Impact of salinity on the kinetics of CO2 fixation by Spirulina platensis cultivated in semi-continuous photobioreactors

In this research, the physiological response of the microalgae Spirulina platensis to salinity stress (1 and 100 g L) was investigated. Spirulina platensis and S. platensis (adapted to high salt concentration) were operated at laboratory scale in a semi-continuous photobioreactors. The responses examined were within 0.5 to 10% CO2 concentration, temperatures from 10 to 40 °C, light intensities from 60 to 200 μmol m s and presented better results in terms of all kinetic parameters. The highest rate of CO2 biofixation for S. platensis was 25.1 gCO2 m h, and the maximum specific growth (μmax) achieved was 0.44 d 0.67 d at 2.5% CO2, 150 μmol m s at 25 °C. Corresponding determined values of S. platensis adapted were 18.2 gCO2 m h, 0.31 d 0.58 d at 2.5% CO2, 60 μmol s m and 28 °C. However, both microalgae exhibited experimental limiting growth factors, CO2 10%, 40 °C and 200 μmol m s, conditions under which photosynthetic CO2 biofixation may be inhibited and photoinhibition of photosynthesis may be enhanced by salinity. The efficiency of 2.5% CO2 removal by S. platensis achieved 99%, whereas S. platensis adapted to 96%, respectively. The kinetic parameters estimated for S. platensis can be used to improve photobioreactor design for reducing of atmospheric carbon dioxide.


Introduction
Global warming is generally attributed to greenhouse gases (GHG) increase in the atmosphere, particularly carbon dioxide (CO2), for which atmospheric concentration has already achieved 387 ppm and needs to get down to 350 ppm or less in order to avoid global climate change consequences 1,2 . By 2100, 26 billion tons of CO2 are estimated to be released into the atmosphere from anthropogenic sources 3 . Photosynthetic organisms such as microalgae species are potent producers of value-added bioactive compounds such as pigments, vitamins and long-chain polyunsaturated fatty acids, when grown under stress conditions can accumulate significant quantities of total lipids [4][5][6] . Recent studies indicated that improvements in culture conditions are needed to obtain adequate productivity of lipid, protein, carbohydrate content. It is well known that numerous parameters influence the growth of these compound content in microalgae: CO2 addition, light, temperature, salinity, nutrient addition, inoculation size, stirring, pH, etc. 4,7 . The National Aeronautics and Space Administration (NASA) was the first institution to become interested in microalgae Spriulina for oxygen production, CO2 reduction and proposed it as one of the primary foods to be cultivated in a future bioregenerative life support system for long-term manned space missions' scenarios such as Moon and Mars bases [4][5][6] . The cyanobacterium Spirulina platensis (S. platensis) is commercially produced as a nutrient source in health food, feed and pharmaceutical industries, especially in developing countries 8 . S. platensis has shown ability of adaptation to quite different habitats and colonizes harsh environments, where life is exceedingly difficult for other organisms. For example, in the lakes containing salt concentrations > 30 g L -1 , the cyanobacterial population became practically monospecific and Spirulina was the only organism present in significant quantities. Indeed S. platensis was found in waters containing from 20 to 270 g L -1 of salt, but growth seemed to be optimal at salt concentrations ranging from 20 to 70 g L -1 and it is possible that the population of S. platensis found at the highest salt concentrations, such as in temporary ponds just before drying, was that of the cyanobacterial biomass established when the concentration of salts was much lower 7 . S. platensis are thermophilic algae with optimal growth temperature between 35 to 37 °C 8 .
When S. platensis was cultivated outdoor under high natural sunlight and salinity-stress, its production was usually accompanied by photosynthesis photoinhibition. Furthermore, it was suggested that salinity-stress enhances photosynthesis photoinhibition in green alga Chlamydomobas reinhardtii 9 . Thus, S. platensis incorporates into a suitable photo-bioreactor configuration that can enhance photosynthesis by increasing growth conditions and controlling exposure of S. platensis to environmental factors, as well as suitable for greenhouse gases attenuation, particularly converting CO2 into biomass in which carbon is biofixed and incorporated into carbohydrates, lipids and proteins 10,11 . The microalgae biomass produced can also be used for various applications, such as biofertilizer, soil conditioner, and biofuels production 8,12 . However, more research and development are necessary on strain of microalgae selection, acclimation, and adaptation with regards to salt tolerance and the impact of other environmental parameters. In this paper, were studied the impacts of light intensity, temperature, and inlet CO2 concentration on the specific growth rate of S. platensis and S. platensis adapted to salinity-stress during biofixation of CO2 in photobioreactors at laboratory scale.

Algal strain and cultivation conditions
Spirulina platensis from the American Type Culture Collection (ATCC) strain 53844 was cultivated in Zarrouk's culture 10 fresh medium described in Tab. 1, adjusted (autoclave medium) to a final pH 9.0 ± 0.2. The stock culture was maintained in a 250 mL Erlenmeyer flask containing 50 mL of the medium at 20 °C under 60 µmol m -2 s -1 of light intensity, and 16/8 h day/night cycle. Every week the culture was transferred to a 500 mL flask containing the respective fresh medium and acclimatized to 0.5% of CO2 mixed with air. Then, for further tests, the acclimated culture was transferred into each photobioreactor and the CO2 concentration was increased gradually by bubbling CO2 (2.5%, 5%, 7.5% and 10%) for 24 h before starting the test at a flowrate of 0.05 L m -1 .
Adaptation of S. platensis to high salinity, an inoculum of microalgae S. platensis were cultivated in a modified Zarrouk's medium 10 Figure 1 shows the graphical abstract scheme of the experimental set up at laboratory scale. The input CO2 gas concentration was controlled by mixing CO2 and air directly to each photobioreactor of 2 L working volume (WV) glass (Pyrex). The input gas mixture was connected by PVC tube (d=0.15 cm). The flowrate of CO2 was measured by a flowmeter (Colepalmer) and a sintered sparge (porous air diffuser) placed into the photobioreactor for bubbling the into the biomass. The exhaust gas from each photobioreactor was connected by a PVC tube of 0.15 cm diameter and the flowrate measured by a flowmeter. Each photobioreactor was inoculated with 200 mL of precultured S. platensis and filled with 1800 mL nutrient solution prepared and mixed 24 h in advance to reach about 0.5 g L -1 concentration of suspension biomass. An experimental factorial design was proposed to study the three factors: CO2 concentration, temperature and light intensity, ranging at five levels using the same experimental procedure, first for S. platensis culture followed by S. platensis adapted to high salt concentration. Five parallel photobioreactors were set up in the same chamber, the initial concentration of the cell biomass was approximately 0.5 g L -1 , the photobioreactors operated at the same temperature and light intensity, but the cell biomass in each photobioreactor received different CO2 concentrations (0.5, 2.5, 5.0, 7.5 and 10%) under continuous bubbling of CO2 at a rate of 0.5 L m -1 , and all run lasted 12 days.
The photobioreactors temperature was maintained constant by immersing the photobioreactors in acrylic open water baths (0.46 x 0.25 x 0.8 m) with immersion circulator analog controller (Isotemp 2100). The light intensity was generated by cool white fluorescent tubes (General Electric 40w to 80w) and measured using a quantum sensor (model LI-190SA), connected to a quantum/radiometer/photometer light meter (model LI-250A, Li Cor Inc. Lincon, NE, USA). The quantum sensor was configured to make measurements of photon flux density (PDF) in the PAR (Photosynthetic Active Radiation, 400-700 nm). The source of CO2 was provided by Airgas Specialty Gases New Jersey, 200 and 300 cubic feet volume cylinders of CO2 concentrations of 0.5, 2.5, 5.0, 7.5 and 10% balance with air, the concentrations certified by the vendor.

Analytical determinations
The concentration of microalgae was measured by the method of filtration, being 10 mL suspension of cell biomass filtered on membrane filter (GF/C filters 1.2 µm, d=47 mm). After dried filters at 80 °C for 24 h, cell biomass weights were determined until achieved constant weight over time. The ratio of carbon in the dry cell biomass [Cc] was determined by ignition, dried cell biomass was ignited at 500 °C in a Thermolyne Furnace (model 62700, Thermolyne Corp. Dubuque, IA) to estimate the dry weight biomass. The cell biomass concentration for both microalgae was also determined as the changes in optical density (OD). The OD of the suspension algal biomass was measured at 680 nm (S. platensis) as absorbance, using a spectrophotometer (UV-VIS Shimadzu-1700). The cell dry weight of S. platensis and optical density (OD680) were established by linear regression (dry cell biomass, g L -1 = 0.477 × OD680 + 0.376; R 2 = 0.957; p = 0.01). Likewise, for S. platensis adapted and cell biomass g L -1 =2.35 × OD680 + 0.32; R 2 =0.94, p=0.01. Triplicate samples of the cell biomass were collected every day until reached maximum microbial growth (4-5 g L -1 ), depending upon the experimental conditions some photobioreactors were shut down earlier than others. The pH of the suspension of cell biomass was measured with a pH meter (Acumet ABIS Plus) calibrated with standard pH solutions of 4, 7 and 11. Analysis of CO2 input/output of each bioreactor was measured with a Gas Chromatograph Shimadzu 17A with TDC detector, gas samples were taken in plastic bags. The CO2 concentration was also monitored analyzing CO2 directly in the gas stream off the bioreactor using colorimetric gas detection tubes RAE systems.

Determination of growth rate and kinetic parameters
Assuming that the microalgae growth can be modeled by a first order dynamic equation: Integrating and re-arranging Eq. 1, the growth coefficient also called specific growth rate (μ, d -1 ) can be calculated using the Eq. 2: where X1 and X2 were the microalgae concentration (g L -1 ) on days t1 and t2, respectively 13 . The biomass productivity rate, also called linear growth (p), is estimated according to Eq. 3: where p (g L -1 d -1 ). Since no organic carbon source is available in culture medium, the CO2 biofixation rate can be indirectly calculated by the carbon content and biomass productivity rate, according to Eq. 4 13 Therefore, the rate of CO2 biofixation per initial inoculation mass of microalgae can be determined by Eq. 5 rCO2 = RCO2 / Xo (5) where RCO2 [gCO2 L -1 d -1 ] is the biofixation rate and rCO2[gCO2 g -1 dry cell].
The average cell carbon content Cc [gC g -1 dry cell] ratio measured experimentally was 0.59 g carbon g -1 dry cell weight, according to the measurement using an elemental analyzer, and MCO2 MC -1 represents the molecular weight of CO2 and C 4,14 , respectively.
Empirical microbial growth kinetic models were explored to describe the impact of environmental factors on specific microbial growth of S. platensis and S. platensis adapted. Mönod model (Eq. 6).

S Ks
where Ks is the Mönod kinetic constant and S, CO2 concentration or light intensity. When the substrate inhibits microbial growth at high concentrations, an optimum at which the highest specific growth rate occurs in this case, the Mönod model can be modified by Andrews model (Eq. 7).
where Ks and S are the same meaning as Mönod, and KI is the inhibition constant.
The effect of the temperature on the maximum specific growth rate is based on Arrhenius model (1889), which implies an exponential increase in growth rate of the cells with rising the temperature. However, it is well recognized that the Arrhenius model fails once the temperature approaches the value of optimum activity, because it cannot represent the decline in rates at higher temperatures. Due to this limitation, alternative models have been proposed, which can predict the decline in rate following the optimum. Mayo model (Eq. 8) modified the Arrhenius equation based on the premise that the active fraction of the enzymes involved in the growth limiting reaction deceases when the temperature exceeds the optimum, this expression is also able to predict a decline in the maximum specific growth rate when the temperature exceeds the optimum 15 .
where A' and k are constants, E1 is the activation energy for cellular multiplication, and E2 is the activation energy for the thermal denaturalization process.

Statistical analysis
The experimental results were evaluated by comparing the specific growth rates of CO2 biofixation by S. platensis and S. platensis adapted to salinity  Figure 2 gives the growth curve of dry cell mass of S. platensis cultivated in photobioreactors at normal salinity concentration for 12 days cultivation period at 2.5% CO2 concentration, 25 °C and µmol m -2 s -1 dry cell mass achieves 4.3 g L -1 , which slowly declines when cultivation conditions change by increasing CO2 concentration (5%), temperature (30 °C) and light intensity (150 µmol m -2 s -1 ). The dry cell mass growth of S. platensis continued to decline even more and photosynthesis of S. platensis is inhibited at cultivation conditions of CO2 concentration (10%), temperature (40 °C), and light intensity (200 µmol m -2 s -1 ). Figure 3 shows the impact of salinity when S. platensis adapted to100 g L -1 NaCl (1.71 mol L -1 ) is cultivated in the same way of S. platensis, for example, dry cell mass achieves 3.2 g L -1 for a 12 days cultivation period at 2.5% CO2 concentration, 25 °C and 100 µmol m -2 s -1 , salinity-stress is usually accompanied by photoinhibition of photosynthesis 10 . Figure 4 illustrates the UV-visible spectrums of S. platensis and S. platensis adapted. An analysis of the spectrum from 600 to 800 nm shows the strongest band centered near 680 nm for S. platensis, which decreases and displaces the maximum to 675 nm for S. platensis adapted, due to the impact of salinity.     Table 3 shows that during the 12 days cultivation period, the highest µ and R values of S. platensis (0.278 d -1 , 25.1 gCO2 m -3 h -1 ) cultivated at 2.5% CO2 was significantly higher (p < 0.0012) than 5% CO2 or higher CO2 concentration runs. In the same cultivation mode, the µ and R of S. platensis adapted decreases (0.15 d -1 , 13.9 gCO2 m -3 h -1 ) because of the effect of salinity, in addition to the high CO2 concentration of 5 to 10% runs, according to the results of Tab. 3. There was no significant difference between μ and R values of S. platensis between 20 and 25 °C runs, but there were significant differences (p < 0.0012) between μ and R values of S. platensis adapted in the same range of temperatures, these values were significantly lower when the cells were exposed to higher temperatures from 30 to 40 °C, respectively. When S. platensis were exhibited to 150 µmol s -1 m -2 μ and R values achieved higher values elevated light intensities e.g. 0.18 d -1 , 16.2 gCO2 m -3 h -1 than corresponding values at 100 and 60 µmol s -1 m -2 . But there was no significant difference between μ and R of S. platensis for 100 and 150 µmol s -1 m -2 runs. Conversely, the μ and R values of S. platensis adapted, increased as the cells were exposed to lower light intensity from 60 µmol s -1 m -2 (0.15 d -1 , 13.2 gCO2 m -3 h -1 ) to 200 µmol s -1 m -2 (0.05 d -1 , 4. 96 gCO2 m -3 h -1 ). But there were no significant differences between μ and R values of S. platensis adapted cultivated with 60 and 100 µmol s -1 m -2 , respectively, indicating that S. platensis adapted to salinity stress is less tolerant to growth at elevated light intensity and temperature, suggesting a decrease in photosynthetic activity of S. platensis adapted (Tab. 2).

Effect of salinity on the kinetics of S. platensis and S. platensis adapted
The relationship between the µ value and environmental cultivation parameters such as CO2 concentration, temperature and light intensity in terms of experimental kinetic models (Monöd, Andrews and Mayo) and the impact of salinity is illustrated for S. platensis (Tab. 3) and S. platensis adapted (Tab. 4), respectively. The impact of light intensity on μ of S. platensis, in principle is described by Mönod well in the range of 60 to 150 µmol m -2 s -1 Eqs. 9 and 10 (Tab. 4). The maximum specific growth (µmax) estimated (0.44 d -1 ) is depleted as CO2 concentration and temperature were increased. Under these conditions, µmax decreased to 0.22 d -1 and Andrew's model described better the kinetics of S. platensis exposed to higher light intensity (Eqs. 11, 12 in Table 4 and Fig.  5). In fact, a decline in the photosynthetic activity was observed after four days of experiment run under 10% CO2, 200 µmol m -2 s -1 , and 40 °C, suggesting these values as potential limiting factors of microalgae growth and photoinhibition 10 . The influence of light intensity on the kinetics of S. platensis adapted was stronger than the impact on the µ of pure culture S. platensis (see Tab. 5: Eqs. 19, 20 and Fig. 5), Andrew's model described this impact, µmax dropped from 0.31 d -1 for 5% CO2 and 25 °C to 0.22 d -1 for light intensities higher than 150 µmol m -2 s -1 and photoinhibition occurred when CO2 and temperature increase to 10% and 40 °C, suggesting that in addition to light intensity salt stress affect its photosynthetic activity and may inhibit completely at 10% CO2 and 40 °C. These results agree with Zeng et al. 10 , who reported that the effect of salinity stress was stronger when cells of S. platensis were grown under higher light intensity 200 µmol m -2 s -1 and showed lower capacity of recovery in the photosynthetic activity after photoinhibition than lower light intensity (100 µmol m -2 s -1 ) grown cells. This was attributed as a result of the fact that stressed cells have lower protein synthesis capacity and thus a slower repair mechanism.
The effect of CO2 concentration on the kinetics of S. platensis, Eqs. 13, 14 and 15 (Tab. 4) and S. platensis adapted Eqs. 21 and 22 (Tab. 5), is depicted by Andrew's kinetic model and is illustrated in Fig. 6. The µmax value estimated for 2.5% CO2, 25 °C and 150 µmol m -2 s -1 , declined from 0.67 d -1 to 0.41 d -1 when CO2 concentration increased in the range of 5 to 10% and the temperature and light intensity rose to 40 °C and 200 µmol m -2 s -1 , respectively, suggesting that CO2 is a limiting factor that inhibited S. platensis growth particularly at 200 µmol m -2 s -1 and 40 °C. An optimal µmax value for S. platensis adapted to salinity stress of 0.58 d -1 determined under the following environmental parameters 2.5 % CO2 (25 °C and 60 µmol m -2 s -1 ) declined to 0.24 d -1 when CO2 concentration rose more than 5% (40 °C and 200 µmol m -2 s -1 ), suggesting that CO2 is a limiting factor of CO2 biofixation in combination with high salt concentration. The effect of temperature on the kinetics of S. platensis (Eqs. 16, 17 and 18 in Tab. 4) and S. platensis adapted to salinity stress (Eq. 23 in Tab. 5) and depicted by Mayo's model (Fig. 7), the optimum temperature for cultivation of S. platensis occurred around 25 °C, 150 µmol s -1 m -2 and 2.5% CO2 but it declined slowly as temperature increased along with CO2 concentration and light intensity to a minimum µ values 10% CO2 and 200 µmol m -2 s -1 . Likewise, as temperature decreased less than 15 °C, µ values decreased, indicating that the temperature is a limited growth factor under the influence of high CO2 concentration and light intensity in spite the fact that S. platensis was characterized as thermophilic microalgae. The optimum temperature of S. platensis adapted occurred at around 25 °C, for 60 µmol m -2 s -1 and 2.5% CO2 but declined slowly as the temperature increased. Despite the fact that S. platensis has been characterized as a thermophilic microalgae, the high salt concentration may have caused strong impact on the cells stress supported by high CO2 concentration (10%) and light intensity (200 µmol m -2 s -1 ) minimize µ values. Likewise, as temperature decreased less than 15 °C, µ values decreased, suggesting that the temperature is a limiting growth factor. In addition, other factors were shown, such as CO2 concentration and light intensity in combination with high salt concentration affect CO2 biofixation rate.  Waste stream flue gases emitted from stationary sources such as power plants, industrial boilers, refineries and others, using fossil fuels for combustion and energy production produce from 4 -14% CO2 and air in a closed space such as a space station or a submarine < 1% CO2 2 . It has been reported that microalgae present one of the few technologies for the capture and utilization of CO2 17 . The results indicate that both S. platensis and S. platensis adapted to high salt concentration can be used for CO2 biofixation of flue gas emitted from stationary sources. In fact, maximum daily evaluation of the efficiency CO2 removal by S. platensis during experiments up to ten days, achieved 92% at 0.5% CO2, 99% at 2.5% CO2, 84% at 5% CO2, 89% at 7.5% and 88% at 10% CO2, respectively. Corresponding values for S. platensis adapted being 96% at 2.5% CO2, 90% at 5% CO2, 78% at 7.5% CO2 and 73% at 10% CO2, respectively (See Appendix).
Small pH changes of S. platensis cell suspension observed during experiments from an initial pH (8.9) in the presence of CO2 ≤ 2.5% and temperatures ≤ 25 °C, after 24 h pH rose to 9.4 ± 0.1 and S. platensis adapted from an initial pH of 8.5 increased to 8.9 ± 0.1. However, in cultivation of S. platensis at CO2 > 5% and temperatures > 25 °C the pH increased to o 9.1 ± 0.2 and S. platensis adapted to 9.2 ± 0.2, after 24 h. These results indicate that there was no formation of carbonic acid but no abrupt decline in pH due to S. platensis and S. platensis adapted were able to metabolize CO2 because both microalgae were previously acclimatized to CO2. S. platensis and S. platensis adapted were able to metabolize CO2 because both microalgae were previously acclimatized to CO2. Therefore, neither was formation of carbonic acid nor decline in pH.
In a three-stage serial tubular photobioreactor were cultivated Scenedesmus obiquus and Spirulina sp. at 30 °C. It was found that, for Spirulina sp. the µmax was 0.44 d -1 , 9.2 g m -3 h -1 with 6% CO2, maximum daily CO2 removal efficiency was 53.3% for 6% CO2 and 45.6% for 12% CO2, the corresponding values for S. obliquus being 28.1% for 6% CO2 and 13.6% for 12% CO2 runs 14 . Yun et al. 18 estimated 26.0 gCO2 m -3 h -1 biofixation rate value when Chlorella vulgaris was cultivated after adaptation in 5% CO2 in wastewater supplemented with nutrients and without pH control at 15% CO2, this value is comparable with the findings of this research and suggests that elevated CO2 concentration may exert effects on the photoinhibitory behavior of the microalgae to different extents according to species 19 . The efficiency of CO2 biofixation by these microalgae strains may have been due to its physiological conditions, such as potential of cell growth and ability of CO2 metabolism. S. platensis was cultivated under different light intensities (100-200 µmol m -2 s -1 ) at 35 °C and adapted to salinity stress up to 0.75 mol L -1 . It was reported that the cells grown in higher light intensity are less tolerant to salinity stress than those grown in lower light intensities, suggesting that salt stress enhances photoinhibition of photosynthesis through a direct effect on PSII reaction center 10 . The results demonstrated that S. platensis adapted to high salinity media 1.71 mol L -1 cultivated at 200 µmol s -1 m -2 , 40 °C and high CO2 concentrations (7.5 and 10%) inhibit microalgae grow, show low photosynthetic activity and consequently photoinhibition 20 . Photoinhibition occurs when the photon flux absorbed by chloroplasts is extremely high, so the concentration of high energy electrons in the cells is too elevated to be consumed in the Calvin cycle. These electrons react with water to form hydrogen peroxide, which is highly harmful to subcellular structures and the cell itself, indicating that salinity enhances photoinhibition of photosynthesis through a direct effect on PSII reaction center, the reason for declination of PSII activity of cells under salinity stress remains open. Moreover, they believed that salinity stress induced damage or inactivation of PSII reaction center as it is in the case of photoinhibition of photosynthesis 10 .

Conclusions
This study shows the potential CO2 biofixation by S. platensis and S. platensis adapted to high salinity, 1.71 mol L -1 NaCl at laboratory scale in photobioreactors. In general, a better rate of CO2 biofixation was achieved by S. platensis as indicated by its kinetic parameters and efficiency CO2 removal, when compared to S. platensis adapted. The impacts of light intensity, CO2 concentration and temperature on the specific growth rate followed the Mönod, Andrews and Mayo kinetic models. For S. platensis the highest dry cell mass concentration was 4.1 g L -1 , cultivated at 2.5% CO2, 25 °C and 150 µmol m -2 s -1 , the highest rate of biofixation 25.1 gCO2 m -3 h -1 and the maximum specific growth (μmax) was 0.44 d -1 -0.67 d -1 . The corresponding values for S. platensis adapted were 3.2 g L -1 at 2.5% CO2, 25 °C and 60 µmol s -1 m -2 , 18 of salinity in combination with environmental grow factors such as elevated CO2 concentration, light intensity and temperature may exert effects on the photoinhibitory behavior of S. platensis adapted, provoking a CO2 biofixation rate depletion. Therefore, the photosynthetic biofixation of CO2 S. platensis and S. platensis adapted showed optimum values at 2.5% CO2 and 25 °C, and more sensitivity to light intensity than S. platensis, suggesting that salinity enhanced photoinhibition of photosynthesis. The efficiency of CO2 removal by S. platensis achieved 99%, whereas for S. platensis adapted 96%, both at 2.5% CO2 concentration and small pH changes exhibited both microalgae cell suspension.