Treatment of organic pollutants in coke plant wastewater by micro-nanometer catalytic ozonation, A/A/O and reverse osmosis membrane Open Access

Coking wastewater is a waste product of coal coking, gas purification, chemical product recovery, and chemical product refining (Zhang et al. 1998). The properties of coking wastewater depend on many factors, including the production of raw materials, the production facilities, the production technology, and the geographical and meteorological environment. These factors can cause large fluctuations in the coking wastewater quality (He et al. 2019). Coking wastewater is typically toxic and contains harmful organics including volatile phenols, polycyclic aromatic hydrocarbons (PAH), and heterocyclic compounds. Toxic and carcinogenic PAHs are difficult to degrade and could cause serious pollution to the environment and are also a direct threat to human health (Kim et al. 2007). Inorganic pollutants in the wastewater such as nitride, cyanide, thiocyanate, and fluoride are also problematic (Ho & Lee 2002; Kong et al. 2018). This highly contaminated wastewater poses a great challenge to society in terms of handling and disposal (Mishra et al. 2021). It is difficult to realize an appropriate degradation of coking wastewater organics in a single biochemical or physicochemical process and it usually requires a complex and tedious treatment process (Kang et al. 2022).

Currently, biological treatment for coking wastewater is still the most popular technology due to its low cost (Zhu et al. 2018). The toxic coking wastewater can seriously impair the microbial activity in the activated sludge (Ye et al. 2018). It is therefore vital to perform pretreatment before biochemical treatment, which could both improve the biodegradability of the wastewater and reduce the load of the subsequent treatment (Du et al. 2022). Coagulation sedimentation, air flotation, and filtration can be used to remove oil, colloids, and suspended matter (Tong et al. 2018). Advanced oxidation technologies are the most important pretreatment method, where organic matter undergoes a strong oxidation reaction with inorganic matter containing hydroxyl radicals. This refractory organic matter is converted into low-toxic or non-toxic small-molecule organic matter such as carbon dioxide and water through oxidative degradation (Liu et al. 2019). Ozone oxidation technology, an advanced oxidation technology, has attracted wide attention due to its high efficiency, safety, and large processing capacity (Szpyrkowicz et al. 2001).

However, high energy costs severely limit the development of ozonation systems. The rate of ozone use in wastewater is also extremely low (Orge et al. 2011; Pugazhenthiran et al. 2011). A micro-nano-bubble (MNB) generator could solve this problem as it can dissolve air, oxygen, ozone, carbon dioxide, and other gases into the water with bubbles between microns and nanometers (Fan et al. 2020; Kwon et al. 2020). MNBs, in comparison to normal bubbles, have a specific surface area, slow rising speed, self-pressurized dissolution, a high surface charge, and high mass transfer efficiency (Seddon et al. 2012; Hu & Xia 2018; Gao et al. 2019). Ozone MNBs can therefore increase the dissolution rate of ozone in water and improve the removal and biodegradability of organic substances in organic wastewater (Xiao et al. 2019).

The pretreatment process of coking wastewater, followed by different forms of biochemical processes is not enough to meet the strict requirements of the Chinese national discharge standard (GB16171-2012) (Zhang et al. 2016). In recent years, membrane separation technology has developed rapidly. Ultrafiltration, filtration, and reverse osmosis technologies have become the research hotspots for depth processing and reuse of coking wastewater (Wang 2014).

Although previous studies show the successful removal of one or two organic pollutants from synthetic wastewater (He et al. 2020), the process of achieving the standard discharge of chemical oxygen demand (COD) of the coking wastewater is still unclear. This work optimized the treatment effects of the coagulation air flotation process, the micro-nanometer catalytic ozonation process, A/A/O process, and the reverse osmosis membrane treatment process on coking wastewater. A combination process was developed to reach the discharge standard in China (GB 16171-2012, COD < 60 mg/L). The results presented here provide a new and efficient combined process for coking wastewater treatment.

The wastewater was collected from the coking plant in Pingdingshan city (Henan Province, China). The raw water quality in terms of organic and inorganic matter is shown in Table 1 and S1 (details supplied in the Supporting Information).

The analyses of pH, COD, ammonia nitrogen biological oxygen demand (BOD₅), and dissolved oxygen were conducted following the standard methods (State Environmental Protection Administration of China 2002). Gas chromatography–mass spectrometry analysis was done to identify the organic pollutant fractions in the raw water.

The water sample for the ozone test was taken from the coagulation air floating water.

The influence of pH involved the following: the initial pH of 500 mL of the coking wastewater was adjusted to pH 4, 6, 8, 10, and 12. The water was then placed in a cylinder of 10 cm in diameter and 40 cm in height. The ozone is produced by an ozone generator with a pure oxygen source (XLK-G10, Changsha Xianglu Environmental Protection Technology Co., Ltd, China). The ozone was introduced into the reactor by continuous inflation. The concentration of ozone in the gas phase was monitored with an ozone analyzer (CY-1A, Beijing Hongchang New Technology Co., China). The average ozone concentration was 80 mg/L, the gas volume was 1.5 L/min, and the bubble generator was a commercial aeration tray (N-50, Shuyang Hongyuan Timber Factory, China). After a reaction time of 0.5 h, the COD value was determined.

The influence of the heterogeneous ozone catalyst involved the following: after the pH was adjusted to 10, the coconut shell activated carbon (YK-AC), apricot shell activated carbon (XK-AC), wooden activated carbon (MZ-AC), and coal activated carbon (US-AC) of the commercial catalysts were soaked in the coking wastewater overnight and then dried at 100 °C until dry to avoid the adsorption of the catalyst. The catalysts were used for the catalytic ozonation reaction at an aeration rate of 1.5 L/min, and a catalyst dosage (liquid:solid in wt%) of 3:1 at a reaction time of 0.5 h. Samples were taken to determine the COD value.

The effect of the dosage of catalyst involved the following under a pH of 10, the aeration rate was 1.5 L/min, and the reaction time was 0.5 h, while the liquid–solid mass ratio of the wastewater and the US-AC catalyst was adjusted to 50:1, 25:1, 15:1, 10:1, 5:1, and 3:1. Samples were taken to determine the COD value.

Before adding the coking wastewater to the reactor for the experiment, the pH was adjusted to 10. The US-AC catalyst dosage (liquid solid, wt%) was 3:1, the average ozone concentration was 80 mg/L, and the gas volume was 1.5 L/min. The bubble generator was a micro-nano gas disperser (HXWNM-J10, Pingdingshan Huaxing flotation Engineering Technical Service Co., Ltd, China). The average diameter of the bubbles according to the supplier was 600 nm, the reaction time was 3 h, and the COD value was measured every 0.5 h.

The biodegradability of the wastewater in the ozone reaction process was determined by using BOD/COD to evaluate the biodegradability of wastewater. The COD was determined by potassium dichromate spectrophotometry and the BOD was determined by the inoculation dilution method.

The biochemical reaction feed water was taken from micro-nano-oxidized ozonated effluent. The test used an inverted A/A/O reaction device. Figure S1 shows the schematic diagram of the A/A/O reaction device. The effective volumes of the anoxic, anaerobic, and aerobic sections were 1.8 L. The filler in each tank was semi-soft, while the filling rate of the anoxic and the anaerobic sections were about 60% while the aerobiotic section was about 20%. The reaction conditions included an influent flow rate of 0.75 mL/min and the total hydraulic retention time of the system was 5 days. The reflux ratio of the nitrification solution was 200% and the dissolved oxygen in the aerobic section was 2–3 mg/L. The dissolved oxygen in the anoxic tank was 0.2–0.5 mg/L, and the dissolved oxygen in the anaerobic tank was less than 0.2 mg/L. The pH was adjusted to about 6.8–7.5 with 10% hydrochloric acid or sodium hydroxide, and the COD value of the effluent from the biochemical reactor was measured every day.

To determine the dominant microbial communities in the reactor, the following nine sludge samples were analyzed: the suspended sludge, the filler-attached sludge, and the bottom sludge in the wastewater from the anoxic reaction tank (labeled Q1, Q2, and Q3), the suspended sludge, the filler-attached sludge, and the bottom sludge in the wastewater from the anaerobic reaction tank (labeled Y1, Y2, and Y3), and the suspended sludge, the filler-attached sludge, and the bottom sludge in the wastewater from the aerobic reaction tank (labeled H1, H2, and H3).

The PowerSoil™ DNA Isolation Kit (Mobio, USA) was used to extract the microbial genomic DNA. The DNA integrity and purity were examined using 1% agarose gel electrophoresis, and the DNA concentration and purity were examined using the NanoDropOne (NanoDrop 2000, Thermo Fisher Scientific, USA). The V4 hypervariable region of the 16S rRNA gene was amplified via polymerase chain reaction (PCR) (in triplicate reactions for each sample) using the primers F515 (5′-GTGCCAGCMGCCGCGGTAA-3′) and R806 (5′-GGAC-TACVSGGGTATCTAAT-3′). The procedures described by Zhu et al. (2019) were followed for this amplification. The volume required for each sample was calculated according to the equal mass principle, and each PCR product was mixed based on the concentration comparison of the PCR products using the GenToolsAnalysisSoftware (Version4.03.05.0, SynGene). The EZNA^® GelExtractionKit gel recovery kit was used to recover the PCR products, while the target DNA fragments were recovered by elution in the Tris-ethylenediaminetetraacetic acid (Tris-EDTA) buffer. The NEBNext^®Ultra™DNALibraryPrepKitforIllumina^® standard procedure was followed to build the library and, after completion of the library, the high-throughput sequencing platform MiSeq was used to perform sequencing.

The reverse osmosis membrane (BONA-GM-19, Jinan Bona Biotechnology Co., Ltd China) was used to treat the biochemical effluent. The wastewater entered the reverse osmosis system, after which a concentrate and a clear solution were obtained at a pump pressure of 0.9 MPa. The COD values of the concentrate and solution were then measured.

The adsorption experiments were carried out using fly ash added to 500 mL of reverse osmosis concentrate. The fly ash was provided by Henan Fly Ash Comprehensive Development and Utilization Center, Henan Province. The fly ash was dried, and then finely ground and passed through a 0.074 mm sieve. The adsorption experiments were done at a fly ash concentration of 20% (mass fraction) and at reaction times of 15, 30, 45, and 60 min. The concentrated water solution containing the fly ash was stirred at 400 rpm on a magnetic stirrer at 30 °C.

The pH of the initial untreated wastewater was adjusted to 9. The 2,600 mg/L of PFS and the 6 mg/L of PAM was added before 20 min of flocculation, then air flotation elapsed for 3 min. The effluent pH from the air flotation was subsequently adjusted to 10 and placed in the micro-nano generator. The catalytic ozonation then ran for 1.5 h with a US-AC catalyst dosage of 3:1. The effluent pH was then adjusted to 6.8–7.5, and this flowed into the A/O/O biochemical reactor and was kept there for 5 days. The effluent from the A/O/O was finally entered into the reverse osmosis membrane, from where the clear liquid could be discharged, and the concentrated water was then returned to the micro-nanometer catalytic ozonation reactor. In all experiments, the A/A/O test was not repeated, n = 1, and the other tests were repeated three times, n = 3.

The range method was used to analyze the orthogonal results (Xuan & Jia 2012). Several factors are responsible for the removal efficiency (Table S3). The COD removal rate of influencing factors are in the order PFS > pH > PAM. The optimal conditions are where PFS is at 2,600 mg/L, at a pH of 9, and a PAM of 6 mg/L. Under these conditions, the COD removal efficiency reached a maximum of 26.09% (with a COD of 2,083.34 mg/L). With an increase in the PFS dosage, the removal efficiency of the COD increased gradually, and with a rise in pH, the COD removal efficiency first increased and then decreased. As the PFS dosing increases, the charge neutralization and adsorption bridging of the flocculants were enhanced and the flocculation removal efficiency of COD was increased (Lei et al. 2018).

The effect of the catalyst dosage was also investigated and can be seen in Figure 2(b). The removal efficiency of the COD in the coking wastewater increased continuously with an increase in the dosage of the ozone catalyst. The removal efficiency of the COD reached a maximum of 69.84 ± 2.68% when the mass ratio of liquid: solid was 3:1. The effective activity position of the catalyst surface increases as the catalyst dosage increases during constant ozone dosage, which is beneficial to the production of high concentrations of hydroxyl radicals from ozone (Chen et al. 2019). Subsequent tests showed that the test catalyst was at a liquid:solid ratio of 3:1.

The BOD/COD of the initial coking wastewater was 0.045, which increased to 0.065 after coagulation and air flotation. These values are far from reaching the biochemical requirements (BOD/COD>0.3). The effluent BOD/COD of different catalytic ozonation times was determined (see Figure 3(b)). When the catalytic oxidation time was prolonged, the BOD/COD first increased from 0.065 to a maximum value of 0.312 at 1.5 h and then decreased, which greatly improved the biodegradability of the wastewater. The improved biodegradability can be attributed to the ozonation that decomposed the refractory organic pollutants (such as benzene rings and aromatics) into small, low-toxicity biodegradable molecular compounds (Amaral-Silva et al. 2016). The biodegradability of the wastewater decreased after 1.5 h, which is possibly because the organic matter is degraded almost without selectivity during the catalytic oxidation of ozone. With increasing catalytic oxidation, some organic substances that biodegrade easily will be oxidized, resulting in a decrease in the overall biodegradability of the wastewater over time. The effluent that was produced after 1.5 h of catalytic ozonation (COD:621.84 mg/L) was therefore selected to be used in the A/A/O reactor.

Table S5 and Figure S3 show the diversity indices (Shannon, Simpson, ACE, Chao indices) and the sparsity curves for nine samples (Q1–3, Y1–3, H1–3). The ACE and Chao indices for microorganisms in the low mud of the anaerobic pond of Y3 were higher than the index values of the other samples, meaning that the microbial community was the most abundant in the anaerobic pond substrate. The microbial community in the H3 aerobic pond substrate had higher Shannon values and lower Simpson values than the other samples. This means that the microbial community diversity was highest in the aerobic pond substrate. This may be because the microorganisms in the aerobic group were exposed to the least amount of toxic substances in the wastewater which increased their survival. Based on the coverage and sparsity curves, the sequencing depth is adequate (Wang et al. 2012a, 2019; Fang et al. 2018).

The principal component analysis at the operational taxonomic units (OTU) level (Figure S4) reveals the microbial community affinities of the nine sludge samples in the A/A/O reactor. New dominant microorganisms have formed in each reaction unit, as indicated by the wide variation of the microbial communities in each reaction unit, and the fact that the microbial communities in different areas of the same reaction unit were similar in structure. The Y1 sample had the smallest point and the lowest diversity. This was consistent with the diversity index analysis.

Figure 5(b) shows the relative abundance of the main microbial groups at the genus level (top 10) for each sample. In Q1, Q2, and Q3, the dominant genera were Thioalkalimicrobium (43.57%), Proteiniphilum (47.97%), and Azoarcu (31.08%). The dominant genus was Azoarcus at 43.65% and 47.69% abundances, respectively, in both Y1 and Y3, and the abundance of Fontibacter was 37.94% in Y2. The dominant genera were Bacillus (16.67%), Fontibacter (26.65%), and Taibaiella (20.20%) in H1, H2, and H3. Azoarcus was ubiquitous in every sample. Azoarcus could denitrify the wastewater and also biodegrade the aromatic hydrocarbons, which was essential in the denitrification and organic degradation of the wastewater (Ma et al. 2015). Nitrincola belongs to Nitrobacter, and mainly occurred in the aerobic reaction units H1, H2, and H3 where it was the main species responsible for the nitrification of the aerobic unit. Bacillus was mainly suspended in the water of the H1 aerobic tank and was mainly responsible for the degradation of thiocyanate and organic matter (Chaudhari & Kodam 2010). Fontibacter, which mainly existed in Y2, could reduce and hydrolyze humic acids under anaerobic conditions. Fontibacter was responsible for the hydrolytic reduction of some organic macromolecules in the anaerobic unit (Ma et al. 2014).

This study shows that the coagulation air flotation-micro-nanometer catalytic ozonation-A/A/O-reverse osmosis membrane integrated system can effectively remove COD in the coking wastewater and meet the new national standard (China) direct discharge standard. When this integrated system operates optimally, it can remove 98.17% of the COD. Micro-nanometer catalytic ozonation can effectively improve the biodegradability of the coking wastewater, and increase the BOD/COD of wastewater from 0.045 to 0.312. This study lays a solid foundation for the treatment of high-concentration and recalcitrant organic wastewater. Even though the effectiveness of this combined treatment system was only demonstrated at the laboratory scale, it provides information for the further development of highly efficient wastewater treatment procedures to remove COD. Further work on developing a fully operational long-term COD treatment system is underway.

This work was supported by National Natural Science Foundation of China (Grant No. 51871250), the Yunnan Major Program of Science and Technology (Grant No. 202102AB080007), the fund of the State Key Laboratory of Advanced Technologies for Comprehensive Utilization of Platinum Metals (Grant No. SKL-SPM-202001), the research and development project of a new technology for the treatment of high-concentration organic wastewater by Chinese enterprises (Grant No. 738010280) and the Fundamental Research Funds for the Central Universities of Central South University (Grant No. 506021729).

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

The authors declare there is no conflict.