Bioresource Technology 132 (2013) 285292

Contents lists available at SciVerse ScienceDirect

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

The culture of Chlorella vulgaris in a recycled supernatant: Effects on

biomass production and medium quality
F. Hadj-Romdhane a, X. Zheng c, P. Jaouen a, J. Pruvost a, D. Grizeau a, J.P. Crou c, P. Bourseau a,b,
a

LUNAM Universit, Universit de Nantes, CNRS, GEPEA UMR 6144, bd de lUniversit, CRTT-BP 406, 44 602 Saint-Nazaire Cedex, France
UEB-Universit de Bretagne Sud, LIMATB, rue de Saint-Maud, BP 92 116, 56 321 Lorient Cedex, France
c
King Abdullah University of Science and Technology (KAUST), Water Desalination and Reuse Center, Saudi Arabia
b

h i g h l i g h t s
" Productivity of Chlorella vulgaris was unaffected during medium recycling over 63 days.
" Organic matter excretion and accumulation by C. vulgaris was clearly observed.
" Molecular weight of excreted molecules ranged in two fractions: 13 kDa and > 40 kDa.
" The major excreted product (BP) was identied as polysaccharides in initial phase.

a r t i c l e
Article history:
Received 4 October
Received in revised
Accepted 5 January
Available online 22

i n f o

2012
form 3 January 2013
2013
January 2013

Keywords:
Chlorella vulgaris
Large-scale culture
Water recycling
Excreted organic matter

a b s t r a c t
Reusing supernatant of microalgae culture medium can have inhibitory or toxic effects on the biomass
production because of the release of organic metabolites by cells in the culture medium during their
growth. This work investigated the impact of Chlorella vulgaris medium recycling on culture productivity,
cells quality and accumulation of excreted metabolites in the culture medium. No signicant impact on
the C. vulgaris growth was observed after 63 days of recycling, the productivity remained stable at around
0.55 kg m3 day1. Organic matters accumulated in supernatant were identied as biopolymers (BP) poor
in nitrogen and with a size above 40 kDa (probably polysaccharides), and small organic molecules (SOM)
richer in nitrogen with a molecular size ranging from 1 to 3 kDa. The concentration of biopolymers in the
supernatant increased till to a maximum and then decreased, possibly consumed by bacteria, whereas
small organic compounds accumulated in the medium.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
Microalgae have a great potential for various applications
including the production of compounds for food, feed and aquaculture, of higher value products for pharmaceutical and cosmetic
industries (Spolaore et al., 2006; Brennan and Owende, 2010).
Some microalgae species containing high lipid contents are being
investigated for biodiesel production (Pruvost et al., 2011). For recent years, large scale cultures are coupled with wastewater treatment or CO2 xation as source of nutrients. Residues of biomass
after metabolites extraction could be converted into biomethane
or bioethanol (Harun et al., 2010).
Microalgae production has greatly increased in recent years,
with a biomass production raised from about 3500 tons per year
in 2004 to about 10,000 tons per year nowadays (Jenck et al.,
Corresponding author at: LUNAM Universit, Universit de Nantes, CNRS,
GEPEA UMR 6144, bd de lUniversit, CRTT-BP 406, 44 602 Saint-Nazaire Cedex,
France. Tel.: +33 (0)2 97 87 45 31; fax: +33 (0)2 97 87 45 00.
E-mail address: patrick.bourseau@univ-ubs.fr (P. Bourseau).
0960-8524/$ - see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.biortech.2013.01.025

2011). Although open ponds have very low average productivity

and biomass concentration, below 10 g m2 day1 and 0.5 kg m3,
respectively, they are largely used for industrial exploitation because they are cheaper to build and operate than closed photobioreactors (Tredici, 2010). As open ponds volume is often several
thousand cubic meters, the culture medium is an important item
in microalgae production costs because huge amounts of water
and nutrients are used. Recycling culture media is therefore a
key issue for the development of large-scale cultures to minimize
water and nutrients consumption (Yang et al., 2011). Lvansky
et al. (1996) reported that recycling the culture medium during
Scenedesmus obliquus growth allowed saving 63% of water volume
and 16% of nutrients. Hadj-Romdhane et al. (2012) estimated that
using an optimized culture medium of Chlorella vulgaris with nutrient inputs limited to physiological requirements of microalgae
would save about 75% of water and 62% of nutrients. Only a few attempts to grow microalgae with medium recycling were however
reported. For example, Lvansky et al. (1996) grew S. obliquus and
Rodol et al. (2003) Nannochloropsis sp. in fed-batch mode.

286

F. Hadj-Romdhane et al. / Bioresource Technology 132 (2013) 285292

It is recognised that organic metabolites are naturally excreted

by microalgae during the growth (Fogg, 1971, 1983; MorineauThomas et al., 2002; Yu et al., 2012) or suddenly released when cell
lysis occurs (Harun et al., 2010). Presence in culture media of
growth inhibitors has already been observed (Pratt and Fong,
1940; Pratt, 1942) and toxic effects have been reported as due to
fatty acids and substances derived by oxidation (Blanchemain
et al., 1994; Wu et al., 2006; Bosma et al., 2008) or occurred in
the case of high efciency photobioreactor with high cells concentration (Javanmardian and Palsson, 1991; Richmond and Zou,
1999; Richmond et al., 2003). When culture medium is recycled,
organic metabolites accumulate and lead to a decrease in biomass
productivity (Lvansky et al., 1996; Rodol et al., 2003). Moreover,
when microalgae are grown with nitrogen starvation to enhance
lipids production, it has been reported that the nutrient limitation
is a stress factor enhancing metabolites excretion (Granum et al.,
2002). The presence of organic matter in the supernatant seems
thus to be unavoidable although undesirable. Moreover, when culture medium supernatant is recycled, the counter ions of nutrients
(e.g., Cl, Na+, Ca2+) will not or poorly be assimilated by microalgae
and will accumulate in the supernatant, leading to a change in
medium salinity which may negatively affect the biomass growth
(Alyabyev et al., 2007). To avoid such mineral accumulation,
Hadj-Romdhane et al. (2012) developed a highly assimilable minimal growth medium (HAMGM) in which counter-ions were replaced by ammonium ion (NH4+).
On the basis of these results, the present work aimed to monitor
the accumulation of organic matter excreted in the culture medium during microalgae growth with recycling of the supernatant
recovered after cells harvesting. Microalgae were grown in HAMGM in order to avoid possible effects resulting from minerals accumulation. Recycling experiments were conducted in a controlled 1L-photobioreactor over nine weeks (63 days). Organic matter (OM)
in supernatant was characterized, the impact of OM accumulation
on culture productivity and cells quality was investigated and the
evolution of OM with recycling was discussed.
2. Methods
2.1. Strain and medium
Recycling experiments were conducted with C. vulgaris (211/19
SAG) in the highly assimilable minimal growth medium (HAMGM)
developed by Hadj-Romdhane et al. (2012). The composition was
as followed (mmol L1): NH4HCO3, 13.9; MgSO47H2O, 0.76; KH2PO4, 0.53; (NH4)2HPO4, 1.24; CaCl22H2O, 0.22 and 0.5 mL L1 of
Hutners trace elements solution (Hutner et al., 1950). The growth
medium was prepared using distilled water.
2.2. Airlift photobioreactor
Experiments were conducted in the photobioreactor (PBR) described in Hadj-Romdhane et al. (2012) and used to develop the
HAMGM medium (see Fig. 1). Photon Flux Density (PFD) was measured by a quantum sensor (Li-250A, Li-COR, Lincoln, NE) and was
set at 270 lmol m2 s1. pH and temperature were controlled by a
pH/temperature probe (sensor Mettler Toledo SG 3253) monitored
by the LabVIEW acquisition software. pH was regulated at 7.5 by
automatic CO2 injection and temperature was maintained at
25 C by air blowing with a blower xed onto the stainless steel
rear side of the PBR. The PBR was operated in a continuous mode
with a constant dilution rate using a membrane dosing pump
(Stepdos, 08/RC, KNF Neuberger). The input ow rate (Q0) was
daily controlled by following the time course of the harvested volume VH (Q0 = VH/t, with t the harvesting time).

2.3. Experimental set-up and recycling protocol

The experimental set-up presented in Fig. 1 was already described in Hadj-Romdhane et al. (2012). A protocol representative
of real operating conditions was retained. Culture medium was fed
into the PBR at a ow rate Q0 = 0.432 L day1 (5  109 m3 s1) corresponding to a retention time sPBR (sPBR = VPBR/Q0) of 2.31 days
(for a PBR volume VPBR of 1 L). The culture was continuously harvested by overowing and stored for 3.5 days before treatment in
an illuminated closed bottle mixed twice a day. In our experimental conditions, the evolution of the stored culture was considered
as negligible as shown by Hadj-Romdhane et al. (2012). In the rst
step, the PBR was maintained in continuous operation with no
recycling until a quasi-steady-state conditions of biomass productivity was reached after 20 days, then recycling was started. Every
3.5 days (twice a week), the entire stored harvested culture
(VH = 1.52 L) was centrifuged (17,600g, 15 min, 15 C) to separate
the supernatant from the biomass. The supernatant was then
replenished with NH4HCO3, MgSO47H2O, KH2PO4, (NH4)2HPO4,
and CaCl2 to offset the consumption of nutrients in the PBR, and
then reintroduced into the feeding medium bottle with no further
treatment. A sufcient volume of feeding medium (VFB = 4 L) was
used to avoid interruption in the PBR feeding. The renewing time
of the feeding bottle sFB (sFB = VFB/Q0) was equal to 9.26 days
(4 L/0.432 L day1). The recycling protocol thus allowed the PBR
to be run in a semi-discontinuous mode with permanent feeding
in recycled medium and to maintain a quasi-steady-state to investigate the system stability.
For each recycling operation, the culture medium was sampled
to characterize the organic matter released in the supernatant (see
Section 2.4) and to measure biomass concentration (CX), and chlorophyll content used as an indicator of the biomass quality. The
volumetric productivity PX was then calculated as:

PX C X  D C X =sPBR

With D = Q0/VPB = 1/sPBR, the dilution rate xed by the feeding

ow rate.
2.4. Analytical methods
Dry biomass concentration was determined by gravimetry, pigment content by UVVis photometry after methanol extraction,
and ion concentrations in the supernatant by ionic chromatography. These analytical methods have been already presented in
Hadj-Romdhane et al. (2012).
Organic matter (OM) in supernatant was characterized by measuring dissolved organic matter (DOM) and total organic matter
(TOM), as described below:
2.4.1. Measurement of dissolved organic matter (DOM)
Samples containing only DOM were obtained by ltration
through a glass-ber lter (Sartorius, 0.45 lm). DOM content was
determined on the sample by combustion. DOM was also fractionated by high performance (liquid) size exclusion chromatography
(HPSEC) and the fractions were characterized by ultraviolet absorbance (UV), and by on-line organic carbon (DOC) and organic nitrogen (DON) detectors.
2.4.1.1. Measurement of the whole dissolved organic carbon by
combustion. DOC was measured by a high temperature combustion
analyser (Shimadzu TOC-5000/5000A). The measure consisted in
oxidizing carbon into CO2 which was carried into a sample cell
set in a non-dispersive infrared gas analyzer, by a carrier gas (here,
pure O2). DOC concentration from combustion was noted
[DOC]comb to differentiate it from HPSEC. [DOC]comb was determined by subtracting the concentration of dissolved inorganic car-

287

F. Hadj-Romdhane et al. / Bioresource Technology 132 (2013) 285292

6

2
4

Culture
medium inlet

7.00
25.0

pH
C

Recovery of
culture medium

Biomass

Harvesting

Sterile air inlet

(mixing)

CO2 injection
(pH regulation)

Supernatant

CO2

Culture medium
recycling
Nutrients
supplement

1. Culture medium feed bottle (VFB = 4 L)

4. pH and temperature probe

7. Solenoid valve

2. Membrane metering pump Stepdos

5. Culture medium harvesting bottle (VH = 1.52 L)

8. Bottle of CO2

3. Airlift PBR (VPBR = 1 L)

6. pH and temperature transmitter

Fig. 1. Schematic diagram of pilot plant for the continuous culture of Chlorella vulgaris (q0 = 270 lmol m2 s1, pH = 7.5, T = 25 C). The PBR used for these experiments is a
at-panel airlift having an optical path (e) of 0.03 m, a working volume (VPBR) of 1 L, and a specic illuminated surface (alight) of 33.3 m1. The PBR was built in a transparent
material, except for the rear side in stainless steel (type 316L). The dashed line represents the medium recycling operation.

bon [DIC]comb from the concentration of dissolved total carbon concentration [DTC]comb . [DIC]comb was measured after the acidication of the sample with phosphoric acid (25% w/w) which converted
all the inorganic carbon into CO2. The [DTC]comb was measured
after the combustion of the sample at 680 C. [DOC]comb is given
in mg C L1.
2.4.1.2. Size exclusion chromatography with UV, carbon and nitrogen
detectors. The HPSEC system used (DOC-LABOR Dr. Huber, Germany) was equipped with a size exclusion column (SEC) TSK HW
50S (250  20 mm, 3000 theoretical plates, MW was calibrated
using polyethylene glycol (PEG) in the scope of 0.140 kDa, Tosoh,
Japan) and three on-line detectors, a xed 254 nm wavelength UV
detector (UVD254), an organic carbon detector (OCD) and a nitrogen
detector (OND). The mobile phase was a phosphate buffer with a
pH of 6.85 (2.5 g L1 of KH2PO4 and 1.5 g L1 of Na2HPO42H2O),
with a ow rate of 1.1 mL min1. Organic molecules are separated
through the column basically according to their hydrodynamic
molecular size, larger molecules being excluded earlier than smaller ones. Fractions obtained out of the column pass through the
UVD254 and then through the OCD and the OND. Detectors operation is detailed in Huber et al. (2011). The UVD254 responds to aromatic and unsaturated structures and the OCD to the content of
organic compounds. The OND responds to both organic and inorganic nitrogen, but the late elution of inorganic salts typical in
HPSEC (e.g. elution time of NH4+ and NO3 is round at 75 min
and 6070 min, respectively) allows the quantication of nitrogen
in biopolymers and natural organic matter in the presence of inorganic nitrogen (Huber et al., 2011). Thus, the system produces

three chromatograms at each run corresponding to the carbon content (dissolved organic carbon, DOC), the nitrogen content (dissolved organic nitrogen, DON), and UV254 absorbance proles of
the dissolved organic matter. The column is bypassed by a small
ow directly towards the detectors, allowing the direct determination of the UVD254 signal, the DOC and the DON of the unfractionated sample.
As shown in Section 3.2.2, the LC-OCD chromatogram of the
recycled supernatant DOM always showed two well separated
peaks corresponding to biopolymers (BP) and small organic compounds (SOM). Thus, the useful quantitative information provided
by HPSEC was given by values of the following parameters:
- [DON]BP and [DON]SOM, the sample concentrations in organic
nitrogen bound, respectively, to the BP and to the SOM, computed by integrating the OND signal, respectively, over BP and
SOM peaks (contribution of inorganic nitrogen to these fractions can be excluded due to their much longer elution time
(>60 min) compared to BP and SOMs,),
- [DOC]BP and [DOC]SOM, the similar concentrations as above, but
in organic carbon,
- [DOC]chrom, the sample concentration in total organic carbon
computed by integrating the OCD signal over the whole chromatogram. [DOC]chrom and [DOC]BP + [DOC]SOM remained very
close, the difference between the two values for all supernatant
being spread over the range [0.29.6%] (results not shown). This
is consistent with the fact that all OM in supernatant are either
SOM or BP as revealed by the chromatograms on Fig. 4 (see
Section 3.2.2),

288

F. Hadj-Romdhane et al. / Bioresource Technology 132 (2013) 285292

- DOCchrom
bypass , the sample concentration in total organic carbon
detected by the OCD on the small ow bypassing the column,
difference between DOCchrom
bypass ,
- and [DOC]chrom gives hydrophobic DOC,
- the abundance of BP and SOM in the sample was expressed with
respect to the DOC detected in the BP and SOM fractions:

DOCBP or SOM
 100%
DOCBP DOCSOM
N=CBP

or SOM

DONBP or SOM
DOCBP or SOM

- the N/C ratio in BP or SOM, can be used to evaluate the content in

polysaccharide-like or protein-like substances (Zheng et al., 2010).
2.4.2. Quantication of the total organic matter
Total organic matter (TOM) including particulate and dissolved
organic matter was quantied by the measurement of total sugars,
as carbohydrates are the major products excreted by phytoplankton (Myklestad, 1995; Biddanda and Benner, 1997).
Total sugars were measured by the phenolsulphuric acid colorimetric method described by Dubois (1956). In the presence of
concentrated sulfuric acid (98%) and phenol (50 g L1), sugars
and their derivatives are transformed into furfural or hydroxymethyl furfural. The orange colour formed is stable for several
hours. The calibration curve was performed with D-glucose.
An aliquot of 0.5 mL of sample was fed into a 15 mL-polypropylene tube, 0.5 mL of phenol (50 g L1) and 2.5 mL of sulfuric acid
(98%) were then added by taking care not to mix samples. After
incubation for 10 min at room temperature, the tubes were vigorously vortexed for 10 s. They were then immersed in a water bath
at 35 C for 30 min. The obtained compounds were quantied by
spectrophotometry (PerkinElmer Lambda10) by reading the
absorbance at 484 nm, the value being noted [PS]TOM.
3. Results and discussion
3.1. Impact of recycling on biomass production
Recycling of C. vulgaris culture medium was conducted during
nine weeks (63 days), corresponding to 18 harvests and medium
recycling operations. Fig. 2 represents the time course of the bio-

mass productivity and the total chlorophyll content during the

whole duration of the experiment including the 20 last days of
the stabilization phase. A small variation of the dilution rate was
observed rate during the 63 days-recycling phase because of uctuations in the feeding ow rate. This affected the biomass concentration as it can be seen on Fig. 2: when the ow rate decreased,
the dilution rate (D) decreased also and the biomass concentration
(CX) increased, and vice versa.
Biomass volumetric productivity PX was not signicantly impacted by medium recycling. PX was maintained at around
0.56 0.07 kg m3 day1 during the 18 recycles. Biomass quality
seemed unaffected as the absorbance ratio (A480/A665) linked to
carotenoid and chlorophyll contents remained constant (results
not shown), as well as the total chlorophyll content in the biomass
with a mean value of 5.5 0.8% (w/w). Thus, a long-term recycling
process revealed here to have no negative impact on the biomass
growth rate and on cells quality. These results are in accordance
with those observed for continuous cultures of C. vulgaris grown
in a recycled medium during 56 days (Hadj-Romdhane et al.,
2012), and of Chlorella pyrenoidosa during 72 days (Leone, 1963).
In both cases, biomass concentration in PBR was below 1.5 g L1
and the authors reported that no inhibition of biomass growth occurred, suggesting there was no inhibiting molecules in the medium or their inuence is negligible. The presence of excreted
growth inhibitors in supernatant is largely mentioned in the literature, particularly in high cell density cultures as reported for Nannochloropsis sp. by Richmond and Zou (1999) and for C. vulgaris by
Javanmardian and Palsson (1991). Rodol et al. (2003) reported
that in fed-batch mode, growth of Nannochloropsis sp. in recycled
medium was completely halted when cells concentration reached
about 3 g L1 (after 16 days of growth), whereas in the fresh medium the biomass concentration increased above 5 g L1 (after
28 days of growth). According to the authors, Nannochloropsis sp.
released soluble inhibitors (not identied) and particulate organic
matter in the supernatant that seemed to be cell walls. Lvansky
et al. (1996) observed also a decrease in the productivity of S. obliquus in fed-batch mode with culture medium recycling, and they
attributed the inhibition to the accumulation of microalgae metabolites. Biomass concentrations achieved in these studies were also
above 3 g L1. This was also mentioned by Yu et al. (2012) who observed that excretion increased with the microalgae growth rate
and with the cell concentration. As a conclusion, we can assume
that, because biomass concentration remained below 1.5 g L1 in

Fig. 2. Experimental time course of biomass concentration (CX), productivity (PX) and total chlorophyll content, during the recycling of Chlorella vulgaris growth medium
(q0 = 270 lmol m2 s1, pH = 7.5, T = 25 C).

F. Hadj-Romdhane et al. / Bioresource Technology 132 (2013) 285292

289

Fig. 3. Time course of total (TOM) and dissolved (DOM) organic matter concentrations in the medium during Chlorella vulgaris growth medium recycling. DOM was quantied
by its DOC concentration measured by combustion, [DOC]comb, and TOM by its polysaccharides content [PS]TOM.

our study, excretion of toxic or inhibiting molecules was limited,

and the biomass was therefore found unaffected by medium
recycling.
Although no decrease in culture productivity and no degradation in cells quality were noted, some biomass precipitates occurred in the culture medium. They can be attributed to the
formation of a microalgal biolm observed on the optical front face
of the PBR. Microscopy observations of C. vulgaris cultures indicated that cells were well dispersed in a fresh medium but that
some cells aggregates appeared in a 63-days-old culture in a recycled medium. Cells aggregation could not result here from change
in medium ionic strength because a highly assimilated medium
was used. Accumulation of ions was thus negligible and ions concentration remained almost constant as it was demonstrated in
Hadj-Romdhane et al. (2012). It can be hypothesized that aggregation resulted here from some biological or physical modications
of cells or from exocellular products which create linkage between
the cells (Morineau-Thomas et al., 2002). In addition, it was observed that the cells size increased in the culture with recycled
medium, which could also be explained by a physiological stress
of the culture.
As a conclusion, the results seem to prove that although the
production of C. vulgaris was not affected by medium recycling, a
drift in biomass quality took place reecting a possible beginning
of culture deterioration.
3.2. Impact of recycling on supernatant composition
3.2.1. Evolution of global organic matter in supernatant
Several changes of the supernatant quality were observed during the 63 days of the recycling experiment. Firstly, the colour
changed with the number of recycles from light to dark yellow.
This trend was already observed during the growth with medium
recycling of C. pyrenoidosa (Leone, 1963) or Nannochloropsis sp.
(Rodol et al., 2003). These authors attributed this fact to biomass
fragments, or the release in the medium of soluble organic matter
such as chlorophyll or other DOM resulting either from natural
microalgae excretion or from sudden cells lysis during centrifugation. Moreover, foam formed at the culture medium surface, suggesting also the excretion and the accumulation in the
supernatant of surfactants such as proteins, amino acids, lipids,
and moreover of polysaccharides resulting of death or decomposition of cells, generally reecting an environmental stress (light,
temperature, pH, salinity or the presence of other microorganisms)
(Richmond, 2004).

Supernatant was thus characterized by HPSEC with DOC and

DON detectors in order to get information on the size and the
chemical nature of the OM they contained. Fig. 3 represents the
time course of polysaccharides [PS]TOM and of dissolved organic
carbon [DOC]comb concentrations in the supernatant during the
recycling experiment. The evolutions with the time of [PS]TOM
and [DOC]comb were found similar: the concentrations increased
initially and reached a maximum after 45 days before decreasing.
Maximal values were, respectively, equal to 230 mg L1 for [PS]TOM
as glucose and 151.5 mg C L1 for [DOC]comb. The initial increase in
[PS]TOM and [DOC]comb in supernatant was consistent with excretion by microalgae and progressive accumulation of metabolites
during recycling and could explain the presence of foam. On the
other hand, the decrease in DOC and in PS may be attributed to
the presence of bacteria in the supernatant, cells density increased
from 4.4  105 to 11  106 CFU mL1, using the DOC as substrate
for their growth. It was already reported that molecules excreted
by microalgae were metabolized by bacteria, in particular in the
case of cyanobacteria/bacteria consortium (Abed, 2010). Although
in the present work bacteria and microalgae did not seem to compete for mineral nutrients (no mineral deciency was observed),
there is a potential risk that bacteria decimate microalgae culture.

3.2.2. DOM fractions in supernatant

Dissolved organic matter (DOM) excreted and accumulated in
supernatant, i.e., biopolymer or SOM, were quantied by HPSEC
with UV254, DOC and DON detectors (see Table 1). Bypass DOC
chrom
and chromatogram DOC, i.e., DOCchrom
, respecbypass and [DOC]
tively, are shown in Table 1-a. The two values correlated quite well
1
and [DOC]chrom was 11% lower than DOCchrom
bypass . This fact was explained by Huber et al. (2011) as a result of the retention of organic
carbon in the column due to hydrophobic interactions with organic
compounds.
In addition, DOC values determined by chromatography correlated well with global parameters. This was the case for
2
DOCchrom
bypass and the DOC determined by combustion, and for the
DOC concentration in the BP fraction with the total sugars content
in total OM [PS]TOM, despite the fact that TOM include not only
DOM but also particulate organic matter3 (see Fig. 3).
1
The correlation was DOCchrom 0:89  DOCchrom
bypass with a correlation coefcient
R2 of 0.97.
2
[DOC]chrom = 0.81  [DOC]comb, R2 of 0.99.
3
([DOC]BP = 0.96 [BP]TOM, R2 = 0.98.

290

F. Hadj-Romdhane et al. / Bioresource Technology 132 (2013) 285292

Table 1
Characterization of the organic matter content by HPSEC: DOC in supernatants (a) (see Section 2.4 for parameters denition); dissolved organic carbon and nitrogen contents, N/C
ratio and abundance in the SOM fraction (b) and the BP fraction (c). Concentrations are in mg L1.
Recycling time (days)

11

35

42

49

56

63

59.4
69.2

114.2
124.9

127.1
134.8

120
138.5

101.1
115.5

93.8
108.6

[DOC]comb***

76

140

151.5

150

127.5

115

(b) SOM fraction

[DOC]SOM
[DON]SOM
N/C ratio
Abundance in OM (%)

10
3
0.30
16.8

24.8
6.9
0.28
21.7

34.4
9.2
0.27
27.1

40.4
9.5
0.24
33.7

42.2
10.7
0.25
41.7

50.6
13.3
0.26
53.9

(c) BP fraction

[DOC]BP
[DON]BP
N/C ratio
Abundance in OM (%)

48
1.65
0.03
80.8

83.2
1.88
0.02
72.9

87.4
1.84
0.02
68.8

79.9
2
0.02
66.6

52.4
1.2
0.02
51.8

34.2
1.58
0.05
36.5

(a) Total OM

[DOC]chrom*
**
DOCchrom
bypass

Computed by integrating the signal given by the organic carbon detector (OCD) on the ow out of the column.
Signal given by the OCD for the ow bypassing the column.
***
Given here for comparison with [DOC]chrom.
**

Fig. 4. LC-OCD chromatograms of dissolved organic matter in the supernatant. The rst peak (on the left) represents the high molecular weight carbon molecules, and the
second one (on the right) represents the small organic molecules.

DOC chromatograms reported in Fig. 4 indicate that DOM was

fractionated by the column in two well-separated peaks for all
supernatants. The rst peak, detected at a retention time around
32 min and eluted with the dead volume of the column, corresponded to high molecular size molecules (BP) with molecular
weight larger than 40 kDa (in terms of PEG). The second peak with
a retention time around 50 min corresponded to small organic
molecules (SOM) with a 1 to 3 kDa MW range (in terms of PEG).
The relative abundance of BP and SOM (expressed as % DOC) varied
during the recycling experiment and ranged from 83% to 40% (BP)
and from 17% to 60% (SOM).
The response to the UVA254 was very low for the two fractions
meaning that both contained small amount of aromatic molecules
(not shown). On the other hand, the nitrogen content in the SOM
fraction was higher than in the BP fraction. N/C Mass ratio ranged
from (20 to 50 for C/N ratio) 0.02 to 0.05 for BP and from 0.24 to
0.30 (or 3.3 to 4.2 for C/N) ratio for SOM. Assuming that nitrogen
mainly refers to proteinacous structures (Huber et al., 2011), low
nitrogen content BP contained few protein-like substances and
can be hypothesized to be mainly polysaccharides. In contrast ,
much higher N/C values for SOM mean that SOM contained excreted nitrogenous molecules, such as small proteins and protein
fragments or amino and nucleic acids. Granum et al. (2002) also

reported that amino acid were the next largest fraction excreted
(5% of excreted DOC) after the mono- and polysaccharides fractions
(respectively, 15% and 33% of excreted DOC) by a diatom (here
Skeletonema costatum).
3.2.3. Evolution of populations with recycling
Fig. 5 gives the evolution of the BP and SOM contents during the
63-days recycling experiment. The N/C ratio gives information on
the nature of organic molecules present in the fractions, while
DOC is a measure of the concentration of OM. The concentration
of BP, [DOC]BP, increased during the rst 42 days from an initial value of 48 mg C L1 to a maximum value of 87.4 mg C L1, before
decreasing rapidly with a similar trend than evolution of global
parameters [DOC]comb and [PS]TOM. The accumulation of microalgae metabolites excreted in the medium, of BP in particular, supports the hypothesis of cells aggregation by exometabolites
observed in the recycled medium (see Section 3.1) (Morineau-Thomas et al., 2002). Moreover, presence of bacteria and polysaccharides will probably lead to an increase in culture viscosity
leading to a more difcult harvesting process operation, as already
observed by Rossi et al. (2008) and Rossignol et al. (1999). On the
other side, the concentration in SOM regularly increased with values of [DOC]SOM rising from to 10 to 51 mg C L1.

F. Hadj-Romdhane et al. / Bioresource Technology 132 (2013) 285292

291

Fig. 5. Time course of the dissolved organic carbon (DOC) in the biopolymers and in small organic matter fractions.

In the rst sample (after 11 days of recycling), 83% of the soluble carbon was bound to BP and only 17% to SOM, which is coherent with the fact that phytoplankton are known to excrete more
exopolysaccharides than small molecules (Myklestad, 1995;
Biddanda and Benner, 1997). The abundance in BP decreased
with a rate initially slow but that increased regularly. A possible
cause of the decrease in the abundance of BP in OM, is bacterial
development. Indeed, recycling experiment was not conducted
in axenic condition but to be close to large-scale cultures condition,
and the presence of bacteria in supernatant was conrmed
(see Section 3.2.1). The biopolymers, abundant in organic carbon,
could be consumed by bacteria as substrate preferentially
to SOM rich in nitrogen (Fogg, 1983). This could also contribute to the regular increase in [DOC]SOM with time (see Fig. 5).
However, no interrelationship was clearly established between
the evolution of the bacterial growth and the decrease in
biopolymers.
Nitrogen content in the BP fraction was very low. It remained
between 1.2 and 2.0 mg L1 and the nitrogen/carbon ratio
(N/C)BP decreased from 0.03 (11th day) to 0.02 (42th day)
before increase up to 0.05 same comment as above (63 days) in
the end.
Regarding SOM, nitrogen-rich organics were abundant in this
fraction during the whole recycling period. N/C was almost constant at a level of 2530%. This phenomenon implies that during
recycling, SOM was accumulated without signicant variation of
its character, which would be amino acid, identied as the largest
fraction of metabolites excreted by microalgae during growth (Granum et al., 2002).
In conclusion, the characterisation by HPSEC of supernatant
conrmed that two fractions can be distinguished in OM released
by microalgae, namely biopolymers (BP) and smaller organic compounds (SOM). Culture of C. vulgaris with medium recycling during
63 days highlighted that organic matter excreted by microalgae
progressively accumulated in the medium. As SOM concentration
increased regularly over the whole experiment time, BP one increased until a maximum when the culture was about 40-daysold, and then decreased. It can be concluded from OM accumulation that recycling culture supernatant introduced a risk of deterioration of medium quality and of culture in the medium-to-long
term. Moreover, presence of bacteria and polysaccharides will
probably lead to an increase in culture viscosity leading to a more
difcult harvesting process operation, as already observed in
Arthrospira platensis by Rossi et al. (2008) and Rossignol et al.
(1999).

4. Conclusion
A regular accumulation of organic matter excreted by microalgae was observed as well as some signs of a drift in biomass quality
that could reect a possible beginning of culture deterioration. Organic matter in supernatant was mainly polysaccharides, with lower amounts of small organic molecules with a nitrogen to carbon
mass ratio very close to that observed in proteineous substances.
The relative proportion of polymers in organic matter decreased
markedly from 80% to 36.5% and the biopolymers population
seemed to be regulated by microalgae excretion and an opposite
phenomenon that can be hypothesized to be consumption by
bacteria.
Acknowledgements
The authors thank Hlne Marec (GEPEA UMR-CNRS 6144) and
Jean-Luc Hauser for photobioreactor design and their technical
support.

References
Abed, R.M.M., 2010. Interaction between cyanobacteria and aerobic heterotrophic
bacteria in the degradation of hydrocarbons. International Biodeterioration &
Biodegradation 64, 5864.
Alyabyev, A.J., Loseva, N.L., Gordon, L.K., Andreyeva, I.N., Rachimova, G.G.,
Tribunskih, V.I., Ponomareva, A.A., Kemp, R.B., 2007. The effect of changes in
salinity on the energy yielding processes of Chlorella vulgaris and Dunaliella
maritima cells. Thermochimica Acta 458, 6570.
Biddanda, B., Benner, R., 1997. Carbon, nitrogen, and carbohydrate uxes during the
production of particulate and dissolved organic matter by marine
phytoplankton. Limnology and Oceanography 42, 506518.
Blanchemain, A., Grizeau, D., Guary, J.C., 1994. Effect of different organic buffers on
the growth of Skeletonema costatum cultures: further evidence for an
autoinhibitory effect. Journal of Plankton Research 16, 14331440.
Bosma, R., Miazek, K., Willemsen, S.M., Vermu, M.H., Wijffels, R.H., 2008. Growth
inhibition of Monodus subterraneus by free fatty acids. Biotechnology and
Bioengineering 101, 11081114.
Brennan, L., Owende, P., 2010. Biofuels from microalgae A review of technologies
for production, processing, and extractions of biofuels and co-products.
Renewable and Sustainable Energy Reviews 14, 557577.
Dubois, M., 1956. Colorimetric method for determination of sugars and related
substances. Analytical Chemistry 28, 350356.
Fogg, G.E., 1971. Extracellular products of algae in freshwater. Archiv fr
hydrobiologie 5, 125.
Fogg, G.E., 1983. The ecological signicance of extracellular products of
phytoplankton photosynthesis. Botanica Marina 25, 314.
Granum, E., Kirkvold, S., Myklestad, S.M., 2002. Cellular and extracellular production
of carbohydrates and amino acids by the marine diatom Skeletonema costatum:
diel variations and effects of N depletion. Marine Ecology Progress Series 242,
8394.

292

F. Hadj-Romdhane et al. / Bioresource Technology 132 (2013) 285292

Hadj-Romdhane, F., Jaouen, P., Pruvost, J., Grizeau, D., Van Vooren, G., Bourseau, P.,
2012. Development and validation of a minimal growth medium for recycling
Chlorella vulgaris culture. Bioresource Technology 123, 366374.
Harun, R., Singh, M., Forde, G.M., Danquah, M.K., 2010. Bioprocess engineering of
microalgae to produce a variety of consumer products. Renewable and
Sustainable Energy Reviews 14, 10371047.
Huber, S.A., Balz, A., Abert, M., Pronk, W., 2011. Characterisation of aquatic humic
and non-humic matter with size-exclusion chromatography organic carbon
detection organic nitrogen detection (LC-OCD-OND). Water Research 45, 879
885.
Hutner, S.H., Provasoli, L., Schatz, A., Haskins, C.P., 1950. Some approaches to the
study of the role of metals in the metabolism of microorganisms. Proceedings of
the American Philosophical Society 94, 152170.
Javanmardian, M., Palsson, B.O., 1991. High-density photoautotrophic algal
cultures: design, construction, and open operation of a novel photobioreactor
system. Biotechnology and Bioengineering 38, 11821189.
Jenck, J., Lpine, O., Legrand, J., Dreno, P., Grizeau, D., Dupr, C., 2011. Valorisation
industrielle des microalgues photosynthtiques. Techniques de lingnieur IN
201, 110.
Leone, D.E., 1963. Growth of Chlorella pyrenoidosa in recycled medium. Applied
Microbiology and Biotechnology 11, 427429.
Lvansky, K., Dedic, K., Bnov, J., Tichy, V., Novotny, P., Doucha, J., 1996. Inuence of
the nutrient solution recycling on the productivity of Scenedesmus obliquus,
utilization of nutrients and water in outdoor cultures. Algological Studies/
Archiv fr Hydrobiologie, Supplement Volumes 81, 105113.
Morineau-Thomas, O.M.-T., Jaouen, P.J., Legentilhomme, P.L., 2002. The role of
exopolysaccharides in fouling phenomenon during ultraltration of microalgae
(Chlorella sp. and Porphyridium purpureum): advantage of a swirling decaying
ow. Bioprocess and Biosystems Engineering 25, 3542.
Myklestad, S.M., 1995. Release of extracellular products by phytoplankton with
special emphasis on polysaccharides. Science of the Total Environment 165,
155164.
Pratt, R., 1942. Studies on Chlorella vulgaris VI. Retardation of photosynthesis by a
growth-inhibiting substance from Chlorella vulgaris. American Journal of Botany
30, 3233.
Pratt, R., Fong, J., 1940. Studies on Chlorella vulgaris II. Further evidence that Chlorella
cells form a growth-inhibiting substance. American Journal of Botany 27, 431
436.
Pruvost, J., Van Vooren, G., Le Gouic, B., Couzinet-Mossion, A., Legrand, J., 2011.
Systematic investigation of biomass and lipid productivity by microalgae in

photobioreactors for biodiesel application. Bioresource Technology 102, 150

158.
Richmond, A., 2004. Handbook of Microalgal Culture: Biotechnology and Applied
Phycology. Blackwell Sciences Ltd., Oxford.
Richmond, A., Zou, N., 1999. Efcient utilisation of high photon irradiance for mass
production of photoautotrophic micro-organisms. Journal of Applied Phycology
11, 123127.
Richmond, A., Cheng-Wu, Z., Zarmi, Y., 2003. Efcient use of strong light for high
photosynthetic productivity: interrelationships between the optical path, the
optimal population density and cell-growth inhibition. Biomolecular
Engineering 20, 229236.
Rodol, L., Zittelli, G.C., Barsanti, L., Rosati, G., Tredici, M.R., 2003. Growth medium
recycling in Nannochloropsis sp. mass cultivation. Biomolecular Engineering 20,
243248.
Rossi, N., Derouiniot-Chaplain, M., Jaouen, P., Legentilhomme, P., Petit, I., 2008.
Arthrospira platensis harvesting with membranes: fouling phenomenon with
limiting and critical ux. Bioresource Technology 99, 61626167.
Rossignol, N., Vandanjon, L., Jaouen, P., Qumneur, F., 1999. Membrane technology
for the continuous separation microalgae/culture medium: compared
performances of cross-ow microltration and ultraltration. Aquacultural
Engineering 20, 191208.
Spolaore, P., Joannis-Cassan, C., Duran, E., Isambert, A., 2006. Commercial
applications of microalgae. Journal of Bioscience and Bioengineering 101, 87
96.
Tredici, M.R., 2010. Photobiology of microalgae mass cultures: understanding the
tools for the next green revolution. Biofuels 1, 143162.
Wu, J.T., Chiang, Y.R., Huang, W.Y., Jane, W.N., 2006. Cytotoxic effects of free fatty
acids on phytoplankton algae and cyanobacteria. Aquatic Toxicology 80, 338
345.
Yang, J., Xu, M., Zhang, X., Hu, Q., Sommerfeld, M., Chen, Y., 2011. Life-cycle analysis
on biodiesel production from microalgae: water footprint and nutrients
balance. Bioresource Technology 102, 159165.
Yu, Y., Hu, H.-Y., Li, X., Wu, Y.-H., Zhang, X., Jia, S.-L., 2012. Accumulation
characteristics of soluble algal products (SAP) by a freshwater microalga
Scenedesmus sp. LX1 during batch cultivation for biofuel production.
Bioresource Technology 110, 184189.
Zheng, X., Ernst, M., Jekel, M., 2010. Pilot-scale investigation on the removal of
organic foulants in secondary efuent by slow sand ltration prior to
ultraltration. Water Research 44, 32033213.