ASABE Logo

Article Request Page ASABE Journal Article

Long Term Outdoor Algae Production on Undiluted Anaerobic Digestate in the Southeastern US

Qichen Wang1,*, Brendan T. Higgins1

Published in Journal of the ASABE 67(1): 181-192 (doi: 10.13031/ja.15727). Copyright 2024 American Society of Agricultural and Biological Engineers.

1Department of Biosystems Engineering, Auburn University, Auburn, Alabama, USA.

*Correspondence: qsw0002@auburn.edu

The authors have paid for open access for this article. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License https://creative commons.org/licenses/by-nc-nd/4.0/

Submitted for review on 5 July 2023 as manuscript number NRES 15727; approved for publication as a Research Article by Associate Editor Dr. Sushant Mehan and Community Editor Dr. Kati Migliaccio of the Natural Resources & Environmental Systems Community of ASABE on 6 November 2023.

Highlights

Abstract. Growing algae on anaerobic digestate could decrease the algal production cost while reducing nutrient pollution. In past work, we developed a successful biological pretreatment for digestate that enables rapid algal growth on digestate without dilution. The objective of this work was to test the effectiveness of this pretreatment in outdoor algae cultures over a year-long timeframe. The study was conducted in semi-continuous, replicated bubble column photobioreactors in Auburn, AL, USA. Algae could grow successfully in pretreated digestate during the fall, spring, and summer, achieving average growth rates of 30, 42, and 66 mg L-1 d-1, respectively. Although the cold temperatures in winter suppressed algal growth, external heating was not required to keep the cultures alive. For two weeks during the summer, the system was challenged with 100% digestate that did not undergo pretreatment, and the algal community suffered a culture collapse with a significant (p < 0.001) decrease in productivity compared to the previous batches in which pretreatment was used. Nitrogen, phosphorus, and sulfur removal were observed during robust algal growth. There was no nitrification in the photobioreactors during the initial 200 days, but evidence of nitrification emerged during the summer and persisted into the fall. Nitrospirae were detected by 16S rRNA sequencing, proving that nitrifying bacteria could adapt to high ammonium (462 - 1502 mg/L). The eukaryotic community was dominated by Coelastrum (>90%) in the cold season, and the dominance transitioned to Chlorella in the warm season (>95%) based on 18S rRNA sequencing. The low relative abundance of cyanobacteria showed that green algae were the favored photosynthetic organisms in the system.

Keywords. Biogas effluent, Microalgae, Nitrification, Nutrient recycling, Wastewater.

Production of microalgal biomass on nutrient-rich wastewater, such as anaerobic digestate, holds the potential to mitigate environmental problems and enable a circular economy (Fuentes-Grünewald et al., 2021). Nutrients in the digestate are assimilated into the algal cells, reducing eutrophication potential toward downstream waterbodies. Algal treatment systems also produce biomass, which can be used in a variety of applications, such as animal feeds and biofuels (Sarwer et al., 2022; Viegas et al., 2021). Algal biomass has been promoted as a third-generation bioenergy feedstock by the Department of Energy in the United States largely because it is more productive than plant crops per unit area (Smith and Crews, 2014). These biological “solar panels” store energy in the form of lipids, proteins, and carbohydrates (Vuppaladadiyam et al., 2018). However, before algal biomass can be widely accepted as a biomass feedstock, more research is needed to focus on solving two major issues: 1. reducing the production cost and 2. improving system resilience toward unstable culture conditions (Kumar et al., 2021).

Utilizing nutrients from anaerobic digestate could reduce the cost and environmental impacts of using fertilizer to grow algae. Consequently, research focused on using anaerobic digestate as algal culture media is gaining momentum. Different types of digestate have been tested using different strains of algae. Ayre et al. (2021) tested Chlorella on swine manure digestate; Bjornsson et al. (2013) tested Scenedesmus on vegetable waste digestate; Chaiprapat et al. (2017) tested Chlorella on food processing waste digestate; Nguyen et al. (2019) tested Scenedesmus on textile wastewater digestate; Feng et al. (2020) tested Chlorella on dairy manure digestate; and Barzee et al. (2022) tested Chlorella on food waste digestate. However, most of the studies focused on maximizing biomass productivity by using robust algal strains, and most grew under indoor conditions. With the exception of Ayre et al. (2017), all of these studies diluted the digestate by 2-10 fold using freshwater to achieve robust algal growth. This is done to overcome algal growth inhibition on full-strength anaerobic digestate. Large-scale implementation of such processes using semi-continuous batch or plug flow reactors (PFR) would lead to massive demand for water resources. Semi-continuous batch or PFR operations are attractive for algae culture because they allow for rapid growth and nutrient removal during periods of moderate algal density (e.g., 0.2-1 g L-1) followed by harvesting at high algal density (>1 g L-1). In contrast, continuous stirred tank reactors enable good effluent quality but require either a trade-off in algal growth (operating at high culture density that blocks light) or harvest efficiency (operating at low culture density).

In past lab-scale studies, we developed a biological pretreatment process that enables rapid algal growth on full-strength anaerobic digestates, including those from municipal sludge, food waste, and dairy manure (Wang et al., 2021b, 2019). This system enabled rapid growth of Chlorella, Auxenochlorella, and a local algae consortium (dominated by Coelastrum) with average 5-day growth rates of up to 500 mg L-1 d-1, >400 mg L-1 d-1, and 184 mg L-1 d-1, respectively (Wang et al., 2021a, 2021b, 2019). These values exceed the growth rates reported by researchers who used similar genera but grew them on diluted digestate (Prandini et al., 2016). Of the ten studies reviewed by Prandini et al. that used real anaerobic digestate, only two had growth rates that exceeded 100 mg L-1 d-1 despite being grown indoors under well-controlled conditions (Prandini et al., 2016). Faster algae growth enables smaller reactor sizes, saving on cost. Moreover, the dilution approach typically requires 2-10 volumes of water per volume of digestate, making it particularly unattractive in water-limited regions. Our system has been proven using indoor algae cultures, but it was never tested year-round in an outdoor environment where seasonal variation can greatly complicate biological system operation.

We hypothesized that the benefits of biological pretreatment would extend to outdoor algal cultures grown on full-strength anaerobic digestate. An outdoor algal production system was operated over a one-year period in Auburn, Alabama, USA. The aims of this study were to (1) test the effectiveness of biological pretreatment for algal growth in an outdoor environment; (2) test the stability of outdoor biomass production and nutrient removal when using full-strength municipal anaerobic digestate across four seasons; and (3) elucidate the dynamics of microbial community succession over time. This study advances knowledge of how algal consortia respond to pretreatment and changing environmental conditions over a long time-horizon.

Materials and Methods

Anaerobic Digestate and Activated Sludge

Both anaerobic digestate and activated sludge were collected from a municipal wastewater treatment plant operated by Columbus Water Works in Georgia, USA, every six months. Raw municipal anaerobic digestate (MAD) was directly pumped from a commercial-scale (~4335 m3) mesophilic sludge digester. Belt press filtered digestate (BMAD) was also collected from the effluent of the digestate belt press filtration unit located at the wastewater treatment plant. The composition of these digestates before and after pretreatment is shown in table 1. Digestate and activated sludge were immediately transported back to Auburn University and stored at 4?.

Table 1. Anaerobic Digestate Compositions.
Concentration
(mg L-1)		MAD [a]BMAD [b]Pretreated MADPretreated BMAD
COD875-1117140-220870-1122103
Ammonium994-1501356-541536-1002205-433
NitriteN.D. [c]N.D.N.D.N.D.
NitrateN.D.N.D.N.D.N.D.
Phosphorate95-619283-355464-544118
Sulfate0.825-47.733.0-46.515.3-42.741.5

    [a]MAD: Municipal anaerobic digestate.

    [b]BMAD: belt press filtered municipal anaerobic digestate.

    [c]N.D. means not detected.

Activated Sludge Pretreatment of Anaerobic Digestate

Both MAD and BMAD were biologically pretreated with activated sludge before addition to the algal reactors, with the exception of a 2-week period from 1 July 2021 to 14 July 2021 when pretreatment was purposefully withheld. For pretreatment, digestate was centrifuged at 4,696 x g for 10 minutes. Digestate supernatant (1 L) was transferred into a 2 L bubble column bioreactor for biological pretreatment. Activated sludge (1% v/v) was added to the reactor to achieve a solids content of around 0.067 g L-1, and the mix was aerated at 0.1 vvm for 1 day or 4 days (Wang et al., 2021b), depending on the experimental plan. Centrifugation (4,649 x g, 5 min) and filtration with VWR No. 415 (25 µm), 413 (5 µm), and then 410 filter paper (1 µm) was used to improve digestate clarity. pH was adjusted to 7.5 - 8.0 with concentrated hydrochloric acid (12 M). For the two weeks when digestate was added to the system without biological pretreatment, only centrifugation and filtration (down to 1 µm) followed by pH adjustment were used.

Algal-Bacterial Consortium Preparation

The outdoor photobioreactors were inoculated with an algal consortium that was preadapted to MAD and described in detail in a previous study (Wang et al., 2021a). Briefly, 500 ml of the original algal consortium was collected from the top of a clarifier that was connected to the fish production tank in Auburn University’s aquaponics system (Kalvakaalva et al., 2023). The consortium was first adapted to biologically pretreated dairy manure digestate, followed by adaptation to pretreated MAD. The adapted consortium was placed under household fluorescent light (164 µmol s-1 m-2 on a 16 h:8 h light/dark cycle) with mixing at room temperature to maintain organism viability.

Semi-Continuous Outdoor Algae Cultivation

The outdoor cultivation was conducted from 6 October 2020 to 19 October 2021. The algal consortium was grown in 2 L polyethylene terephthalate (PET) bottles (without lids) that were operated as bubble column reactors (fig. 1). Each reactor was filled initially with 700 ml of pretreated MAD and 300 ml adapted consortium (dry mass concentration 0.65 g L-1 for the latter), resulting in a 1 L working volume in each bottle. The bottles were oriented at 45° and aligned in a north-south orientation to capture southern sunlight. An aluminum reflector was placed beneath the bottles and the whole apparatus was partially suspended in a 20 L tub of water to modulate temperature swings in the bottles. 500 ml min-1 air with 2% carbon dioxide (CO2) supplementation was aerated into each bottle reactor. Compressed air was moisturized by going through a humidifier bottle. Digestate volume was checked and maintained at 1L in each bubble column reactor during sampling. Deionized water was added to compensate for losses associated with evaporation as needed. Samples (2 ml) were taken from each bottle reactor for the measurement of optical density (550 nm and 680 nm) and pH every 2-3 days. After OD measurement, the samples were centrifuged (13,201 x g, 5 minutes) for solid-liquid separation. The pellets were stored at -80? for DNA extraction. The supernatants were syringe filtered with 0.2 µm Nylon filters for ion chromatography analysis. In fall, spring, and summer, 700 ml of algae culture was harvested whenever OD 550 exceeded 1.5 and replaced with 700 ml of freshly pretreated and filtered digestate. Harvested algal biomass was washed with deionized water three times before freeze drying for the measure of dry weight concentration, followed with an OD – dry weight correlation to estimate the dry weight concentration in each sample point. This generally resulted in the harvesting and replacement of the cultures every 7-14 days based on the growth of biomass. The net effect was a semi-continuous reactor operation with a hydraulic residence time of 10-20 days. Average productivity was calculated for each batch by determining the change in concentration and dividing it by the batch period. Harvest was not conducted during wintertime due to minimal biomass productivity. Table 2 shows a set of naturally occurring and researcher-induced events that were expected to impact algal growth. The researcher-induced events included switching from 4-day to 1-day pretreatment of MAD in March of 2021, withholding pretreatment of digestate for two weeks in July, and a switch from MAD to BMAD starting in the late summer of 2021. The latter was done to save effort on the initial clarification of the digestate, considering that solids removal was already being carried out at the wastewater treatment plant.

Figure 1. Diagram for outdoor photobioreactors.

Analytical Measurements

Ion Chromatography (Cations and Anions)

Suppressed ion chromatography was conducted on water samples using a Prominence High Performance Liquid Chromatography system (Shimadzu, Japan) and ThermoFisher Scientific columns and suppressors for both cations (Na+, K+, NH4+) and anions (Cl-, NO2-, NO3-, PO43-, and SO42-) as described previously (Wang et al., 2019).

16S and 18S rDNA Amplicon Sequencing

DNA was extracted from the preserved biomass pellets collected at the time points shown in table 2. The FastDNA Spin kit (MP biomedicals, USA) was used. The extracted DNA was quantified with the QuanitFluor dsDNA quantification kit (Promega, USA). DNA samples were sent to a sequencing center for 16S rRNA (primers: 515F-806R) and 18S rRNA (primers: 1391F and EukBr) bTEFAP amplicon DNA sequencing using an Illumina MiSeq instrument. Polymerase Chain Reaction (PCR) was performed for 30 cycles using the HotStarTaq Plus Mix Kit (Qiagen, USA) under the following conditions: 95°C for 5 minutes, followed by 30-35 cycles of 95°C for 30 seconds, 53°C for 40 seconds, and 72°C for 1 minute. A final elongation step at 72°C for 10 minutes was performed. PCR products were checked using a 2% agarose gel, and the sequence data was processed using the sequencing center’s ribosomal and functional gene analysis pipeline (MR DNA, TX) as described previously (Wang et al., 2021a). The original FASTQ files from this work have been uploaded to NCBI’s Sequence Read Archive (SRA) under the accessions SUB12956463 for 16S rRNA sequences and under SUB12947560 for 18S rRNA sequences.

Weather Data

The weather data for Auburn, AL, (temperature and precipitation) were downloaded from the website of the National Environmental Satellite Data and Information Service (National Oceanic & Atmospheric Administration). Solar radiation data was downloaded from the website of the National Renewable Energy Lab (National solar radiation database supported by the US Department of Energy).

Statistics

The microbial community diversity (Shannon), richness, and evenness were analyzed using the “vegan” package (Oksanen et al., 2017) in R. Similarity percentage breakdown analysis (SIMPER) was used for comparing the differences in community structure between two selected event points. The comparison between algal growth and nutrient removal was analyzed with Tukey’s multiple comparison test using the “agricolae” (De Mendiburu, 2014) and “car” (Fox and Weisberg, 2018) packages in R, with a significance level of 0.05. Two-tailed independent t-tests were used for analyzing the difference of batch average biomass productivity before and after changes in pretreatments (4-day versus 1-day and pretreated versus not pretreated).

Table 2. List of natural and researcher-induced events affecting the algal system.
EventsDescriptionDate
Initial consortium Samples of the initial consortium which was preadapted in lab scale photobioreactors6/10/2020
Fall 2020 first bloomSamples of the consortium during an algal bloom7/11/2020
4Day pretreated digestate bloomSamples of the consortium cultured in 4-day pretreated digestate23/2/2021
1Day pretreated digestate bloomSamples of the consortium cultured after switch to 1-day pretreated digestate11/3/2021
Intense NitrificationSamples when high nitrate concentration was first observed in the culture.1/7/2021
After 2 weeks of untreated digestateSamples of the consortium cultured in untreated digestate (no pretreatment challenge)14/7/2021
Back to 1Day pretreated digestateSamples of the recovering consortium culture after no pretreatment challenge30/8/2021
1Day pretreated belt press filtered digestateSamples of the consortium cultured in 1-day pretreated digestate with belt press filtration17/9/2021

Results and Discussion

Biomass Production Across Seasons

Algae were able to grow and maintain a high biomass concentration over the full year, with the exception of a period in the early winter and the period following cessation of digestate pretreatment in the early summer (fig. 2). The lag phase lasted around a week in October of 2020, followed by a dramatic increase of biomass concentration through the end of October. One reason for this sharp increase was the addition of a water bath to stabilize the reactor temperature. Initially, the 2 L photobioreactors were directly exposed under sunlight without a water bath. The greenhouse effect inside the 2 L bubble column reactor heated the media to over 39? on a sunny day, even when the highest air temperature was around 28?. After adding around 20 L of water in the secondary container, the temperature fluctuations became milder. The high temperatures during the first week were likely an artifact of the small system size because large open algae systems have been shown to carry enough water to suppress temperature fluctuations (Luo et al., 2020).

It was surprising that biomass concentration was maintained at a relatively high level (~1.5 g L-1) during wintertime, with an average productivity of 23 mg L-1 d-1. The cultures were visibly green despite the low temperatures and low growth. Periods of freezing temperature at night (fig. 3A) resulted in ice layer buildup in the water bath, but continuous culture aeration prevented the bioreactors from freezing. Although the daily lows dipped below the freezing point during cold fronts, the daily highs never fell below 0?. There was a negligible impact on the algal biomass concentration after a cold front passed through Auburn, AL (at day 55). Algal biomass productivity increased in late winter, when the daily high temperatures were typically above 15?, followed by a robust spring algal bloom. The average biomass productivity during spring was 42 mg L-1 d-1. Within a batch cycle (5-7 days), the maximum biomass productivity was 239 mg L-1 d-1, which was recorded during the week of March 21st, 2021. The summer period, excluding the two-week period of no-pretreatment as well as the 3-week recovery that followed, had an average productivity of 66 mg L-1 d-1. The fall periods (combining 2020 and 2021) averaged 30 mg L-1 d-1. These values compare favorably to other outdoor growth systems that use anaerobic digestate. Ayre et al. (2017) observed outdoor productivity of 17.4 mg L-1 day-1 on full-strength swine digestate from late August to late September in Australia (equivalent to Feb-March in US). Pizzera et al. (2019) observed 22.2 mg L-1 day-1 productivity in an outdoor photobioreactor using 5-fold diluted swine digestate to which supplemental N and P nutrients were added. They also tested a raceway pond using the same feed strategy and observed 32.4 mg L-1 day-1 productivity (corresponding to 8.2 g m-2 d-1). These values were obtained over a year of operation in Italy on diluted digestate without chemical or biological pretreatment (only filtration) but with nutrient supplementation. Barzee et al. (2022) observed an average productivity of 11.7 mg VS L-1 d-1 in their outdoor study growing C. sorokiniana on diluted anaerobic digestate. This was observed over an approximately one-year period of growth in Davis, CA. They observed a maximum average productivity of 17.7 mg VS L-1 d-1 during October. All of these values are lower than the overall average annual productivity of 40.2 mg L-1 d-1 observed in this study when excluding the period of no pretreatment and its associated recovery. Not surprisingly, this level of productivity was only 22% of the peak observed productivity of this algal consortium in indoor culture (up to 184 mg L-1 d-1) when using pretreated digestate. Although all the above studies were conducted in temperate climates, it is difficult to directly compare productivities across studies given differences in specific weather and reactor configurations.

A switch from 4-day to 1-day of activated sludge pretreatment was conducted in early March once algal growth started to increase with the warming weather (fig. 2). The switch from 4-day pretreatment to 1-day pretreatment resulted in 37% higher batch average biomass productivity (p = 0.006, paired t-test) based on growth in the two weeks prior to the switch and growth in the two weeks after the switch. This result contrasts with findings from a previous indoor study involving Auxenochlorella protothecoides and Chlorella sorokiniana, where a decrease in algal productivity was linked to shorter durations of pretreatment (Wang et al., 2021b). A possible explanation for the divergence is that outdoor conditions have more variables (changing temperature, light intensity, predators, etc.), which could overshadow the impact of using 4-day versus 1-day pretreatment on anaerobic digestate (figs. 3A and 3B show relevant weather information). Regardless, this result suggests that switching to the less time- and energy-intensive 1-day pretreatment did not cause apparent harm to the system.

Figure 2. Algal consortium biomass concentration and productivity curves over a one-year timeframe. Productivity before (2 batches) and after (2 batches) the transition from 4-day pretreatment to 1-day pretreatment of MAD (left inset); algal consortium biomass concentration curve before (2 batches) and after (2 batches) the transition from 1-day pretreatment to No pretreatment of MAD (right inset). The experiment was carried out in semi-batch photobioreactors. Error bars represent the standard deviation (n = 3).
Figure 3. The weather information in Auburn, Alabama, during the period of experiment. (A) Temperature and precipitation; and (B) solar insulation.

pH was maintained above 7.5 during robust algal production but decreased when algal cultures were growing poorly (fig. 4). One possible explanation was that CO2 consumption exceeded the acidification effect of continuous CO2 supplementation when algae were growing fast (Gao, 2021). When the algae culture was suffering in either cold temperatures or during the no-pretreatment challenge, the cultures acidified quickly due to excess dissolved CO2 in the digestate. Intermittent CO2 bubbling is commonly used for pH control in algal raceway ponds (Eustance et al., 2020). Typically, in open pond systems, the amount of CO2 supplementation is determined by real-time pH monitoring to avoid over-supply of CO2. However, the continuous supply of CO2 is also a critical bottleneck in large algae production systems. One possible future implication of this digestate-algae system is to utilize the CO2 from biogas combustion (Gong and You, 2014). Upcycling both the liquid and gas waste of anaerobic digestion for value-added algal biomass would potentially have economic and environmental benefits as part of a circular bioeconomy.

Figure 4. pH measurements over a one-year period of algae culture. Error bars represent the standard deviation (n = 3).

The highest biomass concentration (2.6 g L-1) was observed during early summer on June 25th, 2021, with 1 day activated sludge pretreatment. A No pretreatment challenge was conducted on the robust algal culture right after two cycles of rapid growth and harvest (fig. 2). Cessation of the pretreatment resulted in a significant loss in productivity (p < 0.001, paired t-test) dropping from an average of 76.3 mg L-1 d-1 to -44 mg L-1 d-1 over the two-week periods before and after cessation of pretreatment, respectively. The batch initial ammonium and phosphorus concentration during the no pretreatment challenge was consistent with the batches before and after the challenge, which indicated the negative impacts to algal biomass productivity were not related to digestate batch effects. This result further proves that the full-strength anaerobic digestate was inhibitory to algal growth and that the biological pretreatment provided benefits to this outdoor culture. Although the period without pretreatment lasted for only 2 weeks, the system took almost a month to recover and required re-inoculation, likely due to the significant loss of the viable algal population due to the restructuring of the microbial community. Figure 2 shows a continuous decline in biomass concentration after the termination of the no pretreatment challenge.

Nutrient Removal

Nitrogen Removal and Transformation

Depending on the feedstock, anaerobic digestate can contain up to 10 g L-1 of ammonium-nitrogen (Jiang et al., 2019), which can strongly inhibit biological wastewater treatment processes (e.g., nitrification) (Meng et al., 2022). In past work, we have shown that species of Chlorella and Auxenochlorella can grow without significant inhibition in ammonium concentrations up to 3,500 mg L-1 and 1,000 mg L-1, respectively. Ammonium concentration decreased in pretreated MAD (~1000 mg L-1) compared the raw MAD (~1500 mg L-1), which was mostly a result of volatilization since nitrification was severely suppressed in pretreatment reactors (Wang et al., 2019). In the present study, no nitrite or nitrate formation was ever observed during pretreatment, so the likelihood of denitrification contributing to N loss was low. Within each semi-batch cycle, average ammonium removal rates of 23.2 and 75.2 mg L-1 d-1 were observed when using pretreated digestate during the spring and summer of 2021, respectively. Late fall to early winter, on the other hand, did not show much ammonium reduction (fig. 5A), possibly due to a lack of algal productivity during the low temperature days. Interestingly, ammonium removal persisted even during the two-week period when digestate was not pretreated and no algal growth occurred, which suggests the importance of non-assimilatory nitrogen removal. Ammonia volatilization and nitrification are two likely pathways (Bankston and Higgins, 2020).

Figure 5. The overall dynamics of soluble inorganic nutrient concentrations: nitrogen (NH4-N, NO2-N, and NO3-N concentrations (A); Phosphate-P and Sulfate-S concentrations (B). Error bars represent the standard deviation (n = 3).

Evidence of nitrification in the algae consortia first appeared in the spring. Nitrite and nitrate became detectable and showed increasing concentrations into the summer, with a maximum nitrite concentration of 384 mg NO2-N L-1 and a maximum nitrate concentrations of 293 mg NO3-N L-1. After each feeding with new digestate, the nitrite concentration would typically drop below the detection limit, followed by another round of accumulation (fig. 5).

The observation of robust nitrification in full-strength municipal digestate was surprising given past observations of complete inhibition of nitrification when switching from 10x-diluted to full-strength digestate (Wang et al., 2019). Ammonium oxidizing bacteria (AOB) are usually inhibited when the free ammonia concentration is above 290 µM, while nitrite oxidizing bacteria (NOB) are even more sensitive to free ammonia, with a typical inhibitory concentration of 46 µM (Park and Bae, 2009). At a pH of 7.5 (as in the present study), these inhibiting concentrations correspond to ~260 mg L-1 and ~41 mg L-1 of total ammonia nitrogen (TAN), respectively. These are generally far lower than what was observed in the digestate. However, Princic et al. (1998) showed that long-term adaptation of a nitrifying community could enable the selection of particular nitrifying organisms that were able to tolerate up to 3,000 mg L-1 of TAN. During the first 200 days of this trial, the high TAN concentration in MAD explained the lack of nitrification, and it was expected that this inhibition would continue throughout the study.

Ammonium removal was around 70% within each fed-batch cycle when nitrate peaks appeared. Thus, the removal of ammonium by algae could have created more favorable conditions for nitrifying bacteria. This finding aligns with past research showing that algae can promote full oxidation of ammonium to nitrate in diluted anaerobic digestate (Bankston et al., 2020; Wang et al., 2023). Algal promotion of nitrification could enable the use of treated digestate in agricultural applications such as hydroponics that are typically poisoned by high ammonia-N levels (Hachiya et al., 2012).

However, nitrification persisted and even accelerated during the algal die-off following the period of no digestate pretreatment. In fact, the highest nitrate concentration was around 293 mg L-1 in July following the No-pretreat test, before algal growth had recovered. Although algal growth was severely impacted without bacterial pretreatment, the nitrifying bacteria (both AOBs and NOBs) were not impaired at all. Moreover, ammonium concentration in No-pretreat digestate are typically higher than those in pretreated digestate, based on the previous work (table 1). That higher substrate (ammonium) concentration may have enabled higher production of nitrite and nitrate during the early summer period, suggests that the nitrifier community had adapted to a high ammonia-N environment, similar to the observations of Princic et al. (1998).

Phosphorus and Sulfur Removal

The removal of phosphorus in traditional wastewater treatment plants usually requires tertiary treatment with chemical additives and/or the use of enhanced biological phosphorus removal (Bunce et al., 2018). In this algal-bacterial system, up to 60% of P was removed from the digestate (fig. 5B). P removal in algae cultures typically occurs through two mechanisms: algal assimilation for growth and precipitation of P induced by increased pH driven by algal photosynthesis (Zhao et al., 2019). There was almost no P removal during the winter, and negative removal (net increase in phosphate) during the no-pretreatment test in the summer. Since a decline in biomass concentration was observed during the no-pretreatment test, it is likely that phosphorus was released from lysing cells. Up to 100% sulfate was also removed through assimilation by algae, and there was a significant (p < 0.001) correlation between sulfate removal and biomass growth (r = 0.69). This suggests that sulfur was a key limiting nutrient for algae growth in this digestate.

Time-Course Restructuring of the Microbial Community

To understand the succession of the algal and bacterial community over the 1-year period, both 16S and 18S rRNA sequencing were carried out. These data were also used to better understand how the community responded to perturbations of the system, such as weather events or a change in digestate pretreatment.

Eukaryotes

Growth in outdoor conditions decreased community diversity over time. The initial eukaryotic community was obtained through adaptation to MAD in a lab-scale study, as documented previously (Wang et al., 2021a). Coelastrum was the dominant eukaryote with a relative abundance of 72% (October 2020), and this increased to 98% by the following spring (March 2021) (fig. 6). Coelastrum, a green algae genus that is part of the family Scenedesmaceae, has been studied for potential carotenoids (such as astaxanthin) production (Liu et al., 2020). The dominance of Coelastrum was observed during the cool weather periods that encompassed the fall of 2020, winter, and spring periods. As the weather warmed, the eukaryotic community completely restructured, and Chlorella came to dominate this community with over 95% relative abundance throughout late spring, summer, and early fall. The higher temperature during the summer was probably the major force that caused this switch from Coelastrum to Chlorella. Chlorella typically prefers temperatures ranging from 25 to 40? (Li et al., 2013), whereas most Coelastrum strains are cultured around 25? (Tharek et al., 2021). Species of Scenedesmus (relatives of Coelastrum) are likewise known to tolerate colder temperatures than Chlorella (Geller et al., 2018; León-Vaz et al., 2023).

Figure 6. Relative 18S rRNA gene abundance at the genus level at selected time points, as noted in the timeline at the bottom.

When the system was challenged with digestate that did not undergo aerobic pretreatment, the relative abundance of Chlorella dropped from 95% to 70% of eukaryotic gene abundance, while Acrasis increased from 2% to 20%. The genus Acrasis has been described as a slime mold (Hohl and Hamamoto, 1969). It is possible that the change in 18S rRNA gene relative abundance underestimates the culture transition given the short duration of the no-pretreatment challenge and the potential for genetic material to persist despite cell death. Without pretreatment, the culture turned brown, and the algae were likely stressed or killed by the inhibitors in the digestate. The increased organic material from lysing cells likely provided essential nutrients for heterotrophic eukaryotes such as slime molds (Hannen et al., 1999). It was also interesting that DNA from Psychoda (drain fly) was also detected at 6% relative abundance during the period without pretreatment. Only one genera of protozoa (Colpoda) was detected at >1% relative abundance in pretreated MAD, but this protozoon was not detected during the no-pretreatment challenge. Another genre of protozoa, Pseudocohnilembus, was only detected after the switch to using BMAD. Although protozoa represent less than 2% of the total eukaryotic relative abundance, they can play a significant role in the algal community as predators (Tillmann, 2004) and, along with suboptimal weather conditions, can explain lower productivity in outdoor systems compared to indoor systems.

Prokaryotes

The microorganisms that made up the prokaryotic community were sourced from the anaerobic digestate, activated sludge (via pretreatment), and the algal consortium that was originally sourced from an aquaponics facility (Wang et al., 2021a). The community composition underwent continuous restructuring over the course of the year (fig. 7). Firmicutes accounted for approximately 50% of the initial community but constituted less than 1% of the prokaryotic community by the end of the study (fig. 7A). Their die-out began in the summer, and the no-pretreatment challenge apparently led to their near elimination. Firmicutes are tough spore-forming gram-positive bacteria that typically allow them to prosper in extreme conditions (Subirats et al., 2022), and they are often abundant in anaerobic digesters, including the municipal anaerobic digestate used in this study (Wang et al., 2021a). The decreased abundance of Firmicutes may stem from sustained competition with gram-negative bacteria phyla such as Proteobacteria and Bacteroidetes under the more aerobic and neutral pH conditions of the algae cultures. Proteobacteria and Bacteroidetes, which were 16% and 13% of the initial community, expanded up to a maximum relative abundance of 81% and 46% in the system, respectively. A similar shift from firmicutes to proteobacteria occurred very rapidly during batch culture testing of a similar indoor system (Wang et al., 2021a). Despite this increase in proteobacteria over time, many proteobacteria genera did not follow this trend. Of particular note, the opportunistic pathogen Pseudomonas made up ~6% of the initial prokaryotic community (inoculum), but it soon died down to <1% of the community once outdoor cultivation commenced. The presence of microalgae may play an important role in this reduction due to increased dissolved oxygen in the reactors as well as the secretion of organic photosynthetic intermediates (Wang et al., 2022). Interestingly, Synergistetes, a gram-negative anaerobic bacterial phylum, was found to increase during the period of intense nitrification (July 2021) and again during the no pretreatment challenge.

Figure 7. Relative 16S rRNA gene abundance at selected time points, as noted in the timeline at the bottom: Phylum level (A); Genus level (B). The relative abundance of nitrification genes is shown at the Phylum and Genus level (C).

The appearance of the genus Nitrosomonas and the phylum Nitrospirae coincided with the emergence of nitrification in the reactors (fig. 7C). These organisms thrived during the no-pretreatment challenge when Nitrosomonas increased in relative abundance by 3-fold (p < 0.001, 2-tailed independent t-test). These findings provide evidence that the nitrifying community could adapt to the high ammonium digestate environment even in the absence of a robust algae culture. The observation of increased nitrification began in May, when temperatures started to increase. González-Camejo et al. (2019) reported that higher temperatures (over 30 °C) favored AOB activities when algae and bacteria were co-cultured in primary municipal effluent. The abundance of Candidatusnitrosoarchaeum, a putative soil nitrifying archaeon, was high during the cold months of the year, but this organism was not associated with any detectable nitrification in the system.

The relative abundance of cyanobacteria remained below 0.3% relative abundance for all sampling points (fig. 7A). In past research, adaptation of this algal consortium to dairy anaerobic digestate in indoor batch cultures led to an increase in the abundance of Limnothrix up to 10% (Wang et al., 2021a). Subsequent adaptation to pretreated MAD further increased Limnothrix abundance up to 14%, which was concerning since this organism has the potential to produce the limnothrixin toxin. In the outdoor cultures of the present study, Limnothrix did not emerge; rather, the dominant cyanobacteria were Arthrospira, which only made up <0.5% of total prokaryotes. This organism is not known to make toxins, and some species are consumed as a nutrition supplement. Overall, it was viewed as positive that the digestate environment did not facilitate the growth of harmful cyanobacteria. It is our long-term aim to use this algae as feed for fish and zooplankton (Hyman et al., 2021).

Conclusions

An algal consortium adapted to anaerobic digestate was cultivated outdoors under semi-continuous operation over a one-year period on full-strength anaerobic digestate. Robust algal biomass productivity (averaging 40 mg L-1 d-1 over one year) could be achieved provided that biological pretreatment of digestate was used. Cessation of this pretreatment resulted in a complete culture collapse, followed by several weeks of recovery. Ammonium nitrogen from anaerobic digestate decreased up to 70% by assimilation, volatilization, and nitrification during the warm months. Up to 60% of PO4-P removal and 100% SO4-S removal were observed within a batch cycle. The eukaryotic community underwent a moderate restructuring during the late fall, winter, and early spring, and this “cold weather” community was dominated by up to 98% Coelastrum. However, a complete community restructuring was observed during late spring, summer, and early fall as Chlorella came to dominate the culture (up to 95% of eukaryotes). Cessation of biological pretreatment of digestate resulted in a large decline in 18S rRNA associated with Chlorella (95% to 70%) and an increase in the slime mold, Acracis, (2% to 20%). The prokaryotic community also showed a moderate restructuring with gradually decreasing Firmicutes (from 50% to <1%), accompanied by an increase in Proteobacteria and Bacteroidetes. Nitrospirae and Nitrosomonas emerged as temperatures warmed and nitrification became apparent. These organisms persisted even when the algae died out during the challenge with non-pretreated digestate. These results suggest the viability of using pretreated full-strength anaerobic digestate for outdoor cultivation of a green-algae dominant community. This study has been proposed as a guiding experiment for a more comprehensive pilot-scale system as part of the circular economy of agricultural production. Future techno-economic analysis (TEA) and life cycle assessment (LCA) will be conducted using the pilot-scale system.

Acknowledgments

The authors acknowledge William Kent and Brian Boswell at Columbus Water Works, Columbus, GA, for material support and assistance in sample collection.

Funding: This work was supported by the USDA Agricultural Research Service No. 6010-32000-028-000-D, the USDA National Institute of Food and Agriculture, and the Alabama Agricultural Experiment Station [Hatch Project ALA0HIGGINS2 and competitive grant number 2020-67021-31145 and 2023-68012-38994].

References

Ayre, J. M., Mickan, B. S., Jenkins, S. N., & Moheimani, N. R. (2021). Batch cultivation of microalgae in anaerobic digestate exhibits functional changes in bacterial communities impacting nitrogen removal and wastewater treatment. Algal Res., 57. https://doi.org/10.1016/j.algal.2021.102338

Ayre, J. M., Moheimani, N. R., & Borowitzka, M. A. (2017). Growth of microalgae on undiluted anaerobic digestate of piggery effluent with high ammonium concentrations. Algal Res., 24, 218-226. https://doi.org/10.1016/j.algal.2017.03.023

Bankston, E. M., & Higgins, B. T. (2020). Anaerobic microbial communities can influence algal growth and nutrient removal from anaerobic digestate. Bioresour. Technol., 297, 122445. https://doi.org/10.1016/j.biortech.2019.122445

Bankston, E. M., Wang, Q., & Higgins, B. (2020). Algae support populations of heterotrophic, nitrifying, and phosphate-accumulating bacteria in the treatment of poultry litter anaerobic digestate. Chem. Eng. J., 398, 125550. https://doi.org/10.1016/j.cej.2020.125550

Barzee, T. J., Yothers, C., Edalati, A., Rude, K., Chio, A., El Mashad, H. M.,... Zhang, R. (2022). Pilot microalgae cultivation using food waste digestate with minimal resource inputs. Bioresour. Technol. Rep., 19, 101200. https://doi.org/10.1016/j.biteb.2022.101200

Bjornsson, W. J., Nicol, R. W., Dickinson, K. E., & McGinn, P. J. (2013). Anaerobic digestates are useful nutrient sources for microalgae cultivation: Functional coupling of energy and biomass production. J. Appl. Phycol., 25(5), 1523-1528. https://doi.org/10.1007/s10811-012-9968-0

Bunce, J. T., Ndam, E., Ofiteru, I. D., Moore, A., & Graham, D. W. (2018). A review of phosphorus removal technologies and their applicability to small-scale domestic wastewater treatment systems. Front. Environ. Sci., 6. https://doi.org/10.3389/fenvs.2018.00008

Chaiprapat, S., Sasibunyarat, T., Charnnok, B., & Cheirsilp, B. (2017). Intensifying clean energy production through cultivating mixotrophic microalgae from digestates of biogas systems: Effects of light intensity, medium dilution, and cultivating time. Bioenergy Res., 10(1), 103-114. https://doi.org/10.1007/s12155-016-9780-9

De Mendiburu, F. (2014). Agricolae: Statistical procedures for agricultural research. R package version. 1-4. Comprehensive R Archive Network.

Eustance, E., Lai, Y.-J. S., Shesh, T., & Rittmann, B. E. (2020). Improved CO2 utilization efficiency using membrane carbonation in outdoor raceways. Algal Res., 51, 102070. https://doi.org/10.1016/j.algal.2020.102070

Feng, S., Liu, F., Zhu, S., Feng, P., Wang, Z., Yuan, Z.,... Chen, H. (2020). Performance of a microalgal-bacterial consortium system for the treatment of dairy-derived liquid digestate and biomass production. Bioresour. Technol., 306, 123101. https://doi.org/10.1016/j.biortech.2020.123101

Fox, J., & Weisberg, S. (2018). An R companion to applied regression (3rd ed.). Sage Publ.

Fuentes-Grünewald, C., Ignacio Gayo-Peláez, J., Ndovela, V., Wood, E., Vijay Kapoore, R., & Anne Llewellyn, C. (2021). Towards a circular economy: A novel microalgal two-step growth approach to treat excess nutrients from digestate and to produce biomass for animal feed. Bioresour. Technol., 320, 124349. https://doi.org/10.1016/j.biortech.2020.124349

Gao, K. (2021). Approaches and involved principles to control pH/pCO2 stability in algal cultures. J. Appl. Phycol., 33(6), 3497-3505. https://doi.org/10.1007/s10811-021-02585-y

Geller, D. P., Das, K. C., Bagby-Moon, T., Singh, M., Hawkins, G., & Kiepper, B. H. (2018). Biomass productivity of snow algae and model production algae under low temperature and low light conditions. Algal Res., 33, 133-141. https://doi.org/10.1016/j.algal.2018.05.005

Gong, J., & You, F. (2014). Optimal design and synthesis of algal biorefinery processes for biological carbon sequestration and utilization with zero direct greenhouse gas emissions: MINLP model and global optimization algorithm. Ind. Eng. Chem. Res., 53(4), 1563-1579. https://doi.org/10.1021/ie403459m

González-Camejo, J., Aparicio, S., Ruano, M. V., Borrás, L., Barat, R., & Ferrer, J. (2019). Effect of ambient temperature variations on an indigenous microalgae-nitrifying bacteria culture dominated by Chlorella. Bioresour. Technol., 290, 121788. https://doi.org/10.1016/j.biortech.2019.121788

Hachiya, T., Watanabe, C. K., Fujimoto, M., Ishikawa, T., Takahara, K., Kawai-Yamada, M.,... Noguchi, K. (2012). Nitrate addition alleviates ammonium toxicity without lessening ammonium accumulation, organic acid depletion and inorganic cation depletion in Arabidopsis thaliana Shoots. Plant Cell Physiol., 53(3), 577-591. https://doi.org/10.1093/pcp/pcs012

Hohl, H. R., & Hamamoto, S. T. (1969). Ultrastructure of Acrasis rosea, a cellular slime mold, during development. J. Protozool., 16(2), 333-44. https://doi.org/10.1111/j.1550-7408.1969.tb02279.x

Hyman, M., Wang, Q., Wilson, A. E., Adhikari, S., & Higgins, B. T. (2021). Production of Daphnia zooplankton on wastewater-grown algae for sustainable conversion of waste nutrients to fish feed. J. Cleaner Prod., 310, 127501. https://doi.org/10.1016/j.jclepro.2021.127501

Jiang, Y., McAdam, E., Zhang, Y., Heaven, S., Banks, C., & Longhurst, P. (2019). Ammonia inhibition and toxicity in anaerobic digestion: A critical review. J. Water Process Eng., 32, 100899. https://doi.org/10.1016/j.jwpe.2019.100899

Kalvakaalva, R., Smith, M., Ayipio, E., Blanchard, C., Prior, S. A., Runion, G. B.,... Higgins, B. T. (2023). Mass-balance process model of a decoupled aquaponics system. J. ASABE, 66(4), 955-967. https://doi.org/10.13031/ja.15468

Kumar, B. R., Mathimani, T., Sudhakar, M. P., Rajendran, K., Nizami, A.-S., Brindhadevi, K., & Pugazhendhi, A. (2021). A state of the art review on the cultivation of algae for energy and other valuable products: Application, challenges, and opportunities. Renew. Sustain. Energy Rev., 138, 110649. https://doi.org/10.1016/j.rser.2020.110649

León-Vaz, A., León, R., Vigara, J., & Funk, C. (2023). Exploring Nordic microalgae as a potential novel source of antioxidant and bioactive compounds. New Biotechnol., 73, 1-8. https://doi.org/10.1016/j.nbt.2022.12.001

Li, T., Zheng, Y., Yu, L., & Chen, S. (2013). High productivity cultivation of a heat-resistant microalga Chlorella sorokiniana for biofuel production. Bioresour. Technol., 131, 60-67. https://doi.org/10.1016/j.biortech.2012.11.121

Liu, Z., Hou, Y., He, C., Wang, X., Chen, S., Huang, Z., & Chen, F. (2020). Enhancement of linoleic acid content stimulates astaxanthin esterification in Coelastrum sp. Bioresour. Technol., 300, 122649. https://doi.org/10.1016/j.biortech.2019.122649

Luo, Y., Yao, J., Wang, X., Zheng, M., Guo, D., & Chen, Y. (2020). Efficient municipal wastewater treatment by oxidation ditch process at low temperature: Bacterial community structure in activated sludge. Sci. Total Environ., 703, 135031. https://doi.org/10.1016/j.scitotenv.2019.135031

Meng, J., Hu, Z., Wang, Z., Hu, S., Liu, Y., Guo, H.,... Zheng, M. (2022). Determining factors for nitrite accumulation in an acidic nitrifying system: Influent ammonium concentration, operational pH, and ammonia-oxidizing community. Environ. Sci. Technol., 56(16), 11578-11588. https://doi.org/10.1021/acs.est.1c07522

Nguyen, M. L., Lin, C. Y., & Lay, C. H. (2019). Microalgae cultivation using biogas and digestate carbon sources. Biomass Bioenergy, 122, 426-432. https://doi.org/10.1016/j.biombioe.2019.01.050

Oksanen, J., Blanchet, F. G., Kindt, R., Legendre, P. M., Minchin, P. R., O’hara, R. B.,... Stevens, H. (2017). Ordination methods, diversity analysis and other functions for community and vegetation ecologists. Vegan: Community Ecol. Package. Comprehensive R Archive Network

Park, S., & Bae, W. (2009). Modeling kinetics of ammonium oxidation and nitrite oxidation under simultaneous inhibition by free ammonia and free nitrous acid. Process Biochem., 44(6), 631-640. https://doi.org/10.1016/j.procbio.2009.02.002

Pizzera, A., Scaglione, D., Bellucci, M., Marazzi, F., Mezzanotte, V., Parati, K., & Ficara, E. (2019). Digestate treatment with algae-bacteria consortia: A field pilot-scale experimentation in a sub-optimal climate area. Bioresour. Technol., 274, 232-243. https://doi.org/10.1016/j.biortech.2018.11.067

Prandini, J. M., da Silva, M. L., Mezzari, M. P., Pirolli, M., Michelon, W., & Soares, H. M. (2016). Enhancement of nutrient removal from swine wastewater digestate coupled to biogas purification by microalgae Scenedesmus spp. Bioresour. Technol., 202, 67-75. http://doi.org/10.1016/j.biortech.2015.11.082

Princic, A., Mahne, I. I., Megusar, F., Paul, E. A., & Tiedje, J. M. (1998). Effects of pH and oxygen and ammonium concentrations on the community structure of nitrifying bacteria from wastewater. Appl. Environ. Microbiol., 64(10), 3584-90. https://doi.org/10.1128/aem.64.10.3584-3590.1998

Sarwer, A., Hamed, S. M., Osman, A. I., Jamil, F., Al-Muhtaseb, A. H., Alhajeri, N. S., & Rooney, D. W. (2022). Algal biomass valorization for biofuel production and carbon sequestration: A review. Environ. Chem. Lett., 20(5), 2797-2851. https://doi.org/10.1007/s10311-022-01458-1

Smith, V. H., & Crews, T. (2014). Applying ecological principles of crop cultivation in large-scale algal biomass production. Algal Res., 4, 23-34. https://doi.org/10.1016/j.algal.2013.11.005

Subirats, J., Sharpe, H., & Topp, E. (2022). Fate of Clostridia and other spore-forming Firmicute bacteria during feedstock anaerobic digestion and aerobic composting. J. Environ. Manag., 309, 114643. https://doi.org/10.1016/j.jenvman.2022.114643

Tharek, A., Yahya, A., Salleh, M. M., Jamaluddin, H., Yoshizaki, S., Hara, H.,... Mohamad, S. E. (2021). Improvement and screening of astaxanthin producing mutants of newly isolated Coelastrum sp. using ethyl methane sulfonate induced mutagenesis technique. Biotechnol. Rep., 32, e00673. https://doi.org/10.1016/j.btre.2021.e00673

Tillmann, U. (2004). Interactions between planktonic microalgae and protozoan grazers. J. Eukaryot. Microbiol., 51(2), 156-168. https://doi.org/10.1111/j.1550-7408.2004.tb00540.x

van Hannen, E. J., Zwart, G., van Agterveld, M. P., Gons, H. J., Ebert, J., & Laanbroek, H. J. (1999). Changes in bacterial and eukaryotic community structure after mass lysis of filamentous cyanobacteria associated with viruses. Appl. Environ. Microbiol., 65(2), 795-801. https://doi.org/10.1128/AEM.65.2.795-801.1999

Viegas, C., Gouveia, L., & Gonçalves, M. (2021). Aquaculture wastewater treatment through microalgal. Biomass potential applications on animal feed, agriculture, and energy. J. Environ. Manag., 286, 112187. https://doi.org/10.1016/j.jenvman.2021.112187

Vuppaladadiyam, A. K., Prinsen, P., Raheem, A., Luque, R., & Zhao, M. (2018). Microalgae cultivation and metabolites production: A comprehensive review. Biofuels Bioprod. Bioref., 12(2), 304-324. https://doi.org/10.1002/bbb.1864

Wang, J., Wang, X., Yu, Z., Huang, S., Yao, D., Xiao, J.,... Zan, F. (2022). Using algae bacteria consortia to effectively treat coking wastewater: Performance, microbial community, and mechanism. J. Cleaner Prod., 334, 130269. https://doi.org/10.1016/j.jclepro.2021.130269

Wang, Q., Cherones, J., & Higgins, B. (2021a). Acclimation of an algal consortium to sequester nutrients from anaerobic digestate. Bioresour. Technol., 342, 125921. https://doi.org/10.1016/j.biortech.2021.125921

Wang, Q., Childree, E., Box, J., López-Vela, M., Sprague, D., Cherones, J., & Higgins, B. T. (2023). Microalgae can promote nitrification in poultry-processing wastewater in the presence and absence of antimicrobial agents. ACS EST Eng., 3(4). https://doi.org/10.1021/acsestengg.2c00360

Wang, Q., Hyman, M., & Higgins, B. T. (2021b). Factors impacting the effectiveness of biological pretreatment for the alleviation of algal growth inhibition on anaerobic digestate. Algal Res., 53, 102129. https://doi.org/10.1016/j.algal.2020.102129

Wang, Q., Prasad, R., & Higgins, B. T. (2019). Aerobic bacterial pretreatment to overcome algal growth inhibition on high-strength anaerobic digestates. Water Res., 162, 420-426. https://doi.org/10.1016/j.watres.2019.07.011

Zhao, Z., Liu, S., Yang, X., Lei, Z., Shimizu, K., Zhang, Z.,... Adachi, Y. (2019). Stability and performance of algal-bacterial granular sludge in shaking photo-sequencing batch reactors with special focus on phosphorus accumulation. Bioresour. Technol., 280, 497-501. https://doi.org/10.1016/j.biortech.2019.02.071