Listen

S2− is one of the common pollutants in heavily polluted rivers. A pump-and-treat ex-situ process with enriched consortium (PEPEC) was used to remove S2− in this study. The kinetic model of S2− removal was developed, and the inflow ratio of the PEPEC was analyzed according to the results of the kinetic models. The results showed that the S2− removal ratio could reach 97.5% ± 0.5%, when the inflow ratio was controlled at 2% for the PEPEC operation. Meanwhile, the removal efficiency and operation performance were assessed for both the simulating ex-situ and in-situ bench-scale tests. Compared with the in-situ processes, the PEPEC showed a stable operation performance during 120 h of bio-treatment, and the concentrations of S2−, COD, NH3-N and TP in the effluent reached approximately 0.5, 20, 0.5 and 0.5 mg/L, respectively. The time consumption (8 h for one batch) and consortium dosage (3 g for the whole operation) in the PEPEC were significantly less than those in the in-situ processes. The PEPEC presented some potential advantages for the bio-treatment of a heavily polluted river.

Listen

  • This article developed a pump-and-treat ex-situ process with enriched consortium (PEPEC) to solve black-odor water problem.

  • This article built a kinetic model and found the optimal in flow ratio of PEPEC.

  • This article applied bench-scale tests to compare PEPEC and in-situ methods and found the advantage of PEPEC.

enriched consortium, dissolved S2−, inflow ratio, initial biomass concentration, pump-and-treat ex-situ process
Listen

Due to the development of rapid urbanization, many rivers have been heavily polluted (Wang et al. 2021). Research has shown that S2− is one of the common pollutants (Song et al. 2017). S2− can form other sulfur-containing pollutants such as hydrogen sulfide and metal sulfides. Hydrogen sulfide is a biologically important molecule with complex physiological functions, which can cause the well-known inhibitory effect on cellular respiration and potential inhalation hazard. Some insoluble metal sulfides like FeS could be suspended in a heavily polluted river, which induces water to be black (Wang et al. 2014). Many in-situ techniques, such as coagulation aeration and biological methods, have been utilized to remove S2− from the water body (Yin et al. 2019). However, these techniques have some disadvantages such as being time-consuming, the high cost, the large space occupation or high microbe dosage (Cao et al. 2020).

In the past decades, ex-situ microbial techniques have rarely been used to treat a heavily polluted river considering the relatively fast stream flow and high flow rate of a river. In fact, in-situ remediation for river often means significant projects and a large amount of investment. However, there is partial retention or even stagnant water in some small sections of a heavily polluted river, which provides the application possibility of the ex-situ process. A pump-and-treat approach with inoculated microbes in a mini-bioreactor is regarded as a promising and developing ex-situ process. This has been successfully used to remediate groundwater and soil (Boal et al. 2015). Biomass concentration is an important factor that affects the treatment efficiency of a bioreactor. Enrichment consortium with a high biomass concentration can shorten the operation time and improve the treatment capacity of a bioreactor, and this technique has been used in the fermentation process (Liu & Zhu 2018). Also, enriched consortia has positive effects on substrate degradation, and higher methane yields were observed in the reactors compared with control reactors set up with standard inoculum (Ozbayram et al. 2017, 2018). An enhanced stability of continuous biomethanation processes has been achieved because enriched consortia could alleviate ammonia toxicity in a continuous stirred tank reactor operating under ammonia-induced, inhibited-steady-state (Fotidis et al. 2017).

Therefore, a pump-and-treat ex-situ process with enriched consortium (PEPEC), namely a simulating ex-situ process, was developed to remove dissolved S2− from a heavily polluted river. Effects of initial biomass and S2− concentration on the removal efficiency of dissolved S2− were firstly investigated, and S2− removal kinetics were then analyzed. Based on the obtained results of kinetic models, the inflow ratio parameter of the PEPEC was investigated. Finally, a comparison of the bio-treatment performance between the PEPEC and the simulating in-situ process was carried out at the lab-scale.

Listen

Water sample

Listen

The original water samples were individually collected at eight sampling sites from Dongsha River that was a heavily polluted urban river located in Beijing, China. The representative water sample was created by the mixture of the eight samples in order to obtain the average values to eliminate the difference among the samples. The characteristics of the water sample were analyzed, and the concentrations of S2−, COD, NH3-N and TP were 20.7 ± 1.2 mg/L, 104.5 ± 8.0 mg/L, 5.3 ± 0.4 mg/L and 3.0 ± 0.1 mg/L, respectively.

Enriched consortium

Listen

Three single strains including Citrobacter sp., Ochrobactrum sp. and Stenotrophomonas sp. were all isolated from Dongsha River and were characterized by 16S rDNA sequence. Their sequences were all uploaded to the National Center for Biotechnology Information, and the GenBank accession numbers were MH181794, MH181795 and MH181796, respectively. Microbial consortium was composed of these three strains, and their mass proportions were set at 1:1:1. The initial cell density of each strain was 8.96 × 105 cells/mL (Colony-Forming Units, cfu). In our previous study, microbial consortium has been proved to be capable of efficient S2−-oxidizing, and more than 60% of dissolved S2− in some urban black-odor water bodies located in Beijing of China were removed (Xu et al. 2019). It was found that S0, S2O32−, SO32− and SO42− were all formed in the whole process of S2− oxidization (Zhang et al. 2022). Enriched consortium was obtained using a liquid medium that consisted of 20 g/L of glucose, 10 g/L of peptone and 10 g/L of yeast extract. The pH of liquid medium was adjusted to 7.0. The incubation was carried out at 25 °C and 120 rpm for 48 h. The enriched consortium harvested with centrifugation was used immediately for the following experiment.

Batch experiment

Listen

Effects of initial biomass and S2− concentration on the bio-treatment efficiency of dissolved S2− were investigated using a bioreactor with 3 L of working volume. According to the reported results of microbial tolerance for S2− (Zhuang et al. 2017), three S2− levels (30, 50 and 100 mg/L) were chosen as the initial concentrations in batch experiment, respectively. Na2S·9H2O was used to increase the initial S2− concentration of each sample. The samples were inoculated at different biomass concentrations, and then were treated at 25 °C in the batch bioreactors. In addition, a group without enriched consortium served as the control. The control was used to obtain the background variation, which was subtracted from that of any other batch experiment.

Continuous experiment

Listen
The effect of inflow ratio on the removal efficiency of dissolved S2− was evaluated in a continuous bioreactor (as shown in Figure 1(a)). The water sample from Dongsha River was directly used as the influent of the continuous experiment. Each bioreactor was inoculated by transferring 1.0 g/L biomass concentration of enriched consortium, and then was operated at 25 °C. After the rapid start-up, each bioreactor was continuously run at various inflow ratios. Each experiment was carried out in triplicate.
Figure 1
The schematic diagram of the PEPEC (a) and in-situ process (b).
View largeDownload slide

The schematic diagram of the PEPEC (a) and in-situ process (b).

Figure 1
The schematic diagram of the PEPEC (a) and in-situ process (b).
View largeDownload slide

The schematic diagram of the PEPEC (a) and in-situ process (b).

Close modal

Simulating the ex-situ process and the in-situ process

Listen

The simulating ex-situ process, namely the PEPEC, was run using the same bioreactor as the continuous experiment. The constructed bioreactor was a cube with an effective volume of 3 L. The influent was pumped from the bottom of the bioreactor into the bioreactor with a flow rate of 0.06 L/min by a peristaltic pump. The effluent was overflowed from the top of the bioreactor. A 2 μm filter membrane was attached to the outlet to retain microorganisms in the bioreactor. For the PEPEC, the inoculum containing 1.0 g/L of biomass concentration was transferred into each bioreactor, and the inflow ratio was set at 2%. The simulating in-situ experiment was carried out using a bench-scale cylindrical bioreactor (as shown in Figure 1(b)). The working volume of cylindrical bioreactor was 30 L. In the simulating in-situ process, two levels of biomass concentrations (0.01 and 1.0 g/L) were chosen to compare the removal efficiency of S2− by enriched consortium. The water sample from Dongsha River was both used as the influent in the ex-situ process and the in-situ process. All experiments were conducted at 25 °C for 120 h. Each experiment was performed in triplicate.

Analytical methods

Listen

The concentrations of S2−, COD, NH3-N and TP were analyzed according to Chinese Water and Wastewater Monitoring and Analysis Methods (MEP 2002b). Both COD and NH3-N were determined using a HACH multi-parameter detector (DR2800, USA); while TP and S2− were measured using a UV–Vis spectrophotometer (UV-1800, Japan). The biomass concentration was determined according to the previously described method (Wang et al. 2014). 30 mL of mixed suspension was sampled during the reaction, and then was centrifuged at 8,000 rpm for 10 min. The supernatant was discarded. The obtained sediment was washed five times with distilled water. After centrifugation, washed sediment was weighted. The biomass concentration was calculated as the wet weight of washed sediment divided by the volume of mixed suspension. At the end of the reaction process, the three single strains were all recovered and sterilized before disposal in this study.

Listen

Removal efficiency of dissolved S2− as a result of initial biomass concentration

Listen
Biomass concentration is one of the most important factors that can greatly influence pollutant removal efficiency. Figure 2 shows the effect of initial biomass concentration on the S2− removal efficiency at different initial S2− concentrations. In each case, the S2− concentration decreased rapidly, and then leveled off. The final S2− concentrations for all groups were similar and the values were all below 0.5 mg/L. However, the time necessary for dissolved S2− removal was obviously different. A positive impact of initial biomass concentration on the spending time was observed in all. When increasing the biomass concentration from 0.01 to 10 g/L, the time spent for S2− removal decreased from 100 to 23 h, and it was significantly shortened by 77% (Figure 2(a)). In contrast to initial biomass concentration, initial S2− concentration had a negative influence on the spending time for S2− removal, as shown in Figure 2. At 30 mg/L of initial S2− concentration, more than 80% of dissolved S2− was removed in the first 16 h of bio-treatment, whereas for 50 and 100 mg/L, the required time was prolonged to 23 and 30 h, respectively.
Figure 2
Effects of the initial biomass concentration on the S2− removal efficiency at different initial S2− concentrations: (a) 30 g/L of initial S2− concentration; (b) 50 g/L of initial S2− concentration; (c) 100 g/L of initial S2− concentration (Xu et al. 2019).
View largeDownload slide

Effects of the initial biomass concentration on the S2− removal efficiency at different initial S2− concentrations: (a) 30 g/L of initial S2− concentration; (b) 50 g/L of initial S2− concentration; (c) 100 g/L of initial S2− concentration (Xu et al. 2019).

Figure 2
Effects of the initial biomass concentration on the S2− removal efficiency at different initial S2− concentrations: (a) 30 g/L of initial S2− concentration; (b) 50 g/L of initial S2− concentration; (c) 100 g/L of initial S2− concentration (Xu et al. 2019).
View largeDownload slide

Effects of the initial biomass concentration on the S2− removal efficiency at different initial S2− concentrations: (a) 30 g/L of initial S2− concentration; (b) 50 g/L of initial S2− concentration; (c) 100 g/L of initial S2− concentration (Xu et al. 2019).

Close modal
Figure 3 shows the variation of average removal rate of dissolved S2− with biomass concentrations at different initial S2− concentrations. The average removal rate of S2− dramatically increased when biomass concentration was raised to 10 g/L, and then remained relatively stable at the higher biomass concentration. 1.2 mg/(L h) of high removal rate was achieved at 10 g/L of biomass concentration.
Figure 3
Variation of the average removal rate of dissolved S2− at various initial S2− concentrations.
View largeDownload slide

Variation of the average removal rate of dissolved S2− at various initial S2− concentrations.

Figure 3
Variation of the average removal rate of dissolved S2− at various initial S2− concentrations.
View largeDownload slide

Variation of the average removal rate of dissolved S2− at various initial S2− concentrations.

Close modal

The appropriate amount of inoculation helped the microorganisms to adapt the new environment and might sustain high metabolic activity, which accelerated the removal of dissolved S2− in the water sample. A further increase of biomass concentration did not contribute to shortening the time due to the competition of S2− among microorganisms, which indicated a ‘saturation’ state of biomass concentration (Pradhan & Rai 2000). This might explain why the removal rate of S2− did not increase further when biomass concentration was increased from 10 to 50 g/L. In addition, it was found that the S2− removal rates were close at various initial S2− concentrations, suggesting that the initial S2− concentration had little influence on the average removal rate in this study.

This finding suggests that other common dissolved pollutants including COD, NH3-N and TP could also be removed by the inoculated microorganisms. Table 1 shows the effect of initial biomass concentration on the removal ratios of COD, NH3-N and TP after 23 h of bio-treatment under the condition of 30 mg/L initial S2− concentration. The removal ratio of COD decreased by 48.9% in the case that biomass concentration increased from 1 to 10 g/L. Compared with other conditions, 50 g/L of initial biomass concentration could not support efficient removal of NH3-N and TP, and the removal ratios were only 32.1 and 50.5%, respectively.

Table 1

Effect of initial biomass concentration on the removal ratios of S2−, COD, NH3-N and TP at 30 mg/L of initial S2− concentration

Initial biomass concentration (g/L)Removal ratio after 23 h of bio-treatment (%)
CODNH3-NTP
0.01 79.5 ± 1.3 93.9 ± 0.4 80.3 ± 0.4 
0.1 80.9 ± 1.0 93.9 ± 0.7 85.2 ± 0.7 
1 81.5 ± 2.3 94.8 ± 0.5 85.1 ± 0.2 
10 41.6 ± 1.2 93.9 ± 0.6 84.0 ± 0.4 
20 29.5 ± 1.7 94.0 ± 0.4 85.6 ± 0.1 
50 24.4 ± 1.9 32.1 ± 2.3 50.5 ± 1.0 
Initial biomass concentration (g/L)Removal ratio after 23 h of bio-treatment (%)
CODNH3-NTP
0.01 79.5 ± 1.3 93.9 ± 0.4 80.3 ± 0.4 
0.1 80.9 ± 1.0 93.9 ± 0.7 85.2 ± 0.7 
1 81.5 ± 2.3 94.8 ± 0.5 85.1 ± 0.2 
10 41.6 ± 1.2 93.9 ± 0.6 84.0 ± 0.4 
20 29.5 ± 1.7 94.0 ± 0.4 85.6 ± 0.1 
50 24.4 ± 1.9 32.1 ± 2.3 50.5 ± 1.0 
View Large

The results shown in Table 1 suggested the optimal range of initial biomass concentration was 0.01–1 g/L. However, the time required for S2− removal was dramatically shortened with the increase of initial biomass concentration from 0.01 to 10 g/L (Figure 2). Considering both the removal ratio and the required time, the optimal initial biomass concentration was 1 g/L. Under this condition, the removal ratios of S2−, COD, NH3-N and TP were obviously enhanced, and they reached 98.5, 81.5, 94.8 and 85.1%, respectively, after 23 h of bio-treatment with the enriched consortium.

Removal kinetics of dissolved S2−

Listen
The S2− removal kinetic model was studied based on the Eckenfelder model (Adams et al. 1975). An ideal condition that the substrate was fully mixed at any time and would be removed completely was assumed in this study. The S2− removal could be described as in Equation (1):
(1)
where t (h) represents the removal time of dissolved S2−, Si (mg/L) is S2− concentration in the influent, Se (mg/L) is S2− concentration in the effluent, X (g/L) is initial biomass concentration and Km (L/(g·h)) refers to specific substrate removal rate coefficient. The kinetics model of S2− removal could also be expressed by the following first-order kinetic Equation (2):
(2)

According to Equation (2), Km depended on X, indicating that the S2− removal kinetic model is special at various X values. On the basis of the obtained experimental data, the special kinetic model and each corresponding Km were calculated, and the results were summarized in Table 2.

Table 2

Summaries of the special kinetic model and the corresponding each Km at various biomass concentrations (Xu et al. 2019)

X (g/L)Km (L/(g·h))R2Kinetic modelsEq.
0.01 28.261 0.9955  (3) 
0.1 4.4567 0.9978  (4) 
1 0.7868 0.9990  (5) 
10 0.1135 0.9971  (6) 
20 0.0667 0.9959  (7) 
50 0.0255 0.9985  (8) 
0 < X ≤ 10 0.7345 0.9996  (9) 
10 ≤ X ≤ 50 1.017 0.9928  (10) 
X (g/L)Km (L/(g·h))R2Kinetic modelsEq.
0.01 28.261 0.9955  (3) 
0.1 4.4567 0.9978  (4) 
1 0.7868 0.9990  (5) 
10 0.1135 0.9971  (6) 
20 0.0667 0.9959  (7) 
50 0.0255 0.9985  (8) 
0 < X ≤ 10 0.7345 0.9996  (9) 
10 ≤ X ≤ 50 1.017 0.9928  (10) 
View Large

It was found that the experimental values at each initial biomass concentration correlated well with the theoretic values, which indicated that this first-order kinetic model could describe the removal of dissolved S2− with the enriched consortium. Km in a bio-treatment system is mainly affected by biomass concentration, substrate concentration and inhibition tendency (Mathur et al. 2008). The Km values decreased with an increase of biomass concentration, as shown in Table 2. The maximum of Km was 28.261 L/(g·h); this value was obtained at 0.01 g/L of the lowest biomass concentration. When biomass concentration increased to 10 g/L, the value of Km significantly decreased to 0.1135 L/(g·h); while with a further increase of biomass concentration to 50 g/L, Km only decreased to 0.0255 (L/(g·h)). Km can indicate the affinity of microorganisms or enzymes for substrates (Mathur et al. 2008). Consequently, the higher the Km value, the lower the affinity is; meanwhile, the substrate is consumed more and more slowly with the decrease of affinity (Shen et al. 2014). Therefore, there was a negative relationship between Km and the average removal rate of dissolved S2−. A low biomass concentration disfavored the rapid adaption of enriched microorganisms to the new environment, and the affinity for substrate was weak, resulting in the low consumption rate of S2−. When the initial biomass concentration gradually increased to 10 g/L, the activity of inoculated microorganisms and the affinity were both enhanced significantly. The dissolved S2− as the substrate became the key limiting factor for the growth of microorganisms, resulting in the intensified competition for S2− among inoculated microorganisms. Therefore, the average removal rate of dissolved S2− obviously increased due to the large consumption of S2− during a short period. With a further increase of biomass concentration, although Km decreased, the provided S2− was not enough for maintaining the growth and metabolism of inoculated consortium, and substrate consumption was shifted to endogenous respiration. This might explain the reason why the average removal rate of S2− did not increase any more when the initial biomass concentration was further raised up to 50 g/L (Figure 3).

Operation stability of the PEPEC as a result of the inflow ratio

Listen
Inflow ratio is an important process parameter, which may significantly influence effluent quality and treatment capacity of a process (Pradhan & Rai 2000). In order to combine with Equation (1), the inflow ratio was calculated from the effective volume of the reactor divided by the influent volume per min in this study:
(11)
where Q (L) is the influent volume per min, V (L) is the effective volume of the reactor and i (%) refers to the inflow ratio. Therefore, Equation (1) could be replaced by the following equation:
(12)

A low inflow ratio can avoid the great fluctuation of water quality due to the effective treatment of pollutant at a long hydraulic retention time (HRT) (Ren et al. 2016). However, a lower value does not necessarily indicate a better result. According to the previous reported results (Yu et al. 2014), an appropriate inflow ratio should meet two requirements: (1) the time spent on treating the pollutant from the original concentration to the target concentration should not exceed the HRT and (2) the inflow ratio should be high enough in order to achieve the maximal treatment capacity.

In addition, HRT was defined as the quotient of the effective volume of the reactor and the influent volume per hour. As such, the relationship between inflow ratio and HRT could be shown as the following equation:
(13)
Therefore, the inflow ratio could be written as follows:
(14)
According to the above results in this study, 1 g/L of initial biomass concentration was optimal compared with other conditions. Thus, the Km and X was considered the best option in Equation (14), and this is shown in Equation (15):
(15)
The theoretic inflow ratio was derived from Equation (15), and its maximal value was 2.5%. The experimental value of inflow ratio was then confirmed within the range of 1.0–3.0%. In order to evaluate the effect of inflow ratio on the operation stability of the PEPEC, the simulating experiment was carried out to investigate the variation of S2− removal efficiency at various inflow ratios, as shown in Figure 4.
Figure 4
Effects of inflow ratio on the operation performance of the PEPEC. (a) Variation of S2− concentration in the effluent. (b) Variation of S2− removal ratio and the ratio of daily treatment volume and the bioreactor volume.
View largeDownload slide

Effects of inflow ratio on the operation performance of the PEPEC. (a) Variation of S2− concentration in the effluent. (b) Variation of S2− removal ratio and the ratio of daily treatment volume and the bioreactor volume.

Figure 4
Effects of inflow ratio on the operation performance of the PEPEC. (a) Variation of S2− concentration in the effluent. (b) Variation of S2− removal ratio and the ratio of daily treatment volume and the bioreactor volume.
View largeDownload slide

Effects of inflow ratio on the operation performance of the PEPEC. (a) Variation of S2− concentration in the effluent. (b) Variation of S2− removal ratio and the ratio of daily treatment volume and the bioreactor volume.

Close modal

Figure 4(a) shows that the concentrations of dissolved S2− in the effluent were different at various inflow ratios during the operation process. When the inflow ratio was below 2.0%, the effluent S2− concentration ranged from 0.3 to 0.4 mg/L. At 2.0% of the inflow ratio, S2− concentration was slightly raised to approximately 0.5 mg/L. However, when the inflow ratio was over 2.0%, the effluent S2− concentrations dramatically increased. This increase was particularly obvious at 3.0% of the inflow ratio, where it exceeded 1.0 mg/L and showed an upward trend. It was concluded that the effluent S2− concentration increased with the increase of the inflow ratio. This was because the required time for S2− removal at 1.0–2.0% of inflow ratio was sufficient so that the S2− could be removed at a low level. At 3.0%, the spending time (0.72 h) surpassed its corresponding HRT (0.56 h), resulting in the accumulation of untreated S2− in the bioreactor and increase of the S2− concentration in the effluent.

Meanwhile, the removal ratio of S2− exhibited a negative relationship to the inflow ratio (Figure 4(b)). The removal ratio of S2− reached more than 97% at 1.0–2.0% of inflow ratio, while it declined at 2.5 and 3.0%. This suggested that a better removal efficiency of S2− would be attributed to the lower inflow ratio. In addition, the higher the inflow ratio selected, the higher the processing capacity obtained. The ratio of daily treatment volume and the bioreactor volume was used to express the processing capacity in this study. As shown in Figure 4(b), the ratio of daily treatment volume and the bioreactor volume was improved with an increase of the inflow ratio. For example, for the 3 L bioreactor, the daily treatment amount could reach 86.4 L/d at 2.0% of inflow ratio, while only 43.2 L/d was obtained at 1.0% of inflow ratio. Although S2− concentration in the effluent was below 1.0 mg/L which reached Grade V of the Surface Water Quality Norm of China (MEP 2002a) at 2.5% of inflow ratio, a higher quality of effluent and processing capacity could be achieved at 2.0% of inflow ratio. Therefore, in order to obtain a better treatment performance, the inflow ratio would be controlled at 2.0% for the PEPEC.

Comparisons of the PEPEC and the in-situ process

Listen
The removal efficiency and processing characteristics of the PEPEC and the in-situ process for treating the original water sample were evaluated, and the results are shown in Figure 5.
Figure 5
Operation performances of three simulating processes. (a) The PEPEC with 1 g/L of initial biomass concentration; (b) the in-situ process with 1 g/L of initial biomass concentration; (c) the in-situ process with 0.01 g/L of initial biomass concentration.
View largeDownload slide

Operation performances of three simulating processes. (a) The PEPEC with 1 g/L of initial biomass concentration; (b) the in-situ process with 1 g/L of initial biomass concentration; (c) the in-situ process with 0.01 g/L of initial biomass concentration.

Figure 5
Operation performances of three simulating processes. (a) The PEPEC with 1 g/L of initial biomass concentration; (b) the in-situ process with 1 g/L of initial biomass concentration; (c) the in-situ process with 0.01 g/L of initial biomass concentration.
View largeDownload slide

Operation performances of three simulating processes. (a) The PEPEC with 1 g/L of initial biomass concentration; (b) the in-situ process with 1 g/L of initial biomass concentration; (c) the in-situ process with 0.01 g/L of initial biomass concentration.

Close modal

The PEPEC was the first process and it was operated at 1 g/L of initial biomass concentration; while the in-situ process was run at 1 g/L (process II) and 0.01 g/L (process III), respectively. No matter which process was used, efficient removal of main dissolved pollutant was achieved. However, the treatment time for these three processes was significantly different. For PEPEC, the concentrations of S2−, COD, NH3-N and TP in the effluent could reach 0.5, 20, 0.5 and 0.5 mg/L after the rapid start-up, respectively; and then they remained relatively stable. As for process II and process III, the spending time (t) in treating dissolved pollutants to obtain the effluent concentration similar to that of PEPEC was approximately 60 and 120 h, respectively. Therefore, 15 batches could be finished during 120 h of operation using PEPEC if the treatment water volume was 30 L for each bath; but for process II and process III, batch number significantly declined to only 2 and 1, respectively. Compared with the in-situ processes, PEPEC showed stable operation performance and good treatment capacity.

Table 3 displays the comparisons of these three processes in treatment efficiency and processing performance. In the case that the target values of effluent concentrations and removal ratios of S2−, COD, NH3-N and TP were similar, each of the processes had different microbial consortium dosages and treatment time when the total water volume was 30 L.

Table 3

Comparisons of treatment efficiency and processing performance among three different processes

ItemsThe PEPECIn-situ process IIIn-situ process III
Water volume (L) 30 30 30 
Enriched consortium dosage (g) 3 30 0.3 
Initial biomass concentration (g/L) 1 1 0.01 
Flow rate (L/min) 0.06 – – 
Treatment time per 30 L of water (h) 8 60 120 
Daily treatment capacity (L/d) 90 12 6 
Effluent concentration of S2− (mg/L) 0.4–0.5 0.4 0.5 
Removal ratio of S2− (%) 97.5–98.0 97.8 97.1 
Effluent concentration of COD (mg/L) 18–23 21 28 
Removal ratio of COD (%) 77.2–82.2 79.8 72.0 
Effluent concentration of NH3-N (mg/L) 0.4–0.6 0.4 0.7 
Removal ratio of NH3-N (%) 88.1–92.1 91.6 84.8 
Effluent concentration of TP (mg/L) 0.4–0.5 0.4 0.4 
Removal ratio of TP (%) 82.7–86.2 87.9 86.1 
ItemsThe PEPECIn-situ process IIIn-situ process III
Water volume (L) 30 30 30 
Enriched consortium dosage (g) 3 30 0.3 
Initial biomass concentration (g/L) 1 1 0.01 
Flow rate (L/min) 0.06 – – 
Treatment time per 30 L of water (h) 8 60 120 
Daily treatment capacity (L/d) 90 12 6 
Effluent concentration of S2− (mg/L) 0.4–0.5 0.4 0.5 
Removal ratio of S2− (%) 97.5–98.0 97.8 97.1 
Effluent concentration of COD (mg/L) 18–23 21 28 
Removal ratio of COD (%) 77.2–82.2 79.8 72.0 
Effluent concentration of NH3-N (mg/L) 0.4–0.6 0.4 0.7 
Removal ratio of NH3-N (%) 88.1–92.1 91.6 84.8 
Effluent concentration of TP (mg/L) 0.4–0.5 0.4 0.4 
Removal ratio of TP (%) 82.7–86.2 87.9 86.1 

Note: The PEPEC was performed at 1 g/L of biomass concentration and 2% of inflow ratio; the in-situ process II and the in-situ process III were performed at 1 and 0.01 g/L of biomass concentration, respectively.

View Large

The dosage of microbial consortium for PEPEC required during the reaction was 10 times that of process III, but the treatment time was only 6.67%. Although the same initial biomass concentration was inoculated, the dosage of microbial consortium and treatment time for PEPEC were only 10 and 13.3% of that of the process II, respectively. Moreover, the ex-situ process and the in-situ process exhibited an obvious difference in the daily treatment capacity. This capacity of PEPEC was 7.5 times and 15 times higher than those of process II and process III, respectively.

In addition, enriched consortium needed to be inoculated only once in PEPEC as a result of the microbe recovery by the bioreactor. As for process II and process III, the dosage would be raised with the increase of operation batches due to the microbial un-recovery for the in-situ process. This might show a potential ‘guerrilla’ capability with less cost of the ex-situ process for treating heavily polluted river water like Dongsha River. The water volume of each retention area along eight sampling sites of Dongsha River was in the range of 50–200 m3. According to the operation performance of PEPEC, theoretically, 10 mini-bioreactors (1 m × 1 m × 1 m) distributed along the river could meet the requirement for the bio-treatment within 1 week. However, there would be still a long way to go from laboratory to scale-up application of the PEPEC. Effects of other factors such as excess sludge, temperature, river sediment, etc., would be investigated in the foregoing research work.

Listen

The initial biomass concentration had an obvious impact on the S2− removal efficiency. The initial S2− concentration could only have an impact on treatment time but had little influence on the average removal rate of S2−. According to the kinetic models of S2− removal, the inflow ratio was optimized at 2%. The removal ratio of dissolved S2− could reach 97.5%±0.5% when the inflow ratio was controlled at 2%. Compared with the in-situ processes, the PEPEC showed a stable operation performance during 120 h of bio-treatment, and the concentrations of S2−, COD, NH3-N and TP in the effluent reached approximately 0.5, 20, 0.5 and 0.5 mg/L, respectively. The PEPEC showed less consortium dosage and time cost, and it also achieved higher daily treatment capacity than the in-situ processes. This work demonstrated that the PEPEC presented some potential advantages for the bio-treatment of a heavily polluted river.

Listen

This work was financially supported by the Central Level, Scientific Research Institutes for Basic R&D Special Fund Business (2019YSKY008) and the Natural Science Foundation of Beijing, China (Grant No. 8182058).

Listen

All relevant data are included in the paper or its Supplementary Information.

Listen

The authors declare there is no conflict.

Adams
E.
,
Eckenfelder
W.
 &
Hovious
C.
1975
A kinetic model for completely-mixed sludge treating variable strength industrial wastewater
.
Water Res.
37
(
9
),
37
–
42
.
Google Scholar
 
Boal
B. A. K.
,
Rhodes
C.
 &
Garcia
S.
2015
Pump-and-treat groundwater remediation using chlorine/ultraviolet advanced oxidation processes
.
Groundwater Monit. Rem.
35
(
2
),
93
–
100
.
https://doi.org/10.1111/gwmr.12095
.
Google Scholar
Crossref
Search ADS
 
Cao
J. X.
,
Sun
Q.
,
Zhao
D. H.
,
Xu
M. Y.
,
Shen
Q. S.
,
Wang
D.
,
Wang
Y.
 &
Ding
S. M. J.
2020
A critical review of the appearance of black-odorous waterbodies in China and treatment methods
.
J. Hazard. Mater.
385
,
121511
.
https://doi.org/10.1016/j.jhazmat.2019.121511
.
Google Scholar
Crossref
Search ADS
PubMed
 
Fotidis
I. A.
,
Treu
L.
 &
Angelidaki
I.
2017
Enriched ammonia-tolerant methanogenic cultures as bioaugmentation inocula in continuous biomethanation processes
.
J. Cleaner Prod.
166
,
1305
–
1313
.
http://dx.doi.org/10.1016/j.jclepro.2017.08.151
.
Google Scholar
Crossref
Search ADS
 
Liu
W. C.
 &
Zhu
P.
2018
Demonstration-scale high-cell-density fermentation of Pichia pastoris
.
Methods Mol. Biol.
1674
,
109
–
116
.
doi: 10.1007/978-1-4939-7312-5_9
.
Google Scholar
Crossref
Search ADS
PubMed
 
Mathur
A. K.
,
Majumder
C. B.
,
Chatterjee
S.
 &
Roy
P.
2008
Biodegradation of pyridine by the new bacterial isolates S. putrefaciens and B. sphaericus
.
J. Hazard. Mater.
157
(
2–3
),
335
.
https://doi.org/10.1016/j.jhazmat.2007.12.112
.
Google Scholar
PubMed
 
MEP
2002a
Environmental Quality Criteria for Surface Water (GB3838-2002)
.
General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China
,
Beijing
,
China
.
MEP (Ministry of Environmental Protection)
2002b
Chinese Water and Wastewater Monitoring and Analysis Methods
, 4th edn.
Chinese Environmental Science Press
,
Beijing
,
China
.
Ozbayram
E. G.
,
Kleinsteuber
S.
,
Nikolausz
M.
,
Ince
B.
 &
Ince
O.
2017
Effect of bioaugmentation by cellulolytic bacteria enriched from sheep rumen on methane production from wheat straw
.
Anaerobe
46
,
122
–
130
.
https://doi.org/10.1016/j.anaerobe.2017.03.013
.
Google Scholar
Crossref
Search ADS
PubMed
 
Ozbayram
E. G.
,
Kleinstruber
S.
,
Nikolausz
M.
,
Ince
B.
 &
Ince
O.
2018
Bioaugmentation of anaerobic digesters treating lignocellulosic feedstock by enriched microbial consortia
.
Eng. Life Sci.
18
,
440
–
446
.
doi: 10.1002/elsc.201700199
.
Google Scholar
Crossref
Search ADS
PubMed
 
Pradhan
S.
 &
Rai
L. C.
2000
Optimization of flow rate, initial metal ion concentration and biomass density for maximum removal of Cu2+ by immobilized microcystis
.
World J. Microbiol. Biotechnol.
16
,
579
–
584
.
https://doi.org/10.1023/A:1008987908001
.
Google Scholar
Crossref
Search ADS
 
Ren
L. J.
,
Xu
L. L.
,
Zhang
Y. Y.
,
Pan
W.
,
Yin
S. L.
,
Zhou
Y.
,
Yu
L. J.
,
Chen
Y. S.
 &
An
S. Q.
2016
Effects of connection mode and hydraulic retention time on wastewater pollutants removal in constructed wetland microcosms
.
Clean (Weinh)
43
(
12
),
1574
–
1581
.
https://doi.org/10.1002/clen.201300842
.
Google Scholar
 
Shen
J. Y.
,
Zhang
X.
,
Chen
D.
,
Liu
X. D.
,
Zhang
L. B.
,
Sun
X. Y.
,
Li
J. S.
,
Bi
H. P.
 &
Wang
L.
2014
Kinetics study of pyridine biodegradation by a novel bacterial strain, Rhizobium sp. NJUST18
.
Bioprocess. Biosyst. Eng.
37
(
6
),
1185
–
1192
.
doi: 10.1007/s00449-013-1089-x
.
Google Scholar
Crossref
Search ADS
PubMed
 
Song
C.
,
Liu
X. L.
,
Song
Y. H.
,
Liu
R. X.
,
Gao
H. J.
,
Han
L.
 &
Peng
J. F.
2017
Key blackening and stinking pollutants in Dongsha river of Beijing: spatial distribution and source identification
.
J. Environ. Manage.
200
,
335
–
346
.
https://doi.org/10.1016/j.jenvman.2017.05.088
.
Google Scholar
Crossref
Search ADS
PubMed
 
Wang
G. F.
,
Li
X. N.
,
Fang
Y.
 &
Huang
R.
2014
Analysis on the formation condition of the algae-induced odorous black water agglomerate
.
Saudi J. Biol. Sci.
21
,
597
–
604
.
doi: 10.1016/j.sjbs.2014.07.002
.
Google Scholar
Crossref
Search ADS
PubMed
 
Wang
L.
,
Yu
L.
,
Xiong
Y.
,
Li
Z.
 &
Geng
J.
2021
Study on the governance of black-odor water in Chinese cities
.
J. Cleaner Prod.
308
,
127290
.
https://doi.org/10.1016/j.jclepro.2021.127290
.
Google Scholar
Crossref
Search ADS
 
Xu
Y.
,
Song
C.
,
Song
N.
,
Wang
J.
,
Yue
Z.
 &
Liu
X.
2019
Condition optimization and kinetic characteristics of S2− bio-oxidation in a black-stinking water body by composite microorganisms
.
Chinese J. Environ. Eng.
13
(
3
),
530
–
540
(in Chinese)
.
doi: 10.12030/j.cjee.201901006
.
Google Scholar
 
Yin
H.
,
Yang
P.
 &
Kong
M.
2019
Effects of nitrate dosing on the migration of reduced sulfur in black odorous river sediment and the influencing factors
.
Chem. Eng. J.
371
,
516
–
523
.
https://doi.org/10.1016/j.cej.2019.04.095
.
Google Scholar
Crossref
Search ADS
 
Yu
H.
,
Liu
J.
,
Zhai
J.
,
Shi
Y.
,
Liu
Z.
 &
Sun
J.
2014
Effect of HRT on treatment of black water from high-speed train using heterotrophic nitrification-aerobic denitrification bacteria
.
Chinese J. Environ. Eng.
8
(
11
),
4715
–
4720
.
Google Scholar
 
Zhuang
R. Y.
,
Lou
Y. J.
,
Qiu
X. T.
,
Zhao
Y. Y.
,
Qian
D.
,
Yan
X. J.
,
He
X. D.
,
Shen
Q. Q.
 &
Qian
L. Z.
2017
Identification of a yeast strain able to oxidize and remove sulfide high efficiently
.
Appl. Microbiol. Biotechnol.
101
,
391
–
400
.
https://doi.org/10.1007/s00253-016-7852-6
.
Google Scholar
Crossref
Search ADS
PubMed
 
© 2022 The Authors
This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/).