Effects of Process Variables in Semi-continuous Cultivation of Chlorella Minutissima on the Biomass Composition

The biomass composition of Chlorella minutissima 26a was evaluated in different conditions for the microalgae cultured in landfill leachate in airlift photobioreactor operated in semi-continuous mode. The effects of the following factors in cells composition were evaluated: landfill leachate concentration (LC), CO 2 percentual flow added in the gas stream (GP), aeration flow in the reactor (AF) and feed flowrate (FR). A Taguchi L 9 orthogonal arrangement was used to evaluate the effects of those four factors on the following response variables: Biomass Productivity (BP), Lipid Productivity (LP), Carbohydrate Productivity (CP) and Proteins Productivity (PP). Results showed that the microalgae had high selectivity to produce proteins, reaching a maximum content of 69,60% in the following conditions: LC: 10%, GP: null, AF: 0,30 vvm and FR: 0.9 µmax vvm. The highest contents of lipid and carbohydrate were 17.4% (LC: 15%, GP: null, AF: 0,45 vvm and FR: 0.7 µmax vvm) and 11.6% (LC: 10%, GP: 15%, AF: 0,15 vvm and FR: 0.7 µmax vvm.), respectively, both values achieved at the two experiment with the highest values of specific growth rate (0.44 d -1 and 0.47 d -1 , respectively). Statistical analysis showed LC was the most influential factor in the cell chemical composition, being significant (p<0,1) in the productivity of lipids, proteins and carbohydrates, with high values observed using between 5 and 10% of LC in the medium. GP did not show significance for any response studied, while the variables AF and FR showed significance in the productivity of proteins. This manuscript is about the use of an airlift photobioreactor operated in semi-continuous mode for cultivation of microalgae Chlorella minutissima . The effects of leachate concentration, percentage of CO 2 added to the gas stream, aeration flow and feed flow on microalgae composition and biomass productivity were evaluated. Studies dealing with the semi-continuous cultivation of microalgae in air transport photobioreactors are scarce in the literature, especially considering the use of Chlorella minutissima. Furthermore, the influence of leachate concentration is not fully understood, with some conflicting data on its effect on biomass composition. Thus, our manuscript contributes to increase knowledge about microalgae cultivation and includes a proposal for semi-continuous mixotrophic cultivation in a high-yield system with advantages such as simple construction, adequate mixing conditions and efficient exposure of cells to light. In addition, our manuscript helps to understand the significance of each factor applied to cultivation by using Taguchi's method and application of analysis of variance (ANOVA).


Introduction
In recent years, microalgae have shown themselves to be highly promising raw materials for the production of various bioproducts, such as food supplements, biofuels, pigments, antioxidants, among others.
Moreover, the search for new resources of energy capable of meeting global demand and, simultaneously, replacing fossil fuels, has increasing annually [3]. Actually, the urgency of combating global warming drives several research works around the planet in order to reduce the amount of polluting gases in the atmosphere, whether in the development of biofuels or in the bio-fixation of gases by plants and microorganisms [4,5].
The concept of biorefinery can be effectively applied to the use of microalgal biomass in industrial processes due to the possibility of production of different interesting compounds or biofuels, also including Actually, the cultivation of microalgae is an easy process with great applicability; however, special attention must be given to the influence of environmental factors. Temperature, pH, lighting and availability of carbon source, aeration flow and feed flowrate are some of the factors that can influence the concentration of microalgae biomass in the process. Therefore, those factors must be carefully controlled so that the biomass produced has the desired quality [15][16][17].
The construction of reactors that enable the cultivation of microalgae on a large scale allowing greater control over operating conditions stands out as an important challenge. Indeed, the high costs of construction and operability of photobioreactors are limiting factors for their commercial use, by not presenting the conditions to support large-scale cultivation in an economically viable manner [18][19][20]. On the other hand, airlift photobioreactors have been studied considering their higher productivity in biomass and due to advantages as low possibility of contamination, better control of the gas-liquid mass transfer rate, greater penetration and exposure to light even with the increase in cell density during cultivation, high productivity per area and low cost of harvesting biomass [14,[21][22][23].
Although microalgae have high lipid productivity, their large-scale cultivation demands a high content of nutrients. In this sense, different types of effluents have been tested as source of water and nutrients for the cultivation of microalgae, as wastewater or landfill leachate, which can reduce production costs [3,24,25].
The use of effluents for microalgae cultivation can also improve the energy balance and reduce environmental impacts associated with their disposal in bodies of water without prior treatment. With this principle, the use of microalgae for the treatment of effluents is a viable and low-cost alternative that can results in great reduction in the effluent pollutant power [26][27][28][29].
In present work, leachate was used as nutrient source and, thus, the evaluated cultivation option can be an interesting for both: use the leachate as nutrient sources to produce valuable biomass and use of the microalgae as a method for the effluent treatment.
Thus, the main objective of this work was to evaluate the production and composition of biomass in airlift bioreactor operated in semi-continuous mode to cultivate the microalgae Chlorella minutissima. The effects of leachate level, CO2 percentage added in the gaseous stream, aeration flow in the reactor, and flowrate of the feed stream, were evaluated in the cultivation performance. The experiments were carried out according to an orthogonal arrangement of Taguchi L9. Cell composition in the different studied conditions in terms of the contents of lipids, proteins, and carbohydrates, was evaluated. The semi-continuous mode was chosen due to its advantages, such as maintenance of the logarithmic phase growth, absence of shading effect caused by high cell density, possibility of constant replacement of fresh culture medium, providing increased productivity [30].

Microalgae specie, maintenance and inoculum preparation
The microalgae Chlorella minutissima 26a was obtained from the Marine Cultures Collection of the Oceanographic Institute at the University of São Paulo (São Paulo, SP, Brazil), located in Cabo Frio city (Rio de Janeiro, Brazil). The strain was stored in a wooden incubator equipped with a photoperiod, controlled by a timer. The maintenance of this cell bank was carried out by replicating the cells in Erlenmeyer flasks (125 mL), with 12 h (light/dark) of photoperiod and an average luminance of 4.8 klux (Fig. 1a). The Guillard medium f/2 [31] was used to grow the strain, with all the reagents used in the preparation of analytical standard grade. The sub-cultures were done every 15 days, in the proportion of 10 mL of previous culture to 90 mL of new culture medium, and the flasks were manually shaken once a day.
The inoculum was grown in a bubble tubular photobioreactor (2.0 L) made up of transparent polyethylene terephthalate (PET) (Fig. 1b), following the same proportions as the used to the growth in Erlenmeyer flasks. In order to avoid contamination of the inoculum, all containers used were previously chlorinated and neutralized with Sodium Thiosulfate (Na₂S₂O₃) [32]. Innoculum growth was carried out in a batch process with aeration of 0.24 vvm, controlled by means of absorbance readings (at 680 nm) until reaching the value of 1.500. This value of absorbance was chosen due to the need to standardize cell concentration which served as a starting point for cultivation in airlift reactors.

Microalgae cultivation in bioreactor
Microalgae growth was carried out in a bench-top airlift photobioreactor with a working volume of 4 liters, continuously illuminated of with luminance of 4.8 klux, and placed in a laboratory with the room temperature kept at 30 °C. The photobioreactor (Fig. 1c) size was based on that used by Tagliaferro et al. [21]. Reactor was charged with inoculum suspension (10% v/v) and culture medium. The culture medium consisted of distilled water, landfill leachate (concentrations according to experimental design, as section 2.2.2) and salt (20 g.L -1 , NaCl). The bioreactor was aerated at flowrate values according to experimental design. Aeration had a percentual volume of CO2 flow, also according to the experimental design.

Determination of the maximum specific growth rate (µmax)
At first, cultivations were carried out in a batch regime until their absorbance reached a value of 0.75 (680 nm), so that the calculation of the maximum specific growth rate (µ max) was performed. To determine the value of the maximum specific growth rate, the graph of dry biomass was plotted on a logarithmic scale as a function of time and the slope of the formed line was calculated at linear region.

Microalgae cultivation in semi-continuous mode
For semi-continuous mode cultivation, firstly the system was operated as a batch (section 2.2.1), until the absorbance reaches 0.75 (680 nm). Then, one third of the working volume of medium was removed from the reactor and fresh medium was added at a feed flowrate (FR, L/h) calculated as Equation 1, using a peristaltic pump. The same volume was removed from the reactor each 8 h, with continuous feeding at FR.
Each removed volume was used as sample for analysis of biomass composition. The total time of the cultivation in each experimental condition was 21 days.  airlift photobioreactor (c).

Biomass harvesting
The biomass was separated from the culture medium by flocculation, using 0.15M aluminum sulfate as the flocculating agent [33]. After flocculation, the biomass was subjected to the vacuum filtration process with quantitative filter paper (15 µm), using a system equipped with Buchner Funnel and a vacuum pump (131 -Type 2 VC, Prismatec). After separation, the biomass was oven dried (LT1000, Terroni) at 60 ºC for 24 hours.

Analytical methods
The humidity was determined by weighing the biomass before and after drying in an oven at 105 °C, according to the AOAC 96811 method [34].
Total lipids (L(%)) were extracted by following the procedure proposed by Bligh and Dyer [35], adapted as following described. The biomass was disposed of in a 125 mL Erlenmeyer and its humidity was adjusted until 64%, by adding distilled water; this humidity value was considered ideal by Zorn et al. [36]. Then, the In which PB is the volumetric productivity in biomass, Vreactor is the reactor volume, t is the time variation and G was lipid, protein or carbohydrate content.
The determination of ash content (A%) was carried out based on the adaptation of the methodology proposed by Van Wychen and Laurens [40], in which the calcination of the samples is used in a muffle furnace. The analysis started with the heating of empty porcelain crucibles and the application of a heating ramp from 50 ° C to 575 ° C for the total combustion of any organic compounds present in the containers.
After reaching the maximum temperature, the crucibles remained in heating at 575ºC for 4 hours. At the end of heating, the crucibles were cooled to room temperature inside a vacuum desiccator, then weighed on an analytical balance (AUW220, Shimadzu). Then, the samples were calcined in which, in each crucible, approximately 0.1 grams of dry biomass were added and the whole (crucible and biomass) was weighed.
The samples were taken to the muffle furnace and underwent the same heating process to which the empty crucibles were subjected. After the end of heating and, once cooled, the crucibles were weighed again and the difference between the mass of the post-calcination set and the mass of the empty set represents the amount of ash present.
Results and discussion . Another study that also evaluated the influence of carbon dioxide on the speed of microalgal growth was developed by Nair, Senthilnathan and Nagendra [42] in which the concentration of CO2 in the feed flow varied between 0.03 and 15%. The results obtained by the authors showed that the highest specific growth rate found was 0.14 d -1 for the experimental conditions that had CO2 percentages equal to 10% and 15%. This fact corroborates with the results presented in this work, since the highest calculated µmax value refers to experimental condition 6, where carbon dioxide concentration is 15%.
The experimental condition 5 presented the highest productivity in biomass and the second highest productivity in carbohydrate. In addition, the experimental condition 2 presented the highest productivity in lipids and the second highest productivity in proteins.
From Fig. 2a, when the LC factor was adjusted at level 2 (10%), GP at level 2 (10%), AF at level 1 (0.15 vvm) and FR at level 2 (0.7µmax), the Biomass Productivity (BP) had its value maximized. In contrast, when the LC factor was set at level 3 (15%), the GP at level 3 (15%), AF at level 2 (0.30 vvm) and FR at level 3 (0.9µmax), this response variable has its value minimized. However, as shown in Table 2 (ANOVA), no significant effects of any of the studied factors were observed for the response biomass productivity (p>0.1), although the landfill leachate seems to be the most influent variable ( Figure 2 and Table 2).  that the value of the lipid content obtained in this work is similar to that found in the literature. However, the higher lipid productivity was observed in experiment 2 (Table 1) Eustance et al. [51] studied the influence of aeration flow on the chemical composition of Scenedesmus dimorphous biomass in relation to protein and carbohydrate content. The authors observed that when no aeration was applied in the process, the protein and carbohydrate contents remained between 20% for proteins and 46% in carbohydrates. After the application of continuous aeration, these values changed to 27% in proteins and 42% in carbohydrates, which exemplifies that the microalgae had greater selectivity for the production of proteins when there was an increase in the aeration flow. Oscillatoria sp, which accredits this species as a good potential for raw material for the incorporation of products in food industries. In the work carried out by Tagliaferro [54], under cultivation conditions similar to those applied in this work, the protein contents remained close to 30% and were independent of the landfill leachate concentration, which remained between 5% and 10% of the total volume of the reactor.
The high protein concentration reached by microalgae in the present work can be interesting for industrial applications, even considering the use of leachate as substrate give restrictions to food applications.
Actually, there are nonfood applications of proteins [55][56][57] and future studies could consider microalgae as raw material in that case.  Observing the effects graph shown in Fig. 2d, when LC was adjusted at level 2 (10%), GP at level 2 (10% vvm), AF at level 1 (0.15 vvm) and the FR at level 2 (0.7µ max) the Carbohydrate Productivity (CP) had its value maximized. The maximum value of carbohydrates content was 11.6%, obtained in Experiment 6 ( Figure 3). Also, Table 2 shows that the Landfill leachate concentration (p<0.1) was the only factor that significantly affected Carbohydrate Productivity. In the work developed by Tagliaferro et al. [58] the highest carbohydrate content found was 15% in the experimental condition that applied a 10% dilution of landfill leachate, coinciding with the level of landfill leachate concentration indicated by the effects graph in the present work. In addition, the authors noted that, for lower landfill leachate concentrations, the maximum carbohydrate content was 9%.
As can be seen in Table 2, the percentage of CO2 was not shown to be a significant factor in the studied range for any analyzed productivity. This fact may be related to the diffusion of carbon dioxide in the culture medium. When cultivating microalgae in large quantities, the diffusion of CO2 from the atmosphere or even from other sources becomes a limiting factor for microalgal growth [59] Actually, the diffusion of CO2 in the microalgal culture medium is aided by the formation of bubbles; however, according to Doucha et al. [60] approximately 50% to 90% of the carbon dioxide injected in the process returns to the environment. Such waste of CO2 causes a reduction in biomass productivity and an increase in cultivation costs, since about 50% of the cost of the raw material used in the process is represented by the addition of carbon dioxide [61].

Fig. 3
Graphs of cellular chemical composition of biomasses produced in the semi-continuous regime.
Experiments conditions according to Table 1.
As also shown in Figure 3, ash content varied in a wide range, from 8.4% to 70%, with maximum value observed by using LC: 10%, GP: 15%, AF: 0,15 vvm and FR: 0.7µmax. It is known that the ash content refers directly to mineral salts and other components whose boiling temperature is very high, as is the case of metals [62][63][64][65], and the values in biomass can vary with species of microalgae, culture conditions, high metallic load present in the leachate [66]. In the work developed by Xu et al. [67], the ash content of the microalgae Chlorella sp was reported as 16.06% after analysis by four hours of heating at 550 ° C. That value was close to those found in experiments 2 and 7 of Figure 3. Ash content variation in literature can also be attributed to different analysis methods. E.g., in the research performed by Cervantes-Urieta et al. [68], the cellular composition of the microalgae Eupyxidicula turris, Trieres mobiliensis and Biddulphia alternans was analyzed and the ash content values found were 50%, 63% and 70%, respectively. For ash content analyses, the biomass of the three species of microalgae were subjected to heating in at 600 ° C for two hours and, according to the authors, the low heating time influenced the high ash content.

Conclusions
Landfill leachate concentration and aeration flow were crucial factors for all the responses variable analyzed in the present study. The landfill leachate concentration was the factor that most influenced the Biomass