Scholarly article on topic 'Baltic Sea microalgae transform cement flue gas into valuable biomass'

Baltic Sea microalgae transform cement flue gas into valuable biomass Academic research paper on "Biological sciences"

CC BY-NC-ND
0
0
Share paper
Academic journal
Algal Research
OECD Field of science
Keywords
{Microalgae / "Baltic Sea" / "Flue gas" / "Biomass composition" / "Natural communities" / Brackish}

Abstract of research paper on Biological sciences, author of scientific article — M. Olofsson, E. Lindehoff, B. Frick, F. Svensson, C. Legrand

Abstract We show high feasibility of using cement industrial flue gas as CO2 source for microalgal cultivation. The toxicity of cement flue gas (12–15% CO2) on algal biomass production and composition (lipids, proteins, carbohydrates) was tested using monocultures (Tetraselmis sp., green algae, Skeletonema marinoi, diatom) and natural brackish communities. The performance of a natural microalgal community dominated by spring diatoms was compared to a highly productive diatom monoculture S. marinoi fed with flue gas or air–CO2 mixture. Flue gas was not toxic to any of the microalgae tested. Instead we show high quality of microalgal biomass (lipids 20–30% DW, proteins 20–28% DW, carbohydrates 15–30% DW) and high production when cultivated with flue gas addition compared to CO2–air. Brackish Baltic Sea microalgal communities performed equally or better in terms of biomass quality and production than documented monocultures of diatom and green algae, often used in algal research and development. Hence, we conclude that microalgae should be included in biological solutions to transform waste into renewable resources in coastal waters.

Academic research paper on topic "Baltic Sea microalgae transform cement flue gas into valuable biomass"

algal

Baltic Sea microalgae transform cement flue gas into valuable biomass

M. Olofsson, E. Lindehoff, B. Frick, F. Svensson, C. Legrand *

Center for Ecology and Evolution in Microbial model Systems (EEMiS), Department of Biology and Environmental Science (BoM), Linnœus University, Kalmar 39182, Sweden

ARTICLE INFO ABSTRACT

We show high feasibility of using cement industrial flue gas as CO2 source for microalgal cultivation. The toxicity of cement flue gas (12-15% CO2) on algal biomass production and composition (lipids, proteins, carbohydrates) was tested using monocultures (Tetraselmis sp., green algae, Skeletonema marinoi, diatom) and natural brackish communities. The performance of a natural microalgal community dominated by spring diatoms was compared to a highly productive diatom monoculture S. marinoi fed with flue gas or air-CO2 mixture. Flue gas was not toxic to any of the microalgae tested. Instead we show high quality of microalgal biomass (lipids 20-30% DW, proteins 20-28% DW, carbohydrates 15-30% DW) and high production when cultivated with flue gas addition compared to CO2-air. Brackish Baltic Sea microalgal communities performed equally or better in terms of biomass quality and production than documented monocultures of diatom and green algae, often used in algal research and development. Hence, we conclude that microalgae should be included in biological solutions to transform waste into renewable resources in coastal waters.

© 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/).

|Wj| CrossMark

Article history:

Received 24 April 2015

Received in revised form 22 June 2015

Accepted 1 July 2015

Available online xxxx

Keywords: Microalgae Baltic Sea Flue gas

Biomass composition Natural communities Brackish

1. Introduction

1.1. Flue gas as CO2 source for microalgae

The cement industry is responsible for approximately 4-5% of global CO2 emissions [1-3]. Using mass cultivation of microalgae is considered environmentally safe and sustainable for the abatement of CO2 from industrial flue gas [4-8]. Recent studies questioned the contribution of algae to global CO2 removal since algae fix CO2 from flue gas but do not offer permanent storage nor are energy efficient [9,10]. However, microalgae biomass can deliver products (biofuels, industrial material etc.) that may replace an equivalent amount of fossil fuels, hence facilitating the sustainability of microalgae-based product development [9]. By 2030 the EC proposes that emissions from sectors covered by emission trade scheme (ETS) will be 43% lower than in 2005 [11]. This proposal, combined with political climate change targets and market forces can provide economic incentive for future company investments in new technology.

Flue gas generally contains 3-15% CO2 (v/v) depending on fuel feedstock and type of operation [12] and thus can be used as a source of CO2 for microalgae cultivation. Microalgae show a good growth potential in CO2 concentrations up to 10-20% regardless of the source, e.g. pure CO2

Abbreviations: Chla, chlorophyll a; DW, dry weight; EC, European Commission; FG, flue gas; KAC, Kalmar Algae Collection; NC, natural community; PBR, photobioreactor; PTFE, polytetrafluoroethylene; Sm, Skeletonema marinoi; SYKE, Finnish Environment Institute; TC, total carbohydrates; TL, total lipids; TP, total proteins; v, volume; w, weight.

* Corresponding author. E-mail address: catherine.legrand@lnu.se (C. Legrand).

[13,14] and industrial flue gas [15-17,6,18,8]. Industrial flue gas contains over 100 substances, of which several are potentially toxic to microalgae (e.g. SOx, NOx, HF, heavy metals) [19]. Numerous studies have weighed opportunities and limitations of microalgal cultivation and based on predictions have showed the potential of a process where industrial waste CO2 is converted to bioproducts through algae [20-24]. Empirical studies on the tolerance of microalgae to industrial flue gas are increasing steadily but rarely include various taxonomic groups of microalgae, and trials with natural or multispecies communities are noticeably lacking.

1.2. Importance of diversity and production for microalgae cultivation

Outdoor mass cultivation of microalgae has generally focused on highly productive monoclonal cultures for biomass production or targeting specific chemicals. Few studies have been using multispecies cultures or natural assemblages of microalgae to improve production yields. Productivity and stability of natural terrestrial ecosystems have been positively linked to diversity and species richness [25-28] but may be applicable to marine habitats [29]. Productivity of agro systems is considered to benefit from intercropping through the establishment of stable and sustainable ecosystems within crop farmlands [30]. Both observational [31] and experimental studies [32-34] indicate that this applies also for microbial communities, including microalgae. The positive relationship between diversity and productivity may be explained by 1) the complementary effect, where resource utilization is higher in a more diverse community [35,29,36] and 2) the selection effect, where one highly productive species, is favored by certain environmental conditions over other species in a diverse community [37]. The two

http: //dx.doi.org/10.1016/j.algal.2015.07.001

2211-9264/© 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

mechanisms may not be complete contrasts but should both be accounted for, while estimating the effect of diversity on productivity [26,37]. The complementary effect in diverse microalgae communities could lead to a more stable and resilient system less prone to invasive species and zooplankton grazing pressure [38]. The same mechanism of decreased pest susceptibility was shown for plant crops [39]. Diverse communities seem to be more productive and resilient in natural variable environments where changing abiotic factors affect productivity.

Additionally, the spectrum of valuable chemicals in terms of lipids, proteins, carbohydrates and pigments produced will most likely be more diverse in a diverse community. Smith et al. [40] suggested that multispecies communities of microalgae in open pond cultivation systems could accumulate more solar energy as lipids due to a more efficient utilization of light from different functional groups of microalgae, in contrast to closed systems with monocultures. Positive effects of species richness (level of two, three and four species) were found on both algal biovolume and lipid content compared to monocultures [41]. These effects were attributed to complementarity rather than selection effects. Stockenreiter et al. [42] found the relationship of functional group richness more strongly linked to lipid content than mere species richness, suggesting that a more efficient light utilization within functionally diverse communities contributed to the higher lipid content. Extrapolating these findings from natural and artificial ecosystems to industrial mass cultivation of microalgae leads to a combination of uncertainties. Nonetheless, the use of natural community as inoculum in large-scale production system could increase the stability, resilience and productivity of the system.

Composition of the flue gas from the cement industry varies with the origin of the raw substrate and the combustion process. For the first time the potential use of flue gas from Cementa AB, Degerhamn, Oland, SE for microalgal biomass production was evaluated using monocultures and natural communities from the Baltic Sea.

This study aims to test the toxicity of cement flue gas on biomass production of a monoculture (Tetraselmis sp., green algae), and to compare biomass composition (lipids, proteins, carbohydrates) and production of a natural microalgal community dominated by spring diatoms to one highly productive diatom monoculture during treatments of flue gas or air-CO2 mixture. The reference monoculture was Skeletonema marinoi (strain SMTV1), a rapidly growing diatom common in the spring bloom community in the Baltic Sea.

2. Material and methods

2.1. Study site

Cementa AB Degerhamn, Oland, southeast Sweden, manufactures 300,000 tons of cement annually and releases approximately 260,000 tons of CO2 through their flue gas. The major components of the flue gas emissions, besides CO2, are SO2, NOx and dust (Table 1). The Cementa HeidelbergCement group is working to reduce the CO2 emissions by 10% by the year 2020 at the plant in Degerhamn, and will strive to achieve 0% by 2050. The possibility of lowering their carbon footprint by using mass cultivation of microalgae to capture the CO2 rich flue gas is currently being evaluated in an academia-industry collaboration. The valorization of the produced algal biomass to generate bulk chemicals, such as lipids, carbohydrates and proteins for conversion to bioenergy or high value products is also assessed.

22. Flue gas and CO2

The cement flue gas was collected from the monitoring sampling point in one of the flue stacks at the Cementa AB Degerhamn factory by using a high-pressure compressor and filling the flue gas into gas cylinders at 150-200 bar. The composition of the flue gas varied 12-15% during the experiments (Table 1). Industrial grade CO2-air mixture (13.5% CO2) was obtained from AGA Gas AB.

Table 1

Composition of the conditioned flue gas collected from Cementa AB, Degerhamn, Sweden for this study. Values are valid from April 2013 to May 2014, metal levels were measured in September 2012.

Source: environmental report HeidelbergCement, Degerhamn 2013.

Temperature 150-200 °C

CO2 12-15%

O2 0-21%

H2O 0-15%

NOx <800 mg/Nm3

SO2 <50 mg/Nm3

CO 0-1000 mg/Nm3

NH3 <50 mg/Nm3

HCl <10 mg/Nm3

HF 0-0.01 mg Nm-

Dust particles <10 mg/Nm3

Metalsa <0.5 mg/Nm3

Mercury (Hg) <0.03 mg/Nm3

Cadmium (Cd) + titanium (Ti) <0.05 mg/Nm3

Metals: Sb + As + Pb + Cr + Co + Cu + Mn + Ni + V.

2.3. Microalgal stock cultures

The green algae Tetraselmis (strain KAC21) and the diatom S. marinoi (strain SMTV1) were grown in filtered Baltic seawater (salinity 7, filtered 0.2 |jm) enriched with original Guillard's f/2 medium and f/2 + Si respectively [43]. Strains were obtained from the Kalmar Algal Collection (KAC) and the Finnish Environment Institute (SYKE). Cultures were grown in 10 L glass flasks, gently bubbled with air at temperature 18 °C (Tetraselmis) and 15 °C (S. marinoi), and at irradiance 300-500 |jmol photons s-1 m-2. Irradiance was measured with a digital scalar irradiance meter (Biospherical Instruments Inc.). A natural microalgal community was sampled during the spring bloom of 2013 in the SW Baltic Sea (PRODIVERSA cruise, station 8,18 April, longitude 17.33342 latitude 56.2559). Diatoms dominated the community, primarily of the genus Chaetoceros spp. dinoflagellates, cryptophytes, euglenophytes and chlorophytes (green algae) were also present in lower abundance.

2.4. Experimental design

2.4.1. Can microalgae use the CO2 in flue gas and how is the quality of the biomass affected?

Tetraselmis was inoculated (5000 cells mL-1) in six cylinders in photobioreactor 1 (PBR1, Fig. 1, Table 2) filled with f/2 Baltic seawater medium. Three replicates were supplied with air (control) and three with cement flue gas daily for 40-120 s at a flow rate of 5 L min-1. Growth performance of Tetraselmis was monitored over 10 days. Samples were taken daily in each replicate cylinder for cell density and pH (days 1-10), and dry weight (DW) from day 3 to 10. Endpoint lipids and inorganic nutrient levels were also measured at day 10. Cells were fixed with Lugol's solution prior to counting. pH was measured with a

Fig. 1. Dry weight (DW) of Tetraselmis sp. during the experiment excluding the initial two-day lag-phase (n = 2).

Table 2

Technical specifications and set-up of the two photobioreactors (PBR) used in the study.

PBR1 PBR2

Dimension (H x d) mm 1000x 100 531.6x117.3

Cylinders 6 12

Material Acrylic Polystyrene

Culture volume L 5.8 4.2

Irradiancea |jmol m-2 s-1 340-550 300-500

Light supply Metal-halide lamps (250 W Osram)

Light:dark 16:8 16:8

Temperature °C 18 15

Air feed (0.2 | m filtered) Perforated pipes Air stone

Air release 20 to 150 mm Air release 0 to 20 mm

from bottom of cylinder from bottom of cylinder

FG/CO2 feed Bottom up through Bottom up through airstone

perforated plug

Daily FG/CO2 distribution Single pulse, 40-120 s Multi pulses (3x), 60 s

Gas flow rate 5 L min- 1 1 L min- 1

a Measured with a digital scalar irradiance meter (Biospherical Instruments Inc.).

compact inoLab Level 1 pH meter. Dry weight was determined by filtering 25 mL onto prewashed (10 mL 0.5 M ammonium formate), dried and preweighed 45 mm glass fiber filters (Whatman GF/C). Maximum specific growth rate was derived from the steepest slope by plotting the natural logarithm of the cell density using at least 3 successive data points (days 1-6). Productivity (mg DW L-1 d-1) was determined using the natural logarithm of DW values during the same time interval. Tetraselmis samples (50 mL) for analysis of total lipids were centrifuged for 20 min at 11,900 xg, at 4 °C (Beckman, Avanti J-25). The supernatant was discarded and the pellet frozen in 50 mL Falcon tubes.

2.4.2. How does the production and chemical composition differ between a monoculture and a natural microalgal community under flue gas treatment?

S. marinoi (Sm) and the natural community (NC) were inoculated (3 |ag Chla L-1) in 3 x 2 x 2 replicate cylinders in photobioreactor 2 (PBR2, Table 2.) filled with f/2 Baltic seawater medium. Triplicates of Sm and NC were repeatedly (3 times during the light period; 08.00, 12.00 and 16.00) sparged with cement flue gas alternatively with industrial grade CO2-air mixture for 1 min at a flow of 1 L min-1. Upon injection of flue gas or industrial grade CO2 pH decreased in the PB2 bottles by 1 to 1.5 units. Recovery of pH to initial values occurred gradually over 3-4 h before the next injection.

Growth performance was monitored over 10 days. Samples were taken daily in each replicate cylinder for Chla concentration, cell abundance and community composition and pH, and DW from days 3 to 10. Endpoint chemical composition (lipids, proteins, carbohydrates) and inorganic nutrient levels were also measured at day 10. Cells were fixed with Lugol's solution prior to counting. pH was measured daily with a H198128 hand pH-meter (Hanna instruments). Dry weight was determined by filtering 25-100 mL onto prewashed (Milli-Q), dried and preweighed 45 mm glass fiber filters (Whatman GF/C). Growth rate was derived from the slope in the growth phase by plotting the natural logarithm of the Chla concentration over days 1-5. Productivity (mg DW L-1 d-1) was calculated using endpoint values of DW over the experimental period (10 days). The theoretical optimum productivity (mg DW L-1 d-1) was calculated using the initial biomass (DW) and the growth rate at maximum yield (day 4). Biomass for analysis of chemical composition was collected by filtration of 1200 mL from each replicate onto a 3 |jm SS Millipore filter and transferred to 50 mL Falcon tubes. The Falcon tubes were centrifuged at 2850 xg (Hettich Universal 16R) and the supernatant was discarded. The pellet was frozen and then vacuum dried using a Labconco bulktray drier and a Scanvac Coolsafe until constant weight. At constant weight, the biomass was divided for analysis of total protein (10 mg) and carbohydrates (10 mg) and the remaining biomass of 96 ± 0.6 mg was used for total lipid analysis.

2.5. Microalgal biomass and composition

Cell density of monoculture Tetraselmis and Sm was measured using a 0.1 mL Palmer Maloney counting chamber. Cell abundance and composition of NC were determined using sedimentation chambers [44]. At least 300 cells were counted in each replicate using an inverted light microscope (Zeiss PrimoVert). Cells were identified to the genus or species level when possible. Cell biovolume was calculated using the method of Olenina et al. [45] and multiplied with cell numbers to obtain the total biovolume (|m3 L-1). Chlorophyll a (Chla) was analyzed fluorometrically (Turner Trilogy fluorometer) on ethanol extracts according to Jespersen & Christoffersen [46]. Depending on concentration, 5-20 mL of microalgal suspension was filtered onto 25 mm Pall AE/E filters prior to extraction in ethanol. Filters for DW were immediately washed after sampling to remove excess salt with 10 mL 0.5 M ammonium-formate and dried in 80 °C (Tetraselmis) or washed with 10 mL MilliQand dried in 100 °C (Sm and NC) until constant weight. Filters were weighed on a Mettler Toledo, College-B B154 scale. Dry weight was calculated by subtracting the weights of the dried filters with biomass and the dried prewashed filters.

2.6. Product analyses

2.6.1. Total lipids

Lipid extractions were performed using a modified Bligh & Dyer method [47]. The lipids were extracted from the algal pellet with Chloroform:MeOH 1:2 (v/v). In order to break cell walls samples were sonicated using a Sonics Vibra-cell (VCX 130) for 5 min at 100% for Tetraselmis and 2 min at 50% for Sm and NC. For Sm and NC the chloroform:MeOH 1:2 extraction procedure was repeated three times by collecting the supernatant in new falcon tubes after centrifugation (10 min, 2850 xg, Hettich Universal 16R). Water and chloroform were then added to the recovered supernatant in the final proportions of chloroform:MeOH:H2O 2:2:1 and the solution was vortexed until homogenous. The solution was then centrifuged (10 min, 2850 xg, Hettich Universal 16R). The aqueous-methanol layer was removed and the remaining chloroform/lipid layer was transferred into pre-weighed glass sampling tubes and dried until constant weight (Mettler Toledo, College-B B154 scale). The empty pre-weighed values were subtracted from the weighed sample values. For the Sm and NC samples the collected supernatant was filtered through a 25 mm syringe filter w/0.2 |m PTFE membrane into pre-weighed glass vials. To remove any traces of water a few drops of MeOH were added and then the samples were evaporated by using a filtered (0.2 |m) stream of compressed air. When the samples appeared dry and the chloroform had completely evaporated the samples were further dried in an oven at 60 °C until constant weight.

2.6.2. Protein analysis

5 mg vacuum-dried biomass was resuspended in 2 mL 1 M NaOH and incubated in a waterbath at 95 °C for 60 min. Samples were then centrifuged (10 min, 2850 xg, Hettich Universal 16R) and 100 |L of the supernatant was transferred to glass tubes. A ready-made assay, Bio-Rad DC Protein assay kit II, using the Folin-Phenol protein quantification method [48] was used to prepare the samples to be read in a spectrophotometer. The sample values were compared to a standard curve with the range of 0.2-1.0 mg mL-1 made from the bovine serum albumin standard included in the assay.

2.63. Carbohydrate analysis

Carbohydrate extraction was carried out using the phenol-sulfuric acid method described by DuBois et al. [49]. The vacuum-dried biomass was dissolved in 5 mL 1 M H2SO4 and incubated in 90 °C for 60 min. After incubation samples were cooled down to room temperature and centrifuged (10 min, 2850 xg, Hettich Universal 16R), 100 |L of the supernatant was added to a glass vial and was let to react with 1 mL

Table 3

Cell density, growth rate, biomass and productivity of Tetraselmis sp. fed with air and flue gas (FG). Slope, confidence interval (CI) and coefficient of determination (R2) of simple regressions of the natural logarithm of cell density were calculated over the experimental period (days). Statistical comparison of slopes for cell density between treatments of Air and FG is represented by p-value. Biomass and productivity are the mean of duplicate values. n.a. not applicable.

Ln cells mL- 1 Growth rate d- 1 CI R2 Day p-Value

Air 0.83 0.67-0.98 0.95 2-6 0.0283

FG 1.04 0.91-1.18 0.98 1-5

Biomass Productivity mg L- 1 d- 1 Range

Air 34.7 31.3-38.0 n.a. 3-9 n.a.

FG 57.3 50.7-64.0 n.a. 3-6 n.a.

of Phenol (5% w/v) and 3 mL of H2SO4 (72 wt.%). A glucose standard curve using Sigma D-( + )-glucose with the range of 0.2-1.0 mg mL-1 was prepared. Samples were incubated in a waterbath at 90 °C for 5 min. Absorbance was measured using a VWR UV-1600 PC spectrophotometer at 490 nm and concentrations were determined by comparing to the standard curve.

2.7. Statistics

Statistical analyses (2-way ANOVAs, paired t-test) followed by Tukey's post-hoc test were performed using Graph Pad prism (version 6.0d for Mac OS X, GraphPad Software, La Jolla California USA). The level of significance was determined to p < 0.05. Multiplicity adjusted p values, accounting for multiple comparisons, are reported. In the experiment with Tetraselmis, biofouling occurred in one replicate per treatment, hence simple regression based on the natural logarithm during linear growth of the exponential phase was performed from two replicates. The slopes of the regressions were then compared statistically between the flue gas and air treatment using Graph Pad prism.

3. Results

3.1. Utilization ofCO2 in flue gas and quality of algal biomass

The FG composition used is shown in Table 1. The main compounds of the FG were CO, NOx and SOx, in addition to CO2. In addition the FG also contained NH3 and various metals that can be used for algal growth, while the presence of dust particles, mercury and cadmium can have a negative effect on algal metabolism. Experiment performed in batch cultures demonstrated that FG can be used as the CO2 source for Tetraselmis increasing growth performance (Fig. 1). The growth of Tetraselmis sp. was significantly higher for FG (1.04 d-1) compared to air control (0.83 d-1) (F = 5.815, p = 0.028, Table 3). Maximum yield (200 mg L- 1) was attained on day 6 with FG compared to day 9 in air control cultures (Fig. 1). Consequently, the highest productivity was achieved in FG treatment (57.3 ± 6.7 mgL-1 d-1) compared to air control (34.7 ± 3.3 mg L-1 d-1), corresponding to a 1.6-fold increase (Table 3).

Addition of FG did not affect the total lipid levels in Tetraselmis sp. regardless of treatment (air: 10.9 ± 5.8%, FG: 10.1 ± 6.8% of DW). No significant differences were found among the treatments (one-way ANOVA, F = 0.433, p = 0.661). The experiment was repeated using reference cultures bubbled with CO2, in which the lipid levels were similar as for the air and FG treatments (data not shown).

3.2. Production and chemical composition of monoculture and a natural microalgal community fed with flue gas

Productivity was significantly higher in NC (~6.5 mg L-1 d-1) compared to Sm (~4.5 mg L-1 d-1) regardless of treatment (2-way ANOVA, community: F = 7.351, p = 0.027, treatment: F = 0.057, p = 0.818) (Table 4). Yield in DW after 10 days was similar for Sm and NC regardless of treatment (2-way ANOVA, community: F = 0.183, p = 0.680, treatment: F = 0.057, p = 0.818). On the other hand, growth rates based on Chla were higher in Sm monoculture (~1 d-1) compared to the NC (~0.8 d-1) (2-way ANOVA, community: F = 57.860, p < 0.0001, treatment: F = 0.028, p = 0.871, Tukey's post-hoc test) (Table 4, Fig. 2). Cell density was twice as high in Sm cultures compared to NC (2-way ANOVA, community: F = 64.100, p < 0.0001, treatment: F = 0.062, p = 0.810). Hence, biovolumes were marginally higher in Sm (biovolumes, 2-way ANOVA, community: F = 6.892, p = 0.030, treatment: F = 0.361, p = 0.564) (Table 4). The ratio between biovolume and cell density showed larger cells in the NC (3.0:1) compared to Sm (1.4:1). This was confirmed by the species composition in NC, dominated by diatoms (90% of the total biovolume) in both treatments, mostly Chaetoceros wighamii (67% of the total biovolume in both treatments) (Fig. 3). All together the genera Chaetoceros and Tabellaria consist of cells 2-10 times larger than Skeletonema cells. Smaller cells (the diatom Cylindrotheca and Skeletonema, the genus Euglena and flagellates) only made up for 10% of the biomass. No significant difference was found between treatments (paired t-test, p > 0.99).

Total lipid (TL) ranged 16-28% DW in Sm and 24-27% DW in NC with no significant difference (Fig. 4a). Total proteins (TP) content of Sm and NC were similar ranging 21-28% DW (Fig. 4b). Total carbohydrates (TC) were higher in Sm (30% DW) compared to NC (20% DW) with a significant difference for FG treatment (2-way ANOVA, community: F = 21.67, p = 0.01, Tukey's post-hoc test) (Fig. 4c).

4. Discussion

4.1. Potential for using flue gas for algal cultivation

Marine and brackish microalgal production provides a sustainable mitigation biofiltration method for the removal of greenhouse gases such as CO2 and various pollutants, nutrients and metals. Due to the composition of the flue gas released at the Cementa AB plant in Degerhamn, SE we needed to test the tolerance limits and biomass productivity of selected microalgal strains and natural communities.

The results reported here demonstrate that flue gas from the Degerhamn plant can be used to produce algae in brackish waters in the Baltic Sea region. Productivity of the Baltic Sea microalgae doubled with flue gas indicating a good source of CO2 and no toxicity to

Table 4

Dry weight, productivity, maximum growth rate, cell density, biovolume and theoretical maximum production ± SD for Skeletonema marinoi (Sm) and a natural microalgae community (NC) fed with CO2 and flue gas (FG). Different letters indicate significant differences for a given parameter.

Treatment

Dry weight (mgL-1)

Productivity (mgL-1 d-1)

Growth rate (d-1) 1-5

Cell density (109 cells L-1)

Biovolume (mm3 L-1 )

Theoretical maximum production (mg L-1 d-1)_

Sm CO2 Sm FG NC CO2 NC FG

77 ± 0.02 81 ± 0.01 77 ± 0.01 75 ± 0.01

4.27 ± 1.70a 4.73 ± 0.61a 6.47 ± 0.99b 6.33 ± 1.29b

1.05 ± 0.04a 1.03 ± 0.04a 0.87 ± 0.12b 0.88 ± 0.02b

3.35 ± 0.88a 3.33 ± 0.37a 1.16±0.05b 1.04 ± 0.12b

462 ± 122a 460 ± 52a 377 ± 23b 329 ± 43b

306 ± 32 291 ± 29 71 ±25 70 ±4

Fig. 2. Natural logarithm of chlorophyll a (Chla) values for a) Skeletonema marinoi (Sm) and b) the natural microalgae community (NC) during the experiment. Trend line marks the days used for calculation of growth rate (n = 3 ± SD).

microalgae. The productivity of microalgae depends heavily on the conditioning of the raw flue gas due to the substantial levels of toxic compounds [19]. We demonstrate that the conditioning of flue gas at Cementa AB, Degerhamn, removing NOx, SOx, dioxins, and particles is satisfactory for microalgal cultivation. Another challenge when using flue gas is the decreased pH due to CO2 and other compounds. The use of brackish seawater in our study provides a natural buffer capacity compared to freshwater and this is particularly relevant on an island where freshwater valuable resources are scarce.

Productivity of microalgae is related to the availability of CO2 in the growth environment and the 12-15% CO2 in the cement flue gas in this study fits the requirements for growth and the CO2 tolerance of brackish microalgae. Green algae and diatoms dominate the natural microalgal community in the Baltic Sea during spring and early summer [50,51]. Both in marine and freshwater, these two algal groups have a broad CO2 tolerance (14-40%) while optimal CO2 for growth is 10-15% [52,53].

Productivity performance of brackish microalgae monocultures and natural communities in this study was in the lower range compared to reported values [54,55]. Worth mentioning is that the primary intention of the present study was not to optimize productivity but provide a comparison of microalgae growth with flue gas. Weis et al. [34] demonstrated that under homogenous conditions a monoculture would perform better than a diverse community. Our results confirm this as maximum growth rate, hence theoretical maximum productivity, achieved in the monocultures exceeded the ones of natural communities four times. Increased species richness can be linked to increased resource use, nutrient recycling and productivity. With increased diversity the cultures can be more adaptable and resilient in a spatial heterogeneous environment [29], which could explain the higher productivity of the natural community at maximum yield through complementary effects [26,41]. Since flue gas had no stress effect on the microalgal

community composition, this finding is encouraging for the use of natural communities in view of a stable production in large-scale cultivation systems.

When evaluating the potential of using flue gas for algal cultivation and product development, it is essential to determine the quality of the microalgal biomass in terms of lipids, proteins and carbohydrates. In this study, lipid, protein and carbohydrate contents in green algae and diatoms were not affected by the cement flue gas (12-15% CO2) compared to standard cultivation methods, using aeration or pure CO2. A moderate increase in CO2 (5-20%) can stimulate the accumulation of lipids [56], while higher CO2 concentrations (>20%) can inhibit growth and decrease lipid content [57,7]. Similar results have been reported for green algae (e.g. Tetraselmis suecica, Chlorella sp.) grown with untreated flue gas from coal fire power plant or industrial CO2 [58]. Furthermore, the diatom Chaetoceros muelleri showed highest yield and lipid accumulation at 10% CO2 [59]. The fatty acid composition in total lipids of Chlorella sorokiniana changed drastically when cultured with flue gas (~10% CO2) instead of pure CO2 [60], suggesting that biomass quality should be investigated beyond bulk measurements. The large variability of reported values of biomass quality between algal groups, species and isolates renders it difficult to identify trends in respect to flue gas effect. In addition, there is a predominant focus on biomass oil content for algal biofuel research, hence neglecting other valuable products.

4.2. Valuable biomass from Baltic Sea microalgae communities

Most of the research in biological solutions using algal biomass is based on the use of monocultures that ensures repeatability, which is paramount in food industry, pharmaceuticals, cosmetics, and biofuel production. Increasing productivity, yield and valuable product content often relies on screening for the optimal strains that should be resilient

100 80

m 40 20 0

sbh Flagellates <10 |jm 1=1 Flagellates 10-20 |m Euglena spp. Tabellaria spp. un Cyiindrotheca spp. tzzi Chaetoceros spp.

С. wighamii 1= Skeletonema spp.

Fig. 3. The contribution (%) of algal taxonomic groups (flagellates) and species to the total biovolume of natural microalgal communities fed with CO2 and flue gas (FG).

Fig. 4. Chemical composition (% DW) of Skeletonema marinoi (Sm) and natural microalgal communities (NC) fed with CO2 and cement flue gas (FG); a) total lipids (TL),b) total proteins (TP) and c) total carbohydrates (TC) (mean, n = 3 ± SD), 'indicates a significant difference between treatments p < 0.05. Two replicates and a dummy value were used to calculate average values in CO2-TP and FG-TC.

to different culture and environmental conditions, contamination and prédation. Ironically, algal monocultures often provide high quality food for predators and pathogens, and are susceptible to contamination by other microbes. Our results show that natural brackish microalgal communities dominated by diatoms are of comparable quality in biochemical composition as other Baltic Sea diatom monocultures. The natural community in this study, dominated by C. wighamii contained high levels of lipids (16-28%) similar to other cold water species; S. marinoi (5-25%, this study), Skeletonema costatum (6-16%) and C. wighamii (3-27%) [61]. Protein levels (20-30%, this study) were also in the range of reported values for other Skeletonema strains [62,63]. Carbohydrates are the main organic compounds produced by algae photosynthesis and accumulate when cells enter stationary growth phase. Our results showed a higher amount of carbohydrates in S. marinoi monocultures compared to S. costatum (5-7%, [62]) or the natural community (20%, this study). This reflected that the monocultures entered stationary phase earlier than the natural community,

due to higher growth rate. However, there was no accumulation of lipids in S. costatum, as a consequence of this stress, as reported in the literature [64,16]. Comparable results to ours are reported in Fig. 3 from Bertozzini et al. [63], regardless of nitrogen stress, carbohydrates were also stored rather than lipids in S. marinoi. Thus, in our study, since nutrients were still available in stationary phase it is likely that light was limiting growth (self shading).

The brackish Baltic Sea algal community performed at the same level as documented species used in algal cultivation systems in terms of biochemical composition. The combination of a high quality biomass and higher productivity at maximum yield emphasizes the potential of using natural communities in algal production.

It remains unclear how flue gas or high CO2 concentrations influence other nutritional values, such as fatty acids, vitamins and pigments. Toxic components in algal biomass, possibly originating from flue gas or recycled nutrients from waste streams, need to be carefully screened in the valorization process.

5. Conclusion

The cement flue gas was not toxic to the microalgae tested, and proved to be a suitable CO2 source with increased productivity relative to air, and comparable quality of biomass composition between flue gas and CO2. The natural brackish microalgal communities dominated by diatoms show high quality of biomass, and higher productivity than other Baltic Sea diatom monocultures at maximum yield. Hence, we demonstrate the potential of using naturally occurring microalgal communities in outdoor large scale algal cultivation systems. It remains to be proven if this applies to other algal communities such as green algae, cyanobacteria, and flagellates along the seasonal succession. For a successful and sustainable large-scale algal biomass production, locally (Baltic Sea) adapted species are sought for, as they are able to cope not only with environmental conditions (temperature, eutrophication, large pH variation), but also with the constraints coupled to the cultivation process (CO2 fluctuations, contamination and pathogens). Therefore, using the natural succession over the year with seasonally adapted species could aid in the stability and sus-tainability of algal cultivation. Hence, microalgae should be included in biological solutions to transform waste into renewable resources in coastal waters.

Acknowledgements

We would like to thank Natalie Henriksson, Jesper Bjork Rengbrandt, and Emmelie Nilsson for laboratory assistance, Lars Malmer for crafting PBR1, Anke Kremp (Finnish Environmental Institute) for providing S. marinoi strain, Carina Bunse (Lnu) for NC sampling, Claes Kollberg and his staff at Cementa AB, Degerhamn, Caroline Littlefield Karlsson for proof reading the manuscript, and Stefan Sandelin (Cementa HeidelbergCement, Sweden), Leif Norlander (SMA Mineral), Max Larsson (Af Group), Kjarstin Hagman Bostrom (Lnu), and Erik Wennerhag (The Swedish Agency for Economic and Regional Growth, Tillvaxtverket) for the support during this project. Funding was provided by the Stiftelsen for kunskaps- och kompetensutveckling (KK- Stiftelsen, project 20120194), the European Regional Development Fund (project 170405), The Regional Council in Kalmar county, the Swedish Research Council FORMAS (Ecochange project 2009-149), Linnsus University, Faculty of Health and Life Science (grant to CL), Aforsk Forskningstiftelsen (project 11-385), and industry partners Cementa HeidelbergCement and SMA Mineral.

References

[1] E. Worrell, L. Price, N. Martin, C. Hendriks, L.O. Meida, Carbon dioxide emissions from the global cement industry, Annu. Rev. Energy Environ. 26 (2001) 303-329.

[4 [5 [6

B. Metz, O. Davidson, P. Bosch, R. Dave, L. Meyer, Climate change 2007—mitigation of [34 climate change, Intergovernmental Panel on Climate Change, Geneva, Working Group III, 2007.

T.A. Boden, G. Marland, R.J. Andres, Global, regional, and national fossil-fuel CO2 [35 emissions, Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, TN, 2011.

M. Huntley, D. Redalje, CO2 mitigation and renewable oil from photosynthetic mi- [36 crobes: a new appraisal, Mitig. Adapt. Strateg. Glob. Chang. 12 (4) (2007) 573-608.

B. Wang, Y.Q. Li, N. Wu, C.Q. Lan, CO2 bio-mitigation using microalgae, Appl. Microbiol. Biotechnol. 79 (5) (2008) 707-718. [37 I. Douskova, J. Doucha, K. Livansky,J. Machat, P. Novak D. Umysova, V. Zachleder, M. Vitova, Simultaneous flue gas bioremediation and reduction of microalgal biomass [38 production costs, Appl. Microbiol. Biotechnol. 82 (1) (2009) 179-185.

C. Yoo, S.Y. Jun, J.Y. Lee, C.Y. Ahn, H.M. Oh, Selection of microalgae for lipid produc- [39 tion under high levels carbon dioxide, Bioresour. Technol. 101 (2010) S71-S74.

K. Kumar, C.N. Dasgupta, B. Nayak, P. Lindblad, D. Das, Development of suitable photobioreactors for CO2 sequestration addressing global warming using green [40 algae and cyanobacteria, Bioresour. Technol. 102 (8) (2011) 4945-4953. F.G.A. Fernandez, C.V. Gonzalez-Lopez, J.M.F. Sevilla, E.M. Grima, Conversion of CO2 [41 into biomass by microalgae: how realistic a contribution may it be to significant CO2 removal? Appl. Microbiol. Biotechnol. 96 (3) (2012) 577-586.

S. Grierson, V. Strezov, J. Bengtsson, Life cycle assessment of a microalgae biomass [42 cultivation, bio-oil extraction and pyrolysis processing regime, Algal Res. 2 (3) (2013) 299-311.

EU, The EU Emissions Trading System (EUETS) Factsheet, 2013. (doi: 102834/55480). [43 M. Packer, Algal capture of carbon dioxide; biomass generation as a tool for greenhouse gas mitigation with reference to New Zealand energy strategy and policy, Energ Policy 37 (9) (2009) 3428-3437. [44

E. Ono, J.L. Cuello, Selection of optimal microalgae species for CO2 sequestration. Second National Conference on Carbon Sequestration, Proceeding on the 2nd Annual [45 Conference on Carbon Sequestration, USA, 2003.

Y.L. Jiang, W. Zhang, J.F. Wang, Y. Chen, S.H. Shen, T.Z. Liu, Utilization of simulated flue gas for cultivation of Scenedesmus dimorphus, Bioresour. Technol. 128 (2013) 359-364. [46

A. Hamasaki, N. Shioji, Y. Ikuta, Y. Hukuda, T. Makita, K. Hirayama, H. Matuzaki, T. Tukamoto, S. Sasaki, Carbon-dioxide fixation by microalgal photosynthesis using ac- [47 tual flue-gas from a power-plant, Appl. Biochem Biotechnol. 45-6 (1994) 799-809.

L.M. Brown, Uptake of carbon dioxide from flue gas by microalgae, Energy Convers. [48 Manag. 37 (6-8) (1996) 1363-1367.

J. Doucha, F. Straka, K. Livansky, Utilization of flue gas for cultivation of microalgae [49

(Chlorella spp.) in an outdoor open thin-layer photobioreactor, J. Appl. Phycol. 17

(5) (2005) 403-412. [50

C.G. Borkenstein, J. Knoblechner, H. Fruhwirth, M. Schagerl, Cultivation of Chlorella emersonii with flue gas derived from a cement plant, J. Appl. Phycol. 23 (1) (2011) 131-135. [51 S. Van Den Hende, H. Vervaeren, H. Boon, Flue gas compounds and microalgae: (bio-) chemical interactions leading to biotechnological opportunities, Biotechnol. Adv. 30 (2012) 1405-1424.

J.R. Benemann, W.J. Oswald, Systems and economic analysis of microalgae ponds for [52 conversion of CO2 to biomass, Final Report. US DOE-NETL No: DOE/PC/93204-T5, Prepared for the Energy Technology Center, Pittsburgh, USA, 1996.

J.R. Benemann, CO2 mitigation with microalgae systems, Energy Convers. Manag. 38 [53 (1997) S475-S479.

K.L. Kadam, Environmental implications of power generation via coal-microalgae [54 cofiring, Energy 27 (10) (2002) 905-922.

R. Sayre, Algal biofuels, a systems approach, In Vitro Dev. Biol. Anim. 46 (2010) [55 S67-S68.

P.J. McGinn, K.E. Dickinson, S. Bhatti, J.C. Frigon, S.R. Guiot, S.J.B. O'Leary, Integration of microalgae cultivation with industrial waste remediation for biofuel and [56 bioenergy production: opportunities and limitations, Photosynth. Res. 109 (1-3) (2011) 231-247. [57

D. Tilman, P.B. Reich, J. Knops, D. Wedin, T. Mielke, C. Lehman, Diversity and productivity in a long-term grassland experiment, Science 294 (5543) (2001) 843-845.

M. Loreau, A. Hector, Partitioning selection and complementarity in biodiversity [58 experiments, Nature 412 (6842) (2001) 72-76.

C.Y. Li, X.H. He, S.S. Zhu, H.P. Zhou, Y.Y. Wang, Y. Li, J. Yang, J.X. Fan, J.C. Yang, G.B. Wang, Y.F. Long, J.Y. Xu, Y.S. Tang, G.H. Zhao, J.R. Yang, L. Liu, Y. Sun, Y. Xie, H.N. [59 Wang, Y.Y. Zhu, Crop diversity for yield increase, PLoS ONE 4(11) (2009).

Y. Zhang, H.Y.H. Chen, P.B. Reich, Forest productivity increases with evenness, species richness and trait variation: a global meta-analysis, J. Ecol. 100 (3) (2012) [60 742-749.

B. Worm, E.B. Barbier, N. Beaumont, J.E. Duffy, C. Folke, B.S. Halpern, J.B.C. Jackson, [61 H.K. Lotze, F. Micheli, S.R. Palumbi, E. Sala, K.A. Seloke, J.J. Stachiwicz, R. Ward, Impacts of biodiversity loss on ocean ecosystem services, Nature 314 (2006) 787-790.

D. Tilman, J. Hill, C. Lehman, Carbon-negative biofuels from low-input high-diversity [62 grassland biomass, Science 314 (5805) (2006) 1598-1600.

D.R. Ptacnik, A.G. Solimini, T. Andersen, T. Tamminen, P. Brettum, L. Lepisto, E. [63 Willen, S. Rekolainen, Diversity predicts stability and resource use efficiency in natural phytoplankton communities, Proc. Natl. Acad. Sci. U. S. A. 105 (13) (2008) 5134-5138. [64

J. McGrady-Steed, P.M. Harris, P.J. Morin, Biodiversity regulates ecosystem predictability, Nature 390 (6656) (1997) 162-165.

T. Bell, J.A. Newman, B.W. Silverman, S.L. Turner, A.K. Lilley, The contribution of species richness and composition to bacterial services, Nature 436 (7054) (2005) 1157-1160.

J.J. Weis, D.S. Madrigal, B.J. Cardinale, Effects of algal diversity on the production of biomass in homogeneous and heterogeneous nutrient environments: a microcosm experiment, PLoS ONE 3 (7) (2008).

B.J. Cardinale, D.S. Srivastava, J.E. Duffy, J.P. Wright, A.L. Downing, M. Sankaran, C. Jouseau, Effects of biodiversity on the functioning of trophic groups and ecosystems, Nature 443 (7114) (2006) 989-992.

K. Gross, B.J. Cardinale, Does species richness drive community production or vice versa? Reconciling historical and contemporary paradigms in competitive communities, Am. Nat. 170 (2) (2007) 207-220.

J.W. Fox, Interpreting the 'selection effect' of biodiversity on ecosystem function, Ecol. Lett. 8 (8) (2005) 846-856.

A.A. Corcoran, W.J. Boeing, Biodiversity increases the productivity and stability of phytoplankton communities, PLoS ONE 7 (11) (2012).

S. Satpathy, D. Mishra, Use of intercrops and antifeedants for management of eggplant shoot and fruit borer Leucinodes orbonalis (Lepidoptera: Pyralidae), Int. J. Trop. Insect Sci. 31 (1-2) (2011) 52-58.

V.H. Smith, B.S.M. Sturm, F.J. de Noyelles, S.A. Billings, The ecology of algal biodiesel production, Trends Ecol. Evol. 25 (5) (2010) 301-309.

M. Stockenreiter, A.K. Graber, F. Haupt, H. Stibor, The effect of species diversity on lipid production by micro-algal communities, J. Appl. Phycol. 24 (1) (2012) 45-54.

M. Stockenreiter, F. Haupt, A.K. Graber, J. Seppala, K. Spilling, T. Tamminen, H. Stibor, Functional group richness: implications of biodiversity for light use and lipid yield in microalgae, J. Phycol. 49 (5) (2013) 838-847.

R.L. Guillard, Culture of phytoplankton for feeding marine invertebrates, in: W. Smith, M. Chanley (Eds.), Culture of Marine Invertebrate Animals, Springer, USA 1975, pp. 29-60.

H. Utermöhl, Zur Vervollkommnung der quantitativen Phytoplankton-Methodik, Mitt. Int. Verein. Limnol. 9 (1958) 1-38.

I. Olenina, S. Hajdu, L. Edler, A Andersson, N. Wasmund, S. Busch, J. Göbel, S. Gromisz, S. Huseby, M. Huttunen, A Jaanus, P. Kokkonen, I. Ledaine, E. Niemkiewicz, Biovolumes and size-classes of phytoplankton in the Baltic Sea. HELCOM, Balt. Sea Environ. Proc. (No. 166) (2006).

A.M. Jespersen, K. Christofferson, Measurements of chlorophyll—a from phytoplankton using ethanol as extraction solvent, Arch. Hydrobiol. 109 (3) (1987) 445-454.

E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification, Can. J. Biochem. Physiol. 37 (8) (1959) 911-917.

O.H. Lowry, N.J. Rosenbrough, A.L. Farr, R.J. Randall, Protein measurements with folin phenol reagent, J. Biol. Chem. 193 (1951) 265-275.

M. Dubois, K.A. Gilles, J.K. Hamilton, P.A. Rebers, F. Smith, Colorimetric method for determination of sugars and related substances, Anal. Chem. 28 (1956) 350-356. S. Suikkanen, S. Pulina, J. Engstrom-Ost, M. Lehtiniemi, S. Lehtinen, A. Brutemark, Climate change and eutrophication induced shifts in northern summer plankton communities, PLoS ONE 8 (6) (2013).

C. Legrand, E. Fridolfsson, M. Bertos-Fortis, E. Lindehoff, P. Larsson, J. Pinhassi, A. Andersson, Interannual variability of phyto-bacterioplankton biomass and production in coastal and offshore waters of the Baltic Sea, AMBIO (2015), http://dx.doi. org/10.1007/s13280-015-0662-8.

Y. Ishida, N. Hiragushi, H. Kitaguchi, A. Mitsutani, S. Nagai, M. Yoshimura, A highly CO2-tolerant diatom, Thalassiosira weissflogii H1, enriched from coastal sea, and its fatty acid composition, Fish. Sci. 66 (2000) 655-659.

F.M. Salih, Microalgae tolerance to high concentrations of carbon dioxide: a review, J. Environ. Prot. 2 (2011) 648-654.

M.R. Tredici, Photobiology of microalgae mass cultures: understanding the tools for the next green revolution, Biofuels 1 (1) (2010) 143-162. M. Olofsson, T. Lamela, E. Nilsson, J.P. Berge, V. del Pino, P. Uronen, C. Legrand, Combined effects of nitrogen concentration and seasonal changes on the production of lipids in Nannochloropsis oculata, Mar. Drugs 12 (4) (2014) 1891-1910.

C. Yoo, G.G. Choi, S.C. Kim, H.M. Oh, Ettlia sp. YC001 showing high growth rate and lipid content under high CO2, Bioresour. Technol. 127 (2013) 482-488.

S.Y. Chiu, C.Y. Kao, M.T. Tsai, S.C. Ong, C.H. Chen, C.S. Lin, Lipid accumulation and CO2 utilization of Nannochloropsis oculata in response to CO2 aeration, Bioresour. Technol. 100 (2) (2009) 833-838.

N.R. Moheimani, Inorganic carbon and pH effect on growth and lipid productivity of Tetraselmis suecica and Chlorella sp. (Chlorophyta) grown outdoors in bag photobioreactors, J. Appl. Phycol. 25 (2013) 387-398.

X.W. Wang, J.R. Liang, C.S. Luo, C.P. Chen, Y.H. Gao, Biomass, total lipid production, and fatty acid composition of the marine diatom Chaetoceros muelleri in response to different CO2 levels, Bioresour. Technol. 161 (2014) 124-130. K. Kumar, D. Banerjee, D. Das, Carbon dioxide sequestration from industrial flue gas by Chlorella sorokiniana, Bioresour. Technol. 152 (2014) 225-233.

D. Schwenk, J. Seppala, K. Spilling, A. Virkki, T. Tamminen, K.M. Oksman-Caldentey, H. Rischer, Lipid content in 19 brackish and marine microalgae: influence of growth phase, salinity and temperature, Aquat. Ecol. 47 (4) (2013) 415-424.

M.R. Brown, S.W. Jeffrey, The amino acid and gross composition of marine diatoms potentially useful for mariculture, J. Appl. Phycol. 7 (6) (1995) 521-527.

E. Bertozzini, L. Galluzzi, F. Ricci, A. Penna, M. Magnani, Neutral lipid content and biomass production in Skeletonema marinoi (Bacillariophyceae) culture in response to nitrate limitation, Appl. Biochem. Biotechnol. 170 (7) (2013) 1624-1636.

G.A. Dunstan, J.K. Volkman, S.M. Barrett, C.D. Garland, Changes in the lipid-composition and maximization of the polyunsaturated fatty-acid content of 3 microalgae grown in mass-culture, J. Appl. Phycol. 5 (1) (1993) 71-83.