Strategy for Rapid Recovery of Simultaneous Sulfide and Nitrite Removal under High Substrate Inhibition


 The paper deals with the strategy for a quick recovery of reactor treating sulfide and nitrite simultaneously under inhibition caused by high substrate concentration. For influent sulfide concentration of 360 mg S/L, respective sulfide and nitrite removal percentages dropped to 74.19% and 14.33% due to inhibition caused by high sulfide and nitrite concentrations. It was found that reduction in the influent substrate concentration (300 mg S/L) could not revive the nitrite removal performance in 4 days’ operation, which still showed a declining tendency from 47.16–18.52%. Regulating the influent pH around 6.70 ± 0.10, it only took 4 days to recover the performance for 300 mg S/L. Furthermore, at influent sulfide concentration increased to 360 mg S/L, respective sulfide and nitrite removal percentages were 99.76 ± 0.27% and 100%. The strategy of regulating influent pH could recover the process performance in a short term, which would provide great convenience for subsequent process research.


Introduction
Pulp production, pharmaceuticals, and oil re ning industries generate large amounts of sulfur and nitrogen-containing wastewater (Show et al. 2013). Simultaneous sul de and nitrite removal process is well-established autotrophic biological denitri cation process, which can achieve sul de and nitrite removal at the same time, and recovers elemental sulfur from wastewater, so this novel process has broad application prospects (Mahmood et al. 2007). High in uent substrate concentration would pose an inhibitory effect on the process performance (Yuan et al. 2020). The process performance would be greatly deteriorated upon introduction of substrate concentration at certain level (Zhang et  However, little has been investigated about the recovery of the process performance once inhibited by high in uent substrate. Generally, reducing the in uent substrate concentration is an effective strategy to solve the problem, sometimes it takes a long time to be restrained by the strategy (Zheng et al. 2013). In case of failure to restore the process performance by reducing the in uent substrate concentration failed to take effect on mitigate the deterioration of the performance, what strategy should be adopted? The objective of the present study was to answer the above-mentioned question. The sul de and nitrite removal performance of was investigated at various sul de concentrations (60-360 mg S/L), and the nature of products was investigated. After the process inhibition due to high in uent sul de concentration, two strategy were applied to recover the performance, which included the reduction of the in uent substrate concentration and regulation of the in uent pH. Based on the Principal Component Analysis (PCA) and multiple comparisons, the effectiveness of regulating the in uent pH for the rapid performance recovery was validated.

Materials And Methods
Page 3/12 2.1 Inoculum, reactor and synthetic wastewater The inoculum was collected from the anaerobic reactor working at Municipal Wastewater Treatment Plant (WWTP) in Hangzhou city. Its volatile suspended solids (VSS) were 6.09 g/L.
A laboratory-scale EGSB reactor was operated during the experiment, which was made of perspex with an effective volume of 6 L. A peristaltic pump was used to inject the synthetic wastewater, and the other pump was used to control the re ux ratio of 1:2. The hydraulic retention time (HRT) was xed at 13.33h, while the temperature was controlled at 25 ± 2°C.

Experimental procedure
The operation of the reactor was divided into three stages, which lasted for 62 d. Table 1 enlists the detailed operating conditions of the bioreactor during different stages. In Stage I, the in uent sul de concentration was increased from 60 mg S/L to 360 mg S/L with a gradient of 60 mg S/L. At in uent sul de concentration of 360 mg/L, the simultaneous sul de and nitrite removal performance process was inhibited due to lower sul de and nitrite removal percentages. During Stage II, decreasing the in uent substrate concentration (300 mg S/L) was tried to recover the performance of the reactor; however, it did not work on short term basis. Finally, decreasing the in uent pH (6.70 ± 0.10) was applied to recover the reactor performance, and the in uent sul de concentration was increased to 360 mg S/L.

Analytical procedures
The in uent and e uent pH, nitrate-nitrogen, nitrite-nitrogen, ammonia-nitrogen and sul de-sulfur were analyzed according to standard methods (APHA et al. 1998). An ion chromatograph (DIONEX-900) analyzed thiosulfate-sulfur, sul te-sulfur and sulfate-sulfur concentration after passing through a 0.45 µm lter. The elemental sulfur and nitrogen gas were estimated according to principle of mass conservation, which were calculated as Eq. (1) and Eq. (2). 3.1 Operation performance by elevating in uent substrate concentration

Rapid recovery strategy for high substrate inhibition
In order to recover the performance of the bioreactor, the strategy of reducing in uent substrate concentration (Stage II) was primarily used. In 4 days' operation, sul de removal percentage quickly recovered to 91.21 ± 1.78%, however, nitrite removal percentage still showed a declining tendency, which descended from 47.16-18.52%. The results showed that reducing the in uent substrate failed to recover the nitrite removal performance on a short-term basis (4 days). During a severe nitrite inhibition (its concentration was higher than inhibition parameter), Tang et al. also found that nitrite removal performance was hard to be recovered by decreasing substrate concentration, only achieved 89% after 39 days, which was due to high inhibition effect of nitrite .
Many studies have shown that pH has a signi cant effect on the denitri cation process (Karanasios et al. 2010). PCA was used to analyze the relation of ΔpH and products of the stage I (60-300 mg S/L). Two principal components were extracted, and the equations related to principal components and ΔpH were established (Eqs. [3][4][5], which suggested an excellent linear relationship between the concentrations of products and ΔpH (Fig.S1, R 2 = 0.9790). It was supposed that the performance of the bioreactor at 360 mg S/L was as good as that at 300 mg S/L, ΔpH was as high as 2.41 according to Eq. (3), which meant that the e uent pH of 360 mg S/L reached 9.41, much higher than inhibited pH of 9.0 for denitri cation (Chung et  Hence, the strategy of regulating the in uent pH was applied during Stage III. The in uent pH was decreased to 6.70 ± 0.10 by keeping the in uent sul de concentration at 300 mg S/L as in Stage II. It only took 4 days to recover the process performance. On the 4th day of Stage III, sul de and nitrate removal percentages were higher than 99.90% and 90.30%, respectively; furthermore, sul de and nitrate removal percentages were even higher than 300 mg S/L in Stage I, without any detection of the e uent sul de and nitrite. With the in uent sul de concentration going up to 360 mg S/L, the bioreactor still showed a good performance, which sul de and nitrite removal percentages were 99.76 ± 0.27% and 100%, respectively. The results provided a strong evidence for the validity of strategy of in uent pH regulation.

Comparison of both strategies
As shown in Table 2, there were signi cant differences in the removal performance (e uent sul de and nitrite), main sulfur-containing products (elemental sulfur and Eff. sulfate) and main nitrogen-containing products (nitrogen gas) between 300 mg S/L in Stage I and that in Stage II, which suggested that reduction in the concentration did not recover the process on short-term basis. No signi cant differences observed on removal performance when the in uent pH was regulated at 6.70 ± 0.10, and main products were intermediate nature of 300 mg S/L in Stage I and those in Stage III; however, signi cantly different products were found between 360 mg S/L in Stage I and those in Stage III. The results proved that regulating the in uent pH effectively succeeded to recover the process on a short-term basis (4 days). It is worth noting that the variations of minor products (thiosulfate and nitrate) and consumed ammonium were a little different from those of removal percentage and main products, which inferred that regulating the in uent pH would cause a slight impact on product types.

Conclusions
The two strategies of reducing in uent substrate concentration and regulating in uent pH were used to restore the process performance of simultaneous sul de and nitrite removal after inhibition caused by high substrate concentration. For a gradual increase in the in uent sul de concentration from 60 mg S/L to 300 mg S/L, the sul de and nitrite removal percentages were 92.65 ± 5.94% and 99.91 ± 0.30%, with in uent pH controlled at 7.0 ± 0.1. Upon increase in the in uent sul de concentration to 360 mg S/L, there was a sudden reduction of sul de and nitrite removal performance, which dropped to 74.19% and 14.33%, respectively. Reducing the in uent substrate concentration to 300 mg S/L, sul de removal percentage quickly returned to 91.21 ± 1.78% in 4 days' operation; however, nitrite removal percentage still showed a declining tendency, descending from 47.16-18.52%. Upon regulating the in uent pH around 6.70 ± 0.10, the performance recovered in 4 days at 300 mg S/L. Furthermore, the sul de and nitrite removal percentages were 99.76 ± 0.27% and 100%, respectively, with in uent sul de concentration going up to 360 mg S/L. Regulating the in uent pH for the rapid performance recovery of was signi cant based on multiple comparisons (p < 0.05). The strategy of regulating in uent pH for the process performance recovery in a short term (4 days) may be greatly convenient for subsequent research.

Declarations
Ethics approval Not applicable.

Consent to participate
Not applicable.

Consent for publication
Not applicable.
Authors' contributions KW performed the experiment and wrote the initial draft. MQ was a major contributor in editing the manuscript. BC participated in the experiment. SL contributed to the conception of the study. YW and ZZ helped perform the analysis with constructive discussions. JC provided nancial support for the experiment, and contributed signi cantly to analysis and manuscript preparation. All authors read and approved the nal manuscript. Sul de removal performance in the simultaneous sul de and nitrite removal process Nitrite removal performance in the simultaneous sul de and nitrite removal process

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. SupplementalMaterial.docx