Electrochemical Degradation of Phenol and Resorcinol Molecules through the Dissolution of Sacrificial Anodes of Macro-Corrosion Galvanic Cells

This paper reports on the processes of phenol and resorcinol electrodegradation carried-out through continuous anodic dissolution of aluminum alloy and carbon steel sacrificial anodes for artificially aerated Cu-Al alloy and Cu-Fe-based galvanic (macro-corrosion) cells and synthetically prepared wastewater solutions. Electrochemical experiments were carried-out by means of a laboratory size, PMMA (Poly-methyl methacrylate)-made electrolyser unit, where significant degrees of phenol (10–89%) and resorcinol (13–37%) decomposition were obtained and visualized through the respective chemical/spectroscopy analyses. In addition, quantitative determination of phenol, as well as resorcinol (and possible electrodegradation products) for the selected experimental conditions was performed by means of instrumental high-performance liquid chromatography/mass spectrometry analysis.


Introduction
Phenolic compounds are widely generated side-products by a number of important industries, including: petrochemical, pharmaceutical and thermoplastic processing plants. Typical oil refinery wastewater could contain phenol in range of 500-1500 mg dm −3 . Phenol and its derivatives (including resorcinol) are highly toxic environmental pollutants [1][2][3][4][5][6], which exhibit low biodegradability; they are also known as being severely toxic to human beings and aquatic life [7][8][9][10]. Electrochemical oxidation processes are very attractive methods for the degradation of phenolic compounds. Some previously published works on this topic covered oxidation of numerous phenolic compounds on a number of noble/semi-noble electrode materials (i.e., Pt, Ir, Ru), as well as on transitional metals and their oxides (Ti, Sb, Sn, Pb), and various forms of carbon materials (glassy carbon, carbon fibre, graphite and boron-doped diamond electrodes), see Refs. [11,12], and articles quoted there for details. Some other important papers published on this topic explored the removal of phenol by means of the dc-driven electrocoagulation process, carried-out through anodic dissolution of Al and Fe electrodes [13][14][15][16][17][18][19]. The electrochemical coagulation method has significant benefits over a conventional, chemical coagulation process, as it allows for full process control through the adjustment of basic electrical (current and voltage) parameters. However, the necessity for application of very expensive, dc (or dc pulse) power supplies is its major disadvantage. Also, employment of anodes made practically of pure aluminum [13,16,18,19] substantially increases electrical energy consumption, due to the effect of Al surface passivation. Therefore, this work is concerned with the removal of phenol and resorcinol molecules from synthetically-prepared wastewater solutions through the process of Al-alloy and Fe-mild steel self-electrocoagulation, realized in aerated, macro-corrosion galvanic cells (with no need for a dc power supply). Here, aluminum alloy and iron electrodes played the role of sacrificial anodes, whereas copper plates were employed as cathodes in order to make a galvanic cell operational [20,21]. Electrooxidation trials were performed for selected synthetic wastewater compositions. Monitoring of phenol/resorcinol decomposition processes (along with a quantitative evaluation of possible reaction products) was performed by means of commonly used, instrumental HPLC/MS analysis [22][23][24], in function of electrolysis time.

Materials and Methods
A PMMA (Poly-methyl methacrylate)-constructed wastewater electrolysis cell with ca. 750 mL of working volume was used in this work (see its schematic representation in Figure 1 below). The cell contained a single sacrificial anode in the centre [Al alloy (surface area: S Al~3 8 cm 2 ; cylinder; diameter: φ = 4 cm, thickness: d = 1 cm): 3.0-5.5 wt.% Zn + 0.016-0.040 wt.% In (MAKROMOR, Gdansk, Poland) or mild steel 1.0332 anode (S Fe~5 6 cm 2 ; plate: 5 × 5 × 0.3 cm)] and two copper cathodes (S Cu~8 4 cm 2 ; two plates: 4 × 5 × 0.1 cm) located at two sides of the electrolyser, with an anode-to-cathode distance of about 20 mm. Prior to being positioned in the cell, both sacrificial anodes were first polished with emery paper down to 2000 grade; then, they were activated for 5 min in 5% HCl solution before being de-greased with acetone and rinsed with ultra-pure water (18.2 MΩ cm Millipore Q3 UV water purification system was used). On the other hand, Cu cathodes were activated in 5% HCl and thoroughly rinsed with ultra-pure water. The electrolyser was operated at room temperature in a wastewater recirculation mode (at a rate of ca. 20 mL min −1 ), carried-out by means of Gilson Minipuls 3 peristaltic pump. Anodic dissolution of Al alloy (Fe electrode) was possible due to external connection of electrodes in order to make respective galvanic cells.
Open-circuit voltages and electrode potentials (recorded vs. SCE: saturated calomel electrode) were always measured prior to and after completion of the electrocoagulation tests. On the other hand, intermittent measurements of corrosion potentials (vs. SCE) and galvanic current readings for aerated (by slow purging of compressed air) cells were conducted every 60 min during the trials. Synthetic wastewater solutions were prepared by dissolving Na 2 SO 4 (electrolyte conductivity: κ = 5.0 and 10.0 mS cm −1 , pH = 7: re-adjusted and HCl-adjusted acidic pH of 3, and t = 20 • C; Polish Chemical Compounds, p.a.) using Millipore ultra-pure water source with initial phenol [Ph] (Sigma-Aldrich, >99%, Poznan, Poland) and resorcinol [Res] (Polish Chemical Compounds, p.a.) concentrations on the order of 1 × 10 −4 and 1 × 10 −3 M. Although the selected concentrations of phenol and resorcinol are indeed quite high, one cannot exclude the possibility of limited phenol discharge to environment at the concentration of 1 × 10 −4 M (or even higher) by a local company. On the other hand, the choice of Na 2 SO 4 salt base (providing the necessary wastewater conductivity for galvanic cells' operation) in this work was primarily (in addition to simplicity) to avoid possible chlorination effects. Initial operation of Cu/Al alloy and Cu/Fe galvanic cells was also tested under analogous experimental conditions, but in the absence of phenol (resorcinol) in the electrolyte by means of the Solartron 12.608 W Full Electrochemical System (ZRA mode). Otherwise, all electrical measurements were conducted with Fluke 85 III true rms multimeter, whereas O 2 content, electrolyte conductivity and pH evaluations were carried-out with HI 9146, HI 9835, and HI 2002-01 meters from Hanna Instruments, respectively.
A major analysis of phenol content in wastewater was carried-out spectro-photometrically with 4-aminoantipyrine [25,26], where for every studied object two independent wastewater samples (5 and 10 mL) were collected for initial solution and after 6, and 12 h of the electrolyser's continuous operation. On the other hand, respective resorcinol analysis was carried-out based on the bromometric determination titration method [27,28].
In addition, both phenol and resorcinol electrodegradation processes were subjected to a quantitative assessment of the reaction products/intermediates by means of the combined HPLC/MS analysis, carried-out for both galvanic cells (Cu/Al alloy and Cu/Fe), at an initial concentration of

Characterization of Macro-Corrosion, Galvanic Cells' Operation
Initially, operation of Cu/Al alloy (in comparison to Cu/Al pure: 6063-T6 alloy) and Cu/Fe galvanic cells was examined in aerated, Na 2 SO 4 -based electrolyte (pH = 7, κ = 5.0 mS cm −1 , t = 20 • C and ca. 10.5 ppm of dissolved oxygen). Hence, after about half an hour, the Cu/Al alloy, the Cu/Al pure and the Cu/Fe galvanic cells exhibited the following steady values of open circuit voltage: 1.034 V, 0.537 V and 0.557 V, correspondingly ( Figure 2a).
Then, the corrosion behavior of the respective galvanic couplings is shown in Figure 2b. Hence, the galvanic couple built by shortening copper cathode with aluminum alloy anode exhibited the largest value of galvanic current (I gc ) of about 55 mA (1.46 mA cm −2 ) after 1 h of exposure time. On the other hand, under analogous experimental conditions, about 28 mA (0.50 mA cm −2 ) and only 2.7 mA (0.12 mA cm −2 ) were recorded for the Cu/Fe and the Cu/Al pure galvanic couples, correspondingly. In all cases, an anodic process corresponds to dissolution of the so-called sacrificial anode (Al or Fe electrode, see Equations (1)-(4) and Figure 2c below), whereas radically inhibited oxidation of pure aluminum anode results from the fact that Al (unlike some of its alloys) is highly-susceptible to the surface formation of thick, electrically insulating and exceptionally corrosion resistant oxide layer.
Hence, cations of the so-called "coagulating agent" (Al or Fe) are released in anodic oxidation process (Equation (1)), which is accompanied by cathodic hydrogen evolution or predominantly by oxygen reduction reaction on the copper electrode surface (Equation (2)), leading to a gradual increase of the solution's pH value. The latter finally results in the formation of aluminum or iron hydroxide (eventually Fe oxide) species, which is shown in Equations (3) and (4) above. Also, prolonged exposure of the galvanic cells under experimental conditions results in a radical modification of the sacrificial anode's surface structure (Figure 2c).

Electrodegradation of Phenol and Resorcinol Molecules
Figure 3a presents comparative efficiency of phenol removal obtained on aluminum alloy and mild steel sacrificial anodes, respectively. Hence, under a baseline set of working parameters ([Ph] = 1 × 10 −4 M, κ = 5.0 mS cm −1 and pH = 7), the copper/aluminum alloy and copper/iron galvanic cells allowed for the removal of 33% and 41% of phenol after 12 h of continuous operation of the electrolysis cell. Then, for a radically increased electrolyte conductivity (κ = 10.0 mS cm −1 ) and under acidic pH of 3, the efficiency of phenol removal was raised to 53% and 89% for the Cu/Al alloy and the Cu/Fe system, respectively.
Finally, for a 10-fold increase in initial concentration of phenol (1 × 10 −3 M), its removal efficiency significantly dropped to reach only about 10% and 21% for these cells, correspondingly. As high concentrations of phenol in the electrolyzed wastewater are strongly detrimental to the efficiency of its electrooxidation, it is also worth noticing in Figure 3a that the process is quite vigorous only during initial hours of its operation. The above could most likely be related to surface electrosorptive nature of phenol molecule and its oxidation intermediates. It should also be stressed here that the results obtained in this work are generally in good agreement with those recorded by other authors in Refs. [13][14][15][16] on the removal of phenolic compounds through the employment of direct current-driven dissolution of Al (or Fe) electrodes, i.e., the so-called electrocoagulation processes. As in this work the process of phenol removal was significantly facilitated with iron electrocoagulant, it could then be assumed that in contrast to Al electrode, Fe provides considerably greater surface area for the resulting flocs.
On the other hand, Figure 3b illustrates comparative efficiency of resorcinol removal from synthetically-prepared wastewater, again for aluminum alloy and mild steel anodes. Thus, for the base conditions ([Res] = 1 × 10 −4 M, κ = 5.0 mS cm −1 and pH = 7), the copper/aluminum alloy and copper/iron galvanic cells exhibited the resorcinol removal levels on the order of 13% and 27% just after 6 h of continuous electrolysis. Then, under a double value of the electrolyte conductivity, the efficiency of resorcinol removal somewhat increased to reach 17% and 30% for the Cu/Al alloy and the Cu/Fe cell, correspondingly. Interestingly, for a 10-fold increase of initial concentration of resorcinol (1 × 10 −3 M), contrary to what was observed for phenol, the removal efficiency of resorcinol significantly increased, reaching very similar values for both examined systems, namely: 35% and 37% for the Cu/Al alloy and the Cu/Fe cells, respectively. The latter effect might be related to the fact that due to the presence of two hydroxyl groups in the meta position, the process of resorcinol electrodegradation could proceed considerably faster than that of phenol under analogous experimental conditions (e.g., see Ref. [29] for details).
Finally, Table 1 presents ranges of open-circuit voltage (V oc ) and electrode (cathode and anode) potential (E 0 ) along with corrosion potential (E cor ) and galvanic current-intensity (I gc ) values for shorted cells, recorded during phenol and resorcinol electrodegradation experiments. Hence, the Cu/Al alloy galvanic cell exhibited V oc values on the order of 1.057-1.195 V, while open-circuit voltages of the Cu/Fe cell were limited to the range 0.612-0.650 V, which is governed by (comparatively to Fe) very anodic potential of the aluminum sacrificial anode (shown in Table 1 vs. SCE). As a result, the Cu/Al alloy couple exhibited considerably larger values of galvanic current (42-56 mA), as compared to 18-28 mA demonstrated by the Cu/Fe galvanic couple. In fact, rather qualitative nature of the recorded galvanic current parameter did not allow for the derivation of any electrical energy or electrode consumption values. In addition, estimated corrosion (mixed) potentials for both galvanic couples oscillated around the recorded open-circuit anode potentials (Table 1), which proved that corrosion processes proceeded with cathodic control governed by the kinetics of cathodic oxygen reduction or hydrogen evolution reactions (Equation (2)). Table 1. Recorded ranges of electrical parameters (open-circuit voltage and electrode potential, mixed-corrosion potential and galvanic current-intensity) for Cu/Al alloy and Cu/Fe galvanic couples, obtained upon phenol and resorcinol electrodegradation experiments.

Phenol and Resorcinol Electrodegradation by HPLC/MS (High Performance Liquid Chromatography/Mass Spectrometry) Analysis
Furthermore, the combined HPLC/MS analysis (important with respect to derivation of the mechanisms for phenol and resorcinol electrodegradation, based on instrumental identification of the reactions by-products) was carried-out on four freshly (within 24 h) pre-electrolyzed wastewater samples (having removed all Fe III and Al hydroxide precipitates), for the following parameters: It should be noted however that the electrodegradation conditions to get the samples for HPLC/MS analysis were not much different from those of Figure 3. Thus, phenol analysis (Samples 1 and 2) resembles that of baseline conditions in Figure 3a (data series: 1). On the other hand, resorcinol analysis (Samples 3 and 4) is similar to that of data series 3 in Figure 3b. Also, a time variation should not be significantly (except for a typical fluctuation) affecting electrolyte's final composition, as: (a) for phenol electrodegradation: 8-h electrolysis time fits plateaus observed in Figure 3a, and (b) for resorcinol electrodegradation: 5-h electrolysis time is close to the maximum recorded analysis time in Figure 3b.
Thus, both phenol, as well as resorcinol electrooxidation on the Al alloy (or Fe) electrode surface (for sulphate-based synthetic wastewater) resulted in lack of ring-based degradation intermediates (see Figure 4a,b, and Table 2 below). In fact, the only quantitatively identified product for these processes was maleic acid, estimated at 8.0 × 10 −7 M for electrodegradation of resorcinol on the surface of iron electrode (Sample 4). Moreover, instrumentally derived (Table 2) concentrations of phenol (Samples 1 and 2) and resorcinol (Samples 3 and 4) could be translated to about 18%, 33%, 11% and 44% of removed phenol (resorcinol), correspondingly. These numbers are in qualitative agreement with those recorded above by the bromometric titration method.
Hence, the most important information obtained from the HPLC/MS instrumental analysis is that both phenol and resorcinol electrooxidation processes, performed through the dissolution of sacrificial Al alloy (or Fe) anode material, proceeded with a total destruction of the aromatic ring structure, where maleic and oxalic acids, and CO 2 are supposedly their key reaction products (see a tentative degradation pathway for phenol and resorcinol molecules in Figure 5 below).

Conclusions
Processes of phenol and resorcinol electrodegradation could successfully be performed through electrocoagulation with Al and Fe ions, produced from sacrificial anodes for aerated, Cu/Al alloy and Cu/Fe galvanic couples. A key advantage of this process over a typical dc-driven Al/Fe electrocoagulation is that its operation is purely based on the theory of a galvanic (corrosion) cell, where dissolution rate of anode is controlled by open-circuit potential difference between the cell's constituents, as well as the presence of a depolarizing agent (typically dissolved oxygen). Here, there is no necessity for expensive, dc rectifier infrastructure and control system. Aluminum alloy sacrificial anode is not susceptible to surface oxidation (passivation) phenomenon, which makes it superior in this application to pure aluminum. Furthermore, an advantage of iron electrode over aluminum is most likely related to considerably greater surface area of the resulting flocs for the former one.
Finally, performed HPLC/MS instrumental analysis implied that both phenol and resorcinol electrooxidation continued with total destruction of the aromatic ring structure, where maleic and oxalic acids, and carbon dioxide species would be the main reaction products. As laboratory size experiments proved general suitability of this electrochemical setup for the degradation of selected phenolic compounds, further work (including optimization of major process parameters: pH, operating time and temperature) should be carried-out on a much bigger, technical scale system, which would allow for the purification of at least 1 m 3 of wastewater per 24 h of continuous operation.