Accepted Manuscript Optimization of flocculation efficiency of lipid-rich marine Chlorella sp. biomass and evaluation of its composition in different cultivation modes Yohanis Irenius Mandik, Benjamas Cheirsilp, Piyarat Boonsawang, Poonsuk Prasertsan PII: DOI: Reference:

S0960-8524(15)00145-5 http://dx.doi.org/10.1016/j.biortech.2015.01.125 BITE 14557

To appear in:

Bioresource Technology

Received Date: Revised Date: Accepted Date:

8 December 2014 27 January 2015 28 January 2015

Please cite this article as: Mandik, Y.I., Cheirsilp, B., Boonsawang, P., Prasertsan, P., Optimization of flocculation efficiency of lipid-rich marine Chlorella sp. biomass and evaluation of its composition in different cultivation modes, Bioresource Technology (2015), doi: http://dx.doi.org/10.1016/j.biortech.2015.01.125

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Optimization of Flocculation Efficiency of Lipid-rich Marine Chlorella sp. Biomass and Evaluation of Its Composition in Different Cultivation Modes

Yohanis Irenius Mandik1,2, Benjamas Cheirsilp2*, Piyarat Boonsawang2, Poonsuk Prasertsan2 1

Department of Chemistry, Faculty of Mathematics and Natural Sciences, University of

Cenderawasih, Jalan Kamp Walker, Jayapura 99358, Indonesia. 2

Department of Industrial Biotechnology, Faculty of Agro-Industry, Prince of Songkla

University, Hat-Yai, 90112, Thailand

*Corresponding author Email address: [email protected]

1

Abstract This study aimed to optimize flocculation efficiency of lipid-rich marine Chlorella sp. biomass and evaluate its composition in different cultivation modes. Among three flocculants including Al3+, Mg2+ and Ca2+ tested, Al3+ was most effective for harvesting microalgal biomass. Four important parameters for flocculation were optimized through response surface methodology. The maximum flocculation efficiency in photoautotrophic culture was achieved at pH 10, flocculation time of 15 min, Al3+ concentration of 2.22 mM and microalgal cells of 0.47 g/L. The flocculation in mixotrophic culture required lower amount of Al3+ (0.74 mM) than that in photoautotrophic and heterotrophic cultures (2.22 mM). The biomass harvested from mixotrophic culture contained lipid at the highest content of 42.08±0.58% followed by photoautotrophic (32.08±3.88%) and heterotrophic (30.42±1.13%) cultures. The lipid-extracted microalgal biomass residues (LMBRs) contained protein as high as 38-44% and several minerals showing their potential use as animal feed and their carbohydrate content were 1629%.

Keywords: Al3+; biodiesel; flocculant; lipid; marine Chlorella sp.

2

1. Introduction Energy security, rocketed oil price, depletive resources, and climate change have been worldwide problems for decades. Nowadays, biofuels and biorefineries are expected, at least to some extent, to mitigate those problems. Microalgae are well known as microorganisms with the capability to produce biofuel. This is because the microalgae have high biomass yields without requiring any arable land for growing (John et al., 2011). Moreover, some microalgal species can grow well in saline, brackish and wastewater environment and some can even accumulate lipid to a high content (>70%) (Mata et al., 2010). These characteristics make microalgae more promising as biodiesel feedstock than terrestrial crops which rely utterly on fresh water (Daroch et al., 2013). So far, most research on algal biofuels has been conducted in two areas i.e. biodiesel synthesis from algal lipids and fermentative ethanol production from algal feedstock (Daroch et al., 2013). However, many steps in the production of biodiesel from microalgae such as biomass harvesting, lipid extraction, and transesterification of microalgal lipid have jeopardized the massive interests of algal biomass due to these costly and energyconsuming processes (Yang et al., 2010; Zheng et al., 2012). Therefore, to be more sustainable in developing microalgal biodiesel industry, and in utilizing renewable energy, the effective downstream process for microalgal biomass should be investigated. Furthermore, the cell debris after lipid extraction should also be characterized and evaluated from the view point of biorefinery. The low concentration and small size of microalgae make the harvesting process difficult. There are several methods that have been used for harvesting of microalgal biomass such as centrifugation, foam fractionation, filtration, flocculation, and gravity sedimentation. Most commercial systems choose centrifugation to harvest microalgae. Filtration could also be used in harvesting process, but membranes will be rapidly fouled by the extracellular organic

3

matter if the medium is filtered directly (Babel and Takizawa, 2010). It has also been known that the use of filtration with pressure and vacuum are suitable methods to concentrate microalgal biomass that are large in size such as Spirulina plantensis. But for recovering of small sized algae strains such as Chlorella sp. and Dunaliella sp., the pressure or vacuum filtration methods are not suitable (Harun et al., 2010). Moreover based on life cycle analysis reported by Sander and Murthy (2010), the harvesting processes using centrifugation or filtration consume large amounts of energy and need intensive maintenance. On the other hand, the cost and energy demand for harvesting microalgae could be significantly reduced if the cultures are preconcentrated. Flocculation is known as a typical preconcentration step that could rapidly reduce the large volumes of culture medium before further dewatering by centrifugation. Flocculation can be induced by pH increase (Wu et al., 2012), chemical flocculants such as inorganic metal salts (Eldridge et al., 2012; Rwehumbiza et al., 2012; Shen et al., 2013; Sanyano et al., 2013) and cationic polymers (Vandamme et al., 2010; Beach et al., 2012), and flocculating microorganisms (Papazi et al., 2009). Physical flocculation induced by ultrasound and electrocoagulation have also been reported (Vandamme et al., 2013). Among these flocculation methods, the flocculations by inorganic metal salts are widely used because of their low-cost and high efficiency. In addition, this process can be easily scaled up and applied for various species of microalgae (Uduman et al., 2010). Wu et al. (2012) investigated the flocculation of three freshwater microalgae and two marine microalgae by pH increase. They explained that Mg2+ in the growth medium might act as flocculant which coagulated microalgal cells at high pH. The pH, Mg2+ dosage and initial biomass concentration were set in the ranges of 8-12, 1.54.5 mg/L and around 1.7 g/L, respectively. The optimal pH for flocculation of freshwater microalgae (pH 10.5) was higher than that for marine microalgae (pH 9.2). Shen et al. (2013)

4

optimized the flocculation of marine alga Nannochloropsis oculata with two cationic salts based on response surface methodology. The optimum flocculation conditions were predicted at microalgal cells of 1.7 g/L, pH 8.3, and flocculant dose of 383.5 µM for aluminum sulfate and at microalgal cells of 2.2 g/L, pH 7.9, and flocculant dose of 438.1 µM for ferric chloride. These results indicate that the optimal condition for flocculation is strongly dependent upon the type of flocculant used, and its dosage varied with different pH levels, microalgae strains and cell concentrations. With the increasing microalgal biodiesel development, the lipid-extracted microalgal biomass residues (LMBRs) would be abundantly produced together with the microalgal lipid. LMBRs are rich in carbohydrates and proteins. Therefore, they can be not only used as a highprotein animal feed but also converted to some products such as fermentable sugars, hydrogen, methane, bioethanol, as well as nutrients for microalgae as a new crop (Harun et al., 2009; Zheng et al., 2012). The conversions of the LMBRs into valuable products not only provide an added bonus to offset the production costs of biodiesel but also lower the treatment or disposal costs of LMBRs. Moreover, this utilization would then give economic and environmental advantages for microalgal biodiesel production. Because marine Chlorella sp. has been identified as a good source for production of lipid and its cultivation has been optimized (Cheirsilp and Torpee, 2012), the further optimization of its harvesting process and the characterization of its biomass as potential feedstocks for valuable products would contribute greatly to its industrialized production. In this study, inorganic metal salts were screened for their effectiveness in flocculation of marine Chlorella sp. biomass and the flocculation parameters were optimized through response surface methodology (RSM). As the microalgae can be cultivated in three different cutivation modes including photoautotrophic, heterotrophic and mixotrophic cutivation modes, these cutivation

5

modes may influence the flocculation efficiency due to the differences in properties of microalgal cell surfaces, cell sizes, and various product formations (Vandamme et al., 2013). Therefore, the flocculation efficiencies of microalgal biomass in different cutivation modes were compared. The harvested microalgal biomass was then characterized for its lipid, carbohydrate, protein and mineral contents.

2. Material and methods 2.1 Microalgae strain and growth medium Marine Chlorella sp. was obtained from the National Institute of Coastal Aquaculture, Thailand. The medium used in this study was BG-11 medium (Cheirsilp and Torpee, 2012). One liter of BG-11 medium contains 1.5 g NaNO3, 0.04 g K2HPO4·3H2O, 0.2 g H2PO4·3H2O, 0.0005 g EDTA g, 0.005 g Fe ammonium citrate, 0.005 g citric acid, 0.02 g Na2CO3 and 1 mL of trace metal solution, pH 7.3. One liter of trace metal solution contains 2.85 g H3BO3, 1.8 g MnCl2·4H2O, 0.02 g ZnSO4·7H2O, 0.08 g CuSO4·5H2O, 0.08 g CoCl2·6H2O and 0.05 g Na2MoO4·2H2O.

2.2 Cultivation of microalgae Microalgae strain was pre-cultured in 400 mL of BG-11 medium in a 500 mL glass bottle. The pre-culture was incubated at 30 oC and air-aerated at a flow rate of 0.01 mL/min under a 3,000 lux light intensity with a 16:8 h light and dark cycle for 3-5 days (Cheirsilp and Torpee, 2012). This was used as a seed culture. The batch cultivation of the microalgae was performed by inoculating 10% (v/v) seed culture into 3 L BG-11 medium in a 3.78 L (1 US gallon) glass bottle. The cultures were incubated at 30 oC, air-aerated at a flow rate of 0.01 mL/min, and illuminated with a 3,000 lux light intensity with a 16:8 h light and dark cycle for

6

5 days. For mixotrophic and heterotrophic cultivation, glucose was used as a carbon source at a concentration of 2 g/L with and without light illumination, respectively. During microalgae cultivation the optical density at 660nm (OD660), pH, dry mass of microalgae, and lipid were determined.

2.3 Harvesting of the microalgal biomass Microalgal biomass were harvested by flocculation method using several flocculants including Al2(SO4)3K2SO4·24H2O, MgSO4·7H2O, and CaCl2·2H2O. The flocculant stock solution was added to the culture medium at the same molar ratio of 2.22 mM then the reaction tubes were vortexed for 5 s. The microalgal suspensions were left to settle for certain time without agitation. Subsequently, the optical density of the supernatant from half the height of the clarified layer and the sludge were measured at 660nm. Agglomerates of microalgae were then washed twice with distilled water and dried at 60 oC in a hot air oven until a constant weight. The dried microalgae were then crushed and sieved using 120 mesh analytical sieve before used for lipid extraction. Concentration factor was calculated using Eq. 1 and flocculation efficiency was calculated using Eq. 2 (Wu et al., 2012): Concentration factor = Final cell concentration / Initial cell concentration Flocculation efficiency (%) = (1-A/B)
100

(1) (2)

A is the OD660 of supernatant from half the height of the clarified layer after flocculation and B is the initial OD660 of the microalgal culture suspension. All experiments were performed with at least three replicates. Analysis of variance was performed to identify any significant differences in the treatment mean values, and the least significant difference (p ≤ 0.05), calculated using SPSS software, was used to separate means.

7

2.4 Optimization through Response Surface Methodology (RSM) Based on the results from screening of flocculant, the suitable flocculant was then selected and used in further study. Factors affecting flocculation process including pH (8-12), flocculation time (5-15 min), flocculant ions concentration (0.74-3.70 mM), and initial concentration of microalgal cells (0.17-0.47 g/L) were varied based on preliminary experimental results and previous reports (Vandamme et al., 2010; Wu et al., 2012; Shen et al., 2013). The effects of these factors on flocculation efficiency of marine Chlorella sp.biomass were evaluated and optimized through Response Surface Methodology (RSM). Design Expert 8.0.6 (Stat-Ease, Inc., Minneapolis, MN, USA) was used for regression analysis of experimental data and to plot response surface with the experimental results of flocculation efficiency from the all of design experimental setup. The experiments were designed based on a Box-Benhken Design (BBD) with a quadratic model employed to study the combined effect of independent variables on flocculation efficiency. Each variable was varied at three levels (1, 0, +1). The flocculation efficiency was a mean of three replications and used as the dependent variable. The second order polynomial quadratic equation (Eq. 3) was fitted to evaluate the main effect and interaction of each independent variable to the response.
=  + ∑    + ∑    + ∑

 



(3)

where Y represents the response of experimental flocculation efficiency, i and j are linear and quadratic coefficients respectively, β is the regression coefficient, n is the number of variables studied in the experiments, and X are factors (independent variables), they are represent pH, flocculation time (min), flocculant ions concentration (mg/L), and initial concentration of microalgal cells (g/L).

8

2.5 Lipid extraction of microalgal biomass Extraction of lipid from the biomass was performed according to the modified procedure of Maxwell et al. (1968). Dried microalgal biomass 40 mg were extracted with 4 mL of n-hexane and sonicated with sonicator (Model Transsonic 460/H, εlma Ltd., Germany) at room temperature. The extraction processes were conducted twice. The suspension was filtered through Whatman No.40 filter paper and the filtrate was taken into pre-weighed glass vial. The hexane solution was evaporated to dryness at 60 oC under vacuum using a rotary evaporator (Eyela, Japan). The extracted lipid was dried and weighed. The lipid content was calculated as percentage of lipid to microalgal biomass. The lipid-extracted microalgal biomass residues (LMBRs) were the retentate after filtration. They were dried at 60 oC in a hot air oven until a constant weight. The dried LMBRs were characterized for their protein, carbohydrate, mineral, and sugar composition.

2.6 Determination of fatty acid composition in microalgal lipid The lipid was converted to fatty acid methyl esters (FAME) before analysis for fatty acid composition (Jham et al., 1982). The first step was conducted with 1 mL of KOH/MeOH (0.5 M) at 100oC for 5 min. In the second step, 400 µL of aq. HCl/MeOH (4:1, v/v) was added to the mixture from the first step and then the mixture was heated in an oil bath for 15 min at 100oC. Then, the tube was cooled and 2 mL of water was added. After that, the mixture was extracted with 2
3 mL of petroleum ether. The organic layer was dried quickly over anhydrous Na2SO4, evaporated and redissolved in 500 µL of CHCl3. The analysis of FAME was conducted using a Gas Chromatograph (HP6850) equipped with a cross-linked capillary FFAP column (length 30 m, 0.32 mm I.D., 0.25 µm film thickness) and a flame ionization

9

detector (FID). The operating conditions were as follows: inlet temperature 290oC, oven initial temperature 210oC held 12 min, then ramped to 250oC at 20 oC/min rate and held 8 min, and detector temperature 300 oC. Fatty acids were identified by comparing their retention times with those of standard ones and calculated as percentage based on their respective peak area using a standard mixture of FAME.

2.7 Characterization of marine Chlorella sp. biomass after lipid extraction 2.7.1

Determination of sugar composition The LMBRs were hydrolyzed by two-step acid hydrolysis (Hoebler et al., 1989). A

small amount of dry LMBRs was added to 3 mL 72% (w/w) sulfuric acid and incubated for 20 min at 30 oC for the primary hydrolysis. The hydrolysate was then diluted to 4% (w/w) sulfuric acid and autoclaved for 20 min at 121 oC as the secondary hydrolysis. The supernatant was neutralized and analyzed for sugar composition. Sugar composition of hydrolyzate was analyzed by using reverse phase high performance liquid chromatography (HPLC, Hewlett Packard, Germany), ZORBAX NH2 5 µm 4.6
250 mm column, mobile phase of acetonitrile:water (75:25), column temperature was 25oC, refractive index detector (RID). 2.7.2 Determination of carbohydrate content Carbohydrate content of LMBRs was calculated based on the measurement of sugar content in LMBRs by the phenol sulfuric acid method after purifying the dried LMBRs using hot alkaline extraction method. Dried LMBRs (10 mg) were resuspended in 0.2 mL deionized water and mixed with 0.4 mL 40% KOH then heated at 90 oC for 1 h. After cooling, 0.2 mL of cold absolute ethanol was added. The sample was then stored overnight in a fridge at -20oC. Furthermore, the sample was centrifuged at 8,000 g and the supernatant was discarded. The pellet was dried overnight in 60 oC oven then the dried pellet (4 mg) was then resuspended in

10

0.5 mL deionized water and added with 0.25 mL of 5% phenol, followed by 1.25 mL of 98% sulfuric acid. The mixture was reacted at 30oC for 5 min. After cooling, the absorbance of yellow-orange color was measured using spectrophotometer (Libra S22 Biochrom, UK) at 490 nm for hexose monosaccharide determination. The blank was prepared by replacing the sample with deionized water. The amount of carbohydrate in LMBRs was calculated using standard curve of glucose and expressed as amount of sugar. 2.7.3 Determination of protein content Protein content of LMBRs was determined using elemental analyzer with thermal conductivity detector (LECO CHN 628 series, USA) by measuring the total kjeldahl nitrogen (TKN) and then the protein content was calculated by multiplying the TKN with factor of 6.25 (Sweeney, 1989). The measurement conditions were as follows: mass of sample 0.2 g, combustion furnace temperature 950 oC, after burner temperature 850 oC, baseline delay time 10 s, minimum analysis time 40 s, comparator level 100, endline time 2 s, IR baseline time 1 s, IR stop flow time 5 s, TC baseline 10 s, burn steps (step 1 time was 15 s at high furnace flow, step 2 time 300 s at medium furnace flow, step 3 time 5 s at high furnace flow), ballast equilibrate time 30 s, ballast not filled timeout 600 s, aliquot loop fill pressure drop 200 mmHg, aliquot loop equilibrate pressure time 8 s, oxygen flow 40 psi, and helium flow 40 psi. 2.7.4 Determination of mineral content Mineral content in LMBRs were analyzed by an inductively coupled plasma optical emission spectrometry (ICP-OES) using optical emission spectrometer (Optima 4300 DV, Perkin Elmer Instruments, USA). The method for sample preparation referred to atomic absorption spectroscopy analytical methods (Perkin-Elmer, 1996). One gram of dried LMBRs were placed in a porcelain crucible, and then placed in a cool muffle furnace and ashed at 600 o

C for 3 h. The ash was then cooled and dissolved in 5 mL of 20% HNO3. The solution was

11

filtered through an acid-washed filter paper into a 50 mL volumetric flask. The solution was diluted to appropriate volume with deionized water and mixed well. This sample solution was used for mineral analysis using ICP-OES referring to WI-RES-ICP-OES-001 test method. The wave length of each element was as follows: Pb (220.353 nm), As (188.979 nm), Fe (238.204 nm), Cu (327.393 nm), Cd (228.802 nm), Cr (267.716 nm), Zn (206.200 nm), Na (589.592 nm), Ba (233.527 nm), Mn (257.610 nm), Ni (231.604 nm), K (766.490 nm), Ca (317.933 nm), Mg (285.213 nm), Al (396.153 nm), Co (228.616 nm), Sr (460.733 nm), and Tl (190.801 nm).

3. Results and Discussion 3.1 Screening of flocculants In this study, the suitable flocculants for harvesting of marine Chlorella sp. in photoautotrophic culture were screened. The unadjusted final culture pH was around 10. Three flocculant ions including Mg2+, Ca2+, and Al3+ each was added at 2.22 mM. The flocculation efficiency with and without (control) addition of flocculants were compared (Figure 1). After 15 min of flocculation time, Al3+ ions showed the highest flocculation efficiency of 56.48%, while Ca2+ and Mg2+ ions showed lower flocculation efficiency at 31.19% and 30.22%, respectively. The culture medium without flocculant showed the lowest flocculation efficiency of 5.42%. Although flocculation could occur spontaneously in microalgal cultures when pH increases above 9 (Spilling et al., 2011), this property also depends on the medium composition which have to contain certain amount of flocculants to induce precipitation. After 60 min of flocculation time, the flocculation efficiency using Al3+, Ca2+, and Mg2+ ions slightly increased up to 64.98%, 32.43%, and 32.06%, respectively. The concentration factors using these ions were 4.33, 2.16, and 2.14, respectively. The mechanism of flocculation by Al3+ could be explained as follows: the presence of Al3+ in the microalgae cultures would form

12

Al(OH)3 which agglomerates at alkaline pH. The agglomerates then induce flocculation of microalgal cells by charge neutralization and/or sweep flocculation (Wu et al., 2012). The culture pH influences the charge of not only the microalgal cell surface but often also of chemical flocculants, and is therefore an important parameter to be considered. Wu et al. (2012) reported that when pH was increased Mg2+ in the growth media might produce Mg(OH)2 which precipitated at high pH and induced the flocculation of the microalgae. Vandamme et al. (2010) also reported that the presence of Mg2+ in the growth medium was essential for flocculation and a condition of pH more than 10 could induce self-flocculation of Chlorella vulgaris. Since there are several factors affecting the flocculation of the microalgal biomass such as medium components, microalgae species, pH, the appropriate ions for flocculation could be different. In the present study, the trivalent cation Al3+, which showed the highest flocculation efficiency, might have a higher capacity to neutralize the charge of marine Chlorella sp. cells and then sweep marine Chlorella sp. cells than did the bivalent cations (Mg2+ and Ca2+). Al3+ ions have also been reported to be effective in flocculation of marine microalgae such as Tetraselmis suecica, Chlorococcum sp., and Nannochloropsis salina (Eldridge et al., 2012; Rwehumbiza et al., 2012).

3.2 Maximizing flocculation efficiency of marine Chlorella sp. biomass in photoautophic culture through RSM 3.2.1 Statistical analysis on flocculation efficiency Based on the results of flocculant screening, Al3+ ion was selected and used as flocculant in optimization study. Four important parameters involved in the flocculation process were optimized. They were pH, flocculation time, flocculant ions concentration, and initial concentration of microalgal cells. The experimental design with the observed responses

13

for flocculation efficiency are presented in Table 1. Based on the experimental design, flocculation efficiencies were in range from 0 to 78.4%. It was found that Run 24 (pH of 10, flocculation time of 15 min, Al3+ ions oncentration of 2.22 mM, and initial concentration of microalgal cells of 0.47 g/L) gave the maximum flocculation efficiency of 78.4% with a concentration factor of 5.23. With the same initial concentration of microalgal cells (0.47 g/L) and Al3+ ions concentration (2.22 mM), the comparable flocculation efficiency (76.9%) with a concentration factor of 5.13 could be obtained when using a higher pH of 12 with a shorter flocculation time of 10 min (Run 9). The condition with a high pH of 12 and a high Al3+ ions concentration of 3.70 mM also showed high flocculation efficiency of 78.0% with a concentration factor of 5.20 (Run 23). Furthermore, the experimental data were analyzed using Design Expert software (8.0.6 version) for analysis of variance (ANOVA), regression coefficient, and regression equation. Sequential P-value was used to select the highest order polynomial where the additional terms are significant and the model is not aliased. The suitable model that have insignificant lack-offit and maximum R-squared and maximum adjusted R-squared was selected (Table 2). The response functions in the term actual factor to predict flocculation efficiency (Yi) given in Eq.4.

Yi = – 592.55 + 119.39A + 2.11B – 14.31C– 230.20D + 0.32AB + 5.94AC + 25.70AD – 1.62BC + 0.25BD + 30.24CD – 6.52A2 – 0.01B2– 7.29C2 – 40.52D2

(4)

where independent variables A is pH, B is flocculation time, C is Al3+ ions concentration, and D is initial concentration of microalgal cells. The design of the present experiment including dependent variables response Yi, was flocculation efficiency, A, B, C, D are referred as the

14

main effort linear terms while AB, AC, AD, BC, BD, CD are the interaction terms, and A2, B2, C2, D2 are the quadratic terms involved in the process. The statistical significance of the model equation was analyzed by the F-test for analysis of variance (ANOVA), to the fitted model. The P-values < 0.05 implies that model was statistically valid. The highest determination coefficient (R2) obtained was 0.9333. This indicated that up to 93.3% of the variations in response can be explained by the model. The regression of linear term A (pH) was the most significant factor of flocculation efficiency (Pvalue < 0.0001), followed by linear term C (Al3+ ions concentration) with P-value of 0.0006, linear term D (initial concentration of microalgal cells) with P-value of 0.0031, and linear term B (flocculation time) with P-value of 0.0143, respectively (Table 2). The interaction between pH (A) vs. Al3+ ions concentration (C) with P-value of 0.0042, and the interaction between flocculation time (B) vs. Al3+ ions concentration (C) with P-value of 0.0332 significantly influenced the flocculation efficiency. On the other hand, the interaction between pH (A) vs. flocculation time (B), pH (A) vs. initial concentration of microalgal cells (D), flocculation time (B) vs. Initial concentration of microalgal cells (D), and Al3+ ions concentration (C) vs. initial concentration of microalgal cells (D) were not significant (P-value > 0.05). For quadratic terms A2 with P-value less than 0.0001 and C2 with P-value of 0.003 were significant.

3.2.2 Response surface plots for flocculation efficiency The three-dimensional response surface and two-dimensional contour plots from the calculated responses in which two variables were kept constant at their center points and the other two variables were varied within their experimental range, are shown in Figure 2. These graphs were plotted in order to investigate the interactions between independent variables and determine the optimal level of each variable for a desired response.

15

Figure 2a presents the interaction effect between pH (A) vs. Al3+ ions concentration (C). Because the effectiveness of ions flocculant depends strongly on the pH of the medium, pH adjustment is needed to obtain maximum flocculation efficiency. At a low Al3+ ions concentration (< 2.22 mM), the optimal pH for maximum flocculation efficiency was found at 10.0. With increasing Al3+ ions concentration > 2.22 mM, the optimal pH was shifted to a higher level of pH 12 and this gave higher flocculation efficiency. The highest flocculation efficiency was then obtained at the maximum pH level tested of 12 and Al3+ ions concentration tested of 3.70 mM. It was possible that high pH and excess flocculants may promote the precipitation of flocculants, and these precipitates that carry a large adsorptive surface area and a positive superficial charge may enhance the flocculation efficiency (Wu et al., 2012). Figure 2b shows the interaction effect between flocculation time (B) vs. Al3+ ions concentration (C). As the flocculant concentration influences both the extent and rate of flocculation, at low Al3+ ions concentration there is a need to increase the flocculation time up to 15 min to reach the highest flocculation efficiency. However, with increasing the Al3+ ions concentration >2.22 mM the flocculation time could be reduced to the minimum level tested of 5 min. This result suggests the possible adjustment between Al3+ ions concentration and flocculation time. The higher Al3+ ions concentration used, the faster flocculation would occur. However, increasing Al3+ ions concentration is costly and may limit the further use of microalgal biomass. Slow flocculation also negatively affects the cost of flocculation. It has been reported that rapid flocculation requires a smaller harvesting unit and thus incurs lower investment costs (Vandamme et al., 2013). Therefore, the acceptable flocculation time might be set before the minimum effective dose of Al3+ ions is determined. Figures 2c and 2d show the interaction effects between pH (A) vs. flocculation time (B) and pH (A) vs. initial concentration of microalgal cells (D), respectively. The flocculation

16

efficiency was less affected by flocculation time and initial concentration of biomass when using Al3+ ions concentration at the center point of 2.22 mM. The maximum flocculation efficiency could be obtained at any flocculation time and initial concentration of microalgal cells if the pH was set at the level > 10. In addition, there was less interaction effect between pH vs. flocculation time and pH vs. initial concentration of biomass. Figure 2e also shows that the flocculation efficiency was less affected by flocculation time and initial concentration of microalgal cells when using Al3+ ions concentration and pH at the center points of 2.22 mM and pH 10, respectively. Figure 2f shows the interaction effect between initial concentration of microalgal cells and Al3+ ions concentration. This figure also shows that when flocculation time and pH are set at the center points, there is a need to increase Al3+ ions concentration to harvest a high concentration of microalgal cells. However, increasing the Al3+ ions concentration also increases the flocculation cost. Therefore, the pH should be first adjusted to the maximum level and the flocculation time should be extended to the acceptable level before determination of the minimum effective dose of Al3+ions.

3.3 Flocculation efficiency in different cultivation modes The flocculation efficiency depends on not only pH and cell concentration but also the properties of microalgal cell surfaces, cell sizes and various product formations, and these variables may vary within a species depending on culture conditions (Vandamme et al., 2013). Therefore, the flocculation efficiency using Al3+ ions as flocculant for harvesting of marine Chlorella sp. in photoautotrophic, heterotrophic, and mixotrophic cultivation modes were compared (Figure 3). As the microalgal cell concentration and pH in each cultivation mode were different, to eliminate these effects the initial concentration of microalgal cells and pH were adjusted to be the same at 0.47 g/L and 10, respectively. It was obvious that the suitable

17

Al3+ ions amount and time needed for harvesting of biomass were different in different cultivation modes. The minimum amount of Al3+ ions required for harvesting biomass from photoautotrophic and heterotrophic cultures was the same at 2.22 mM, whereas harvesting biomass from mixotrophic culture needed only 0.74 mM Al3+ ions. Interestingly, the flocculation time needed for mixotrophic culture was also shortest at 5 min followed by heterotrophic and photoautotrophic cultures (15 min and 30 min, respectively). The highest flocculation efficiencies in each mode were 97.37±1.62 %, 70.92±0.25% and 85.92±2.06%, respectively. After flocculation, the final cell concentrations in each mode were 3.05 g/L, 2.22 g/L and 2.69 g/L with the concentration factors of 6.49, 4.73 and 5.73, respectively. This is the first report for the flocculation efficiencies in different cultivation modes. As initial cell concentration and pH were adjusted to be the same, the different flocculation efficiencies in each cultivation mode would be due to the differences in properties of microalgal cell surfaces, cell sizes, and product formations. As it has been reported that smaller cell size requires a higher flocculant amount (Schenk et al., 2008), the larger cell size of microalgae observed in mixotrophic culture (data not shown) would be one reason for the higher flocculation efficiency. In addition, microalgae often excrete significant quantities of organic matter such as polysaccharides and proteins into the growth medium (Hulatt and Thomas, 2010). These matters may enhance the interaction between microalgal cell surface and flocculants and thus reduces the chemical demand for flocculation. However, the excessive addition of flocculants (more than the optimum dosage for flocculation efficiency) seem to destabilize charge neutralization and agglomerate formation then affect the sweeping flocculation process.

18

3.4 Characterization of marine Chlorella sp. biomass There are many factors that determine the characteristics of microalgal biomass such as microalgae strain, growth medium, growth conditions, cultivation modes and time of harvesting. Lipid content and composition of marine Chlorella sp. LMBRs obtained from different cultivation modes are summarized in Table 3. The mixotrophic culture gave the highest lipid content (42.08±0.58% based on dry biomass) followed by photoautotrophic (32.08±3.88%) and heterotrophic (30.42±1.13%) cultures. The lipid contents determined in this research were similar to those previously reported by Cheirsilp and Torpee (2012). After lipid extraction, the LMBRs were determined for their protein and carbohydrate contents. The protein content of LMBRs from heterotrophic culture was highest at 44.6±0.03% based on dry LMBRs followed by mixotrophic (40.1±0.15%) and photoautotrophic (38.9±0.05%) cultures. The protein content of marine Chlorella sp. LMBRs in this study was comparable to that of Haematococcus pluvialis LMBRs (40.3%) (Ju et al., 2012) and higher than that of Scenedesmus sp. LMBRs (32.4%) (Yang et al., 2010) and Dunaliella tertiolecta (UTEX LB 999) (13.4%) (Goo et al., 2013). The LMBRs from mixotrophic culture contained carbohydrate at the highest content of 29.2±0.03% while those in photoautotrophic and heterotrophic cultures contained only about 16%. The carbohydrate content in this study was close to that of Scenedesmus sp. LMBRs (24.7%) (Yang et al., 2010). The percentage of fatty acid compositions in lipid extracted from marine Chlorella sp. cultivated in different modes are shown in Table 4. Table 4 shows that marine Chlorella sp. generally accumulated fatty acids ranging from C12:0 to C24:1 with the majority of palmitic and oleic fatty acids. Palmitic acid was the highest fatty acid found in three cultivation modes (48.9% in heterotrophic, 46.3% in mixotrophic and 44.8% in photoautotrophic cultures). Meanwhile, very long chain fatty acid

19

(C24:1) was found in biomass cultivated in heterotrophic mode (2.57%). The highest content of saturated fatty acids (SFA) was also found in biomass cultivated in heterotrophic mode (80.75%) followed by mixotrophic (71.9%) and photoautotrophic (60.5%) cultures. The marine Chlorella sp. fatty acids are more saturated than fatty acids of soybean oil which is used as feedstock for biodiesel production in the US and the EU. It has been known that the more saturated oil could provide biodiesel with higher cetane number (CN), decreased NOx emissions, a shorter ignition delay time, and oxidative stability (Antolin et al., 2002). Other research also found that palmitic and oleic acids were the main components in microalgae (Dayananda et al., 2006; Halim et al., 2012). The sugar composition of LMBRs determined using HPLC showed that glucose is the main monosaccharide constituent in marine Chlorella sp. carbohydrate. The results are consistent with the results of Goo et al. (2013) who reported that Dunaliella tertiolecta (UTEX LB 999) carbohydrate consisted of mainly glucose and it is then a promising candidate for the source of biorefinery. Meanwhile, the carbohydrates from Dunaliella salina were mainly composed of galactose, xylose, and glucose (Mishrav and Jha, 2009). The determination of mineral content using ICP-OES showed that the marine Chlorella sp. LMBRs contained macro minerals such as Mg 101 g/kg, K 5,815 mg/kg, Na 2,537 mg/kg and Ca 187.76 mg/kg, and micro minerals content such as Fe 2,360 mg/kg, Mn 869.12 mg/kg, Cu 76.28 mg/kg, and Zn 93.03 mg/kg. It should be noted that although the microalgae was flocculated using Al3+ and the agglomerates of microalgal biomass might contain some extent of aluminium hydroxide precipitate, the amount of Al found in marine Chlorella sp. LMBRs (172.37 mg/kg) was lower than the maximum tolerable levels of Al in the feed for poultry and swine (1,000 mg/kg) (Mineral Tolerance of Animals, 2005). Several minerals that were not added to the growth medium were also found in the marine Chlorella sp. LMBRs. These

20

include Pb (3.46 mg/kg), Cr (0.72 mg/kg), Ba (1.85 mg/kg), Ni (5.24 mg/kg), Sr (4.19 mg/kg), and Tl (0.56 mg/kg). Ju et al. (2012) also found that Haematococcus pluvialis LMBRs comprised macro minerals such as Mg 8,200 mg/kg, K 4,300 mg/kg, Ca 4,800 mg/kg, and Na 2,000 mg/kg, and micro minerals content such as Fe 1287 mg/kg, Mn 87 mg/kg, Zn 396 mg/kg, B 23 mg/kg, Cu 11 mg/kg. Based on these mineral compositions, there is a great potential of marine Chlorella sp. LMBRs to be used as feedstock for the production of animal supplements (Cromwell, 1997).

4. Conclusions Lipid-rich marine Chlorella sp. biomass can be efficiently harvested using Al3+ ions. The flocculation efficiency was optimized through Response Surface Methodolody. The minimum requirement of Al3+ ions for harvesting of biomass from mixotrophic culture was lower than those from photoautotrophic and heterotrophic cultures. The marine Chlorella sp. lipid contained mainly saturated fatty acids that are suitable to be used as biodiesel feedstock. The microalgal biomass after lipid extraction having a high content of protein and several minerals shows its great potential to be used as animal feed and the sugar composition of its carbohydrate can be used as sugar feedstock.

Acknowledgement The first author would like to convey his heartfelt gratitude to the Directorate General of Higher Education-Ministry of Education and Culture, the Republic of Indonesia, Rector of University of Cenderawasih Indonesia. This work was financially supported by Graduate School of Prince of Songkla University and Thai government under Grant No. AGR560094S. This work was also supported by Thai Research Fund under Grant No. RTA5780002.

21

References 1. Antolin, G., Tinaut, F.V., Briceno, Y. 2002. Optimisation of biodiesel production by sunflower oil transesterification. Bioresour. Technol. 83, 111–114. 2. Babel, S., Takizawa, S. 2010. Microfiltration membrane fouling and cake behavior during algal filtration. Desalination. 261, 46–51. 3. Beach, E.S., Eckelman, M.J., Cui, Z., Brentner, L., Zimmerman, J.B. 2012. Preferential technological and life cycle environmental performance of chitosan flocculation for harvesting of the green algae Neochloris oleoabundans. Bioresour. Technol. 121, 445-449. 4. Cheirsilp, B., Torpee, S. 2012. Enhanced growth and lipid production of microalgae under mixotrophic culture condition: Effect of light intensity, glucose concentration and fedbatch cultivation. Bioresour. Technol. 110, 510–516. 5. Cromwell, G.L., 1997. Trace elements and animal production. In: Fisher, P.W.F., L’Abbé, M.R., Cockeli, K.A., Gibson, R.S. (Eds.), Trace elements in man and animals-9: Proceedings of the ninth international symposium of trace elements in man and animals, NRC Research press, Ottawa, pp. 290-293. 6. Daroch, M., Geng, S., Wang, G. 2013. Recent advances in liquid biofuel production from algal feedstocks. Appl. Energy. 102, 1371-1381. 7. Goo, B.G., Baek, G., Choi, D.J., Park, Y.I., Synytsya, A., Bleha, R., Seong, D.H., Lee, C.G., Park, J.K. 2013. Characterization of a renewable extracellular polysaccharide from defatted microalgae Dunaliella tertiolecta. Bioresour. Technol. 129, 343–350 8. Halim, R., Danquah, M.K., Webley, P.A. 2012. Extraction of oil from microalgae for biodiesel production: A review. Biotechnol. Adv. 30, 709–732.

22

9. Harun, R., Danquah, M.K., Forde, G.M. 2009. Microalgal biomass as a fermentation feedstock for bioethanol production, J. Chem. Technol. Biotechnol. 85, 199–203. 10. Harun, R., Singh, M., Forde, G.M., Danquah, M.K. 2010. Bioprocess engineering of microalgae to produce a variety of consumer products. Renew. Sust. Energ. Rev. 14, 1037–1047. 11. Hoebler, C., Barry, J.L., David, A., Delort-Laval, J., 1989. Rapid acid hydrolysis of plant cell wall polysaccharides and simplified quantitative determination of their neutral monosaccharides by gas-liquid chromatography. J. Agricult. Food Chem. 37, 360–367. 12. Hulatt, C.J., Thomas, D.N. 2010. Dissolved organic matter (DOM) in microalgal photobioreactors: a potential loss in solar energy conversion? Bioresour. Technol. 101, 8690-8697. 13. Jham, G. N., Teles, F. F. F., Campos, L. G. 1982. Use of aqueous HCl/MeOH as esterification reagent for analysis of fatty acids derived from soybean lipids. J. Am. Oil Chem. Soc. 59(3), 132-133. 14. John, R.P., Anisha, G.S., Nampoothiri, K.M., Pandey, A. 2011. Micro and macroalgal biomass: A renewable source for bioethanol. Bioresour. Technol. 102, 186-193. 15. Ju, Z-Y., Deng, D-F., Dominy, W. 2012. A defatted microalgae (Haematococcus pluvialis) meal as a protein ingredient to partially replace fishmeal in diets of Pacific white shrimp (Litopenaeus vannamei, Boone, 1931). Aquaculture 354–355, 50–55. 16. Mata, T.M., Martins, A.A., Caetano, N.S. 2010. Microalgae for biodiesel production and other applications, a review. Renew. Sust. Energ. Rev. 14, 217–232. 17. Maxwell, J. R., Douglas, A. G., Eglinton, G., McCormick, A. 1968. The botryococceneshydrocarbons of novel structure from the algae Botryococcus braunii Kutzing. Phytochem. 7, 2157-2171.

23

18. Mineral Tolerance of Animals. 2005. Committee on Minerals and Toxic Substances in Diets and Water for Animals, Board on Agriculture and Natural Resources, Division on Earth and Life Studies, National Research Council, U.S. 19. Mishrav, A., Jha, B. 2009. Isolation and characterization of extracellular polymeric substances from microalgae Dunaliella salina under salt stress. Bioresour. Technol. 100, 3382-3386. 20. Papazi, A., Makridis, P., Divanach, P. 2009. Harvesting Chlorella minutissima using cell coagulants. J. Appl. Phycol. 9, 1573–1576. 21. Perkin-Elmer Corporation. 1996. Analytical Methods for Atomic Absorption Spectroscopy. USA. 22. Rwehumbiza, V.M., Harrison, R., Thomsen, L. 2012. Alum-induced flocculation of preconcentrated Nannochloropsis salina: residual aluminium in the biomass, FAMEs and its effects on microalgae growth upon media recycling. Chem. Eng. J. 200–202, 168–175. 23. Sander, K., Murthy, G. 2010. Life cycle analysis of algae biodiesel. Int. J. Life Cycle Ass. 5, 704–714. 24. Sanyano, N., Chetpattananondh, P., Chongkhong, S. 2013. Coagulation–flocculation of marine Chlorella sp. for biodiesel production. Bioresour. Technol. 147, 471–476. 25. Schenk, P.M., Thomas-Hall, S.R., Stephens, E., Marx, U.C., Mussgnug, J.H., Posten, C., Kruse, O., Hankamer, B. 2008. Second generation biofuels: high-efficiency microalgae for biodiesel production. Bioenerg. Res. 1, 20-43. 26. Shen, Y., Cui, Y., Yuan, W. 2013. Flocculation optimization of microalga Nannochloropsis oculata. Appl. Biochem. Biotechnol. 169, 2049-2063. 27. Spilling, K., Seppälä, J., Tamminen, T. 2011. Inducing autoflocculation in the diatom Phaeodactylum tricornutum through CO2 regulation. J. Appl. Phycol. 23, 959-966.

24

28. Sukenik, A., Shelef, G. 1984. Algal autoflocculation–verification and proposed mechanism. Biotechnol. Bioeng. 26, 142–147. 29. Sweeney, R.A. 1989. Generic combustion method for determination of crude protein in feeds: collaborative study. J. Assoc. Off. Anal. Chem. 72, 770-774. 30. Uduman, N., Qi, Y., Danquah, M.K., Forde, G.M., Hoadley, A. 2010. Dewatering of microalgal cultures: a major bottleneck to algae-based fuels. J. Renew. Sustain. Energy 2, 0127011–0127015. 31. Vandamme, D., Foubert, I., Meesschaert, B., Muylaert, K. 2010. Flocculation of microalgae using cationic starch. J. Appl. Phycol. 22, 525–530. 32. Vandamme, D., Foubert, I., Muylaert, K. 2013. Flocculation as a low-cost method for harvesting microalgae for bulk biomass production. Trends Biotechnol. 31, 233-239. 33. Wu, Z., Zhu, Y., Huang, W., Zhang, C., Li, T., Zhang, Y., Li, A. 2012. Evaluation of flocculation induced by pH increase for harvesting microalgae and reuse of flocculated medium. Bioresour. Technol. 110, 496–502. 34. Yang, Z., Guo, R., Xu, X., Fan, X., Li, X. 2010. Enhanced hydrogen production from lipidextracted microalgal biomass residues through pretreatment. Int. J. Hydrogen Energy. 35, 9618-9623. 35. Zheng, H.L., Gao, Z., Yin, F.W., Ji, X.J., Huang, H. 2012. Lipid production of Chlorella vulgaris from lipid-extracted microalgal biomass residues through two-step enzymatic hydrolysis. Bioresour. Technol. 117, 1–6.

25

Figure legends Figure 1 Comparison of flocculation efficiency using three flocculants at the same molar concentration of 2.22 mM in harvesting of marine Chlorella sp. biomass at various times. Figure 2 Contour and 3D of response surface plots of flocculation efficiency. (a) Effect of pH (A) and Al3+ ions concentration (C) (b) Effect of flocculation time (B) and Al3+ ions concentration (C) (c) Effect of pH (A) and flocculation time (B) (d) Effect of pH (A) and initial concentration of microalgal cells (D) (e) Effect of flocculation time(B) and initial concentration of microalgal cells (D) (f) Effect of Al3+ ions concentration (C) and initial concentration of microalgal cells (D) Figure 3 The flocculation efficiency using Al3+ ions as flocculant for harvesting of marine Chlorella sp. in photoautotrophic (a), heterotrophic (b), and mixotrophic (c) cultivation modes

26

Flocculation efficiency (%)

100 80 60 40 20 0 0

15 30 45 Flocculation time (min) Mg2+

Ca2+

60

Al3+

Figure 1 Comparison of flocculation efficiency using three flocculants at the same molar concentration of 2.22 mM in harvesting of marine Chlorella sp. biomass at various times.

27

Figure 2 Contour and 3D of response surface plots of flocculation efficiency. (a) Effect of pH (A) and Al3+ ions concentration (C) (b) Effect of flocculation time (B) and Al3+ ions concentration (C)

28

(c) Effect of pH (A) and flocculation time (B) (d) Effect of pH (A) and initial concentration of microalgal cells (D) (e) Effect of flocculation time(B) and initial concentration of microalgal cells (D) (f) Effect of Al3+ ions concentration (C) and initial concentration of microalgal cells (D)

29

(b) Heterotrophic culture 100 Flocculation efficiency (%)

Flocculation efficiency (%)

(a) Photoautotrophic culture 100 80 60 40 5 min 15 min 30 min 60 min

20 0

80

5 min 15 min 30 min 60 min

60 40 20 0

00

1.5 3.0 4.5 6.0 20 40 60 80 Al3+ concentration (mM)

7.5 100

00

1.5 3.0 4.5 6.0 20 40 60 80 Al3+ concentration (mM)

7.5 100

Flocculation efficiency (%)

(c) Mixotrophic culture 100 80 60 40 5 min 15 min 30 min 60 min

20 0 0

1.5 3.0 4.5 6.0 20 40 60 80 Al3+ concentration (mM)

7.5 100

Figure 3 The flocculation efficiency using Al3+ ions as flocculant for harvesting of marine Chlorella sp. in photoautotrophic (a), heterotrophic (b), and mixotrophic (c) cultivation modes

30

Table 1 BBD with the experimental responses Factor 1 Run A: pH

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

-1(8) 0(10) 0(10) 0(10) 0(10) 0(10) 0(10) -1(8) +1(12) +1(12) 0(10) 0(10) 0(10) +1(12) 0(10) -1(8) -1(8) 0(10) +1(12) +1(12) -1(8) 0(10) +1(12) 0(10) -1(8) 0(10) 0(10) 0(10) 0(10)

Factor 2

Factor 3

B: Flocculation time

C: Al3+ ions concentration

(min) 0(10) 0(10) 0(10) -1(5) 0(10) -1(5) 0(10) 0(10) 0(10) +1(15) 0(10) -1(5) 0(10) 0(10) -1(5) 0(10) +1(15) +1(15) -1(5) 0(10) -1(5) +1(15) 0(10) +1(15) 0(10) 0(10) 0(10) 0(10) +1(15)

(mM) +1(3.70) -1(0.74) +1(3.70) -1(0.74) 0(2.22) 0(2.22) 0(2.22) 0(2.22) 0(2.22) 0(2.22) +1(3.70) +1(3.70) 0(2.22) -1(0.74) 0(2.22) 0(2.22) 0(2.22) 0(2.22) 0(2.22) 0(2.22) 0(2.22) +1(3.70) +1(3.70) 0(2.22) -1(0.74) 0(2.22) 0(2.22) -1(0.74) -1(0.74)

Factor 4 D: Initial concentration of microalgal cells (g/L) 0(0.32) +1(0.47) +1(0.47) 0(0.32) 0(0.32) +1(0.47) 0(0.32) -1(0.17) +1(0.47) 0(0.32) -1(0.17) 0(0.32) 0(0.32) 0(0.32) -1(0.17) +1(0.47) 0(0.32) -1(0.17) 0(0.32) -1(0.17) 0(0.32) 0(0.32) 0(0.32) +1(0.47) 0(0.32) 0(0.32) 0(0.32) -1(0.17) 0(0.32)

Response Flocculation efficiency (%) 2.51 47.1 65.1 0 58.7 62.6 58.9 1.48 76.9 66.4 21.2 59.9 58.7 5.45 43.9 0 0 58.9 53.5 47.6 0 63.6 78.0 78.4 0 59.1 59.3 30.1 52.1

Note: Values in parentheses are uncoded independent variables

31

Table 2 Regression coefficients of response variables Flocculation efficiency Source

Coefficients

P-value

Model intercept

-592.55

< 0.0001

A-pH

119.39

< 0.0001*

B-Flocculation time

2.11

0.0143*

C-Al3+ ions concentration

-14.31

0.0006*

D-Initial concentration of microalgal cells

-230.2

0.0031*

AB

0.32

0.5399

AC

5.94

0.0042*

AD

25.7

0.1553

BC

-1.62

0.0332*

BD

0.25

0.9710

CD

30.24

0.2120

A2

-6.52

< 0.0001*

B2

-0.01

0.9698

C2

-7.29

0.0030*

-40.52

0.8243

D

2

* means significant at 5% level

32

Table 3 Lipid content and LMBRs composition of marine Chlorella sp. biomass obtained from different culture modes LMBRs composition Lipid (% based Culture mode

Protein (% based

Carbohydrate (% based on

on dry LMBRs)

dry LMBRs)

on dry biomass)

Photoautotrophic

32.08±3.88b

38.9±0.05 b

16.6±4.6b

Heterotrophic

30.42±1.13b

44.6±0.03 a

16.7±0.1b

Mixotrophic

42.08±0.58a

40.1±0.15 b

29.2±0.03a

Different superscript letters indicate significant difference between culture modes.

33

Table 4 Percentage fatty acid compositions of lipid extracted from marine Chlorella sp. cultivated in different culture modes Culture mode

C12:0 C14:0 C16:0 C16:1 C17:0 C18:0 C18:1 C18:2 C24:1 USFA SFA

Photoautotrophic

3.18

3.38

44.8

2.43

2.39

6.74

29.7

7.38

0.00

39.5

60.5

Heterotrophic

8.09

5.69

48.9

4.11

6.51

11.5

11.2

1.34

2.57

19.3

80.8

Mixotrophic

0.29

2.82

46.3

2.28

12.8

9.79

13.5

12.23

0.00

28.0

71.9

34

•

Downstream process for lipid-rich marine Chlorella sp. biomass was optimized

•

Microalgal biomass in mixotrophic culture was easier harvested than other cultures

•

Cell debris after lipid extraction contained high protein and minerals

35