Periodic CO2 Dosing Strategy for Dunaliella salina Batch Culture

A periodic CO2 dosing strategy for D. salina 19/30 batch culture is proposed. A model of periodic CO2 dosing including dosing time calculation, dosing interval estimation and final chlorophyll yield prediction was established. In experiments, 5% CO2/95% N2 gas was periodically dosed into D. salina culture. Two different gas dosing flow rates were tested. The corresponding dosing time for each flow rate was estimated via the model (10 min·d−1 for 0.7 L·min−1 and 36 min·d−1 for 0.3 L·min−1). Daily pH measurements showed that the pH of these cultures dosed periodically was always kept between 7.5 and 9.5, which highlights that periodic gas supply can maintain a suitable range of pH for microalgal growth without expensive buffers. Notably the culture dosed for set daily intervals was seen to have similar growth to the culture supplied constantly, but with much higher CO2 capture efficiency (11%–18%) compared to continuous dosing (0.25%). It shows great potential for using periodic gas supply to reduce cost, wasted gas and energy use.


Introduction
Currently, one emerging application of microalgae is for the fixation of CO2 [1], for it may offer a way to reduce the levels of unwanted CO2 whilst also allowing the production of useful by-products from the algae such as bio-oils, chemicals, fatty acids and substances, etc. [2]. However, the large-scale cultivation of microalgae still faces many problems and barriers. These include the energy required to continuous mix, dewater and process the algal biomass, and the large areas of land needed for such projects [3]. CO2 dosing is one of the concerns highlighted in this study. The major problem with CO2 being dosed into microalgae cultures is that conventionally the bubbles used are fairly large. As Zimmerman et al. [4] discuss, such bubbles will reduce the interfacial area between bubbles and the liquid, and will reduce the overall mass transfer for both CO2 dissolution and O2 stripping. To counteract the resulting low mass transfer, large flow rates of CO2 enriched gas are often continuously bubbled into the cultures. While the mixing effect on the culture is beneficial for ensuring the even distribution of light and nutrients, the intense turbulence produced by large flow rate can damage the algal cells within the culture and reduce the productivity [5]. A further disadvantage of using high flow rate, especially with low mass transfer, is that most of the gas bubbled into the culture will pass through and be wasted. Carvalho and Malcata [6] agree that when CO2 is bubbled into algal cultures, the mass transfer is not particularly effective and considerable gas is wasted, adding to operational costs. Even with some high mass transfer dosing techniques (e.g., Dissolved Air Flotation systems), hypothetically, continuous dosing is still not a wise option, because when the concentration of dissolved CO2 has reached an equilibrium value, the gas-liquid mass transfer process stops. Any additional input beyond this point would not increase the total amount of dissolved CO2, but cost energy and waste CO2.
An innovative CO2 microbubble-dosing technology was studied by Ying et al. [7][8][9], which has proved that the microbubble-dosing technology can enable a higher CO2 mass transfer and consequently lead to a higher microalgal growth rate and a greater CO2 capture efficiency. Nonetheless, the operational parameters (e.g., dosing time, dosing interval and flow rate, etc.) for microbubble-dosing technology still need to be engineered. In this study, an optimal periodic CO2 dosing strategy is proposed and a model established based on D. salina (19/30) cultures. The main hypothesis for this study is that by using a microbubble driven airlift bioreactor, high mass transfer can be attained in a batch culture by supplying gas periodically (with little CO2 wasted and less energy cost) and achieve similar algal growth compared to when the gas is supplied continuously.

Model of Periodical CO2 Dosing
For an optimal periodical dosing strategy, three major principles need to be followed. Firstly, despite the exclusion of buffer solution and pH auto-regulating system, the pH of the culture needs to be controlled in a suitable range by periodic CO2 dosing. Secondly, the dosing time should only be long enough for CO2 to reach its equilibrium concentration. Meanwhile, the equilibrium pH (corresponding to CO2 equilibrium concentration) is expected to be the lower limit of the suitable pH range for the microalgal species being utilized. Thirdly, the time period without dosing (dosing interval) should ensure that the microalgae use up the dosed CO2, whilst ensuring that the pH increase to the upper limit is within the suitable pH range. Following these three principles, a culture with periodic dosing, compared to one with continuous dosing, should have sufficient (but not excess) CO2 and an optimal pH range to support optimal growth (achieving similar growth as with continuous dosing), while with minimal amount of CO2 wasted and less energy input. Based on these three principles, the dosing time, dosing interval and final algal yield can be estimated.

Estimation of Dosing Time
Assuming the suitable pH range for the culture of a particular microalga species is given between A and B, of which the corresponding concentration of [CO2] can be calculated by Equation (1)  (mol/L) 10 2 10 To control the culture in a target pH range the amount of CO2 needed to be transferred to the medium can simply be estimated as the difference between dissolved CO2 level at pH=B and pH=A (assuming the CO2 uptake rate is negligible compared with CO2 gas-liquid mass transfer rate). The dosing time is thereby calculated as the amount of CO2 needed to be transferred to the medium divided by the average CO2 mass transfer rate, shown in Equation (2), where ν'MTR represents CO2 average mass transfer rate (mol·L −1 ·min −1 ) and can be calculated as Equation (3). The derivation of Equation (3) was explained in Ying et al. [9]. Finally, assuming the pH before dosing was controlled at B which gives the initial dissolved CO2 level as [CO2]pH=B, the optimal valid dosing time needed to drop the pH from B to A can then be estimated through Equation (4), which is obtained by combining Equations (2) and (3). As long as the suitable pH range for a particular type of algae is given and the KLa for a certain dosing condition is known, the optimal dosing time can be estimated. 2 From the previous study carried out by Ying et al. [9], it was found that KLa estimation is more accurate based on the changes in [CT], especially when pH > 8.4. Therefore, Equation (4) can also be written in terms of [CT], shown as Equation (5).

Estimation of Dosing Interval
The dosing interval here is defined as the time period without gas dosing. During this period, pH increases gradually because of the uptake of CO2 by microalgae, until the pH achieves the upper limit of the suitable range, then dosing needs to be started again. Therefore, the effective estimation of dosing interval is crucial for periodic dosing, either too long or too short could cause the pH to exceed the upper or lower limit of the suitable range and adversely affect the algal growth.
The simplest way to estimate the dosing interval is to divide the amount of CO2 expected to be absorbed by the CO2 uptake rate. However, the instantaneous growth rate differs with the concentration of the algae [12], which leads to changes in CO2 uptake rate. Instead of instantaneous CO2 uptake rate, the average CO2 uptake rate for the whole active growth period is therefore used to simplify the estimation of dosing interval Equation (6).
Ying et al. [9] reported that when the pH was less than 8.4, the changes in the amount of total carbon almost all come from the changes in dissolved CO2 [13], however, when pH was more than 8.4, the changes in dissolved CO2 cannot fairly represent the CO2 uptake by algae, as both HCO3 − and CO3 2− would generate dissolved CO2 to compensate for the consumption of CO2. In other words, the amount of CO2 consumed by algae should be more than the changes in dissolved CO2, as HCO3 − and CO3 2− would also contribute to the amount of CO2 consumption. Therefore, the changes in total carbon [CT] should be considered instead of the changes in [CO2]. The [CT] can be calculated by Equation (7) [9]. Equation (6) should be converted into Equation (8). The dosing interval can then be estimated as long as the average CO2 uptake rate is known.
Since the periodic dosing strategy is proposed to achieve similar algal growth to when gas is supplied continuously, the average CO2 uptake rate is assumed to be the same as in the culture with continuous or excessive CO2 dosing. According to the information from previous D. salina cultures [7], the correlations for CO2 uptake rate versus algal biomass concentration (measured as chlorophyll content) was described by Equation (9) [7], and the relation between total chlorophyll content increase and total CO2 uptake was given (based on the cultures with excessive gas dosing) as Equation (10) For the same time period, Equation (10) can be transformed into Equation (11): The average CO2 uptake rate can be fairly described as: where tc is the selected culture time period (tc2-tc1), beginning from tc1 (the start of log growth phase) and ending by tc2 (the end of log growth phase).
Assuming the chlorophyll content ([Chl], mg·L −1 ) is equal to the initial concentration ([Chl]0, mg·L −1 ) plus the amount of its increase (ν'Chl × t), The average CO2 uptake rate can then be obtained by solving Equation (9), Equation (11)-(13), which gives: The dosing interval is then given by: Theoretically the dosing interval is better to be shortened as the algae grows, which may in practice increase the complexity of the time control process. Using a constant dosing interval through the whole log growth period can simplify the operating process. By doing so, one of the major concerns is that the pH level may exceed the upper limit of a target range. However, one magnitude of difference in the concentration (mol·L −1 ) of dissolved CO2 only changes the pH by one unit [7], while the CO2 uptake rates for D. salina cultures are in the range of 10 −4 to 10 −3 mol·L −1 ·d −1 , therefore, the pH value may only increase over the upper limit of a target range by no more than 0.1 unit. Plus empirically the target pH range for algal culture can be slightly narrowed down during the dosing interval calculation in order to prevent the pH value exceeding the real upper limit. The second concern is dissolved oxygen accumulation. According to the basic photosynthetic reaction equation, the O2 generation rate equals to CO2 uptake rate. Therefore, 10 −3 mol·L −1 ·d −1 of CO2 uptake rate could result in around 32 mg·L −1 of daily O2 accumulation (300% over saturation with respect to air), which will diminish the rate of photosynthesis [14]. Thus, the dosing interval has to be limited to no more than 1 day so that Dissolved Oxygen (DO) can be removed by CO2 dosing in time. In terms of O2 stripping, it takes less than 10 min to reduce the dissolved oxygen from 32 to 0.03 mg L −1 by microbubble (d32 = 388 μm) dosing of 5% CO2/95% N2 under 0.7 L·min −1 , according to the previous mass transfer study [7].
Statistically, from previous D. salina studies [7], the active growth phase usually lasted for 8 days (due to the light limitation). According to the literature, D. salina can tolerate a pH range of 5.5 to 10 [15]. A target pH range 7.5-9.5 was selected in this study. The dosing interval for D. salina cultures was estimated via Equation (15) to give approximately 1.5 days (based on Chl0 = 5 mg·L −1 and tc = 8 days). Conservatively, the dosing interval of 1 day was suggested for best results in practice, so that DO can be removed by CO2 dosing in time.

Prediction of Final Concentration of Chlorophyll Content
Once the dosing time td and dosing interval ti are known, the total dosing time (tTotal) through the culture period tc can be calculated as: td × tc/(td + ti). The total amount of CO2 uptake can be predicted by Equation (16): The final chlorophyll content can be estimated by Equation (17), By combining Equation (11), Equations (16) and (17), it gives: Equations (16) and (18)

Experimental Results
A set of D. salina batch cultures were carried out to test the hypothesis that sufficient CO2 can be attained in a culture by "micro-bubbling" gas periodically (with little CO2 wasted and less energy cost) whilst similar algal growth can be achieved compared to when the gas is supplied continuously, but with higher CO2 capture efficiency. Figure 1 shows the daily chlorophyll content of D. salina cultures supplied continuously and periodically with 5% CO2. As can be seen, the growth of these cultures, indicated by their chlorophyll content, appears to be fairly similar. This strongly supports the idea that D. salina growth is proportional to the total amount of CO2 that has been effectively transferred from gas phase into liquid phase, while extra CO2 dosing beyond the valid range does not improve the productivity. The final chlorophyll contents for the cultures with different dosing conditions were expected to be the same, as the total amount of CO2 mass transfer was kept identical. The contrast between the growth of the cultures fed with 5% CO2 and the control cultures can also be seen from Figure 1. Unsurprisingly the chlorophyll content of the control cultures remained much lower than other cultures, and a similar phenomenon was also observed in the study of Ying et al. [7].
Arguably the most important finding is the comparison that can be drawn between the growth of the D. salina cultures supplied continuously and daily with gas. It appears that there was no significant difference between the two types of culture. The daily chlorophyll content of continuously dosed culture seemed slightly higher between day 4 and day 11, which indicates the D. salina grew a bit faster under continuous dosing (approximately 1 day in advance of the periodic dosed cultures). However, the cultures engaged with periodic dosing model were still competitive to the continuously dosed culture, as they achieved a similar level of final chlorophyll content although with one day of delay. Regarding the CO2 capture efficiency, the cultures with different dosing conditions were compared ( Table 1). As can be seen, by applying a periodic dosing strategy, CO2 capture efficiency achieved is about 10%-20%. It is expected that the capture efficiency could be further enhanced by improving the CO2 mass transfer (e.g., by further reducing the microbubble size). In contrast, with continuous dosing, capture efficiency was only 0.25% of CO2 supplied, which indicates that most of the CO2 was wasted rather than been captured. With a view to using microalgae for CO2 sequestration, this will mean not only wasted energy to dose CO2, but also any CO2 that was prevented from entering the atmosphere by fixation in the algae culture, will be greatly exceeded by the amount of CO2 passing straight through the culture into the atmosphere. Therefore, this result shows the potential for both economic and energy savings by adopting a periodic dosing strategy, as it appears that similar algal growth to a culture supplied continuously with CO2 can be achieved with periodic dosing, but with minimal CO2 waste and minimal energy consumption on dosing. conditions. The cultures with 0.7 L·min −1 of periodic dosing were conducted in parallel, as were the control cultures. Therefore, the daily chlorophyll content for these two dosing conditions shown in this graph is the average value, with the error bars given separately.  Additionally, the pH control achieved using periodic dosing was also observed during these experiments (Figure 2). The pH in the culture supplied periodically with gas was maintained in the target region of 7.5-9.5 without the use of expensive buffers. This also indicates the periodic dosing model for dosing time and dosing interval estimation are accurate, so that the pH was controlled in an expected range. These results agree with the previous studies by Ying et al. [7] who conducted a similar experiment culturing D. salina in Airlift Loop Bioreactors (ALBs) proved 30 min·d −1 of gas (5% CO2, 95% N2).
Finally, the predicted final concentrations of chlorophyll content for different periodic dosing conditions Equation (18) were compared with the experimental results, shown in Table 2. The errors between theoretical values and experimental values were about 2%-3%, which indicates the accuracy of Equation (18).

Figure 2.
Daily pH values for D. salina cultures supplied periodically with 5% CO2. There are two pH values for each day, a higher one and a lower one, representing the pH values before and after CO2 dosing, respectively.

Experimental Setup
The experimental setup is illustrated in Figure 3. Samples of the D. salina were pre-cultured in shake flask (100 mL culture in 250 mL flasks) in a 25 ± 2 °C growth room. The growth medium composition is identical to the one described in Ying et al. [7], shown in Table 3. For the start of the main culture, 50 mL of this D. salina was added to 2.5 L of fresh culture medium in ALB. The dosing time (td) and dosing interval (ti) were estimated based on Equations (5) and (15) and applied for the cultures engaged with periodic dosing strategy (No. 2, 4 and 5). pH was measured for each culture daily (for the culture dosed periodically, pH was measured twice per day, before and after dosing). 15 mL of sample for each culture was taken after gas dosing, followed by topping up the culture with 15 mL of fresh medium (For flask cultures, a sterilized glass stick was inserted into the culture for a proper stirring. Samples were taken after that.). The chlorophyll content for each sample was determined by measuring each sample's optical density for wavelengths of 645 and 663 nm using the same method to that used in Zimmerman et al. [16] and Ying et al. [7].   Table 3.

Estimation of the Dosing Time and Dosing Interval
In Table 4, the dosing time for 0.3 and 0.7 L·min −1 were calculated based on Equation (5). The detailed calculation was shown as follows.
The KLa for CO2 mass transfer in the real algal culture was previously determined to be 0.044 min −1 for 0.3 L·min −1 of bubbling flow rate and 0.174 min −1 for 0.7 L·min −1 [9]. Finally, td was calculated to be about 36 min for 0.3 L·min −1 of bubbling flow rate and 10 min for 0.7 L·min −1 .
For dosing interval estimation, it can be calculated by Equation (15). Based on the assumption that Chl0 = 5 mg·L −1 and tc = 8 days, ti is calculated to be 1.5 days. Conservatively, 1 day of dosing interval was used in the experiment so that DO can be removed by CO2 dosing in time.

Conclusions
A periodic CO2 dosing strategy for D. salina culture is proposed, with a model of periodic CO2 dosing including dosing time calculation, dosing interval estimation and final chlorophyll yield prediction established. The cultures applying periodic CO2 dosing strategy provide a similar productivity to the culture with continuous dosing, but with a greater CO2 sequestration efficiency. Due to the time limitation, only two different gas dosing flow rates were tested in this study. More flow rates can be tested to explore the relation between dosing flow rate and CO2 capture efficiency in the future studies.