Figure 3 shows the proposed reaction pathway for the electrochemical oxidation of phenol for the conditions being studied. Two main pathways are followed during degradation: direct/indirect oxidation, also known as the quinone pathway; and electrophilic aromatic halogenation, also known as the chlorophenol pathway. The quinone pathway is similar to previously published pathways 5, 7, 8, 11. Two main mechanisms are at work to degrade phenol through the oxidative pathway: direct oxidation at the electrode surface and oxidation by hydroxyl radicals (). Equation 4 shows the reaction for the production of hydroxyl radicals.
Figure 3. Proposed reaction pathway for electrochemical oxidation of phenol. Compounds in boxes are known to be products but have not been confirmed by experimentation. *Present only when using an NaCl electrolyte. [Color figure can be seen in the online version of this article, available at wileyonlinelibrary.com.]
Download figure to PowerPoint
Hydroxyl radicals are adsorbed to the surface of the electrode or can diffuse into the bulk solution around the electrode; however, they are very short lived because of their reactivity 11. These radicals can directly oxidize the organic pollutants into intermediate products. Hydroxyl radicals can also react with each other to form water and oxygen, shown in Equation 5.
Direct oxidation at the electrode is also known as electrochemical cold combustion 7, 22. Organics that diffuse to the electrode surface can be converted directly to carbon dioxide and water.
Good agreement exists among researchers with respect to what products are produced along the oxidative pathway of phenol degradation, regardless of the type of electrode used. Considering only the oxidative pathway, it seems that the electrode type does not impact the specific products but only their distribution and rates of production and degradation. Because of this observation, it is possible to produce a general reaction pathway for the direct electrochemical oxidation of phenol, and that can be seen in Figure 3.
The phenol's hydroxyl group activates the ortho and para positions for oxidation by free chlorine 23. As long as free chlorine is present in the bulk solution, the degree of chlorination will continue to increase until eventually the chlorinated phenols are oxidized by the breaking of the phenol ring. The chlorophenol pathway can be seen in Figure 3.
An interesting note is that only the ortho and para positions on the phenol ring are eligible for substitution. Because of this, tetra- and pentachlorophenol are extremely unlikely to be formed. Published work on chlorination of phenols showed that the tendency for rupture of the phenol ring increases with an increase in chlorination, with 2,4,6-TCP being the highest chlorinated phenol possible 24. The data in the present study support this observation. The reactor was run beyond the point at which all of the lower chlorinated phenols had been removed completely and the free chlorine level started to increase rapidly. No PCP was found in the effluent. More experimentation may be necessary to rule out PCP as a potential product, but, under the conditions of the present study, no production of PCP was observed.
In the case of either pathway of degradation, the ideal end products are carbon dioxide and water. One of the main aspects of the oxidation is the rupture of the phenol ring. Cleavage of the phenol ring converts the intermediates into organic acids, which are easily biodegradable into carbon dioxide and water.
Electrochemistry with sodium sulfate electrolyte
When a graphite anode was used and no chloride was present in the electrolyte matrix, as shown in Figure 4, the measured toxicity increased rapidly once the electrochemistry began. The only compound produced in significant quantities under these conditions was BQ. Benzoquinone is, by a wide margin, the most toxic compound observed in the present study and, according to Maluleke and Linkov 25, is considered to be one of the most toxic xenobiotics. With BQ being much more toxic than the other degradation products, it controls the toxicity of the effluent. If the contribution of an individual compound to the effluent can be determined using Equation 3, BQ alone contributes 98.5 of the 100.4 TU present in the effluent, or 98%, starting at 518 coulombs (72 min). From this time until the end of the experiment, BQ makes up more than 99% of the total TU present. Prior research by Olmstead and LeBlanc 26 in the area of mixture toxicity with Daphnia magna showed a similar effect when one compound (chlorpyrifos) was much more toxic than the others.
Figure 4. Degradation and toxicity evolution of effluent from treatment with graphite/stainless steel and Na2SO4 electrolyte. Sum represents the millimolar sum of all measured compounds. Limits of quantification for phenol, benzoquinone, hydroquinone, and catechol were 0.006, 0.004, 0.002, and 0.001 mM.
Download figure to PowerPoint
It is also interesting to note that the predicted toxicity aligns well with the measured toxicity with respect to the data shown in Figure 4. This alignment is due in part to the presence of a toxicity-controlling component such as BQ. This alignment also supports the premise that the quantified compounds, BQ, HQ, phenol, and catechol, are the major representative compounds responsible for effluent toxicity. Although organic acids and other ring cleavage products were not quanitified in the present study, their contribution to toxicity in this organism appears to be minimal.
Figure 5 shows the difference in toxicity and composition change when using a BDD anode as opposed to graphite. All other experimental conditions were the same except for the size and shape of the electrodes. Similarly to the graphite experiment from Figure 4, an increase in toxicity was due mostly to the formation of BQ as an intermediate. However, intermediates were formed at lower concentrations when using a BDD anode than with the graphite anode. Note that the concentrations of intermediates shown in Figure 5 are below the limits of quantification of the HPLC method; therefore, they are not accurate but do show the general trend. This difference is most important from an effluent toxicity standpoint. As stated earlier, BQ controls the toxicity, and reducing BQ production has a significant impact on toxicity reduction. Boron-doped diamond likely produces less BQ than graphite because of the increased oxidation efficiency with BDD. With the greater oxidation efficiency, intermediates are produced in much lower quantities. It should be noted that differing electrode shape and current density between the graphite and BDD experiments might also have an effect on the results, but these effects are believed to be minimal. At the final sampling time of the experiment, the same number of coulombs for the experiments shown in Figures 4 and 5, there were approximately seven times fewer TU with BDD than with graphite. In comparing the two different anode types, it can be seen that BDD is the better choice from the standpoint of toxicity reduction. Boron-doped diamond is already known to be a superior electrode material from a reactivity standpoint 11, 13, and these findings only confirm that these benefits lead to an overall reduction in effluent toxicity compared with graphite.
Figure 5. Degradation and toxicity evolution of effluent from treatment with boron-doped diamond/stainless steel and Na2SO4 electrolyte. Sum represents the millimolar sum of all measured compounds. Limits of quantification for phenol, benzoquinone, and catechol were 0.004, 0.009, and 0.003 mM.
Download figure to PowerPoint
This finding has an impact on regulating and managing toxic effluents. Although regulating all constituents of an effluent is not possible or practical, more consideration should be given to those compounds that control toxicity. It has been shown that changing electrode type can affect the composition of the effluent, meaning that it is possible to manipulate the reaction conditions to favor a lower-toxicity product. Another point for consideration is the environmental persistence of the compounds present in the effluent. Although a certain compound may be extremely toxic, the length of time it will exert that toxicity is significant. From the results of the present study, in the case of a nonreactive electrolyte such as sodium sulfate, BDD produces a lower-toxicity effluent if the reaction is to be stopped before complete mineralization of the pollutants. With respect to the Microtox test organism, BDD is the better electrode type compared with graphite because of the production of a lower-toxicity effluent.
Electrochemistry with sodium chloride electrolyte
Figure 6 shows the degradation of phenol and the toxicity change of an effluent from a BDD/stainless-steel system using a chloride electrolyte. In this reaction situation, no compounds from the quinone pathway were observed, likely because of the high oxidizing power of BDD combined with free chlorine. The distribution of toxicity in the effluent is relatively balanced. No one compound had direct control over the effluent toxicity in the same way as BQ did when it was present. 4-Chlorophenol and 2,4-dichlorophenol make up a majority of the effluent toxicity from 259 to 777 coulombs. It can be seen in Figure 6 that at 518 coulombs the predicted and measured toxicities are very similar. When the predicted toxicity is the same as the measured toxicity, it is likely that a majority of the compounds significant to the effluent toxicity have been quantified. At 1,296 and 1,815 coulombs, however, there is a discrepancy between the predicted and measured toxicities. This discrepancy indicates that another compound is present and is having a significant effect on the effluent toxicity, one that has not been identified or quantified. It is likely that this unidentified compound is a di- or trichlorinated aromatic compound. Onodera et al. 27 suggested that, after the formation of 2,4,6-TCP, potential exists for the production of 2,6-dichloro-p-benzoquinone; this hypothesis will be investigated in the future. Nonetheless, this unidentified compound is removed by the end of the experiment, and the predicted and measured toxicities reconverge. It should be noted that in Figure 6 the last data point for TUmeas is a worst-case scenario. The maximum concentration that can be tested using the basic Microtox test protocol is 45%, corresponding to an IC50 of 45 and 2.222 TU. The sample actually contains less than 2.222 TU, but the level cannot be accurately determined using the basic test protocol. Therefore, for the Microtox basic test protocol, 2.222 TU acts similarly to a limit of quantification when measuring the toxicity of an effluent as IC50. The limit of quantification is shown in Figure 6 as a dashed line.
Figure 6. Degradation and toxicity evolution of effluent from treatment with boron-doped diamond/stainless steel and NaCl electrolyte. Sum is the millimolar sum of all measured compounds. Error bars indicate the range of the data. Limits of quantification for phenol, 2-chlorophenol, 4-chlorophenol, 2,6-dichlorophenol, 2,4-dichlorophenol, and 2,4,6-trichlorophenol were 0.006, 0.006, 0.002, 0.004, 0.002, and 0.002 mM.
Download figure to PowerPoint
The comparison between electrochemistry using a sodium chloride electrolyte versus sodium sulfate is important. In terms of composition, the oxidative intermediates such as BQ, HQ, and catechol were not found when using sodium chloride. This absence has a significant effect on toxicity, because the chlorinated phenols are significantly less toxic than BQ. The BDD and sodium chloride system produces an effluent that, at the most toxic point, is one-third as toxic as BDD and sodium sulfate and 1/25 as toxic as graphite and sodium sulfate. From a strict toxicity standpoint, a clear benefit exists with treatment with BDD and sodium chloride. The problem with this conclusion is that chlorophenols are environmentally persistent 28, and, if the reaction is not taken to completion, there is a risk of adding persistent toxicity to the effluent. Electrochemistry with sodium chloride can also produce unwanted byproducts such as chlorate, and the present study shows that significant amounts of chlorate were produced. Chlorate has a very low Microtox toxicity relative to other potential products, with an EC50 determined to be greater than 11.98 mM (1,000 mg/L). However, chlorate has been found to be a potential carcinogen in mice 29, so it should be regarded with some caution; it is also environmentally persistent. These unwanted products are not produced during electrochemistry with sodium sulfate. However, chloride may naturally be present in an effluent requiring treatment. These concerns bring up the importance of using more than just toxicity readings to judge and compare effluents. Consideration must be given to the environmental persistence and degradability of an effluent as well as the potential for production of unwanted byproducts. As important as the effluent's toxicity is the length of time for which toxicity will be present in the environment. Further study is recommended to analyze the toxicity and persistence of compounds once they are released into the environment. Consideration of toxicity should go beyond the point of discharge.