Introduction

Managing waste streams generated during technological processes is widely recognized as a significant challenge for the industrial sector. Recently, much attention is being paid to the use of anaerobic digestion (AD) for organic waste valorization, biogas production, and digestate generation [1]. Over the last 20 years, the production of biogas obtained in the AD process has increased significantly. According to the European Biogas Association (EBA), in 2020, as many as 18,943 biogas plants and 725 biomethane plants were recorded across Europe, producing around 167 TWh (15.8 bcm) of biogas and 26 TWh (2.5 bcm) of biomethane. The EBA projects production to reach a minimum of 380 TWh by 2030, showing an increasing trend in the near future and indicating an increased interest in the anaerobic digestion of organic waste [2]. However, apart from biogas, the main product of the anaerobic treatment process is digestate, which is rich in burdensome nutrients and organic contaminants [1]. In accordance with OJ L 170, 25.6.2019 [3] post-digestion products obtained, inter alia from fresh plants or their parts can be used as fertilizing products in UE. However, such digestate must meet certain criteria, and the anaerobic digestion process has to be carried out in specific conditions [3]. On the other hand, due to a high ammonia content, an uncontrolled management of digestate may lead to water eutrophication disturbing the ecological stability of ecosystems. Moreover, digestate may also enhance biological contamination due to the presence of numerous pathogens and phytotoxins. Additionally, digestate storage may increase greenhouse gas emissions (N2O, CH4) as anaerobic digestion rarely decomposes all volatile solids. On the other hand, a high concentration of nutrients and macroelements (phosphorus, nitrogen, potassium, sodium, magnesium, sulfur, calcium) and the presence of various trace elements (e.g. iron, copper, nickel, cobalt) make the digestate an attractive material for reuse. Digestate can be separated into a solid (constituting 10–20% by mass) and a liquid (80–90% by mass) fraction, allowing to treat them independently [1, 4]. The most common separation devices for digestates used in full-scale plants are decanter centrifuges, screening drum presses, and screw presses. Compared to liquid digestate, solid digestate has attracted more attention in research and development opportunities due to efficient post-treatment processes and valorization purposes. In recent years, new routes have been proposed for solid fraction valorization, including biochar production, biofuel generation (e.g., for applying in domestic furnaces), and treatments for methane recovery. The liquid fraction is more difficult to manage and also creates greater environmental problems than the solid digestate. This results from a low C/N ratio and a high content of refractory substances, including phosphorus, nitrogen, and often heavy metals. The increased concentration of nitrogen and other compounds can restrict the application of liquid digestate in the soil as described by the European Nitrates Directive [1, 5]. Furthermore, the high water content makes the liquid digestate difficult to store and transport as well as limits its direct use as a soil conditioner or fertilizer due to high costs [6]. So far, several chemical and mechanical techniques have been developed to treat the liquid digestate, including ammonia stripping, reverse osmosis, or struvite precipitation [7]. However, these technologies are characterized by low efficiency, high costs, and a generation of difficult-to-treat by-products. To overcome these problems, new biological methods are being introduced and studied. One of the alternatives is the application of microalgae capable of growing in a low C/N environment typical for liquid digestates. These organisms are known to utilize ammonium nitrogen, nitrates, and phosphates, as well as some organic compounds [8]. Both pure cultures and complex communities have been studied in the literature. They have been successfully applied to treat anaerobic digestates obtained from various waste processing, including food waste[9, 10], food waste and animal manure [11], agricultural waste [12], or piggery waste [4, 13]. However, in contrast to mono- and axenic cultures, the use of microalgae and microalgae-bacteria consortia for digestate treatment is believed to be a more appropriate approach since microbial consortia are considered more stable due to their more diverse metabolism and different light absorption capabilities, which can contribute to the increased nutrient removal efficiency as well as greater biomass productivity. In mixed systems, bacteria produce CO2, which is then consumed by algae and converted to O2 and, in turn, absorbed by ammonium oxidizing and heterotrophic bacteria. In addition, dissolved organic carbon released by microalgae can be used by heterotrophic bacteria. Moreover, nitrifying bacteria oxidize ammonia that could be toxic to algae and produce nitrates better available by the latter microbes [14, 15]. Liquid digestate containing high nutrient contents is therefore a potentially attractive material for the cultivation of microalgae or algae-bacteria communities. With biomass separation, treated liquid digestate could be applied as process water, animal drinking water, or used for irrigation in areas of drought [15]. Additionally, biomass created during the treatment is a promising substrate for a sustainable production of various compounds and value-added products. The bioenergetic potential of algae is currently the main subject of diverse studies for the algal-derived fuels (bioethanol, biodiesel, biomethane, or biohydrogen). Moreover, algal biomass is a rich source of bioactive compounds including lipids, fatty acids, pigments, proteins, amino acids, and polysaccharides. Recent studies have also shown the antibacterial, antiviral, and antioxidant potential of properly processed biomass. This approach is consistent with the overall idea of a zero-waste and circular economy [16].

It should be noted that most of the experiments described in the literature have been performed using preliminarily treated digestate in order to avoid biological contamination of algae consortia and primarily reduce ammonia concentrations and turbidity. High turbidity, elevated ammonia contents, and unfavorable nitrogen-to-phosphorous ratio (N/P) can hinder the development of microalgae despite the high tolerance of many strains to toxic components [17, 18]. High pH of some digestates combined with progressing algal development (which increases pH during respiration processes) favors the formation of the un-ionized form of ammonia, which is more toxic and difficult to uptake than ionized ammonium nitrogen. Therefore, many studies point to the need for pH adjustment in order to improve ammonium intake and boost digestate remediation [19]. Although preliminary treatment of digestate (substrate dilution [17], sterilization [18], pH adjustment [19]) can improve microalgae growth, at the same time, it generates additional costs and could be impractical on an industrial scale [18].

The objective of this research was to assess the efficiency of the non-sterilized and non-filtered liquid digestate treatment conducted with algal, environmental isolates using natural sunlight for illumination. For the treatment process, Tetradesmus obliquus and Microglena sp. were used, with Microglena applied previously only in a few, singular studies [20, 21]. In comparison to analogous research, liquid digestate, as a substrate less optimal for algal growth, was not subjected to sterilization, filtration, or pH correction. A bioreactor, in which the treatment process was conducted, was illuminated only with natural sunlight despite the experiments performed in a transitional climate zone, where cloudy days prevail. Abovementioned conditions (application of non-commercial, environmental isolates, lack of sterilization, filtration and pH correction, and no additional illumination) were tested for achieving satisfactory treatment efficiency with the potential to reduce operating costs.

Materials and Methods

Liquid Digestate and Microorganisms

The digestate used in the study was obtained from the Warmia Fruit and Vegetable Processing Company Ltd. (Kwidzyn, Poland), treating vegetable wastes (corn and green peas). The substrate originated from the bioreactor working under an organic loading rate equal to 1.5 kgVS/m3/d. The digestate was subjected to centrifugation (10 min, 4800 G) using Rotina 420 (Hettich, Switzerland) to obtain the liquid fraction further used for the research (Table 1). The liquid digestate was not subjected to other pre-treatments such as sterilization or filtration. Apart from initial, gradual introduction of digestate to the bioreactor until the set HRT was achieved, the substrate was not diluted.

Table 1 Physicochemical characteristic of the liquid digestate fraction
Full size table

A nonaxenic consortium containing Tetradesmus obliquus and Microglena sp. green algae (each with an initial cell density of 1.0 × 104 cells/mL) was applied for the experiment. These algae were isolated and identified in the previous study showing the ability to grow in the liquid digestate [20]. Microalgae were initially cultivated using a modified BBM (Bold’s Basal Medium) with a threefold nitrogen concentration (3N-BBM) prepared in accordance with Andersen [22].

Experiment Design

A 350 mL volume photobioreactor working in semi-continuous mode was used in the study (Fig. 1). The experiment was conducted from the end of May to the beginning of September in laboratory conditions. The bioreactor was illuminated with natural sunlight of an average intensity of 6.0 kLux. The natural cycle of day and night was therefore kept. The average temperature of 30 ± 2 °C was registered throughout the experimental run. The bioreactor was supplied with the constant airflow maintaining the microalgal cultures in suspension. Air was provided with a capacity of 0.7 L/L/min using a membrane blower, model AP-80 H (Thomas, Yamasuga, Japan). A consortium of algae was initially cultivated in a 3N-BBM medium for 4 weeks, after which the digestate started being introduced to the bioreactor for another 112 days. Hydraulic retention time (HRT) equal to solid retention time (SRT) was set at 30 days. Specifically, twice a week, 39 mL of the effluent was withdrawn from the bioreactor, and the same volume of fresh liquid digestate was introduced. Feeding and discharge of the bioreactor were performed manually using a syringe. Total and soluble chemical oxygen demand (tCOD/ sCOD), microalgal biomass parameters defined by chlorophyll a concentration, optical density, the number of algal cells, as well as light intensity were measured once a week. Other indicators such as pH, nitrates, nitrites, ammonium nitrogen, orthophosphates, turbidity, and total suspended solids (TSS) were determined twice a week. Microbiological parameters, turbidity, TSS, and tCOD were measured directly in the reactor effluent, whereas the remaining indicators were determined after filtration. Treatment efficiency (the removal rate of nutrients and COD) was calculated for the steady state period, which was established between the 56th and 112th day of the run.

Fig. 1
figure 1

Laboratory installation used for semi-continuous experiment (1 bioreactor, 2 air outlet, 3 sample collection port, 4 liquid digestate inlet, 5 source of light, 6 air inlet, 7 rotameter, 8 membrane air pump)

Full size image

Analytical Methods

Physicochemical Analysis

A DR6000 spectrophotometer and the following HACH-Lange tests: PhosVer 3 (no. 8048), NitraVer 5 (no. 8039), NitriVer 3 (no. 8507), and modified Nessler (no. 8038) were used to determine orthophosphates, nitrates, nitrites, and ammonium nitrogen, respectively. All nitrogen forms measured during the experiment were expressed as mgN/L allowing to determine the nitrogen removal rate. Volatile fatty acids and COD were determined with HACH-Lange tests LCK 365 and LCK 214, respectively. Turbidity, total suspended solids, and color were measured according to EN ISO 7027, the photometric method (8006 method), and the Platinum-Cobalt standard method (8025 method), respectively. TSS was then used to calculate the overall algae-bacteria biomass productivity in accordance with Min et al. [23]. Total solids, volatile solids, and pH were analyzed with the standard methods [24]. To determine the concentrations of selected metals, the following HACH-Lange methods were applied: FerroVer no. 8008 for iron, Zincon no. 8009 for zinc, AluVer no. 8012 for aluminum, and CuVer 1 no. 8506 for copper. The concentrations of sulfates, sulfides, and chlorides were determined with the respective HACH-Lange tests: SulfaVer 4 method no. 8051, a methylene blue method no. 8131, thiocyanate method no. 8113. Light intensity was measured directly on external walls of the bioreactor using the LXP-10B illuminance meter (Sonel, Poland). Surveys were taken three times a day—in the morning, at noon, and in the evening. The illumination rate inside the bioreactor (internal illumination) was not measured.

Microalgal Biomass Analysis

Microalgal biomass was determined by measuring the level of chlorophyll a (chl-a), the optical density, and the algal cell density.

The concentration of chl-a is one of the basic indicators of algal biomass growth. As all taxonomical algal groups produce this photosynthetic pigment, chl-a concentration can reflect changes and/or disturbances in algal biomass. The chl-a concentration was determined according to the methodology described previously in Sobolewska et al. [20] using 90% methanol during extraction. The final chl-a concentration was calculated with formula 1.

$$Chl-a\left[mg/L\right]=\left(16.5\times {A}_{665}\right)-\left(8.3\times {A}_{650}\right)$$
(1)

where A665—absorbance measured at 665 nm; A650—absorbance measured at 650 nm.

The biomass growth was also evaluated by optical density measured at 680 nm (OD680), a wavelength corresponding to the second chlorophyll absorption peak and allowing to indirectly estimate biomass levels in suspensions [21].

Cell density of each algal taxa was determined by applying cell counting on a microscopic slide using a droplet method. Together with quantitative analysis, taxonomic identification was performed based on morphological characteristics. The number of cells per 1 mL of a sample was calculated using formula 2 enabling the further evaluation of algal taxa percentage share.

$$Nx[cells/mL]=\left(xi\times Ap\right)/\left(Si\times V\right)$$
(2)

where Nx is the algal cell density per 1 mL of sample; xi is the cell number of individual taxa; Ap is the total microscopic slide surface area = 400 mm2; Si is the surface area used to count taxa (mm2), and V is the drop volume (cm2).

Determination of Bacterial Communities Using Metagenomic Analysis

At the end of the experiment, V3-V4 16S rRNA metagenomic analysis was performed on the biomass collected from the photobioreactor. Analysis was executed in accordance with the procedure described in Sobolewska et al. [21]. Acquired data was deposited in the NCBI database. Using the Sequence Read Archive tool, a bioproject and biosample with respective access numbers of PRJNA922144 and SAMN32643824 were prepared. To determine the phylogenetic bacterial biodiversity, 16S rRNA fragments were matched against their corresponding taxonomical levels from phylum to species. The presence of bacteria, confirmed with metagenomic analysis (“Bacterial Community in Biorector” section) resulted from the lack of digestate sterilization and use on nonaxenic algal cultures.

Statistical Analysis

During the experimental run, physicochemical indicators and microbiological parameters were measured in triplicates. The average values, standard deviation (SD), error bars, and correlation coefficients (where 0.0–0.2 very weak correlation, 0.2–0.4 weak correlation, 0.4–0.7 moderate correlation, 0.7–0.9 strong correlation, 0.9–1.0 very strong correlation) were calculated in Microsoft Excel 365 version 2211.

Results and Discussion

Treatment Efficiency

The photobioreactor was exclusively operated under sunlight conditions with no additional light sources. The variations of physical–chemical indicators in the course of the run are plotted in Fig. 2. The post-digestion effluent contains vast amounts of nitrogen, phosphorous, and organic carbon (Table 1), which can be utilized by microalgae for their growth and other metabolic processes [1]. Considering the above, the treatment process was mainly evaluated based on the removal of orthophosphates, nitrogen forms, and COD with simultaneous control of biomass productivity as well as microalgal biomass parameters.

Fig. 2
figure 2

Changes in physicochemical indicators during semi-continuous experimental period (a pH; b phosphates; c nitrogen; d chemical oxygen demand; e total suspended solids/ turbidity); the figure does not include 4 initial weeks of microalgae multiplication

Full size image

During the experimental run, the concentration of nitrates increased until the 31st day of the process. This can be related to the overall biomass growth as reflected by the increase of optical density, turbidity, and TSS. Due to no changes observed in pH and no decrease in ammonium nitrogen, the participation of nitrification bacteria (confirmed in small share with Nitrosomonas sp. and Nitrospira sp. by metagenomic analysis) in a cultivation period is unlikely. After this period, the concentration of N-NO3 significantly dropped to stabilize at a relatively low level since the 38th day of the run. In contrast, the ammonium nitrogen content in the treated digestate showed much lower fluctuations throughout the experiment. Importantly, the concentration of nitrites was close to zero, which could indicate oxygen-sufficient conditions in the reactor (data not included in Fig. 2c). There is an indication that the presence of Microglena sp. could slightly affect changes in ammonium nitrogen concentration. After the 49th day of the experiment, the trend for the slow and gradual rise in ammonium nitrogen concentration was observed, corresponding to the decrease in Microglena sp. cell abundance. This trend was offset for the period of 84th and 105th day, correlating to the small increase in cell density. Moreover, after the complete disappearance of Microglena sp. registered at the end of the experimental run, the ammonium nitrogen increased relatively sharply again. Nitrogen plays a significant role in the growth of living organisms. Eukaryotic algae can directly absorb inorganic nitrogen only in the form of ammonium nitrogen, nitrates, and nitrites. Due to lower energy requirements, ammonium nitrogen is usually assimilated faster than nitrates by algae. On the other hand, an excessive ammonium concentration might have a repressive growth effect. Besides, nitrogen in the form of nitrates is thermodynamically more stable in aqueous, oxidized environments and therefore might be dominant in the substrate [25]. However, it should be noted that the assimilation capabilities as well as a tolerated nutrient concentration are usually species- and strain-dependent [26]. Nitrogen transformations did not considerably affect pH. On one hand, ammonium nitrogen increases the pH of the environment, whereas the appearance of nitrates through nitrification has a reverse effect. However, as depicted in Fig. 2a, the pH value showed greater variations only in the first part of the experiment, and since the 38th day of the run, it stabilized at 9.5. The high pH level could be attributed to the significant activity of microalgae, which consume CO2 and carbonates thus increasing buffering capacity of the environment. On the other hand, increased pH results in conversion of ammonium nitrogen into free ammonia, which, in turn, could inhibit the growth of microalgae [27], but the effect of ammonia reduction occurring with volatilization during aeration of culture media was not measured. It can be therefore assumed that the pH increase could be the reason for the microalgae density decrease observed after 42 days of the process (Fig. 3). Additionally, during the experiment, high intensity of sunlight (the average of 6000 Lux) and high the temperature (30 ± 2 °C) could inhibit nitrification process with photooxidation of bacterial cytochrome as described in Lavrentyev and Gardner [28]. The calculated total nitrogen concentration (as the sum of nitrogen from ammonium nitrogen, nitrates, and nitrites) in the reactor effluent in the steady-state period was equal to 95 mgN/L. This gave the average nitrogen removal rate of 86% (compared to raw liquid digestate and observed for this period. Hence, a complete nitrogen removal was not achieved, suggesting the presence of remaining organic compounds not assimilable by microalgae and their native, accompanying bacteria. Generally, the recorded removal rate was greater than the values reported in the literature. In a raceway pond treatment system operated with food waste digestate, over 80% reduction in total nitrogen was reached [10]. In the cited study, unfiltered digestate was used as a substrate, and nitrogen was removed not only through microbial transformation but also by ammonia oxidation. In another study, a consortium of Scenedesmus spp. and Chlorella spp. was applied for the treatment of agricultural digestate [12]. The authors documented only 20–29% of nitrogen removal rate with a significant impact of nitrification in the overall biological process. In turn, Tan et al. [29] reported total nitrogen reduction at approximately 84%, which was achieved for initially filtered starch wastewater treated in an outdoor photobioreactor using Chlorella pyrenoidosa.

Fig. 3
figure 3

Changes in cell density and percentage share of algal taxa

Full size image

Phosphorus is another key macronutrient for living organisms, and algae are believed to possess a great capacity to store and use phosphorus. Inorganic phosphates (PO4-P) are the dominant phosphorus forms in the digestate (up to 90% contribution), and they play a significant role in the overall microalgal growth [1, 29]. As illustrated in Fig. 2b, initially, the phosphate concentration in the treated digestate showed a downward trend until day 38 of the experiment, after which it stabilized at 49 mg/L. It can be assumed that phosphates could be removed from the digestate not only as a result of microalgal activity but also through precipitation, which is intensified at a high pH of 8–10 [30]. This gave the average phosphate removal rate of 86%, which was comparable to the findings of Xu et al. [13] where Scenedesmus obliquus (now also known as Tetradesmus obliquus) was applied for the treatment of sterilized piggery anaerobic effluent. A similar efficiency was also documented in a study of Tan et al. [29], where filtered starch wastewater was treated by Chlorella pyrenoidosa.

Algae are also known to be able to utilize both inorganic and organic carbon [25]. The soluble COD indicator reflects the presence of organic compounds in the liquid digestate, whereas total COD also includes the concentration of organic compounds in biomass. Changes in total and soluble COD are plotted in Fig. 2d. The value of total COD increased twice in the initial experimental stage, which could be linked to the biomass growth in this period, as this indicator moderately correlates to TSS (Table 2). In contrast, soluble COD remained at a relatively constant level throughout the experiment and did not correlate with tCOD. The calculated removal rate of soluble COD in a steady-state period reached 91%. For comparison, Abu Hajar et al. [11] reached a 78–82% COD removal rate using Scenedesmus dimorphus to treat unsterilized animal manure and food waste. Likewise, the algal treatment of piggery anaerobic liquid digestate allowed to reduce 62–75% sCOD [13]. In another study, the sCOD removal efficiency oscillated between 40 and 75%, which was dependent on the digestate dilution [9]. In our study, the reduction rate of total COD (tCOD) was lower (70%), probably due to the biomass growth. In algae-bacteria systems, both microbial groups can be involved in the removal of organic compounds via anabolic and catabolic reactions [31]. However, a complete removal of COD is very unlikely, because hardly degradable compounds and the products of cell lysis remain in the effluent [29, 32]. As mentioned earlier, the growth of biomass was indicated by both tCOD and TSS increase as well as turbidity, and a strong correlation between these indicators was observed. Total suspended solids also correlated with optical density; however, the strongest correlation of over 0.99 was noted between TSS and turbidity (Table 2). In a steady-state period, the mean levels of TSS and turbidity were 3793 mg/L for TSS and 2462 FAU, respectively.

Table 2 Correlation coefficients between physicochemical indicators
Full size table

The TSS-based overall biomass productivity was equal to 22 mg/L/d and corresponded to the lower limit values obtained in analogous digestate treatment studies. It should be noted that biomass productivity can be influenced by inter alia a type of substrate, its characteristics and pretreatment methods, environmental conditions such as lightning and temperature [10], and strain-dependent microbial capabilities. The strains used in this study were environmental isolates, acquired from substrates other than liquid digestate, and therefore the natural ability of algal cells to develop in such substrates may have been lower. The applied process conditions and specific substrate (liquid digestate) could have the greatest impact on the results obtained, which differ from those reported in the literature. For instance, Xu et al. [13] achieved 6–14 times higher biomass productivity, whereas the treatment efficiencies were 62–75% for COD, 58–75% for TN, and 70–89% for TP. Similar values (78–82% of COD removal, 65–72% of TN removal, and 63–100% of TP reduction) were reported by Abu Hajar et al. [11], with biomass productivity higher when the substrate dilution rate was increased. On the other hand, the research of Barzee et al. [10] shows that with the equally low biomass productivity (17.7 ± 1.8 mg VS/L/d), high treatment efficiencies can still be achieved (total nitrogen removal over 80%). It should also be noted that, despite the relatively low biomass productivity, the other parameters characterizing cell growth, i.e., the chlorophyll a concentration and optical density suggested the development of the algal culture allowing to achieve satisfactory treatment results. The analogous biomass values (chl-a = 19.04 mg/L) were also reported by Ansari et al. [33], who applied Scenedesmus sp. and achieved high removal rates of COD (95%), phosphate (over 80%), and nitrates (over 99%).

Changes in the Structure of Algal Community During Treatment

From the beginning of the experiment until the 98th day, the only detected algae were Tetradesmus obliquus and Microglena sp. Within the first month, the algal cell number increased after which a decrease in cell density and its further stabilization was observed (Fig. 3). The initial rapid growth of algae could be induced by phosphates that remained in the environment from the 3N-BBM medium used before the digestate was introduced to the bioreactor. Moreover, in the first weeks, the growth of algae was unaffected by the biomass density, which was initially low. Conversely, determined cell density began to decrease after the 42nd day of the process, despite high, increasing chl-a and OD680 values suggesting a potential mutual shading effect (see more in the “Mutual Shading Effect and Influence of Light Intensity” section). The determination of cell density indicates the presence of a significant number of Microglena sp. cells, mainly in the first half of the experiment, after which the share of this taxon decreased significantly till complete disappearance occurring between the 105th and 112th day. Starting from the 56th day, a drop in average light intensity was noted suggesting the influence on Microglena sp. regression. This effect was enhanced by the occurrence of mutual shading evidenced inter alia by progressive turbidity and TSS increase. Although this phenomenon most likely affected both microalgal strains, obtained results indicate that Microglena cells were less resistant. It should also be noted that the adaptive capabilities of Microglena sp. to substrates such as digestate have not yet been described in scientific literature. The growth of algae could be affected by pH and vice versa. It is commonly known that pH strongly impacts algal cell division and metabolic processes, and the maximal growth of these organisms can be observed in neutral environments [22]. As noted in the “Treatment Efficiency” section, alkaline pH favors result in a release of free ammonia, which is much more inhibitory to microalgae cells than ammonium ions. The growth of algae in the first month of the run corresponded to the increase in both optical density and chl-a concentration. Interestingly, after this period, the algae cell density significantly dropped as discussed above, but the concentration of chl-a remained at a constant level of 23.0 mg/L until the end of the experimental run (Fig. 4). The correlation factor between these indicators suggested weak correlation (R > 0.36) in contrast to the literature findings [34]. Additionally, the chl-a concentration can be affected by multiple factors such as phosphorous and nitrogen concentrations or temperature [35]. With the slow increase of phosphates (registered from 77 to 112 days of the process), a small increase in nitrates concentration, and highly probable daily temperature rise (occurring naturally for the summer period, especially in August), maintaining high chlorophyll a concentration despite cell density reduction was possible. This observation may also be explained by the presence of other phototrophic microorganisms in the nonaxenic community of the photobioreactor as discussed below. Another interesting observation is the appearance of the Chlorella sp. algae determined metagenomically as Chlorella sorokiniana at the end of the experimental run. Their growth could be induced by the decrease in sunlight intensity, occurring naturally at the beginning of September (corresponding to the end of the experiment). It is highly probable that singular C. sorokiniana cells were present in the substrate or reactor since the beginning of the experimental run but were not detected analytically, and the high light intensity could have inhibited their photosystem II activity resulting in overall photoinhibition. Gao et al. [36] documented that C. sorokiniana growth is being inhibited when light intensity exceeds 150 µmol/m2/s, while Li et al. [37] described that Chlorella kessleri biomass growth stops at 200 µmol/m2/s. Hence, the decrease in light intensity to the levels tolerated by Chlorella cells could lead to their multiplication and taxonomical domination at the end of the run. Moreover, between the 105th and 112th day of the experiment, a rising trend in ammonium nitrogen concentration (correlating with a pH decrease) was observed, which might be the cause of a rapid development of Chlorella sp. [38].

Fig. 4
figure 4

Changes in optical density and chlorophyll a concentration

Full size image

Bacterial Community in Bioreactor

The bacteria presence, determined at the end of the experiment using metagenomic analysis, resulted from the lack of digestate sterilization and the use of nonaxenic algal cultures. It was found that the most dominant bacteria at the phylum level were Firmicutes (26.31%) and Patescibacteria (17.38%) as well as Actinobacteriota (14.86%). This may suggest that bacteria were mainly responsible for organic contaminants oxidation, phosphorus removal, and nitrogen transformation in the digestate [39]. Furthermore, a significant abundance (around 8%) of Chloroflexi, Cyanobacteria, and Proteobacteria was also detected. Bacteria from the Proteobacteria phylum could be responsible for the degradation of organic matter and detoxification of inorganic pollutants [40]. Moreover, bacteria from the Chloroflexi phylum could contribute to phosphorous removal [39]. At the order level, Bacillales, Dojkabacteria, Microtrichales, and Chloroplast record dominated, whereas a high relative abundance of Rhizobiales, Oligoflexales, Caldilineales, and Deinococcales was also noted (Fig. S1 -Supplementary material). Metagenomic analysis revealed that Tetradesmus obliquus and Chlorella sorokiniana observed at the end of the experiment (see the “Changes in the structure of algal community during treatment” section) were assigned to the Chloroplast record (Cyanobacteria phylum). In turn, members of Bacillales are known to be important denitrifying and hydrocarbon-degrading bacteria [41], while Microtrichales represented mostly by bacteria Iamia sp. (relative abundance higher than 8%) are heterotrophs responsible for the mineralization of organic carbon [42]. Likewise, the activity of Trupera sp. from the order Deinococcales could be linked to the removal of COD from the digestate [43]. In contrast, the family Rhizobiaceae includes bacteria responsible for nitrogen fixation [44]. Additionally, an important role could be exhibited by Chloroflexaceae autotrophic denitrifiers [45] and chemoorganotrophic Blastocatellaceae showing an ability to decompose complex carbon compounds [46] (Fig. 5).

Fig. 5
figure 5

Bacterial community and bacteria cell abundance at the level of family

Full size image

Mutual Shading Effect and Influence of Light Intensity

The average light intensity over the experimental period is shown in Fig. 6. It should be noted that illumination was measured on the external walls of the bioreactor and therefore mutual shading effect was not directly evaluated. Throughout the first 63 days of the experiment, the mean light intensity was similar. The semi-continuous experimental period started in spring (May) in which the number of sunny days is relatively high [47]. A significant decrease of illumination was registered in the 70th, 91st, and 105th days (with mean values of 0.103–0.106 kLux) suggesting cloudy weather conditions. A decrease in illumination was noted since the beginning of August as a result of lower insolation, more cloudy weather, and more abundant in rainy days [47]. However, no correlation between lower illumination and biological parameters such as cell density, optical density, and chlorophyll a was found proving that the decrease in sunlight was most likely the result of single cloudy days, which did not significantly affect the deterioration of the quantitative parameters of the algae (see more in the “Changes in the structure of algal community during treatment” section). Light parameters inside the bioreactor and the ability of light to penetrate and reach microalgal cells also depend on the mutual shading effect. Its occurrence may be indicated by a decrease in cell density after the 42nd day of the experimental run (“Changes in the structure of algal community during treatment” section) and confirmed by visual observations. The phenomenon of biofilm overgrowing the internal walls and the biomass gathered in the bottom of the bioreactor (despite mixing) became visible at the beginning of August (about the 70th day of the process) and lasted till the end of the experiment (105th day). A significant impact of mutual shading on the development and stability of microalgal cultures as well as treatment efficiency has already been described in the literature [48,49,50]. Guieysse et al. [48] reported a negative correlation between cell density, photosynthetic activity, and increased O2 consumption. The shading effect might be indirectly evidenced by a significant increase in the TSS concentrations from the 49th day of the experiment with the highest value (4270 mg/L) reported on the 105th day. The relationship between TSS and the mutual shading effect was described by Foladori et al. [49] and Uggetti et al. [50]. The increased TSS concentrations (and thus the shading effect) could have been the reason for the upward trend in phosphates concentration starting from the 77th day, as shown in Fig. 2b. However, it should be noted that liquid digestate had dark color, which might have affected penetration and spread of light intensifying mutual shading effect. The decrease in sunlight intensity enhanced by the mutual shading effect as well as TSS and turbidity rise observed from the 49th day forward were probably the main factors causing the decrease and disappearance of Microglena sp. Overall, this taxa showed less resistance to changes registered for light conditions than Tetradesmus obliquus.

Fig. 6
figure 6

Changes in light intensity occurring during the experimental run (orange dots represent mean values; orange bars show maximal and minimal values; red dotted line represents mean light intensity calculated for all measurements); for the 70th, 91st, and 105th day, registered values were lower than 0.5 kLux

Full size image

Conclusions

The study showed that natural sunlight can be successfully used for efficient liquid digestate treatment with algae-bacteria systems even in temperate climate zones. As high as 91% sCOD and 86% nutrients were removed from non-sterile and unfiltered liquid digestate proving that initial substrate pre-treatment is not required to achieve high treatment efficiency. Microglena sp. not used previously for digestate treatment showed low resistance to light intensity reduction and effects of mutual shading. The results also indicated that Microglena sp. can potentially promote ammonium nitrogen removal.