Electrochemical oxidation of guaiacol to increase its biodegradability or just remove COD in terms of anodes and electrolytes

Abstract

To assess the role of electrochemical oxidation as a single mineralization technique or a pretreatment (for biotreatment) of lignin-containing wastewater, Ti/Sb-SnO₂ and Ti/Pb₃O₄ were used as comparable anode materials to degrade the lignin monomer guaiacol. The electrode catalytic activity was examined by cyclic voltammetry. Then the guaiacol degradation was analyzed. Finally, the main intermediates and products were identified. The results show that Ti/Sb-SnO₂ is non-selective in treating guaiacol and its intermediates while Ti/Pb₃O₄ has a weaker ability in degrading the intermediates. A chlorine-containing media is suitable to perform total organic removal with a high rate, but the active chlorine and organochlorine may ruin the follow-up biological treatment. Electrochemical oxidation (EO) pretreatment could save 50–70% electric energy and reduce the burden of subsequent biotreatment. Quinones and dimers are important intermediates that could be used to distinguish the capability of anodes and the function of Cl⁻.

---

1. Introduction

Lignin is one of the major components of wood, which is a prior contaminant in water used in the paper and pulp industry.^1,2 The lignin is concentrated during the water recycling process in paper making, causing the deterioration of water quality. Therefore lignin must be removed in time to ensure the quality of paper/pulp for the downstream manufacture. Lignin is a natural material, but the direct discharging of lignin is not advisable because of its high chemical oxygen demand (COD) and color.^3,4 To save the water consumption in this industry, the lignin-containing water is generally treated and reused to supplement fresh manufacturing water.

Due to its phenolic-polymer structure, lignin is more biorefractory than other common organics.^5,6 Lignin is generally present in the form of salt with high solubility, therefore the lignin-containing solution is stable and expensive to remove by physical methods. Among various chemical and biotreatment (BT) methods, electrochemical oxidation (EO) may be a suitable approach to treat lignin-containing solutions, where the anodic oxidants could attack the linkage of monomer and the benzene ring in priority.^7,8 In recent years, many researchers devoted themselves into utilizing this technique to treat lignin containing solution under practical conditions. The recent development as well as an interesting issue is to combine EO with other methods, to increase the efficiency and lower the load imposed to each method.^9–12 The electrochemical oxidation–biotreatment (EO–BT) is a promising approach. The idea is to pretreat lignin using EO to destroy the recalcitrant sections of lignin, and then, the effluents of EO would be more amenable to the following biodegradation. The objective role of EO in sole-EO and EO–BT processes are different. The sole EO is aimed for removing target organics, color and COD to achieve standard discharge. The EO in EO–BT is aimed for enhancing the biodegradability as higher as possible on the premise of energy saving. Our previous study has confirmed the feasibility of using EO as the pretreatment of BT (EO–BT), where Ti/Sb-SnO₂ and Ti/PbO₂ electrodes exhibited good performances of degrading sodium lignosulphonate and the biodegradability of the treated solution was successfully increased from biorefractory level to biodegradable level.¹²

Lignin has no specific molecular weight. Therefore it is difficult to investigate its exact degradation mechanism directly. Guaiacol (2-methoxy-phenol) is a typical lignin monomer, hence it was chosen as the probe to examine lignin degradation mechanism. Moreover, few reports revealed the lignin degradation results in chloride-containing wastewater, which is important for areas using sea water in their flushing systems (e.g. Hong Kong). Therefore the assistance of electro-generated active chlorine was investigated as well. In this paper, two typical anodes Ti/Sb-SnO₂ and Ti/Pb₃O₄ were used to treat guaiacol solution. Ti/Sb-SnO₂ is well known for its high activity of generating strong oxidizing hydroxyl radical (˙OH),^13–15 while Ti/Pb₃O₄ is a good anode material for chlorine evolution in chloride containing solution as previously reported.¹⁶ Therefore in this experiment their unique capabilities are expected to be utilized properly based on the specific media and purpose.

2. Experimental

All reagents were analytical grade. The deionized-distilled water (pure water) used in all the experiments was obtained from a Millipore Waters Milli-Q water purification system.

2.1 Electrode preparation

Ti/Sb-SnO₂ and Ti/Pb₃O₄ electrodes were fabricated as stated previously.^16,17 In brief, Ti plate or stick was first pretreated using sand blasting, alkali degreasing and acid etching in turn. Then the Ti substrate was electrochemical reduced in aqueous solution and a TiH_x layer was formed. Then the metal Sb and Sn were electrodeposited on this TiH_x layer. Then the sample was heated in an oven at 773 K for 1 h. After natural cooling, the fabrication of Ti/Sb-SnO₂ electrode was completed. To prepare Ti/Pb₃O₄ electrode, the metal Pb was electrodeposited on Ti/Sb-SnO₂ basement, and then the sample was heated in an oven at 773 K for 1 h. After natural cooling, Ti/Sb-SnO₂/Pb₃O₄ electrode (i.e. Ti/Pb₃O₄ in short) was made.

2.2 Electrode characterization

The electrochemical property of the two anodes were characterized by cyclic voltammetry (CV) using a potential-galvanostat (CHI 660d, Chenhua, China). The test cell was a beaker containing 0.2 L salt solution (Na₂SO₄ or NaCl). The fabricated electrode was used as the working electrode (effective electrode area 12 cm²). Copper plate was used as the counter electrode. Ag/AgCl was used as the reference electrode.

2.3 Guaiacol degradation

Guaiacol (2-methoxy-phenol, C₇H₈O₂, CAS no. 90-05-1, Sinopharm, China) was dissolved in the pure water at a concentration of 500 ppm. The solution volume during the test is 0.2 L. Solution was stirred by magnet (300 rpm). The fabricated electrodes were used as anodes with an effective area 43 cm². Graphite sticks with the same size were used as cathodes. A potential/galvano-stat (PS-305D, Shanghai Yutai, China) was utilized to perform the galvanostatic condition. The anodic current density was defined at a moderate level of 20 mA cm⁻². The sample was taken every 0.5 h for further tests. No thermostat was used because the solution temperature remained at ∼25 °C (room temperature).

2.4 Evaluation of degradation result

An UV-vis spectrophotometer (Agilent 8453, Agilent) was employed to monitor the UV-vis absorption for evaluation of guaiacol removal. Aerobic biodegradability of the solution was evaluated by BOD₅/COD ratio. The COD and 5 days BOD were determined according to the Standard Methods (A. D. Eaton et al., APHA, AWWA, WEF, Baltimore, 2005).

2.5 Identification of the intermediates

The separation and identification of guaiacol and its degradation intermediates was performed at a Thermo Quest Finnigen LCQ Duo Mass spectrometer system, which was equipped with Thermo P4000 pumps, Thermo AS3000 autosampler, UV6000LP photodiode array UV detector and electrospray ionization with a quadruple ion-trap mass spectrometer operating at a negative mode (LC-ESI/MS). For the intermediates separation, a Varian Pursuit XRs C18 column (3.0 μm, 2.0 × 150 mm) was utilized. The mobile phase was comprised with two solutions: (A) ultra-pure water (MS grade), and (B) acetonitrile. The composition of the mobile phase was changed according to the following gradient: 100% of water during the first 2 min; from 2 to 16 min, B was linearly increased from 0 to 50%; from 16 to 20 min, B was continuous linearly decreased to 0%. The flow rate of mobile phase was kept at 0.1 mL min⁻¹. GC-MS (Thermo Fisher, USA) was also used to separate and identify the intermediates and products with small molecular weight. The equipment used is a Thermo Trace GC Ultra coupled to an ISQ mass spectrometer equipped with a HP-5 capillary column (30 m × 0.25 mm i.d., 0.25 μm film). Exact 1 μL sample was injected in the gas chromatography for analysis. SPDE technique was used to operate the auto sampler, while the NIST database was used to identify the compounds. The detected compounds were considered to be positively identified when their mass spectra and GC retention times agreed with those of authentic samples.

3. Results and discussion

3.1 Electrode catalytic activity for guaiacol degradation

The electrochemical oxidation can generally be achieved in a direct and/or indirect way. Anode may directly oxidize the organics on the material surface (i.e. direct oxidation).¹⁸ Simultaneously, the electrogenerated oxidants may oxidize the organics in bulk solution via indirect oxidation.¹⁹ The portion of the two ways depends on the anode material and the media. Cyclic voltammetry may be a simple but visual approach to analyze the tendency of direct oxidation. Fig. 1 shows the cyclic voltammogram curves of both electrodes under different conditions.

Fig. 1 Cyclic voltammograms of Ti/Sb-SnO₂ and Ti/Pb₃O₄ in different solutions with or without guaiacol (abbreviated as R) (a) Ti/Sb-SnO₂ in 5 wt% Na₂SO₄ (b) Ti/Sb-SnO₂ in 5 wt% NaCl (c) Ti/Pb₃O₄ in 5 wt% Na₂SO₄ (d) Ti/Pb₃O₄ in 5 wt% NaCl.

In chlorine-free media, the oxygen evolution potential (OEP) of Ti/Sb-SnO₂ and Ti/Pb₃O₄ is around 1.9 V and 1.6 V (vs. Ag/AgCl), respectively (calculated from Fig. 1a and c). Since oxygen evolution and organic oxidation are competitive reactions, the above result indicates that the oxidation power of Ti/Sb-SnO₂ is better than Ti/Pb₃O₄. When guaiacol was added, the CV curve of Ti/Sb-SnO₂ slightly changed (current decrease at high potential) (Fig. 1a), indicating only a fraction of guaiacol adsorbed on the electrode surface (direct oxidation). Guaiacol does not seem to be electroactive in the potential window of the Ti/Sb-SnO₂. The indirect oxidation by hydroxyl radicals may play a major role under this circumstance. However, the current response of Ti/Pb₃O₄ increased significantly after guaiacol was added (Fig. 1c), indicating a large amount of guaiacol adsorbed on Ti/Pb₃O₄ electrode surface and guaiacol seem to be electroactive on this anode material (direct oxidation).

In chlorine-containing media, the current response of both electrodes (Fig. 1b and d) was higher than before (Fig. 2a), indicating the additional contribution of chlorine evolution. When guaiacol was added, the direct oxidation of guaiacol was obvious for Ti/Pb₃O₄, and the current response was even larger than that in Fig. 2c. But this phenomenon was still not obvious for Ti/Sb-SnO₂.

Fig. 2 Important indexes in electrochemical oxidation of 500 ppm guaiacol solution by Ti/Sb-SnO₂ and Ti/Pb₃O₄ (5 wt% Na₂SO₄ or 5 wt% NaCl as the supporting electrolyte) (a) guaiacol removal rate% analyzed by liquid chromatography (b) UV-vis absorption spectra versus time (diluted solutions) (c) UV_{275 nm} removal rate% according to UV-vis absorption spectra (d) COD removal rate%.

Thus, the oxidation modes on the two anodes are revealed after incorporating the result of CV tests and review on literature.^18,19

The possible anodic reactions (depending on anode nature and supporting electrolyte) are proposed as follows.

Despite of anode types or solutes, at high anode potentials, the discharge of water occurs all the time:

H₂O → ˙OH + H⁺ + e⁻ (1)

˙OH_ads → O₂↑ (side reaction of electrochemical oxidation) (2)

For Ti/Sb-SnO₂, anodic generated hydroxyl radicals are basically physical adsorbed,²⁰ and indirect oxidation is the main oxidation mode:

˙OH_ads + R → products (indirect oxidation) (3)

For Ti/Pb₃O₄, anodic generated active oxygen is mostly chemically adsorbed (i.e. Pb₃O₄ lattice is incorporated by oxygen¹⁶), so direct oxidation is main oxidation mode for this anode:²¹

Pb₃O₄ + x˙OH_ads → Pb₃O_4+x + xH⁺ + xe⁻ (4)

Pb₃O_4+x + R → Pb₃O₄ + products (direct oxidation) (5)

When Cl⁻ is present, chlorine evolution happens on both electrodes and generates chlorine-containing radical (Cl_ads), active chlorine (Cl_act), such as Cl₂ and HOCl:²²

Cl⁻ → Cl_ads + e⁻ (6)

Cl_ads + R → Cl⁻ + products (indirect oxidation) (7)

2Cl_ads → Cl₂↑ (chlorine evolution) (8)

Cl_ads + H₂O → HOCl + H⁺ + e⁻ (9)

Cl_act + R → products (indirect oxidation in bulk solution) (10)

In summary, when Cl⁻ is present, series of reactions are involved in guaiacol degradation, including direct oxidation, indirect oxidation by adsorbed oxygen species, indirect oxidation by adsorbed chlorine species, and indirect oxidation by active chlorine in bulk solution. In the absence of Cl⁻, only the direct oxidation and the indirect oxidation by adsorbed oxygen species are possible. Scheme 1 illustrates these reaction pathways when using the two electrodes. Diverse results may be obtained due to the different oxidative power of the oxidants in each pathway (different anodes and oxidizing intermediates).²³

Scheme 1 Electrochemical oxidation mechanism of guaiacol (R in short) by Ti/Sb-SnO₂ (a) and Ti/Pb₃O₄ (b) in chlorine-free or chlorine-containing media.

3.2 Electrochemical oxidation of guaiacol

3.2.1 Evaluation of the treatment.
(1) Guaiacol removal. Result of quantitative analysis of guaiacol removal by LC is shown in Fig. 2a. Component of electrolyte played a more significant role than the anode material in guaiacol removal rate. In NaCl, the degradations of guaiacol removal on both electrodes were much faster; more than 90% removal was observed for both anodes in 2.5 h.
(2) UV-vis absorption decrease. The UV-vis absorption variations of the guaiacol solution (5 wt% Na₂SO₄ or NaCl) versus time are shown in Fig. 2b. The characteristic absorption value of guaiacol at 275 nm (UV_{275 nm}) decreased gradually during EO treatment, indicating the degradation of guaiacol (benzene structure). Fig. 2c illustrates the UV_{275 nm} removal rate% versus time according to UV-vis absorption spectra, which is slightly different from the LC analysis, due to the involvement of intermediates that also absorb UV_{275 nm}. Ti/Pb₃O₄ generally had better performance than Ti/Sb-SnO₂ in both electrolytes. In Na₂SO₄, the absorption peak reduced to 55% (Ti/Pb₃O₄) and 45% (Ti/Sb-SnO₂), respectively after 2.5 h. In NaCl, a much faster reduction of the absorption peak to 90% was observed for both electrodes in 2.5 h. By comparing Fig. 2a and c, it could be concluded that UV_{275 nm} is not an exact indicator to determine guaiacol removal, but it could be used as a quick test index.
(3) COD removal. The anodes' performance in COD removal was also examined (Fig. 2d). In Na₂SO₄, Ti/Pb₃O₄ had a faster initial COD decay than that of Ti/Sb-SnO₂, however the latter anode had a more durable and consistent performance, and a clear cross of performance curves was observed at around 1.75 h. The overall COD removal for Ti/Pb₃O₄ and Ti/Sb-SnO₂ was 50% and 60%, respectively, at 2.5 h. In NaCl, COD removal rates were much faster than those in Na₂SO₄ for both electrodes, the final COD removal were 80% at 2.5 h, in which Ti/Sb-SnO₂ had a better overall performance than Ti/Pb₃O₄.

From the above results, the two anodes have similar performance in guaiacol removal (LC analysis), while the Ti/Sb-SnO₂ and Ti/Pb₃O₄ has good overall and initial COD removal performance, respectively. This observation indicates that Ti/Pb₃O₄ may have deficient ability in eliminating intermediates. Finally, the presence of Cl⁻ could accelerate the degradation process, demonstrating the effective assistance of indirect oxidation by the active chlorine. The details of chloride in the process were therefore further investigated.

3.2.2 Kinetics. For a fixed set of electrochemical devices, when anode and solution are fixed, the current density, pH, temperature are the main important operating parameters. But in considering the possible biotreatment in real applications, a neutral condition (i.e. without pH adjustment) was used for the remaining investigations of other parameters. For the same reason, the solution temperature was fixed at ∼25 °C (room temperature), even though the lignin degradation was faster at higher temperature, as reported in previous study.¹² Fig. 3 shows the effect of anode material and supporting electrolyte on the constant rate K (first order kinetics, h⁻¹). The K value could be doubled when Na₂SO₄ was replaced by NaCl, indicating the significant assistance of indirect oxidation. There was not much difference between the two anode materials for degrading guaiacol, but according the COD measurement, their difference may be found in treating the intermediates, which will be discussed later.

Fig. 3 First order kinetics elimination of guaiacol (5 wt% supporting electrolyte, 25 °C).

The current density has a strong relationship on the degradation kinetics because the current is the direct driving force. In this experiment we found the K value increased almost linearly with the moderate current density (10 mA cm⁻² to 50 mA cm⁻²). The degradation was under kinetic control rather than mass transfer control under this circumstance. A moderate current density at 20 mA cm⁻² was adopted in the following experiments to ensure the details of reaction mechanisms can be observed in details.

3.2.3 Effect of chloride concentration. The indirect oxidation by active chlorine could be attributed as a homogeneous catalysis, in which the mass transfer is generally not a limiting step. The chlorine evolution, however, is a heterogeneous process (under relative high anode potential), the chlorine evolution efficiency mainly depends on the anode nature and chloride concentration. Several chloride concentration gradients were introduced to investigate the effect of chloride concentration on guaiacol degradation. Compared with the 5 wt% NaCl as used previously, the 1 wt% NaCl clearly increased chlorine evolution efficiency. However, the organic removal efficiency decreased simultaneously, especially for Ti/Pb₃O₄ electrode (Fig. S1 in ESI†). It could be deduced that if the chlorine evolution is predominant, the effective current for direct oxidation is decreased inevitably. Therefore too high the chlorine evolution ability of the anode, the overall organic degradation is negatively affected due to the inhibition of direct oxidation. A suitable chloride concentration region to ensure an optimized balance between the direct and indirect oxidation is therefore necessary. In this study, it is found that ∼5 wt% NaCl concentration is the appropriate dosage for this purpose.

In addition, it was interesting to find when the chloride concentration was overdosed, a special characteristic UV-vis absorption peak was recognized around 290 nm (ESI Fig. S1b†). It could be inferred that intermediates accumulated more obviously with reinforced indirect chlorination.

3.2.4 Current efficiency. The anode materials can be categorized into nonselective and selective anode by oxidation selectivity toward reactant and/or intermediates. The nonselective anode treats reactant and its intermediates equally, and results in less accumulation of intermediates. The selective anode has better oxidation ability toward the reactant; more intermediate therefore could accumulate in the solution. In this test, the average current efficiency (CE) of COD removal during each time period could be used to infer the oxidation ability of anode toward guaiacol and its intermediates. Fig. S2 (in ESI†) shows the CE of Ti/Sb-SnO₂ and Ti/Pb₃O₄ versus time in different media. In 5 wt% Na₂SO₄, the initial CE on Ti/Sb-SnO₂ was only ∼15%, but it increased to ∼40% at 2 h, indicating this anode material is performing like a nonselective anode. While the initial CE on Ti/Pb₃O₄ was high at ∼50%, nevertheless, it progressively dropped to around 10%, indicating Ti/Pb₃O₄ could be considered as a selective anode. In NaCl, due to the presence of the indirect oxidation, anode selectivity could not be distinguished. In 5 wt% NaCl, the CEs on both electrodes were volatile. When the chlorination became more dominant (in 1 wt% NaCl), the CEs on both electrodes decreased with time, indicating the insufficient oxidation ability of active chlorine and anode toward the possible chlorinated residuals.²⁴

3.2.5 Colority variation. During the EO treatment of guaiacol, the color of the solution varied with time. The guaiacol solution was initially colorless, then turned into brown, and ended up to be colorless again. The color change depends on the anode material, the supporting electrolyte, and reaction time, as shown in Fig. 4. In Na₂SO₄, the color of samples treated by Ti/Pb₃O₄ had more obvious color accumulation during the treatment, which is likely attributed to the accumulation of quinones.^3,25 Since quinones are the most common oxidized/degraded intermediates of phenol, the color variation can further justify that the Ti/Sb-SnO₂ and Ti/Pb₃O₄ is likely a nonselective and selective anode, respectively. However in NaCl, it was interesting to note that Ti/Sb-SnO₂ had an intensive color build-up initially, which was not observed for Ti/Pb₃O₄ at all. Since Ti/Sb-SnO₂ is not good at chlorine evolution as Ti/Pb₃O₄ does, it is apparently that the adsorbed chlorine or indirect oxidation generated by Ti/Pb₃O₄ may have an effective oxidation and/or bleaching effect on those colored intermediates. Another possibility is that some specific intermediates with high color may be generated by using Ti/Sb-SnO₂, but they can still be oxidized into colorless products in due course.

Fig. 4 Photos of the solution samples versus time during the treatment (in 5 wt% Na₂SO₄ or in 5 wt% NaCl).

In real paper/pulp wastewater treatment, the color removal is a critical issue. The above results indicate that the electrolyte, reaction time and anode selection are the three main factors to control the color of the effluent.

3.3 Aerobic biodegradability enhancement

During the EO treatment, the low biocompatible guaiacol is decomposed into various intermediates. Usually the intermediates may have higher biodegradability than guaiacol, for better cost-effectiveness, there is no need to remove COD thoroughly by EO. The EO treated effluent (to an appropriate level of treatment) could be send to the biotreatment (BT) plants in an EO–BT process. The BOD₅/COD is a simple but effective indicator to access the biodegradability. Generally if BOD₅/COD value is above 0.3, the solution is regarded as biodegradable. Fig. 5 illustrated the BOD₅/COD variations during EO treatment. In 5 wt% Na₂SO₄, Ti/Sb-SnO₂ exhibited better ability in biodegradability enhancement of the solution. The BOD₅/COD value increased from less than 0.1 to 0.3 in 2 h, and then further increased to 0.6 in 5 h. However Ti/Pb₃O₄ could not improve the biodegradability effectively, the BOD₅/COD could only reach to 0.2 after 5 h. Effluents with this low BOD₅/COD value could not be properly treated by the following biotreatment.

Fig. 5 Variations of BOD₅ and BOD₅/COD of guaiacol solution versus time during EO treatment by Ti/Sb-SnO₂ and Ti/Pb₃O₄ (5 wt% Na₂SO₄ or 5 wt% NaCl as the supporting electrolyte).

In chlorine containing media, it is necessary to be aware of the possible organochloride's effect on metabolism of the biomass. In 5 wt% NaCl, Ti/Pb₃O₄ could not improve the biodegradability as before. The performance of Ti/Sb-SnO₂ was unfortunately lowered as well, with a best BOD₅/COD of 0.4 at 1.5 h. Beyond this reaction time, the BOD₅/COD will be reduced. Therefore, Ti/Sb-SnO₂ is a proper anode material in EO–BT and Ti/Pb₃O₄ is not. The chlorine-free solution is more suitable to be treated by EO–BT combined process.

The above results could be attributed to the different components of intermediates under different conditions. BOD₅ could reflect the composition of the solution in some extent. The variations of BOD₅ are also shown in Fig. 5. In Na₂SO₄ the BOD₅ remained high by Ti/Sb-SnO₂. In NaCl the BOD₅ decreased all the way from the beginning to the end by both anodes. The identification of intermediates could be helpful to justify this.

3.4 Electric energy consumption in EO and EO–BT

The energy consumption (E) of EO technique is a general concern; the process optimization is necessary for application. Tables 1 and 2 listed the necessary energy consumption for the purpose of total COD removal and just elevating aerobic biodegradability, respectively. The calculation of E is shown as follows:where U is cell voltage, V; I is the current, A; t is treating time, s; V is solution volume, m³; ΔCOD is COD removal, g m⁻³. Several key observations could be drawn from the data. (1) The electrolyte concentration is an important factor for the energy saving. In general, the lower the ohmic-drop across the electrochemical cell (due to higher electrolyte concentration), the better the energy saving. (2) Since the total retention time of EO–BT was reduced, the energy consumption of biodegradability enhancement saved additional 50–70% energy compared to that of COD removal by using sole-EO. (3) Appropriate amount of Cl⁻ could save energy of COD removal and 5 wt% NaCl was found to be suitable in this study. (4) In chlorine containing media, it took more time to increase BOD/COD₅ value to above 0.3, making the energy saving of EO–BT less effective than that in chlorine-free media.

Table 1 Energy consumption (E) necessary for the COD removal and aerobic biodegradability enhancement of 500 ppm guaiacol solution at 20 mA cm⁻² on the Ti/Sb-SnO₂ electrode

Supporting electrolyte	COD			BOD5/COD
	Time/h	Removal/%	E/kW h m^−3	Time/h	Value	E/kW h m^−3
1 wt% Na2SO4	5.0	97.72	139.75	2.0	0.382	55.90
5 wt% Na2SO4	4.5	96.24	106.43	2.0	0.369	47.30
2 wt% Na2SO4	4.5	97.65	94.82	2.0	0.338	42.14
1 wt% NaCl	3.5	95.21	123.41	2.0	0.385	70.52
5 wt% NaCl	3.0	95.07	90.30	2.5	0.330	75.25
2 wt% NaCl	5.0	93.38	141.90	2.5	0.207	—
2 wt% Na2SO4
+5 wt% NaCl	3.0	97.84	58.05	—	—	—

---

Table 2 Energy consumption necessary for the COD removal and aerobic biodegradability enhancement of 500 ppm guaiacol solution at 20 mA cm⁻² on the Ti/Pb₃O₄ electrode

Supporting electrolyte	COD			BOD5/COD
	Time/h	Removal/%	E/kW h m^−3	Time/h	Value	E/kW h m^−3
1 wt% Na2SO4	6.0	89.95	203.82	—	—	—
5 wt% Na2SO4	7.0	91.70	201.67	—	—	—
2 wt% Na2SO4	7.0	93.40	180.60	—	—	—
1 wt% NaCl	4.5	93.91	164.48	2.5	0.294	91.38
5 wt% NaCl	3.0	94.34	96.75	2.5	0.251	80.63
2 wt% NaCl	5.5	91.02	151.36	3.0	0.166	—
2 wt% Na2SO4
+5 wt% NaCl	3.0	96.56	70.95	—	—	—

---

Ti/Sb-SnO₂ seems to be an energy saving electrode, especially in EO–BT combined process. Ti/Pb₃O₄ is slightly inferior in energy saving, but its electrode stability is one order of magnitude higher than Ti/Sb-SnO₂, especially in chlorine-containing media.¹⁶ Therefore, a balance between cost and efficiency must be fully considered in the actual application of these electrodes.

3.5 Degradation pathway

The identification of intermediates is very useful to reveal the degradation pathway and functional mechanism of the two anodes in different media. By GC-MS method (include SPDE technique), the trace amount of volatile intermediates could be gathered and detected. The main detected compounds are listed in Tables S1 and S2 (in ESI†). It should be noted that 1-chloro-2-methoxy-benzene, a major impurity, coexisted with the guaiacol before treatment. In 5 wt% Na₂SO₄, the guaiacol degradation pathway on both electrodes is initiated by forming the quinones. Then quinones are further oxidized to give ring-opening products, such as maleic acid. It should be noted that the oxidation kinetics of the two electrodes are different. Ti/Sb-SnO₂ exhibited a faster speed to degrade intermediates. After 5 h of treatment, the detected types of intermediates by Ti/Sb-SnO₂ were more than those of Ti/Pb₃O₄. The former also results in intermediates with smaller molecular weights, which are easier to be uptaken/utilized by biomass and therefore improved the biodegradability.

The reason to have a lower biodegradability enhancement in the presence of NaCl is exposed after examining Table S2† (the intermediates in NaCl). The aromatic organic halides (AOX) and small organic halides (RX) such as 2-chloro-maleic acid constituted part of intermediates. These substances with higher biotoxicity were reported to inhibit the metabolism of the biomass.^26,27

By LC-MS method, content of quinones (1,2-benzoquinone and 1,4-benzoquinone) in the treated solution versus time could be quantified (Fig. 6) through the integration of each MS peak. It was found that the quinones were accumulated during the first 2.5 h, and the level of accumulation depended on the anode material and electrolytes (i.e. NaCl > Na₂SO₄; Ti/Pb₃O₄ > Ti/Sb-SnO₂). This result is consistent with the above results involving the oxidizing selectivity of the anode towards guaiacol and its intermediates.

Fig. 6 Profiles of relative content of quinones and dimers during EO treatment.

Besides the quinones, the dimer structures of guaiacol were also detected, identified and quantified by LC-MS. Two types of dimers were identified (dimer 1: m/z 231; dimer 2: m/z 245, with an additional methyl than dimer 1). The dimers were found only in samples treated by Ti/Pb₃O₄ in Na₂SO₄ and samples treated by Ti/Sb-SnO₂ in NaCl. The discovery of electropolymerization is associated with the previous CV results (i.e. current increased when guaiacol was added). The formation of dimers may also contribute to coloring of the solution during the earlier stage of reaction (Fig. 3, such as Ti/Sn-SnO₂ in NaCl).

Combining the results of GC-MS and LC-MS, the proposed guaiacol degradation pathways were illustrated in Fig. 7. The final product CO₂ was confirmed by determination of total organic carbon (TOC). Samples' TOC decreased gradually from the beginning to the end during the EO treatment, and about 65% and 40% TOC were removed after 2.5 h EO treatment in NaCl and Na₂SO₄, respectively.

Fig. 7 Possible degradation pathway of guaiacol by EO treatment.

4. Conclusions

The types of anode and electrolyte determine the performance of EO in removing guaiacol and its intermediates, and resulting in the diversity of the results. The main findings of this investigation could be summarized as below.

(1) The presence of Cl⁻ could accelerate the degradation process but it can also hamper the following biodegradability in BT.

(2) EO–BT combined process could shorten the EO treating time and could save 50% or more electric energy compared with sole-EO process.

(3) Ti/Sb-SnO₂ is good at removing COD, which is non-selective in treating guaiacol and its intermediates. Ti/Pb₃O₄ is more selective in degrading guaiacol with a weaker ability in degrading the intermediates.

(4) The degradation mechanism of guaiacol consists of dimers formation, quinones formation and the following ring opening processes. The presence of Cl⁻ will produce undesirable aromatic organic halides (AOX) and small organic halides (RX).

Acknowledgements

The authors are grateful for the financial support of the research grant from the HK Polytechnic University (G-SB16).

References

1. K. Pan, M. Tian, Z.-H. Jiang, B. Kjartanson and A. Chen, Electrochim. Acta, 2012, 60, 147–153.
2. H. Xu, Q. S. Yuan, D. Shao, H. H. Yang, J. D. Liang, J. T. Feng and W. Yan, J. Hazard. Mater., 2015, 286, 509–516.
3. A. Dominguez-Ramos, R. Aldaco and A. Irabien, Ind. Eng. Chem. Res., 2008, 47, 9848–9853.
4. H. Lange, S. Decina and C. Crestini, Eur. Polym. J., 2013, 49, 1151–1173.
5. M. Tian, J. Wen, D. MacDonald, R. M. Asmussen and A. Chen, Electrochem. Commun., 2010, 12, 527–530.
6. M. T. Rasha Tolba, J. Wen, Z.-H. Jiang and A. Chen, J. Electroanal. Chem., 2010, 649, 9–15.
7. A. Dominguez-Ramos and A. Irabien, Ind. Eng. Chem. Res., 2013, 52, 7534–7540.
8. R. Pelegrini, J. Reyes, N. Durán, P. P. Zamora and A. R. de Andrade, J. Appl. Electrochem., 2000, 30, 953–958.
9. I. Oller, S. Malato and J. A. Sánchez-Pérez, Sci. Total Environ., 2011, 409, 4141–4166.
10. M. Tuomela, M. Vikman, A. Hatakka and M. Itävaara, Bioresour. Technol., 2000, 72, 169–183.
11. X. Cui, J. Liang and D. Wang, J. Chem. Technol. Biotechnol., 2015, 90, 747–753.
12. D. Shao, J. D. Liang, X. M. Cui, H. Xu and W. Yan, Chem. Eng. J., 2014, 244, 288–295.
13. C. Comninellis and G. Chen, Electrochemistry for the Environment, Springer Science+Business Media, 2010.
14. D. Shao, W. Yan, X. Li and H. Xu, ACS Sustainable Chem. Eng., 2015, 3, 1777–1785.
15. W. Y. Wu, Z. H. Huang and T. T. Lim, Appl. Catal., A, 2014, 480, 58–78.
16. D. Shao, W. Yan, L. Cao, X. L. Li and H. Xu, J. Hazard. Mater., 2014, 267, 238–244.
17. D. Shao, W. Yan, X. L. Li, H. H. Yang and H. Xu, Ind. Eng. Chem. Res., 2014, 53, 3898–3907.
18. C. A. Martinez-Huitle and S. Ferro, Chem. Soc. Rev., 2006, 35, 1324–1340.
19. M. Panizza and G. Cerisola, Chem. Rev., 2009, 109, 6541–6569.
20. C. A. Martínez-Huitle, A. de Battisti, S. Ferro, S. Reyna, M. N. Cerro-López and M. A. Quiro, Environ. Sci. Technol., 2008, 42, 6929–6935.
21. C. Comninellis and C. Pulgarin, J. Appl. Electrochem., 1993, 23, 108–112.
22. K. L. Rule, V. R. Ebbett and P. J. Vikesland, Environ. Sci. Technol., 2005, 39, 3176–3185.
23. S. Kim, S. K. Choi, B. Y. Yoon, S. K. Lim and H. Park, Appl. Catal., B, 2010, 97, 135–141.
24. M. Murugananthan, S. S. Latha, G. Bhaskar Raju and S. Yoshihara, Sep. Purif. Technol., 2011, 79, 56–62.
25. Z. Lu, B. Tu and F. Chen, J. Wood Chem. Technol., 2003, 23, 261–277.
26. S. Aquino Neto and A. R. de Andrade, J. Appl. Electrochem., 2009, 39, 1863–1870.
27. J. Jeong, C. Kim and J. Yoon, Water Res., 2009, 43, 895–901.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra23381j

---