Vinasse as a Sustainable Medium for the Production of Chlorella vulgaris UTEX 1803

: This study investigates distillery wastewater, commonly known as vinasse, as a potential culture medium for the production of Chlorella vulgaris and its most relevant metabolites. The e ﬀ ect of vinasse concentration on the composition of the biomass (proteins, carbohydrates, and lipids) was evaluated in treatments performed in 6-L tubular air-lift reactors. The reactors were operated at 25 ◦ C for 18 days, in total darkness, under a continuous ﬂow of air. Results showed a rapid growth of microalgae in the ﬁrst ten days, when an average production of 0.87 g / L was reached. Then, the daily biomass productivity began to decrease, up to an average value of 11.8 g / L at the 16th day. For all treatments, there was a signiﬁcant reduction in the concentration of most metabolites in the ﬁrst eight days. This was likely due to the adaptation of the biomass to the new conditions, with a transition from autotrophic to heterotrophic metabolism. From the 10th day, the concentration of metabolites in the biomass began to increase, reaching a nearly constant value at the 16th day. The observed maximum concentrations (% w / w ) were: 48.95% proteins, 2.88% xylose, 7.82% glucose, 4.54% arabinose, 8.28% fructose, and 4.82% lipids. These values were only marginally a ﬀ ected by the type of treatment. Overall, the results obtained suggest that vinasse is a promising and sustainable medium for the growth of C. vulgaris and the production of valuable metabolites.


Introduction
Large-scale production of wastewater is an inevitable consequence of contemporary societies. It leaves a trace on global biochemical cycles, mainly nitrogen, and phosphorus [1], as well as high concentrations of carbon and other nutrients. This has significantly reduced the self-purification capacity of natural water bodies [2,3]. As a result, several environmental regulations have been established to control this problem and regulate the levels of organic load, nitrogen, and phosphorus in the treated water [4].
The bioethanol industry produces large volumes (12)(13)(14) liters per liter of ethanol) of a wastewater known as vinasse [5]. Its characteristics depend on various factors such as variety and maturation degree of sugarcane, efficiency of the fermentation process and the conditions of the distillation of microalgae biomass. Glucose was the main monosaccharide detected, ranging between 46% and 76% of the total carbohydrate content.
From the above considerations, it clearly emerges that a biorefinery approach combining microalgae cultivation with the biological degradation of vinasse can be a sustainable option for the production of value-added compounds for the biofuel and other sectors. However, due to its chemical composition, high organic content and low pH, not all microalgae strains are able to grow in media containing vinasses. The main objective of this study was to evaluate whether the C. vulgaris strain UTEX 1803 could be grown in a culture medium supplemented with vinasse. This strain was selected because of its ability to grow in wastewater and its relatively high productivity of valuable metabolites, such as proteins, carbohydrates, and lipids. We were also interested in determining the biomass productivity and the evolution of the main microalgae components during the investigated treatments.

Vinasse Production
The vinasse used was obtained from the fermentation of molasses in the Laboratory of Chemical Processes at Universidad Industrial de Santander. For the fermentation process, 45 kg of commercial molasses were diluted in 151 L of water until reaching approximately 18 • Brix. The mixture was pasteurized at 80 • C for 1 h, then it was cooled to 40 • C and the pH was adjusted to 4.2 by addition of sulfuric acid (95 wt%). The inoculum was prepared using 20 L of the mixture, to which ammonium chloride (144 g), magnesium sulfate (24 g), urea (24 g), phosphate rock (10 g), and 500 g of commercial yeast Saccharomyces cerevisiae (Levapan, Colombia) were added. The inoculum was transferred to the tank together with the other diluted molasses. After 1 h aeration, the tank was covered to allow the fermentation process to proceed. The process was carried out for three days. Then, the mixture was evaporated by an evaporator operating at 94 • C in two stages, each lasting 210 min.

Vinasse Characterization and Nutrient Consumption
Once obtained, the vinasses were assayed for total nitrogen, sodium, potassium, phosphorus and organic carbon. Other parameters such as alcoholic degree and the concentration of total sugars and solids were also determined. Total antioxidants content was assessed following procedure of Zuorro et al. [24,25]. Total nitrogen (TN) was evaluated by the Kjeldahl method; according to SM 4500-N (BUCHI, Flawil, Switzerland). Sodium and potassium were determined by atomic absorption, according to SM 3111 B (Thermo Fisher Scientific, Waltham, MA, USA) and EPA 3050 (Thermo Fisher Scientific, Waltham, MA, USA), respectively. Phosphorus was determined spectrophotometrically by SM 4500-P C (Thermo Fisher Scientific, Waltham, MA, USA) and total organic carbon (TOC) by NTC 5167 (Thermo Fisher Scientific, Waltham, MA, USA). These measurements were made at the Laboratory of Technical Consultations, Universidad Industrial de Santander (Colombia). Finally, sugar content (AOAC 932.14), total solids (AOAC 925.23), and alcoholic degree (ICONTEC 74) were determined by the Food Laboratory CICTA of Universidad Industrial de Santander.

Inoculum Preparation
Chlorella vulgaris UTEX 1803 was obtained from the UTEX collection (Austin, TX, USA). The alga was grown in in tubular glass reactors with a culture volume of 2 L containing Bold's Basal Medium [26]. Each reactor was equipped with a bubble aeration system for the injection of air with 1% (v/v) CO 2 at a flow-rate of 0.6 L min −1 .

Microalgae Cultivation
Tubular 6-L air-lift reactors with a culture volume of 5 L were used. The reactors were equipped with a bubble aeration system for the injection of air at a flow-rate of 0.6 L min −1 . The temperature was 25 ± 1 • C and the pH was not controlled. Each reactor was coated with an aluminum foil to provide a  To test the effect of vinasse on microalgae growth and  composition, different dilutions of inoculum in vinasse were made (Tables 1-3). Each treatment was carried out in triplicate for 18 days.

Quantification of Biomass Growth and Biomass Components
Several methods are available in the literature for measuring biomass growth. The most widely used procedures include cell counting, absorbance or turbidity measurements, dry weight determination and quantification of some growth-related metabolites such as chlorophyll [27]. We used a dry-weight method. In particular, every two days, a 10-mL sample was removed from each reactor, centrifuged at 3400 rpm for 20 min and the supernatant was withdrawn. The solid fraction was resuspended in 10 mL of distilled water, filtered on pre-combusted CF/C glass fiber filters and dried overnight at 60 • C in an oven containing a bed of silica gel.
The main biomass components were quantified every two days. Protein content was determined by the Kjeldahl method using a nitrogen conversion factor of 6.25 [27]. Lipids were determined by the modified Bligh and Dyer method described by Barajas-Solano et al. [28]. Total carbohydrates were evaluated by the colorimetric method of Dubois et al. [29] modified by Jerez-Mogollón et al. [30]. This procedure is based on the spectrophotometric detection of the complex formed by the reaction of carbohydrates with phenol in concentrated sulfuric acid.

Vinasse Characterization
Vinasse is mainly composed of organic matter, potassium, sulfur, magnesium, nitrogen, and calcium, in amounts depending on the effluent source. It also contains phenolic dyes, caramel and melanoidins, which are responsible for its characteristic dark color. Studies carried out by Gloria and Filho [31], showed that vinasse from molasses had higher contents of organic matter and mineral elements. Table 4 shows the concentration ranges for the main compounds identified in the vinasse used as culture medium in this study.

Biomass Production
The biomass growth rates determined in the investigated treatments are reported in Table 5.  Figure 1 displays the time variation of biomass production at different vinasse concentrations. As can be seen, the highest growth rate was achieved using undiluted vinasse (75% v/v), which led to a biomass concentration of 5.11 g/L after 18 days. This is probably due to the higher concentration of nutrients in the medium.

Biomass Production
The biomass growth rates determined in the investigated treatments are reported in Table 5. 0.67 ± 0.08 Figure 1 displays the time variation of biomass production at different vinasse concentrations. As can be seen, the highest growth rate was achieved using undiluted vinasse (75% v/v), which led to a biomass concentration of 5.11 g/L after 18 days. This is probably due to the higher concentration of nutrients in the medium. In Figure 2, the pH changes in media with different concentrations of vinasse are shown. A first point to note is that in all media, including the most acidic one (CM3), a steady-state pH value close In Figure 2, the pH changes in media with different concentrations of vinasse are shown. A first point to note is that in all media, including the most acidic one (CM3), a steady-state pH value close to seven was reached. This can be attributed to different factors, such as the uptake of acidic vinasse components by the microalgae, the production of extracellular compounds capable of complexing or reacting with those components and the continuous injection of air in the medium, which contributes to the removal of the carbon dioxide produced during cellular respiration. Another important consideration to be made is that, despite the optimum pH for the growth of the investigated microalgal strain is around 7-7.5, the biomass is capable of growing even at pH value close to 4-4.5. This is what was observed during the first four days of growth in CM3, the medium with the highest concentration of vinasse and hence of nutrients. Since the growth rate of microalgae depends on both the pH and the amount of nutrients, the above results suggest that the reduction in biomass growth due to the low pH value is compensated for by the increase resulting from the higher amount of nutrients. to seven was reached. This can be attributed to different factors, such as the uptake of acidic vinasse components by the microalgae, the production of extracellular compounds capable of complexing or reacting with those components and the continuous injection of air in the medium, which contributes to the removal of the carbon dioxide produced during cellular respiration. Another important consideration to be made is that, despite the optimum pH for the growth of the investigated microalgal strain is around 7-7.5, the biomass is capable of growing even at pH value close to 4-4.5. This is what was observed during the first four days of growth in CM3, the medium with the highest concentration of vinasse and hence of nutrients. Since the growth rate of microalgae depends on both the pH and the amount of nutrients, the above results suggest that the reduction in biomass growth due to the low pH value is compensated for by the increase resulting from the higher amount of nutrients. The five treatments performed at different biomass/vinasse ratios (see Tables 2 and 3) gave the results presented in Figure 3. The trends were qualitatively similar for all treatments, with a progressive increase in biomass concentration during the first 8-10 days, followed by a nearly stationary phase. At day 18, treatment C2 resulted in the highest biomass concentration (8.7 g/L) and productivity (0.48 g L −1 d −1 ), while the lowest values (6.1 g/L and 0.34 g L −1 d −1 ) were observed for C1. The five treatments performed at different biomass/vinasse ratios (see Tables 2 and 3) gave the results presented in Figure 3. The trends were qualitatively similar for all treatments, with a progressive increase in biomass concentration during the first 8-10 days, followed by a nearly stationary phase. At day 18, treatment C2 resulted in the highest biomass concentration (8.7 g/L) and productivity (0.48 g L −1 d −1 ), while the lowest values (6.1 g/L and 0.34 g L −1 d −1 ) were observed for C1.  Tables 2 and 3).
In the present work, C. vulgaris was not affected by the concentration of the vinasse evaluated. Liu et al. [21] used industrial cane molasses as a carbon source for the production of Chlorella zofingiensis. When the microalgae were subjected to an ion-exchange pretreatment to remove metal  Tables 2 and 3). In the present work, C. vulgaris was not affected by the concentration of the vinasse evaluated. Liu et al. [21] used industrial cane molasses as a carbon source for the production of Chlorella zofingiensis. When the microalgae were subjected to an ion-exchange pretreatment to remove metal ions, a biomass productivity of 1.55 g L −1 d −1 was obtained. This value was slightly higher than that found using glucose instead of molasses. At higher nutrient levels, the production of biomass was decreased, which was attributed to substrate inhibition. Doucha and Livansky [32] obtained up to 117.2 g/L of C. vulgaris under heterotrophic conditions with glucose as carbon and energy source. This high production of biomass was achieved with an initial glucose concentration of 65 g/L. However, the specific growth rate decreased approximately linearly with increasing glucose concentration, which again was attributed to substrate inhibition phenomena. It was also shown that the content of some microalgae components, such as proteins, chlorophylls, and β-carotene, could be increased by keeping the biomass in the fermenter for an additional time after the cell growth was arrested due to glucose deficiency. Candido and Lombardi [33] used conventional and biodigested vinasses pretreated by filtration or centrifugation as a culture medium for C. vulgaris. The highest growth rates were obtained in 60% (v/v) filtered conventional and 80% (v/v) biodigested vinasses. The authors pointed out that filtration or centrifugation are essential treatments to be performed on vinasses before their use as a culture medium in order to eliminate toxic components and improve light penetration into the medium. In another study on a novel strain of Micractinium sp., sugarcane vinasse was used as a nutrient source for the heterotrophic and mixotrophic growth of the biomass [34]. Mixotrophic cultures resulted in higher specific growth rates and productivities, compared to the heterotrophic ones. For both of them, the highest biomass concentration and productivity were obtained with 10% (v/v) vinasse. Higher vinasse concentrations caused a decrease in microalgal growth, which was ascribed to toxic effects of vinasse components and/or light intensity reduction due to their presence. These considerations provide a possible explanation for the results of the present study displayed in Figure 3. In particular, there appears to be an optimal vinasse concentration (C2 treatment) at which the negative effects on biomass growth and productivity due to the presence of inhibiting and/or toxic compounds are balanced by the positive effects resulting from the higher nutrient levels. Figure 4 shows the consumption of the primary nutrients (phosphorus, potassium, and nitrogen) and carbon present in vinasse by the microalgae in the C2 treatment. This consumption is strictly related to the amount of biomass produced.  Figure 4 shows the consumption of the primary nutrients (phosphorus, potassium, and nitrogen) and carbon present in vinasse by the microalgae in the C2 treatment. This consumption is strictly related to the amount of biomass produced.

Nutrient Consumption by the Microalgae
Carbon is the major component of the biomass, reaching up to 50% of the total weight [35]. At day 18, it was reduced by about 85%. The amount of nitrogen consumed was approximately 38%, corresponding to about 950 mg/L. Travieso et al. [36] tested the suitability of pretreated vinasse as a culture medium for C. vulgaris in a continuous photobioreactor. They found a total nitrogen reduction of up to 1000 mg/L, which is very close to that of the present study, although the vinasse we used was not subjected to any pretreatment. From Figure 4, it can also be seen that 25% of all phosphorous present in the vinasse was consumed by the biomass at day 18, which corresponds to 14.25 mg/L of its initial content. Similar results were obtained by Valderrama et al. [20] using C. vulgaris and pretreated vinasse. In vinasse, two carbon sources are present: Ethanol and residual sugars. They can be consumed through both autotrophic and heterotrophic pathways ( Figure 5). Due to its low content (0.05-0.07 % v/v), ethanol can be expected to be consumed very quickly by C. vulgaris without exerting any inhibitory effect on it. Regarding sugars, their content decreased from 11.3-11.5% to about 6% in the first eight days. Then, they increased up to 8.3% at day 18. This increase, and also that observed for total solids, can be explained by the secretion of exopolysaccharides by the microalgae. As is known, Carbon is the major component of the biomass, reaching up to 50% of the total weight [35]. At day 18, it was reduced by about 85%. The amount of nitrogen consumed was approximately 38%, corresponding to about 950 mg/L. Travieso et al. [36] tested the suitability of pretreated vinasse as a culture medium for C. vulgaris in a continuous photobioreactor. They found a total nitrogen reduction of up to 1000 mg/L, which is very close to that of the present study, although the vinasse we used was not subjected to any pretreatment. From Figure 4, it can also be seen that 25% of all phosphorous present in the vinasse was consumed by the biomass at day 18, which corresponds to 14.25 mg/L of its initial content. Similar results were obtained by Valderrama et al. [20] using C. vulgaris and pretreated vinasse.
In vinasse, two carbon sources are present: Ethanol and residual sugars. They can be consumed through both autotrophic and heterotrophic pathways ( Figure 5). Due to its low content (0.05-0.07% v/v), ethanol can be expected to be consumed very quickly by C. vulgaris without exerting any inhibitory effect on it. Regarding sugars, their content decreased from 11.3-11.5% to about 6% in the first eight days. Then, they increased up to 8.3% at day 18. This increase, and also that observed for total solids, can be explained by the secretion of exopolysaccharides by the microalgae. As is known, depending on the growth phase and culture conditions, microalgae can produce biofilms containing carbohydrates and proteins [37,38].

Biomass Composition.
During each of the five investigated treatments (D1, D2, C1, C2, C3, C4, and C5) the biomass was characterized in terms of protein, carbohydrate, and lipid content. The results of measurements performed at intervals of two days are shown in Figure 6.
Regarding the protein content, in all treatments a decrease was observed during the first six days. After this period, the amount of proteins began to increase until reaching a nearly constant value around the 14th day. At the 6th day, the following values were determined (% w/w): 20.47% (D1), 26.37% (D2), 27.78% (C1), 17.90% (C2), and 15.90% (C3), while at the 18th day they were, respectively, equal to: 47.53%, 48.95%, 46.67%, 45.95%, and 47.37%. Changes in the protein content occurring during the first ten days can be attributed to a progressive shift of the metabolism from photoautotrophic to mixotrophic and to fully heterotrophic. In a study by Coca et al. [39], Spirulina platensis was grown in an airlift photobioreactor using a culture medium supplemented with 1 or 2 g/L of beet vinasse. At the lowest vinasse concentration, both biomass and protein productivities were increased, while at higher concentration they showed a decrease, although the protein content of the biomass remained relatively stable, at about 53%. In another study on Spirulina maxima grown in media supplemented with 3% (v/v) sugarcane vinasse under cyclic two-stage cultivation (12-h autotrophic conditions followed by 12-h fed-batch heterotrophic conditions during the dark phase), a protein content ranging from 74.3 to 77.3% (w/w) was determined [22]. In yet another study on the cultivation of selected microalgae strains in sugarcane vinasse, a protein content between 34 and 40% (w/w) was obtained [23]. Furthermore, an inverse correlation was observed between protein and carbohydrate contents. According to the authors, this could be related to the high availability of

Biomass Composition
During each of the five investigated treatments (D1, D2, C1, C2, C3, C4, and C5) the biomass was characterized in terms of protein, carbohydrate, and lipid content. The results of measurements performed at intervals of two days are shown in Figure 6. Regarding the protein content, in all treatments a decrease was observed during the first six days. After this period, the amount of proteins began to increase until reaching a nearly constant value around the 14th day. At the 6th day, the following values were determined (% w/w): 20.47% (D1), 26.37% (D2), 27.78% (C1), 17.90% (C2), and 15.90% (C3), while at the 18th day they were, respectively, equal to: 47.53%, 48.95%, 46.67%, 45.95%, and 47.37%. Changes in the protein content occurring during the first ten days can be attributed to a progressive shift of the metabolism from photoautotrophic to mixotrophic and to fully heterotrophic. In a study by Coca et al. [39], Spirulina platensis was grown in an airlift photobioreactor using a culture medium supplemented with 1 or 2 g/L of beet vinasse. At the lowest vinasse concentration, both biomass and protein productivities were increased, while at higher concentration they showed a decrease, although the protein content of the biomass remained relatively stable, at about 53%. In another study on Spirulina maxima grown in media supplemented with 3% (v/v) sugarcane vinasse under cyclic two-stage cultivation (12-h autotrophic conditions followed by 12-h fed-batch heterotrophic conditions during the dark phase), a protein content ranging from 74.3 to 77.3% (w/w) was determined [22]. In yet another study on the cultivation of selected microalgae strains in sugarcane vinasse, a protein content between 34 and 40% (w/w) was obtained [23]. Furthermore, an inverse correlation was observed between protein and carbohydrate contents. According to the authors, this could be related to the high availability of a nitrogen source in vinasses, which would favor the accumulation of proteins over that of carbohydrates.
In previous studies on different strains of C. vulgaris carbohydrate levels ranging between 41% and 55% were found [18,40,41]. Monosaccharides were up to 22% of the total biomass on day 18, with a biomass production of 13.17 g/L. The presence of carbohydrates in low amount in the biomass is likely due to the high level of nitrogen in the culture medium, as these two quantities appear to be inversely related [40,42]. Jerez-Mogollón et al. [30] obtained a maximum content of xylose of 9.70% (w/w) in a culture of C. vulgaris UTEX 1803 enriched with sodium acetate. In the same study, a glucose content of up to of 9.30% (w/w) was determined [30].
Finally, from Figure 6 it can be seen that lipids were the metabolites at the lowest concentration, their amounts (%w/w) being: 1.72% (D1), 1.63% (D2), 1.79% (C1), 1.67% (C2), and 1.83% (C3). Similar low levels of these compounds were found by Engin et al. [34], who used a culture medium supplemented with vinasse to grow Micractinium sp., although in other studies significantly higher lipid contents were achieved [43,44]. However, as already mentioned, there seems to be an inverse correlation between protein accumulation in algal biomass and carbohydrates and/or lipid content [23], consistently with the results of the present study.
Overall, it can be concluded that incorporation of appropriate amounts of vinasse in culture media is not only an effective remediation strategy for the treatment of this effluent but also an attractive alternative for the production of valuable metabolites such as carbohydrates and proteins from C. vulgaris [45,46]. This approach can be included in the wider context of the sustainable reuse of waste materials for reducing the consumption of natural resources and the environmental impact of human activities [47][48][49]. Furthermore, it has the potential of being easily transferred to the industrial scale, as the conditions necessary for the heterotrophic growth of microalgae are similar to those employed for the large-scale production of yeast and bacteria [50,51]. In principle, vinasse could be treated in either open or closed bioreactors, with suspended or non-suspended biomass [52]. The experience gained so far from similar systems suggests that a closed photobioreactor with a periodic harvesting of the biomass is the best option, as this reactor configuration allows better pH and temperature control, better mixing, higher cell densities, better protection against culture contamination, and lower evaporative losses, compared to the open one [53], and it will be optimized in a response surface methodology study [54].

Conclusions
The results of this study indicate that vinasse represents a suitable and effective medium for the growth of C. vulgaris UTEX 1803. The algal biomass can be used as a source of value-added compounds for food, nutraceutical, cosmetic, and biofuel applications. Concerning the inoculation of the strain, there appear to be no significant differences in the amount of metabolites produced by the biomass in the five treatments examined.
Future studies should be focused on the optimization of the treatment conditions for the vinasse-biomass system. Another important issue to be addressed is to evaluate whether the production of a specific metabolite or class of metabolites can be controlled by proper selection of process conditions.