Evaluation of internally illuminated photobioreactor for improving energy ratio

Journal of Bioscience and Bioengineering VOL. 117 No. 1, 92e98, 2014 www.elsevier.com/locate/jbiosc

Evaluation of internally illuminated photobioreactor for improving energy ratio Ambica Koushik Pegallapati, Nagamany Nirmalakhandan,* Barry Dungan, F. Omar Holguin, and Tanner Schaub Civil Engineering Department, New Mexico State University, Las Cruces, NM 88003, USA Received 22 April 2013; accepted 20 June 2013 Available online 6 August 2013

The internally illuminated photobioreactor (IIPBR) design has been shown to be more efficient in utilizing the incident light energy than the externally illuminated designs. This study evaluated (i) optimal sparging of the IIPBR with CO2enriched air (CEA) to enhance biomass productivity; and, (ii) single-stage and two-stage operation of the IIPBR to enhance lipid productivity. Growth data from two algal cultures-Scenedesmus sp. and Nannochloropsis salina, cultivated in an 18-L prototype version of the IIPBR were used to establish the optimal conditions for the two goals in terms of the energy ratio. Based on the optimized results under sparging with CEA, the energy ratio in the IIPBR in the first stage with Nannochloropsis salina was at least 6 times higher due to optimal performance of the IIPBR at lower energy input than typical literature results for other PBR designs, whereas the energy ratios in the second stage were comparable to literature results. Ó 2013, The Society for Biotechnology, Japan. All rights reserved. [Key words: Internally illuminated photobioreactor; Net energy ratio; Lipid content; Fatty acid methyl esters content; Two stage operation]

Compared to traditional fuel crops, microalgae have been identified as a more viable feedstock for biodiesel production due to their higher oil content, faster growth rates, and lower land requirements for cultivation. Besides its potency as a renewable fuel, combustion of algal biodiesel results in lower net carbon emissions than petroleum-based fuels and soybean-derived biodiesel (1). Despite such advantages, microalgal biodiesel is not commercially viable yet (2,3) due to poor photosynthetic efficiency and low oil yields. It has been recommended that energy-efficient photobioreactors with high light harvesting capability coupled with algal strains capable of higher lipid accumulation and faster growth rate under optimized supply of carbon/nutrients can render algal biodiesel production economically feasible (4). Traditionally, performance of PBRs has been assessed and compared on the basis of volumetric or areal biomass productivity, with little regard to the energy associated with the process. Recent studies (5e7) have begun to consider the energy input to the process and the energy that can be harvested from the biomass in assessing and optimizing PBRs for biofuel production. These studies have suggested energetic measures such as biomass productivity per unit energy input, net energy ratio, and net energy gain as more appropriate ones to evaluate PBRs and to assess the returns from energy intensive cultivation practices (such as CO2-enrichment, nutrient enrichment and starvation) that have been suggested to maximize biomass and oil production. In this study, the performance of an internally illuminated photobioreactor (IIPBR) is optimized in terms of the energy ratio. The design features and advantages of this IIPBR, which can serve as * Corresponding author. Tel.: þ1 (575) 646 5378; fax: þ1 (575) 646 6049. E-mail address: [email protected] (N. Nirmalakhandan).

an energy-efficient parent reactor for seeding mass scale systems, have been presented earlier (8). The energy ratio that we propose to adapt in this study is calculated as the ratio of the energy output to the energy input. Energy input to the cultivation process includes energy expended for illuminating the cultures and that for mixing the cultures and providing the CO2 supply. In artificially illuminated PBRs, the former is significantly higher than the latter. For a given incident light energy and a given algal species, efficiency of conversion of light energy to biomass is a function of the reactor geometry; higher conversions could be achieved with higher incident area per unit culture volume and shorter light path length. For example, for a given incident area per unit culture volume, the IIPBR geometry has been shown to have a smaller footprint, shorter light path length, higher biomass density, and higher biomass productivity per unit energy input than the traditional externally illuminated bubble column PBR (BCPBR) (8). From the perspective of algal biodiesel production, the energy that can be harvested from the biomass can be quantified in terms of lipid content or, more specifically in terms of fatty acid methyl esters (FAME) content. Microalgal strains are known to accumulate lipids [mostly triacylglycerols (TAGs), which are saturated and mono-unsaturated fatty acids] when subjected to stressing conditions such as nutrient limitation or altered growth conditions such as fluctuations in light intensity or temperature (9e11). However, such stressing conditions are known to retard biomass growth rate. Since lipid (or FAME) productivity is equal to the lipid content (or FAME content) times biomass productivity, maximizing lipid (or FAME) productivity has remained a challenge. To achieve high biomass productivity and high lipid content, two-stage cultivation has been proposed, where growth-stimulating conditions are

1389-1723/$ e see front matter Ó 2013, The Society for Biotechnology, Japan. All rights reserved. http://dx.doi.org/10.1016/j.jbiosc.2013.06.020

VOL. 117, 2014

EVALUATION OF INTERNALLY ILLUMINATED PHOTOBIOREACTOR

maintained in the first stage followed by stressing conditions in the second stage to raise the lipid content (12). For example, Su et al. (12) had evaluated two-stage cultivation of Nannochloropsis oculata, with the following stressing approaches in the second stage to maximize lipid content: nitrogen limitation, varied inoculum concentrations, salinity, and irradiance. The maximum lipid content achieved in the second stage was 2.8 times that in the first stage at an inoculum concentration of 2.3 g L1, salinity of 35 g L1, and illumination of 500 m Einsteins m2 s1 (12). Solovchenko et al. (10) had investigated accumulation of fatty acid content under nutrient stressing at three light intensities of 35, 200 and 400 m Einsteins m2 s1. The total fatty acid content (TFA) of the cultures grown under 400 m Einsteins m2 s1 and sufficient nitrogen was found to be higher than those under nitrogen starved cultures at lower light intensities (10). Under environmental stresses, microalgae have the ability to modify lipid metabolism allowing them to endure extreme conditions (13e16). During unfavorable nutrient limitation such as nitrogen deprivation, microalgae modify their metabolism to support the synthesis of lipid bodies, which are largely comprised of triacylglycerols (15,17,18). Both lipid production and algal biomass compete for photosynthetic assimilate (15). While similar studies had evaluated the effect of nutrient-stress on algal growth and lipid accumulation, not many have studied the feasibility of accumulation of lipid and fatty acids and compositional changes under carbon-limitation in the second stage. We evaluated the energy ratio during lipid productivity under CO2 limitation reducing de novo synthesis of fatty acids and assessing the lipid accumulation based on the organism’s ability to undergo lipid remodeling of its resources. The ultimate goal of this study was to improve the energy ratio in the IIPBR by evaluating the following two premises: (i) biomass productivity could be maximized by optimizing the carbon supply; and (ii) lipid content of biomass could be improved through stressing the cultures under nutrient starvation. To validate the first premise, growth of two algal species in the IIPBR under sparging at various CO2-air ratios was evaluated. Studies in the past have identified carbon, which constitutes 40e50% by mass of algae, as one of the main limitations to algal growth besides light (19). Carbon supply should be optimized so that it neither limits nor inhibits the growth; the level is strainspecific. For example, in a bubble columns study, Hsueh et al. (20) reported increase of 135% in the biomass under sparging with 5% CO2-enriched air (CEA) compared to sparging with ambient air; and, increase of 200% under sparging with 8% CEA. Further increase in enrichment to 10% resulted in inhibition, with pH falling below 5. In another study, Ryu et al. (21) evaluated the effect of CEA of 0.5%, 1%, 2% and 5% with Chlorella sp. cultures and reported that, compared to the biomass obtained at CEA of 0.5%, biomass increased by 34% at 1% CEA; by 55% at 2% CEA; and by 75% at 5% CEA. While sparging with CO2-enriched air is seen to increase productivity, the related costs also will increase. As such, an optimal enrichment has to be provided so that growth can be maximized. Thus, the first part of this study was to optimize the operation of the IIPBR in terms of the CO2-air ratio in the sparging gas. To validate the second premise, two alternate strategies were evaluated: (i) stressing under carbon-rich/nutrient-limiting condition in a single-stage; and (ii) carbon-limited/nutrient-limited condition in the second stage of a two-stage cultivation scheme. Several studies have shown that light intensity and nutrients such as nitrates, phosphates, CO2 supply not only aid in cell and chlorophyll formation, but also alter the biochemical pathways and formation of cellular components such as proteins, lipids, and carbohydrates (15,22). Stressing the cultures by limiting the light and nutrients can alter the biochemical composition of algae, favoring storage of energy-rich metabolites (15). To the best of our

93

knowledge none of the previous studies had considered carbonlimitation as a means of stressing and the duration of stressing as a means of altering the fatty acid composition and maximizing the energy ratio. Since the intent of this study was to evaluate and compare the effects of specific operating conditions, the tests were conducted under laboratory conditions under fixed light input to avoid the variations typically encountered under outdoor conditions. As such, the energy ratios estimated in this study including the light energy input, are quite low. Nevertheless, the absolute values of the energy ratios served as a rational measure, not only to assess the outcomes of different test conditions in this study, but also to compare the results of this study with those from several laboratory studies reported in the literature. MATERIALS AND METHODS Algal strains The two algal strains chosen in this study were Scenedesmus sp. and Nannochloropsis salina. Growth media for two species were prepared as described elsewhere (6) and autoclaved. The inocula for both algal cultures were grown in an incubator. The first-stage reactor (IIPBR) was seeded with the inoculum diluted with the nutrient media to achieve initial optical density at 750 nm (OD750nm) greater than 0.15. Internally illuminated PBR The details of the IIPBR tested in this study and its principle of operation have been described previously (6). In this study, the culture volume was fixed at 18 L; mixing was provided by sparging with CO2enriched air from the bottom of the annular space via four porous silica air diffusers. The air supply was sterilized (0.2 mm Millipore filter paper) and measured with gas proportioner (EW-03218-50 Cole Parmer flow meter system). CO2 gas flow rate was measured by a CO2 mass flow meter (00261BY, Cole Parmer) in standard mL min1. The CO2-air ratio was a variable in the experiments. Algal growth measurements Algal growth was tracked daily in terms of OD750nm and converted to dry weight through previously established correlations. Culture samples from the IIPBR were diluted with deionized water (23,24) prior to O.D. measurements to ensure that the spectrophotometer readings were below 0.5. Algal dry weight was determined by centrifuging the wet algal cells as described previously (6). In the case of the marine algae, N. salina, same procedure was repeated except for washing and centrifuging the settled algal samples twice to minimize salts. Besides growth measurement, pH and temperature were monitored using Mettler Toledo M300 pH transmitter placed in the culture. During all the tests under laboratory conditions, the temperature ranged between 26  C and 27  C. Light measurements were taken daily as described previously (6). Lipid and fatty acid methyl ester (FAME) analysis Total lipids, which are otherwise defined as organic solvent (such as hexane, ether) soluble fraction of a matrix of the algal cultures were estimated by the extraction of organic soluble solutes from dried tissue using a mixture of CHCl3 and MeOH based on the Folch method of lipid extraction (25). The lipid extraction is performed using an accelerated solvent extraction (ASE) system using Dionex 350 (Dionex Corporation, Salt Lake City, UT, USA) by Mulbry et al. (26). Briefly, 0.25 g of dried algal samples were mixed with 30 g of Ottawa sand and loaded into 33 mL sample cells. The mixtures of algae and sand were extracted using chloroform:methanol (2:1, v/v) at 120  C and pressure of w1500 psi for 5 min and transferred to a pre-weighed 60 mL collection vials. The increased temperature was adapted from ASE method (26). The extraction is performed under an inert atmosphere to prevent lipid oxidation. The collection vials with extract solutions were dried under a stream of nitrogen to estimate the lipid content gravimetrically. Within the context of this study we opted for optimal lipid yield and productivity with disregard to co-product production. All lipid extracts were stored under nitrogen at 20 C. The fatty acid profiles of the cultures were analyzed base catalyzed direct transesterification of algal tissue. The results represent the fatty acid profile of the bound esterified fatty acids and not the free fatty acid content. Briefly, 10 mL of glycerol tritridecanoate (13:0 FAME standard at 20 mg mL1 in hexane) is added to 50 mg of dry algal tissue. Then 5 mL of 0.2 N KOH in MeOH was added and the mixture vortexed for 20 s and finally placed in hot water bath at 65 C for 10 min (with additional vortexing for 30 s). These last two steps were repeated three times total. To quench the reaction, 1 mL of 1-M acetic acid was added to each sample and vortexed for 20 s. Two milliliters of hexane with an internal standard methyl tricosanoate at 50 mg L1 was added to each sample vial. Each sample was vortexed for 20 s and two phases are separated by centrifugation. The top hexane layer was taken for the GCeMS analysis. For GC/MS, helium was used as the carrier gas with a 2-ml injection volume. The temperature ramp started at 80 C and ramped 20 C min1e220 C and held for 6 min for a total run time of 13.3 min. The instrument was tuned with a standard spectra auto tune method, and a calibration curve was made from a Supelco component FAME mixture (cat no. 47885-U, Sigma Aldrich, St. Louis, MO, USA).

94

PEGALLAPATI ET AL.

J. BIOSCI. BIOENG.,

Experimental scheme To verify the first premise that biomass productivity in the IIPBR could be optimized in terms of carbon supply, growth of the two test cultures was evaluated in the IIPBR sparged with CO2-air mixtures at various CO2-air ratios (% vol/vol). Both species were sparged at the same gas flow rate of 800 mL min1 and at the same gas-to-culture volume (Q/L) ratio of 0.044 min1. One set of batch experiments was conducted for each species under sparging with ambient air to serve as the baseline. Scenedesmus sp., was tested in batch mode, sparged with CEA of 1.0% and in fed-batch mode sparged at CO2-air ratios of 2.0%, 3.0%, 4.0%, and 5.0%. N. salina was tested in fed-batch mode, with CO2-air ratios of 0.5, 1.0 and 2.0%. In the fed-batch mode, 10% of the culture volume was harvested, and the reactor volume was replenished with fresh media. Harvesting cycle was repeated when the biomass concentration (27) following a harvest equaled or surpassed the concentration prior to the harvest. To verify the second premise that lipid productivity could be improved by stressing, the two strategies mentioned earlier were evaluated. In the first one, carbon-rich/nutrient-limiting condition in single stage was simulated by extending the batch operation of the IIPBR into fed-batch mode with continuous sparging with CEA but without addition of any nutrients. In the second strategy, carbon-limiting condition in the second stage of a twostage system was simulated by cultivating the cultures harvested from the first stage under sparging with ambient air in 2-L beakers serving as the second stage. Each day’s harvest from the first stage was placed in a separate second stage reactor as indicated in Fig. 1; for example, volume harvested on day n was cultivated in beaker B-n, and stressed for (13 e n) days under sparging with ambient air to induce carbon limitation. All the second stage reactors were illuminated at the same intensity and sparged at the same rate as the first stage (0.044 min1), but with atmospheric air. Algal samples from the first stage and the second stage reactors were analyzed for lipid and TFA content at the end of 13 days of the experiment to assess the potential for lipid accumulation as a function of carbon-limited stressing time in the second stage. Lipid and FAME productivity Lipid productivity, L (g lipid L1 d1) and FAME productivity F (g FAME L1 d1) were estimated from  L ¼ B

LC 100

 F ¼ B



TFA 100

(1)  (2)

where B is biomass productivity estimated following Pegallapati et al. (8); LC is the gravimetric lipid content (%); and TFA is the gravimetric total FAME content (%) of the algal cells. Energy ratio In this study, energy ratio, ER, was estimated as the ratio of energy produced to the energy input. Energy produced was estimated by two methods: as that equivalent to the lipids produced (EL, W m3) using the typical calorific value of lipids, GL (¼38.93 kJ g1 lipid) (28); or, as that equivalent to the total FAMEs produced (EF, W m3) using the typical calorific value of FAMEs, GF (¼39.30 kJ g1 FAME) (29): EL ¼ GL L ¼ 38:93  L  1000  0:0115 ¼ 447:7L

(3)

EF ¼ GF F ¼ 39:30  F  1000  0:0115 ¼ 452:0F

(4)

Energy input was estimated as the sum of that expended for mixing (Em, W m3) and that expended for illuminating the cultures (Ei, W m3) as described previously (6): Em ¼

Ei ¼

gQh 60V

(5)

0:22SIi V

(6) 3

where g is the specific weight of the culture broth (¼9810 N m ); Q is the gas flow rate (m3 min1); h is the height of the culture (m); V is culture volume (m3); S is the illumination surface area of the reactor (m2); and Ii is the incident illumination (m Einsteins m2 s1). Thus, the energy ratio, ER () can be expressed either as ERL ¼ EL =ðEm þ Ei Þ

or

ERF ¼ EF =ðEm þ Ei Þ

(7) (8)

where ERL is lipid-based energy ratio and ERF is the FAME-based energy ratio.

RESULTS Optimal CO2 for biomass productivity Algal growth was expressed as dry biomass concentration using correlation developed between optical density and dry weight of the algal cells previously (8,30). Productivities and lipid contents achieved under sparging with ambient air and with CO2-air mixtures of various CO2-air ratios are summarized in Fig. 2 for the two test species. In both cases, biomass productivities were significantly higher when sparged with CEA than with ambient air, suggesting that the carbon supply rate was certainly limiting the growth in the latter case, and productivity could be maximized by selecting optimal CO2-air ratio. Under all runs with CO2-enrichment with both algal strains, pH was stabilized with no significant effects on algal growth, details of which were discussed previously (30). While we agree that comparison across different reactors and species is not accurate, it is impossible to find comparable literature studies with identical reactor configuration, algal species, operating conditions as in the present study. Results were compared and discussed based on at least one similar condition such as either CO2 supplementation or principle of operation of the reactor (e.g., bubble columns or airlift). Nevertheless, the energy ratio offers a common basis to compare the performance of different reactors, species, and operating conditions to identify the best combination to develop further.

FIG. 1. Plan view of first stage and second stage reactors for two-stage cultivation.

VOL. 117, 2014

EVALUATION OF INTERNALLY ILLUMINATED PHOTOBIOREACTOR

Lipid content

10

30 20

Productivity [g/L-d]

1 0.40 0.22

0.20 0.075

0.032

0.03

0.02

0.01

0.046

0.038

0.006 0.003

0.002 CO2-air ratio [%]

0.035

1.0

2.0

3.0

4.0

5.0

CO2 applied [g/L-d] 0.041

1.17

2.33

3.45

4.67

5.83

B

10

60 50 40

1 Productivity [g/L-d]

10

0.12

0.1

30 0.09

0.1

0.037

0.10 0.055

0.026

0.02

0.01

0.07

Lipid content [%]

A

Lipid

Lipid content [%]

Biomass

20

-10

0.002 CO2-air ratio [%]

0.035

0.5

1.0

2.0

CO2 applied [g/L-d]

0.041

0.58

1.17

2.33

in Table 1. When compared to Chiu et al. (34), Chiu et al. (35), Ryu et al. (21), and Loubiere et al. (36), the optimal productivity with Scenedesmus sp. obtained at CO2-air ratio of 4% in this study is as good, but at much lower gas-to-culture volume ratio (0.25 min1, 0.2 min1, and 0.13 min1 vs. 0.044 min1 in this study). Since gasto-culture volume ratio is directly proportional to the sparging energy input, the above results confirm the advantage of the IIPBR over the bubble columns used by Chiu et al. (34), Chiu et al. (35), and Ryu et al. (21). Comparing the results from the annular column design of Zittelli et al. (27), the maximum biomass productivity achieved in this study with N. salina (¼0.104 g L1 d1) is lower, but is deemed more energy-efficient because of the lower light intensity (91.4 vs. 175 and 259 m Einsteins m2 s1); and the lower gas-to culture volume ratio (0.044 min1 vs. 0.5 min1) utilized in the current study. The higher productivities reported by Zittelli et al. (37) in their flat panel reactor (0.61e0.85 g L1 d1) are due to the shorter light path length of 1.2 cm compared to that of 4.765 cm in the current study. Even though the comparisons of biomass productivities in Table 1 cover diverse algal species, a more equitable comparison could be made in terms of the energy ratio as discussed later.

10 0

0.007

95

-20

FIG. 2. Lipid content (total lipids), biomass productivity, and lipid productivity as a function of CO2-air ratio for (A) Scenedesmus sp. and (B) Nannochloropsis salina.

In the case of Scenedesmus sp., optimal CO2eenrichment for maximizing biomass productivity was observed at a CO2-air ratio of 4% with a biomass productivity of 0.401  0.07 g L1 d1 (averaged during harvesting) in the fed-batch mode, with an average biomass concentration of 1.4  0.04 g (dry) L1. An increasing trend in lipids was noted with CO2-air ratio up to 2%; thereafter, CO2-enrichment did not result in any increase in lipid content (Fig. 2A). Lower lipid contents obtained with 3% and 4% CEA runs can be attributed to lipid analyses of algal sample during active growth phase, which is the onset of steady state. Similar trend of lower lipids with higher biomass density was noticed in studies by Li et al. (31) and Lv et al. (22), suggesting that the optimal CO2-air ratio for lipid content and lipid productivity might not be the same. Lipid contents achieved in this study are comparable to those reported in the literature with Scenedesmus sp.: 23% in 250-mL Erlenmeyer flasks at CO2-air ratio of 5% (32); and 25% under sparging with ambient air (33). In the case of N. salina, optimal CO2-enrichment for maximizing biomass productivity was observed at a CO2-air ratio of 1%, with biomass productivity of 0.104  0.0076 g L1 d1 in the fed-batch mode with an average biomass concentration of 0.516  0.012 g (dry) L1. Lipid contents of N. salina obtained at the end of the experiments are presented in Fig. 2B. CO2-air ratio of 1% resulted in maximum lipid content of 52.7%. In this case, CO2-enrichment beyond 1% did not affect the lipids positively; similar finding was reported by Chiu et al. (34), though the CO2-enrichments adopted in their study were much higher (2%e15%). Lipid contents obtained in this study are comparable to those by Chiu et al. (35) with Nannochloropsis Oculata sparged with CO2-enriched air in fedbatch mode with daily harvesting: 29.7  2%, 26.2  1.9%, 24.6  1.7% and 22.7  1.9% at CO2-air ratios of 2%, 5%, 10% and 15%, respectively. These findings underscore the need to optimize CO2enrichment considering the lipid productivity to minimize the cost of carbon supply. Biomass productivities achieved in this study for the two species are compared against those reported in selected literature studies

Optimizing lipid accumulation in second stage Two-stage cultivation of Scenedesmus sp. and N. salina was evaluated with the respective optimal CO2-air ratios (of 4% and 1%) established in the first stage. The two-stage experiment lasted 13 days, at which point, total lipid (LC) and total lipid bound fatty acid (TFA) contents were analyzed in each of the second stage reactors and the first stage reactor (IIPBR). In the case of Scenedesmus sp., lipid and TFA contents of 11.3% and 3.9% were recorded in the first stage, while those in the second stage ranged from 15% to 32.4% and 10.47% to 25.01% depending on the stressing period as shown in Fig. 3A. The maximum lipid content of 32.4% (2.8 fold higher than that in the first stage) and maximum TFA of 25.01% (6.4 fold higher than that in the first stage) were achieved after 7 days of stressing under CO2 limitation in the second stage. In the case of N. salina, lipid and TFA contents of 39.2% and 11.2% were recorded in the first stage; those with CO2 limitation in the second stage ranged from 28% to 40% and 16.3% to 35.2% (Fig. 3B). Though no significant accumulation of lipids was noted as a result of two-stage cultivation of N. salina, approximately three-fold increase in TFA content (noted from 6 through 9 days of stressing) was observed during second stage (Fig. 3B). Lipid contents of 39%e 40% obtained in this study with N. salina are better than the lipid content of 22% reported by Boussiba et al. (38) obtained under sparging with ambient air; and 27% reported by Griffiths and Harrison (39), under nutrient replete conditions, for the same species.

TABLE 1. Comparison of biomass productivity: IIPBR vs. PBRs from literature. Study

This study This study Chiu et al. (34) Chiu et al. (35) Ryu et al. (21) Ryu et al. (21) Zittelli et al. (27) Zittelli et al. (27) Zittelli et al. (37) Zittelli et al. (37) Loubiere et al. (36)

Speciesa Modeb

S Ns C No C C N N N N Ch

FB FB FB FB B B FB FB FB FB B

Biomass CO2-air Q/L ratio PAR ratio (%) (min-1) (m E m-2 productivity (g L-1 d-1) s-1) 4 1 2 2 2 5 2 2 3 3 2

0.044 0.044 0.25 0.25 0.2 0.2 0.1 0.1 0.5 0.5 0.13

91.4 91.4 300 300 100 100 175 89 115 230 141

0.401 0.104 0.422 0.480 0.295 0.335 0.170 0.20 0.61 0.85 0.300

a S-Scendesmus sp.; Ns-N. salina; No-Nannochloropsis oculata; C-Chlorella sp.; N-Nannochloropsis sp.; Ch- Chlamydomonas reinhardti. b B- batch; FB- fed-batch.

PEGALLAPATI ET AL.

J. BIOSCI. BIOENG.,

A 35

32.4 29.7

30 25

25 22.2 20

20 15

15.1 14.9

20.1

28.4

28.2

26.9

24

23.8 21.7

19.8

19.4 16.9

16.2

14.9

21.8

21.2

15.4

14.3

14.1

12

10

10.4

5 0 1

2

3

B

4 5 6 7 8 9 10 11 Days of cultivtaion under CO2 limitation Total Lipids %

12

13

Total FAMEs %

45 40

40.7

39.6 37.4

39.9 37.9

39

35

33.6

35.1

35.5

33.9

33.8 30.6

30

28.8

27.9 26.1

25

29 23.9

27.5

30 28 20.5

20 16.2

15 10 5 0 1

2

3 6 7 8 9 10 11 Days of cultivtaion under CO2 limitation

12

13

FIG. 3. Lipid and total FAME contents achieved in stage 2 at the end of the experiment. (A) Scenedesmus sp., (B) Nannochlorposis salina.

While literature reports on carbon-sufficient and carbon-limited conditions in two-stage cultivation for lipid accumulation are scarce, additional generic comparisons with results reported under carbon-supplementation with the two other species can be made. In a study on Chlorella vulgaris, Lv et al. (22) had reported lipid content of 20% at CO2-air ratio of 1%; in a study on Botryococcus braunii, Ge et al. (40) had reported lipid content of 10.4% at CO2-air ratio of 2%. For the two algal species tested, lipid and TFA contents followed a similar trend with multiple peaks with no defined peak during the days limited with CO2. Similar trend with culture age was noticed under nutrient restrictions in the media with C. vulgaris and Chlamydomonas reinhardtii by Deng et al. (41) and with N. oculata seasonally outdoors by Olofsson et al. (42). In the present study, each data point in Fig. 3A and B represents a beaker subjected to stress, the trend obtained with lipids and TFA is inexplicable at this time. However, the underlying reasons behind such trend can be augmented by daily monitoring experiments for lipids in future work. The normalized distribution of the fatty acids show that under CO2 limitation there is a decrease in the ratio of saturated FAs to TFA for both species (Tables S1 and S2). Also to note is that the compositional ratio of the FAs did not remain the same throughout CO2 limitation indicating that there is lipid-remodeling occurring during the stress. A previous study in Chlorella vulgaris showed that FAME distribution is effectively modulated by varying CO2 concentrations (43) and was more pronounced in C18, thus making it possible to modify fatty acid content by fluctuating CO2 levels. In this study, a general trend for the relative abundance of the mono-

unsaturated species was noted to increase with CO2 limitation in N. salina and a similarly with Scenedesmus sp. until day 10 of stressing later to which the observations slowly decreased in abundance. There seems to be no apparent trends in the relative abundance in sums of the saturated or polyunsaturated species due to CO2 limitation. Further research is warranted to identify lipid species responsible for lipid remodeling. Energy ratio The energy ratio analysis was conducted to establish the optimal stressing period using the results obtained with N. salina. From the energy ratios plotted in Fig. 4, the optimal duration of CO2-limited growth to achieve the highest energy ratio can be seen as 10 days. Optimal energy ratios found in this study for the two stages are compared with those calculated from selected literature studies in Table 2. The lipid-based energy ratio, ERL, in the IIPBR under single-stage operation is 4e6 times higher than that in the bubble column studies by Chiu et al. (35) and Feng et al. (44). Though biomass productivities and lipid productivities reported by Chiu et al. (35) and Feng et al. (44) are higher than those found in the present study, their lower energy ratio compared to that of the IIPBR can be attributed to the sub-optimal utilization of the light energy in the bubble columns, CO2-enrichment, and the type of algal strain. As demonstrated earlier (6,8), the energy-efficiency of the IIPBR over other reactor geometries is reflected in the energetic measure, biomass production per unit energy input, B/E. Even though ERL did not improve in the second stage, result of this study is higher than that of Su et al. (12) under two-stage growth but with nitrate deprivation (0.009 vs. 0.002). The FAME-based energy ratio, ERF, achieved in this study under CO2 limitation is comparable to but lower than those reported by Rodolfi et al. (32) under nitrogen deprivation and phosphate starvation (0.005 vs. 0.006e0.014). Though higher biomass productivities were reported in the airlift flat plate reactors under nutrient replete conditions, the higher light energy input (over an incident surface area of 3.4 m2) in those studies lowered their energy ratios (32). As noted earlier, the above energy ratios are rather low because of the light energy input. Nevertheless this energetic measure enabled rational comparison of different reactors and species. Additionally, this energetic measure underscores the proposed approach to evaluate PBRs and cultivation systems on the basis of energy input and energy output rather than on the basis of biomass productivity or lipid productivity as had been done previously. The

0.01 0.009

0.009 0.008 Energy ratio [unit less]

Lipids and FAMEs at the end of the indicated days [%]

Lipids and FAMEs at the end of the indicated days [%]

96

0.007

0.007 0.006

0.006

0.005

0.005 0.004

0.004

0.003

0.003

0.006 0.005

0.004

0.007 0.006 0.006

0.005 0.004

0.004

0.004 0.004

0.003

0.002 0.001

0.001 0.001 0.001 0.0003

0 1

2 3 6 7 8 9 10 11 12 No. of days with CO2 Limitation, Nannochloropsis salina ERL

ERF

FIG. 4. Energy ratios obtained in second stage under CO2 limitation.

13

VOL. 117, 2014

EVALUATION OF INTERNALLY ILLUMINATED PHOTOBIOREACTOR

97

TABLE 2. Comparison of energy ratio: this study with literature results. a

Algae /Reactor

b

This study Ns/IIPBR Ns/B Rodolfi et al. (32) N/AFP N/AFP N/AFP N/AFP Feng et al. (44) Cz/BC Cz/BC Cz/BC Chiu et al. (35) No/BC Su et al. (12) No/Rec No/FP a b c d e f

B (g Le1 de1)

LC (%)

L (g Le1 de1)

TFA (%)

F (g Le1 de1)

ERLe

ERFf

CO2 limited

0.09 0.04

39.1 35.5

0.035 0.015

3.9 27.5

0.004 0.012

0.054 0.009

0.005 0.007

3 3 3 3

N deprived N sufficient P starved P sufficient

0.30 0.60 0.38 0.60

20 11 37 13.2

0.06 0.066 0.14 0.079

100 100 100

Ambient air Ambient air Ambient air

Without N, with P With N and without P With N and P

0.12 0.09 0.20

65.1 44.7 33.5

0.077 0.039 0.065

0.011 0.006 0.009

I

300

2

CO2 sufficient

0.48

29.7

0.143

0.012

I II

300 300

Ambient air 2

N sufficient Without N

0.11 0.07

10.0 48.2

0.011 0.034

0.015 0.002

c

Stage

PAR (m E me2 se1)

CO2-air ratio (%)

I II

91.4 91

1 Ambient air

NA NA NA NA

115 115 115 115

NA NA NA

Stressd

0.006 0.006 0.014 0.008

Ns, N. salina; N, Nannochloropsis sp.; Cz, Chlorella zofingiensis; No, Nannochloropsis oculta. IIPBR, internally illuminated photobioreactor; B, beakers; AFP, alveolar flat plate reactor; BC, bubble columns; FP, flat plate reactor; Rec, rectangular reactor. I, first stage; II, second stage; NA, not applicable, only one stage. N, nitrate; P, phosphate. Energy equivalent of lipid per unit energy input (Eq. 7). Energy equivalent of FAMEs per energy input (Eq. 8).

proposed approach can be applied to outdoor reactors under natural illumination by sunlight to estimate more realistic energy ratios. For example, using the results of outdoor studies on the green wall panel reactor (GWP) by Rodolfi et al. (32) and the flat plate reactor (FP) by Feng et al. (44), the following energy ratios were estimated: 1.05 under nitrate-sufficient conditions in GWP; 1.84 under nitrate-deprived conditions in GWP; and 1.83 under nitratedeprived conditions in FP. Though the energy ratios achieved in this study in the two stages are comparable with literature results, the energy ratios obtained in second stage were lower than those found in the first stage. Similar results were noted from two-stage study by Su et al. (12). Though the lipid content in that study improved approximately by five fold in the second stage, the energy ratio declined. Lower growth, non-identical reactor, and sub-optimal light input per unit volume of the culture might have resulted in higher energy inputs in the second stage. However, the benefits of second stage operation have to be assessed further by studying the effect of other limiting nutrients such as nitrate or phosphates outdoors using the IIPBR thereby minimizing the energy inputs (such as light) to the cultivation system. Use of energy ratio based on lipid content and total fatty acid content as an energetic measure to optimize growth of Scenedesmus sp. and N. salina was demonstrated in an internally illuminated photobioreatcor. The premise of improving lipids and TFA accumulation under CO2 limitation resulted in favorable results with Scenedesmus sp., but not with N. salina. Comparing the energy ratios found under the single-stage and two-stage strategies, it can be concluded that a single-stage operation may be sufficient. Twostage cultivation can be used to produce alternative FAME distributions from a single culture, thus allowing it to preferentially produce desired fatty acids for various commodities. Optimal CO2to-air ratios found in this study for improving productivity and energy efficiency of the two species were found comparable with those reported in the literature. While this paper presents the laboratory findings from the development of process to optimize the process, based on these results, we plan to extend the studies to outdoor tests. Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.jbiosc.2013.06.020.

ACKNOWLEDGMENTS This study was supported in part by a grant from the DOE National Alliance for Advanced Biofuels and Bioproducts (NAABB), a grant from the US Air Force Research Laboratory (AFRL), by the NSF Engineering Research Center, ReNUWIt, and by the Ed and Harold Foreman Endowed Chair. References 1. Batan, L., Quinn, J., Wilson, B., and Bradley, T.: Net energy and greenhouse gas emission evaluation of biodiesel derived from microalgae, Environ. Sci. Technol., 44, 7975e7980 (2010). 2. Hu, Q., Sommerfeld, M., Jarvis, E., Ghirardi, M., Posewitz, M., Seibert, M., and Darzins, A.: Microalgal triacylglycerols as feedstocks for biofuel production: perspectives and advances, Plant J., 54, 621e639 (2008). 3. Rittmann, B. E.: Opportunities for renewable bioenergy using microorganisms, Biotechnol. Bioeng., 100, 203e212 (2008). 4. Darzins, A., Pienkos, P., and Edye, L.: Current status and potential for algal biofuels production. Report T39-T2. BioIndustry Partners and National Renewable Energy Laboratory, IEA Bioenergy Task 39 (2010). 5. Hulatt, C.-J. and Thomas, D.-N.: Productivity, carbon dioxide uptake and net energy return of microalgal bubble column photobioreactors, Bioresour. Technol., 102, 5775e5787 (2011). 6. Pegallapati, A.-K. and Nirmalakhandan, N.: Energetic evaluation of an internally illuminated photobioreactor for algal cultivation, Biotechnol. Lett., 33, 2161e2167 (2011). 7. Arudchelvam, Y. and Nirmalakhandan, N.: Optimizing net energy gain in algal cultivation for biodiesel production, Bioresour. Technol., 114, 294e302 (2012). 8. Pegallapati, A.-K., Arudchelvam, Y., and Nirmalakhandan, N.: Energy-efficient photobioreactor configuration for algal biomass production, Bioresour. Technol., 126, 266e273 (2012). 9. Sandnes, J.-M., Kallqvist, T., Wenner, D., and Gislerød, H.-R.: Combined influence of light and temperature on growth rates of Nannochloropsis oceanica: linking cellular responses to large-scale biomass production, J. Appl. Phycol., 17, 515e525 (2005). 10. Solovchenko, A.-E., Khozin-Goldberg, I., Didi-Cohen, S., Cohen, Z., and Merzlyak, M.-N.: Effects of light intensity and nitrogen starvation on growth, total fatty acids and arachidonic acid in the green microalga Parietochloris incise, J. Appl. Phycol., 20, 245e251 (2008). 11. Xin, L., Hong-ying, H., Ke, G., and Ying-xue, S.: Effects of different nitrogen and phosphorus concentrations on the growth, nutrient uptake, and lipid accumulation of a freshwater microalga Scenedesmus sp., Bioresour. Technol., 101, 5494e5500 (2010). 12. Su, C.-H., Chien, L.-J., and Gomes, J.: Factors affecting lipid accumulation by Nannochloropsis oculata in a two-stage cultivation process, J. Appl. Phycol., 23, 903e908 (2011).

98

PEGALLAPATI ET AL.

13. Guschina, I.-A. and Harwood, J.-L.: Lipids and lipid metabolism in eukaryotic algae, Prog. Lipid Res., 45, 160e186 (2006). 14. Sato, N., Hagio, M., Wada, H., and Tsuzuki, M.: Environmental effects on acidic lipids of thylakoid membranes, Biochem. Soc. Trans., 28, 912e914 (2000). 15. Sharma, K.-K., Schuhmann, H., and Schenk, P.-M.: High lipid induction in microalgae for biodiesel production, Energies, 5, 1532e1553 (2012). 16. Thompson, G.-A.: Lipids and membrane function in green algae, Biochim. Biophys. Acta Lipids Lipid Metab., 1302, 17e45 (1996). 17. Wang, Z.-T., Ullrich, N., Joo, S., Waffenschmidt, S., and Goodenough, U.: Algal lipid bodies: stress induction, purification, and biochemical characterization in wild-type and starchless Chlamydomonas reinhardtii, Eukaryot. Cell, 8, 1856e1868 (2009). 18. Merchant, S.-S., Kropat, J., Liu, B., Shaw, J., and Warakanont, J.: TAG, you’re it! Chlamydomonas as a reference organism for understanding algal triacylglycerol accumulation, Curr. Opin. Biotechnol., 23, 352e363 (2012). 19. Chisti, Y.: Biodiesel from microalgae, Biotechnol. Advan., 25, 294e306 (2007). 20. Hsueh, H.-T., Li, W.-J., Chen, H.-H., and Chu, H.: Carbon biofixation by photosynthesis Thermosynechoccous sp. CL-1 and Nanochloropsis oculata, J. Photochem. Photobiol., 95, 33e39 (2009). 21. Ryu, H.-J., Oh, K.-K., and Kim, Y.-S.: Optimization of the influential factors for the improvement of CO2 utilization efficiency and CO2 mass transfer rate, J. Ind. Eng. Chem. Res., 15, 471e475 (2009). 22. Lv, J.-M., Cheng, L.-H., Xu, X.-H., Zhang, L., and Chen, H.-L.: Enhanced lipid production of Chlorella vulgaris by adjustment of cultivation conditions, Bioresour. Technol., 101, 6797e6804 (2010). 23. Yeh, K.-L. and Chang, J.-S.: Effect of light supply and carbon source on cell growth and cellular composition of a newly isolated microalga Chlorella vulgaris ESP-31, Eng. Life Sci., 10, 201e208 (2010). 24. Yeh, K.-L. and Chang, J.-S.: Nitrogen starvation strategies and photobioreactor design for enhancing lipid content and lipid production of a newly isolated microalga Chlorella vulgaris ESP 31: implications for biofuels, Biotechnol. J., 6, 1358e1366 (2011). 25. Folch, J., Lees, M., and Sloane-Stanley, G.-H.: A simple method for the isolation and purification of total lipids from animal tissues, J. Biol. Chem, 226, 497e509 (1957). 26. Mulbry, W., Kondrad, S., Buyer, J., and Luthria, D.-L.: Optimization of an oil extraction process for algae from the treatment of manure effluent, J. Am. Oil Chem. Soc., 86, 909e915 (2009). 27. Zittelli, G.-C., Rodolfi, L., and Tredici, M.-R.: Mass cultivation of Nannochloropsis sp. in annular reactors, J. Appl. Phycol., 15, 107e114 (2003). 28. Feinberg, D.-A.: Fuel options from microalgae with representative chemical compositions. SERI/TR-231e2427. Solar Energy Research Institute, Golden, CO (1984). 29. Williams, P.-J.-B. and Laurens, M.-L.: Microalgae as biodiesel & biomass feedstocks: review & analysis of the biochemistry, energetics & economics, Energy Environ. Sci., 3, 554e590 (2010).

J. BIOSCI. BIOENG., 30. Pegallapati, A.-K. and Nirmalakhandan, N.: Internally illuminated photobioreactor for algal cultivation under carbon dioxide-supplementation: performance evaluation, Renew. Energy, 56, 129e135 (2013). 31. Li, Y., Horsman, M., Wang, B., Wu, N., and Lan, C.-Q.: Effects of nitrogen sources on cell growth and lipid accumulation of green alga Neochloris oleoabundans, Appl. Microbiol. Biotechnol., 81, 629e636 (2008). 32. Rodolfi, L., Zittelli, G.-C., Bassi, N., Padovani, G., Biondi, N., Bonini, G., and Tredici, M.-R.: Microalgae for oil: strain selection, induction of lipid synthesis and outdoor mass cultivation in a low-cost photobioreactor, Biotechnol. Bioeng., 102, 101e112 (2008). 33. Xin, L., Hong-ying, H., and Yu-ping, Z.: Growth and lipid accumulation properties of a freshwater microalga Scenedesmus sp. under different cultivation temperature, Bioresour. Technol., 102, 3098e3102 (2011). 34. Chiu, S.-Y., Kao, C.-Y., Chen, C.-H., Kuan, T.-C., Ong, S.-C., and Lin, C.-S.: Reduction of CO2 by a high-density culture of Chlorella sp. in a semi continuous photobioreactor, Bioresour. Technol., 99, 3389e3396 (2008). 35. Chiu, S.-Y., Kao, C.-Y., Tsai, M.-T., Ong, S.-C., Chen, C.-H., and Lin, C.-S.: Lipid accumulation and CO2 utilization of Nannochloropsis oculta in response to CO2 aeration, Bioresour. Technol., 100, 833e838 (2009). 36. Loubiere, K., Pruvost, J., Aloui, F., and Legrand, J.: Investigations in an external-loop airlift photobioreactor with annular light chambers and swirling flow, Chem. Eng. Res. Des., 89, 164e171 (2010). 37. Zittelli, G.-C., Pastorelli, R., and Tredici, M. R.: A modular flat panel photobioreactor (MFPP) for indoor mass cultivation of Nannochloropsis sp. under artificial illumination, J. Appl. Phycol., 12, 521e526 (2000). 38. Boussiba, S., Vonshak, A., Cohen, Z., Avissar, Y., and Richmond, A.: Lipid and biomass production by the halotolerant microalga Nanochloropsis salina, Biomass, 12, 37e47 (1987). 39. Griffiths, M.-J. and Harrison, S.-T.-L.: Lipid productivity as a key characteristic for choosing algal species for biodiesel production, J. Appl. Phycol., 21, 493e507 (2009). 40. Ge, Y., Liu, J., and Tian, G.: Growth characteristics of Botryococcus braunii 765 under high CO2 concentration in photobioreactor, Bioresour. Technol., 102, 130e134 (2011). 41. Deng, X., Fei1, X., and Li, Y.: The effects of nutritional restriction on neutral lipid accumulation in Chlamydomonas and Chlorella, Afr. J. Microbiol. Res., 5, 260e270 (2011). 42. Olofsson, M., Lamela, T., Nilsson, E., Bergé, J.-P., Pino, V.-D., Uronen, P., and Legrand, C.: Seasonal variation of lipids and fatty acids of the microalgae Nannochloropsis oculata grown in outdoor large-scale photobioreactors, Energies, 5, 1577e1592 (2012). 43. Yusof, Y.-A.-M., Basari, J.-M.-H., Mukti, N.-A., Sabuddin, R., Muda, A.-R., Sulaiman, S., Makpol, S., and Ngah, W.-Z.-W.: Fatty acid composition of microalgae Chlorella vulgaris can be modulated by varying carbon dioxide concentration in outdoor culture, Afr. J. Biotechnol., 10, 13536e13542 (2011). 44. Feng, P., Deng, F., Fan, L., and Hu, Z.: Lipid accumulation and growth characteristics of Chlorella zofingiensis under different nitrate and phosphate concentrations, J. Biosci. Bioeng., 114, 405e410 (2012).