The treatment of phenolic contaminants from shale gas drilling wastewater: a comparison with UV-Fenton and modified UV-Fenton processes at neutral pH

Abstract

In a shale gas field, wastewater from drilling processes contains many recalcitrant and toxic pollutants. These pollutants are difficult to dispose in a classical wastewater treatment system. In this study, a comparative experiment was conducted to check the treatment efficiency of a UV-Fenton process at pH = 3 and a modified UV-Fenton process at neutral pH with real drilling wastewater, which was collected by a coagulative effluent in a sewage treatment plant in a shale gas field, Chongqing. The consumption of Fenton reagents and the rate of removal of COD and BOD₅ during three processes were evaluated in order to obtain an optimal process. In addition, phenolic contaminants in the effluent were measured by GC-MS and 7 of them were singled out to serve as an evaluation standard for effective treatment, which has not been reported in previous experiments. The results showed that three UV-Fenton processes had the ability to treat drilling wastewater and a modified UV-Fenton process with carboxymethyl-β-cyclodextrin (CMCD) had positive effects on the treatment of phenolic contaminants at neutral pH with minimal reagent consumption (Fe²⁺ = 14 mg L⁻¹ and H₂O₂ = 768 mg L⁻¹). It was demonstrated that modified UV-Fenton with CMCD could extend the pH range in the treatment process and avoid negative effects in UV-Fenton process at acidic pH.

---

1. Introduction

With a rapid development of the society and economy in China, shale mining has become a new source of energy.¹ According to a report by the National Energy Administration, China plans to build the first domestic ten billion shale gas field in 2017.^2,3 However, some environmental problems are produced with the development of shale gas. During the drilling operation, a lot of wastewater is generated as a main source of pollution in the exploitation of shale gas.^4,5 Drilling wastewater is mainly produced from the loss of drilling mud in drilling process, the leakage of drilling mud circulatory system and the clean of ground equipment with dark color, high chemical oxygen demand (COD) level and low biodegradability.

Some feasible treatment systems have been employed to dispose the drilling wastewater, such as a coagulation method.⁶ However, the over standard of COD and chromaticity is still retained after coagulation processes in the effluent. Meanwhile phenolic contaminants as good antioxidants and sanitizers always remained in the effluent of the coagulation operation. The removal of phenolic contaminants by a traditional wastewater treatment system is considered as an incomplete method and it has been observed from the laboratory research that some residual soluble phenolic pollutants were barely removed. Moreover, after coagulation, biological processes have been commonly used to further oxidize pollutants to reduce costs, but the residual phenolic pollutants would serve as a negative component in the biodegradability of wastewater.⁷

During the coagulation process, the hydrophobicity of the phenolic pollutants promotes their adsorption on solid particles, and these pollutants are transported to activated sludge, which will transforms into a new source of pollution. Particularly, bisphenol A is a popular topic in water treatment and is detrimental at low concentrations,⁸ which is limited under 0.1 mg L⁻¹ according to petroleum industry effluent standard (GB31571-2015). Thus, an effective method is imminently needed for the treatment of drilling sewage to overcome these shortcomings, and to meet the demands of drilling wastewater effluent standards.

Among the current advanced treatment technologies for the degradation of phenolic pollutants, advance oxidation processes (AOPs) are an option for producing strong nonspecific oxidant hydroxyl radicals, which can oxidize most organic compounds. Among them, the photo-Fenton process is one of the AOPs that has attracted a lot of interest in the scientific community for lower process costs by utilizing the power of sunlight.^9–11 It can generate HO· when H₂O₂ reacts with iron, by the Fenton chain-reaction according to eqn (1) and (2).¹² Unlike the conventional Fenton process, the UV-Fenton process has some photocatalytic steps where Fe³⁺ converted to Fe²⁺ (eqn (3)) greatly increases the utilization of iron, and H₂O₂ decomposes to HO· (eqn (4)) under UV irradiation.¹³ Nevertheless, to initiate the photo-Fenton process, the solution must be penetrated by UV light and any suspended matter is found to naturally affect the removal efficiency. As a result, these treatments commonly need a lower pH to avoid the formation of Fe³⁺ oxyhydroxides at neutral pH. However, a strongly acidic medium, necessary neutralization and a high salt content would drastically limit their application.

Fe²⁺ + H₂O₂ → Fe³⁺ + HO⁻ + HO· (k = 63 M⁻¹ s⁻¹) (1)

Fe³⁺ + H₂O₂ → Fe²⁺ + H⁺ + HO₂˙ (k = 0.01 M⁻¹ s⁻¹) (2)

Fe³⁺ + H₂O + hv → Fe²⁺ + H⁺ + HO· (k = 3.33 × 10⁻⁶ M⁻¹ s⁻¹) (3)

H₂O₂ + hv → 2OH· (k = 4.13 × 10⁻⁵ s⁻¹) (4)

In this study, two different UV-Fenton processes were investigated to single out a more suitable one to degrade phenolic pollutants from drilling sewage that was treated by a coagulation operation. A classical UV-Fenton process was conducted at pH 3 and a modified UV-Fenton process was performed at neutral pH. For the modified UV-Fenton process, the addition of complexing agents which were able to form photoactive species were considered as an effective way to solve the problem of the formation of inert iron precipitates at neutral pH. The most commonly used complexing agent is ethylenediaminetetraacetic acid (EDTA) which can form stable soluble complexes with iron in a wide pH range.¹⁴ However, as an industrial synthetic chemical, EDTA could serve as a recalcitrant in biodegradation and pose secondary environmental risks because it can mobilize and enrich toxic metals.¹⁵ It needs a natural and biodegradable ligand as an alternative. Hence, this study selected two natural organic additives to modify the photo-Fenton process and compared their degradation effects on phenolic compounds at the same doses of UV energy.

Supramolecular photochemistry, as a new branch of supramolecular chemistry, studies photo induction in host–guest systems.¹⁶ Cyclodextrins (CDs) are supramolecular hosts that have been widely certified and can serve as a medium in photochemical reactions. They are cyclic oligosaccharides consisting of 6–12 glucopyranose units and have been extensively employed in environmental remediation.¹⁷ CDs and their derivatives have been reported as an enhancing reagent in the Fenton process because of their lower toxicity. The formation of a ternary complex (pollutant-CD-iron) could enhance the rate of removal of pollutants through the direct attack by OH·. It has been reported that CDs and their derivatives could enhance the Fenton reaction to degrade polycyclic aromatic hydrocarbons.¹⁸ Moreover, there have been a number of studies that have attempted to use CDs to enhance the photodegradation of phenolic pollutants in water,^19,20 but there are few studies that used CDs in the photo-Fenton system. In this study, carboxymethyl β-cyclodextrin (CMCD), which shows a great ability to chelate metals, was selected as a complexing agent to assist the UV-Fenton process to degrade phenolic pollutants in drilling sewage at neutral pH.

Additionally, humic acids (HA) are ubiquitous in natural water and play an important role in the photochemical processes occurring in surface water.²¹ HA can produce some reactive species that assist in the Fenton process as photosensitizers or photoinitiators in the degradation of organic pollutants.²² Thus, a HA/UV-Fenton system was selected as another modified UV-Fenton process.

Previous experiments have studied these processes in the degradation of phenolic pollutants under UV irradiation, but the majority of the studies were conducted on a laboratory scale.^21,23,24 There is also little data on the in-depth comparison of their efficiency in real petroleum wastewater effluents. In this study, three treatment processes were evaluated and monitored for 50 different residual phenols and the results were examined using SPE/GC-MS. The Fenton reagent's consumption by the three processes was evaluated in order to determine an optimal process.

2. Materials and methods

2.1 Materials

UV-Fenton experiments were performed with analytical grade hydrogen peroxide (H₂O₂, 30% w/v), ferrous sulfate (FeSO₄·7H₂O, 99%), sodium hydroxide, and sulfuric acid. They all were purchased from KeLong (China) without any further purification. Carboxymethyl-β-cyclodextrin (CMCD, 98%) was purchased from JiangLai, Shanghai. Humic acid (FA > 90%) was purchased from Mackilin, Shanghai (CAS: 1415-93-6). All reagents used for chromatographic analyses from KeLong (China), including phenol, p-cresol, m-cresol, 4-ethyphenol, bisphenol A, 1-naphthol, and 2-naphthol were analytical grade; and acetonitrile and isopropyl alcohol were chromatographic grade.

2.2 Drilling wastewater

Wastewater was used from the wastewater treatment plant located in the Chongqing shale gas field where drilling wastewater was disposed by an independent process flow. The effluent of the coagulation section was collected every two hours a day to serve as experimental samples. The collected wastewater samples were separately stored in a circulated tank. The average range of the initial COD was 450–569 mg L⁻¹, BOD₅ was 20–25 mg L⁻¹ and the ratio of BOD₅ to COD (B/C) was lower than 0.5 (approximately 0.045). Other physicochemical composition parameters of drilling wastewater are shown in Table 1, where the measurement methods were all based on standard Methods for Water and Wastewater Analysis from the Ministry of the Environmental Protection Agency of China. The initial pH was 7.6–8.8 which was adjusted by H₂SO₄ (0.1 M) and set to neutral pH or a final pH of 3. The content of phenolic contaminants in drilling wastewater varied with the drilling process, but the main components of wastewater did not change. The results from the analysis of effluents showed that there were 56 phenolic pollutants and only 23 of them were detected in 12 samples as shown in Table 2. 7 indices of phenolic pollutants were beyond the effluent standard (marked with an asterisk in Table 2) and the change in contaminant concentrations, as one of evaluation criteria, was monitored during a comparative study.

Table 1 Physicochemical composition of a drilling wastewater sample

Property parameter	Original value
Density (g L^−1)	1.08
Viscosity (Pa s)	1.092
Ca^2+ (mg L^−1)	582 ± 25
Na^2+ (mg L^−1)	2230 ± 107
Cl^− (mg L^−1)	2530 ± 236
Br^−(mg L^−1)	1385 ± 183
SO4^2−	1132 ± 153
HCO3^−	431 ± 25
CO3^2−	11.2 ± 4.4

---

Table 2 Average, minimum and maximum concentrations of detected phenols

Contaminant	Max (μg L^−1)	Min (μg L^−1)	Average (μg L^−1)
Phenol*	20 120	4671	14 123.4
o-Cresol	872	312	523.2
p-Cresol*	6579	1387	4939
m-Cresol*	8380	1080	7621.4
1-Hydroxy-2,4-dimethylbenzene	680	359	550.2
2-Ethylphenol	55	23	33.3
1-Hydroxy-2,6-dimethylbenzene	134	92	123.1
1-Hydroxy-2,6-dimethylbenzene	123	21	45.6
1-Hydroxy-2,3-dimethylbenzene	558	349	436.2
3-Ethylphenol	192	22	76.5
4-Ethylphenol*	3923	332	2314.4
4-Isopropylphenol	231	44	123
3,5-Dimethylphenol	157	50	113.0
4,4′-Dihydroxybenzophenone	27	6.75	16.4
Bisphenol A*	652	212	523
2-Phenylphenol	421	123	231.3
4,4′-Dihydroxy diphenylmethane	46	21	33
1-Naphthol*	190	23	265.2
2-Naphthol*	2098	88	1865.2
5-Amino-1-naphthol	156	11	89
8-Amino-2-naphthol	125	32	68
2-Amino-1-naphthol	52	12	32
6-Amino-1-naphthol	331	143	251
Sum concentration mg L^−1	46.10	9.41	34.40

---

2.3 Preparation of reagents

All reagents were prepared freshly before use in the UV-Fenton process. FeSO₄ was prepared as a solution. The required mass of the iron salt was dissolved in deionized water and the pH was adjusted to 2.0–3.0 using H₂SO₄ (0.1 M). A certain mass of CMCD and HA was dissolved in deionized water. Hydrogen peroxide (30% w/v) was diluted with deionized water and its concentration was determined by titration with 0.1 M Na₂S₂O₃.

2.4 Experimental setup

Four jacked stirred tank reactors (12 cm × 8 cm, 0.5 L volume) which could control the water temperature were used and the temperature was kept at 25 °C. UV radiation was provided using a low-pressure mercury lamp (30 W, λ = 254 nm, irradiation intensity = 90 μW cm⁻²) on the top of the reactors. The UV energy accumulation per unit of volume of treated water (Q_UV, kJ L⁻¹) was quantified by eqn (5) during the UV illumination. Ferrioxalate actinometry was selected to determine the irradiation area for ensuring the same light energy accumulation in each reactor. The schematic of the reaction facility system is shown in Scheme 1.

Scheme 1 Schematic of the treatment facility.

In eqn (5), UV is the average of the radiation intensity which was measured by a UV radiometer (Linshang Technology, LS126C) and the incident radiation was kept almost constant at 10 W m⁻², (t_n − t_n−1) was the exposure time for sample n, Ar was the illuminated surface area of reactors (m²) and Vt was the volume of the treated water for each sample (control in 0.5 L).

2.5 Comparison experiments

Three processes were evaluated in a dark room to rule out any environmental light interference. A volume of 300 mL of effluents were poured into the reactors. The theoretical addition of oxidizing reagent and catalyst were calculated as follows (eqn (6)), where C_H2O2 mg L⁻¹ is the H₂O₂ concentration of the solution:

C_H2O2 = 2.125COD mg L⁻¹ (6)

However, according to the previous case,²⁵ H₂O₂ and iron concentration were used at lower levels of 0.6C_H2O2 and 0.015C_H2O2, respectively. Four reactors were operated with three different processes and one control experiment (original sample) under parallel irradiation. Every process was carried out three times to guarantee high statistical significance for the results.

The pH of effluents was set at 3 ± 0.3 in the classical UV-Fenton process, which had been demonstrated as an optimal pH in each experiment. Then, H₂O₂ and Fe²⁺ were added to the reactors at the same time. Before the UV-Fenton process began, the reactors were covered for 15 min to evaluate the Fenton reaction. In the UV-Fenton modified HA process at neutral pH (6.5 ± 0.3), the pH was adjusted after mixing 10 mg L⁻¹ of HA in the effluents, then H₂O₂ and Fe²⁺ were added. The UV-Fenton with CMCD process followed the same procedure with the exception of the order of the addition of CMCD. CMCD (10 mg L⁻¹) and FeSO₄ were prepared in water and then added into the effluents with H₂O₂. After 15 min, the reactors were exposed under the UV irradiation. During the reaction, the amount of H₂O₂ was added or eliminated according to eqn (6) during every sample analysis period.

2.6 Sample preparation and analysis method

All samples were taken every 15 min. After sampling, COD was measured through the dichromate method. The details of the method were obtained from Standard Methods for Water and Wastewater Analysis from the Ministry of the Environmental Protection Agency of China. It is worth noting that before the determination of COD, the water samples were adjusted to pH 13–14 using NaOH (0.1 M) and heated to 60 °C for 1 h. This step helps in avoiding the interference of ferrous ion and residual H₂O₂. BOD₅ was carried out according to Standard Methods (5210-D test), using an OxiTop® (manometric respirometry) system. The H₂O₂ concentration was measured with a titanium(IV) oxysulphate spectrophotometric method at 410 nm. The Fe²⁺ concentration was analyzed using the o-phenanthroline standardized method according to ISO 6332. Under the same conditions, three sets of parallel analyses were conducted to reduce the measurement error to 5%. Data are displayed as the mean ± standard deviation (n = 3).

After excess H₂O₂ was eliminated by catalase, phenolic pollutants were extracted by solid phase extraction (SPE), using a Cleanert® PEP tube (3 mL, 100 mg) from Agela Technologies (China). Firstly, the columns were washed with 5 mL of water–acetonitrile (99 : 1 v/v). Next, water samples were extracted into the cartridges at 10 L min⁻¹. Finally, cartridges were washed with water–acetonitrile (99 : 1 v/v), and eluted with 6 mL of acetonitrile followed by 6 mL of isopropanol. This step was used to concentrate pollutants as well as to reduce insoluble material.

The quantitative analysis of target contaminants was performed with gas chromatography-mass spectrometry (GC-MS). The samples were filtered with a 0.2 μm syringe before they were injected into the GC-MS. GC-MS (Agilent Technologies, USA) analysis was performed with a DB-WAX column (30 m × 0.2 m × 0.25 μm) that was interfaced to an electron ionization source. The operation details were adapted from Thacker et al.²⁶ Briefly, GC operational conditions were as follows: injection volume of 1 μL, injection port temperature of 250 °C, linear velocity of 1 mL min⁻¹ of He carrier gas, and split ratio of 15 : 1. The oven temperature ramp program included an initial temperature of 60 °C (2 min), with an increase at 8 °C min⁻¹ and a final temperature of 300 °C (8 min). The MS was operated in electron ionization mode with an ion source temperature of 250 °C, transfer line temperature of 250 °C and ionization energy of 70 eV.

3. Results

The content of contaminant phenols in treated drilling wastewater in the three processes was so varied that seven main phenolic contaminants were individually evaluated in this study, and their initial concentrations (IC) and residual concentrations (RC) are shown in Table 3 and the total removal rate was calculated according to the following equation:

Table 3 The concentration of phenolic pollutants in the three processes

Items	UV-Fenton (pH = 3 ± 0.3)		UV-Fenton/HA (pH = 6.5 ± 0.3)		UV-Fenton/CMCD (pH = 6.5 ± 0.3)
	Initial/residual concentration (μg L^−1)
Phenol	14 123.4	34.2	17 123.4	147.7	12 123.4	241.1
p-Cresol	4939.3	2.8	3939.0	87.3	5939.3	93.0
m-Cresol	7621.4	212.4	8621.4	431.1	7321.4	209.2
4-Ethylphenol	2314.4	0	2214.4	155.8	2014.4	143.1
Bisphenol A	523.1	43.5	623.0	131.0	403.1	33.0
1-Naphthol	365.2	49.3	345.2	149.2	325.2	21.3
2-Naphthol	1865.2	165.3	1565.2	323.3	1465.2	98.1
Sum (mg L^−1)	31.75	0.51	34.43	1.43	29.59	0.84
Total removal (%)		98.40		95.85		93.16

---

According to the results shown for the three processes, the content of contaminant phenols in treated drilling wastewater had mostly changed and met the effluent criteria even in low Fenton reagent concentrations. However, there were some marked differences in the three processes.

3.1 Classical UV-Fenton

The drilling wastewater treatment with the UV-Fenton process at pH = 3 successfully removed the phenolic pollutants. During this treatment process, the dark Fenton reaction only converted 19.4% of the COD and a growth of BOD₅ was observed during the first 15 min (Fig. 1). This phenomenon could be interpreted by the fact that many oxidation intermediates were generated during the degradation of contaminants with the Fenton reagents. In addition, it is worth noting that these three processes were all conducted in the dark for 15 min before the UV-Fenton reaction. This was a necessary step because the low light transmittance of the original sample would affect the result of the UV-Fenton process. According to the UV transmittance assessment, the transmittance of the original was about 5% at 254 nm. After the dark Fenton degraded the sample, the transmittance increased to at least 70%. The colored colloidal particles that had not been removed by the coagulation treatment were almost broken down by the dark Fenton process.

Fig. 1 Comparison of the removal of COD and BOD₅ in three different UV-Fenton processes.

The monitoring curves of the Fe²⁺ concentration and H₂O₂ consumption are shown in Fig. 2. The Fe²⁺ concentration was stable at 2 mg L⁻¹ and the H₂O₂ consumption was 801 mg L⁻¹ after irradiation for 120 min (Q_UV = 1.18 kJ L⁻¹). COD (51.77%) was removed at the end of experiment. Compared with a previous study which treated model wastewater with a single substrate, this result was unsatisfactory.²⁷ The reason for this could be attributed to the effect of the presence of inorganic anions (Table 1). Many studies investigated that some inorganic anions, especially halide ions, might alter the reaction dynamics of the photo-Fenton process and the extent of the change depended on the type and concentration of ions in solution.²⁸ Yuan et al., confirmed that compared with Cl⁻, Br⁻ had a greater inhibitory effect on the bleaching rate of azo dye.²⁹ In this system, there were several halide ions in the original wastewater and their effect should be considered. Moreover, in contrast with the literature, the concentration of Cl⁻ (<100 mM) was lower than the effective inhibitory concentration and the generation of some radicals like ClOH⁻ could play a role in pollutant oxidation. On the contrary, many studies reported that the degradation of pollutants was improved by existing chlorine ions during the photo-Fenton process.^30,31 Therefore, the inhibitory effect in the UV-Fenton process could not be attributed to Cl⁻. Furthermore, along with the literature,³² the inhibitory effect of Br⁻ was more significant in this system, although the concentration of Br⁻ was less than that of Cl⁻.

Fig. 2 Comparison of the consumption of H₂O₂ and Fe²⁺ concentrations in solution in three different UV-Fenton processes.

However, the UV-Fenton process was still successful in the conversion of phenolic pollutants, and the phenolic pollutants were all up to the effluent standard after irradiation for 60 min (Q_UV = 0.59 kJ L⁻¹) as shown in Fig. 3. The residual concentrations of phenol, p-cresol and 4-ethylphenol were below 0.24% compared with the original values. The recalcitrants m-cresol, bisphenol A, 1-naphthol and 2-naphthol were below 13.4%. COD/BOD₅ reached 0.53 at the end of experiment which means that the effluent was suitable for the biodegradation process.

Fig. 3 UV-Fenton degradation of 7 phenolic contaminants in drilling wastewater at pH = 3 ± 0.3.

Moreover, to describe the degradation kinetics, the experimental data from Fig. 3 was fitted to a pseudo-first-order kinetic model. The kinetic rate constant k was calculated from the following equation.

The kinetic constants and regression coefficients of the degradation of phenols are summarized in Table 4. The results showed that the rate of degradation depended on the structure and property of the phenolic pollutants. Phenol, p-cresol and 4-ethylphenol, due to their simple structures and active chemical properties, were easier to remove with the UV-Fenton reaction than m-cresol, bisphenol A, 1-naphthol and 2-naphthol.

Table 4 Pseudo-first-order rate constants of UV-Fenton degradation of 7 phenolic contaminants in drilling wastewater at pH = 3

Items	UV-Fenton (pH = 3)
	k (min^−1)	R^2
a Standard error.
Phenol	−0.0347 ± 0.003^a	0.9602
p-Cresol	−0.0416 ± 0.001^a	0.9958
m-Cresol	−0.0246 ± 0.002^a	0.9583
4-Ethylphenol	−0.0642 ± 0.003^a	0.9798
Bisphenol A	−0.0221 ± 0.005^a	0.9228
1-Naphthol	−0.0151 ± 0.001^a	0.9909
2-Naphthol	−0.0158 ± 0.001^a	0.9937

---

3.2 Modified UV-Fenton process with HA

The result of drilling wastewater treatment by UV-Fenton with HA modified at pH = 3 was more unfavorable than the other processes, because of the lower B/C (around 0.19) and longer irradiation time required to reach the same residual concentration of the phenolic pollutant. The COD concentration remained around 307 mg L⁻¹ after irradiation for 120 min (Q_UV = 1.22 kJ L⁻¹), and the removal rate was only 33%. Meanwhile, H₂O₂ was consumed (544.4 mg L⁻¹) less than that of the classical UV-Fenton process. During the Fenton process, the precursor Fe²⁺ rapidly reacted with H₂O₂ and was converted to Fe³⁺ which favored the formation of iron hydroxides at neutral pH as a limiting factor for iron cycling. Indeed, the Fenton reaction stopped when Fe²⁺ converted to Fe³⁺. Nonetheless, Ruales-Lonfat et al. reported that the iron hydroxides in suspension when UV illumination started were still able to continue the redox reaction and consume H₂O₂ to generate OH· as catalysts of a heterogeneous UV-Fenton process (eqn (9) and (10)).³³

Fe³⁺OH(solid) + hv → Fe²⁺ + HO· (9)

Fe²⁺(solid) + H₂O₂ → Fe³⁺ + HO⁻ + HO· (10)

At the end of experiment, the phenolic pollutants were removed but a longer exposure time was required to meet the goal of the effluent standard (Fig. 4). Among them, the residual concentrations of bisphenol A, 1-naphthol, and 2-naphthol were lower than the classical UV-Fenton process, with only 0.131 mg L⁻¹ (residual rate: 21%), 0.149 mg L⁻¹ (43%), and 0.323 mg L⁻¹ (20%), respectively. However, they were considered as the main monitoring pollutants in the new petroleum industry standard of sewage discharge (GB31571-2015), where the maximum concentration limit of bisphenol A was 0.1 mg L⁻¹ and that of 2-naphthol was 1 mg L⁻¹. Additionally, the data in Fig. 4 were fitted to a pseudo-first-order kinetic model for further analysis. The kinetic rate constant and regression coefficients of the degradation of phenols were calculated according to eqn (8) and are shown in Table 5. The results showed that the kinetic rate constant dropped significantly compared with the UV-Fenton process. Therefore, the introduction of HA in the UV-Fenton process evidently slowed down the degradation rate of phenolic pollutants. On other hand, there might exist another approach for phonic pollutant with HA rather than the effect of Fenton reagent.

Fig. 4 UV-Fenton degradation of 7 phenolic contaminants in drilling wastewater with HA at pH = 6.5 ± 0.3.

Table 5 Pseudo-first-order rate constants of the UV-Fenton degradation of 7 phenolic contaminants with HA in drilling wastewater at pH = 6.5 ± 0.3

Items	UV-Fenton/HA (pH = 6.5 ± 0.3)
	k (min^−1)	R^2
a Standard error.
Phenol	−0.0257 ± 0.007^a	0.8382
p-Cresol	−0.0196 ± 0.003^a	0.9515
m-Cresol	−0.0156 ± 0.004^a	0.9268
4-Ethylphenol	−0.0165 ± 0.006^a	0.8992
Bisphenol A	−0.0091 ± 0.001^a	0.9940
1-Naphthol	−0.0052 ± 0.004^a	0.9540
2-Naphthol	−0.009 ± 0.002^a	0.9951

---

The promoting mechanism of the UV-Fenton with HA process was reported stating that under lower concentrations HA could generate an exited triplet state under UV irradiation and generate more free radicals with dissolved oxygen than the classic UV-Fenton reaction.³⁴ However, there were other works that indicated that higher concentration of HA would inhibit the degradation of micro-pollutants which are lower in concentration than HA.³⁵ This was reported to be caused by the competing oxidation of HA with pollutants where HA acted as an OH· scavenger. The radical scavenging effects did not merely eliminate the present radicals but also limited the generation of the radical. In this case, the Fenton process degraded a significant number of phenolic pollutants, but did not reach the goal for the removal of the micro-pollutants. This is due to the fact that HA became a competitor in this process with decreasing pollutant concentrations at longer treatment times.

3.3 Modified UV-Fenton process with CMCD

The UV-Fenton process conducted with CMCD at neutral pH showed satisfactory results. The COD was removed around 61.33% with a high biodegradability ratio (B/C = 0.61) after 120 min of irradiation (Q_UV = 1.16 kJ L⁻¹). Previous studies reported that CMCD had more biodegradability than other cyclodextrin derivatives which could serve as a nontoxic carbon source for the biodegradation processes,³⁶ so BOD₅ was sharply increased to 110 mg L⁻¹ after CMCD was added in the end of this process. Moreover, it was demonstrated that CMCD was able to avoid the interference of inorganic anions in the Fenton process and effectively avoid the decrease of OH· by hydroxyl radical scavengers.¹⁸ The Fe²⁺ concentration remained at 8.15 mg L⁻¹ and the H₂O₂ consumption reached 768 mg L⁻¹ which is less than the classical UV-Fenton process at the same UV irradiation energy dose. For this system, CMCD was insufficient to bind all iron (approximately 14 mg L⁻¹) at low concentration (10 mg L⁻¹) and a fraction of the Fe²⁺ converted to Fe³⁺ was lost during in the Fenton reaction at neutral pH with the decrease of Fe²⁺ concentration in the first 30 min. However, when UV irradiation began, the iron–CMCD complex as a soluble complex ensured a stable iron supply to catalyze H₂O₂. Additionally, the results for the removal of phenolic pollutants were beyond expectations (Fig. 5). A dramatic decline of phenolic pollutants in the Fenton reaction did not occur like other processes but after UV irradiation the degradation of phenolic pollutants was accelerated. Particularly, the final concentrations of bisphenol A, 1-naphthol, 2-naphthol were 0.033 mg L⁻¹ (residual rate: 8.19%), 0.021 mg L⁻¹ (6.45%) and 0.098 mg L⁻¹ (6.3%), respectively. Moreover, the degradation kinetic analyses were made on the basis of the data in Fig. 5. The result showed that the degradation process of phenolic pollutants followed pseudo-first order kinetics. The specific results for kinetic rate constants were calculated from eqn (8) and are summarized in Table 6. The kinetic analyses showed that the degradation reaction in the UV-Fenton process with CMCD was milder than the classic process with a lower kinetic rate constant, whereas the degradation of phenolic pollutants was more complete than the classic Fenton process. This interesting difference was attributed to the role of cyclodextrins in the Fenton process.

Fig. 5 UV-Fenton degradation of 7 phenolic contaminants in drilling wastewater with CMCD at pH = 6.5 ± 0.3.

Table 6 Pseudo-first-order rate constants of the UV-Fenton degradation of 7 phenolic contaminants with CMCD in drilling wastewater at pH = 6.5 ± 0.3

Items	UV-Fenton/CMCD (pH = 6.5 ± 0.3)
	k (min^−1)	R^2
a Standard error.
Phenol	−0.0285 ± 0.005^a	0.9182
p-Cresol	−0.029 ± 0.004^a	0.9779
m-Cresol	−0.0248 ± 0.004^a	0.9448
4-Ethylphenol	−0.0204 ± 0.001^a	0.9935
Bisphenol A	−0.0151 ± 0.003^a	0.9716
1-Naphthol	−0.0177 ± 0.002^a	0.9870
2-Naphthol	−0.0178 ± 0.007^a	0.9170

---

There are many studies that include a deep discussion of the role of cyclodextrins in the Fenton process, but they lack examples on the application of the UV-Fenton process with cyclodextrins. However, it has been indicated that CMCD has a good ability to bind Fe²⁺ with many carboxylic acid groups.³⁷ According to the work on organic ligands to improve the UV-Fenton process, the positive effects of the CMCD could be comparable to EDTA.^38,39 Additionally, CMCD as a supermolecular host compound has been reported, where it can selectively degrade some aromatic hydrocarbons by forming a ternary complex which leads to the formation of hydroxyl radicals around the target pollutants and reacts with them at high rate constants.¹⁸ In recent work, some studies showed that some oxidation intermediates of phenolic pollutants could form a complex with β-CD which could act as cavity-confined co-catalyst for the acceleration of iron cycle under UV irradiation.⁴⁰

4. Conclusions

In this study, three UV-Fenton processes were compared for the treatment of shale gas wastewater which was collected from a wastewater treatment plant in the Chongqing shell gas field. The results showed that under the same irradiation dose, the UV-Fenton process at acidic pH and the modified UV-Fenton process at neutral pH successfully treated wastewater and effluent BOD₅/COD met the values of 0.53 and 0.61, respectively. The UV-Fenton with CMCD system was considered more suitable for the biological oxidation process, because of the lower consumption of H₂O₂ and a moderate reaction process. Additionally, the removal efficiency of the phenolic pollutant that were degraded was compared in this study. The three processes had the ability to oxidize phenolic pollutants, but the UV-Fenton with HA process at neutral pH had limitations in the removal of micro-pollutants.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China and the China National Petroleum Corporation Petrochemical Unit Funded Project (U1262111).

Notes and references

1. J. Y. Gao and F. Q. You, ACS Sustainable Chem. Eng., 2015, 3, 1282–1291.
2. Y. H. Chang, X. J. Liu and P. Christie, Environ. Sci. Technol., 2012, 46, 12281–12282.
3. G. Meiyu, L. Xi, P. N. Chris, B. M. Michael, S. Wenrui, C. Yuntian and X. Yuan, Renewable Sustainable Energy Rev., 2016, 66, 742–750.
4. H. Yang, R. J. Flower and J. R. Thompson, Science, 2013, 340, 1288.
5. C. E. Clark, R. M. Horner and C. B. Harto, Environ. Sci. Technol., 2013, 47, 11829–11836.
6. B. Chen, H. Yan and H. J. He, Progress in Environmental Protection and Processing of Resource, Treatment of Drilling Wastewater in Oilfield by Coagulation/Iron Chipping Micro-Electrolysis Two Process, 2013, 295–298, pp. 1273–1278.
7. G. M. S. ElShafei, F. Z. Yehia, O. I. H. Dimitry, A. M. Badawi and G. Eshaq, Appl. Catal., B, 2010, 99, 242–247.
8. D. Kong, L. N. Xing, R. Liu, J. J. Jiang, W. Y. Wang, L. Q. Shang, X. T. Wei and W. D. Hao, Food Chem. Toxicol., 2013, 60, 497–505.
9. J. Rodriguez-Chueca, C. Amor, J. R. Fernandes, P. B. Tavares, M. S. Lucas and J. A. Peres, Chemosphere, 2016, 145, 351–359.
10. J. Sharma, I. M. Mishra and V. Kumar, J. Environ. Manage., 2015, 156, 266–275.
11. G. Pliego, J. A. Zazo, P. Garcia-Munoz, M. Munoz, J. A. Casas and J. J. Rodriguez, Crit. Rev. Environ. Sci. Technol., 2015, 45, 2611–2692.
12. N. Kang, D. S. Lee and J. Yoon, Chemosphere, 2002, 47, 915–924.
13. J. C. Crittenden, S. Hu, D. W. Hand and S. A. Green, Water Res., 1999, 33, 2315–2328.
14. A. De Luca, R. F. Dantas and S. Esplugas, Water Res., 2014, 61, 232–242.
15. X. Liu, J. H. Fan, Y. Hao and L. M. Ma, Chem. Eng. J., 2014, 250, 354–365.
16. V. Balzani and F. Scandola, Tetrahedron, 1992, 48, 10443–10514.
17. Y. Liu, C. J. Zou, C. Q. Li, L. Lin and W. J. Chen, Desalination, 2016, 377, 28–33.
18. M. E. Lindsey, G. Xu, J. Lu and M. A. Tarr, Sci. Total Environ., 2003, 307, 215–229.
19. Y. B. Zhou, X. C. Gu, R. Z. Zhang and J. Lu, Ind. Eng. Chem. Res., 2015, 54, 426–433.
20. X. Zhang, X. K. Li and N. S. Deng, Ind. Eng. Chem. Res., 2012, 51, 704–709.
21. M. Kępczyński, A. Czosnyka and M. Nowakowska, J. Photochem. Photobiol., A, 2007, 185, 198–205.
22. N. Klamerth, S. Malato, M. I. Maldonado, A. Aguera and A. Fernandez-Alba, Catal. Today, 2011, 161, 241–246.
23. J. M. Monteagudo, A. Duran, I. San Martin and A. Carnicer, Appl. Catal., B, 2011, 106, 242–249.
24. L. Guo, B. B. Wang, W. Y. Huang, F. Wu and J. Huang, 2009 International Conference on Environmental Science and Information Application Technology, Vol Iii, Proceedings, 2009, pp. 322–324,  DOI:10.1109/Esiat.2009.189.
25. R. H. Lu, G. H. Huang, H. Y. Zhang and S. H. Guo, J. Environ. Eng., 2010, 136, 281–287.
26. J. B. Thacker, D. D. Carlton, Z. L. Hildenbrand, A. F. Kadjo and K. A. Schug, Water, 2015, 7, 1568–1579.
27. S. R. Pouran, A. R. A. Aziz and W. M. A. W. Daud, J. Ind. Eng. Chem., 2015, 21, 53–69.
28. G. L. Truong, J. D. Laat and B. Legube, Water Res., 2004, 38, 2383–2393.
29. R. X. Yuan, S. N. Ramjaun, Z. H. Wang and J. S. Liu, Chem. Eng. J., 2012, 209, 38–45.
30. R. X. Yuan, S. N. Ramjaun, Z. H. Wang and J. S. Liu, Chem. Eng. J., 2012, 192, 171–178.
31. M. M. Mico, J. Bacardit, J. Malfeito and C. Sans, Appl. Catal., B, 2013, 132, 162–169.
32. X. Y. Lou, D. X. Xiao, C. L. Fang, Z. H. Wang, J. S. Liu, Y. G. Guo and S. Y. Lu, Environ. Sci. Pollut. Res., 2016, 23, 4778–4785.
33. C. Ruales-Lonfat, J. F. Barona, A. Sienkiewicz, M. Bensimon, J. Velez-Colmenares, N. Benitez and C. Pulgarin, Appl. Catal., B, 2015, 166, 497–508.
34. J. Wenk, U. von Gunten and S. Canonica, Environ. Sci. Technol., 2011, 45, 1334–1340.
35. I. Michael-Kordatou, C. Michael, X. Duan, X. He, D. D. Dionysiou, M. A. Mills and D. Fatta-Kassinos, Water Res., 2015, 77, 213–248.
36. E. Mousset, M. A. Oturan, E. D. Van Hullebusch, G. Guibaud and G. Esposito, Crit. Rev. Environ. Sci. Technol., 2014, 44, 705–795.
37. W. Zheng and M. A. Tarr, Spectrochim. Acta, Part A, 2006, 65, 1098–1103.
38. X. F. Xue, K. Hanna, C. Despas, F. Wu and N. S. Deng, J. Mol. Catal. A: Chem., 2009, 311, 29–35.
39. S. Rahim Pouran, A. R. Abdul Aziz and W. M. A. Wan Daud, J. Ind. Eng. Chem., 2015, 21, 53–69.
40. L. T. Fu, Z. W. Zhao, J. H. Ma and X. F. Hu, Catal. Commun., 2015, 65, 96–101.

---