Effects of Nitrogen Availability on the Bioenergy Production Potential and CO 2 Fixation of Thermosynechococcus CL-1 under Continuous Cultivation

Nitrogen availability directly affects the microalgal metabolism. A thermophilic cyanobacterium named Thermosynechococcus CL-1 was cultivated in a continuous system to evaluate the effects of NO3 fluxes on the biomass production, bioenergy production, and CO2 fixation. The results show that decreasing the NO3 flux to a N-deprived level (1.01 mM/d) enhances the carbohydrate content in TCL-1 to 45%, accompanied by a decrease in the lipid content. However, increasing the NO3 flux from a N-deprived level (1.01 mM/d) decreases the carbohydrate content dramatically, accompanied by a slight increase in the lipid content. No matter whether the NO3 flux decreases from a N-replete level (8.35 mM/d) to a N-deprived one (1.01 mM/d), or increases from a N-deprived level to higher one, the peak biomass yield occurs at the same NO3 flux level, 4.18 mM/d. In addition, the peak lipid yield, carbohydrate yield, and CO2 fixation rate were recorded at 482 and 660 mg/L/d, and 3.9 g/L/d, respectively, under the same NO3 flux level (4.18 mM/d). Although cultivating TCL-1 under the 4.18 mM/d NO3 flux level exhibits great biomass production, CO2 fixation, and bioenergy production potential, different procedures (the NO3 flux decreasing from the N-replete level and the NO3 flux increasing from the N-deprived level) could influence the final quantity of bioenergy production and CO2 fixation. The NO3 flux, NO3 concentration in the bioreactor, and the NO3 flux variation routes are all important factors to determine the nitrogen availability of TCL-1 in continuous cultivation.


INTRODUCTION
Global warming and climate change result from the excess emissions of CO 2 in the atmosphere.Adopting fossil fuels as the primary energy source is responsible for the excess emissions of CO 2 .To reduce the CO 2 level in the atmosphere, both of developing CO 2 sequestration technologies as well as searching sustainable energy are important.Photosynthetic organisms, e.g., cyanobacteria and microalgae, harvest the light energy to fix CO 2 and store the energy in the biomass as bioenergy.Bioenergy is renewable and sustainable (John et al., 2011), and is considered to be substituted for fossil fuels as the energy source.Cultivating cyanobacteria or microalgae exhibits great potential for reducing the CO 2 levels in the atmosphere as well as producing the bioenergy.
Instead of purging CO 2 into microalgae cultivation system, capturing the CO 2 from the gas phase into liquid phase via alkaline solution (as bicarbonate and carbonate) followed by cultivating thermophilic cyanobacteria (Thermosynechococcus CL-1; TCL-1) has been investigated to avoid CO 2 leakage and acidification of the culture medium (Hsueh et al., 2007b).Cyanobacteria can assimilate bicarbonate as the carbon source (Hsueh et al., 2009;Su et al., 2012), and the assimilation rate of bicarbonate is higher than that of CO 2 (Hsueh et al., 2007a).
Lipid or carbohydrate content in the biomass as well as the biomass productivity is the key to apply the biomass for biodiesel or bioethanol production (Griffiths and Harrison, 2009;Branyikova et al., 2011).To obtain higher lipid or carbohydrate content in the biomass, nitrogen deprivation is a feasible strategy.It is well known that microalgae decline the protein content but enhance lipid (Merzlyak et al., 2007) or carbohydrate content (Branyikova et al., 2011) when nitrogen is exhausted.Some microalgae can accumulate lipid from 50 to 60% (dry weight) (Griffiths and Harrison, 2009) or carbohydrate from 38 to 51% (dry weight) (Branyikova et al., 2011;Ho et al., 2012), but cyanobacteria enhance the carbohydrate content more under nitrogen deprived conditions (Griffiths and Harrison, 2009).Although the strategy of nitrogen deprivation can be applied to enhance the lipid or carbohydrate content in the biomass, the changes of morphology (Krasikov et al., 2012), photosynthetic efficiency (Frada et al., 2013), growth rate (Krasikov et al., 2012), and the cell composition (Merzlyak et al., 2007;Ho et al., 2012) have been revealed.As a result, the decline of the bioenergy yield is inevitable due to the decrease of the growth rate.Due to aforementioned reasonings, Griffiths and Harrison (2009) suggests that both the composition and the biomass yield should be taken into consideration.Cultivating fast growing microalgae can improve the biomass yield, and thus improve the bioenergy yield.Since only adopting the nutrient deprived strategy can't enhance the energy production efficiently, other strategies for increasing the microalgal growth rate or the biomass yield should be considered and adopted.
Resupplying the nitrogen source to the microalgae which have gone through N-deprived conditions may have the potential for enhancing the bioenergy yield.Microalgae hold the essential pigment and protein to keep the cell alive under N-deprived conditions, and wait for the nitrogen source replenishing to the environment.Once the nitrogen replenishing to the environment, the enhancement of the growth rate (Voronova et al., 2009) and the inorganic carbon assimilation rate (Livne and Sukenik, 1992) have been observed.However, the effects of N-deprived conditions and resupplying the nitrogen source to the cultivation system on the biomass yield, CO 2 fixation rate, and the bioenergy yield of the thermophilic cyanobacteria have not been investigated.
The present study focuses on the effects of N-deprived conditions and replenishing the nitrogen source from Ndeprived conditions on thermophilic cyanobacterium, Thermosynchecoccous CL1 (TCL-1).TCL-1 exhibits high growth rate (3.5 d -1 in Hsueh et al., 2009), and being considered to have great potential for CO 2 fixation and bioenergy production.Decreasing the NO 3 -flux from Nreplete to N-deprived level followed by increasing the NO 3 flux from N-deprived to N-replete level in continuous cultivation system was adopted to evaluate the effect of nitrogen availability on the biomass, bioenergy yield, and CO 2 fixation of TCL-1.Both of the lipid and carbohydrate yield were examined, and the calorific and energy generation rate of TCL-1 were also included to evaluate the bioenergy production potential.

Strain and Culture Medium
The unicellular fresh water cyanobacterium strain TCL-1 was isolated from the Chin-Lun hot spring in eastern Taiwan as described previously (Hsueh et al, 2007a).The modified Fitzgerald medium (Takeuchi et al., 1992) was the influent medium in the continuous cultivation system, and the ingredients (in mg/L) were as follows: 39 K 2 HPO 4 , 75 MgSO 4 •7H 2 O, 36 CaCl 2 •2H 2 O, 58 Na 2 SiO 3 •9H 2 O, 6 FeC 6 H 5 O 7 , 6 citric acid, 1 EDTA, and 1 mL/L Gaffron solution (Takeuchi et al., 1992).Sodium bicarbonate and sodium carbonate were added as carbon source (pH 9.5) in the influent medium, and the concentrations of the dissolved inorganic carbon (DIC) were fixed at 113 mM (as CO 3 -2 ).Sodium nitrate was the nitrogen source.The nitrate concentrations in the feeding medium (DIN in ) were altered after the cultivation system reached steady state.The nitrate concentrations decreased gradually from 5.8 mM (N-replete) to 4.4, 2.9, 1.5, and 0.7 mM (N-deprived) followed by increasing gradually from 0.7 mM (N-deprived) to 2.9, 5.8, and 11.6 mM.All the chemicals used in the culture medium were of analytical grades.

Continuous System
A tubular bioreactor with 45 cm height, 5.9 cm inner diameter, and 1 L working volume was used in this study.A valve to harvest samples for daily analyses was located at the bottom of the bioreactor.In addition, a hole for effluent was set at 37 cm height.N 2 was purged into the bioreactor under constant flow rate to prevent the accumulation of O 2 that would otherwise inhibit photosynthesis of TCL-1.Mixing was performed by magnetic stirrer under constant rotational speed.Continuous illumination was accomplished by fluorescent lamp with 15 klx (measured by lx meter; TM 50000, TOMEI).
After preparation and sterilization, the medium was mixed homogeneously by magnetic stirrer and balanced in the incubator at 50°C before using.TCL-1 was pre-cultured in a 250 mL glass bioreactor, and was inoculated into the 1 L tubular bioreactor to reach the OD 680nm (optical density at 680 nm) at 0.1.The medium was fed into the bioreactor via the peristaltic pump at 1 mL/min constant flow rate, i.e., the dilution rate was 1.4 d -1 (the maximum growth rate of TCL-1 was 3.5 d -1 ; Hsueh et al., 2009).Consequently the NO 3 -fluxes were 8.35 (N-replete) to 6.34, 4.18, 2.16, and 1.01 mM/d (N-deprived) followed by 1.01 to 4. 18, 8.35, and 16.85 mM/d (as shown in Table 1).All the equipment was kept at 50°C inside an incubator (HIPOINT,.The data presented in all the Tables and Figures were the results under steady states (the steady states reached after 5 days later when the NO 3 -flux changed).The biomass yield presented in the Figs. 2 (a) and 2(d) were the mean and the standard deviation values under each steady state (at least 5 samples).The data in Table 2, Figs. 1, 3, and 4 are the mean values of two experimental samples taken at each steady state.

Analytical Procedures
Samples were taken from the bioreactor and from the effluent.Microalgal biomass productivity is the product of the microalgal cell density by the dilution rate (1.4 d -1 ).To determine the cell density, 10 mL samples were taken from the bioreactor, filtered (by 0.45 μm glass fiber filters; Advantec, GC-50), dried (put the filter into 105 °C oven one day) and weighted by analytical balance (AND, GR-202).The aliquot from filtration was collected, quantified to 100 mL, and preserved the 30 mL in a 4 °C refrigerator.The DIC and DIN in the bioreactor were analyzed by ion chromatograph (DX-100, DIONEX, KOH was the eluent,).The C/N ratio was calculated by dividing "DIC in the bioreactor" by "DIN in the bioreactor".
The biomass in the effluent was harvested, washed with where DIC in the bioreactor stands for the dissolved inorganic carbon in the bioreactor; DIC in stands for the dissolved inorganic carbon in the influent medium.The time course profiles of the biomass concentrations, DIC and DIN concentrations can be found in the supplementary file (Figs.S1 to S3).
To evaluate the TCL-1 bioenergy production potential, the lipid and carbohydrate content were analyzed.The method for analyzing the lipid content was as Holm-Hansen et al. (1967) described.After the biomass broken by sonication (SONIFIER-450, BRANSON), the methanol/ chloroform (1:2 v/v) solution was used for lipid extraction.The method for analyzing the carbohydrate content was as DuBois et al. (1956) described.The protein content was evaluated by the product of nitrogen content in TCL-1 times 6.25 (Batista et al., 2013).

Effects of the Nitrogen Deprivation on the Cellular Compositions
The nitrogen and carbon contents in TCL-1 are affected obviously by the NO 3 -fluxes (as shown in Table 2).Contrast to decrease the nitrogen content slightly while the NO 3 -flux decreases from 8.35 to 6.34 mM/d, the nitrogen content declines dramatically once the NO 3 -flux lower than 6.34 mM/d.The nitrogen content under 1.01 mM/d NO 3 -flux is 2.9% only.It suggests that the nitrogen content in TCL-1 reflects the NO 3 -flux directly in the system.Moreover, the nitrogen content in TCL-1 declines accompanied with a decrease of the protein content.The protein content decreases from 51.9 to 18.1% while the NO 3 -flux decreases from 8.35 to 1.01 mM/d.This indicates that the thermophilic cyanobacteria may decrease the protein content under Ndeprived conditions.
An obvious decline of the carbon content in TCL-1 is observed while the NO 3 -flux is less than 4.18 mM/d.It suggests that the carbon content in TCL-1 remains stable unless the NO 3 -flux decreases substantially (< 4.18 mM/d).When the NO 3 -flux is low, the mechanism of carbon fixation by TCL-1 may be damaged by the insufficient of nitrogen.
Lower lipid content in TCL-1 under lower NO 3 -flux is observed (Fig. 1(a)).Contrast to the obvious enhancement of the lipid content in microalgae, there is no dramatic enhancement of the lipid content in TCL-1 while the NO 3 flux decreases.Similar result can be observed while the mesophilic cyanobacteria are cultivated under N-deprived conditions.The lipid content in Synechococcus sp.can't be increased under N-deprived conditions (Griffiths and Harrison, 2009).Even though the thermophilic cyanobacteria are cultivated in this study, the lipid content cannot be enhanced under N-deprived conditions.However, it is noted that the lipid contents in TCL-1 (range from 18.3 to 20.6%) under nitrogen-replete conditions (8.35 to 6.34 mM/d) are higher than the lipid contents in mesophilic cyanobacteria (with an average of 8%), and are comparable to the lipid contents in green algae under nitrogen-replete conditions (with an average of 23%; Griffiths and Harrison, 2009).The lipid contents in thermophilic cyanobacteria are higher than the lipid contents in mesophilic cyanobacteria, e.g., the lipid content in thermophilic cyanobacterium Synechococcus elongatus PCC 6301 is 18%, in Chroococcidiopsis is ranged from 10 to 15% (Ishida et al., 1997), in Thermosynechococcus elongates is approximate 20% (Eberly and Ely, 2012).It shows high potential of adopting thermophilic cyanobacteria as the feedstocks for biodiesel production.
Carbohydrate content in TCL-1 is very sensitive to the nitrogen availability (Fig. 1(b)).Even though the NO 3 -flux 42.9 ± 0.0 7.8 ± 0.0 6.9 ± 0.0 104.7 ± 8.9 decreases from 8.35 to 6.34 mM/d, the carbohydrate content increases from 17.2 to 25.4%.When the NO 3 -flux further decreases to 4.18 mM/d, an obvious enhancement of the carbohydrate content (from 25.4 to 43.4%) is discovered.However, the carbohydrate content maintains at about 45% even though the NO 3 -flux decreases to 1.01 mM/d.
However, the increasing of the glycogen content in cyanobacteria under N-deprived conditions has been presented (Allen, 1984;Lehmann and Wober, 1978;Hickman et al., 2013).The glycogen content can be accumulated to more than 50% in Anacystis nidulans (Synechococcus PCC 6301) under N-deprived conditions in batch and continuous systems (while the influent KNO 3 is 0.2 g/L and the dilution rate is lower than 0.05 h -1 ; Lehmann and Wober, 1978).The glycogen content in Synechococcus elongates can be accumulated to 40-60% under N-deprived conditions (Allen, 1984;Hickman et al., 2013).Although the increased carbohydrate content in TCL-1 can be observed, the actual carbohydrate (starch, glycogen, cellulose, or other accumulating carbohydrates) is worth for further investigation.

Effects of Resupplying the Nitrogen from Deprivation on the Cellular Compositions
TCL-1 cellular compositions are varied after increasing the NO 3 -flux to N-deprived conditions, i.e., increasing the NO 3 -flux from 1.01 to 4.18 mM/d (Table 2).Enhancing the NO 3 -flux from 1.01 to 4.18 mM increases the nitrogen content from 2.9 to 6.9%.However, the nitrogen content increases slightly while the NO 3 -flux increases continually.Similar to the nitrogen contents in TCL-1, the carbon contents in TCL-1 are also enhanced by increasing the NO 3 -flux from 1.01, 4.18, to 8.35 mM/d (Table 2).The carbon content is restored to 44% while the NO 3 -flux increases to 8.35 mM/d.It suggests that increasing the NO 3 -flux from 1.01 to 4.18 mM/d help the carbon fixed in the TCL-1.It may be resulted from the formation of proteins, e.g., Rubisco or DIC transporters, by the resupply of the nitrogen, which increases the carbon fixation efficiently.Nevertheless, the carbon content can't be enhanced anymore after the NO 3 flux increases to 8.35 mM/d.Therefore, the nitrogen is not a limited factor for carbon fixation while the NO 3 -flux is greater than 8.35 mM.
The lipid and carbohydrate contents in TCL-1 show different patterns as increasing the NO 3 -flux from Ndeprived level (1.01 to 4.18 mM/d) (Figs.1(c) and (d)).Slight enhancement of the lipid content is found, contrast to the sharp decline of the carbohydrate content.Owing to the glucose (carbohydrate) is the initial product from photosynthesis and the lipid should be synthetized in the following metabolism, the inadequate NO 3 -flux may cause the enzyme insufficient for synthesizing the lipids.Therefore, increasing the NO 3 -flux from 1.01 to 4.18 mM/d may repair the enzyme that is related to the lipid synthesize but degraded by the nitrogen deprivation.
Once the NO 3 -flux is higher than 4.18 mM/d, the variations between individual lipids or carbohydrates contents are small, and indicates that the nitrogen supply is not a dominate factor for affecting the lipids or carbohydrates content.This can be examined by the excess DIN in the bioreactor under 8.35 and 16.85 mM/d (Table 1).

Effects of Nitrogen Availability on the Bioenergy Production and CO 2 Fixation
High biomass yield as well as the right content in the biomass is the key to obtain high bioenergy yield.The biomass yield is affected obviously by the NO 3 -fluxes (Figs.2(a) and 2(d)).It is noted that the sharp enhancement of biomass yield under 4.18 mM/d is observed no matter the processes are decreasing or increasing the NO 3 -flux.The highest biomass yield is 2,557 and 1,521 mg/L/d under the same NO 3 -flux (4.18 mM/d).It is noted that even though the DIN in the bioreactor under 4.18 mM/d NO 3 flux are extreme low (0.19 or 0.17 mM), the biomass yield increases sharply.Maintaining low DIN level in the bioreactor and feeding the NO 3 -slowly by reducing the NO 3 -flux may keep TCL-1 hunger for nitrogen source, and thus utilize the nitrogen source efficiently as well as stimulate cell division.In addition, Brussaard et al. (1998) demonstrates resupplying the nitrogen source to N-deprived cell may enhance the cell lysis of diatom.The low DIN concentration in the bioreactor and low NO 3 -flux may avoid cell lysis from replenishing the nitrogen source to N-deprived conditions.Therefore, the enhancement of the biomass yield may come from the eager to the nitrogen source used in cell division and from the avoidance of the cell lysis.
The nitrogen availability can't be assessed by the DIN concentration in the bioreactor or NO 3 -flux alone.Although the DIN levels are close to 0.1-0.2mM under 4.18 and 1.01 mM/d NO 3 -fluxes, the biomass yields of TCL-1 are extremely different.Even though the NO 3 -fluxes are varied to the same level while other nutrients are sufficient, the way to reach the particular NO 3 -flux is the key to affect the biomass yield of TCL-1.For example, the biomass yield is 1,521 mg/L/d while the initial NO 3 -flux is set at 8.35 mM/d followed by decreasing the NO 3 -flux to 6.34 and 4.18 mM/d.However, the biomass yield is 2,557 mg/L/d while the NO 3 -flux increases from 1.01 to 4.18 mM/d.The ambient nutrients including DIN (approximate 0.1-0.2mM) in the bioreactor are similar under 4.18 mM/d NO 3 flux, but the processes that TCL-1 experienced are different (decreasing or increasing the NO 3 -fluxes).Therefore, the DIN concentration in the bioreactor as well as the NO 3 flux can't perfectly explain the relationship between the nitrogen availability and properties of TCL-1.Two procedures, namely, the NO 3 -flux decreasing from the Nreplete level and the NO 3 -flux increasing from the Ndeprived level, as well as NO 3 -level could also influence the biomass yield.Similar results can also be observed in Lehmann and Wober's study (1978).Although the same nitrogen fluxes and the same procedures of chemostat are set in their study, the different biomass yields of Anacystis nidulans (Synechococcus PCC 6301) can be observed due to the different nitrogen availability (the degree of nitrogen deprivation in batch cultivation before the chemostat starts).Integrating their result and the result presented in this study, we hypothesis that all of NO 3 -flux, nitrogen concentration in the bioreactor, and the variation routes of NO 3 -flux during microalgal cultivation determine the nitrogen availability of microalgae.
Since the biomass yield is affected by the NO 3 -flux, nitrogen concentration in the bioreactor, and the variation routes of NO 3 -flux during cultivation, these factors also influence the bioenergy yield of TCL-1.Both of the lipid  et al., 2010), it suggests that the processes of increasing the NO 3 -flux from N-deprived conditions can be another strategy to obtain high lipid yield.Increasing the biomass yield instead of increasing the lipid content only, especially under nitrogen deprivation conditions with sacrificing the lipid yield, may fulfill the high lipid yield demand.Contrast to the lipid contents vary slightly under various NO 3 -fluxes, the carbohydrate contents vary apparently.As a result, both of the carbohydrate content and the biomass yield are the keys to obtain high carbohydrate yield.The strategy for enhancing the carbohydrate content or the biomass yield alone may not adequate for getting high carbohydrate yield.While the NO 3 -flux decreases gradually from 8.35 to 6.34, and 4.18 mM/d, both of carbohydrate content and the biomass yield reaches to outstanding levels under 4.18 mM/d, and thus the peak carbohydrate yield (660 mg/L/d) are reached.While the NO 3 -flux increases gradually from 1.01 to 4.18 mM/d (from nitrogen deprivation conditions), the high carbohydrate yield of 562 mg/L/d is reached.The carbohydrate yields in some references and this work are revealed in Table 3.The distinguished carbohydrate yields in this study may result from the difference in analytical methods, i.e., the total carbohydrate yield in this work may higher than the individual carbohydrate yield in other references.According to the results of lipid and carbohydrate yield, setting the NO 3 -flux at 4.18 mM/d may represent an excellent level for bioenergy production by TCL-1.
The calorific values of TCL-1 are close to an average value, 21.8 kJ/g, under various NO 3 -fluxes (Figs.3(a) and 3(c)).Therefore, the energy generation rate also depends strongly on the biomass yield.The calorific values of TCL-1 are similar with the calorific values of Spirulina (21.2 kJ/g; Biller and Ross, 2011).The peak energy generation rate is 55.9 kJ/L/d under 4.18 mM/d NO 3 -flux and increases the NO 3 -flux from nitrogen deprivation conditions.The highest CO 2 fixation rate in this study (Figs. 4(a) and 4(b)) was recorded at 3.9 g/L/d under 4.18 mM/d NO 3 flux while the NO 3 -flux increased from the low level (nitrogen deprivation conditions; 1.01 mM/d NO 3 -flux).On the other hand, another CO 2 fixation rate, 2.4 g/L/d, was recorded under the same NO 3 -flux (4.18 mM/d) while NO 3 -flux decreased from 8.35 mM/d.Both of the CO 2 fixation rates (3.9 or 2.4 g/L/d) of this species exhibit greater potential than other cyanobacteria while the CO 2 fixation rates of Synechocystis aquatulis, Anabaena sp., Microcystis aeruginosa, Microcystis ichthyoblabe, and Spirulina sp. are 1.5, 1.4, 0.52, 0.49, and 0.38 mg/L/d, respectively (Ho et al., 2011).It reveals that 4.18 mM/d NO 3 -flux exhibits distinguished characteristic of bioenergy production as well as CO 2 fixation.

CONCLUSIONS
Effects of decreasing the NO 3 -flux from N-replete to Ndeprived level and increasing the NO 3 -flux from Ndeprived to N-deprived level on TCL-1 were examined in a continuous cultivation system.Decreasing the NO 3 -flux to N-deprived level can enhance the carbohydrate content to 45% accompanied with a decrease of the lipid content.However, increasing the NO 3 -flux from N-deprived level (1.01 mM/d) decreases the carbohydrate content dramatically accompanied with an increase of the lipid content slightly.
No matter the NO 3 -flux decreasing from N-replete level (8.35 mM/d) to N-deprived level (1.01 mM/d) or increasing from N-deprived level to higher N-level, the peak biomass yield occurs at the same NO 3 -flux level, 4.18 mM/d.The peak lipid yield, carbohydrate yield, and the CO 2 fixation rate were recorded at 482 and 660 mg/L/d, and 3.9 g/L/d, respectively, under the same NO 3 -flux level (4.18 mM/d).Two procedures, namely, the NO 3 -flux decreasing from the N-replete level and the NO 3 -flux increasing from the N-deprived level, as well as NO 3 -level could also influence the final quantity of bioenergy production and CO 2 fixation.All the NO 3 -flux, NO 3 -concentration in the bioreactor, and the NO 3 -flux variation routes are important factors to determine the nitrogen availability of TCL-1 in continuous cultivation.

Fig. 1 .
Fig. 1.Effects of decreasing the NO 3 -fluxes gradually on the (a) lipid content and (b) carbohydrate content of TCL-1, and increasing the NO 3 -fluxes gradually on the (c) lipid content and (d) carbohydrate content of TCL-1.

Fig. 2 .
Fig. 2. Effects of decreasing the NO 3 -fluxes gradually on the (a) biomass (b) lipid (c) carbohydrate yield of TCL-1, and increasing the NO 3 -fluxes gradually on the (d) biomass (e) lipid (f) carbohydrate yield of TCL-1.

Fig. 3 .Fig. 4 .
Fig. 3. Effects of decreasing the NO 3 -fluxes gradually on the (a) calorific value (b) energy generation rate of TCL-1, and increasing the NO 3 -fluxes gradually on the (c) calorific value (d) energy generation rate of TCL-1.

Table 1 .
The effect of various NO 3 -fluxes on the DIC and DIN concentrations in a bioreactor under steady state.
rotation speed of a centrifuge at 5,000 rpm; U-32R, BOECO), and dried with a freeze dryer (FDU-1200, EYELA).The dried biomass was preserved in -20°C refrigerator before analysis.To determine the C, N, H content in TCL-1, about 3.5 mg dried biomass was analyzed by elemental analyzer (Vario EL III, Elementar).The CO 2 fixation rate was the product of the carbon content and the biomass yield times 3.67 (from 44/12).The C recovery (%) was calculated by the following equation: in DIC in the bioreactor (cell density C%) Crecovery (%) DIC 100%

Table 2 .
The effect of various NO 3 -fluxes on the cellular compositions of TCL-1.

Table 3 .
Carbohydrate yields of some cyanobacteria or microalgae species in references and this work.Calculated by (biomass concentration × (starch content in individual species)/20d) b Calculated by (starch content in Chlorella vulgaris (0.8 g/L)/18d) c Calculated by (glycogen concentration variations (0.08 g/L)/3d)