Residual sugar from microalgae biomass harvested from phycoremediation of swine wastewater digestate

Concerns about the uncertain availability of fossil fuels in the near future have motivated the scientific community to search for alternative sources of renewable energy. For instance, ethanol derived from crops, such as corn and sugar cane, has become a commodity with a multibillion-dollar industry that still threatens to push up the price of these plants for food (Ho et al. 2013a; Brasil et al. 2015). There are several environmental implications associated with current production of energy crops, such as atmospheric emissions associated with the use of fire in sugar-cane fields; the excessive use of water for irrigation; contamination of groundwater and soil by pesticides; territorial expansion; and soil erosion, among others (Abbasi & Abbasi 2010). In this regard, lignocellulosic materials have been considered, although production of sufficient biomass requires logging and deforestation (Cheng & Timilsina 2011). Production of ethanol from lignocellulosic biomass that does not compete with the food industry is still struggling to take off due to high costs (Cheng & Timilsina 2011; Khoo 2015).

The use of microalgae for biofuel production has been discussed extensively. Compared to conventional crops, the growth rates and yield of microalgae are significantly superior. Other advantages include less demand for consumable resources (e.g. water and soil) (Mata et al. 2010; Khan et al. 2018) and the increased potential for CO₂ mitigation (Mu et al. 2014; Ullah et al. 2014). Microalgae biomass is composed of large amounts of carbohydrates (polysaccharides) in cell walls and across the intracellular matrix that can be converted into fermentable sugars (Harun et al. 2011). Residual sugar concentrations (wt wt⁻¹) of 80% (Spirulina platensis (Markou et al. 2013) and Synechococcus sp. (Möllers et al. 2014)), 45–70% (Chlorella vulgaris KMMCC-9; (Kim et al. 2014)), and 37.9–44.3% (Scenedesmus sp. CCNM 1077; (Pancha et al. 2016)) were reported.

It is worth noting, however, that the economic feasibility of microalgae production in an industry-relevant setting is largely influenced by the availability and costs of water and/or nutrients. In an attempt to reduce costs, the use of wastewater has been considered (Service 2011; Popp et al. 2014). Many industries are contemplating the use of algae-based phycoremediation treatment to remove nutrients from wastewater effluents while simultaneously producing valuable microalgae biomass (Brasil et al. 2015). The microalgae produced in this process can have different biochemical compositions depending on the nutrient concentration present in the wastewater used as growth medium (Lee et al. 2015a, 2015b). For instance, nutrient-rich wastewaters such as those generated from confined swine production may constitute an alternative growth medium to produce microalgal biomass rich in carbohydrates (Michelon et al. 2015; Özçimen & İnan 2015). Variations in carbohydrate content from microalgae can also occur depending on the harvesting method; that is, mechanical centrifugation or chemical coagulation/flocculation (Lee et al. 1998; Borges et al. 2011; Coward et al. 2014). This effect, however, is not always observed (Ndikubwimana et al. 2016), thus suggesting the need for assessment on a case-by-case basis.

Also, considering that microalgae biochemical composition can vary depending on the harvesting method used, ancillary investigation was performed to determine whether centrifugation or coagulation-flocculation (the two most conventional harvesting methods) could affect the total amount or residual sugar present in the biomass.

Microalgae inoculum was obtained from a field-scale lagoon used to remove nutrients from swine wastewater digestate originating from an anaerobic biodigester (Brazilian Agricultural Research Corporation, EMBRAPA, Concórdia, SC, Brazil). The inoculum was composed of a consortium dominated by Chlorella spp. as previously identified (Michelon et al. 2015). Experiments were performed at pilot scale using 500-L reactors (121.2 cm internal Ø; 58.4 cm height) placed inside a greenhouse, exposed to direct sunlight (photosynthetic photon flux density average and standard deviation of 321.5 ± 411.4 µmol m⁻² s⁻¹) and under ambient average temperature of 31.7 °C ±16.3 °C. These measurements (n = 3) were taken in the morning (8am), mid-day (12pm) and afternoon (4pm). Reactors were inoculated with 30% of inoculum (volume-based) containing 70 mg dry weight microalgae L⁻¹. The growth medium was continuously mixed in the reactor using a submersible aquarium pump (flow rate of 1,200 L h⁻¹). The growth medium consisted of 6% v/v of raw digestate effluent diluted in the reactor's total volume of water. Dilution of digestate was necessary to enhance light penetration and microalgae growth. The chemical composition of the growth medium used in the reactor at the beginning of the experiments (i.e. at time zero) was (average mg L⁻¹ ± standard deviation): total organic carbon (100 ± 5.2), biological oxygen demand (BOD₅ 90.8 ± 0.9), alkalinity as CaCO₃ (190 ± 10), total nitrogen (50.3 ± 0.9), ammonia-N (40.1 ± 0.7) and phosphate-P (10.5 ± 4.6). pH was 7.9 ± 0.6.

In this work, we focused on the microalgae only. The efficiency of phycoremediation as a treatment approach to remove nutrients from swine wastewater digestate was discussed elsewhere (Mezzari et al. 2013; Michelon et al. 2015; Prandini et al. 2016).

After 5 days of cultivation, the biomass in the reactor reached 0.3–0.4 g dry weight microalgae L⁻¹. At this point, biomass was harvested either by centrifugation (3,000 × g, at 25 °C for 30 min; EVODOS, T10, The Netherlands) or chemical coagulation-flocculation. A tannin-based cationic polyphenolic organic polymer produced through ammonium chloride and formaldehyde reaction was used as coagulant. The use of tannin was chosen due to its biodegradability (Beuckels et al. 2013; Vandamme et al. 2013) as opposed to other types of coagulants (e.g. aluminum) that may jeopardize water quality (Rosseland et al. 1990). The tannin used was extracted from Acacia tree (A. mearnsii) bark and is available commercially in liquid form with 30% w/v of tannic acid solution (flavan- 3,4-diol) and weight distribution of 830–1940Da (CAS # 85029-52-3; Veta Organic^TM, Brazilian Wattle Extracts, Canoas, Brazil). This tannin was chemically modified through the Mannich reaction to improve cationic strength properties by adding an ammonium quaternary functional group. Coagulation was performed directly in the reservoirs by adding 0.01% v v⁻¹ of tannin. The coagulation/flocculation method proved to recover >95% of microalgae biomass at neutral pH (Mezzari et al. 2014).

The concentration of sugars present in microalgae can decrease significantly within days at room temperature or when stored in the refrigerator (4 °C) (Adamson 2015). For this reason, the harvested microalgae biomass was immediately frozen (−40 °C) and lyophilized (Model 030-JJ LJI Scientific) on the same day.

The cellular lipid content was determined by ether extraction (Ankom XT15) (AOCS 2013). Protein content was measured by the combustion method (Leco FP-528) (AOAC 1990). Ash content was determined according to the Brazilian Compendium of Animal Nutrition, method 36 (BCAA 2009). Carbohydrate was determined by subtracting total cell dry weight from the measured lipid, protein and ash concentrations (Bi & He 2013).

Acid hydrolysis was used to recover sugar from microalgae biomass. A fixed amount of collected biomass (15 g L⁻¹ re-suspended in distilled H₂O) was used as substrate for reaction assays. Different concentrations of sulfuric acid were tested (i.e. 47, 94, 188, 281 and 563 mM) to determine the most effective concentration. Hydrolysis assays were conducted in Erlenmeyer flasks. Reaction took place at 100 °C for 30 min (Waiser Lab. Products NC EST – 011). Samples were cooled at room temperature and then centrifuged at 3,200 × g at 20 °C for 8 min (Excelsa^® II model 206 BL). The supernatant containing the residual sugars was collected and the pH adjusted to 5.5 using 1 M NaOH. The residual sugar concentration was analyzed using the DNS (dinitrosalicylic acid) method with glucose as standard for calibration curves (Miller 1959). After mixing 0.75 mL of glucose with 0.5 mL of DNS reagent, samples were heated at 100 °C for 5 min. Samples were cooled at room temperature and then 3 mL of water was added. Sugar concentrations were determined spectrophotometrically (Varian, Inc. Cary^® 50 UV-Vis) at 540 nm.

Prior to fermentation tests, the supernatant containing the residual sugar was sterilized in an autoclave at 121 °C for 15 min (Phoenix^® Av-75/2). Fermentation assays were performed in triplicates using two different strains of yeasts: Saccharomyces cerevisiae (AEB Fermol^®) and Saccharomyces cerevisiae chardonnay (Proenol^®). Pre-inoculum was prepared in sterile Erlenmeyer flasks by adding 20 g L⁻¹ of yeast into sterile deionized water containing 0.2 g L⁻¹ nutrient medium YPD broth medium (Himedia^®). After approximately 1 h of incubation, yeast suspensions were washed three times in phosphate buffer and transferred to 500 mL (3% w v⁻¹) Erlenmeyer flasks containing 200 mL of the sterile hydrolyzed sugar solution. Incubation took place at 30 °C for 48 h. Samples were taken over time for determination of sugar concentration as described above.

Experiments were performed in triplicate (n = 3). The results were presented as mean ± standard error. Data were tested for normality and homoskedasticity and the statistical differences between group means were determined by one-way ANOVA. Tukey's honestly significant difference (HSD) post hoc test was conducted after the determination of the homogeneity of variances (p ≥ 0.05). Statistical analyses were performed using SAS^TM (2012). The level of significance considered for all the analyses was 5% (p ≤ 0.05).

Microalgae can accumulate considerable amounts of lipids and carbohydrates under different nutrient-deficient conditions, making them one of the most versatile and sustainable sources for biofuel production (Fan et al. 2014). Most microalgae have carbohydrate contents ranging between 5–25% of the total cell biomass, depending on the species (Bruton et al. 2009; Biller & Ross 2011; Prajapati et al. 2014). The amount of carbohydrate content in microalgae biomass can be increased once cells are deprived of nutrients and/or exposed to additional sources of atmospheric CO₂ (Chen et al. 2013; Ho et al. 2013b). Our previous studies corroborate these findings (Michelon et al. 2015). Table 1 compares the carbohydrate content from different species of microalgae grown under controlled conditions using synthetic growth medium amended or not with CO₂. Most of the experimental conditions tested may not realistically represent the environmental dynamics (e.g. variations in light and temperature) expected at field scale. In this work, the microalgae produced from a pilot-scale experiment exposed to field conditions were rich in proteins (50.3%) and carbohydrates (41%) but low in lipids (1.3%). Thus, the cultivation of microalgae biomass as proposed here may not be attractive for industries seeking oil and its derivates such as Omega-3; Omega-6, etc.

Acid pre-treatment is a method used regularly to disrupt microalgae cell walls before proceeding to enzymatic hydrolysis. The disruption facilitates the release of entrapped carbohydrates present in the cell wall. To determine the most adequate concentration of acid for hydrolysis pretreatment, different concentrations of sulfuric acid ranging from 47 to 563 mM were tested (Figure 1). More diluted acid concentrations are always preferred for hydrolysis because the process becomes less harsh and costly. The use of low concentrations of acids for hydrolysis can still be more effective than other hydrolysis methods (e.g. enzymatic hydrolysis) (Ho et al. 2013c). The optimum concentration of sulfuric acid, which led to a significant (p < 0.01) increased sugar recovery (0.496 g-sugar g⁻¹ microalgae-DW; Table 2), was 188 mM (Figure 1). The lowest and highest concentrations of acid tested; that is, 47 and 563 mM, were not as effective in recovering sugar, with only 0.13 and 0.22 g-sugar g⁻¹ microalgae-DW, respectively. The sugar yield obtained here was comparable to other specific microalgae strains grown under controlled laboratory conditions using synthetic medium amended with CO₂ (29.5 to 98.2%) (Xu et al. 2013; Coward et al. 2014; Möllers et al. 2014; Lee et al. 2015a, 2015b; Wan et al. 2015) (Table 1).

There are various mechanical and chemical methods available for harvesting microalgae, such as centrifugation, flocculation, filtration and screening, gravity sedimentation, and flotation (Coward et al. 2014). Among these approaches, coagulation and flocculation with organic polymers are the most appropriate and cost-effective option for large-scale operations (Xu et al. 2013; Mezzari et al. 2014; Wan et al. 2015). There are conflicting lines of evidence showing that different harvesting processes result in changes to microalgae biochemical composition. (Borges et al. 2011; Coward et al. 2014; Michelon et al. 2015; Ndikubwimana et al. 2016). Thus, it is important to determine on a case-by-case basis if the harvesting method of choice can ultimately affect microalgae composition and residual sugar concentration. To address this question, the microalgae residual sugar yield harvested from centrifugation was compared to microalgae collected via coagulation-flocculation. No significant (p ≤ 0.05) differences in sugar concentration were observed independently of the method of harvesting used (Figure 2).

One limitation of bioethanol production from microalgae carbohydrates is that not all residual sugars are suitable for yeast fermentation (Lee et al. 2015a, 2015b). In this regard, less complex sugars such as glucose or fructose are usually preferred (Markou et al. 2013). Two different commercially available strains of Saccharomyces were used in the fermentation assays to investigate which yeast could lead to higher sugar consumption. These yeast strains were used because of their broad metabolic capabilities and capacity to adapt in response to changes in the environmental conditions, ultimately increasing bioethanol yield (Sharma et al. 2016; Mohd Azhar et al. 2017). Suspended cells of S. cerevisiae (Fermol Aromatic Group – AEB^®) and S. cerevisiae (Fermol Chardonnay Group – AEB^®) consumed 39.8% and 70.6%, respectively of the initial glucose concentration after 12 h of experiment. S. cerevisiae chardonnay was capable of removing 91.6% of the initial glucose present in the medium after 24 h of incubation (Table 2). The rate of sugar consumption was significantly higher (p < 0.0001) for S. cerevisiae chardonnay during the exponential growth phase (between 50 min and 6 h of incubation) (Table 2).

Table 3 shows the theoretical ethanol yield expected from microalgae biomass in comparison with other conventional feedstocks. Data from microalgae were estimated based on microalgae yield coefficient (0.3 g L⁻¹ obtained every 5 days of cultivation) and the concentration of sugar recovered from biomass. Microalgae biomass yield was eight-fold higher than corn and two-fold lower than sugarcane. However, the higher concentration of carbohydrate present in microalgae biomass outweighs its lower yield in comparison to sugar cane. Hence, the potential for sugar production from microalgae (43.4 ton ha⁻¹ yr⁻¹) is comparatively superior to sugarcane (18 ton ha⁻¹ yr⁻¹). In this regard, microalgae biomass can play an important role in the development of sustainable biorefineries that are less dependent on arable land and water for irrigation. The existing concerns about the use of arable land for production of biofuels instead of food are diminished because microalgae can be produced in areas unsuitable for agricultural practices (e.g. desert, sand, etc.) (Mussgnug et al. 2010). The water footprint to produce microalgae can range between 200 to 1,000 m³ ton⁻¹ (assuming typical yields of 1–5 g fresh weight L⁻¹) which is considerably higher than the estimated global average water footprint of sugar cane; that is, 209 m³ ton⁻¹ (Gerbens-Leenes & Hoekstra 2012). However, the water used for microalgae growth can be reused postharvest (Mezzari et al. 2014), thus significantly minimizing the amount of water needed. Another advantage of the use of microalgae is the short harvesting cycle (1–20 days) compared to sugarcane, which is harvested once or twice a year. The frequent harvesting provides uninterrupted supply of raw material to meet the constant demands imposed by industries. Once in the industry, the operational costs associated with biomass pretreatment are expected to be lower with the use of microalgae because cells lack hemicellulose and lignin, facilitating saccharification (Babadzhanov et al. 2004; Carrieri et al. 2010). Despite these advantages, however, the industry of microalgae for biofuels is still struggling to take off, mostly due to high operating costs with harvesting and dewatering, as well as low cell productivity (Khan et al. 2018). Consequently, the production costs of microalgae are still much higher (673 to 700 US$ ton biomass⁻¹) (Kang et al. 2015; Hoffman 2016) than sugarcane (20–26 US$ ton biomass⁻¹) (Cardoso et al. 2019). Yet, the high cost to produce microalgae can be offset by extracting and marketing residual byproducts of high added value for the nutraceutical and pharmaceutical industries. Although the production of microalgae biomass can unfold a promising feedstock alternative to ethanol production, it still needs to be further investigated by encompassing more comprehensive techno-economic analysis.

In this work, field scale experiments simulating phycoremediation of swine wastewater produced microalgae rich in proteins (50.3%) and carbohydrates (41.0%). Among the concentrations of sulfuric acid tested for the recovery of sugar from biomass, the concentration of 188 mM showed best results with 0.497 ± 0.001 g sugar g algae⁻¹. The use of mechanical or chemical coagulation-flocculation for harvesting the microalgae biomass had insignificant effect on the biomass residual sugar. Compared to S. cerevisiae, S. cerevisiae chardonnay showed significantly faster consumption of sugar during the exponential growth phase, consuming 92% of the total sugar added (8 g L⁻¹) within 24 h. These results support the notion that phycoremediation used as tertiary treatment system for removal of nutrients from wastewaters could provide valuable microalgae feedstock rich in fermentable sugars.

This research was supported by the Brazilian Agricultural Research Corporation (EMBRAPA) grant no. 02.12.08.004.00.05 and CAPES Foundation, Ministry of Education of Brazil.

Conflict of interest: The authors declare that they have no conflicts of interest.