Hybrid System of Fenton Process and Sequencing Batch Reactor for Coking Wastewater Treatment

Abstract

The aim of the work was to investigate the treatment efficiency of coking wastewater in a hybrid system combining the Fenton process with an SBR reactor. The Fenton reaction was optimised using variable reagent doses of 0.75, 1.0, 1.25 and 1.5 g/L for iron ions and 750, 1000, 1250, and 1500 mg/L for H₂O₂. The effects of Fe²⁺ and H₂O₂ concentration on BOD, COD, TOC, TN N-NH₄⁺ and BOD/COD ratio were studied in detail to optimise the pretreatment performance. The selection of the most favourable parameters for the Fenton reaction was based on the frequency of occurrence of a different combination of the chemical reagents. The most beneficial doses were found to be 0.75 g/L of iron (II) ion and 1000 mg/L of hydrogen peroxide, at which the COD reduction rate was about 40% and a high increase in the BOD₅/COD ratio from 0.1 to 0.31 was observed. Moreover, the obtained results showed that the efficiency of removing organic pollutants and nitrogen compounds was higher in the SBR reactor fed with pretreated wastewater. However, the relatively low efficiency of removing TKN (25%) and NH₄⁺ (21%) indicates the presence of toxic substances in them that may inhibit the removal of nitrogen compounds.

1. Introduction

The coking industry is a strategic branch of the European and global economy, as metallurgical coke is an important fuel and raw material for steel production. The production of coke and the processing of coke by-products utilise numerous water streams, resulting in the generation of various types of wastewater streams. Some of these are related to the coking processes themselves (e.g., coke quenching), while others to the purification of the coke oven gas. The wastewater streams generated in coking plants include those from wet coke quenching, benzene condensation and rectification, tar processing, gas cooling, the processing of hydrogen sulphide to sulphuric acid and the steam condensates used to recover liquid coke products. The amount of wastewater produced depends on the technological process, the production volume and the internal water recirculation. It is estimated that 0.15–0.35 m³ of technological wastewater is produced per 1 Mg of processed coal, corresponding to a wastewater volume of 0.35 to 0.45 m³ per 1 Mg of coke. Coking wastewater belongs to the most polluted and environmentally hazardous industrial wastewater [1,2]. It contains significant amounts of ammonium salts and chemical compounds such as phenols, oils, tars, suspensions, polycyclic aromatic hydrocarbons (PAHs), toxic organic nitrogen compounds, cyanides, ammonia and hydrogen sulphide as well as the following anions: chlorides, sulphates, sulphides, thiosulphates, thiocyanate [3,4,5,6]. The high concentration of organic and inorganic pollutants and the low biodegradability make this wastewater a danger to the external environment. Since these substances are carcinogenic and can cause reproductive toxicity, genotoxicity, immunotoxicity and respiratory problems, the discharge of untreated wastewater into the natural environment can negatively affect the quality of surface and groundwater, aquatic organisms and indirectly the food chain [7]. Therefore, it cannot be discharged directly into receiving waters without prior treatment [8,9,10]. Coking wastewater is usually treated first onsite. It is channelled into purification installations located in the coking plant [1]. Biological processing is the most commonly used method for wastewater treatment in such installations [11]. However, the efficiency of these systems is often limited, so it is recommended to implement integrated systems that combine different process units (biological, chemical and physical process units) [1,3,12], e.g., the combination of biological processes with coagulation and adsorption [13]. The most commonly employed wastewater treatment systems include sand filtration, chemical neutralisation and the activated sludge process [14]. Among activated sludge processes, prevalently used are A/O (anoxic/oxic), A2/O (anaerobic/anoxic/oxic) and SBR (sequencing batch reactor) [15]. A number of new biological methods for the treatment of coking wastewater are also being tested, e.g., denitrification on biocathodes [16], bioagumentation [17] or the use of microalgae [18,19]. Conventional biological treatment processes are considered easy to operate, simple to maintain and relatively cheap. However, coking wastewater is inherently complex and contains many toxic and recalcitrant substances, which can greatly inhibit the activity of microorganisms and negatively affect treatment efficiency [20]. As a result, there is increasing interest in technologies for the pre-treatment of coking wastewater before treatment in biological reactors. These processes aim to degrade toxic and harmful substances and improve the biodegradability of the wastewater, which is crucial for the efficiency of the biological process [21]. Membrane processes, stripping, ozone oxidation, air flotation, coagulation, and a large group of advanced oxidation processes (AOP) are potential methods for the pretreatment of coking plant wastewater [22]. Among them, AOPs deserve special attention due to their high efficiency in removing non-biodegradable and emerging pollutants resulting from the formation of reactive oxygen species (ROSs). Numerous literature reports confirm the effectiveness of various AOP methods in the treatment of coking wastewater, e.g., Fenton and Fenton-based processes [23,24,25,26,27]: photocatalysis [28,29]; catalytic ozonation [30]; wet air oxidation [31]; and electrochemical oxidation [32]. One of the most representative AOP methods for removing persistent and toxic contaminants from coke wastewater is the Fenton process, which takes place in the presence of H₂O₂ and Fe²⁺ ions as process catalysts. The reaction mechanism leads to the catalytic decomposition of hydrogen peroxide in the presence of iron ions, resulting in reactive hydroxyl radicals OH• with a very high oxidation potential [33]. The process involves a series of interconnected reactions. In the initial step, Fe²⁺ ions react with H₂O₂ to generate hydroxyl radicals, ferric ions (Fe³⁺), and hydroxide ions (OH⁻). However, excess H₂O₂ can act as a radical scavenger, and Fe³⁺ can precipitate as Fe(OH)₃ outside the optimal pH range of 2.5–3.0, reducing efficiency. Additionally, H₂O₂ can react with hydroxyl radicals to produce less reactive hydroperoxyl radicals (HO₂^•), which may reduce the overall efficiency of the process. In turn, ferric ions (Fe³⁺) are regenerated into ferrous ions (Fe²⁺) through reactions with hydroperoxyl radicals, allowing the reaction cycle to continue. Furthermore, excess H₂O₂ can decompose into water and oxygen, while excess hydroxyl radicals may recombine to form H₂O₂ again. Moreover, hydroxyl radicals also react with organic pollutants (R), oxidizing them and generating intermediate organic radicals (

Table 1. The main chemical reactions of the Fenton process based on [34].

Reaction		Reaction Rate Constant, L/(mol × s)
Fenton system	Fe2+ + H2O2 → Fe3+ + HO• + OH−	53–76
	Fe2+ + HO• → Fe 3+ + OH−	2.6–5.8 × 108
	H2O2 + HO• → HO2• + H2O	1.7–4.5 × 107
	Fe2+ + HO2• → Fe 3+ + OH2−	0.72–1.5 × 108
	Fe3+ + HO2• → Fe2+ + O2 + H+	0.33–2.1 × 106
	HO• + HO• → H2O	5–8 × 109
	HO2• + HO2• → H2O2 + O2	0.8–2.2 × 106
	HO• + HO2• → H2O + O2	1.4 × 10 10
	2 H2O2 → 2H2O + O2
	Fe3+ + H2O2 → [Fe(HO2)]2+ + H+	3.1 × 10−3
	[Fe(HO2)]2+ → Fe2+ + HO•
	Fe2+ + HO2• → Fe3+ + HO2−	0.75–1.5 × 106
	R + HO• → R• + H2O	107–1010
	R + HO• → HOR•	107–1010
	R• → RR
	R• + Fe2+ → R− + Fe3+
	R• + Fe3+ → R+ + Fe2+
	R+ + OH− → R-OH

). This process effectively breaks down persistent organic pollutants, but its efficiency relies on precise control of parameters such as reagent concentrations, pH, and the dosing of H₂O₂ [34].

The Fenton process has attracted increasing attention due to its numerous advantages, including high efficiency at room temperature and atmospheric pressure; use of readily available and easy-to-handle chemicals; simple operation, no need for specialised equipment; and short reaction time [35]. Although the Fenton process is commonly used, there is limited research on its effectiveness as a pretreatment method in combination with the biological treatment of coking wastewater. The process must be further optimised to reduce the consumption of chemicals and lower operating costs.

Although some work has already been published on the possibility of treating coke oven effluent using the Fenton reaction and the SBR reactor, their results are inconclusive. This is mainly due to the fact that the authors used wastewater from different coking processes and thus differed significantly in chemical composition. This necessitates the need to optimize process parameters each time for wastewater from a specific production plant, and this is true for both the Fenton process as a pretreatment method, as well as the biological treatment in the SBR reactor. For this reason, this paper presents the results of a study on treating real industrial wastewater from coking plants in southern Poland.

The effects of Fe²⁺ and H₂O₂ concentration were studied in detail to optimise the pre-treatment performance. Subsequently, SBR systems with and without Fenton pretreatment were comparatively analysed to verify the treatment efficiency of the Fenton/SBR hybrid process. The results obtained in this study provide technical support for the pre-treatment technology of coking wastewater, which is then treated using biological methods in SBR reactors.

2. Materials and Methods

The following parameters were analysed in the study: chemical oxygen demand (COD) (chemical method according to Hach, spectrophotometer Hach DR/4000, Loveland, CO, USA), total organic carbon (TOC) (dry combustion with the Multi N/C H3100, Analytykjena, Jena, Germany) pH value (pH meter Cole Parmer model no. 59002-00, Bunker Court, Vernon Hills, IL, USA), Kjeldahl nitrogen (TKN) (steam distillation with the BÜCHI K-355 after mineralisation of the sample by the digestion unit K-435, BÜCHI, Flawil, Switzerland), total suspended solids (TSS) (measured by oven drying at 105 °C with the SL115 drying oven, POL-EKO, Wodzislaw Slaski, Poland), volatile suspended solids (VSS) (measured by combustion at 550 °C with the MFC, CZYLOK, Jastrzębie-Zdrój, Poland), ammonium nitrogen (N-NH₄⁺) (steam distillation, BÜCHI K-355, Flawil, Switzerland), nitrate (N-NO₃), nitrite (N-NO₂) (both nitrogen forms measured using the Hach chemical method, Hach DR/4000 spectrophotometer, Loveland, CO, USA), 5-day biochemical oxygen demand (BOD₅) (respirometric method, OxiTop^® Control System, WTW, Weilheim, Germany). All analyses mentioned were carried out in accordance with the APHA Standard Methods for the Examination of Water and Wastewater [36]. Samples for determination of COD and TOC were prepared as follows: (1) filtration through a membrane filter with a pore diameter of 4.4 µm (the action allowed the removal of suspended organic compounds from the sample); (2) filtration through a membrane filter with a pore diameter of 0.45 µm (the filtrate obtained was the dissolved fraction of COD). Sample preparation for the other analyses consisted of centrifugation (11,200 rpm, 15 min, Eppendorf 5804 centrifuge) and filtration through a soft filter (Munktell 3w, Ahlstrom-Munksjö, Falun, Sweden). The concentration of dissolved oxygen (DO) was measured by a portable DO meter (Elmetron CPO-401, Gliwice, Poland).

2.1. Wastewater Composition and Sludge Inoculum

The coking wastewater originated from a coke plant located in the Silesian Voivodeship (Poland) with a concentration of 4500–5500 mg/L for COD, 250–680 mg/L BOD₅, 500–600 ammonia-nitrogen (NH₄⁺), 1200–1400 mg/L TOC, the pH value was about 8.6–9.0.

The activated sludge was collected from a bioreactor at the coke plant’s own wastewater treatment plant and then inoculated directly into a laboratory-scale SBR reactor without additional acclimatisation. The initial concentration of suspended solids in the mixed liquor (MLSS) was 5.0 g/L with a volatile suspended solids (VSS)/SS ratio of 0.75.

2.2. Pretreatment of the Coking Wastewater

The pretreatment of the coking wastewater by the Fenton method was carried out in a 1000 mL beaker with a magnetic stirrer (MST Digital, Velp Scientifica, Usmate, Italy). In the first part of the work, the Fenton process was optimized with variable doses of iron ions and H₂O₂ (

Table 2. The dosage of reagents used and the conditions for the Fenton process.

Fe2+ (g/L)	H2O2 (mg/L)
0.75	750	1000	1250	1500
1.0
1.25
1.50

Notes: Source of iron (II) ions—hydrated iron (II) sulfate (VI) (FeSO₄x 7H₂O), Source of H₂O₂—hydrogen peroxide (30% hydrogen peroxide solution).

). Each time, the pH of the solution was first lowered to 3.5 using H₂SO₄ (1 mol/L), and then the reagents (sulfate, hydrogen peroxide) were added. The fast mixing phase (660 rpm) lasted 5 min, followed by the slow mixing phase (330 rpm) for 15 min. After that, the pH was raised to 7.5 using NaOH (3 mol/L) and the sedimentation phase was carried out for 120 min. Then, the decanted liquid was used for further tests. During the described conditions, the sample was not exposed to light.

All chemical reagents used in the analyses were characterized by high purity and were purchased at Sigma-Aldrich (Merck KGaA, Darmstadt, Germany).

2.3. SBR Set up and Operation

The laboratory model of the SBR reactor was a glass cylinder (27.5 cm high and 14.5 cm in diameter) with a working volume of 3 L. The bioreactor was stirred with a magnetic stirrer (MST Digital, Velp Scientifica) and aerated with an air compressor which supplied an air flow of 100 L/h through a fine bubble diffuser located at the bottom. A peristaltic pump (MU-01, Major Science, Taoyuan City, Taiwan) was used to fill and empty the reactor. The coking wastewater was fed into two bioreactors, called SBR and FEN/SBR, respectively, to domesticate the activated sludge. The bioreactors were previously inoculated with sludge from the industrial wastewater treatment facility of the coke plant. The concentration of MLSS in both reactors was 4.8 g/L. The pH value of the wastewater was adjusted to 7 with H₂SO₄. During the acclimatisation phase, the 24-h operating cycle of the SBR reactor was as follows: 2 h wastewater feed, 18 h aeration, 2 h mixing, 1.5 h sedimentation and 15 min discharge, with a wastewater exchange ratio of 20%. The operating cycle was established based on the results of research conducted at the Czestochowa University of Technology, which concerned the treatment of coking wastewater in an SBR reactor after prior pre-treatment with ultrasound [37]. The hydraulic retention time (HRT) was 2 days; the average wastewater flow was 1.5 L/d, while 1.22 L of treated wastewater and 0.28 L of sludge were discharged per day. DO content in the aeration phase ranged from 2 to 4 mg/L while in the denitrification phase, it was maintained at 0.2 to 2 mg/L. Figure 1 presents a scheme of the experiment.

2.4. Statistical Analysis

The results obtained were subjected to statistical analysis using the following statistical methods: (a) one-way analysis of variance (p < 0.05) or if its assumptions were not met, the Kruskal-Wallis ANOVA rank test; (b) multifactor analysis of variance (p < 0.05). In addition, for data in which the analysis of variance revealed significant differences between the variables, further analyses were carried out using multiple comparison tests, so-called post hoc tests, and in particular Tukey’s HSD test (Tukey’s honest significant difference test). The homogeneity of variance was checked using the Levene test. All analyses were performed in Statistica 7.0. The scheme of the statistical analyses carried out as part of the study is shown in Figure 2.

3. Results and Discussion

The aim of the research conducted was to determine the efficiency of wastewater treatment in a hybrid system that combines the Fenton process with biological wastewater treatment in an SBR reactor. The industrial wastewater used for the study was characterised by a high content of organic pollutants (COD in the range of 4500–5500 mg/L), a high concentration of total nitrogen (TN > 610 mg/L), which mainly occurred in the form of ammonium nitrate, and a very low biodegradability. To determine the biodegradability of wastewater in biological reactors, the BOD₅/COD ratio is most commonly used, which should assume a value of about 0.5 [38]. The very low BOD₅ to COD ratio in the tested coking wastewater (0.1) proves that it is completely resistant to degradation in biological processes.

3.1. Fenton Pretreatment of Coking Wastewater

The research to optimise the process parameters for the pretreatment of coking wastewater with the Fenton reaction was conducted in four research series using raw coking wastewater with variable contamination characteristics. For this reason, in the first phase of the work, a one-way analysis of variance was performed for all contamination indicators in all these series. The results of the ANOVA showed that there were no statistically significant differences between the indicators in the different research series (p < 0.05) Figure 2. However, due to the large number of variables in the form of reagent dosages and only four control samples, performing the factorial analysis of variance required the estimation of the difference between the value of the parameter obtained after conditioning in a given experimental series and the value of this parameter for the control sample in this experimental series, i.e., the estimation of the Δ’ defined in the methodological part. Without the mathematical operations described, statistical analysis would be impossible due to the unequal number of variables in the groups.

Figure 3 shows the changes in selected parameters (Δ’) estimated as the difference between the value of the parameter obtained after pretreatment in a given research series and the value of this parameter for the complementary control sample (without Fenton pretreatment). Of all the indicators monitored in the study, the dose of hydrogen peroxide had the greatest effect only on the changes in Δ’BOD₅. The changes in Δ’TOC, Δ’TN and Δ’N-NH₄⁺ were most influenced by the iron ion dose, to a lesser extent by the hydrogen peroxide dose and least by the interaction between the chemical reagents used in the study. The influence of the above factors on Δ’pH, Δ’COD, Δ’BOD₅, Δ’BOD₅/COD was as follows: dose of Fe²⁺ > interaction between the doses of the reagents > dose of H₂O₂.

After the Fenton process, a drop in the pH value of the conditioned wastewater was observed. With the exception of the iron ions at a dose of 1.25 g/L, the pH changes were small, as they were between 1.64 and 1.33 g/L regardless of the dose of hydrogen peroxide. At the above-mentioned dose of Fe²⁺, the pH changes were between 2.12 and 1.51. A significant reduction in COD was observed in the wastewater pre-treated with Fenton. Compared to the control sample, a reduction of 24 to 39% was observed. A similar trend was observed for TOC, TN and ammonium nitrogen, for which a reduction of 28–48%, 35–48%, 73–86% and 16–21.5%, respectively, was observed compared to the control sample.

The changes in BOD₅ and the BOD₅/COD ratio were again strongly dependent on the dose of chemical reagents used. At iron ion doses of more than 0.75 g/L, after reaching a local maximum, a decrease in the parameter was observed with an increase in the hydrogen peroxide dose.

When selecting the optimum dosage of reagents to carry out the Fenton reaction for the treatment of coking wastewater in a Fenton process-SBR reactor hybrid system, the highest possible reduction in COD, TOC, TN, N-NH₄⁺ and an increase in the BOD₅ and BOD₅/COD ratio were taken into account. Table 2 shows the most advantageous reagent dosages for the Fenton reaction based on the assumptions described above and the results of the Tukey test carried out with the Statistica programme.

With the above adopted pretreatment process parameters, the COD reduction rate was approximately 40% and the BOD/COD ratio increased to 0.31. In the studies of Chu et al. [23], a higher COD reduction of 44 to 50% was achieved in Fenton-pretreated coking wastewater, which can be explained by a significantly higher dose of hydrogen peroxide and a longer reaction time (H₂O₂ dose of 0.3 M and reaction time of 2 h).

A high COD removal rate after 30 min of reaction time (>70%) was also obtained by Kim et al. [39], who conducted the Fenton process for coking wastewater with an iron ion dose of 0.75 g/L and a hydrogen peroxide dose of 10.5 g/L. However, the dose of hydrogen peroxide in the cited article was 10 times higher than in the studies presented in this article. This confirms the fact that the Fe²⁺/H₂O₂ ratio is one of the key parameters for the efficiency of the Fenton reaction. Iron, which serves as a catalyst in the Fenton reaction, is necessary for the generation of hydroxyl radicals. At low Fe²⁺ concentrations, hydrogen peroxide is more difficult to decompose and produces fewer free radicals. In contrast, when the iron ion concentration is high, H₂O₂ decomposition is enhanced and free radicals are generated, leading to a scavenging effect [40]. Therefore, to avoid the scavenging effect, it is necessary to optimise the catalyst dosage to increase the reaction efficiency, resulting in lower wastewater treatment costs [41].

The selection of the most favourable parameters for Fenton pretreatment was based on the frequency of occurrence of a particular combination of iron ions and hydrogen peroxide doses (

Table 3. Selection of the most favourable parameters for performing the Fenton reaction based on the results of the statistical analysis performed for the second stage of the study (the table summarises the most favourable results displayed by the Statistica programme after performing the Tukey test). The table summarises the changes in the values of the indicators compared to the control (unconditioned) sample. In this study, pretreatment was carried out for 16 combinations (4 iron and 4 doses of H₂O₂). For each of the parameters analysed, the 6 most favourable changes from the point of view of the biological process were selected. For example, in the case of the COD, TOC and nitrogen indices, those with the most significant reduction in indicators were selected, and in the case of the BOD/COD ratio, those with the greatest increase. The number of occurrences of a given pretreatment condition in this narrowed list was then counted. In order to systematise and facilitate the comparison, different colours were assigned to the iron doses (e.g. green for an iron dose of 0.75 g/L, orange for an iron dose of 1.0 g/L, blue for 1.25 g/L and grey for an iron dose of 1.5 g/L), and a colour gradient was also intro-duced related to the H₂O₂ dose (as its concentration increases, the intensity of the colour increases). We hope that this systematisation will facilitate understanding and interpretation of the table.

Δ’BOD			Δ’COD			Δ’BOD5/COD
Sample	time	results	Sample	time	results	Sample	time	results
Fe1.0	P1500	301	Fe1.0	P1250	−2335	Fe1.25	P1250	0.16
Fe0.75	P750	347	Fe1.0	P1500	−2286	Fe0.75	P750	0.17
Fe1.0	P1000	372	Fe1.25	P1250	−2104	Fe1.0	P1500	0.19
Fe1.0	P1250	394	Fe0.75	P750	−2067	Fe0.75	P1000	0.22
Fe0.75	P1000	454	Fe0.75	P1000	−2032	Fe1.0	P1250	0.23
Δ’TOC			Δ’pH
Sample	time	results	Sample	time	results
Fe1.25	P1250	−0.71	Fe1.0	P750	−1.49
Fe1.0	P1250	−0.64	Fe1.5	P1000	−1.45
Fe1.25	P1000	−0.61	Fe1.5	P1250	−1.44
Fe1.0	P1000	−0.56	Fe0.75	P1500	−1.42
Fe1.25	P1500	−0.56	Fe1.0	P1000	−1.33
Δ’TN			Δ’N-NH4+
Sample	time	results	Sample	time	results
Fe0.75	P1000	−155.2	Fe0.75	P1000	−162.6
Fe0.75	P750	−149.7	Fe0.75	P750	−152.7
Fe0.75	P1250	−139.4	Fe0.75	P1250	−146.8
Fe1.0	P1000	−139.3	Fe1.0	P1000	−145.9
Fe1.0	P1250	−138.8	Fe1.0	P1250	−141.6
Frequency of Sample Appearance in the Analysed Set
Fe0.75	P750	5	Fe1.0	P750	1	Fe1.25	P1000	1
Fe0.75	P1000	5	Fe1.0	P1250	6	Fe1.25	P1250	2
Fe0.75	P1250	2	Fe1.0	P1000	5	Fe1.25	P1500	1
Fe0.75	P1500	1	Fe1.0	P1500	3	Fe1.5	P1000	1
						Fe1.5	P1250	1

). When analysing the frequency of occurrence of the samples in the sets and taking into account economic considerations, it was decided that the wastewater would be conditioned using an iron (II) dose of 0.75 g/L and a hydrogen peroxide dose of 1000 mg/L prior to biological treatment. With the above pre-treatment process parameters, the COD reduction rate was approximately 40%. In addition, the factor that determined the selection of the Fenton reaction conditions was the highest increase in the BOD₅/COD ratio.

3.2. Effectiveness of the Fenton-SBR Hybrid System

After Fenton pretreatment, the performance of biodegradation of coking wastewater in the SBR was evaluated. Two SBR reactors were operated simultaneously for 51 days. The FEN-SBR reactor was fed with Fenton pretreated wastewater, while the second (control) reactor treated raw wastewater without pretreatment. During the treatment process in the reactor, no significant differences in the pH value of the wastewater were observed (Figure 4). The differences on the first day of the process were due to the fact that the pH of the Fenton pretreated wastewater was increased from 3 to 7.5 to inhibit the Fenton reaction. The following figures show the concentration of organic compounds in the treated wastewater and the efficiency of their removal during the treatment process, expressed by COD (Figure 5), BOD₅ (Figure 6), the BOD₅/COD ratio (Figure 7) and TOC (Figure 8).

During biological treatment in the SBR, the COD concentration of the Fenton pre-treated and untreated coking wastewater fell from 2631 and 4593 mg/Lin the influent to 550 and 1300 mg/L in the effluent and remained relatively constant from 40 d onwards. This corresponds to a COD removal of approximately 79 and 71%, respectively (Figure 5). A similar trend was observed in the efficiency of TOC removal, which reached almost 91 in the case of the FEN-SBR (Figure 8).

The initial BOD concentration and BOD₅/COD ratio of the Fenton- pretreated coking wastewater of 870 mg/L and 0.33, respectively, were not only higher than those of the untreated wastewater (502 mg/L and 0.11) but also the removal efficiency of BOD₅ of the Fenton-pretreated coking wastewater was higher. The result shows that the biodegradability of the coking wastewater in the SBR was improved after Fenton oxidation pretreatment. The BOD₅ removal efficiency reached almost 99% and the BOD₅/COD ratio decreased to 0.02, which means that all biodegradable compounds were removed. This can be explained by the fact that the Fenton reaction is considered an effective method for the removal of toxic substances from wastewater and improves its biodegradability expressed by the BOD₅/COD ratio [42].

Despite the relatively high efficiency in the removal of organic pollutants in the FEN-SBR reactor, several compounds resistant to biodegradation remained in the treated wastewater. In addition, toxic substances that inhibited the biological oxidation process probably remained in the wastewater even after pretreatment with the Fenton process as well.

This hypothesis can be confirmed by monitoring the removal of nitrogen compounds in SBR and FEN-SBR reactors. The TN removal efficiency in the FEN-SBR reactor was higher than in the reactor fed with unconditioned wastewater (Figure 9). From about day 21, TN concentrations in wastewater treated in the SBR and FEN-SBR reactors stabilised in the range of 557–570 mg/L and 460–492 mg/L, with final removal rates of 21 and 25%, respectively, demonstrating the stable treatment efficiency of the reactors, but at the same time their low TN removal efficiency. In the influents, nitrogen compounds were mainly present in the form of ammonium nitrogen and accounted for 81% and 80% of the total nitrogen in the SBR and FEN-SBR reactors, respectively. The NH₄⁺ concentrations in the treated wastewater in the SBR and FEN-SBR reactors were 450–464 mg/L and 459–492 mg/L (Figure 10), with an effective removal efficiency of about 18% and 21%, respectively. At the same time, the nitrate concentration in the effluent of both reactors was 1–2 mg/L, while N-NO₂ was 81–122 mg/L for SBR and 70–75 mg/L for FEN/SBR. This observation shows that the removal of nitrogen compounds in the proposed hybrid system is not effective. The high concentration of N-NH₄⁺ and N-NO₂ in the wastewater indicates that toxic substances, which are denitrification inhibitors, are still present in the wastewater flowing into the FEN-SBR reactor despite the use of the Fenton reaction in the pretreatment process. A similar phenomenon was observed by Wang, et al. [26]. The authors demonstrated that the remaining organic matter after the aeration phase could not be optimised by the heterotrophic denitrifying bacteria. The authors suggested that in addition to the typical refractory impurities of coking wastewater, the presence of toxic oxidation intermediates after Fenton pretreatment is also possible.

4. Conclusions

The coking wastewater was treated using a combination of Fenton oxidation and SBR processes. In order to optimize the Fenton reaction variable reagent doses of Fe²⁺ and H₂O₂ were tested. These doses were 0.75, 1.0, 1.25 and 1.5 g/L for iron ions, and 750, 1000, 1250 and 1500 mg/L for H₂O₂, respectively. The effects of Fe²⁺ and H₂O₂ concentration on BOD, COD, TOC, TN N-NH₄⁺ and BOD/COD ratio were studied in detail to optimize the pre-treatment performance. The most beneficial doses were found to be 0.75 g/L of iron (II) ion and 1000 mg/L of hydrogen peroxide, at which the COD reduction rate was about 40% and a high increase in the BOD₅/COD ratio from 0.1 to 0.31 was observed. The increase in the BOD₅/COD ratio confirmed the positive effect of the Fenton reaction on improving the biodegradability of the tested wastewater. Moreover, the obtained results showed that the efficiency of removing organic pollutants and nitrogen compounds was higher in the SBR reactor fed with pretreated wastewater. However, no satisfactory results were obtained for the removal of nitrogen compounds, which were respectively 25% for TKN and 21% for NH₄⁺. High concentrations of ammonium nitrogen and nitrates in the treated wastewater indicate the presence of nitrification and denitrification inhibitors in the wastewater. Further investigations are necessary to optimize the process of wastewater pretreatment with the Fenton reaction and to confirm the suitability of the proposed hybrid method on a large scale.