Advanced electrochemical treatment of real biotreated petrochemical wastewater by boron doped diamond anode: performance, kinetics, and degradation mechanism

Petrochemical wastewater (PCWW) refers to the wastewater discharged by petrochemical plants. It is produced in large quantities, including not only produced wastewater, but also cooling water and water for other uses. Because of the variety of petrochemical products, as well as the complex reaction processes and unit operations involved, the composition and nature of PCWW is complex and changeable. Generally, PCWW is very difficult to treat because of the high content of organic compounds, especially aromatic and aliphatic hydrocarbons, including their derivatives, and a large number of heavy metals (Jia et al. 2019).

At present, biochemical treatment technology, which is economical in investment, low in operation cost, convenient in management, and stable in treatment effect, is commonly used to treat PCWW in practice (Huang et al. 2017). Liang et al. (2019) reported on a super-sized industrial-scale plant with an expanded granular sludge bed bioreactor and aerobic-activated sludge to treat PCWW containing 4,649 ± 651 mg·L⁻¹ chemical oxygen demand (COD) and more than 50 mg·L⁻¹ of petrochemicals with removal efficiencies of 85.6 ± 2.5% and 81.5 ± 4.8%, respectively. Yang et al. (Yang et al. 2015) developed a microaerobic hydrolysis–acidification–anoxic/oxic process that had an overall COD removal efficiency of 72–79% with an influent COD of 348 ± 59 mg·L⁻¹. Although biochemical treatments have high efficiency and are the most widely used in industrial wastewater treatment, their effluent CODs are still high. With the improvement in the emission standards for industrial sewage effluent in China (especially when the effluent COD is required to be less than 50 mg·L⁻¹ to meet the 1A effluent discharge standard), simple biochemical treatment technology may have difficulty in meeting these standards, so coupling with other advanced treatment technologies to further reduce COD is required.

At present, there are already several reports on the BDD anodic oxidation of biochemically treated industrial wastewaters. Martin de Vidales et al. (Martin de Vidales et al. 2015) reported the application of a BDD electrode in the electrochemical oxidation of caffeine-intensified biologically treated urban wastewater to obtain almost complete mineralization. Garcia-Segura et al. (Garcia-Segura et al. 2015) reported the use of the BDD anode to treat secondary effluent from a municipal sewage treatment plant containing 29 target pharmaceuticals and pesticides and achieved almost complete mineralization. However, compared with these wastewaters, the composition of PCWW is more complex and the treatment is more difficult. Moreover, most of the reported electrochemical treatments of PCWW are direct oxidation, which always suffer from high effluent COD or high energy consumption (Rocha et al. 2012; dos Santos et al. 2014b), making them unsuitable for practical application. There are few reports focusing on the combining of biochemical treatment and BDD anodic oxidation technology for the treatment of PCWW to produce ultra-low COD effluent.

In this study, the electrochemical degradation of real biotreated petrochemical wastewater (BTPCWW) by a BDD anode was conducted to achieve an effluent COD lower than 50 mg·L⁻¹. The effects of the applied current density, pH of the solution, agitation rate, and type of anode materials were investigated. The main contaminants in the influent and the oxidation intermediates were determined by gas chromatography coupled to mass spectrometry (GC-MS). Moreover, the possible degradation mechanism was proposed according to the results of the GC-MS analysis. Finally, the energy consumption of the whole process was also estimated.

A real BTPCWW sample was derived from the effluent of an industrial sewage treatment plant in Ningbo, China. The influent contained dyestuff wastewater, paint wastewater, organic phosphorus wastewater, organic cyanide wastewater, halogenated organic wastewater, ether-containing wastewater, and organic resin wastewater produced by companies in a chemical industrial district in Ningbo. The wastewater had strong toxicity and a complex composition, which may have contained oil, cyanide, phenols, sulfur, arsenic, aromatic hydrocarbons, ketones, and other pollutants. The wastewater was treated first by an anaerobic–aerobic biochemical technique in the industrial sewage treatment plant. The pH of the secondary effluent was about 7.4 and the COD was about 120 mg·L⁻¹. It should be noted that the wastewater used in this study was the raw effluent without any additional treatment.

The BDD electrode was supplied by Condias (Germany). The coatings were double-sized deposited on an Nb plate (20 × 20 × 2 mm) by hot filament chemical vapor deposition. A graphite plate electrode was purchased from Chuxi Industrial Co., Ltd, China. All chemicals at the analytical grade were purchased from Aladdin Industrial Inc., China, and used without further purification. All aqueous solutions were prepared with ultrapure water (resistivity >18 MΩ·cm) at room temperature.

The degradation of the BTPCWW was conducted in batch within an undivided electrochemical cell with a capacity of 300 mL. The solution was stirred by a magnetic stirrer. Of the real BTPCWW, 200 mL was stored in the electrochemical cell each time and the electrolysis was conducted under galvanostatic condition by a DC power supply (MS-305D, Maisheng, China). A BDD or graphite electrode with a surface area of 8 cm² was used as the anode while a same-sized stainless-steel plate was used as the cathode. In all conditions, the distance between the anode and cathode was maintained at 2 cm and the solution was kept at 25 °C by a thermostat water bath. The adjustment of the pH value was conducted using 1 mol·L⁻¹ H₂SO₄ or 1 mol·L⁻¹ NaOH solution. The applied current was 40–120 mA, pH value was 3–10, and agitation rate was 200–600 rpm.

The compounds in BTPCWW and corresponding intermediates in the degradation process were determined by GC-MS (Shimadzu QP2010 SE, Japan) coupled with an Agilent DB-1 column (30 m × 0.25 mm, 1 μm film thickness). The injection volume was 1 μL. The temperatures of the detector and the ion source were both 250 °C.

The concentration of chloride ion was determined by titrimetry with K₂CrO₄ solution as the indicator and AgNO₃ solution as the titrator.

As a key variable affecting anodic oxidation performance and energy consumption, applied current density has been reported to be the most important factor governing the production of •OH at a BDD anode and determining the degradation rate and effectiveness. Therefore, the effect of the current density on the anodic oxidation of BTPCWW was analyzed. As shown in Figure 1(a), the COD removal efficiency increased with the current density from 5 to 15 mA·cm⁻², which was mainly due to the larger amount of •OH produced by the BDD anode under a higher applied current. However, COD could not be completely removed under all applied currents. After 3 h of electrolysis, the final COD removal efficiencies were 70.6%, 71.7%, 74.9%, 76.8%, and 82.6% at the current densities of 5, 7.5, 10, 12.5, and 15 mA·cm⁻², respectively. Moreover, the inset of Figure 1(a) shows the pseudo-first-order kinetics for all applied current densities. The observed pseudo-first-order kinetic rate constant (k_obs) increased from 0.412 to 0.636 h⁻¹ when the current density increased from 5 to 15 mA·cm⁻², further proving that the increase in current density could accelerate the BDD anodic oxidation of BTPCWW.

Unlike generated H₂O₂ and O₂, the generated and ClO⁻ are helpful, and can also contribute to the electrochemical degradation of organic compounds through indirect oxidation (Cotillas et al. 2018a).

The occurrences of side reactions can be further confirmed by the ICE values (shown in Figure 1(b)), which were below 1 under all applied current densities. Moreover, ICE decreased with increasing electrolysis time, which is a typical electrochemical mass transfer control process (Cotillas et al. 2018b). However, the highest initial ICE value was obtained at the lowest applied current. For current densities of 5, 7.5, 10, 12.5, and 15 mA·cm⁻², the corresponding initial ICE values were 0.723, 0.595, 0.564, 0.568, and 0.590, respectively, mainly because the increasing current was used in parasitic reactions at the anode rather than in the degradation of the organic compounds.

Overall, the degradation of BTPCWW followed the pseudo-first-order kinetic. With the increase in current density, a higher degradation rate and higher COD removal efficiency were obtained, whereas the corresponding current efficiency decreased because of the occurrence of the side reactions.

The effect of pH on the anodic oxidation of organic compounds was contradictory, since it is closely related to the chemical properties of the treated organic compounds. Therefore, the effect of pH on the anodic oxidation of BTPCWW with a BDD anode was studied. The results are shown in Figure 2. The fastest degradation rate was observed at a pH of 3 (k_obs= 0.591 h⁻¹) with a final COD removal efficiency of 83.5% (COD = 19.7 mg·L⁻¹). At a pH of 10, k_obs decreased to 0.382 h⁻¹ with a final COD removal efficiency of only 65.8%. The pH can affect the rate of the electrogeneration of •OH (Kouskouki et al. 2019), therefore, it can be inferred that a large amount of physisorbed •OH's being produced at low pH values. On the other hand, increasing the pH value will reduce the O₂ over-potential, leading to the promotion of the oxygen evolution reaction and reduction of the degradation's performance (Ammar et al. 2016). It is noteworthy that after 3 h of electrolysis in the alkaline medium, the solution became a bit turbid with a white flocculent. This may have been due to the formation of polymer intermediates, which make the degradation process more difficult (Saad et al. 2016).

The variation of the ICE values under different pH values was also examined. The results are displayed in Figure 2(b). Similar to COD removal efficiency, the highest ICE value was obtained at a pH of 3. For pH of 3, 7.4, and 10, the initial ICE values were 0.670, 0.564, and 0.489, respectively. Typically, ICE at pH of 3 is 1.37 times of that at pH of 10.

Accordingly, an acidic medium not only increased COD removal efficiency, but also had higher current efficiency, as compared with the neutral and alkaline media.

To verify the role of the mass transfer process in the electrochemical oxidation of BTPCWW with the BDD anode, the effect of the agitation rate, which ranged from 200 to 600 rpm, was studied. As shown in Figure 3(a), the agitation rate had a significant impact on the COD removal efficiency, when the agitation rate was below 400 rpm, the COD removal efficiency increased with the agitation rate. Under a low agitation rate, mass transportation would be limited and would lead to low COD removal efficiency (de Oliveira Marcionilio et al. 2019). However, at an agitation rate of 600 rpm, an unexpected performance of the COD removal was seen. In the first 1.5 h, the COD removal rate at 600 rpm was faster than that at 400 rpm. Afterward, the rate increased extremely slowly until the final COD removal efficiency was only 65.3%, which was much lower than that at 400 rpm (74.9%). This result may have been due to the electro-oxidation with the BDD anode's having been controlled mainly by direct oxidation rather than indirect oxidation with the other oxidants (Moreira et al. 2015). Under such a mechanism and with the decrease in the concentrations of the pollutants in the solution, the amount of adsorbed pollutants on the surface of the electrode decreased dramatically when the agitation rate was too fast, resulting in a significant decrease in the COD removal efficiency.

The variation of the ICE values also shows a similar trend (shown in Figure 3(b)). The highest initial ICE value of 0.879 was obtained at an agitation rate of 600 rpm, which was 1.56 times that at 400 rpm. But the value decreased very rapidly and became lower than that at 400 rpm after about 0.7 h and reached 0.016 after 3 h of electrolysis, which was only 19.1% of that at 400 rpm, signifying that the side reactions, including O₂ evolution, had become serious at 600 rpm as a result of the low quantities of adsorbed organic pollutants on the surfaces of the electrode.

Accordingly, too fast or too slow agitation rate was detrimental to the degradation of BTPCWW. In this study, agitation rate of 400 rpm is considered as the most appropriate value.

The anode plays an important role in electrochemical oxidation, since it has a great influence on the oxidation efficiency and reaction mechanism. This study comparatively analyzed the electrochemical degradation of BTPCWW on the BDD and graphite electrodes. As shown in Figure 4(a), the COD removal efficiency on the BDD electrode was significantly higher than that on the graphite electrode. After 3 h of electrolysis, COD removal efficiencies of 74.9% and 49.6% were obtained from the BDD and graphite electrodes, respectively. Moreover, the inset of Figure 4(a) shows the pseudo-first-order kinetic with k_obs of 0.518 h⁻¹ and 0.217 h⁻¹ for the BDD and graphite electrodes, respectively. The higher oxidation efficiency of the BDD electrode could be explained by a large amount of electrogenerated •OH. The BDD electrode was classified as an inert electrode. Therefore, a large amount of •OH was physiosorbed on the surface of the electrode, resulting in a higher degradation rate and efficiency. On the contrary, the •OH produced on the graphite electrode was less and its oxidation depended mainly on the chemisorbed ‘active oxygen’ (oxygen in the oxide lattice, MO_x+1) and other strong oxidants (ClO⁻ and ⁠) (Fil et al. 2014). However, •OH had the strongest oxidation ability, so the BDD electrode was more efficient for the anodic oxidation of BTPCWW.

The variation of ICE also shows a similar trend (shown in Figure 4(b)). The initial ICE values were 56.4% and 29.1% for the BDD and graphite electrodes, respectively, with the former's being 1.94 times larger than the latter, signifying that most of the applied current had been used to oxidize organic pollutants on the surface of the BDD electrode instead of creating side reactions, as compared with the graphite electrode.

To investigate the possible electrochemical degradation pathway of BTPCWW on the BDD electrode, the electrolysis of 200 mL of wastewater was conducted under a current density of 10 mA·cm⁻², pH of 7.4, and agitation rate of 400 rpm. The corresponding intermediates and main products were determined by GC-MS analysis.

Firstly, the GC-MS analysis of the raw wastewater revealed that it was mainly composed of biochemical refractory compounds, including nitrous oxide, sulfonates (i.e. aminomethanesulfonic acid), alcohol, ammonium chloride, chloroalkanes, benzenes (i.e. p-methylaminoethyl phenol, nitrobenzene, nitrophenyl sulfonate etc), and small amounts of formic acid and acetic acid. During electrolysis, the main intermediates detected included benzoquinone, methylphenylethylamine, methylsulfonyl acetic acid, cyclohexylethanamine, pyruvic acid, ethanol, ammonium chloride, chloramines, malonic acid, formic acid, acetic acid, and carbon dioxide. After 3 h of electrolysis, the main products in the effluent were ammonium chloride, formic acid, acetic acid, and malonic acid.

After consideration of the peak heights, as well as the components of the final products, carbon dioxide, malonic acid, ammonium chloride, formic acid, and acetic acid were selected for quantitative analysis. The results are shown in Figure 5. The concentration of CO₂ increased rapidly to 33.3 mg·L⁻¹ in the first half-hour, then began to decline gradually until no CO₂ was detected after 3 h of electrolysis, indicating that the mineralization of the organic compounds had been achieved during the electrochemical oxidation of BTPCWW. In the initial stage of the reaction, the current efficiency is relatively high and the mineralization of the organic compounds dominated with only a few side reactions having occurred. Thus, a large amount of CO₂ was generated. As the reaction proceeded, the concentrations of the organic compounds declined continuously, the occurrences of the side reactions increased, and the generation of CO₂ by the mineralization of the organic compounds decreased. The variation of NH₄Cl behaved oppositely by decreasing rapidly from 66.9 to 14.5 mg·L⁻¹ in the first half-hour, then increasing gradually to a final concentration of 61.0 mg·L⁻¹. The initial decrease of ammonium was due to the formation of nitrates through oxidation via •OH. The subsequent increase in ammonium could be explained by the reduction of nitrate to ammonium at the stainless-steel cathode (de Oliveira Marcionilio et al. 2019). Furthermore, the concentration of Cl⁻ was also determined, because the presence of chloride compounds had been detected by GC-MS, and chloride ions were also considered to be one of the main oxidants in electrochemical oxidation (Li et al. 2017). In this study, the detected initial and final concentrations of Cl⁻ were 221.7 and 162.4 mg·L⁻¹, respectively, signifying that a part of Cl⁻ may have been oxidized to active chlorine, chlorate, and perchlorate, proving that indirect oxidation also occurred. In addition, the degradation of the chemicals on the graphite anode may have depended mainly on Cl⁻ in wastewater. CH₂O₂, C₂H₄O₂, and C₃H₄O₄ were the main organic compounds in the final effluent. During the whole electrolysis process, they exhibited a similar evolution, which increased in the first few hours but then decreased gradually. However, their concentration peaks appeared at different times. C₂H₄O₂ and C₃H₄O₄ reached their peaks at 0.5 h, while CH₂O₂ reached its peak in 1 h, confirming that CH₂O₂, C₂H₄O₂, and C₃H₄O₄ were intermediates from the oxidation of the other organic compounds. Accordingly, the possible degradation pathways of the anodic oxidation of BTPCWW with the BDD electrode are displayed in Figure 6.

Since the treated wastewater in this study was real industrial wastewater, it was particularly important to estimate the cost. The evolution of energy consumption against COD removal efficiency under different current densities was calculated. The results are displayed in Figure 7. Generally, the energy consumption increased with the current density under the same COD removal efficiency and the variation was more significant at higher COD removal efficiency. When the current density was 5, 7.5, 10, 12.5, and 15 mA·cm⁻², the corresponding average voltage was 4.71, 4.94, 5.20, 5.65, and 5.96 V, respectively. The voltage growth was faster than the increase in COD removal efficiency caused by the increasing applied current. Thus, the result was higher energy consumption at high current density. However, it should be noted that the COD removal rate and efficiency was relatively deficient at low current density. Therefore, it was necessary to select an appropriate applied current (i.e. an applied current density of 10 mA·cm⁻² was used).

Some researchers have also studied the energy consumption in the treatment of real wastewater with BDD electrodes. The results are listed in Table 2. The direct electrochemical oxidation of PCWW was difficult, the reported energy consumption in Rocha's study was as high as 109 kWh·dm⁻³, signifying that this treatment was unsuitable for industrial application. Although, in Gargouri's study, the energy consumption was much lower than that of Rocha, it was still too high for practical application, and the most important thing is that the effluent COD was still as high as 794 mg·L⁻¹. However, the treatment of biotreated wastewaters (both coking and winery wastewater) showed a significant reduction of energy consumption with low effluent COD. In this study, on the basis of the existing biochemical process, electrochemical oxidation with BDD anode was considered to treat the BTPCWW, in order to reduce the effluent COD and improve the effluent quality. Without chemical supplements and pH adjustment, the energy consumption was only 6.24 kWh·m⁻³ with a final COD of 30.2 mg·L⁻¹ at a current density of 10 mA·cm⁻², which was basically the same as that of treatment of biotreated winery wastewater. These results indicate that BDD anodic oxidation has a relatively low cost and it has good potential applications for existing biochemically treated industrial petrochemical wastewater treatment plants, which would need to further improve the effects of their treatments.

The electrochemical degradation of real BTPCWW with a BDD anode was conducted. The results showed that the degradation process followed the pseudo-first-order kinetic. With the increase in the applied current density, a higher degradation rate and higher COD removal efficiency were obtained, whereas the corresponding current efficiency decreased during degradation. An analysis of the effect of the pH value revealed that the degradation of BTPCWW favored an acidic medium by faster COD removal efficiency and higher current efficiency. The effect of the agitation rate was variable. The highest COD removal efficiency was achieved under 400 rpm and a further increase in the agitation rate would only increase COD removal in the first few hours but lower the total COD removal efficiency. In addition, a comparative study of the degradation effect of the BDD and graphite anodes was conducted and a better COD removal efficiency was obtained with the BDD electrode. A GC-MS analysis revealed that after 3 h of electrolysis with the BDD electrode, the main products were ammonium chloride, formic acid, acetic acid, and malonic acid. Finally, an estimation of the energy consumption was 6.24 kWh·m⁻³ with a final COD of 30.2 mg·L⁻¹ at a current density of 10 mA·cm⁻² without the need for a chemical supplement.

This work was supported by the Ningbo Natural Science Foundation (grant number 2018A610280) and Zhejiang Provincial Natural Science Foundation of China (grant number LQ20B070004).

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