3.1 soil pH
The H+ produced at the anode and the OH− produced at the cathode causes the pH of the electrolyte to decrease and increase respectively. The pH change of catholyte and anolyte during electrokinetic remediation were shown in Fig. 2, and it could be seen that, when distilled water, citric acid, and EDTA were used as electrolyte, the pH of anolyte decreased to 2.23, 1.66, and 3.16, respectively, and the catholyte increased to 11.6, 3.35, and 9.05, respectively. Soil pH changed in the electromigration of H+ and diffusion of electrolyte, and the results were shown in Fig. 3. As shown in Fig. 4, when deionized water was used as the electrolyte, the pH of S1 zone of ZR0 (original soil pH 7.73) decreased to 4.86, while S5 increased to 8.71; the pH of S1 zone of HR0 (original soil pH 5.47) decreased to 3.1 after remediation, while the pH of S5 increased to 5.82; the pH of the soil in S1 and S5 zones of BR0 (original soil pH 6.67) after remediation was less altered when compared to ZR0 and HR0. When citric acid was used as electrolyte, the pH of the S1 decreased to 3.12 after ZR1 remediation, and the pH of the S5 decreased slightly to 7.05 under the effect of free diffusion of catholyte. BR1 and HR1 soil pH both changed dramatically after remediation. In test of BR1, the soil pH of S1 decreased to 2.52, meanwhile, the soil pH of S2 ~ S4 decreased dramatically to 4 ~ 4.6, and there was almost no change in the pH of S5. When EDTA was used as the electrolyte, the pH of soil samples in S1 ~ S5 showed a small increasing trend after remediation in ZR2, HR2 and BR2 tests, among which, the pH of the soil in S2 ~ S4 area increased more in BR2. In general, the pH of soil sections from S2 to S4 in ZR did not show any significant changes, indicating that ZR had a strong buffering capacity and a small effect on soil pH during the 4d operating time; while in the BR experiment, soil pH was greatly influenced by electrolyte, and S2 to S4 changed under different restoration conditions; For HR test, due to its strong acidic nature, the soil pH was greatly affected by citric acid as the electrolyte, while it was less affected by EDTA.
3.3 Pb, Cd removal
The Cd residual rates after electrokinetic remediation were shown in Fig. 5. As shown in Fig. 5(a), the average residual rates of Cd in ZR0, HR0 and BR0 were 97.3%, 81.6% and 94%, respectively; among them, the lowest Cd residual rate was 44.9% in the S1 zone in HR0; as shown in Fig. 5(b), the residual rates from S1 to S5 in the experiment of HR1 showed a gradual increase, with an average residual rate of 66.2%, and the Cd residual rate in the S1 zone (pH 2.8) only reached 17.9%; compared with the blank control group, the soil residue rate in BR1, ZR1 and HR1 decreased significantly, especially in the S1 zone. It could be seen that when deionized water and citric acid were used as electrolytes, soil pH was an important factor affecting the Cd removal rate in ZR, HR and BR, and the lower the soil pH was, the higher the Cd removal rate was. The average residual rate of Cd in ZR2, HR2, and BR2 was 82.%, 82.0%, and 78.8%, respectively, which indicated that the background value of soil pH had less influence on the Cd removal rate when EDTA was used as electrolyte. After electrokinetic remediation, the Cd residual rate in the soil showed a trend of "low at both ends and high in the middle", and there was a heavy metal aggregation phenomenon in the S3 section, especially, the Cd residual rate in the S5 and S4 sections in the BR2 was 2.4% and 16.8%, which was much lower than that in the ZR2 and HR2, which showed that when EDTA was used as the electrolyte, anionic complexes with Cd in BR more easily formed (e.g., [Cd-EDTA]2−) under the effect of electromigration to anode[12, 13].
The Pb removal rates after electrokinetic remediation were shown in Fig. 6. As seen in Fig. 6(a), the lowest value of Pb residual rate in HR0 occurred in the S1 zone (pH = 3.1) with 84.1%, while the average Pb residual rate in the S2 ~ S5 zone was as high as 99.5%. As shown in Fig. 6(b), comparing ZR1 and BR1, the average residual rate of Pb in HR1 was the lowest at 80.0%, and the removal of Pb in S1 (pH = 2.82), S2 (pH = 3.28), S4 (pH = 5.50), and S5 (pH = 5.80) zones were 60.9%, 82.1%, 84.3%, and 84.5%, respectively, with a focusing in S3 section phenomenon, indicating that citric acid and Pb form anionic complexes (e.g., Pb-(Citrate)24−) that migrated toward the anode under the action of electric field. Researches shown that Pb was desorbed from the soil particle surface only when the soil pH was < 4[25], which explains the low removal of Pb in ZR0, ZR1, BR0 and BR1 after 4 d of electrokinetic remediation. As could be seen in Fig. 6(c), when EDTA was used as the remediation solution, ZR2 and BR2 showed a focusing phenomenon in S3, and HR2 showed a focusing phenomenon at S1, and both had the lowest residual rate in S5, followed by the S4, and this phenomenon was especially prominent in BR2, indicating that when the soil pH was higher than 5.4, it was mainly through the formation of complexes with anionic with EDTA ( such as [Pb-EDTA]2−) to remove Pb[13].
In comparison, it was found that the removal of Pb and Cd in ZR were lower than those in HR and BR. There were several reasons to explain this phenomenon: firstly, the change of soil pH during electrokinetic remediation caused a series of physicochemical changes, such as precipitation/dissolution of minerals, adsorption/desorption of pollutants, and ion exchange between the solid and liquid phases of the soil[26]. Among the three soils, ZR has the largest pH background value of 7.73, which is alkaline (Table 1), with a strong soil buffering capacity and no significant change in soil pH after remediation (Fig. 4), and the high pH value made the desorption of heavy metals from soil particles longer; secondly, ZR had a high organic matter content, which was dominated metal binding and generated insoluble complexes with heavy metals through electrostatic adsorption or complexation (chelation)[17]; finally, the combined effect of soil pH and organic matter content resulted in the firm adsorption of heavy metals Pb-Cd in the soil compared to either single factor[27]. In addition, factors such as soil cation exchange capacity, mucilage content, and redox status also affect the heavy metals removal. In conclusion, the strong buffering capacity of ZR soil, the slow change of soil pH, and the strong binding ability of soil particles with heavy metals Pb-Cd made the electrokinetic remediation ineffective, and it was recommended to combine with other remediation techniques (e.g., leaching[28, 29], PRB[4, 30], chemical amendment[5, 31], etc.) to improve the remediation efficiency.
Time played a crucial role in the migration of substances in the soil and the desorption/dissolution of contaminants[32]. Extending the remediation time of HR and BR tests, the removal of Cd and Pb in the soil after 10 d of remediation was shown in Fig. 7 and Fig. 8. As seen in Fig. 7: extending the remediation time could significantly improve the removal of Cd; the average removal rate of Cd in HR1 was 91.1%, the highest removal rate was in the S1 area, 99.5% and the lowest removal rate was in S5 area, 71.2%,, and at same time, the pH values of S1 ~ S5 zones were lower than 3.5. The pH values of S1 ~ S5 zones in BR1 were 2.21, 2.81, 3.32, 5.54, and 7.98, and the removal of S1 ~ S3 zones were 99%, 90%, and 66%, respectively, whereas the focusing phenomenon appeared in the S4 and S5 zones, which resulted in an average Cd removal rate of only 22%. This was because with the prolongation of time, citrate ions in the cathodic liquid were neutralized by OH− produced by electrolysis, and OH− entered into the soil aqueous solution under the sway of electromigration, which made heavy metals and other metal ions meet OH− to form a precipitation (the realization of which revealed that there was a slate phenomenon in the soil of S5 zone with white precipitates), which could be used to increase the removal rate of heavy metals through the cathodic control method. The average removal rate of Cd in both HR2 and BR2 was 44%, and the migration pattern of Cd in both soils was consistent, with the removal rates in the order of S5 > S4 > S3 > S1 > S2. Cd in the soil of S1 zone migrated to the cathode in the form of Cd2+ at low pH (< 4.5), while in S5, S4 and S3 zones migrated to the anode as a complex [Cd-EDTA]2− at high pH, and focusing in the S2 region. Therefore, the method of controlling the anode pH could be used to prevent the focusing phenomenon and improve removal. In conclusion, low soil pH could significantly increase the removal of Cd, while EDTA could also promote the desorption of Cd in high pH soil, and the method of controlling the pH value of anolyte could be used to improve its removal rate.
The average removal rate of Pb in HR1 and BR1 were 40% and 20%, respectively, and the highest removal appeared in S1 section, which were 84.8% and 59.6%, respectively, while the lowest removal appeared in S5 (5%) and S3 section (7.6%), which were mainly affected by the pH value and organic matter content of the soil, and the existence of different morphologies of Pb in the soil and its complexation with electrolytic solution led to the differences in its migration pattern. The removal rate of Pb in HR1 was much lower than that of Cd (91.1%), which was due to the competitive adsorption capacity of Cd in soil was poor, and it was easy to be replaced and migrated into the soil aqueous solution and be removed; when EDTA was used as the electrolyte, the average value of the removal rate of Pb in HR2 was 63%, and the rate of the removal rate of the S5 to S1 section decreased in order, and the change of the pH value of soil was small compared with that of HR1 at this time. The average value of Pb removal rate in BR2 was 42%, and the aggregation effect appeared in S1 area, and the overall removal rate was lower than that of HR, which was because the pH value of BR soil was higher compared with that of HR, and its organic matter content was nearly three times of that of HR, and the heavy metal Pb formed insoluble chelates with the organic matter, which was more difficult to be desorbed from the surface of the soil particles.
In comparison, it was found that citric acid and EDTA as electrolytes have different mechanisms of action on heavy metals Pb and Cd. Citric acid accelerated soil acidification and desorbed heavy metals from the surface of soil particles, or formed negatively charged complexes with metal ions, which was more effective than Pb in the removal of Cd from the soil, while EDTA mainly formed negatively charged complexes by complexing with heavy metals, which was more effective than Cd in the removal of Pb from the soil. It was further demonstrated that EDTA formed anionic complexes (e.g. [Pb-EDTA]2−) to remove heavy metal Pb.