Nutrient removal from high strength nitrate containing industrial wastewater using Chlorella sp. strain ACUF_802

Abstract

the research aim of this study was to characterize an isolated native strain of Chlorella sp. ACUF_802, well adapted to a high nitrate concentration environment and to investigate its potential to nitrate and phosphate removal from industrial wastewaters with the minimal addition of chemical reagents and energy. The isolated strain was identified and evaluated for its capability to support biomass growth and nutrient removal from synthetic wastewater in batch tests using different concentrations of carbon and nitrogen, different carbon sources and N:P ratios. The strain was isolated via the plating method from the settler of a pilot scale moving bed biofilm reactor performing a nitrification process. The strain was identified using molecular analysis with rDNA primers. Using sodium bicarbonate as carbon source, the batch productivity (71.43 mg L⁻¹ day⁻¹) of the strain Chlorella sp. ACUF_802 was calculated with a logistic model and compared to the values reported in the literature. Assays on the effect of the N:P ratio indicated that the productivity was increased 36% when the N:P ratio was close to 1 (111.96 mg L⁻¹ day⁻¹), but for a complete phosphorus removal a 5:1 N:P ratio with nitrate concentrations ≤125 mg∙L⁻¹ is recommended. The isolated microalgae strain Chlorella sp. ACUF_802 showed versatility to grow in the synthetic industrial wastewaters tested and can be considered as an appropriate organism for nitrogen removal from industrial wastewaters in the presence of an organic or inorganic carbon source.

Introduction

It is widely recognized that water treatment and sanitation are essential to the achievement of many sustainable development goals (WWAP 2015). As a consequence, stricter government-issued regulatory wastewater effluent standards that give special attention to inorganic pollutants have been recently developed (Ibisch et al. 2016), especially nutrients such as nitrogen (N) and phosphorus (P), which contribute to eutrophication. Ammonium, nitrate, and phosphate are the primary forms of N and P in wastewater (Renuka et al. 2015). To date, biological and chemical systems for the removal of nutrients from wastewater are available. Chemical systems are expensive due to the addition of chemicals, while the main cost of biological systems is due to the required aeration and the sludge management and disposal (Mallick 2002). Under this scope, many efforts have been devoted to the research of alternative wastewater treatment approaches aiming to set sustainable and innovative solutions to remove N and P compounds from wastewater (Ruiz et al. 2011; Ge and Champagne 2016).

In wastewater treatment, biological systems for nutrient removal are the base of the secondary treatment process, which aims to reduce the biochemical oxygen demand (BOD) by reducing the organic matter content in advanced oxidation processes, aerated lagoons, aerobic granular reactors, and activated sludge plants. This is mediated, primarily, by a mixed population of heterotrophic bacteria and protozoa that utilize the organic compounds for energy and growth (Abdel-Raouf et al. 2012). However, in comparison to the entirely heterotrophic systems, the primary attraction of algal systems stands in the low-grade technology and in the saving of energy, since photosynthetic oxygen production can replace mechanical aeration (Mallick 2002). Furthermore, algal systems present other beneficial effects like valuable biomass and co-product generation and nutrient removal from wastewaters (Oswald et al. 1957; Gonzalez et al. 1997; Mallick 2002; Cai et al. 2013; Renuka et al. 2013; Arbib et al. 2014). More precisely, the Chlorella genus has shown promising properties for wastewater treatment (Mata et al. 2010; Abdel-Raouf et al. 2012). However, only a limited number of researchers have focused on the performance and cultivation of these microalgae associated to the treatment of nutrient-rich industrial wastewaters (Table 1).

Table 1 Treatment of concentrated wastewater by different strains of the Chlorella genus reported in the literature

In addition, industrial wastewaters can contain very high N and P concentrations, up to three orders of magnitude higher than the limits in water bodies (De la Noüe et al. 1992) and can vary highly depending on the industry source (Arbib et al. 2013; Zhang et al. 2014a). For example, anaerobic digestion effluents from different bioreactors treating poultry litter can contain 1570 mg L⁻¹ of total nitrogen (TN) and 154 mg L⁻¹ of total phosphorous (TP) (Singh et al. 2011), citric acid production effluent can contain 305 mg L⁻¹ TN and 35 mg L⁻¹ TP (Li et al. 2013), carpet mill industrial effluent can contain 0.1–28.1 mg L⁻¹ N-NO₃⁻∙and 20–35 mg L⁻¹ P-PO₄³⁻ (Chinnasamy et al. 2010), and cattle manure can contain 2313 mg L⁻¹ TN and 1119 mg L⁻¹ TP (Kobayashi et al. 2013).

The ability of microalgae to adapt their metabolism to different culture conditions provides opportunities to several biotechnological applications (Bumbak et al. 2011; Mirzaie et al. 2016). Currently, natural photoautotrophic conditions for biomass production are implemented with CO₂ supplementation (Gupta et al. 2016). However, these cultivation conditions are limited due to low growth rates, low light penetration and photoinhibition (Liang et al. 2009). Under mixotrophic conditions, microalgae can grow using both light as energy source and CO₂ as carbon source (autotrophic growth) or using organic carbon as energy and carbon source (heterotrophic growth) (Zhang et al. 2014a). The higher growth rates normally achieved under mixotrophic or heterotrophic conditions are advantageous for industrial biomass production (Cerón García et al. 2006; Liang et al. 2009). Although mixotrophic cultivation of microalgae can provide higher biomass concentrations than under autotrophic conditions, the selection of the organic carbon source plays a key role in biomass production systems (Yeh and Chang 2012; Zhang et al. 2014a, b; Gupta et al. 2016). The effects of organic compounds on the growth of microalgae to produce biomass and simultaneously perform wastewater treatment by mixotrophic cultivation offer interesting opportunities. However, little is known about the effect of elevated nutrient concentrations on mixotrophic algal growth (Zhang et al. 2014a).

Each microalgal strain has its optimal growth conditions, making the strain selection process an important step in the establishment of a microalgal treatment system (Richmond 2004). To date, a considerable number of algae species have been screened and numerous microalgae species libraries have been established (Ji et al. 2015). However, microalgae species that tolerate elevated nutrient concentrations, as in the case of using industrial wastewaters as the culture medium, still need to be screened (Mata et al. 2010; Ji et al. 2015).

The aim of this study is, therefore, to characterize an isolated native strain of Chlorella sp. ACUF_802 (Electronic supplementary material Fig. S1), well adapted to a high nitrate concentration environment (the nitrification stage of an industrial wastewater treatment plant), for its potential application in nitrate and phosphate removal from industrial wastewater with minimal addition of chemical reagents and energy. The characterization was carried out by analyzing the biomass growth rate under specific operational conditions and different nutrient concentrations and by taking into account the biomass productivity (P_b) and the nutrient removal capability with two specific strategies: cultivate Chlorella sp. ACUF_802 growth in synthetic industrial wastewater under (1) autotrophic and mixotrophic conditions and (2) with only atmospheric air at different N:P ratios. Furthermore, in order to evaluate the biomass productivity, a microalgal growth curve has been described through a logistic regression model (Verhulst 1838).

Material and methods

Isolation and growth of the microalgae strain from wastewater

The strain Chlorella sp. ACUF_802 was isolated from the settler of a pilot scale Moving Bed Biofilm Reactor (MBBR). The MBBR was used for the treatment of industrial wastewaters generating a nitrate rich effluent, i.e., 2500 mg L⁻¹ NO₃⁻ and 1–2 mg L⁻¹ PO₄³⁻ (Vannecke et al. 2016). An initial sample from the algal biofilm attached on the wall of the settler was taken and cultivated in a 2-L bioreactor (Pyrex bottle). A modified Bold’s basal medium (BBM) was used to stimulate the microalgae growth and activity (Nichols and Bold 1965). The bioreactor was continuously exposed to artificial light (Phillips TLA, 30W/55), and the modified BBM was refreshed every 2 weeks.

To establish a pure culture of the microalga species, 10 mL of sample was collected from the bioreactor, poured in a closed 15 mL falcon tube, and exposed for 1 week to continuous artificial white light (95 μmol photons m⁻² s⁻¹). Then, standard plating methods were used to separate the algal population. One hundred microliters from the sample were plated on Petri dishes containing 25 mL of agar BBM and 200 μg mL⁻¹ of antifungal and antibacterial combos (amphotericin + rifampicin, 1:1). Petri dishes were incubated for 2 weeks at 22 °C under continuous illumination. Single green colonies were transferred to a new Petri dish with agar BBM without antibiotics. After an additional week, each single colony was inoculated in liquid BBM and algal growth was monitored daily by optical density measurements at 680 nm wavelength and by cell counting using a Bürker counter chamber.

Strain identification

The strain was identified by light microscopic examination as belonging to the Chlorella genus (Fig. 1) using a Nikon Eclipse E800 (Nikon Instruments Europe B.V. Düsseldorf, Germany). Further molecular analyses were conducted on Chlorella sp. ACUF_802 at the algal collection of the Department of Biology, University of Naples “Federico II” (www.acuf.net), where the strain was deposited and included in a special microalgae strain collection for industrial applications. DNA was extracted from liquid culture with the protocol by Doyle and Doyle (1990) and used for a polymerase chain reaction (PCR) with primers targeting the internal transcribed region of rDNA (ITS1_F 5′-TCCGTAGGTGAACCTGCGG-3 (White et al. 1990); ITS_rev_R 5′-TTCAAAGATTCGATGATTCAC-3′). PCR was carried out in a 25-μL aliquot containing approximately 100 ng DNA, a deoxynucleoside triphosphate mixture (0.2 mM each), buffer (1/10 volume of the supplied 10x buffer), supplemented to give a final concentration of 2.5 mM MgCl₂, 1.25 U of Taq polymerase (EconoTaq, Lucigen), and 0.5 pmol of each primer. Amplification was run in an Applied Biosystem 2720 thermal cycler. The profile used was 5 min at 95 °C, 33 cycles of 95 °C for 30 s, 54 °C for 45 s, and 72 °C for 45 s and a final elongation step of 10 min at 72 °C. The amplification product (423 bp) was evaluated on 1.2% (w/v) agarose gel in a electrophoretic run and purified using the QIAquick® PCR Purification kit (Qiagen Inc., Valencia, CA, USA).

Fig. 1

Microscopic photograph of the Chlorella sp. ACUF_802 strain, isolated from a high nitrate concentration environment (the nitrification stage of an industrial wastewater treatment plant)

Sequence reaction was obtained with the BigDye Terminator Cycle Sequencing technology (Applied Biosystems, Foster City, CA), purified in automation using the Agencourt CleanSEQ Dye Terminator Removal Kit (Agencourt Bioscience Corporation, 500 Cummins Center, Suite 2450, Beverly, MA 01915, USA), and a robotic station Biomek FX (Beckman Coulter, Fullerton, CA). Product was analyzed on an Automated Capillary Electrophoresis Sequencer 3730 DNA Analyzer (Applied Biosystems). The amplification primers were used as the sequencing primers. Nucleotide sequence similarity was determined by using BLAST version 2.0 (National Center for Biotechnology Information databases) (Boratyn et al. 2013). To identify the isolate to the specie level, rDNA (ITS) sequences were aligned and compared with 87 sequences belonging to different strains of Chlorella and other genera (see Supplementary data) which showed a high identity percentage by BLAST analysis. The sequences were downloaded in the GenBank nucleotide database. The multiple nucleotide alignment was obtained by UGENE software (Okonechnikov et al. 2012) with the addition of the Chlorella sp. ACUF_802 strain sequence. The alignment, consisting of 326 sites, was manually inspected for quality and the end gaps trimmed. Manual inspection is a widely used technique for reducing the number of sequencing errors and improving quality (Alachiotis et al. 2013).

Bayesian inference was obtained with MrBayes 3.2.0, running two million generations and a sample frequency of 200, and using the general time reversible model (Tavaré 1986) with an invariable four gamma-distributed substitution rate category to correct for among site rate variation (GTR + G + I). The analysis was stopped at an average standard deviation of split frequencies of 0.00608. The first 25% of sampled trees were discarded as burn-in before calculating posterior probabilities (Electronic supplementary material Fig. S2).

Culture media and experimental procedure

Two different culture media were tested to simulate different types of wastewaters and to study the performance of the microalgae growth under different nutrient concentrations and different carbon sources (Electronic supplementary material Fig. S1). The composition of the synthetic industrial wastewater used as the basis of the cell culture consisted of a modified version of the BBM. BBM is commonly used for green algae (Nichols and Bold 1965). The modified version of BBM was prepared as described in Table 2. The medium simulates a wastewater from the phosphorus manufacturing process with a phosphorus concentration of 200 mg L⁻¹ (Yapijakis and Wang 2006). Two different carbon sources were added to the synthetic wastewater (SW): either sodium bicarbonate (NaHCO₃, SB), to simulate an alkaline rich wastewater, denoted as SWSB, or ammonium acetate (CH₃COONH₄, AA), denoted as SWAA (Table 3). The latter simulated the condition of mixing diverse wastewaters resulting in the presence of two different inorganic nitrogen and carbon sources, similarly to the experiments carried out by Kobayashi et al. (2013) for the treatment of the anaerobic digestion effluents.

Table 2 Initial nitrate, phosphate, sodium bicarbonate and ammonium acetate concentrations used in tests to investigate: (I) performance with different carbon source, (II) effect of the N:P ratio in cultures with medium at different N and P concentrations with a N:P ratio close to 1, and (III) at different N and P concentrations with a 5:1 N:P ratio

Table 3 Biomass kinetic parameters of Chlorella sp. ACUF_802 determined by the Verhulst logistic model for the carbon sources and the different N:P ratios tested. X₀ (mg biomass L⁻¹): initial biomass concentration; X_m (mg biomass L⁻¹): maximum final biomass concentration; μ_m (day⁻¹): maximum specific growth rate; P_b (mg L⁻¹ day⁻¹): batch productivity

The stock culture of the strain was maintained in modified BBM. Experiments were carried out in triplicate in 250-mL flasks closed with a cotton plug in an illuminated culture room at 25 (± 2) °C with agitation (mixing) supplied by a table shaker at 100 rpm and constant illumination (95 μmol photons m⁻² s⁻¹). Fluorescent lights are commonly used in algal bioreactors and have been shown to satisfactorily simulate photosynthetically active radiation (Zippel et al. 2007). In this study, the batch reactors were inoculated with a specific volume of microalgae in order to obtain a similar initial biomass concentration (X₀) in all the assays (initial optical density at 680 nm close to 0.10 corresponding to 2.5 × 10⁶ cells mL⁻¹). The experiment lasted 2 weeks without additional pH adjustment along the experiments.

Effect of the different carbon and nitrogen sources and different nutrient availabilities

Three different sets of experiments were carried out to evaluate the capability of metabolic adaptation and the nutrient removal efficiency of Chlorella sp. ACUF_802 from different media composition. The first set was performed by using a different nitrogen and carbon source. The second one was executed by lowering the N:P ratio, below 5:1 (molN:molP), and by focusing on nitrogen limiting conditions (see Table 3). The third set of experiments was performed by keeping a higher N:P ratio, i.e., higher than 5:1, with phosphorous limiting conditions, as reported for aerobic treatment of industrial wastewater (Ammary 2004; Klausmeier et al. 2004; Li et al. 2010). Specifically, the initial N concentration of the medium was increased. The nutrient concentrations are detailed in Table 3.

Analytical methods and data analysis

The biomass concentration (X) was determined according to Moheimani et al. (2013). Samples were diluted by appropriate ratios to ensure that absorbance values were in the range of 0.1–1 (dimensionless). To convert the OD₆₈₀ values to biomass as dry weight (mg L⁻¹), a calibration curve was determined. Biomass dry weight (DW) was measured by filtering an aliquot of an algal sample on pre-weighed glass-fiber filter paper with 0.45-μm pore size. The filters were then dried at 105 °C in an oven for 12 h. Algal biomass DW was determined by the difference of the two weights. Chlorophyll α (Chl α) was measured in vivo with an Aquafluor (Turner Designs) fluorimeter as described by Gargano et al. (2016), and pH values were measured with a pH meter Bench model AD-1030 (Adwa Instruments Inc., Szeged, Hungary).

The Verhulst logistic kinetic model (Verhulst 1838) was used to predict the evolution of the experimental biomass concentration. This model can accurately describe the growth of biomass in different culture conditions occurring in many batch bioreactors as a sinusoidal curve (Gong and Lun 1996, Arbib et al. 2013). The microbial growth could be expressed as described by Eq. (1), where μ_m is the maximum specific growth rate (day⁻¹), and X₀, X, and X_m are the concentration of biomass (mg L⁻¹) at an operation time equal to 0, t, and infinite (i.e., the maximum concentration) respectively.

$$ X=\frac{X_0{X}_{\mathrm{m}}{e}^{\mu_{\mathrm{m}}\ f}}{X_{\mathrm{m}}-{X}_0+{X}_0{e}^{\mu_{\mathrm{m}}\ f}} $$

(1)

To estimate the P_b with a standardized methodology, the lag phase should be minimized or even not be considered due to its high variability related to the experimental conditions and the experimental methodology (Arbib et al. 2013). The biomass productivity (mg L⁻¹ day⁻¹) was then calculated as in Eq. (2).

$$ {P}_{\mathrm{b}}=\frac{\mu_{\mathrm{m}}\left(0.9\cdot {X}_{\mathrm{m}}-1.1\cdot {X}_0\right)}{\ln \left(\frac{9\cdot \left({X}_{\mathrm{m}}-1.1\cdot {X}_0\right)}{1.1\cdot {X}_0}\right)} $$

(2)

Phosphorus was measured as orthophosphate ions (PO₄³⁻) and nitrogen as nitrate (NO₃⁻) and ammonium (NH₄⁺, where present). Analyses of nitrate and phosphate were performed with ion chromatography using a761 compacts IC analyzer (Metrohm, Herisau, Switzerland) according to APHA (2005). Amonium was measured by a titrimetric method as described by Wagner (1940). The nutrient removal rates were calculated considering the total nitrogen and phosphorus amounts removed during the incubation period (days). Kinetic modeling was performed using the SOLVER tool of Microsoft Excel 2011(Microsoft®), and the values were estimated by least squares. A confidence interval for p ≤ 0.05 (C.I.) was used for the estimation of the determination coefficient.

Results

Isolation and identification of the microalgae strain

The strain Chlorella sp. ACUF_802 was isolated and purified from the effluent of a nitrification reactor. The strain demonstrated tolerance to high nitrate concentrations growing spontaneously in this environment. The cells were planktonic and spherical with a diameter range of 3–6 μm, single chromatophore, and a parietal mantle-shaped covering nearly the whole of the cell wall. A conspicuous round pyrenoid was present in each chromatophore (Fig. 1). The ITS phylogenetic tree suggests a close relation (0.99) of this strain with other Chlorella sp. isolated from wastewater (Chlorella sp. Iso4) (Electronic supplementary material Fig. S2).

Microalgae growth and nutrient removal from different wastewaters

Chlorella sp. ACUF_802 was able to grow in the synthetic industrial wastewaters tested. A typical evolution of growth in batch cultures was observed for all experiments. The Verhulst kinetic model showed a good alignment (lines in Fig. 2a) to the experimental data. In particular, SWSB showed the best alignment with R² of 0.98 (Table 3). SWSB showed the highest maximal growth rate (μ_m) (1.19 day⁻¹) and P_b (71.43 mg L⁻¹ day⁻¹) (Fig. 2; Table 3) during the experiments. The maximum biomass concentration (X_m) of 485.21 mg L⁻¹ was achieved by the cultures growing in the SWAA medium (Fig. 2a).

Fig. 2

Variation of biomass growth (a), chlorophyll α concentration (b), and pH (c) of Chlorella sp. ACUF_802 growing at different nitrogen concentrations with N:P ratio close to 1. SWSB (circles) and SWAA (squares). Error bars indicate standard deviation, where n = 3

The cultures grown in SWSB and SWAA showed a similar growth trend during the first 4 days. Afterwards, the growth rate of SWAA reached the highest biomass concentration prior to achieving the stationary phase (after 11 days of incubation). The Chl α concentration of Chlorella sp. ACUF_802 cells growing in SWAA medium increased until the sixth day but declined at day 7. In contrast, cultures growing in SWSB globally showed a low Chl α concentration (Fig. 2b).

The pH evolution over time showed a typical increase in both the SWSB and SWAA media reaching a maximum value of 10.2 and 8.0, respectively. This difference is partially due to the presence of different carbon sources in the modified BBM media tested, leading to the different initial pH values of 8.8 and 6.8, respectively. In both experiments, the pH slightly decreased during the last monitoring days (Fig. 2c).

The nitrogen removal rate by Chlorella sp. ACUF_802 was faster in SWSB medium that contains an inorganic carbon source (Fig. 3a). A concentration lower than 8 mg L⁻¹ N was achieved at the second day of the experiment. On the other hand, for the SWAA medium, low NO₃⁻ concentrations were achieved only at the 6th day (Fig. 3b). Indeed, in this experiment, the nitrogen sources were both nitrate and ammonium, as ammonium acetate was added as the carbon source. The experiment showed that ammonium was the preferred N source, with 71.6% removal obtained in the first 2 days. During this incubation time, only 8.5% of the nitrate was removed, while during the next 6 days, 82% of the remaining nitrate was consumed (Fig. 3b). The complete uptake of phosphate was not achieved in both experiments (Fig. 3c). Only a modest decrease of the phosphate concentrations was observed: SWSB gave a PO₄³⁻ removal efficiency of 17.8% and SWAA only 11.4% from the high strength (220 mg L⁻¹ PO₄³⁻) synthetic industrial wastewater.

Fig. 3

Nutrient removal performance of Chlorella sp. ACUF_802. Ammonium removal from SWAA media (a), nitrate removal from SWAA (b), and phosphate removal (c) in SWSB and SWAA. Error bars indicate standard deviation, where n = 3

Nitrogen limiting conditions (low N:P ratio)

To simulate the composition of a phosphate-rich wastewater, the first N:P set of experiments was carried out by fixing the N:P ratio lower than 5:1 and by varying the initial nitrogen and phosphorous concentrations (Table 2). During the exponential phase, the four experiments exhibited a similar growth trend, and variations were observed after the 6th day when algal cultures with a lower nitrogen concentration (T 1.1) showed a slightly lower growth rate (Fig. 4a). The highest X_m was observed in the T 1.2 assay (1237.19 mg L⁻¹), and the maximum specific growth rate was observed during the experiment T 1.1 (0.75 day⁻¹) (Table 3).

Fig. 4

Variation of biomass growth (a), chlorophyll α concentration (b), and pH (c) of Chlorella sp. ACUF_802 growing at different nitrogen concentrations with a N:P ratio close to 1. Treatment symbols: T 1.1 (diamonds), T 1.2 (squares), T 1.3 (triangle), and T 1.4 (circles). Error bars indicate standard deviation, where n = 3

The Chl α concentration had a strong relation with the initial nutrient concentrations. Experiments T 1:3 and T 1.4, with the highest N and P concentration yielded the highest Chl α concentration (4784.8 ± 213.6 and 4012.8 ± 21.6 μg L⁻¹ Chl α, respectively), while the experiment T 1.1 gave the lowest Chl α concentration (3289.3 ± 87.9 μg L⁻¹ Chl α) (Fig. 4b). A similar pH trend was observed in the T 1.2 and T 1.4 experiments: both reached the maximum pH value at the third day of incubation (pH > 11), which decreased after the 6th day (Fig. 4c). The nitrate removal rate was faster in experiment T 1.1 (depleted on day 2) than in the other experiments, which showed a complete removal around the 6th day of incubation (Fig. 5a). The highest phosphate removal efficiency was achieved in the experiments T 1.2 and T 1.1, 37.4 and 31.7%, respectively (Fig. 5b).

Fig. 5

Nitrate (a) and phosphate (b) removal with different nitrogen concentrations when incubating Chlorella sp. ACUF_802 in media with a N:P ratio close to 1. T 1.1 (diamonds), T 1.2 (squares), T 1.3 (triangle), and T 1.4 (circles). Error bars indicate standard deviation, where n = 3

Phosphorous limiting conditions (high N:P ratio)

Chlorella sp. ACUF_802 was incubated in a medium characterized by a N:P ratio higher than 5:1 (initial P and N for the 3 conditions reported in Table 2). The exponential growth phase was similar for the different experiments: the medium with a higher nitrate concentration (T 2.1) exhibited a lower growth rate than the other experiments until day 8. The exponential growth phase lasted until day 10 (Fig. 6a).

Fig. 6

Variation of biomass growth (a), chlorophyll α concentration (b), and pH (c) of Chlorella sp. ACUF_802 growing at different nutrient concentrations at a 5:1 NP ratio. Treatment symbols: T 2.3 (diamonds), T 2.2 (squares), and T 2.1 (triangle). Error bars indicate standard deviation, where n = 3

The Chl α variation showed a significant relationship with nutrient availability: a higher nutrient concentration resulted in higher Chl α accumulation (T 2.1; Fig. 6b). Nevertheless, T 2.1 did not achieve the highest biomass concentrations in this experiment. T 2.3 biomass concentration was higher at the 10th day of incubation than in the T 2.1 experiment. In the case of X_m and P_b, the higher values were achieved in the experiment T 2.3 (1488.09 mg L⁻¹ and 96.85 mg L⁻¹ day⁻¹) than in all the others assays. On the contrary, T 2.2 showed the highest μ_m (0.80 day⁻¹) (Table 3). As expected, due to the different initial nitrogen concentrations, a different incubation time was required to achieve a complete nitrate removal from the media: experiment T 2.1 exhibited a longer nitrate removal time (14 days), whereas a higher nitrogen removal efficiency was achieved in T 2.3 at the 4th day (Fig. 7a). Complete phosphate removal was observed in this assay, but at a different time: experiment T 2.2 showed the highest phosphate removal efficiency in 4 days (Fig. 7b).

Fig. 7

Nitrate (a) and phosphate (b) removal of Chlorella sp. ACUF_802 growing at different nutrient concentrations at a 5:1 N:P ratio. Treatment symbols: T 2.3 (diamonds), T 2.2 (squares), and T 2.1 (triangle). Error bars indicate standard deviation, where n = 3

Discussion

Species identification

Chlorella (green algae; Chlorophyta) is a cosmopolitan genus living in both aquatic and terrestrial habitats (Masojídek and Torzillo 2008). Furthermore, Chlorella and a few other green algae of simple morphology are often summarized as the so-called airborne algae (Sharma and Rai 2010), which are ubiquitous in terrestrial ecosystems (Rindi 2011). In the systematics of coccoid green algae, the identification of the Chlorella species is a difficult task (Krienitz et al. 2004). There are a large number of algae from various taxonomic groups showing the same morphological characteristics and the classification of coccoid unicellular algae belonging to Chlorophyta requires a molecular approach. However, a recent study carried out on 400 internal transcribed spacer 2 (ITS2) and/or 18S ribosomal RNA sequences of Chlorella and related taxa has confirmed that Chlorellaceae consist of a Parachlorella and a Chlorella-clade (Krienitz et al. 2004) and that Chlorella is polyphyletic (Heeg and Wolf 2015). For this reason, a clear-cut species designation can be problematic, as in the case of the Chlorella strain ACUF_802. The close relation obtained with the ITS phylogenetic tree suggests that our isolated strain can correspond to another Chlorella sp. isolated from wastewater (Chlorella sp. Iso4) (Electronic supplementary material Fig. S2), which still requires an appropriate taxonomical assignment; therefore, for a cautious designation, our isolate was determined as Chlorella sp. ACUF_802.

Growth kinetics and nutrient removal performance on different wastewaters

Chlorella sp. ACUF_802 can change its metabolic pathway according to the supply of organic substrates, suggesting that it can sustain heterotrophic growth in addition to common autotrophic growth (Tan et al. 2014). In our study, the ability of the isolated strain to perform mixotrophic and autotrophic growth was confirmed. This adaptive behavior could be very promising in the field of wastewater treatment as differences in wastewater composition can relatively affect the nutrient removal performed by adaptive microalgae strains. The growth rate in the assay with the SWAA medium was higher than in SWSB. However, SWSB showed a higher μ_m and P_b (Fig. 2; Table 3). In another study, Chlorella vulgaris biomass growth was extremely different under diverse carbon supply; in particular, the growth on acetate salts was higher than all other examined carbon sources (Battah et al. 2013). The biomass concentrations of C. vulgaris (2.62 g L⁻¹) in mixotrophic growth were 140% higher than autotrophic growth (Mirzaie et al. 2016). Moreover, the total consumption of nitrate from the SWSB medium was obtained at the second day of the assay (Fig. 3), which means that nitrogen depletion was a limiting factor in this experiment.

In this study, the initial pH values were between 6.8 and 8.9 (Fig. 2), which are optimal conditions for the establishment of the cultures (Park et al. 2011). However, the pH values declined at the end of the exponential growth phase in both the SWSB and SWAA experiments, afters day 6 and 8, respectively, due to a decrease in photosynthetic activity (Fig. 2). In addition, HCO₃⁻ consumption during photosynthesis in autotrophic conditions usually raises the pH value of the growth medium (Lin et al. 2007; Park et al. 2011). Cell suspensions with high cell densities and without addition of CO₂ gas may attain pH values as high as 11 (Grobbelaar 2007). In this study, pH values as high as 10.5 were observed in SWSB medium without any pH adjustment (Fig. 2). Zhang et al. (2014b) and Su et al. (2012) observed a similar pH trend of algal reactors, including Phormidium sp., Chlamydomonas reinhardtii, Chlorella sp., C. vulgaris, and Scenedesmus rubescens, which changed from the initial value of 7.0 to around 10.0 after 7 days of cultivation for the five algae species.

The elevated pH can enhance the nitrogen removal from the medium via ammonia volatilization and the phosphorus removal through phosphate precipitation (Craggs et al. 2005). Therefore, the pH can influence both the algal growth and the nitrogen and phosphorus removal efficiency of wastewater treatment systems (Issa et al. 2015). Nevertheless, the highest phosphate removal percentage in SWSB experiments with high P concentration (200 mg L⁻¹) did not exceed 17.8% (Fig. 3). Ammonia volatilization was not observed in the SWAA test due to the lower maximum pH achieved. It is known that the pH rises when sodium or potassium salts of acetate are used as the substrate (Pérez-Garcia et al. 2011). This occurs because the remaining Na⁺ or K⁺ couple with hydroxyl ions (OH⁻) or other anions to form alkali (Pérez-Garcia et al. 2011). This phenomenon was also observed in this study (Fig. 2), although the conditions were not heterotrophic (without dark exposure), but mixotrophic. Nevertheless, the effect on the increase in pH was low compared to the other assay conditions.

Two drawbacks about using bicarbonate as a carbon source have been addressed: the high cost (more than three times that of gaseous CO₂) and the need of specific microalgae species which tolerate high pH and high ionic strength (Markou et al. 2014). However, bicarbonate salts have a higher solubility when compared to CO₂ (NaHCO₃ solubility is > 90 g L⁻¹ at 25 °C) and it is expected that its application efficiency would be higher than that of CO₂ application. In the study of Chi et al. (2013), the cyanobacterium biomass productivity in a reactor was in the range of 0.11–0.13 g L⁻¹ day⁻¹, which is low for lab-scale photobioreactors. In this study, the Chlorella sp. strain ACUF_802 gained three times higher (than Chi et al. 2013) biomass concentration using the same carbon source (SWSB) (Table 3), even if the initial carbon source concentration was six times lower (0.024 M) than the lowest concentration (0.16 M) tested by Chi et al. (2013).

A decline of the biomass and chlorophyll α production was observed in the SWSB experiment when the culture reached pH values over 10 (Fig. 2). Simultaneously, a change in the color of the microalgae culture from green to yellow was observed and it became lighter until the end of the assessment (data not shown). The pH affects the growth of the microalgae, and different species and strains have different optimal pH values at which the fastest growth is achieved (Gong et al. 2014). The general optimum pH for most freshwater algal species is between 7 and 9 (Gong et al. 2014). Failure to keep the correct pH can lead to slow growth or the culture to collapse. During the first 3 to 6 days, a fast growth and nutrient consumption in the SWSB experiment was observed that rapidly increased the pH values (Figs. 2 and 5) and caused a reduction of the availability of the carbon source and the precipitation of the phosphate in the medium (Fig. 3).

Wu et al. (2014) reported that C. vulgaris is able to grow in nitrogen and phosphorus rich wastewaters. These microalgae can exhibit autotrophic and mixotrophic growth showing differences in biomass production. The authors summarized that for phototrophic growth, the X_m can vary from 700 to 1500 mg L⁻¹ and for mixotrophic growth from 400 to 2000 mg L⁻¹. These values were higher compared to the modeled data in this study under SWSB and SWAA conditions, 350.2 mg L⁻¹ and 485.21 mg L⁻¹, respectively (Table 3). Additionally, Wu et al. (2014) reported a biomass concentration in the range from 100 to 600 mg L⁻¹ for low N and P concentration experiments (12.76 mg L⁻¹ NH₄⁺∙and 2.0 mg L⁻¹ PO₄³⁻∙). These values are similar to SWSB in the current study, probably due to the fast assimilation of the nitrogen. The preference for ammonia in the SWAA confirms the observations in other studies: similar results were observed by Liang et al. (2009) in their study on C. vulgaris under mixotrophic conditions, which showed improved biomass production with 1 and 2% glycerol, but inhibition with 5 and 10% glucose. Lin et al. (2007) found that an isolated strain of C. pyrenoidosa from landfill leachate was well adapted to high ammonium concentrations. Inhibition of biomass growth was observed only in cultures with an ammonia concentration exceeding 135 mg L⁻¹ (Lin et al. 2007).

BBM and anaerobic digestion effluent from beef steers cattle waste were tested by Kobayashi et al. (2013) to cultivate three Chlorella sorokiniana strains aiming for biomass production (Table 1). The authors found that the biomass productivities in all the strains were 16–21 mg L⁻¹ day⁻¹. However, the results indicated that anaerobically digested effluents marginally suppressed the growth of two strains and inhibited the growth of the third one. In this study, the highest maximum biomass productivity and batch productivity were obtained when Chlorella sp. ACUF_802 was grown in SWSB (416.51 and 71.43 mg L⁻¹ day⁻¹, respectively). On the other hand, for the two C. sorokiniana strains studied by Kobayashi et al. (2013), the removal of TP and phosphate was 12% and 24%, respectively, and the nitrogen (TN) removal was 90%. Comparably, Kong et al. (2010) cultivated Chlamydomonas reinhardtii during 10 days in a centrated municipal wastewater and found 83% and 15% N and P removal, respectively. Similar nitrate (> 95%) and phosphate (16.2%) removal efficiencies were observed in this study, which suggests a similar tolerance of Chlorella sp. ACUF_802 to high nutrient concentrations (Fig. 3). Kobayashi et al. (2013) suggested that the degree of nutrient removal may be influenced by the effluent type and Chlorella strain. Therefore, the low nutrient removal and biomass growth rate of the three C. sorokiniana strains tested could be due to the inability of these strains to effectively utilize the available nutrients.

Some strains like C. pyrenoidosa generate more ATP from glucose than the autotrophic and mixotrophic cultures with light as energy supply (Yang et al. 2000). In the present study, mixotrophic and autotrophic growths at high nitrate and phosphate concentrations were observed (Table 3). Some microalgae species are not truly mixotrophic but have the ability of switching between phototrophic and heterotrophic metabolism depending on the environmental conditions (Zhang et al. 2014a; Pérez-Garcia and Bashan 2015). This could in part explain the difference in the carbon source preference observed in the SWAA assay.

Low biomass concentration values were measured for Chlorella sp. ACUF_802 tested with the different carbon sources (Fig. 2a). On the contrary, high biomass concentration and productivity were observed in axenic cultures of these strains. The authors found for Scenedesmus obliquus, Chlorella vulgaris, and Chlorella kessleri and a natural bloom a higher biomass concentration in real wastewater effluent than in the synthetic wastewater. After 10 days, C. vulgaris showed a maximum biomass concentration of 821 (± 88) mg L⁻¹ in synthetic wastewater and 1303 (± 270) mg L⁻¹ in real wastewater and 5% CO₂ in air (Arbib et al. 2014). The dissimilarity with the results presented in this study can be related to the low N:P ratios adopted, due to the use of P as a non-limiting nutrient and the pH control applied by Arbib et al. (2014).

The variability in productivity values reported by different authors can be due to the wide multiplicity of the experimental conditions employed: light intensity, light/dark cycles, culture medium, microalgae species, temperature, aeration, source, and proportion of carbon dioxide and photobioreactor type. As mentioned by Arbib et al. (2013), the culture time is different between the studies or the criteria used to stop a test are not indicated, which leads to productivity values depending on the duration of the stationary phase. On the other hand, some authors stop the experiments before the stationary phase; in those cases, the productivities are only partial, and the maximum biomass concentration achieved in the reactors is unknown.

Effect of nitrogen and phosphorus ratios on the growth kinetics and nutrient removal efficiency

It is generally assumed that the standard composition for microalgae is the Redfield ratio C₁₀₆H₁₈₁O₄₅N₁₆P (Redfield 1958). This would mean that a culture medium with a N:P ratio (molar) of 16 is required for a balanced growth of microalgae. This formula is usually considered as the baseline of possible nutrient limitation measurements (Arbib et al. 2013). It is noteworthy that many commercial nutrient solutions for growing microalgae are far from this N:P ratio proposed by Redfield. Other researchers report an optimum N:P ratio between 5:1 and 8:1 (molar) for microalgae growth and conclude that the Redfield ratio is not an universal biochemical optimum (Klausmeier et al. 2004; Li et al. 2010). Determination of the proper N:P ratio for a microalgae system for wastewater treatment is a research topic that still requires attention. Table 3 shows the productivity increased four times when adjusting the N:P ratio to 2.79 (T 1.2). This was a higher increase than the one obtained with different concentrations maintaining a N:P ratio range between 5:1 and 8:1, as recommended by the literature (Ammary 2004; Klausmeier et al. 2004).

Chlorella sp. ACUF_802 consumed the nitrogen source fast, resulting in a low biomass productivity (Fig. 4a) and decreasing of photosynthetic activity after day 6 of incubation as observed in the T 1.1 experiment (Fig. 4b). This photosynthetic behavior is supported by the results obtained by Da Silva et al. (2009), where a decrease in the concentration of the nitrogen sources led to a decrease in biomass growth, chlorophyll α concentration and microalgae biomass productivity of a marine microalga (Rhodomonas sp.). Nitrogen depletion triggered a rapid decline in nitrogen containing compounds such as photosynthetic pigments (Da Silva et al. 2009), causing strong decline of the photosynthetic efficiency. Wang et al. (2013) reported that among four different initial nitrogen concentrations applied, Oedogonium sp. grew most rapidly in the 1/4BG-11 nitrogen (64 mg L⁻¹ N) culture medium during the first 3 days. This is in agreement with the trend observed in the T 2.3 predicted growth by the logistic model in the present study (Fig. 4). The logistic curve growth, which represents a sinusoidal curve, fitted the experimental data on the initial day of cultivation until day 11. After that, the predicted model trend showed a higher biomass concentration at day 14. Experimentally, the growth of Chlorella sp. ACUF_802 showed a lag phase starting on day zero of cultivation until day 4 and increased linearly until the maximum growth (on day 7of the assay T2.2 and on day 10 in the assays T2.1 and T2.3) was achieved. However, from day 10 onwards, Chlorella sp. ACUF_802 growth decreased until day 14. A similar growth trend was reported by Jamaian et al. (2017) and Gani et al. (2017) for other green algae (i.e., Botryococcus sp.) utilizing domestic wastewater. The lack of further sampling days during the stationary phase (later than day 10 for the assays T2.1 and T2.3) might be the reason of the higher maximum biomass concentrations predicted at the end of the experiments by the model compared with the experimental data.

The increase of the nitrogen concentration (higher N:P ratios) can enhance the maximum final biomass concentration achieved (Table 3), while the phosphorous increase did not support an enhancement of the final biomass concentration (Da Silva et al. 2009). This means that not phosphorous, but nitrogen, can be considered as the limiting nutrient in wastewater. The proper N:P ratio for nitrogen removal (usually around N:P = 9) has to ensure simultaneously a high removal efficiency and a short lag phase. Moreover, an important effect of the N:P ratio is the capability of biomass generation. Variations of the N:P ratio of a wastewater greatly affects the nitrogen removal efficiency (Arbib et al. 2013). At N:P ratios between 1 and 13, Arbib et al. (2013) obtained an average nitrogen removal efficiency of 91.6 (± 2.19)%. A similar maximum removal efficiency (95%) was reported for Scenedesmus dimorphus in diluted agro-industrial wastewater (N:P = 0.57) by Gonzalez et al. (1997). In this study, it was observed that with a N:P ratio of 5:1, complete removal of nitrogen can be achieved in 14 days using the microalgae strain Chlorella sp. ACUF_802 (Fig. 7) when the N and P concentrations were 56 mg N L⁻¹ and 16 mg P L⁻¹, respectively (T 2.1).

Conclusions

The strain Chlorella sp. ACUF_802, native of a nitrate rich wastewater, is well adapted to high nutrient concentrations, and it can perform nutrient removal from wastewater. Nitrogen in synthetic wastewaters simulating highly concentrated industrial wastewaters was effectively utilized by Chlorella sp. ACUF_802 to achieve a batch productivity of 71.43 mg L⁻¹ day⁻¹. Bicarbonate-based synthetic wastewater showed a faster nitrate removal, although the maximal P removal efficiency (achieved using the ammonium acetate based synthetic wastewater) was only 16.2%. The productivity of Chlorella sp. ACUF_802 could be increased four times by adjusting the N:P ratio to 1.3 (T 1.2). The strain showed a high biomass growth and productivity in the SWAA medium. Chlorella sp. ACUF_802 could be used to treat highly concentrated industrial wastewaters, thus combining nutrient removal from wastewater with the reduction of the biomass production cost.