Advanced treatment of secondary effluent by the integration of heterogeneous catalytic ozonation and biological aerated filter

In the municipal wastewater treatment plant, the conventional treatment processes, e.g. the mainstream activated sludge anaerobic–anoxic–oxic (A²/O) and sequencing batch reactor (SBR) systems, are commonly used (Su et al. 2021; Singh et al. 2022). However, limited by the biodegradability of pollutants, there are still some refractory or toxic organic pollutants that remained in the biological effluent, including pesticides, antibiotics, sunscreen, etc. (Wu et al. 2018; Shreve & Brennan 2019; Tang et al. 2022). The pollutants that cannot be effectively removed are discharged into the environment directly with sewage. Although they usually exist in the water ecosystem at light concentration, long-term exposure will damage the micro-ecosystem and cause the presence of drug-resistant pathogens. Therefore, the advanced treatment of secondary effluent has received extensive attention (Gonder et al. 2020; Satayavibul & Ratanatamskul 2021).

Nowadays, advanced oxidation technologies (AOTs) have been intensely investigated and often used for advanced sewage treatment, e.g. ozonation, Fenton/Fenton-like reactions, photo-catalysis, and the activated persulfate method (Chavez et al. 2020; Su et al. 2021; Tang et al. 2021b). Among these advanced sewage treatment methods, ozonation is considered a promising advanced technology, because it saves equipment space, produces less sludge, reacts faster, and does not require extra energy or soluble chemicals. The catalytic ozonation can improve the biodegradability of wastewater, reduce the refractory pollutants and also partly decrease chemical oxygen demand (COD). For example, Zhang et al. (2014) found that biochemical oxygen demand (BOD₅)/COD of coking wastewater could increase from 0.09 to 0.43, indicating that the biodegradability of the wastewater was improved, and the residual refractory pollutants with large molecules could be transformed into smaller ones after catalytic ozonation treatment (Liu et al. 2021, 2022a). Qiu et al. (2022) used the catalytic ozonation system to treat refractory organics from secondary effluent, and the different pollutants had different reaction rates with ozone/hydroxyl radicals (•OH). Also, these products were difficult to be completely mineralized in a short period of time (Chen et al. 2022; Wang et al. 2022). Excitedly, the generated small molecules could be easily removed by the biological treatment process. The biological aerated filter (BAF), as an alternative fixed-film process, had been widely used to treat various wastewater, including municipal and industrial wastewaters (Vatankhah et al. 2019; Li et al. 2020), due to its excellent characteristics, involving a small footprint, simple operation and maintenance, higher biomass concentration, less sludge production, and effective pollutants removal (Zhou & Xu 2020). Many studies investigated the pollutants removal ratios, operational parameters optimization, and microbial community structures for established BAF systems (Ren et al. 2019; Xiang et al. 2021). BAF was with a vertical column, thus the bio-community could readily adapt to the different levels of dissolved oxygen (DO) and/or renewable water quality. Therefore, an integration of catalytic ozonation and the BAF system would be a satisfactory method for enhancing the removal capacity of pollutants. Also, the BAF structures and types of filter media were related to installation costs and removal ratios of pollutants (Xiang et al. 2021). BAF with various structures and different types of filter media had been prepared, and it was impractical and cost-prohibitive to evaluate their performance individually (Xiang et al. 2021; Nikoonahad et al. 2022). If there was a way to relate influent COD to effluent COD and reactor height for specific media, it would be critical for providing important information for the design and optimization of water treatment processes towards cost-effectiveness.

In this study, the effects of catalytic ozonation coupled with up-flow BAF on the advanced treatment of secondary effluent were investigated. The influences of catalytic ozonation on secondary effluent treatment are described in detail, including water quality parameters (e.g. COD, UV₂₅₄, -N, and total organic carbon (TOC) and economic analysis. Importantly, the COD removal kinetics behavior as a function of influent COD concentration in the BAF was investigated. This study provides a reference for the advanced treatment of secondary effluents for practical purposes.

The raw water sample was the secondary effluent obtained from a conventional activated sludge system (anaerobic–anoxic–oxic (A²/O) process) in a municipal wastewater treatment plant in Changsha, China. The removal of soluble substances was deeply researched, therefore, to avoid the negative effect of suspended solid (SS), the raw water was filtered by 0.45 μm glass fiber film in advance. The water characteristics are shown in Table 1.

Table 1

The raw water characteristics of secondary effluent

Parameter .	COD (mg/L) .	TOC (mg/L) .	TNs (mg/L) .	-N (mg/L) .	UV254 (Abs) .	pH .
Secondary effluent	52.2–55.3	4.15–4.34	–	74.3–77.8	0.13–0.16	7.2–7.8

Parameter .	COD (mg/L) .	TOC (mg/L) .	TNs (mg/L) .	-N (mg/L) .	UV254 (Abs) .	pH .
Secondary effluent	52.2–55.3	4.15–4.34	–	74.3–77.8	0.13–0.16	7.2–7.8

View Large

The catalytic ozonation reactor held 1 L of reaction solution with 0.5 g/L catalysts (Mn_xCu_yO_z/γ-Fe₂O₃), which had been proven to possess excellent catalytic activity against ozone by a previous study (Liu et al. 2022a). The synthesis method of Mn_xCu_yO_z/γ-Fe₂O₃ is shown in Supplementary Materials, Text S1. Furthermore, the characteristics including crystal structure and surface morphology are supplied in Text S2. The generated ozone via the ozone generator entered the reaction solution through the porous aeration plate and the gas flow rate and ozone dosage were about 200 mL/min and 10 mg/L, respectively. For the BAF reactor, the ozone-pretreated secondary effluent/secondary effluent entered from the bottom of the up-flow BAF, which mixed well with the air generated by the micro-porous air diffuser at the bottom of the column. The effective volume and the hydraulic retention time (HRT) were about 2 L and 2 h, respectively. The flow rate of air was controlled according to the requirement of a gas-water ratio of 3:1.

The concentrations of water characteristics including COD, BOD, and -N were detected via the standard methods (APHA 2015). The DO and pH were routinely monitored by a DO meter (HQ30d, HACH, USA) and pH meter (pHSJ-3F, Leici, China), respectively. The TOC was measured by the TOC analyzer (TOC-V_CPH, Shimadzu, Japan). The UV₂₅₄ was detected by a UV-visible spectrophotometer (TU-1810, PERSEE, China). In addition, the soluble organic matter was measured by a three-dimensional fluorescence spectrometer (HORIBA, Canada). The concentrations of ozone in the gas phase and solution were determined by iodometry and indigo methods, respectively (Liu et al. 2018).

Biofilm samples were detached from the surface of ceramsite in the BAF reactor. The samples were obtained from the single BAF (Sample 1) and catalytic ozonation coupled BAF (Sample 2) systems. DNA was extracted using a PowerBiofilm™ DNA Isolation Kit for Biofilm (Mo Bio, Carlsbad, CA, USA) and stored at −80 °C. The microbial community structure was further analyzed with high-throughput sequencing technology, using Illumina Hiseq 2000 from the Shanghai Meiji Biomedical Technology Co., Ltd (Shanghai, China).

By plotting ln(C/C_i) against H, the slope K can be obtained for different values of C_i. The slope m and the intercept lnk^∗ can be determined from the plot of ln(KQ/A) against ln(C_i), respectively.

Figure 5(b) displays the variation of -N with operating time. In the first 30 days, the removal ratio of -N fluctuated with the increasing operation time, while, after 30 days, the removal ratio remained relatively stable. In the process of the established BAF, -N decreased from 77.43 to 22.69 mg/L with a removal ratio of about 70%. In the catalytic ozonation coupled with the BAF system, at the same condition, the -N decreased from 77.43 to 15.73 mg/L in the catalytic ozonation coupled with the BAF system and the removal ratio could reach about 78%. In the initial stage, the nitrifying and denitrifying bacteria had an adaptative process to water quality with residual ozone (Zhou et al. 2019). With longer operation times, the bacteria gradually accumulated and enriched on the surface of the ceramsite to form a stable biofilm, which facilitated the removal of -N. According to a review study by Jin et al. (2023), the microorganisms with significant degradability were in the BAF system. Most of the denitrifying bacteria were nitrite-oxidizing bacteria and ammonia-oxidizing bacteria. In detail, in the aerobic environment, the -N was oxidized to by the ammonia-oxidizing bacteria, including Nitrosomonas, Nitrosococcus, Nitrosolobus, and Nitrosovibrio, etc. (Ushiki et al. 2018). The was further oxidized to generate the by the nitrite-oxidizing bacteria, including Nitrobacter, Nitrospina, Nitrococcus, etc. (Friedrich et al. 2020). At last, the generated was reduced to N₂ under the anoxic conditions by the heterotrophic denitrifying bacteria (Daims et al. 2015). Therefore, the -N in the solution was supposed to be transformed into N₂ in this study. However, the study by Bao et al. (2016) used commercially available ceramsite as the filter media for BAF to remove -N and the removal ratio was about 67.29%. In this study, the BAF with homemade ceramsite media could remove 70% -N. Therefore, the homemade sludge base ceramsite might be suitable for use as filter media for the simultaneous removal of -N and COD.

The TOC and UV₂₅₄ of the solution are shown in Figure 5(c). TOC showed a trend of decreasing volatility. In the catalytic ozonation coupled with the BAF system, the TOC had slightly increased, which was caused by the shedding of part of the biofilm in the initial stage. At last, the TOC removal ratio tended to stabilize. The changing trend was consistent with that of the COD. In addition, the UV₂₅₄ was also volatile in both systems (0.04–0.085 Abs in BAF and 0.02–0.075 Abs in catalytic ozonation coupled with BAF). The pH and DO had important influences on the growth of microorganisms. Figure 5(d) shows the change in pH and DO during the operation. The pH of the solution was about 7–8, and the DO was about 4–5 mg/L. The above conditions were suitable for the growth of microbial films, which could promote the removal of pollutants by microorganisms (Tang et al. 2021a).

To evaluate the removal effect of the established integration of heterogeneous catalytic ozonation and the BAF system on pollutants, Table S1 summarizes the system performance of BAF combined with other processes. Zhang et al. (2014) used the ozonation-coupled BAF system to treat the bio-treated coking wastewater with 59 and 76% removal of COD and -N, respectively. Farabegoli et al. (2009) adopted the chemical precipitation coupled BAF system to dispose of domestic sewage with 33, 96, and 66% removal of COD, -N, and total suspended solid (TSS), respectively. In addition, Xu et al. (2015) used heterogeneous Fenton oxidation (HFO) with an anoxic moving bed biofilm reactor to treat the biologically pretreated coal gasification wastewater, and the removal ratios of COD, total phenols and total nitrogen (TN) were 82, 83, and 85%, respectively. In this study, we used the integration of heterogeneous catalytic ozonation and BAF to dispose of the secondary effluent from the waste water treatment plant (WWTP), and the removal ratios of COD and -N were 73 and 70%, respectively. Based on these results, it would be a trend to combine biofilters with advanced oxidation processes to improve pollutant degradation efficiency with low energy consumption were a tendency. The BAF treatment was once very popular due to its stable discharge for high standards of effluent.

The Shannon, Simpson, and Chao1 index could all reflect the microbial diversity variation (Fu et al. 2021). The high Shannon index indicated rich community diversity, and the low Simpson index indicated rich community diversity (Ding et al. 2018). The Chao1 index was the same as OTUs, they all could account for changing species in community abundance and biofilms. As shown in Table 2, the Shannon, Simpson, and Chao1 indices of the microbial film in BAF were 5.2037, 0.0178, and 772.176, respectively, and the indices were 4.2132,0.0368, and 576.782, respectively, in catalytic ozonation coupled BAF. These indices indicated that the biofilms in BAF exhibited high community richness, while the biofilms in catalytic ozonation coupled BAF had the advantage of community diversity. The residual ozone in the solution could screen microorganisms and the available carbon sources promoted thriving microbial growth.

Figure 6(b) shows the Venn diagram of OTUs for Samples 1 and 2. The common OTUs between Samples 1 and 2 were 380, while the proprietary OTUs were 380 and 119, respectively. The differences in the microbial communities in Samples 1 and 2 were small. The reason for this difference might be the screening effect of the residual ozone in the sewage on the microorganisms (Zhang et al. 2018).

A BAF equipped with homemade ceramsite was established for the efficient and stable pretreatment of secondary effluents. The surface of the filter was conducive to the attachment of microorganisms, which could effectively remove the COD and -N from the solution. In the BAF system, the COD (⁠-N) could decrease from 54.26 to 32.56 mg/L (decrease from 77.43 to 22.69 mg/L). Based on the experimental conditions and equipment, the COD values at different BAF heights were expressed as a function of influent COD concentration, lnC/C_i = −2.53H/(⁠ ). Importantly, the catalytic ozonation system could improve the biodegradability of secondary effluents, which facilitated subsequent microbial treatment. In the catalytic ozonation coupled with the BAF system, the COD and -N could decrease to 14.65 and 15.73 mg/L, respectively. In addition, the microbial community variation illustrated that the residual ozone in the solution could screen microorganisms and the available carbon sources in the ozone-catalyzed solution promoted thriving microbial growth. Therefore, the combined catalytic ozonation and BAF process as a promising technology can improve the overall treatment efficiency of secondary effluent.

The financial support from the Hunan Provincial Key Research and Development Program (2022SK2068) is acknowledged.

X.L. investigated the study, did data curation, and wrote the original draft. L.C. and J.P. edited the draft. H.L. and Y.Y. conceptualized and supervised the study, wrote the review and edited the manuscript. Z.Y. and L.Y. did the project administration.

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.

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

The authors declare there is no conflict.