Bioresource Technology 135 (2013) 128136

Contents lists available at SciVerse ScienceDirect

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

A biorenery from Nannochloropsis sp. microalga Extraction of oils

and pigments. Production of biohydrogen from the leftover biomass
B.P. Nobre a,c, F. Villalobos b, B.E. Barragn b, A.C. Oliveira a, A.P. Batista a, P.A.S.S. Marques a, R.L. Mendes a,
H. Sovov d, A.F. Palavra c, L. Gouveia a,
a
Laboratrio Nacional de Energia e Geologia (LNEG), Unidade de Bioenergia, Estrada do Pao do Lumiar, 1649-038 Lisboa, Portugal
b
Escuela Nacional de Ciencias Biolgicas, Instituto Politcnico Nacional, Mxico, DF, Mexico
c
IST, Centro Qumica Estrutural, DEQB, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal
d
Institute of Chemical Process Fundamentals of the AS CR, Rozvojov 135, 165 02 Prague, Czech Republic

h i g h l i g h t s

" Nannochloropsis sp. was studied as a biorenery context.

" Production of fatty acids for biodiesel, high added-value compounds and biohydrogen.
" Extracts fractionation by SFE (oils for biodiesel and pigments for food).
" Fermentation of biomass leftover by E. aerogenes yielded maximum 60.6 mL H2/gdry biomass alga.
" Best SFE conditions extracted 45 glipids/100 gdry biomass (40 C, 300 bar, CO2 + ethanol).

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

Article history: The microalga Nannochloropsis sp. was used in this study, in a biorenery context, as biomass feedstock
Available online 28 November 2012 for the production of fatty acids for biodiesel, biohydrogen and high added-value compounds. The mic-
roalgal biomass, which has a high lipid and pigment content (mainly carotenoids), was submitted to
Keywords: supercritical CO2 extraction. The temperature, pressure and solvent ow-rate were evaluated to check
Nannochloropsis their effect on the extraction yield. The best operational conditions to extract 33 glipids/100 gdry biomass
Oils were found to be at 40 C, 300 bar and a CO2 ow-rate of 0.62 g/min. The effect of adding a co-solvent
Carotenoids
(ethanol) was also studied. When supercritical CO2 doped with 20% (w/w) ethanol was used, it was
Biohydrogen
Biorenery
possible to extract 45 glipids/100 gdry biomass of lipids and recover 70% of the pigments. Furthermore,
the remaining biomass after extraction was effectively used as feedstock to produce biohydrogen
through dark fermentation by Enterobacter aerogenes resulting in a hydrogen production yield of
60.6 mL/gdry biomass.
2012 Elsevier Ltd. All rights reserved.

1. Introduction drawbacks. The main constraints include the limited amount of

Renewable carbon-neutral liquid biofuels are needed to replace
petroleum-derived transport fuels in the near future as they con-
tribute to global warming and are of a limited availability. A prom-
ising alternative is microalgae, because of their much higher
photosynthetic efciency, areal productivity and oil content and
also that they do not compete with food cultures, arable land,
and potable water, and they have the possibility of being harvested
on a daily basis (Gouveia and Oliveira, 2009; Gouveia, 2011). How-
ever, the current implementation of microalga-based systems has
been economically constrained and still has some technological
Corresponding author. Fax: +351 21 7127195.
E-mail address: luisa.gouveia@lneg.pt (L. Gouveia).
biomass which can be obtained with currently available photobior-
eactors, the low biomass productivity, the low harvesting ef-
ciency and the relatively low microalga intrinsic lipid content
(Amaro et al., 2011).
In terms of microalgal lipids, the total amount and the type of
fatty acids are specic for each species of microalgae and this is
linked to environmental factors (e.g. light intensity, pH, salinity,
temperature, concentration of nitrogen, and other nutrients;
Gouveia and Oliveira, 2009; Gouveia et al., 2009). The lipid content
of microalgae is approximately 2050% of its dry weight with a
possibility to attain up to 80% (Spolaore et al., 2006).
In addition to the energy content of the microalgal biomass,
these microorganisms have the capacity to synthesize bioactive
molecules, such as carotenoids, antioxidants, anti-inammatory

0960-8524/$ - see front matter 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biortech.2012.11.084
 B.P. Nobre et al. / Bioresource Technology 135 (2013) 128136
and other valuable organic compounds, which can be used in food,
feed, cosmetics, biomaterials, nanostructures and the pharmaceu-
tical industry. In particular, the carotenoids such as astaxanthin,
canthaxanthin and zeaxanthin have been regarded as free-radical
scavengers and cancer preventives and are believed to play an
important role in the protection against a large number of chronic
and acute health conditions (e.g. Stahl and Sies, 2005).
All these applications are in addition to the original energy pur-
poses. The biorenery approach consists in the production of a
wide range of biofuels and chemicals. This could be done through
the use of various technologies in a cost-effective and environmen-
tally sustainable way. The concept is not new, however, it contrib-
utes to making biofuel production economically viable (Gouveia,
2011).
The extraction of lipids and carotenoids from microalgae is usu-
ally carried out using classical extraction methods, such as using
organic solvents and Soxhlet extractions (e.g. Gouveia and Oliveira,
2009; Bai et al., 2011). However, the selectivity is low and a large
amount of toxic solvent is needed. In recent years, supercritical
uid extraction (SFE) has become an important alternative to the
conventional separation methods. CO2 is the most used supercrit-
ical solvent because the compounds can be obtained without con-
tamination by toxic organic solvents and thermal degradation. The
product is solvent free and compatible with the use of the label
natural (Bruno et al., 1993). On the other hand, supercritical
CO2 has a high diffusivity, a low viscosity, and a low surface tension
which improves the mass transfer inside the matrix. Many applica-
tions involving supercritical uid extraction of lipids and bioactive
compounds (e.g. carotenoids, fatty acids) from microalgae have
been studied (Andrich et al., 2005; Macas-Snchez et al., 2005;
Nobre et al., 2006; Gouveia et al., 2007; Liau et al., 2010). The
use of a polar co-solvent like ethanol can improve CO2 solvent
properties for more polar compounds, increasing the extraction
rate, as well as the recovery yield. This is particularly interesting
when applying supercritical CO2 to the bioactive compounds ex-
tracted from microalgae (Nobre et al., 2006). Furthermore, other
authors (Andrich et al., 2005; Macas-Snchez et al., 2005; Liau
et al., 2010) have used SFE with success to obtain lipids and carote-
noids from the microalga Nannochloropsis, but none of these works
have studied the fractionation of the extracts in order to separate
lipids from carotenoids.
After the extraction of oils and pigments from the microalga
biomass, the remaining biomass could be used in a fermentation
process as a substrate to produce hydrogen. The conversion of
lipid-extracted microalgal biomass residues into hydrogen plays
a dual role in renewable energy production and sustainable devel-
opment of the microalgal biodiesel industry (Yang et al., 2011).
Dark and photo-fermentations are considered the more advanta-
geous biological processes, due to the possibility of using several
different feedstocks as a substrate, namely organic wastes and
microalgal biomass (Ferreira et al., 2011, 2012; Yang et al., 2011)
coupled with the wastewater treatment and CO2 mitigation to
obtain clean energy (Lindblad et al., 2002; Hallenbeck, 2009).
Nannochloropsis sp., a marine microalga, commonly used in
aquaculture, has recently become more widely recognized as a
potential source of lipids for biodiesel production (Gouveia and
Oliveira, 2009; Moazami et al., 2012). It has also been considered
as a potential source of carotenoids (Lubin et al., 2000), such as
astaxanthin, canthaxanthin and zeaxanthin.
Several studies have been done with the microalga Nannochlor-
opsis. Examples can be shown from past work in the eighties,
where the effect of environmental factors on growth rate, cell-lipid
content and productivities in laboratory and outdoor cultures were
studied (Boussiba et al., 1987). Recently, the production metrics
were studied in a scalable outdoor photobioreactor for commercial
applications (Quinn et al., 2012), in terms of the biomass and lipid
129
productivities, as well as evaluating the process of mixing the cul-
ture and its inuence on the energy consumption. The supercritical
uid extraction of the bioactive lipids (Andrich et al., 2005), the
factors and strategies that affect fatty acid accumulation, such as
light intensity and N starvation (van Vooren et al., 2012) and the
net energy analysis of the production of biodiesel and biogas from
Nannochloropsis (Razon and Tan, 2011) are only some of the many
examples of the studies done with this microalga.
The aim of this work was to use the biomass of the Nannochlor-
opsis sp. and apply the biorenery approach for the production of
oil (for biodiesel production), high added-value compounds
(carotenoids) and biohydrogen. This work deals with the culture
of the microalga, the supercritical uid extraction of the oil and
high added-value compounds (carotenoids), and the production
of biohydrogen using the remaining biomass. Furthermore, for
comparison the supercritical uid extraction process with conven-
tional methods, the oil and carotenoids from the microalga were
also obtained by Soxhlet and Bligh & Dyer, acetone and ethyl ace-
tate extractions, respectively.
2. Methods
2.1. Microalga
Nannochloropsis sp. (NANNO-2 from SERI algotec) was grown in
modied GPM medium (0.200 g/L KNO3, 0.038 g/L K2HPO4,
0.034 g/L H3BO3, 0.030 g/L Na2EDTA, 4.30 mg/L MnCl24H2O,
1.45 mg/L FeCl36H2O, 0.30 mg/L ZnCl2, 0.13 mg/L CoCl26H2O) in
75% ltered seawater (GF/C lter 1.2 lm pore) and 25% de-ionized
water.
The microalga was cultivated in polyethylene bags (PBR) of 10 L
capacity, agitated by bubbling air (1 vvm/mL L1 min1), at a con-
stant temperature of 25 1 C and under 25.7 lmol/lEinsteins
light intensity (measured at the surface of the PBR with Phywe
Lux-Meter) by uorescence lamps (Philips TL-D 36 W/54-765).
Each culture was grown for 40 days, in order to guarantee high
lipid accumulation during the stationary phase (20 days). The
growth of the cultures was measured through O.D.540 nm in a cuv-
ettes path length of 1 cm (Hitachi U-2000) and dry weight concen-
tration (GF/C lter 1.2 lm pore).
The harvesting of the biomass was carried out by stopping agi-
tation, followed by centrifugation at 10,000 rpm for 10 min (Beck-
man Avanti J-25I) and dried in an oven at 70 C.
2.2. Extraction
Previously to extractions, the microalga Nannochloropsis sp. was
ground using a ball mill Retsch model MM400. 0.5 g of dry bio-
mass was milled for 3.5 min, using 8 balls (10 mm ) and speed
of 25 s1. This procedure allowed to homogenize the samples, as
well as increasing the contact area of the microalgae with the
extraction solvent.
2.2.1. Total lipid content of biomass
Total lipid content in the microalga biomass was assessed by
two different methods: Soxhlet extraction and Bligh and Dyer
(1959). The Soxhlet extraction was carried out using 1 g of micro-
algae for 6 h. The amount of total lipids was determined gravimet-
rically. Two solvents were tested: hexane and ethanol.
In the Bligh and Dyer extraction (Bligh and Dyer, 1959) 150 mg
of ground microalga biomass was added to a mixture of methanol,
chloroform and water (10:5:4, v/v/v) and stirred for 24 h. After-
ward, the sample was centrifuged, removing the liquid and stored
in a separating funnel. This process was repeated two more times
using a stirring time of 2 h. Chloroform and water were added to
130 B.P. Nobre et al. / Bioresource Technology 135 (2013) 128136

the total liquid collected so that the proportions of methanol, chlo- Table 1
roform and water were of 10:9:9 (v/v/v), respectively. The bottom Operational conditions for supercritical uid extraction experiments.

was collected and evaporated. The total lipid content was deter- Experiments Solvent T P Flow-rate
mined gravimetrically. (C) (bar) (g/min)
SFE1 CO2 40 125 0.35
2.2.2. Carotenoid content SFE2 CO2 40 200 0.35
SFE3 CO2 40 300 0.35
2.2.2.1. Hot extraction. One gram of microalga was mixed with SFE4 CO2 60 300 0.35
19 mL ethyl acetate during 24 min at a temperature of 65 C. This SFE5 CO2 40 300 0.62
method was performed both with whole (unbroken) and ground SFE6 CO2 modied with 5 wt.% ethanol 40 300 0.62
microalga. SFE7 CO2 modied with 10 wt.% ethanol 40 300 0.62
SFE8 CO2 modied with 20 wt.% ethanol 40 300 0.62
SFE9 CO2 combined with CO2 modied with 40 300 0.62
2.2.2.2. Cold extraction. Twenty ve milligrams of microalga was 20 wt.% ethanol
mixed with 2 mL acetone, 1 g of glass beads and kept at 22 C
for 10 min, followed by stirring for 30 s using a vortex. Afterward,
the sample was centrifuged and the supernatant was collected. The
Supercritical uid experiments were carried out with ground
operation was repeated until the supernatant had no color. Two
Nannochloropsis sp. at the conditions described in Table 1.
solvents were tested: ethyl acetate and acetone. This method
The total carotenoids obtained with each extraction method
was performed with whole (unbroken) and ground microalga.
were quantied by spectrophotometry and identied by using Thin
Layer Chromatography (TLC) and High Performance Liquid Chro-
2.2.3. Supercritical uid extraction (SFE) of lipids and carotenoids matography (HPLC) (see Section 2.5).
The SFE apparatus used in the study was detailedly described by
Mendes et al. (1995) and modied to include a co-solvent addition
system (Fig. 1). CO2 (99.998% purity, Air Liquide, Portugal) is 2.3. Analysis of fatty acids (transesterication)
pumped to the system by a mini-pump (5) and the pressure is con-
trolled by a back-pressure regulator (8). The target temperature of The fatty acids analysis was determined by gas chromatography
CO2 is achieved with the help of a heat exchanger (13), before (GC). Fatty acids methyl esters were prepared according to Lepage
reaching the extraction vessel (5 cm3) (14), which is lled with and Roy (1986) with modications prior to GC analysis. A known
about 1.25 g of dry microalga biomass mixed with glass beads weight of extract (about 20 mg of lipids) was added to 2 mL of a
(3 mm ), put between two layers of glass wool. The ethanol is mixture of methanol:acetyl chloride (95:5, v/v) and a volume of
pumped to the system by an HPLC pump (10) and a calculated 0.2 mL of a solution of an internal standard heptacanoic acid in
amount of co-solvent is injected, in order to achieve the desired petroleum ether 80100 (5 mg/mg) was added. Nitrogen was
composition. Afterward, the system is allowed to equilibrate for added to the vials, in order to obtain inert atmosphere, and the
1 h and then a purge was carried out, in order to obtain the desired samples were heated at 80 C for 1 h in the dark. Afterward, the
composition at the entrance of the extraction vessel. After this, samples were allowed to cool down to room temperature and di-
extraction could be carried out and CO2 and ethanol were pumped luted with 2 mL of hexane (containing 2,6-di-tert-butyl-4-methyl-
to the system at the desired ow-rates, in order to maintain the phenol-BHT 0.01%) and 1 mL of water, in order to extract the
working concentration. After leaving the extraction vessel the uid methyl esters. The organic phase was dried over anhydrous sodium
is expanded to atmospheric pressure in a three-way valve (15) and sulfate and collected in vials under nitrogen atmosphere. The sam-
the extract precipitates in a cooled glass U tube (16) lled with ples were analyzed by gas chromatography using a CP-3800 GC
glass wool. Total amount of CO2 is measured by a wet test meter (Varian, USA) equipped with 30 m SUPELCOWAX 10 capillary col-
(18). The extract is collected washing the glass wool, the inside umn (0.32 mm of internal diameter and 0.25 lm of lm thickness).
of the U-tube and the expanding tubing with acetone. Quantica- Injector (split 1:50) and detector (ame ionization) temperatures
tion of the total lipids was carried out gravimetrically concentrat- were kept constant at 250 C. The oven temperature program
ing the collected solution under vacuum and drying the extract started at 200 C for 20 min, increased at 20 C/min until 220 C
under nitrogen. and kept constant of this temperature for 20 min. Carrier gas, He,

Fig. 1. Schematic diagram of the supercritical uid extraction apparatus 1, CO2 cylinder; 2, check valve; 3, ice cooler; 4, lter; 5, CO2 mini-pump; 6 and 7, manometers; 8,
back-pressure regulator; 9, co-solvent supply tank; 10, co-solvent HPLC pump; 11, manometer; 12, water bath; 13, heat exchanger; 14, extraction vessel; 15, three-way valve;
16, glass U-tube; 17, rotameter; 18, wet test meter; 1927, valves.
 B.P. Nobre et al. / Bioresource Technology 135 (2013) 128136
was kept at a constant rate of 0.9 mL/min. Fatty acid methyl esters
standards injections were performed in order to identify their pres-
ence in the extracts from Soxhlet and SFE. Fatty acids composition
was calculated as percentage of the total fatty acids present in the
sample, determined from the peak areas. Fatty acid content was
calculated according to EN 14103 (2003).
2.4. Identication and quantication of the extracted carotenoids
Total carotenoids were quantied by spectrophotometry (Hit-
achi-2000). Spectra were run between 380 and 700 nm. Calcula-
tions were performed based on the equivalents of lutein/
zeaxanthin using the Beer Lambert equation with a value of
234 L/(g cm) for the specic optical coefcient at the wavelength
of the maximum absorbance of lutein/zeaxanthin in the solvent
(450 nm for all solvents tested) (Rodriguez-Amaya, 2005).
The carotenoids were identied by TLC. A characterization of
the carotenoids extracted was made using a silica gel plate, which
was previously subjected to heating in an oven at 80 C for 1.5 h.
The eluent consisted of a mixture of petroleum ether 4060 C:ace-
tone:diethylamine in the ratio 10:4:1 (v/v/v), respectively.
Moreover, HPLC was used to identify and quantify the major ex-
tracted carotenoids. The HPLC system consisted of a liquid chro-
matograph, Hewlett Packard 1100 series, with UV/VIS detector
adjusted to 450 nm. A mobile phase of (methanol and 0.2%
H2O):acetonitrile (75:25, v/v) was used at 1 mL/min with the re-
versed phase column, 250 mm  4.0 mm, lbondapack. Standards
of astaxanthin (98% Sigma), lutein (90% Sigma), trans-b-carotene
(95% Type I, Sigma) echinenone (98% Roche) and canthaxanthin
(10% Roche) were used to identify the major carotenoids, by
comparing the retention times of the carotenoid with those of
the standard compound. Each carotenoid was quantied by using
the HPLC relative areas.
2.5. Experimental procedure for biohydrogen fermentative production
from Nannochloropsis sp. leftover biomass
2.5.1. Inoculum culture conditions
Enterobacter aerogenes ATCC 13048, harvested from exponen-
tially grown cultures, was used for fermentation experiments.
The original culture was kept at 4 C in solid CASO Agar. The bac-
teria synthetic growth media used was a 20 g/L peptone solution
(with 5 g/L NaCl), while the fermentation media for the biohydro-
gen production assays contained K2HPO4 (7.0 g/L), KH2PO4 (5.5 g/
L), tryptone (5 g/L), yeast extract (5 g/L), (NH4)2SO4 (1.0 g/L),
MgSO47H2O (0.25 g/L), CaCl22H2O (0.021 g/L), Na2MoO42H2O
(0.12 g/L), nicotinic acid (0.02 g/L), Na2SeO3 (0.172 mg/L) and NiCl2
(0.02 g/L). The initial medium pH was constant, with values around
6.8.
The different media were always sterilized before use.
2.5.2. Biohydrogen dark fermentation experiments
Batch fermentation assays were performed in 159 mL glass
asks containing a phases volumetric ratio (gaseous head-
space:liquid fermentation medium) of 5 under orbital shaking
(220 rpm). The microalga biomass residues remaining after the
oil and pigments extraction (SFE3, SFE8, SFE9, Soxhlet), as well as
the microalgal biomass before oil extraction (whole/unbroken
and ground in ball mill as referred in Section 2.3), were used as
substrate, at 2.5 and 10 gdry biomass/L concentrations. The microalga
biomass was characterized in terms of volatile solids (VS) content.
Firstly, moisture was determined by drying the sample in an oven
at 105 C until constant weight; following, total ash was deter-
mined by incineration at 550 C in a mufe furnace. The VS (%,
w/w) was calculated by the difference between dry and ash
weights.
131
The bioreactors containing fermentation medium and microal-
gal biomass were sterilized in autoclave at 121 C/15 min before
starting the fermentative process.
The reactors and fermentation medium were purged with bub-
bling N2 for 2 min to eliminate O2, before inoculation with expo-
nentially grown E. aerogenes at 10% (v/v) (corresponding to
0.1 dry weight biomass/L culture media), and sealed with butyl
rubber stoppers. The process was carried out under orbital shaking
(220 rpm), at 30 C, for 6 h (previously determined as sufcient
time to reach equilibrium process). After 6 h, gas samples were
collected directly from the headspace of the reactor and analyzed
for H2 and CO2 content by gas chromatography. The nal concen-
trations of ethanol and organic acids were also determined by
HPLC.
A control assay, without microalgal addition, was also car-
ried out, not being detected any biohydrogen volumetric produc-
tion.
2.5.3. Analysis of the gas and liquid phases from fermentative process
The concentrations of H2 and CO2 were determined by gas chro-
matography in a Varian 430-GC gas chromatograph with TCD and a
fused silica column (select Permanent gases/CO2-Molsieve 5A/Bor-
abound Q tandem #CP 7430). The gases injector and column were
at 80 C and the detector at 120 C. Argon was the carrier gas at
32.4 mL/min ow rate. A gas-tight syringe was used to directly
withdraw 0.5 mL of the gaseous headspace which was injected into
the gas chromatograph.
The supernatants of the fermentation trials, obtained by centri-
fugation (15,000 rpm/2 min) and ltration of the liquid phases
(0.45 lm), were analyzed by HPLC in terms of ethanol and organic
acids concentration. The analysis was performed in a Merck Hit-
achi HPLC system (Darmstadt, Germany) equipped with an Aminex
HPX-87H column and a refraction index detector. The temperature
of the column was set to 50 C, and the eluent consisted of H2SO4
5 mM at 0.5 mL/min ow rate. The volume of the injection was
20 lL.
3. Results and discussion
3.1. Microalga oil and pigment content
3.1.1. Total pigment content in the microalga
The yield of pigment extraction from the whole biomass
(unbroken) was 0.23 0.02 mgtotal pigments/gdry biomass for the hot
extraction process, being null for the extraction in the cold condi-
tion. In addition, the yield of pigment extraction from the ground
biomass was 0.97 0.06 (hot extraction, with ethyl acetate);
1.34 0.02 and 1.47 0.10 mgtotal pigments/gdry biomass (cold extrac-
tion, with ethyl acetate and acetone, respectively).
The increase in the amount of pigment extracted for the ground
microalga is the result of the increase in the contact area between
microalga and the solvent.
The cold extraction leads to a higher yield than that of the hot
extraction process and it is the highest value obtained using ace-
tone. This could possibly be due to the fact that acetone is a more
polar solvent than ethyl acetate, hence more suitable for the polar
carotenoids present in the microalga. On the other hand, the high
temperature used in the hot extraction process (65 C) could pro-
mote the degradation of the carotenoids, since these compounds
are sensitive to heat and therefore a lower extraction yield was
obtained.
The cold extraction using ethyl acetate reported an increase of
38% in the extracted pigments compared with the hot extraction
and this process is benecial from an energetic and economical
point of view.
132 B.P. Nobre et al. / Bioresource Technology 135 (2013) 128136

3.1.1.1. Identication of carotenoids. TLC was used to identify the Table 3

pigments present in Nannochloropsis sp. The resulting pigments Fatty acid composition of Nannochloropsis sp. extracts obtained by Soxhlet extraction
with n-hexane and SFE.
from the cold acetone extraction showed the presence of astaxan-
thin, lutein/zeaxanthin, canthaxanthin and b-carotene, which is Fatty Soxhlet (n- SFE 9, extract 1 SFE 9, extracts 2, 3 and 4
consistent with data reported by Lubin et al. (2000). acid hexane) (%) CO2 (%) CO2 + EtOH (20 wt.%) (%)

HPLC analysis of carotenoids extracted from Nannochloropsis sp. C13:0 0.28 0.30 0.26
was carried out in order to identify the major pigments. The peaks C14:0 3.94 3.77 4.63
C15:0 0.52 0.55 0.57
were identied by comparing their retention time with those of the C16:0 34.99 34.69 34.60
standards of astaxanthin, lutein, trans-b-carotene, echinenone and C16:1 37.03 36.32 34.85
canthaxanthin. The HPLC chromatogram of the extract from cold C17:1 0.27 0.24 0.19
extraction with acetone conrms the presence of astaxanthin, can- C18:0 0.72 0.79 0.74
C18:1 17.63 17.81 16.71
thaxanthin, lutein/zeaxanthin and b-carotene (see Supplementary
C18:2 0.12 0.15 0.05
data). Further identication of peaks was based on a comparison C18:3 0.66 0.76 0.81
with published data by other researchers (Lubin et al., 2000). A C20:4 1.04 1.24 1.45
possible identication shows the presence of violaxanthin/neoxan- C20:5 2.67 2.33 4.75
thin, vaucheriaxanthin and chlorophyll a. Other peaks could be Others 0.12 1.04 0.37

mono and di-esters of astaxanthin. Table 2 shows the carotenoid

and chorophyll a prole for the extract obtained by cold extraction
with the acetone process and supercritical uid extraction (SFE). It
of palmitic, palmitoleic and oleic acids. The polyunsaturated fatty
can be seen that vaucheriaxanthin (30%) is the major carotenoid in
acid content observed was higher than the one specied in the bio-
the extract obtained by cold extraction with acetone, followed by
diesel standard (EN 14214, 2008) and this implies that these oils
astaxanthin (19%) and lutein/zeaxanthin (18%). Cantaxanthin and
should be mixed with other crops or microalgae oils to obtain a
b-carotene contribute with only 9 and 8%, respectively, for the total
biodiesel within the specications. The fatty acid content of the li-
percent of pigment content in the extracts.
pid extract obtained by Soxhlet was 54.0 2.8 gfatty acids/100 glipids,
which corresponds to 21.6 1.1 gfatty acids/100 gdry biomass.
3.1.2. Total lipid content of the microalga
The total lipid content for the Soxhlet extraction of Nannochlor-
opsis sp. using n-hexane was 40.7% which is within the range of 3.2. Supercritical uid extraction (SFE)
3168% reported by Spolaore et al. (2006).
Another solvent used for the Soxhlet extraction of lipids from 3.2.1. Extraction of lipids
the microalga Nannochloropsis sp. was ethanol. A value of 50.6% Supercritical CO2 extraction of lipids and pigments from ground
for the total lipid content was obtained, which is higher than that Nannochloropsis sp. biomass was carried out. The obtained results
obtained using hexane, possibly because with this solvent polar of the yield of the extraction as a function of solvent amount are
lipids are also extracted. shown in Figs. 2 and 3 for total lipids and total pigments,
Furthermore, extraction of lipids with the Bligh and Dyer meth- respectively.
od was also performed. The total amount of lipids obtained was The effect of pressure on the extraction yield, when pure CO2
25.3% which is lower than that reported by Spolaore et al. (2006) was used as a solvent, was studied at 40 C and pressures of 125,
and the ones obtained with Soxhlet extraction. 200 and 300 bar. From Fig. 2 it can be seen that the yield of extrac-
tion increases considerably with the increase in pressure from 125
to 200 bar and a further rise to 300 bar leads to a faster extraction
3.1.2.1. Oil characterization. The analysis of the fatty acids from the
and therefore, a higher yield is reached more quickly. This behavior
microalga Nannochloropsis sp. biomass, revealed the presence of
was expected since the solubility of the compounds increases with
C13:0, C14:0, C15:0, C16:0, C16:1, C17:1, C18:0, C:18:1, C18:2,
pressure (at constant temperature) due to the increase in the den-
C18:3, C20:4 and C20:5 (Table 3), which is in accordance with
sity of supercritical solvent. From Fig. 2 it can also be seen that,
studies produced by Gouveia and Oliveira (2009) and Andrich
when pure CO2 was used as a solvent, the yield of the extraction
et al. (2005). As it can be seen, the fatty acid composition of
was almost the same for both temperatures 40 and 60 C, at
Nannochloropsis sp. lipids obtained by Soxhlet extraction with n-
300 bar. On the other hand, the ow-rate showed to have no inu-
hexane and by SFE was similar, and the extracts were composed
ence on the extraction yield. This result could mean that the mass
transfer process is controlled by the equilibrium between the solid
and the uid phase. Therefore, the following experiments were car-
Table 2 ried out at a temperature of 40 C, a pressure of 300 bar, and a sol-
Relative proportion of each carotenoid and chlorophyll a to total pigment content, vent ow-rate of 0.62 g/min. At 300 bar, and for both
determined by HPLC analysis at 450 nm, of Nannochloropsis sp. extracts obtained by temperatures, the yield of extraction was almost 34 glipids/
cold extraction with acetone and supercritical uid extraction. 100 gdry biomass. This value was near that obtained by the Soxhlet
Pigment Cold extraction SFE9, extract SFE9, extract 2 extraction using hexane and in accordance with the results ob-
with acetone (%) 1 CO2 (%) CO2 + EtOH 20 wt.% tained by Andrich et al. (2005), showing that supercritical CO2
(%) can be a suitable solvent for the extraction of lipids from Nanno-
Violaxanthin/ 7.89 n.d. 13.20 chloropsis sp.
neoxanthin When ethanol was added as a co-solvent to supercritical CO2,
Astaxanthin 19.44 27.05 13.71
the yield of extraction increased (Fig. 2) and the rate of extraction
Vaucheriaxanthin 30.44 30.05 34.30
Lutein/ 18.04 12.81 22.35 was higher in the beginning of the extraction. This fact could pos-
zeaxanthin sibly be due to a higher solubility of the lipids in the modied CO2
Canthaxanthin 9.25 13.49 4.71 with ethanol, which would increase the rate of extraction. The
Chlorophyll a 2.91 2.84 3.37 higher yield of extraction obtained with CO2 doped with ethanol
b-Carotene 8.35 11.82 5.06
Others 3.62 1.84 3.25
could be due to the fact that polar lipids could also be extracted.
Moreover, the yield of extraction increased with the ethanol
 B.P. Nobre et al. / Bioresource Technology 135 (2013) 128136 133

Fig. 2. Yield of the extraction of lipids from Nannochloropsis sp. as a function of solvent amount for the several experimental conditions studied.

Fig. 3. Yield of the extraction of pigments from Nannochloropsis sp. as a function of solvent amount for the several experimental conditions studied.

weight fraction in the CO2 + ethanol mixture and a value of which is near the value obtained with the Soxhlet extraction with
45 glipids/100 gdry biomass was obtained when using 20 wt.% ethanol, this solvent.
134 B.P. Nobre et al. / Bioresource Technology 135 (2013) 128136
3.2.2. Extraction of pigments
In what concerns the extraction of pigments (Fig. 3), a similar
behavior was observed to that of the extraction of lipids. When
pure CO2 was used (Fig. 3), the extraction yield slightly increased
with the temperature from 40 C to 60 C, at 300 bar, and a value
of 55 mgtotal pigments/100 gdry biomass (at 60 C and 300 bar), could
be obtained, which is much lower than that reached with the cold
extraction using organic solvents. On the other hand, the yield of
extraction increased dramatically with pressure, being almost
insignicant at 125 bar (1.58 mgtotal pigment/100 gdry biomass). Fur-
thermore, the yield of the extraction of pigments was not affected
by the solvent ow-rate.
The use of ethanol as a co-solvent (Fig. 3) increased the yield of
extraction of pigments with a faster extraction of the compounds.
Furthermore, the increase in ethanol weight fraction improved the
yield of the extraction, and a value of 100 mgtotal pigments/
100 gdry biomass could be obtained when using 20 wt.% ethanol.
3.2.3. Fractionation of lipids and pigments
The extraction of lipids with supercritical CO2, at 40 C and
300 bar, lead to the recovery of almost all lipids (85% of the lipids
were extracted, value determined by comparison with the Soxhlet
extraction with n-hexane) and the pigments were extracted in low
amounts (34% of the total pigments were recovered). On the other
hand, the extraction using supercritical CO2 modied with ethanol
(20 wt.%) allowed almost 70% of the total pigments to be recov-
ered. Therefore a fractionation of lipids and pigments using CO2
combined with ethanol (20 wt.%) at 40 C and 300 bar was pro-
posed and tested. This was performed in order to obtain a fraction
of oil with the potential for biodiesel production and an oil en-
riched carotenoid which could be suitable for the food and/or
nutraceutical industries. Furthermore, a supercritical uid extrac-
tion and fractionation experiment was carried out, SFE9 (see Figs. 2
and 3), in which the rst extract (SFE9, extract 1) was obtained
using pure supercritical CO2 as the solvent and extracts 2, 3 and
4 were obtained using supercritical CO2 modied with ethanol
(20 wt.%). As a result, in extract 1 it was possible to obtain
33 glipids/100 gdry biomass, corresponding to a recovery of 78% of
the lipids, and 38 mgpigments/100 gdry biomass (e.g. 26% of the total
pigments). The obtained oil presented a pigment concentration of
0.11 gpigment/100 goil, a fatty acid content of 50.8 3.7 gfatty acid/
100 goil and a composition (see Table 3) similar to that obtained
by the Soxhlet extraction with n-hexane. Andrich et al. (2005) also
veried that there were no particular differences in the fatty acid
composition of the Soxhlet and supercritical uid extracts, con-
rming the use of SFE as an alternative for the extraction of lipids
for biodiesel production from microalgae.
On the other hand, in the fraction corresponding to extracts 2, 3
and 4 it was possible to obtain 12 glipids/100 gdry biomass, which cor-
responds to the remaining 28% of the oil in the biomass, as well as
53 mgpigments/100 gdry biomass, equivalent to the extraction of 50% of
the total pigments remaining in the biomass. Although the total
amount of extracted lipids was the same as in experiment SFE8,
where supercritical CO2 modied with ethanol (20 wt.%) was used
in the same operational conditions, the total amount of pigments
extracted was lower. This could possibly be due to the fact that
as the amount of available lipids for extraction is lower (lipids
remaining after SFE9 extract 1), the solubility of the carotenoids
in the supercritical solvent decreases, as the lipids act as an entrai-
ner for the extraction of pigments. Furthermore, when the micro-
alga biomass was exhausted of lipids the amount of extracted
pigments was very low, although 35% of the carotenoids still re-
mained in the microalga biomass (this is evident in extract 4).
Nevertheless, it was possible to obtain an oil with a pigment
concentration of 0.44 gpigment/100 goil (four times higher than that
obtained in extract 1, and twice the value obtained in SFE8), which
contained 48.9 gfatty acid/100 goil, with a prole (see Table 3) show-
ing an enrichment in good bioactive fatty acids (C20:5 5%). The
pigment composition of this oil (see Table 2) showed the presence
of astaxanthin, zeaxanthin/lutein, canthaxanthin and b-carotene in
a mixture corresponding to almost 50% of the total pigments,
which would make it suitable for its use in the food and/or nutra-
ceutical industries.
3.3. Biohydrogen fermentative production from the remaining biomass
of Nannochloropsis sp.
After the extraction of oils and pigments, under different condi-
tions, the remaining biomass was used as a substrate (2.5 and
10 gdry biomass/L) in a dark fermentation process by a strain of E. aer-
ogenes. Similar experiments were carried out using the microalgal
biomass before oil extraction (whole/unbroken and ground) at the
same initial concentrations for comparison.
The results showed that microalgal residues can be successfully
used to produce hydrogen by E. aerogenes, leading to specic
hydrogen yields which are higher than 47 mL H2/gdry biomass for
an initial concentration of 2.5 gdry biomass/L (Fig. 4). For all experi-
ments, the gas purity (H2/CO2) attained was between 1.0 and 1.4.
Although higher volumetric H2 production was obtained
when using 10 gdry biomass/L, the specic hydrogen yields were
signicantly lower (45%) than the ones obtained with only
2.5 gdry biomass/L. This behavior is in accordance with similar studies
carried out by our research team with the microalga Scenedesmus
obliquus (results not shown). This suggests that the E. aerogenes
strain used in this work shows an increased fermentative efciency
with a decreasing initial microalgal concentration, as a substrate.
Furthermore, Yang et al. (2011) also observed for S. obliquus lipid
extracted residues fermented by an anaerobic digested sludge, a
decrease in the hydrogen yield production from 40.3 to 25.6 mL/
VS when the initial substrate concentration increases from 4.5 to
45 g VS/L.
The process yield values achieved with the microalgal biomass
before and after oil extraction were of the same order of magni-
tude, with slightly higher values for the lipid-extracted residual
biomass. This is a very positive outcome, from a sustainable and
renewable point of view, considering the biorenery approach in-
tended in the present work. In fact, after the extraction of oil, suit-
able for biodiesel production, and added-value carotenoid
molecules, it is still possible to value the residual biomass obtain-
ing a clean energy carrier.
Experiments were also carried out using, as a substrate, whole
(before lipid extraction) Nannochloropsis sp., submitted or not to
a previous milling process, for comparison. The application of this
prior step is needed for an efcient oil extraction. The results ob-
tained showed no difference in the performance of bacteria, lead-
ing to very similar process yields (4748 mL H2/gdry biomass). This
can be advantageous, considering that it is possible to eliminate
an energy consuming processing step, when it is only intended
to produce biohydrogen.
As referred to earlier in this paper, the fermentation of lipid-
extracted microalgal biomass yielded higher hydrogen values
(5061 mL H2/gdry biomass) than the fermentation of the whole
biomass (4748 mL H2/gdry biomass). This can be related to a higher
effective sugar concentration (fermentable compounds) in lipid-
extracted residues due to the oil extraction (3445%).
Regarding the effect of the different oil and pigment extraction
methods applied, on the efciency of the leftover biomass, as a
substrate, it was veried that the highest hydrogen yield
(60.6 mL H2/gdry biomass) was attained using the remaining biomass
obtained by SFE9 extraction conditions, described in Section 3.2.
This is a benecial result considering that these extraction con-
ditions allow an optimal outcome both for biodiesel, pigment and
 B.P. Nobre et al. / Bioresource Technology 135 (2013) 128136
Fig. 4. Hydrogen specic yield (mL H2/g alga) obtained from the fermentation of Nannochloropsis sp. biomass (before and after oil and pigments extraction) by a strain of the
bacteria Enterobacter aerogenes.
hydrogen production, in an integrated biorenery process, show-
ing complementarity between both processes.
The use of microalgal biomass as a substrate for fermentative
hydrogen production has recently been published, substantiating
the emerging potential for this process.
A previous study was carried out by the authors (Ferreira et al.,
2011) that used E. aerogenes and the biomass of the cyanobacteria
Anabaena sp. after photoautotrophic H2 production (10 gdry biomass/L),
as fermentative bacteria and substrate, respectively. Using the
same operational and physico-chemical condition of the present
study, the yield of H2 production was 1.26 kg H2/kgdry biomass or
15.2 mL/gdry biomass, which are much lower (about half) than that
obtained with Nannochloropsis sp. (Fig. 4).
Efremenko et al. (2012) used Clostridium acetobutylicum immo-
bilized cells to ferment various microalgae (Arthrospira platensis,
Nannochloropsis sp., Dunaliella tertiolecta, Galdieria partita, Chlorella
vulgaris, Cosmarium sp., Nostoc sp.). The highest productivity
attained was for Nannochloropsis sp. with a result of 0.35 mmol
H2/L medium/h. Our results are far better since 1.07 mmol H2/L
medium/h productivity was obtained when using 2.5 g
Nannochloropsis with a shorter thermal treatment and without
the addition of an acid.
S. obliquus lipid extracted microalgal biomass residues, submit-
ted to different pre-treatments (alkaline and thermal) yielded a
maximum 45.5 mL H2/g VS when inoculated with an anaerobic di-
gested sludge by Yang et al. (2010). In the present study, the max-
imum yield attained was 97.7 mL H2/g VS, considering a total
volatile solid content of 62% VS (w/w) in the samples after oil
extraction.
In other studies, C. vulgaris and D. tertiolecta yielded 10.8 mL H2/g
VS and 12.6 mL H2/g VS, respectively, using an anaerobic digested
sludge (not treated) as inoculum (Lakaniemi et al., 2011); while
Liu et al. (2012) obtained 81 mL/g alga along 65 h fermentation
of C. vulgaris (acid hydrolyzed) by Clostridium butyricum.
Hydrogen production is usually accompanied by organic acids
and solvent production. These intermediate products directly re-
ect changes in metabolic pathways (Khanal et al., 2004), with
an impact on the biohydrogen produced. In the end of the process,
liquid samples were collected and analyzed to determine their or-
ganic acids and solvent contents. In all trials only trace amounts of
135
ethanol were detected (0.10.6 g/L). This behavior was also ob-
served by Yang et al. (2010), using S. obliquus lipid extracted resi-
dues as a substrate for a fermentative process by an anaerobic
digested sludge. Concerning organic acids, all nal assayed concen-
trations were also veried to be very low, namely succinate
(0.10.2 g/L), acetate (0.20.4 g/L) and propionate (0.4 g/L, from
fermentation media). Propionic acid was not produced along fer-
mentation as it was already present in the liquid media (0.5 g/L).
These results are consistent with the slight variations observed
for initial and nal media pH values which were around 6.66.8.
The results obtained in this work prove that Nannochloropsis sp.
has the potential to serve as an effective feedstock for dark fermen-
tative H2 production. Moreover, the conversion of lipid-extracted
microalgal biomass residue into hydrogen plays the dual role in a
sustainable microalgal biodiesel industry.
4. Conclusions
This study highlighted the potential of the microalga Nanno-
chloropsis sp. as a feedstock for the production of fatty acids for
biodiesel, high added-value compounds (carotenoids) and biohy-
drogen, in a biorenery context.
The more effective extraction method was CO2 SFE plus ethanol
(20 wt.%), being possible to extract 45 g/100 gdry biomass of lipids
and recover 70% of the pigments.
The remaining biomass was even harnessed as substrate to pro-
duce biohydrogen, by dark fermentation, at a yield of 60.6 mL/
gdry biomass.
The assessment of energy balance, CO2 emissions and economic
impact associated to this global process, is an ongoing parallel
study.
Acknowledgements
This research was part of Microalgae as a sustainable raw mate-
rial for biofuel production (Biodiesel, Bioethanol, Bio-H2 and Biogas)
(PTDC/AAC-AMB/100354/2008), a project sponsored by the Portu-
guese Foundation for Science and Technology (Fundao para a
Cincia e a Tecnologia FCT). Beatriz P. Nobre thanks FCT for the
136 B.P. Nobre et al. / Bioresource Technology 135 (2013) 128136
research grant (SFRH/BPD/42004/2007). F. Villalobos thank IPN
(Project SIP 20110775) and CONACyT for a graduate scholarship.
The authors would also like to thank Dr. Stephanie Seddon-Brown
for the English proofreading and the reviewers for their valuable
comments that have improved the manuscript quality.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.biortech.2012.11.
084.
References
Amaro, H.M., Guedes, A.C., Malcata, F.X., 2011. Advances and perspectives in using
microalgae to produce biodiesel. Applied Energy 88, 34023410.
Andrich, G., Nesti, U., Venturi, F., Zinnai, A., Florentini, R., 2005. Supercritical uid
extraction of bioactive lipids from the microalga Nannochloropsis sp.. European
Journal of Lipid Science and Technology 107, 381386.
Bai, M., Cheng, c., Wan, H., Lin, Y., 2011. Microalgal pigments potential as
byproducts in lipid production. Journal of the Taiwan Institute of Chemical
Engineers 42, 783786.
Bligh, G.E., Dyer, J.W., 1959. A rapid method of total lipid extraction and
purication. Canadian Journal of Biochemistry and Physiology 3, 912917.
Boussiba, S., Vonshak, A., Cohen, Z., Avissar, Y., Richmond, A., 1987. Lipid and
biomass production by the halotolerant microalga Nannochloropsis salina.
Biomass 12, 3747.
Bruno, T., Castro, C.A.N., Hammel, J.F., Palavra, A.M.F., 1993. Supercritical uid
extraction of biological products. In: Kennedy, J.F., Cabral, J.M.S. (Eds.), Recovery
Processes for Biological Materials. J. Wiley & Sons, Chichester.
Efremenko, E.N., Nikolskaya, A.B., Lyagin, I.V., Senko, O.V., Makhlis, T.A., Stepanov,
N.A., Maslova, O.V., Mamedova, F., Varfolomeev, S.D., 2012. Production of
biofuels from pretreated microalgae biomass by anaerobic fermentation with
immobilized Clostridium acetobutylicum cells. Bioresource Technology 114,
342348.
EN 14103, 2003. Fat and oil derivatives fatty acid methyl esters (FAME)
determination of ester and linolenic acid methyl ester contents.
EN 14214, 2008. Automotive fuels fatty acid methyl esters (FAME) for diesel
engines requirements and test methods.
Ferreira, A.F., Marques, A., Batista, A.P., Marques, P.A.S.S., Gouveia, L., Silva, C., 2012.
Biological hydrogen production by Anabaena sp. yield, energy and CO2
analysis including fermentative biomass recovery. International Journal of
Hydrogen Energy 37, 179190.
Ferreira, A.F., Ribau, J.P., Silva, C.M., 2011. Energy consumption and CO2 emissions of
potato peel and sugarcane biohydrogen production pathways, applied to
Portuguese road transportation. International Journal of Hydrogen Energy 36,
1354713558.
Gouveia, L., 2011. Microalgae as a feedstock for biofuels. Springer Briefs in
Microbiology. Springer, Heidelberg, Germany.
Gouveia, L., Marques, A.E., Silva, T.L., Reis, A., 2009. Neochloris oleabundans UTEX
#1185: a suitable renewable lipid source for biofuel production. Journal of
Industrial Microbiology and Biotechnology 36, 821826.
Gouveia, L., Nobre, B., Mrejen, M., Cardoso, A., Mendes, R., 2007. Food functional oil
coloured by pigments extracted from microalga with supercritical CO2. Food
Chemistry 101, 717723.
Gouveia, L., Oliveira, A.C., 2009. Microalgae as a raw material for biofuels
production. Journal of Industrial Microbiology and Biotechnology 36, 269274.
Hallenbeck, P.C., 2009. Fermentative hydrogen production: principles, progress, and
prognosis. International Journal of Hydrogen Energy 34, 73797389.
Khanal, S.K., Chen, W.H., Li, L., Sung, S., 2004. Biological hydrogen production:
effects of pH and intermediate products. International Journal of Hydrogen
Energy 29, 11231131.
Lakaniemi, A.M., Hulatt, C.J., Thomas, D.N., Tuovinen, O.H., Puhakka, J.A., 2011.
Biogenic hydrogen and methane production from Chlorella vulgaris and
Duanliella tertiolecta. Biotechnology of Biofuels 4, 34.
Lepage, G., Roy, C., 1986. Direct transesterication of all classes of lipids in a one-
step reaction. Journal of Lipid Research 27, 114119.
Liau, B.C., Shen, C.T., Liang, f.P., Hong, S.E., Hsu, S.L., Jong, T.T., Chang, C.M.J., 2010.
Supercritical uids extraction and antisolvent purication of carotenoids from
microalgae and associated bioactivity. Journal of Supercritical Fluids 55, 169
175.
Lindblad, P., Christensson, K., Lindberg, P., Pinto, F., Tsygankov, A., 2002.
Photoproduction of H2 by wildtype Anabaena PCC 7120 and a hydrogen
uptake decient mutant: from laboratory experiments to outdoor culture.
International Journal of Hydrogen Energy 27, 12711281.
Liu, C.H., Chang, C.Y., Cheng, C.L., Lee, D.J., Chang, J.S., 2012. Fermentative hydrogen
production byClostridium butyricum CGS5 using carbohydrate-rich microalgal
biomass as feedstock. International Journal of Hydrogen Energy 37, 15458
15464.
Lubin, L.M., Montero, O., Moreno-Garrido, I., Huertas, I.E., Sobrino, C., Gonzlez-
del-Valle, M., Pars, G., 2000. Nannochloropsis (Eustigmatophyceae) as source
of commercially valuable pigments. Journal of Applied Phycology 12,
249255.
Macas-Snchez, M.D., Mantell, C., Rodriguez, M., Martinez de la Ossa, E., Lubian,
L.M., Montero, O., 2005. Supercritical uid extraction of carotenoids and
chlorophyll. Journal of Food Engineering 66, 245251.
Mendes, R.L., Coelho, J.P., Fernandes, H.L., Marrucho, I.J., Cabral, J.M.S., Novais, J.M.,
Palavra, A.F., 1995. Applications of supercritical CO2 extraction to microalgae
and plants. Journal of Chemical Technology and Biotechnology 62, 5359.
Moazami, N., Ashori, A., Ranjbar, R., Tangestani, M., Eghtesadi, R., Nejad, A.S., 2012.
Large-scale biodiesel production using microalgae biomass of Nannochloropsis.
Biomass & Bioenergy 39, 449453.
Nobre, B., Marcelo, F., Passos, R., Beiro, L., Palavra, A., Gouveia, L., Mendes, R., 2006.
Supercritical carbon dioxide extraction of astaxanthin and other carotenoids
from the microalga Haematococcus pluvialis. European Food Research and
Technology 223, 787790.
Quinn, J.C., Yates, T., Douglas, N., Weyer, K., Butler, J., Bradley, T.H., Lammers, P.J.,
2012. Nannochloropsis production metrics in a scalable outdoor photobioreactor
for commercial applications. Bioresource Technology 117, 164171.
Razon, R.F., Tan, R.R., 2011. Net energy analysis of the production of biodiesel and
biogas from the microalgae: Haematococcus pluvialis and Nannochloropsis.
Applied Energy 88, 35073514.
Rodriguez-Amaya, D.B., 2005. A Guide to Carotenoid Analysis in Foods. ILSI Press,
Washington, USA, ISBN 1-57881-072-8.
Spolaore, P., Joannis-Cassan, C., Duran, E., Isambert, A., 2006. Commercial
applications of microalgae. Journal of Bioscience and Bioengineering 101, 87
96.
Stahl, W., Sies, H., 2005. Bioactivity and protective effects of natural carotenoids.
Biochemistry Biophysics Acta 1740, 101107.
van Vooren, G., Grand, F., Legrand, J., Cuin, S., Peltier, G., Pruvost, J., 2012.
Investigation of fatty acids accumulation in Nannochloropsis oculata for
biodiesel application. Bioresource Technology 124, 421432.
Yang, Z., Guo, R., Xu, X., Fan, X., Li, X., 2010. Enhanced hydrogen production from
lipid-extracted microalgal biomass residues through pretreatment. Interna-
tional Journal of Hydrogen Energy 35, 96189623.
Yang, Z., Guo, R., Xu, X., Fan, X., Luo, S., 2011. Fermentative hydrogen production
from lipid-extracted microalgal biomass residues. Applied Energy 88, 3468
3472.