Co-cultivation of fungal-microalgal strains in biogas slurry and biogas purification under different initial CO2 concentrations

The effects of five different microalgae-fungi on nutrient removal and CO2 removal were investigated under three different CO2 contents (35%, 45% and 55%). The results showed that the highest nutrient and CO2 removal efficiency were found at 55% CO2 by cocultivation of different microalgae and fungi. The effect of different initial CO2 concentration on the removal of CO2 from microalgae was significant, and the order of CO2 removal efficiency was 55% (v/v) >45% (v/v) >35% (v/v). The best nutrient removal and biogas purification could be achieved by co-cultivation of C. vulgaris and G. lucidum with 55% initial CO2 content. The maximum mean COD, TN, TP and CO2 removal efficiency can reach 68.29%, 61.75%, 64.21% and 64.68%, respectively under this condition. All highest COD, TN, TP and CO2 removal efficiency were more than 85%. The analysis of energy consumption economic efficiency revealed that this strategy resulted in the highest economic efficiency. The results of this work can promote simultaneously biological purification of wastewater and biogas using microalgal-fungal symbiosis.


Results and Discussion
The five selected strains growth at different CO 2 concentration. These five fungal-microalgal mixture strains survived all treatments. Table 1 shows the results of cell growth and average daily productivity of the selected five different fungal-microalgal mixture (i.e., P. geesteranus/C. vulgaris, G. lucidum/C. vulgaris, P. ostreatus/C. vulgaris, G. lucidum/S. obliquus and G. lucidum/S. capricornutum) under different CO 2 concentration treatments (35%, 45% and 55%, v/v). From Table 1, it was concluded that the growth rates under 55% CO 2 concentration treatment were higher than 45% and 35% CO 2 concentration treatment. Probably reason is that biomass production of the fungal-microalgal mixture depends on CO 2 consumption as the carbon source under phototropic condition 20 . This statement can also be supported by the results of the mean daily productivity data shown in Table 1. Notably, the growth rate and mean daily productivity under 55% CO 2 concentration were recorded highest by G. lucidum/C. vulgaris mixture (0.352 d −1 , 0.174 gL −1 d −1 ), follow by G. lucidum/S. obliquus, P. geesteranus/C. vulgaris, P. ostreatus/C. vulgaris and G. lucidum/S. capricornutum. Hence, high CO 2 concentration (55%, v/v) was chosen as the most effective treatment and G. lucidum/C. vulgaris strain can be ranked as the optimal fungal-microalgal mixture according to its high biomass production. However, what were the possible reasons that G. lucidum/C. vulgaris co-cultivation mixture was superior to the other four similar mixtures? Firstly, biomass production ability of microalgae strain C. vulgaris was relatively high than some other microalgae based on previous studies [21][22][23][24][25] . For example, Zhao et al. 21,23 reported that the growth rate and mean daily productivity of high-yield strain C. vulgaris can reach 0.363 d −1 and 0.112 g L −1 d −1 with optimal wavelength mixing ratios treatments, and reach 0.372 d −1 and 0.183 g L −1 d −1 with treatment of synthetic high-strength wastewater 21,23 . Secondly, based on the molecular mechanism of filamentous fungal-based bio-flocculation, fungal cell capacity for self-pelletization may be significantly different as it is strain-specific and not all filamentous fungal strains can form pellets during growth 26 . In this study, after co-cultivation, green-colored pellets were found by interaction between fungal strain G. lucidum and microalgal strain C. vulgaris, instead of milky white-colored pellets like other four fungal-microalgal mixture, which indicate that Table 1. Growth rates and mean daily productivity of the five selected strains under different CO 2 concentration treatments. pelletization capacity of G. lucidum/C. vulgaris mixture was relatively strong. It is unavoidable to form biofilm on the wall of reactor, the final treatment efficiencies will be affected by some parameters such as the decreased illumination intensity or shortage of nutrient 27 . However, according to our experiences 28,29 , algal-fungal symbionts achieved relatively high biomass for 10 days and the removal efficiencies of pollutants decreased after 10 days, which will lead to poor economic efficiency if last for a longer period of cultivation time.
For P. geesteranus/C. vulgaris, and P. ostreatus/C. vulgaris mixtures, simple adherence or entrapment mode was found for the interaction between microalgal cells and the fungal pellets 30 . Likewise, Linder 31 reported that fungal cell capacity for self-pelletization was correlated to the accumulation of a family of low molecular weight amphipathic, hydrophobic proteins accumulated on the hyphal surface 31 . These hydrophobic proteins are potentially involved in hyphae adherence to solid substrates 32 . Thirdly, pH was the key factor affecting formation of fungal-algal pellet 15,20 . For G. lucidum/C.vulgaris mixture under 55% CO 2 concentration treatment in the experiment, final pH value was 7.16 after 10 days co-cultivation period, which is slightly high than the other four fungal-microalgal mixtures. Therefore, enhanced solubility of CO 2 in the alkalescent biogas slurry was found, which act as the carbon source in the nutrient solution. It is similar to the previous conclusion that pH serviced as the key factor to induce the pelletization of M. circillenous alone 33 . Likewise, Liu et al. 34 used pH adjustment to induce the formation of fungal cell pelletization, providing a simplified method by which to facilitate the cell harvest of oleaginous cells 34 . Nutrient removal efficiencies at different CO 2 concentration. Based on Table 1, a considerably high average COD remove efficiency (55.72%-68.29%) was achieved with 55% CO 2 concentration treatment, followed by 54.26%-66.29% and 46.62%-60.52% with 45% and 35% CO 2 concentration treatments. This variation trend can further confirmed the conclusion that organic carbon is the basic ingredient of microalgae, which accounts for about half of microalgal biomass and can be utilize for heterotrophic or mixotrophic growth [35][36][37] . In this study, under 55% CO 2 concentration treatment, the five fungal-microalgal strains grown at autotrophic and heterotrophic conditions using CO 2 as the only carbon source. Besides, the corresponding average COD remove efficiency ranked: G. lucidum/C. vulgaris (68.29%) > P. geesteranus/C. vulgaris (63.92%) > P. ostreatus/C. vulgaris (62.45%) > G. lucidum/S. obliquus (59.17%) > G. lucidum/S. capricornutum (55.72%). Furthermore, Fig. 1 depicts the changes in COD removal during 10 days of the experimental period and maximum COD removal efficiency even reached 87.37% by G. lucidum/C. vulgaris, which is slightly high than most previous report conclusions. For instance, Zhao et al. 38 demonstrated that the highest COD removal efficiency can reach 85.35% 38 . Similarly, Yan and Zheng 18,39 carried out series researches and reported that 86% of COD could be removed by Chlorella sp. within 24 h with optimal photoperiods, while 78.9% with optimal mixed wavelength ratio (red:blue = 5:5) 18,39 . These different results are closely related to different influent CO 2 concentrations, photoperiod of the experiment. Besides, all the results imply that the screening of microalgal, fungal and fungal-microalgal strains is effective to reduce the COD in biogas slurry.
According to Table 2, TN in the biogas slurry was removed by the five fungal-microalgal strains significantly under the three CO 2 concentration treatments. But the TN removal efficiencies were a little different during 10 days experimental schedule as displayed in Fig. 2. The highest average TN remove efficiency was obtained under 55% CO 2 concentration treatment for the five fungal-microalgal mixtures containing P. geesteranus/C. vulgaris, G. lucidum/C. vulgaris, P.ostreatus/C. vulgaris, G. lucidum/S. obliquus and G. lucidum/S. capricornutum, reaching 54.07%, 61.75%, 51.32%, 63.93% and 59.83%, respectively. Especially, the mixed culture containing G. lucidum/S. obliquus demonstrated highest N removal efficiency, while P.ostreatus/C. vulgaris showed relatively low TN removal efficiency. Coincidentally, it showed very consistent with previous reported conclusion that the highest TN remove efficiency was obtained for the bacterial-microalgal mixture containing C. vulgaris, S. obliquus, and N.oleoabundans, reaching 61.49%, 63.13%, and 55.26%, respectively 2 . In addition, Zhao et al. 21 also recommended that S.obliquus can contribute to a high nitrogen RE than C. vulgaris or N. oleoabundans and nitrogen RE even reached 76% given appropriate mixed ratio of wavelength (red:blue = 7:3) 21 . Xu et al. 40 investigated an integrated approach that combined freshwater microalgae Scenedesmus obliquus cultivation with piggery anaerobic digestate liquid treatment and revealed that average nitrogen RE was 58.39-74.63% 40 . Assimilation within microalgal biomass may be the key mechanism of nitrogen removal because reproduction of fungal/microalgal need sufficient nitrogen source to produce nucleic acids 41 .
As far as phosphorus was concerned, five fungal-microalgal mixtures survived in the co-cultivation and phosphorus REs were recorded as 56.29%, 64.21%, 53.74%, 61.98% and 52.29%, respectively, with 55% CO 2 concentration treatment, which is higher than under 45% and 35% CO 2 concentration treatment. Obviously, the trend of TP removal efficiency (Fig. 3) is consistent with those of fungal-microalgal growth rate, but not the same to COD and nitrogen RE (Table 1 and Table 2). Phosphorus is a key element in microalgae culture and is an important component of cell membrane phospholipids and adenosine triphosphate 22 . Furthermore, the presence of Ca 2+ and Mg 2+ in the biogas slurry and the alkaline conditions caused by fungal-microalgal growth promoted phosphorus precipitation and the formed deposits were helpful for phosphorus removal from biogas slurry 42,43 . Zhao et al. 23 reported that the TP RE by the three microalgae (i.e., C. vulgaris, S. capricornutum, and S. obliquus) was 97.01%, 95.40%, and 95.87% for high C loading waste water with initial P concentration of 0.4 mg L −1 23 . Similarly, Powell 44 reported that more than 95% of the soluble P in the primary effluent was removed by Chlorella when the initial P concentrations were 4 mg L −1 for the primary effluent 44 . It is worth noting that the phosphorus RE observed in this work with fungal-algal mixture seems slightly lower than above-mentioned previous works. Most significant influence factor for such strange phenomenon was the initial phosphate concentration in the biogas slurry, which had a strong influence on the accumulation of polyphosphate in the microalgae. Thus, the high initial phosphate concentration (20 mg L −1 ) was probably responsible for the uncompleted consumption of TP in this study 23,44 .
SCientifiC RepORts | (2018) 8:7786 | DOI:10.1038/s41598-018-26141-w Above all, the biogas slurry nutrient (i.e., COD, TN, and TP) were reduced efficiently and similarly by the five fungal-microalgal mixtures under the different CO 2 concentration treatments (35%, 45% and 55%, v/v) for 10 days batch culture, and the nutrient remove efficiency (REs) were presented in Table 2 and Figs 1-3 The nutrient removal efficiencies with 55% CO 2 treatments were higher than those of the other treatments and achieved the highest COD, TN, and TP removal efficiencies of 68.29 ± 4.73%, 63.93 ± 5.13%, and 64.21 ± 5.36%, respectively. However, no consistent relationship was found between COD remove efficiency, TN remove efficiency, TP remove efficiency and fungal/microalgal growth rates in this study. In other words, the fungal-microalgal growth rate is not proportional to the nutrients (COD, TN and TP) removal from biogas slurry, which is consistent with the conclusion previous reported before by Yan et al. 39 and Wang et al. 22 . These nutrient removals were mainly achieved via an assimilation process of microalgal/fungal reproduction as they require abundant carbon, nitrogen, and phosphorous sources for heterotrophic or mixotrophic growth 22,36 . But, it was not consistent for COD, TN and TP remove efficiency as detailed mechanism was concerned. Therefore, selection and optimization of fungal-microalgal strains is very important for biogas upgrading, biogas slurry nutrients removal, microalgal metabolism and greenhouse gas reduction. In this study, the optimal fungal-microalgal mixture for biogas slurry nutrients removal was G. lucidum/C. vulgaris under 55% CO 2 concentration treatments.
Biogas upgrading. Average CO 2 removal rates (%) were investigated as a function of operating time to evaluate differences in biogas upgrading with different CO 2 influent concentrations for the five fungal-microalgal strains ( Table 2, Fig. 4). Specifically, G. lucidum/C. vulgaris strain recorded high average CO 2 removal rate of 64.21 ± 5.36%, followed by P. geesteranus/C. vulgaris for 64.21 ± 5.34%, P. ostreatus/C.vulgaris for 58.53 ± 4.87%, G. lucidum/S. obliquus for 55.62 ± 4.63% and G. lucidum/S. capricornutum for 54.84 ± 3.99%. This result agrees with the findings of previous studies by Sun et al. 2 , who reported that CO 2 can be reduced up to 49.95%-62.31% by bacterial-microalgal co-cultivation containing S. obliquus, C. vulgaris, N. oleoabundans and activated sludge 2 . At the end of experimental duration, the highest CO 2 -remove efficiency was recorded as 86.97 ± 5.38% by G. lucidum/C. vulgaris strain (Fig. 4), which was higher than most conclusions that reported before 18,21,23,39 . It can further confirmed that selection and optimization of fungal-microalgal strain can significantly address such issues as CO 2 sequestration, biomass production, nutrient removal of biogas slurry, and simultaneously biogas purification for engineering progress in the future. Moreover, the effect of biogas upgrading in this study agreed with the variation trends of the growth rates and mean daily productivity for the fungal-microalgal strains ( Table 1). Half of such biomass reproduction was derived from CO 2 sequestration 45 . If took algal-fungal biomass production based on 1 Kg CO 2 removal as a measurable indicator, G. lucidum/C. vulgaris had the highest the biomass production, which were 644.33 g/L, 529.76 g/L and 484.24 g/L, respectively. G. lucidum/S. capricornutum had the lowest biomass production, which was consistent with the analysis of the growth characteristics of the algae. In addition, all the algal-fungal biomass production decreased as initial CO 2 concentration increased from 35% to 55%. This finding implied that high CO 2 could inhibit growth of the algal-fungal biomass. This is consist with Sun's research 2 .
Though numerous influence factors, such as mix wavelength ratio, photosynthetic photon flux density, different photoperiod treatments, initial CO 2 influent concentrations, algal strains and different C/N ratios, were expatiated deeply for reduction of CO 2 in the biogas, investigation about effect of pH for biogas upgrading was still incomplete in the previous research 2, 18,[21][22][23]38,39 . In this study, pH was detected and recorded every day during 10 days experimental duration for every fungal-microalgal strain under all initial CO 2 influent concentration ( Table 3). Similar variation trend of pH was found and was proportion to biomass reproduction and biogas slurry nutrients removal. It was induced that elevated pH or slight alkalescent biogas slurry contributed to enhance sequestration of CO 2 by solution of assimilation. In addition, O 2 and H 2 O (v/v) concentrations in the biogas almost unvaried during the experimental period (data not shown). O 2 concentration (v/v) was increased from 0.12% ± 0.02% to 0.52% ± 0.04% (data not shown), whereas H 2 O concentrations (v/v) were in the range of 1.12% ± 0.16% and 3.03% ± 0.22% (data not shown). Since raw biogas always contains saturated steam, the presence of H 2 O in the upgraded biogas does not negatively affect the growth of microalgae. In addition, H 2 O and O 2 can also be applied for microalgal photosynthesis and respiration 46 .
The COD of the biogas slurry can be efficiently removed during the biogas purification, which was in line with that of biogas CO 2 removal and microalgae growth. These results were affirmed in the work of Tongprawhan et al. 47 , who suggested that CO 2 fixation with microalgae was environmentally sustainable in wastewater purification. The COD reduction is attributed to the assimilation process of microalgae, which involves cell growth of microalgae and microalgal-fungal pellets, the CO 2 uptaking of the microalgal and microalgal-fungal pellets was positively related to the microalgal cell growth and COD removal 48 . The microalgae cells assimilation requires abundant carbon from biogas slurry and biogas for producing nucleic acid 49 . Furthermore, Chisti 45 reported that approximately half of the microalgae cell was carbon derived from CO 2 uptaking. Especially, the synthetic materials in the pollutant removal process could be used during the microalgal autotrophic metabolism. They are acted as an enzyme activator or energy (ATP), and as the components of microalgae 50 .Therefore, the COD in biogas slurry can promote the CO 2 removal efficiency of biogas using microalgae. Energy consumption economic efficiency. Table 2 also showed the energy consumption economic efficiency for biogas CO 2 removal efficiency and the biogas slurry nutrient removal efficiency using different microalgae with different CO 2 content. The results show that, for 35% CO 2 , G. lucidum/C. vulgaris has the highest energy efficiency in these five cultures. Although the TN removal of G. lucidum/S. obliquus was 2.3% higher than G. lucidum/C. vulgaris, the difference in TN removal rate was not significant (p > 0.05). Similarly, G. lucidum/C. vulgaris can achieve relatively high energy efficiency with 45% and 55% CO 2 . For the same reason, there was no significant difference between G. lucidum/C. vulgaris and G. lucidum/S. obliquus for energy efficiency of TN and TP removal with 45% CO 2 (p > 0.05), as well as energy efficiency of TN removal between G. lucidum/S. obliquus and G. lucidum/C. vulgaris with 55% CO 2 (p > 0.05). These results are consistent with the analysis of the microalgal growth and nutrient removal mentioned above. As a result, G. lucidum/C. vulgaris can achieve high energy efficiency with 55% CO 2 . The reason can be conclude that, CO 2 can provide an important and sufficient carbon source for photosynthesis of microalgae and promote its growth with 55% CO 2 . The Ganoderma lucidum can provide a carrier for microalgal growth that promotes their growths. The symbiont resulted in removing nutrient in the sewage and CO 2 in biogas efficiently 2,28,51 . According to the Eq.(3), the energy efficiency depend on the removal rate of nutrients or CO 2 . According to the results of Table 2, the removal rate of G. lucidum/C. vulgaris is superior to other cultures and lead to high energy efficiency.

Conclusions
Five different microalgae-fungi had significant effects on nutrient and CO 2 removal. The removal of pollutants and biogas purification increased as the increasing of CO 2 content in biogas. G.lucidum/C. vulgaris was selected as the better biological treatment with the initial 55% CO 2 content because of its high pollutant purification efficiency. The mean COD, TN, TP and CO 2 removal efficiency were 68.29%, 61.75%, 64.21% and 64.68%, respectively. The analysis of the energy consumption economic efficiency demonstrated that cocultivation of microalgae and fungi experienced the highest economic efficiency. Similarly, three fungal strains obtained from China General Microbiological Culture Collection Center were selected in this study for the further research as they have high growth rate and high pelletization ability (namely, P. geesteranus, G. lucidum and P. ostreatus). To form pellets, spore solutions were cultivated at 25 ± 0.1 °C for 7 d on 500 mL synthetic growth medium (glucose, 10 g L −1 ; NH 4 NO 3 , 2.0 g L −1 ; K 2 HPO 4 , 1.0 g L −1 ; NaH 2 PO 4 ·H 2 O, 0.4 g L −1 ; MgSO 4 ·7H 2 O, 0.5 g L −1 ; and yeast extract, 2.0 g L −1 ; pH 6.5). The obtained biomass was washed and homogenized with 100 mL of sterile distilled water in a laboratory blender. Subsequently, these obtained strains were used for the co-cultivation with microalgae.
As far as fungal-microalgal co-cultivation was concerned, microalgal suspensions (100 mL) of C. vulgaris, S. obliquus, and S. capricornutum were obtained after preparation and then each suspension was mixed with 5 mL of P. geesteranus, G. lucidum or P. ostreatus pellet suspension. The co-culture conditions for fungal-microalgal mixtures were as follows: constant light 200 μmol m −2 s −1 , 25 ± 0.5 °C, artificial intermittent shaking at 160 rpm approximately for 168 h. All of the biogas upgrading and wastewater purification experiments were biologically conducted in triplicated and the daily biomass concentrations were measured during operational periods in 10 days. Biogas slurry and biogas. The CO 2 content in synthetic biogas were 35.26 ± 2.19% (vol.%), 45.28 ± 1.92% (vol.%), 55.13 ± 3.11% (vol.%). The biogas slurry was obtained from an anaerobic digester in Hongmao Hacienda, Kunshan City, Jiangsu Province, PR China. The raw biogas slurry was pretreated by passing through a glass microfiber filter (GF/C; Whatman, USA) and ultraviolet sterilizer (SKW-UVU01; SKYUV Water Treatment Co. Ltd, China) for 2 minutes to prevent potential interference from sediment and some microorganisms 18 . The characteristics of the raw biogas slurry before and after pretreatment were listed in Table 4, which revealed that the characteristics of biogas slurry almost unchanged before and after pretreatments.
Experimental procedure. According to our previous studies, mono-microalgal strain C. vulgaris and mono-fungal stain G. lucidum already showed great ability on biogas upgrading and simultaneously biogas slurry nutrients removal by itself in the bioreactor 21,23,28 . Hence, three above-mentioned fungal strains were co-cultivated with C. vulgaris, and three selected microalgal strains were co-cultivated with G. lucidum in this study for the further study. In view of one mixture was double counted, there are five fungal-microalgal mixtures were co-cultivated for next step in this experiment, such as P. geesteranus/C. vulgaris, G. lucidum/C. vulgaris, P. ostreatus/C. vulgaris, G. lucidum/S. obliquus and G. lucidum/S. capricornutum.
Detailed procedures were as follows based on research design above: 100 mL of microalgal suspensions of C. vulgaris, S. obliquus and S. capricornutum (about 118 mg L −1 of all the dry weight) were cultured, then each suspension was mixed with 5 mL of P. geesteranus, G. lucidum and P. ostreatus pellet suspension (about 83 mg L −1 of dry weight). The initial density of the microalgae co-cultivated with fungal cells was maintained at about 123.52 ± 3.46 mg·L −1 for the five fungal-microalgal pellets. The following conditions were used: the light intensity was 200 μmol m −2 s −1 , the experimental period was 10 d, the temperature was 25 ± 0.5 °C and the 1ight:dark cycles was 12 h:12 h. The growth rates, mean daily productivity, nutrient removal and CO 2 content with different fungal-microalgal co-cultivation types were evaluated daily and the optimal CO 2 concentration was selected by analyzing the economic efficiencies of the biogas CO 2 and the biogas slurry nutrient removal.
Sampling and analyses. The biogas slurry in photobioreactors was sampled daily for determination of COD, total nitrogen (TN) and total phosphate (TP). The biogas was sampled for component analysis (CH 4 , CO 2 , O 2 and H 2 O, v/v) using a circulating gas analyzer (GA94; ONUEE Co. Ltd, China). Dry weights of microalgae were measured through exsiccation after being filtered with a glass microfiber filter (GF/C, Whatman, USA). The filtrates were used for nutrient determination according to the standard methods 52 .  Table 3. Variations in pH under various CO 2 concentration treatments for the five selected strains. Biogas CO 2 and total biogas slurry nutrient removal efficiency (RE, %) was calculated based on the following equation: where C i is the biogas CO 2 content or total nutrient concentration (g L −1 ) in cultures at time t i and C 0 is the initial biogas CO 2 content or total nutrient concentration (g L −1 ) at time t 0 (day). Specific growth rates (μ) were derived from the growth phase using the following equations: where D i stand for the biomass concentration (g L −1 ) at time t i (d) and D 0 is the biomass concentration (g L −1 ) at time t 0 (d).
The CO 2 or biogas slurry nutrient removal economic efficiency was evaluated based on the following equation: where E is the biogas CO 2 or biogas slurry nutrient removal economic efficiency (USD −1 ), R is the removal efficiency of pollutant (%) in Eq. (1), k is the electric power charge per unit of power consumption (USD kW −1 h −1 ), which is around 0.645 RMB kW −1 h −1 in local, i.e. around 0.097 USD kW −1 h −1 ; T is the light application time (h), and P is the LED electric quantity (W).

Statistical analyses.
Statistic analysis was carried out using Statistic Package for Social Science (SPSS, V19.0). One-way analysis of variance (ANOVA) was used to determine whether the impact of various factors on the test indicators is significant. Duncan's multiple range tests was used to analyze the significant difference between groups. The value p = 0.05 was regarded as the threshold for statistical significance.