J Appl Phycol (2009) 21:559567 DOI 10.

1007/s10811-008-9391-8

Microbial flocculation, a potentially low-cost harvesting technique for marine microalgae for the production of biodiesel
Andrew K. Lee & David M. Lewis & Peter J. Ashman

Received: 24 July 2008 / Revised and accepted: 7 November 2008 / Published online: 13 December 2008 # Springer Science + Business Media B.V. 2008

Abstract Microbial flocculation is investigated as a separation technique for harvesting marine microalgae for the production of biodiesel. Organic carbon (acetate, glucose or glycerine) was used as substrate for the growth of flocculating microbes in situ. Under stress, due to nutrient depletion, these microbes produced extracellular polymeric substances that promote flocculation of the coccolithophorid alga, Pleurochrysis carterae. Maximum recovery efficiency was achieved at low concentration of organic substrate (0.1 g L1) and with a long mixing time (24 h); an average recovery efficiency of over 90% and a concentration factor of 226 were achieved. The recovery efficiency is positively correlated with mixing time (R2 =0.90). The concentration factor is negatively correlated to the product of substrate concentration and mixing time (R2 =0.73). The microalgae cells were not under stress and remained viable, thus potentially allowing media to be reused in large-scale processes without further treatment. Other advantages of the process are that no metallic flocculants were required and the organic substrates are readily available, e.g. glycerine is a byproduct of biodiesel production and acetic acid may be produced by anaerobic digestion of the biomass residue after lipid extraction. Further research is required to optimise the process.
Paper presented at the 3rd Congress of the International Society for Applied Phycology, Galway, Ireland. A. K. Lee : D. M. Lewis : P. J. Ashman Microalgal Engineering Research Group, School of Chemical Engineering, University of Adelaide, Adelaide, SA, Australia 5000 A. K. Lee (*) School of Chemical Engineering, University of Adelaide, Adelaide, SA, Australia 5000 e-mail: alee@chemeng.adelaide.edu.au

Keywords Concentration factor . Extracellular polymeric substances (EPS) . Pleurochrysis carterae . Recovery efficiency . Separation

Nomenclature
C0 Ct c co ci cij CF RE t V Xi Y

initial concentration of cells (cells mL1) concentration of cells in the media at time, t (cells mL1) substrate concentration (g L1) intercept regression coefficient for the ith factor regression coefficient for the interaction between the ith and jth factors concentration factor recovery efficiency mixing time (h) volume level of the ith factor predicted response (RE or CF)

Introduction Biodiesel production from vegetable oils is a proven technology and is widely available in Europe and the United States of America (Ahmann and Dorgan 2007); however, plantation oil crops, waste vegetable oil and animal fat are only available in limited amounts, and the diversion of food crops for the production of biofuels is a major concern (Kleiner 2007). Some species of microalgae, with high growth rate and high lipid content, appear to be attractive alternatives as a feedstock for biodiesel production (Chisti 2007). The marine coccolithophorid microalga Pleurochrysis carterae offers advantages such as a high

560

J Appl Phycol (2009) 21:559567

lipid production rate (21.9 t ha1 year1), the ability to grow in arid areas that are not suitable for farming, and a high range of pH tolerance (pH 7 to 11) which reduces the likelihood of biological contaminants (Moheimani and Borowitzka 2006). However, low-cost harvesting, which would be necessary for the economic production of microalgal lipids from P. carterae, is technically challenging. Major obstacles include: the small size of P. carterae (8~19 m; Winter and Siesser 1994), small density differences between the microalgae and the growth media, the ability of microalgae to remain in suspension, low cell concentrations (0.05% dry mass) in the media and the high ionic strength of sea water. All of these factors make harvesting of P. carterae one of the most challenging aspects of successful commercial production (Sukenik et al. 1988; Benemann and Oswald 1996). In addition to the low cost, the harvesting of microalgal lipids for biodiesel production must also be reliable and yield a product that is free from metallic contaminants. Microalgae may be harvested commercially using centrifugation, filtration, and flocculation, either used individually or in combination. However, these methods are not suitable for biodiesel production with P. carterae as a feedstock. Centrifugation requires high capital, energy and running costs and is therefore only suitable for highvalue products (Becker 1995). Filtration is suitable for filamentous or colony forming microalgae such as Spirulina sp. or Micractinium sp. Flocculation using multivalent metal salts will contaminate the algal biomass (Mohn 1988) while flocculation using cationic polymers is inhibited by the high ionic strength of sea water (Bilanovic and Shelef 1988). Alternatively, bioflocculation, induced by environmental stresses such as extreme pH, temperature or nutrient depletion, may cause cell composition changes and is generally considered as too unreliable to be economical on a commercial scale (Benemann and Oswald 1996). Furthermore, the presence of sodium in a marine environment would reduce the exocellular protein responsible for bioflocculation (Higgins and Novak 1997). Microbial flocculation had been practised in processes such as wastewater treatment (Al-Shahwani et al. 1986; Noe et al. 1992) and fermentation (Esser and Kues 1983; Hantula and Bamford 1991). The process was also suggested as a harvesting technique for microalgae by Benemann and Oswald (1996). Further evidence supporting its use in the present application may be found in the literature: for example, it has been reported that all known microalgae produce exopolysaccharides such as uronic or pyruvic acid, which are indistinguishable from those produced by bacteria (Shipin et al. 1999), and that these polymers are responsible for the adhesion of cells (Frlund et al. 1996). Furthermore, microalgae embedded in the polymer matrix did not show signs of stress or lysis over an

extended period (Shipin et al. 1999). Although the exact composition of the polymers may vary according to the species (Choi et al. 1998) or extraction methods (Comte et al. 2006), the fact that all bacterial and microalgal surfaces are negatively charged with similar polysaccharides (Becker 1995; Liao et al. 2002) suggests that the same flocculants may be used to aggregate microalgae. Soil and activated sludge samples are the best sources of organisms producing extracellular polymeric substances (EPS; Toeda and Kurane 1991). The method of Toeda and Kurane (1991) involved isolating and culturing the bacteria, Alcaligenes cupidus KT201, then extracting the bioflocculant and applying it in a purified form. Similar techniques were used by others with the flocculating bacteria Enterobacter sp. and Bacillus coagulans strain (Yokoi et al. 1997; Salehizadeh et al. 2000). Bioflocculant has also been produced from glucose using the bacteria Pullularia pullulans (Zajic and Leduy 1973) and Rhodococcus erythropolis (Kurane et al. 1986). All of these procedures were complex and expensive (Liu and Fang 2002; Lian et al. 2007) and there was no direct correlation between the amount of polymer and the floc settleability (Li and Yang 2007). Therefore, the methods described by these reports are not suitable for harvesting microalgae for low-cost commodities such as biodiesel. Furthermore, many of the results described are also not applicable in a marine environment (Zhang et al. 2002). In an aquatic environment, non-living surfaces are colonised by bacteria which eventually form biofilms with other microbes. However, living surfaces are usually relatively free from bacteria due to surface chemistry incompatibility (Rosowski 1992), the production of repellents (Gerhart et al. 1988), or sloughing of surface layers (Mittelman 1996). Therefore, the attachment mechanism and the microbes involved will likely be species specific. However, when bacterial concentrations are high enough, some of the functional groups from the EPS formed will be suitable for cell adhesion (Liao et al. 2002). An organic carbon substrate is required for bacterial growth to occur. Potentially cheap and readily available sources of organic carbon include glucose, acetate or glycerine. Acetates and propionates are major products of the anaerobic digestion of organic biomass (Fujita et al. 2000) and could potentially be obtained by the digestion of the microalgal biomass remaining after lipid extraction. Glycerol is a by-product of the transesterification of lipids during the production of biodiesel (Ma and Hanna 1999). The aim of this paper is to investigate the feasibility of flocculation, initiated by natural colonies of microbial species grown with glucose, acetate or glycerine, as a low-cost harvesting technique for the marine microalgae, P. carterae.

J Appl Phycol (2009) 21:559567

561

Materials and methods P. carterae (CCMP647) was obtained from the Murdoch University microalgae culture collection. P. carterae was grown at 22C on a shake table in a culture tube under a light intensity of 300 mol photons m2 s1. The growth medium was BG-11 medium (Rippka et al. 1979) modified by the addition of (g L1) NaCl 18, NaNO 3 1.50, K2HPO43H2O 0.040, MgSO47H2O 0.075, CaCl22H2O 0.036, citric acid combined with ferric ammonium citrate 0.006, EDTA 0.001, (g L1) H3BO3 2.86, MnCl24H2O 1.81, ZnSO 4 7H 2 O 0.222, NaMoO 4 5H 2 O 0.390, CuSO45H2O 0.079, Co(NO3)26H2O 0.0494, cyanocobalamin and biotin 1 ng L1 and thiamine 2 ng L1. The medium was first autoclaved and vitamins were added after cooling. The pH was adjusted to pH 8.2 by the addition of sodium bicarbonate prior to the inoculation of the culture. The contents were later introduced to a 250-mL Erlenmeyer flask on a shake table and finally to a 20-L carboy photobioreactor. Mixing was provided by sparging air from the bottom of the carboy; lighting was supplied by four cool-white fluorescent tubes with an intensity 400 mol photons m2 s1 with a 12/12 h light/dark cycle. Cultures near the end of log phase growth were used for the flocculating experiments. Preparation of microbe culture Modified BG11 was used for culturing the microbes, with non-sterile water used to introduce bacterial growth. The culture was then exposed to atmosphere for a few hours and cultivated in 1-L flasks under the same conditions as described previously. Initially, P. carterae was the only form of organic carbon added, this ensured only those microbes that were able to coexist with P. carterae would be selected. Microalgal samples that were observed to flocculate were then isolated and subcultured. A small aliquot (0.5 mL) of stock solution (100 g L1 of glucose, glycerine or acetate) was added about once a week during the stationary phase to sustain the growth of various microbes. This mixture of microalgae and microbes was used as the flocculating culture for the experiments. Selection and preparation of the non-acetate-utilising microbe culture Cultures of microalgae that were observed to flocculate were selected and diluted by factors of 106, 107 and 108. Small samples (0.2 mL) of the diluted microalgae culture were then spread onto nutrient agar plates containing 10 g L1 Oxoid peptone, 10 g L1, Oxoid lab Lemco powder and agar 15 g L1 in modified BG11 media. Observable microbial colonies on the agar were then aseptically transferred to culture tubes containing Oxoid nutrient broth of the same recipe but without agar. Cultures were grown on a shake table at 22C and 70 rpm for up to 48 h prior to use.

Determination of flocculation effectiveness The effectiveness of flocculation was determined by the recovery efficiency and the concentration factor during microalgal separation. Recovery efficiency (RE) is defined as the ratio of the mass of cells recovered to the total mass of cells. The recovery efficiency is determined as: RE 1 Ct =C0 1

Concentration factor (CF) is the ratio of the final product concentration to the initial concentration. The concentration factor is determined as:  CF  V0 RE Vf 2

Microalgal cell concentration determination Cell counts were determined by using a Sedgewick Rafter counting slide and Olympus IX 50 microscope coupled with Analysis LS Research software. As it is difficult to achieve a representative sample in counting objects of a single class when mixed with other classes (Russ and Dehoff 2000), a minimum of six readings were taken for each sample and the mean of these values used in calculations.

Experimental procedure for microbial flocculation of P. carterae A 0.1 mL aliquot of the microbe culture, or the non-acetateutilising microbe culture, was added to 100 mL of P. carterae culture. The mixture was shaken vigorously and the initial microalgae concentration (C0) determined by cell counting. Samples (100 mL) were then introduced into 250-mL flasks on a shake table at 70 rpm. An organic substrate of acetate, glucose or glycerine was added to the shake flasks with a final substrate concentration of either 0.1 or 0.5 g L1, as shown in Table 1. For the experiments with the non-acetate-utilising microbe culture, only acetate (at either 0.1 or 0.5 g L1) is added to the shake flask. The flasks were left on the shake table under normal growth conditions, but without illumination, for either a 6 or 24-h duration, as shown in Table 1. At the end of this period, 90 mL of the supernatant was collected as soon as possible to prevent settling and the concentration (Ct) was determined by cell counting. Any flocs, together with a small amount of media, were gently transferred to a measuring cylinder with 0.1 mL graduations and allowed to settle under gravity for 30 min. Floc volumes were thus recorded. The

562 Table 1 Levels of each factor in the factorial design Factor

J Appl Phycol (2009) 21:559567

Levels Low level () High level (+)

(A) Substrate (B) Concentration of substrate (g L1) (C) Mixing time (h)

Acetate (1)

Glucose (2)

Glycerine (3) 0.1 6 0.5 24

recovery efficiency and concentration factor for each treatment were calculated using equations (1) and (2). Experimental design The effectiveness of the microbe culture in promoting microbial flocculation was investigated using a factorial design with three factors: organic substrate, concentration of substrate and mixing time. The levels for each factor are shown in Table 1. The experimental treatments were randomised and repeated in triplicate. The effectiveness of the non-acetate-utilising microbe culture was investigated similarly except that only acetate was used as the organic substrate.

The recovery efficiency and concentration factor for the flocculation of P. carterae using the non-acetate-utilising microbe culture are presented in Table 3. In this case, only acetate was used as the substrate, while the factors of acetate concentration and mixing time were investigated using a factorial design with two factors and three replicates. When microbes were cultured that were unable

Results Quality of polymers on flocculation During the experiments, two main types of flocs were observed, with one type appearing to be darker and denser than the other. Figure 1(a) shows EPS forming a dense floc with a large number of microalgal cells embedded and with relatively few microalgae cells left in the surrounding media. Figure 1(b) shows a much looser floc with a large amount of EPS but with very few microalgal cells attached (some microalgal cells appeared to be attached but were in fact positioned under the floc and then compressed by the glass slide). This second type of floc was usually observed with higher substrate concentration (>0.5 g L1). Effectiveness of flocculation The recovery efficiency and concentration factor for each treatment and replicate of the factorial experiment are presented in Table 2 for flocculation using the standard microbe culture. It is clear that microbial flocculation of P. carterae is substantially enhanced at long mixing times. There was no apparent difference between the flocculation recovery efficiency with varying substrate type. Larger volumes of flocs were observed to form at higher doses of organic carbon, but the recovery efficiency remained relatively constant. Thus, the concentration factor decreases with increasing substrate concentration at long mixing times.

Fig. 1 Photomicrograph showing extracellular polymeric substance (EPS) with a good attachment of microalgal cells, and b poor attachment of microalgal cells

J Appl Phycol (2009) 21:559567

563

Table 2 Average recovery efficiency (RE) and concentration factor (CF) for the microbial flocculation of P. carterae as a function of substrate type, substrate concentration and mixing time. (Values in brackets indicate the range of results) Substrate Substrate concentration (c) Mixing time (t) Acetate Glucose Glycerine Recovery efficiency (%) Concentration factor Recovery efficiency (%) Concentration factor Recovery efficiency (%) Concentration factor 0.1 g L1 6h 52 149 53 167 45 179 (5153) (146151) (4470) (146182) (3654) (143216) 24 h 88 240 90 236 94 204 (7596) (215264) (8396) (232239) (9196) (192228) 0.5 g L1 6h 52 133 48 181 50 171 (5053) (106150) (3560) (141242) (3861) (152210) 24 h 90 100 93 109 93 131 (8195) (89106) (8997) (89121) (9397) (100185)

to utilise acetate, and when acetate is the only substrate provided to this culture, the flocs were formed loosely and appeared paler in colour, indicating a lack of flocculating activity. Furthermore, a significantly lower average recovery efficiency is obtained than for the standard microbe culture A summary of analysis of variance (ANOVA) for recovery efficiency and concentration factor are presented in Tables 4 and 5. Only variables (indicated by # in the tables) with p values under 0.05 are deemed to be significant. They are the mixing time from Table 4 and both the mixing time and the interaction of mixing time and concentration from Table 5. The ANOVAs showed that the nature of the organic substrate did not have a significant effect on either the recovery efficiency or the concentration factor. Furthermore, the ANOVAs demonstrated that the interactions between the substrate and both substrate concentration and mixing time were also insignificant. Thus, the recovery efficiency and concentration factor data (Table 1) were fitted to the following linear regression model for a 22 data set in which the data for all of the different carbon substrates were pooled: Y co cB XB cC XC cBC XB XC The coefficients, co, ci, and cij, for each linear regression model were calculated using Microsoft Excel 2003. The

regression analysis for recovery efficiency gave the following relationship with an R2 value of 0.89:

RE 0:36 0:023 t The regression analysis for the concentration factor gave the following relationship with an R2 value of 0.73: CF 136 4:95 t 15:23 tc

Effect of microbes on microalgal flocculation It was clear from these experiments that the addition of an organic carbon substrate to a culture of P. carterae that was previously dosed with the microbe culture promoted the flocculation of the microalgae. However, it is not clear whether the flocculation of the microalgae is promoted by the presence of microbes or by an effect of carbon substrate addition on the microalgae cells themselves. Further experiments using the non-acetate-utilising microbe culture, with the addition of acetate, demonstrated that the microalgae cultures flocculated with lower recovery efficiency than cultures containing the standard microbe culture. Thus, it was concluded that the action of the microbes in the culture enhanced flocculation and was not an effect of the carbon substrate on the microalgae cells.

Table 3 Average RE and CF for the microbial flocculation of P. carterae using non-acetate-utilising microbes as a function of substrate type, substrate concentration and mixing time (values in brackets indicate the range of results) Substrate Substrate concentration(c) Mixing time (t) Acetate Recovery efficiency (%) Concentration factor 0.1 g L1 6h 45 (3849) 130 (110141) 24 h 65 (4394) 159 (144234) 0.5 g L1 6h 49 (4753) 138 (131157) 24 h 73 (4494) 156 (125205)

564 Table 4 ANOVA for recovery efficiency due to microbial flocculation of P. carterae with factors (A) substrate type, (B) substrate concentration and (C) mixing time

J Appl Phycol (2009) 21:559567

SUMMARY OUTPUT Regression Statistics Multiple R R Square Adjusted R Square Standard Error Observations ANOVA df Regression Residual Total 6.00 29.00 35.00 0.95 0.90 0.87 0.08 36.00

RE vs A B C AB AC BC

SS 1.57 0.18 1.76 Standard Error 0.09 0.04 0.21 0.00 0.08 0.00 0.01

MS 0.26 0.01

F 41.54

Coefficients Intercept # X Variable A X Variable B X Variable C # X Variable AB X Variable AC X Variable BC 0.45 -0.04 -0.05 0.02 0.02 0.00 0.00

t Stat 4.87 -1.08 -0.22 3.91 0.23 1.43 0.23

P-value 0.00 0.29 0.83 0.00 0.82 0.16 0.82

Effect of the nature of substrates The three organic substrates had similar flocculating activities when the appropriate bacterial cultures were used. The apparent differences in recovery efficiency and concentration factor were not of statistical significance as their p values of 0.29 and 0.53 (X Variable A, Tables 4 and 5) were much greater than 0.05 Characterisation of flocculating bacteria The bacteria isolated were identified by their morphology, biochemistry and 16S rRNA sequences (Murray et al. 2003). They are found to be Pseudomonas stutzeri and Bacillus cereus. However, readers should bear in mind that a mixed culture has been used throughout the experiments and the bacteria isolated from agar plates may not be a true representation of those present in the culture.

was dependent upon the amount of flocculating bacteria and their stress level, optimum dosage occurred when the growth of bacteria was stimulated but depleted soon afterward (Bossier and Verstraete 1996; Li et al. 2006). This observation agrees well with previous work on production of yeast, Saccharomyces cerevisiae, where an increase in glucose concentration did not increase flocculating activities (Chang et al. 2005). While a supply of organic carbon is essential for the growth of microbes, it is the depletion of nutrients that is essential for the aggregation of cells. The lack of nutrients is the major stress factor inducing the microbes to form extracellular polymeric substances (EPS). Effect of mixing time The recovery efficiency increased with mixing time, from an average of 0.5 for 6 h to 0.9 for 24 h. Increasing the mixing time increased the number of bacteria up to a level which was limited by the available substrate, and also increased the probability of binding contact between microalgal cells and the polymers. However, the rate of recovery was expected to plateau due to the decreasing concentration of cells in the media, and hence the recovery efficiency was expected to have a non-linear response to mixing time. In practice, shorter mixing times may be necessary to optimise energy consumption during separation; further

Discussion Effect of substrate concentration There was no appreciable increase in the recovery efficiency when excess carbon substrates were added. Similarly, the concentration factor was reduced as the excess substrate produced loose polymers, adding mass to the flocs. As the aggregation

J Appl Phycol (2009) 21:559567 Table 5 ANOVA for concentration factor due to microbial flocculation of P. carterae with factors (A) substrate type, (B) substrate concentration and (C) mixing time

565

SUMMARY OUTPUT Regression Statistics Multiple R R Square Adjusted R Square Standard Error Observations ANOVA df Regression Residual Total 6.00 29.00 35.00 0.86 0.73 0.68 28.26 36.00

CF vs A B C AB AC BC

SS 63735.66 23154.55 86890.21 Standard Error 32.78 14.16 73.63 1.59 28.84 0.64 2.62

MS 10622.61 798.43

F 13.30

Coefficients Intercept # X Variable A X Variable B X Variable C # X Variable AB X Variable AC X Variable BC # 118.22 8.94 -12.03 6.98 47.39 -1.01 -15.23

t Stat 3.61 0.63 -0.16 4.38 1.64 -1.58 -5.82

P-value 0.00 0.53 0.87 0.00 0.11 0.12 0.00

laboratory trials at larger scale would be necessary to allow the balance between recovery and cost to be determined. Effect of hydrodynamics Proper mixing was expected to increase the frequency of contact between microalgae and the EPS, and hence to increase recovery efficiency and concentration factor. Thus, too little mixing was likely to result in poor flocculation, while over-mixing could cause the flocs to break up into smaller fragments. In a controlled experiment, where P. carterae was allowed to settle under gravity without mixing, the flocs which formed were extremely loose and difficult to separate properly from the culture media. The concentration factor was estimated to only be in the order of 10. In the laboratory, mixing was achieved by shake tables. This method of mixing was clearly inefficient in terms of energy usage. A more energy-efficient method to promote mixing is to use baffled channels, which is widely practiced in flocculation and mixing during wastewater management (McConnachie 1993). Proper mixing together with an optimised mixing time will minimise energy usage in larger scale processes. Effect of floc quality on microalgal separation The volume of EPS produced was not related to the flocculation and separation processes. For maximum cell recovery, it is

desirable if the organic carbon is converted into a tightly bounded EPS. This can be achieved by reducing the organic carbon concentration, increasing mixing time and using an appropriate mix of microbes. Effect of gravity The regression analysis calculated positive values of co for each of the predicted responses which indicated the effect that gravity has on the settling behaviour of P. carterae. However, it was observed that without proper mixing to supply the nutrients required by the bacteria, the flocs formed were loose, difficult to collect and the settling behaviour was unreliable and unpredictable. Comparison with other harvesting methods For the production of biodiesel, microbial flocculation is potentially a significant improvement over other commercial harvesting methods. In these experiments, the microalgal cells were not damaged and maintained their integrity, and the flocculation and settling behaviour was observed to be predictable and reliable. At larger scales, media could be reused to minimise the cost of nutrients and the demand for water; the organic carbon substrates used in the process are relatively low-cost and potentially readily available since, in the case of acetate and glycerol, they may be produced as by-products of biodiesel production. Additionally the

566

J Appl Phycol (2009) 21:559567 exchange resin. Water Res Oxf 30:1749. doi:10.1016/00431354(95)00323-1 Fujita M, Ike M, Tachibana S, Kitada G, Kim SM, Inoue Z (2000) Characterization of a bioflocculant produced by Citrobacter sp. TKF04 from acetic and propionic acids. J Biosci Bioeng 89:40 46. doi:10.1016/S1389-1723(00)88048-2 Gerhart D, Rittschof D, Mayo SW (1988) Chemical ecology and the search for marine antifoulants. J Chem Ecol 14:1905. doi:10.1007/ BF01013485 Hantula J, Bamford DH (1991) The efficiency of the proteindependent flocculation of Flavobacterium sp. is sensitive to the composition of growth medium. Appl Microbiol Biotechnol 36:100104. doi:10.1007/BF00164707 Higgins M, Novak J (1997) Characterisation of exocellular protein and its role in bioflocculation. J Environ Eng 123:479485. doi:10.1061/(ASCE)0733-9372(1997)123:5(479) Kleiner K (2007) The backlash against biofuels. Nature Reports. Clim Change 2(Jan):2008. New York.. doi:10.1038/climate.2007.71 Kurane R, Toeda K, Takeda K, Suzuki T (1986) Culture conditions for production of microbial flocculant by Rhodococcus erythropolis. Agric Biol Chem 50:23092313 Li X, Yang S (2007) Influence of loosely bound extracellular polymeric substances (EPS) on the flocculation, sedimentation and dewaterability of activated sludge. Water Res 41:10221030. doi:10.1016/j.watres.2006.06.037 Li Z, Kuba T, Kusuda T (2006) The influence of starvation phase on the properties and the development of aerobic granules. Enzyme Microb Technol 38:670674. doi:10.1016/j.enzmictec.2005.07.020 Lian B, Chen Y, Zhao J, Teng H, Zhu L, Yuan S (2007) Microbial flocculation by Bacillus mucilaginosus: applications and mechanisms. Bioresour Technol 99:48254831. doi:10.1016/j.biortech.2007. 09.045 Liao BQ, Allen DG, Leppard GG, Droppo IG, Liss SN (2002) Interparticle interactions affecting the stability of sludge flocs. J Colloid Interface Sci 249:372380. doi:10.1006/jcis.2002.8305 Liu H, Fang H (2002) Extraction of extracellular polymeric substances (EPS) of sludges. J Biotechnol 95:249256. doi:10.1016/S01681656(02)00025-1 Ma F, Hanna M (1999) Biodiesel production: a review. Bioresour Technol 70:115. doi:10.1016/S0960-8524(99)00025-5 McConnachie G (1993) Water treatment for developing countries using baffled-channel hydraulic flocculation. Proceedings of the Institution of Civil Engineers, Water Maritime Energy 101:5561 Mittelman M (1996) Bacterial adhesionmolecular and ecological diversity. Wiley-Liss, New York, pp 101108 Moheimani N, Borowitzka MA (2006) The long-term culture of the coccolithophore Pleurochrysis carterae (Haptophyta) in outdoor raceway ponds. J Appl Phycol 18:703712. doi:10.1007/s10811006-9075-1 Mohn F (1988) Harvesting of microalgal biomass. Microalgal biotechnology. ed. Borowitzka. Cambridge University Press, Cambridge. Ch. 15 Murray PB, Barron EJ, Jorgensen JH, Pfaller MA, Yolken RH (2003) Manual of clinical microbiology. ASM, Washington, D.C Noe J, Lalibert G, Proulx D (1992) Algae and waste water. J Appl Phycol 4:247254. doi:10.1007/BF02161210 Rippka R, Deruelles J, Waterbury J, Herdman M, Stanier R (1979) Generic assignments, strain histories and properties of pure cultures of cyanobacteria. J Gen Microbiol 111:161 Rosowski JR (1992) Specificity of bacterial attachment sites on the filamentous diatom Navicula confervacea. Can J Microbiol 38:676686 Russ J, Dehoff RT (2000) Practical stereology. Kluwer/Plenum, New York, p 149181 Salehizadeh H, Vossoughi M, Alemzadeh I (2000) Some investigations on bioflocculant producing bacteria. Biochem Eng J 5:3944. doi:10.1016/S1369-703X(99)00066-2

microalgal cells were not stressed so that the lipid composition is likely to remain steady and no metallic flocculants were used which might contaminate the product. Finally, the minimum concentration of organic carbon substrate used in these experiments (0.1 g L1) is still high in comparison to the dry mass concentration of the microalgal suspension, which is in the order of 0.5 g L1. Therefore, further research is required to reduce the level of substrate and to minimise the mixing energy required for this process.

Acknowledgement The authors would like to thank the following people; Prof Michael Borowitzka for supplying P. carterae, the Infectious Diseases Laboratory from the Institute of Medical and Veterinary Science, Adelaide, South Australia, for the identification of the bacteria involved in the flocculation and Mr. Steven Amos for technical assistance with microalgae culturing.

References
Ahmann D, Dorgan J (2007) Bioengineering for pollution prevention through development of biobased energy and materials. National Center for Environmental Research, U.S. Environmental Protection Agency, Office of Research and Development, Washington, Report no. EPA/600/R-07/028, pp 117129 Al-Shahwani MF, Jazrawi SF, Al-Rawi EH (1986) Effects of bacterial communities on floc sizes and numbers in industrial and domestic effluents. Agric Wastes 16:303311. doi:10.1016/0141-4607(86) 90060-0 Becker E (1995) Microalgae biotechnology and microbiology. Cambridge University Press, Cambridge, pp 158160 Benemann J, Oswald W (1996) Systems and economic analysis of microalgae ponds for conversion of carbon dioxide to biomass. 4th Quarterly Technical Progress Report, Department of Energy, Pittsburgh Energy Technology Centre, Report CONF-9409207-2. pp 105109 Bilanovic D, Shelef G (1988) Flocculation of microalgae with cationic polymerseffects of medium salinity. Biomass 17:6576. doi:10.1016/0144-4565(88)90071-6 Bossier P, Verstraete W (1996) Triggers for microbial aggregation in activated sludge. Appl Microbiol Biotechnol 45:16. doi:10.1007/ s002530050640 Chang YI, Shih LH, Chen SW (2005) Effects of glucose and mannose on the flocculation behavior of Saccharomyces cerevisiae at different life stages. Colloids Surf B Biointerfaces 44:614. doi:10.1016/j.colsurfb.2005.05.008 Chisti Y (2007) Biodiesel from microalgae. Biotechnol Adv 25:294 306. doi:10.1016/j.biotechadv.2007.02.001 Choi CW, Yoo S, Oh IH, Park SH (1998) Characterization of an extracellular flocculating substance produced by a planktonic cyanobacterium, Anabaena sp. J Biotechnol Lett 20:643646. doi:10.1023/A:1005358204636 Comte S, Guibaud G, Baudu M (2006) Relations between extraction protocols for activated sludge extracellular polymeric substances (EPS) and EPS complexation properties. Part I, Comparison of the efficiency of eight EPS extraction methods. Enz Microb Tech 38:237245. doi:10.1016/j.enzmictec.2005.06.016 Esser K, Kues U (1983) Flocculation and its implication for biotechnology. Proc Bioch 18:2123 Frlund B, Palmgren R, Keiding K, Nielsen PH (1996) Extraction of extracellular polymers from activated sludge using a cation

J Appl Phycol (2009) 21:559567 Shipin O, Meiring P, Phaswana R, Kluever H (1999) Integrating ponds and activated sludge process in the PETRO concept. Water Res 33:17671774. doi:10.1016/S0043-1354(98)00389-3 Sukenik A, Bilanovic D, Shelef G (1988) Flocculation of microalgae in brackish and sea waters. Biomass 15:187199. doi:10.1016/ 0144-4565(88)90084-4 Toeda K, Kurane R (1991) Microbial flocculant from Alcaligenes cupidus KT201. Agric Biol Chem 55:27932799 Winter A, Siesser WG (1994) Coccolithophores. Cambridge University Press, Cambridge, UK, pp 1520

567 Yokoi H, Yoshida T, Mori S, Hirose J, Hayashi S, Takasaki Y (1997) Biopolymer flocculant produced by an Enterobacter sp. Biotechnol Lett 19:569573. doi:10.1023/A:1018301807009 Zajic J, Leduy A (1973) Flocculant and chemical properties of a polysaccharide from Pullularia pullulans. Appl Microbiol 25:628635 Zhang J, Liu Z, Wang S, Jiang P (2002) Characterization of a bioflocculant produced by the marine myxobacterium Nannocystis sp. NU-2. J Appl Microbiol Biotechnol 59:517522. doi:10.1007/ s00253-002-1023-7