual species of inorganic carbon on the algae growth and fatty 2.4. Growth determination of P. tricornutum
acids profiles under low concentration of CO2 in the bubbling air
(0.03–0.06%) with diffusion conditions caused by different gas flow Microalgae growth was measured every 24 h by determining
rates. biomass concentration (g L1) as described in our previous work
Phaeodactylum tricornutum, a potential strain for biofuels pro- (Song et al., 2013). The specific growth rate (d1) of P. tricornutum
duction in terms of our previous paper (Song et al., 2013), was in each experiment was calculated according to the following
selected as a research object in this study as well. The effect of dif- equation (Lim et al., 2010):
ferent gas mixing on P. tricornutum for the growth, lipid accumula-
k ¼ ðln N1  ln N2 Þ=ðt 1  t2 Þ ð1Þ
tion and fatty acid profiles under the optimum light were
1
investigated. Relationship between total lipid content, FAME con- where k (d ) is the specific growth rate in exponential growth
tent per lipid and specific growth rate to gas liquid ratios, respec- phase, N1 and N2 are the biomass concentration at day t1 and t2.
tively were studied. Biomass productivity was obtained according to the following
calculation (Song et al., 2013):

2. Methods PDMDW ¼ DM  k ð2Þ

1 1
where PDMDW is the biomass productivity (mg L d ) in the expo-
2.1. Algal strain
nential growth phase and DM (mg L1) is the biomass concentration
at the end of the exponential phase.
The microalga P. tricornutum used in this study was obtained
from freshwater algae culture collection of the Institute of Hydro-
2.5. Lipid analysis of P. tricornutum
biology in China (FACHB-Collection). The algae strain was main-
tained in artificial sea water enriched with f/2 nutrients to obtain
The total lipid contents were extracted with a chloroform/
a salinity value of 3.34 wt.% (Guillard and Ryther, 1962).
methanol mixture (2:1, v/v) mixture and were quantified gravi-
metrically (Song et al., 2013). The lipid productivity (PL) was calcu-
2.2. Photobioreactors and culture conditions lated according to the following equation:
PL ¼ PDMDW  LW ð3Þ
Photobioreactors used for the indoor pilot cultivation of P. tricor-
1 1
nutum were designed by our team (Patent No. 201220592057.1). where PL is the lipid productivity (mg L d ) in the exponential
They were identical bubble columns and made of plexiglass. The growth phase. LW stands by lipid content based on dry weight.
working volume was 3 L. And the columns were designed as Fatty acid content and composition analysis were determined
300 mm tall with an inner diameter 120 mm and wall thickness by two steps including preparation of fatty acids methyl ester
5 mm. The air flow was controlled and the gas inlet was located (FAME) and Gas Chromatography–Mass Spectrometry analysis.
at the bottom of the photobioreactor. Air was injected through FAME was prepared by acid-catalyzed esterification in our previ-
two symmetrical aeration heads. And the outlet and standby feed- ous study. Dried algae samples (about 300 mg) were reacted
ing port were set up at the top of the column respectively. directly with 1.5 ml mixture of methanol and solid acid (30 mg)
Three fluorescent lamps were placed on one side of each (Amberlyst 15, Sigma–Aldrich, USA). Transesterification was firstly
column. The columns and auxiliary devices were sterilized by carried out in a 60 °C water bath for 40 min and then 1.5 ml KOH–
washing with a 5% solution of sodium hypochlorite and then under Methanol solution (4% of m:v) was added the solution for 60 °C
ultraviolet radiation of 15 min before beginning the culture. The water bath for 60 min. Upon completion of the reaction, hexane
algae culture in the photobioreactor was aerated by an electromag- (1 mL) was added into the solution, and the whole solution was
netic air compressor with filtered air containing 0.03–0.06% CO2 homogenized (Ultrasonic Homogenizer SCIENTZ-IID, China) for
(Mata et al., 2010). The P. tricornutum was cultivated in an artificial 10 min, and then separated into two layers. Heptadecanoic acid
climate chamber with a temperature of 25 ± 1 °C, continuous illu- methyl ester (50 lL, 2 mg mL1) was added to upper layer solution
 M. Song et al. / Bioresource Technology 165 (2014) 31–37 33

(200 lL) for methyl ester analysis on GC–MS (Trace GC-DSQII,

Thermo Fisher, USA). Solution (l lL) was injected under splitless
mode and the injector temperature was 210 °C. The oven temper-
ature was set at 140 °C, and held for 2 min, then raised to 200 °C at
a rate of 3 °C min1, and held at 200 °C for 10 min. Mass spectra
was recorded under electron ionization (70 eV) at a frequency of
5 scans for 1 s and the ion source was 250 °C and full scan was
150–650 amus. In order to calculate the FAME yield of samples, a
mixture of FAME standards (Sigma Aldrich 47,885, 10 mg mL1)
was analyzed at the same GC–MS condition as the depicted above.
FAME yield were calculated by Eq. (4), where fi is the correction
factor of section i and this data is obtained from the analysis of
the mixture of FAME standards, Ai is the peak area of section i, As
is the peak area of internal standard, Cs is the concentration of
internal standard and v is the volume of upper layer:
X  Ai 
FAME yields ¼ fi v ð4Þ
As =C s

Fig. 2. Specific growth rate and biomass productivity of Phaeodactylum tricornutum

for different gas liquid ratios (vvm). The linear relationship between specific growth
3. Results and discussion
rate (d1) and gas liquid ratio (vvm) is shown in inset.

3.1. Effect of gas liquid ratio on P. tricornutum growth

between specific growth rate and GLR was observed. The positive
To study the growth properties of P. tricornutum under different effect of increasing the GLR on the growth rate has been previously
gas liquid ratios, the growth rate and biomass productivity were reported (Alias et al., 2004; Mirón et al., 2003), however, the quan-
determined. Biomass concentration as a function of cultivation tification of the GLR has not been clearly established. P. tricornutum
time at 25 ± 1 °C and 54 lmol-photons m2 s1 under different achieved the highest specific growth rate (0.5 d1) and highest bio-
gas liquid ratios is plotted in Fig. 1. Growth curves include logarith- mass productivity (227.09 mg L1 d1) under GLR of 1.5 vvm in
mic growth and stationary phases were observed in each trial. P. logarithmic growth phase. The biomass and specific growth rate
tricornutum achieved stable growth phase at the third day under decreased at GLR above 1.5 vvm. This was consistent with the
different gas liquid ratios (GLRs) with an exception that under report that excessive aeration rates could reduce the microalgae
GLR of 1.5 vvm, the biomass of P. tricornutum kept a slow rising. growth (Mirón et al., 2003). The optimum GLR of 1.5 vvm achieved
The increasing biomass and specific growth rate with increasing in this study for P. tricornutum growth showed differences with
GLR to the value of 1.5 vvm (Fig. 2) were attributed to improved others (Alias et al., 2004; Mirón et al., 2003). Alias et al. (2004)
radial mixing, which caused higher mass transfer and higher fre- reported that the biomass productivity strongly decreased at GLR
quency of light exposition. Certain degree of mass transfer and above 2.0 vvm. And Mirón et al. (2003) demonstrated that a GLR
light availability in the culture medium were beneficial to increase of greater than about 1.3 vvm could damage to the cells; conse-
the photosynthetic efficiency of algal cells and, consequently, the quently, the biomass productivity of P. tricornutum was reduced.
biomass productivity (Ota et al., 2009a; Alias et al., 2004; Mirón The possible explanation of the difference may be the difference
et al., 2003). of the P. tricornutum strain, the bubble column and light intensity.
The specific growth rate calculated from the growth curve in The maximum growth rates of P. tricornutum under gas liquid
Fig. 1 as a function of GLR is plotted in Fig. 2. A linear relationship ratios of 0–2.5 vvm were 0.42 d1, 0.55 d1, 0.38 d1, 0.45 d1,
0.50 d1, 0.49 d1, 0.43 d1, respectively. Under the GLR of
0–0.1 vvm, the maximum growth rate was achieved on the first
culturing day. Under higher GLR of 0.5–2 vvm, it was attained on
the second culturing day. And at the highest GLR of 2.5 vvm, it
was reached on the third day. These results indicated that at lower
GLR, the maximum growth rate was easier to be achieved at earlier
time. The highest maximum growth rate was got under GLR of
0.1 vvm at the first culturing day. And the maximum growth rate
in the culture system without air supply was higher than others
except 0.1 vvm at first culturing day. These results may be
explained that in the first incubation time, low concentration of
initial cell was easier to adapt the slight aeration or not. So altering
aeration rates are being considered for the further study to
enhance the biomass and lipid productions simultaneously reduce
the energy.

3.2. Effect of gas liquid ratio on P. tricornutum lipid accumulation

To determine the effect of different gas liquid ratios on lipid

accumulation properties of P. tricornutum, the variations of lipid
content per microalgal biomass, fatty acid methyl ester (FAME)
Fig. 1. Growth curves of Phaeodactylum tricornutum in growth medium for different content per lipid and lipid productivity under different gas liquid
gas liquid ratios (vvm). Values are expressed as means ± s.d. (n = 3). ratios were presented in Fig. 3(a). Total lipid content increased
34 M. Song et al. / Bioresource Technology 165 (2014) 31–37

with increasing GLR to 2 vvm and kept constant (about 25%) up to oils. Therefore, it is necessary to further assess the FAME content
GLR of 2.5 vvm. However, compared to GLR of 1.5 vvm, the algae at of species to determine the potential for algae biodiesel
GLR of 2 vvm and 2.5 vvm only attained a lower lipid productivity production.
of 47.52 mg L1 d1, 42.83 mg L1 d1, respectively, because of As the variations of biomass productivity, the lipid productivity
non-competitive biomass concentration. That may be caused by and FAME content also increased with the air supply up to the GLR
excessive degree of GLR, which may have a negative effect on pho- value of 1.5 vvm, then slightly decreased at GLR above 1.5 vvm.
tosynthesis in algae and resulted in impairment in growth, conse- This positive relation between biomass productivity and FAME
quently, induced the lipid accumulation by shifting carbon flux content was consistent with the previous report that biomass con-
from starch to the lipid synthesis pathway (Eriksen, 2008; Li centration after logarithmic growth was significantly related to the
et al., 2010; Mata et al., 2010). It was interesting to find that under fatty acids production (Ota et al., 2009a). The FAME content per
no air supply, the lipid content of 13.17% was higher than those lipid achieved the highest value of 20.24% (dry weight) under
under slightly aeration of 0.1 vvm and 0.5 vvm. That may be GLR of 1.5 vvm, then decreased in the range of 2–2.5 vvm. Decrease
explained that without air supply, low biomass was attained pos- of the FAME content under high GLR may be caused by decrease of
sibly due to the self-shading of algal cells without mixing. As a triglycerides (TAGs) as FAME is mainly made by the TAGs (Chisti,
result, the production of total lipid was higher than in the other 2007). High GLR may limit the triacylglycerol biosynthesis by the
cells (Li et al., 2010). glycerol pathway (Olofsson et al., 2012; Hu et al., 2008). Fig. 3
Beside intracellular lipid content, lipid productivity has been showed that the variations trend of FAME was different with that
widely recognized as one of the most obvious and easily measur- of the total lipid content. The main reason is that total lipid content
able characters for biodiesel production (Griffiths et al., 2011), not only contain neutral storage lipids such as monoglycerides
because it depends on both biomass productivity and lipid content. (MAGs), diglycerides (DAGs) and triglycerides, free fatty acids,
Recently, fatty acid methyl ester, as the main component of biodie- hydrocarbons and pigments but also contain polar structural lipids
sel, has been paid attentions by some researchers (Han et al., 2013; such as phospholipids and glycolipids (Olofsson et al., 2012; Hu
Wu et al., 2013). However, there is still insufficient quantitative et al., 2008).
data in the report about the fatty acid profiles of the microalgae The total lipid content and FAME content as functions of GLR
were plotted in Fig. 3(b). The linear relationships among lipid con-
tent, FAME content and GLR were observed. The positive effect of
increasing the GLR on the lipid properties has not been reported
before. P. tricornutum achieved the highest FAME content per lipid
(20.24% dry weight) and highest lipid productivity (48.48 mg L1
d1) under GLR of 1.5 vvm, which was higher than the value of
44.8 mg L1 d1 (Mata et al., 2010). And it achieved the highest
lipid content (25.19% dry weight) under GLR of 2.5 vvm.

3.3. Fatty acid compositions of P. tricornutum under different gas

liquid ratios

Fatty acid compositional profiles of the P. tricornutum strains

under different GLRs were shown on Table 1. It is generally recog-
nized that fatty acid profiles play most important role in determin-
ing fuel properties (Tan and Lin, 2011; Hoekman et al., 2012). And
C16–C18 fatty acids are considered as the most suitable composi-
tions for biodiesel production (Knothe, 2009). Table 1 showed that
under different GLRs, the P. tricornutum possessed considerable
amounts of C16 and C18 species, ranging from 91.05% to 95.36%.
Some reports agreed that low levels of polyunsaturated and low
levels of saturated FAs are favorable for biodiesel (Knothe, 2009;
Hoekman et al., 2012). And monounsaturated fatty acids of palmit-
oleic acid (16:1) and oleic acid (18:1) were considered as the most
favorable biodiesel composition, as they can give the finest com-
promise between oxidative stability and cold flow properties
(Knothe, 2009; Hoekman et al., 2012). It was interesting to find
that monounsaturated fatty acids (MUFAs) presented a major per-

Table 1
Fatty acid compositional profiles of Phaeodactylum tricornutum (% of total FAME)
cultivated under different gas liquid ratios (vvm).

Fatty acid profiles Gas liquid ratio (vvm)

0 0.1 0.5 1 1.5 2 2.5
C16–18 91.05 95.02 95.36 94.94 92.90 93.04 94.05
SFA 44.91 42.20 40.69 50.65 50.16 51.90 53.97
MUFA 55.09 57.80 59.31 49.20 48.79 47.85 45.75
Fig. 3. (a) Lipid productivity, lipid content per microalgal biomass (%, g/g, dry PUFA 0.00 0.00 0.00 0.14 1.05 0.25 0.28
weight) and FAME content per lipid (%, g/g, dry weight) of Phaeodactylum
tricornutum for different gas liquid ratios (vvm). (b) The linear relationships among SFA = saturated fatty acids (14:0, 16:0, 18:0, 24:0); MUFA = monounsaturated fatty
lipid content per microalgal biomass (%, g/g, dry weight), FAME content per lipid (%, acids (16:1, 18:1); PUFA = polyunsaturated fatty acids (16:2, 16:3, 18:2, 18:3, 20:4,
g/g, dry weight) and gas liquid ratio (vvm). 20:5, 22:6).
 M. Song et al. / Bioresource Technology 165 (2014) 31–37 35

centage of 45.75–59.31%, followed by saturated fatty acids (SFAs)

(40.69–53.97%). These results indicated that P. tricornutum oils
would achieve high cetane number and low iodine value to meet
the requirements in standards of ASTM D6751 in the US and EN
14214 in Europe (Hoekman et al., 2012; Francisco et al., 2010;
Refaat, 2009). The content of PUFA made up only a small percent-
age of (0.14–1.05%) of the total FAME. Predominantly, PUFA largely
comprise the structural lipid fraction which mainly composed the
membranes of organelles in the cell (Olofsson et al., 2012). Table 1
showed that polyunsaturated fatty acid (PUFA) began to emerge at
higher GLR above 1 vvm, which indicated the small part of cell
division occurred in the stable phase, this was consistent with
the growth curve showed in Fig. 1. Highest content of MUFA
(59.31%) and lowest SFA (40.69%) under different GLRs were
achieved at 0.5 vvm, the FA compositions would give the finest
compromise between oxidative stability and cold flow properties.
However, there is no significant difference between GLR of
1.5 vvm and GLR of 0.5 vvm on the content of SFA and MUFA. FA
profiles under GLR of 1.5 vvm also showed favorable properties Fig. 4. pH variations under different gas liquid ratios at 298 K and a light intensity
for biodiesel production. Combined with biomass and lipid produc- of 54 lmol-photons m2 s1.
tivity, GLR of 1.5 vvm was considered as the optimum condition for
P. tricornutum growth and oil production. In the further study,
incubation time of the algae at GLR of 1.5 vvm would be extended promoted higher CO2 absorption rate from the air to the medium
to induce the transfer of polar structural lipids to natural storage than the CO2 uptake rate by algae for photosynthesis (Bitaubé
lipids. Pérez et al., 2008).
Overall, optimum biodiesel composition of palmitoleic acid Generally, when CO2 dissolves in water, CO2 hydrolyses to H+
+ 2
(16:1) was the most common FAs in the P. tricornutum oils under and HCO 
3 , and HCO3 ionizes to H and CO3 , and the above two
different GLRs, followed by palmitic acid (C16:0). Which consistent reactions are both significant in both directions. The balance
with the report that the major fatty acids are C16:0 and C16:1 in between these opposing processes determines the relative num-
the Bacillariophyceae (Hu et al., 2008). The average values of bers of ions and neutral molecules. The relative numbers of each
C16:1 (48.36%) and C16:0 (41.46%) of P. tricornutum were higher type of ion or molecule in the reaction are constant or not will
than the mean contents of the two fatty acids in the previous determine whether the state of chemical equilibrium is produced.
report (Hoekman et al., 2012). The fatty acids profile in some mic- So aeration allows a continuous CO2 absorption from the atmo-
roalgae could be easily changed through altering the cultivation sphere to the medium, which provokes a displacement of the car-
2
conditions (Damiani et al., 2010; Mata et al., 2010). The fatty acid bonate balance, increasing the HCO 3 and CO3 concentration. And
compositions also fluctuated under different GLRs. For example, the dissociation constants for carbonic acid in seawater are given
C16:1 ranged from about 40% to 57%, C16:0 fluctuated from about as follows:
36% to 47%, and C18:1 alternated from 2.27% to 5.66%. As GLR
increased from 0.5 vvm to 2.5 vvm, the content (%) of C16:0 and K 1 ¼ ½Hþ ½HCO3 =½CO2  ð5Þ
C18:1 increased, but the content of C16:1 decreased dramatically.
These results indicated that GLR had important effect on the fatty
K 2 ¼ ½Hþ ½CO2 
3 =½HCO3  ð6Þ
acid contents. Air flow (GLR in this study) can produce degrees of
mixing stress which influenced the fatty acid content. And pH var- The constants (K⁄1
and K⁄2)
can be calculated by a thermody-
iation controlled by GLR may be another important impact factor. namic model, (Millero et al., 2006; Ota et al., 2009b) which allows
estimates of the dissociation constants of inorganic carbon in sea-
3.4. Effect of CO2 
3 /HCO3 on the fatty acid compositions of P. water as a function of temperature, pH and salinity at atmospheric
tricornutum pressure.
According to the thermodynamic model (Millero et al., 2006),
During the study, pH was measured every 24 h in the culturing under the conditions of temperature of 298 K and salinity of
time. Fig. 4 showed the pH variations under different gas liquid 3.34 wt.%, pK⁄1, pK⁄2 were calculated as 5.85 and 8.98, respectively.
ratios at 298 K and a light intensity of 54 lmol-photons m2 s1. According to the function of fraction of dissociation of inorganic
Under the condition of aeration or not, the pH showed significant carbon and pH in the culture medium (Ota et al., 2009b), in place
difference. The pH increased dramatically in the first 3 days with- of the pH region from 8 to 9 corresponding to the different GLRs
2
out air supply and kept average pH of 9.1 in the stable phase. How- studied, HCO 3 and CO3 were considered as the major species of
ever, pH decreased rapidly on the first day with air supply and kept inorganic carbon. In this study, the effect of CO2 
3 /HCO3 ratio on
2 
average pH of 8.0–8.1 in the stable phase. The increase of pH under the growth of algae was investigated. The CO3 /HCO3 ratios were
no aeration mainly caused by the CO2 uptake for algal photosyn- calculated from pH data in Fig. 4. The calculated CO2 
3 /HCO3 ratios
thesis (Ota et al., 2009a), then the stable pH value was achieved were1.318, 0.132. 0.125, 0.118, 0.112, 0.108, 0.102 under corre-
as the algae grew into stationary phase on the third day. However, sponding GLRs of 0, 0.1, 0.5, 1, 1.5, 2, 2.5 vvm, respectively.
with air supply, besides photosynthesis in algae culture medium, CO2 
3 /HCO3 seem to have some effect on the algae growth and
air bubbling velocity was another important factor to influence fatty acid synthesis (Fig. 5). The lower CO2 
3 /HCO3 ratio, the higher
the pH (Ota et al., 2009a; Bitaubé Pérez et al., 2008), which was specific growth rate and FAME content. In addition, some research
consistent with the results in Fig. 4 that as air bubbling velocity have reported that high value of HCO 3 /CO2 may contribute to the
caused by GLR increasing, average pH decreased in the stable enhancement of cell growth and the beginning of fatty acid synthe-
phase. Especially in the first culturing day with low initial cell con- sis (Ota et al., 2009a; Sanchez and Harwood, 2002). These results
centration, the reducing pH mainly caused by air bubbling which indicated that bicarbonate ion HCO 3 was an important substrate
36 M. Song et al. / Bioresource Technology 165 (2014) 31–37

60 thank Findlay A Nicol of Shandong University of Finance and Eco-

nomics for revising the English in the manuscript.
Fatty acid profiles (% of total FAME)

50

40 C16:0 References
C16:1
C16:2 Alias, C.B., Garcia-Malea Lopez, M.C., Acien Fernandez, F.G., Fernindez Sevilla, J.M.,
30 Garcia Sanchez, J.L., Molina Grima, E., 2004. Influence of power supply in the
C18:0
feasibility of Phaeodactylum tricornutum cultures. Biotechnol. Bioeng. 87 (6),
C18:1 723–733.
20
C18:2 Barbosa, M.J.G.V., 2003. Microalgal photobioreactors: scale-up and optimisation
C18:3 (Ph.D. thesis). Wageningen University, The Netherlands.
10 Bitaubé Pérez, E., Caro Pina, I., Pérez Rodríguez, L., 2008. Kinetic model for growth of
Phaeodactylum tricornutum in intensive culture photobioreactor. Biochem. Eng.
0 J. 40 (3), 520–525.
Boyer, J.S., 1982. Plant productivity and environment. Science 218, 443–448.
Brennan, L., Owende, P., 2010. Biofuels from microalgae – a review of technologies
0.10 0.11 0.12 0.13 1.2 1.4
for production, processing, and extractions of biofuels and co-products. Renew.
2- -
CO3 /HCO3 ratio Sust. Energy Rev. 14 (2), 557–577.
Chisti, Y., 2007. Biodiesel from microalgae. Biotechnol. Adv. 25 (3), 294–306.
Damiani, M.C., Popovich, C.A., Constenla, D., Leonardi, P.I., 2010. Lipid analysis in
Fig. 5. Each fatty acid content as a function of CO2 
3 /HCO3 at 298 K and a light
Haematococcus pluvialis to assess its potential use as a biodiesel feedstock.
intensity of 54 lmol-photons m2 s1 for 9 days incubation time.
Bioresour. Technol. 101 (11), 3801–3807.
Eriksen, N.T., 2008. The technology of microalgal culturing. Biotechnol. Lett. 30 (9),
1525–1536.
for algal growth and FAME accumulation. Therefore, it was demon- Francisco, E.C., Neves, D.B., Jacob-Lopes, E., Franco, T.T., 2010. Microalgaes feedstock
for biodiesel production: carbon dioxide sequestration, lipid production and
strated that HCO 3 was a promoter at low CO2 concentrations
biofuel quality. J. Chem. Technol. Biotechnol. 85, 395–403.
(0.03–0.06%) during the algae cultivation, which further supple- Griffiths, M.J., Hille, R.P., Harrison, S.T.L., 2011. Lipid productivity, settling potential
ment the previous report which only confirmed HCO 3 was good and fatty acid profile of 11 microalgal species grown under nitrogen replete and
limited conditions. J. Appl. Phycol. 24 (5), 989–1001.
for cell growth at the high CO2 concentrations of 1–50% (Ota
Guillard, R.R., Ryther, J.H., 1962. Studies of marine planktonic diatoms. I. Cyclotella
et al., 2009b). nana Hustedt and Detonula confervacea (cleve) Gran. Can. J. Microbiol. 8, 229–
The effect of CO2 
3 /HCO3 ratio on the fatty acids profiles (% of 239.
total FAME) was quantitatively investigated. Fig. 5 showed total Han, F., Wang, W., Li, Y., Shen, G., Wan, M., Wang, J., 2013. Changes of biomass, lipid
content and fatty acids composition under a light-dark cyclic culture of Chlorella
fatty acid profiles content in nine culture days plotted against pyrenoidosa in response to different temperature. Bioresour. Technol. 132, 182–
CO2  2 
3 /HCO3 ratio and the CO3 /HCO3 ratio seem to have effect on 189.
the content of fatty acid profiles. As the CO2 
3 /HCO3 ratio increased
Hoekman, S.K., Broch, A., Robbins, C., Ceniceros, E., Natarajan, M., 2012. Review of
biodiesel composition, properties, and specifications. Renew. Sust. Energy Rev.
to 0.125, the content of C16:1 increased to the highest value of 16, 143–169.
57.04%, then slightly decreased in the ratios of 0.132–1.318. In Hu, Q., Sommerfeld, M., Jarvis, E., Ghirardi, M., Posewitz, M., Seibert, M., Darzins, A.,
the contrary, the content of C16:0 decreased to 36.05%, and kept 2008. Microalgal triacylglycerols as feedstocks for biofuel production:
perspectives and advances. Plant J. 54 (4), 621–639.
constant (about 36%) up to ratio of 1.318. In addition, the content Knothe, G., 2009. Improving biodiesel fuel properties by modifying fatty ester
of C18:1 decreased slightly. During the de novo biosynthesis path- composition. Energy Environ. Sci. 2 (7), 759.
way of fatty acid, saturated palmitoyl-ACP (16:0-ACP) and stea- Lam, M.K., Lee, K.T., 2012. Microalgae biofuels: a critical review of issues, problems
and the way forward. Biotechnol. Adv. 30, 673–690.
royl-ACP (18:0-ACP) are produced. After the elongation of the
Larkum, A.W.D., Ross, I.L., Kruse, O., Hankamer, B., 2012. Selection, breeding and
16:0-ACP to 18:0-ACP proceeds, generated fatty acids are modified engineering of microalgae for bioenergy and biofuel production. Trends
by desaturase to produce unsaturated fatty acids (Hu et al., 2008; Biotechnol. 30, 198–205.
Li, Y., Han, D., Hu, G., Sommerfeld, M., Hu, Q., 2010. Inhibition of starch synthesis
Ota et al., 2009a). As the CO2 
3 /HCO3 ratio increased, the content
results in overproduction of lipids in Chlamydomonas reinhardtii. Biotechnol.
of C16:0 and C18:0 decreased may be explained that the desatura- Bioeng. 107 (2), 258–268.
tion rate was higher than the elongation rate, which finally Lim, S.L., Chu, W.L., Phang, S.M., 2010. Use of Chlorella vulgaris for bioremediation of
enhanced the production of C16:1. textile wastewater. Bioresour. Technol. 101, 7314–7322.
Liu, T., Wang, J., Hu, Q., Cheng, P., Ji, B., Liu, J., Chen, Y., Zhang, W., Chen, X., Chen, L.,
Gao, L., Ji, C., Wang, H., 2013. Attached cultivation technology of microalgae for
efficient biomass feedstock production. Bioresour. Technol. 127, 216–222.
4. Conclusions Mata, T.M., Martins, A.A., Caetano, N.S., 2010. Microalgae for biodiesel production
and other applications: a review. Renew. Sust. Energy Rev. 14 (1), 217–232.
The study demonstrated gas flow rate plays an important role Millero, F.J., Graham, T.B., Huang, F., Bustos-Serrano, H., Pierrot, D., 2006.
Dissociation constants of carbonic acid in seawater as a function of salinity
on P. tricornutum growth and lipid production. GLR of 1.5 vvm and temperature. Mar. Chem. 100 (1–2), 80–94.
was considered as the optimum GLR for algae growth and yielding Mirón, A.S., Garcı´a, M.C.C., Gómez, A.C., Camacho, F.G.a., Grima, E.M., Chisti, Y., 2003.
oil, with specific growth rate (0.5 d1), biomass productivity Shear stress tolerance and biochemical characterization of Phaeodactylum
tricornutum in quasi steady-state continuous culture in outdoor
(227.09 mg L1 d1), FAME content per lipid (20.24% dry weight), photobioreactors. Biochem. Eng. J. 16 (3), 287–297.
lipid productivity (48.48 mg L1 d1) and the favorable FA profiles Olofsson, M., Lamela, T., Nilsson, E., Bergé, J.P., del Pino, V., Uronen, P., Legrand, C.,
of C16–C18 (92.90%). The linear relationships among specific 2012. Seasonal variation of lipids and fatty acids of the microalgae
Nannochloropsis oculata grown in outdoor large-scale photobioreactors.
growth rate, lipid content, FAME content and GLR were observed.
Energies 5 (12), 1577–1592.
CO2 
3 /HCO3 ratio had important effect on the fatty acids synthesis, Ota, M., Kato, Y., Watanabe, H., Watanabe, M., Sato, Y., Smith Jr., R.L., Inomata, H.,
especially the component of C16:1. 2009a. Fatty acid production from a highly CO2 tolerant alga, Chlorocuccum
littorale, in the presence of inorganic carbon and nitrate. Bioresour. Technol. 100
(21), 5237–5242.
Acknowledgements Ota, M., Watanabe, H., Kato, Y., Watanabe, M., Sato, Y., Smith Jr., R.L., Inomata, H.,
2009b. Effect of inorganic carbon on photoautotrophic growth of a microalga
Chlorococcum littorale. Biotechnol. Prog. 25, 492–498.
Funding for this research was provided by Natural Science Refaat, A.A., 2009. Correlation between the chemical structure of biodiesel and its
Foundation of China (51078221), Science and Technology Develop- physical properties. Int. J. Environ. Sci. Technol. 6 (4), 677–694.
ment Planning of Shandong Province (2012GGE27027), The Sanchez, J., Harwood, J.L., 2002. Biosynthesis of triacylglycerols and volatiles in
olives. Eur. J. Lipid Sci. Technol. 104, 564–573.
Program for New Century Excellent Talents in University of Minis- Song, M., Pei, H., Hu, W., Ma, G., 2013. Evaluation of the potential of 10 microalgal
try of Education of China (Grant No. NCET-12-0341). The authors strains for biodiesel production. Bioresour. Technol. 141, 245–251.
 M. Song et al. / Bioresource Technology 165 (2014) 31–37 37

Stephens, E., Ross, I.L., Mussgnug, J.H., Wagner, L.D., Borowitzka, M.A., Posten, C., Wijffels, R., Barbosa, M., 2010. An outlook on microalgal biofuels. Science 329, 796–799.
Kruse, O., Hankamer, B., 2010. Future prospects of microalgal biofuel production Wu, C., Wang, W., Yue, L., Yang, Z., Fu, Q., Ye, Q., 2013. Enhancement effect of
systems. Trends Plant Sci. 15, 554–564. ethanol on lipid and fatty acid accumulation and composition of Scenedesmus
Tan, Y.X., Lin, J., 2011. Biomass production and fatty acid profile of a sp. Bioresour. Technol. 140, 120–125.
Scenedesmusrubescens-like microalga. Bioresour. Technol. 102, 10131–10135. Zhu, X.G., Long, S.P., Ort, D.R., 2008. What is the maximum efficiency with which
Tredici, M.R., 2010. Photobiology of microalgae mass cultures: understanding the photosynthesis can convert solar energy into biomass? Curr. Opin. Biotechnol.
tools for the next green revolution. Biofuels, 143–162. 19, 153–159.