- Email: [email protected]
Degradation of trichloroethene by nanoscale zero-valent iron (nZVI) and nZVI activated persulfate in the absence and presence of EDTA
Chemical Engineering Journal 316 (2017) 410–418
Contents lists available at ScienceDirect
Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej
Degradation of trichloroethene by nanoscale zero-valent iron (nZVI) and nZVI activated persulfate in the absence and presence of EDTA Haoran Dong ⇑, Qi He, Guangming Zeng, Lin Tang, Lihua Zhang, Yankai Xie, Yalan Zeng, Feng Zhao College of Environmental Science and Engineering, Hunan University, Changsha, Hunan 410082, China Key Laboratory of Environmental Biology and Pollution Control (Hunan University), Ministry of Education, Changsha, Hunan 410082, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
The kinetics of TCE degradation by
nZVI/nZVI activated persulfate were compared. The final products of TCE degradation in both the two systems were analyzed. The effects of EDTA on TCE degradation by nZVI and nZVI/ persulfate were examined. Either positive or negative effect of EDTA was observed in different systems.
a r t i c l e
i n f o
Article history: Received 11 November 2016 Received in revised form 18 January 2017 Accepted 30 January 2017 Available online 1 February 2017 Keywords: Persulfate Nanoscale zero-valent iron EDTA ISCO Trichloroethene
a b s t r a c t This study investigated the degradation of trichloroethene (TCE) by nanoscale zero-valent iron (nZVI) w/o persulfate (PS) in the absence and presence of ethylenediaminetetraacetic acid (EDTA). In the absence of EDTA, the degradation of TCE obeyed the pseudo-first-order kinetics, and the rate constants increased exponentially with increasing molar ratio of PS and nZVI. Ethene was found to be the main reaction products in either the single nZVI system or nZVI/PS system. The presence of EDTA increased TCE degradation by nZVI in a short reaction period (within 360 min), whereas the TCE degradation was almost ceased after 1 day in a long-term (6 days) experiment, resulting in a lower TCE degradation than that in the absence of EDTA. However, EDTA drastically decreased TCE degradation in nZVI/PS system. The positive effect of EDTA in the nZVI system should be attributed to its possible inhibition of the precipitation of Fe2+/ Fe3+, which could reduce the surface passivation of nZVI. On the other hand, the adverse effect should be due to the rapid corrosion of nZVI by EDTA and the generation of a large amount of Fe2+ ions in short time, which could possibly lead to a rapid consumption of sulfate radicals originally for the degradation of TCE. Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction In recent years, the persulfate-based in-situ chemical oxidation (ISCO) technologies have been widely applied in the remediation of contaminated soils and groundwater [1–4]. Activation of persulfate ⇑ Corresponding author at: College of Environmental Science and Engineering, Hunan University, Changsha, Hunan 410082, China. E-mail address: [email protected] (H. Dong). http://dx.doi.org/10.1016/j.cej.2017.01.118 1385-8947/Ó 2017 Elsevier B.V. All rights reserved.
(PS) can be achieved by a variety of reactions, including heat, base, or transition metals (Eqs. 1–3) [5–8]. Sulfate radicals (SO 4 ) are generated from the activation of persulfate, and they are very reactive with a wide range of contaminants, such as trichloroethene (TCE) and many other halogenated organic contaminants (HOCs) [9]. heat
S2 O2 8 ! 2SO4
ð1Þ
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418 OH
2 þ 2S2 O2 8 þ 2H 2 O ! SO4 þ 3SO4 þ O2 þ 4H
ð2Þ
2 nþ nþ1 S2 O2 ! SO 8 þM 4 þ SO4 þ M
ð3Þ
Recently, zero-valent iron (ZVI) and nanoscale zero-valent iron (nZVI) have been reported to be able to activate persulfate for the degradation of contaminants [10–17]. Firstly, Fe0 corrosion occurs via reaction with persulfate (Eq. (4)) [10] and can also be initiated by waters under anaerobic conditions (Eq. (5)) [18–20], resulting in the release of Fe2+:
411
tion. Pristine nZVI particles (Nanofer 25, produced from nanosized ferrihydrite) were purchased from the NANOIRONÒ Company (Czech Republic, EU). The mixed gas standards (8.10% acetylene, 12.3% ethylene, 3.11% ethane; argon was used as the balance gas) were prepared using gasses (the purities of these gases were >99.99%) purchased from Airichem Specialty Gases & Chemicals Co., Ltd. (Dalian, China). The chemical stock solutions were prepared by dissolving chemicals into ultrapure water. Ultrapure water was purged with argon for 30 min prior to usage.
2þ Fe0 þ S2 O2 þ 2SO2 8 ! Fe 4
ð4Þ
2.2. Degradation of TCE
Fe0 þ 2H2 O ! Fe2þ þ 2OH þ H2
ð5Þ
The TCE degradation was tested in a series of duplicated batch experiments. Batch experiments were conducted in 43 mL serum vials capped by Teflon Mininert valves. Each vial contained 20 mL of 2 g/L nZVI w/o EDTA (at molar ratio of 1:1) suspension (with 23 mL of headspace). All the samples were prepared in the anaerobic glovebox so the vial headspace was initially filled with argon. TCE degradation was initiated by spiking 0.1 mL of a TCE stock solution (48 g/L TCE in methanol) into the nZVI or nZVI-EDTA suspension, resulting in an initial TCE concentration of 240 mg/L at pH 7.3. After complete mixing, 0.3 mL of 23.5 M persulfate solution was quickly added into the vials. Then, the vials were sealed and kept shaking at 250 rpm on a reciprocating shaker at room temperature (22 ± 1 °C). Control experiments were also carried out with only persulfate or EDTA added into the TCE solutions. During the reaction, 0.1 mL of the aqueous sample was withdrawn using a 250 lL gastight syringe at selected time intervals. The sample was then transferred into a 2 mL GC (gas chromatograph) vial containing 1 mL of hexane for the extraction of TCE. After phase separation, the TCE in the extract was analyzed by using a Shimadzu QP2010 ultra GC equipped with a mass spectrometer (MS) and a DB-5 capillary column. The gas samples (100 lL) in the headspace were withdrawn using a tight gas syringe for the analysis of the main gas products (using Shimadzu QP2010 ultra GC–MS). The scan range of MS was from 2 to 100 m/z, where the sum of these ions is referred to as the total ion count (TIC). Quantification of the gas products was performed by integrating the TIC and comparing the peak areas with the calibration curves obtained with the mixed gas standards (8.10% acetylene, 12.3% ethylene, 3.11% ethane; argon was used as the balance gas).
Then, the Fe2+ ions released from Fe0 corrosion can activate persulfate to generate sulfate radicals (Eq. (6)) [10,13,19,20]: 2 2þ 3þ S2 O2 ! SO 8 þ Fe 4 þ SO4 þ Fe
ð6Þ
However, Fe0 corrosion usually occurs rapidly and can produce a large amount of Fe2+ ions in a short time. The high Fe2+ concentration in the system of persulfate activation could possibly induce the scavenging of sulfate radicals (SO 4 ), and thus result in the decrease in the removal of contaminants (Eq. (7)) [10,13,19]: 3þ Fe2þ þ SO þ SO2 4 ! Fe 4
ð7Þ
To prevent the sulfate radical scavenging due to the presence of excess Fe2+, Fe2+ can be chelated by some organic ligands to form a stable metal chelate, which is effective in activating persulfate [21–24]. Among a variety of chelating agents, ethylenediaminetetraacetic acid (EDTA) is most widely used in industrial processes such as metal finishing, textile manufacturing, pulp and paper production [25–27]. EDTA is a well-known chelating agent with six potential sites (four carboxyl and two amino groups) available for binding with metal cations. It has been successfully used as a chelating agent to avoid the rapid conversion of Fe2+ to Fe3+ and to provide a Fe2+Fe3+ redox couple to facilitate persulfate activation [23]. Besides, it was reported that the organic ligands (e.g., EDTA) could prevent the formation of passivation layer of Fe3+ (hydro)oxides on the external surface of microscale ZVI through chelating of EDTA with Fe3+, which maintained the exposure of active sites on the ZVI surface [25]. Accordingly, it is expected that EDTA could influence the performance of both nZVI and nZVI/persulfate systems. However, no studies have yet systematically investigated the influence of EDTA on the degradation of organic contaminants by nZVI or nZVI activated persulfate. In this study, we used nZVI to activate persulfate and examined its potential environmental application for the degradation of trichloroethene (TCE). EDTA as a chelating agent was added into the system to explore its influence on the removal of TCE. The specific research objectives were to (i) quantify the reaction kinetics and main products in the system of nZVI w/o persulfate; (ii) investigate the influence of EDTA on the degradation of TCE by single nZVI or nZVI activated persulfate; and (iii) gain insights into the reaction rates, pathways and underlying mechanisms of TCE degradation in different systems. 2. Experimental section 2.1. Chemicals and Materials All chemical reagents used in this study, including sodium persulfate, TCE, cis-1,2-dichloroethene (c-DCE), trans-1,2dichloroethene (t-DCE), 1,1-dichloroethene (1,1-DCE), vinyl chloride (VC), methanol, hexane, K2Cr2O7, HCl, NaOH and EDTA were analytical reagent grade and used without further purifica-
2.3. Removal of Cr(VI) To compare the removal efficiency of Cr(VI) by nZVI in the absence and presence of EDTA, the kinetics experiments were conducted using a 500 mg/L nZVI suspension and 50 mg/L Cr(VI) in deionized water in the absence and presence of EDTA at pH 7.3. The final solutions (200 mL) were placed on a rotary shaker at 22 ± 1 °C and 250 rpm. At selected time intervals, the mixed suspensions were sampled and filtered immediately (through 0.45 lm membranes) for the measurement of Cr(VI) and soluble Fe concentration. The Cr(VI) concentrations were measured following the 1,5diphenylcarbazide method with a UV–visible spectrophotometer (UV-2550) at wavelength of 540 nm [28]. The concentration of soluble Fe ions in the samples was measured following a phenanthroline spectrophotometry method with UV–visible spectrophotometer (UV-2550) at wavelength of 510 nm [29]. The EDTA concentration was measured following the Standard Test Method for Tetrasodium Salt of EDTA in Water (ASTM D3113-80). Besides, the EDTA degradation was also determined by the measurement of the total organic carbon (TOC) with TOC-VCPH/CPN (Shimadzu Corporation).
412
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
3. Results and discussion 3.1. Kinetics of TCE Degradation by nZVI/ nZVI-Activated Persulfate Fig. 1a depicts the kinetics of TCE degradation by nZVI and nZVI with different mass of persulfate (PS) in batch tests under anaerobic conditions. Only 21% of TCE was removed by nZVI alone in a 360 min period, while the TCE removal significantly increased with the increasing concentration of persulfate. 72% and 84% of TCE were removed at nZVI/PS molar ratio of 1:0.5 and 1:1, respectively. Approximately 97% of TCE was removed at nZVI/PS molar ratio of 1:2. The control experiment shows that the single addition of persulfate could not degrade TCE. Thus, the remarkable degradation of TCE in the nZVI/PS system should be due to the formation of sulfate radicals [10–17]. The reaction kinetics for the reductive degrada-
tion of TCE by nZVI was able to be described by the following pseudo-first-order kinetic model.
dC TCE ¼ kobsTCE C TCE dt
ð8Þ
where C TCE is the concentration of TCE at sampling time and kobsTCE is the observed pseudo-first-order rate constant. The kinetic rate constants kobsTCE were determined by the pseudo-first-order kinetic model using the data collected within 360 min. The kobsTCE of TCE degradation by nZVI without persulfate was just 0.036 h1. Much faster degradation kinetics for the dechlorination of TCE by nZVI was observed with the presence of persulfate. The kobsTCE of nZVI with persulfate was 0.234 h1, 0.324 h1, 0.702 h1, respectively, at PS/nZVI ratio of 0.5, 1, and 2. The rate constant increased exponentially with the increasing
Fig. 1. (a) Effect of the molar ratio of persulfate and nZVI on the degradation of TCE. The solid lines show the pseudo-first-order fit curve for the reaction. Error bars indicate a relative error of ±5% for concentration measurements. (b) Changes in the kinetic rate constants for the degradation of TCE with respect to the PS/nZVI molar ratio. TCE = 36.5 lM, nZVI = 2 g/L.
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
molar ratio of persulfate and nZVI (Fig. 1b). Liang et al. [23] also reported the pseudo-first-order kinetic kobs of TCE in the PS/Fe3+/ EDTA oxidation system at different solution pH, which were 0.096 h1, 0.142 h1 and 0.502 h1 at pH 3, 7 and 10, respectively. Their study showed that the TCE degradation by persulfate activated by Fe3+/EDTA was pH dependent and the higher pH increased the activation strength in the Fe3+/EDTA activated persulfate system. In this study, all the experiments were carried out at the same initial pH (7.3), however, the solution pH were not controlled and varied during the reactions. The final pHs in the PS/nZVI systems were recorded and found to be around pH 3.5. In comparison, it can be observed that the kobsTCE of nZVI/PS were generally higher than that in Liang’s study [23] at the similar pH levels (pH 3 7). However, it should be noted that higher solution pH might con-
413
tribute to the better TCE degradation by nZVI/PS system based on the findings of Liang et al. [23], which needs to be further investigated in the future study. 3.2. Analysis of Reaction Products The kinetic studies indicate that nZVI-activated persulfate could be a promising approach to remove TCE compared to the single nZVI reduction method. To further verify the effectiveness of TCE declorination by nZVI-activated persulfate, the final reaction products were analyzed and compared with that in the single nZVI reduction system. TCE dechlorination and the major reaction products are shown in Fig. 2 (a-b). It was found that ethene was the main product in all experiments, and a small amount of ethane
Fig. 2. Degradation of TCE and formation of main products by (a) nZVI and (b) nZVI activated persulfate, Fe:PS = 1:1 (molar ratio). TCE = 36.5 lM, nZVI = 2 g/L. Error bars indicate a relative error of ±5% for concentration measurements.
414
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
and acetylene was also yielded. Besides, the trivial amounts of higher coupling products (propane, propene, butanes, butenes, pentanes, etc.) and less-chlorinated intermediates (cis-DCE, VC) were also detected. However, these reaction intermediates (i.e., the higher coupling products, cis-DCE and VC) were not quantified partly due to the deficits in the molar balance, which might result from losses during sampling caused by the overpressure generated from the reduction of water to H2 by nZVI. Besides, it was suspected that these reaction intermediates were rather reactive and quickly converted to the final products (i.e., ethane, ethane and acetylene). Hence, no apparent accumulation of the intermediates occurred during the reactions. Under the reaction conditions of
nZVI alone, it took 6 days to dechlorinate 31.2 lM TCE, and the final products were ethane (4.3 lM), ethene (16.1 lM), and acetylene (1.5 lM). In the system of nZVI-activated persulfate (PS/ Fe = 1), 29.8 lM TCE were transformed within reaction period of 360 min, primarily to two carbon unsaturated compounds, including acetylene (2.1 lM) and ethene (15.0 lM), and ethane (2.5 lM). 3.3. Possible Factors influencing TCE Degradation The above results reveal that the presence of persulfate radically accelerated the degradation of TCE. However, it should be noted that the degradation of TCE by nZVI-activated persulfate
Fig. 3. (a) Degradation of TCE by nZVI/nZVI + persulfate in the absence and presence of EDTA in a reaction period of 360 min; (b) Degradation of TCE by nZVI/nZVI + persulfate in the absence and presence of EDTA in a reaction period of 6 days. TCE = 36.5 lM, nZVI = 2 g/L. nZVI:EDTA = 1:1 (molar ratio); nZVI:EDTA:PS = 1:1:1 (molar ratio). Error bars indicate a relative error of ±5% for concentration measurements.
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
was rapid in the first 2–3 h and became slower afterwards (as demonstrated in Fig. 1a and Fig. 2b). The change in TCE reaction rate in nZVI-activated persulfate system is suspected to be caused by consumption of persulfate, depletion of nZVI particles, or passivation of reaction sites on the nZVI particles. The increasing concentration of persulfate contributed to the higher TCE degradation, which indicates that the concentration of persulfate is one factor that determined the reaction rate and the nZVI particles should not be depleted. Besides, Al-Shamsi and Thomson [12] found that the nZVI particles were passivated quickly following exposure to persulfate, causing the decrease of reaction rate with time. As mentioned previously, nZVI can activate persulfate indirectly by releasing Fe2+ into the aqueous systems (Eqs. (4) and (5)), and then the released Fe2+ can activate persulfate directly to
415
generate free radicals (Eq. (6)). The final iron species in this system is Fe3+. Theoretically, all Fe0 should be converted into Fe3+ as an indicator of the complete depletion of nZVI [30]. However, it was observed that some precipitates were formed during the reaction, which is likely a result of the following reaction pathway (Eq. (9)):
Fe3þ þ 3H2 O ! FeðOHÞ3 # þ3Hþ
ð9Þ
Thus, it is presumed that the passivation of nZVI surface could also possibly be one reason for the decreasing degradation rate of TCE with reaction time. Precious studies have reported that Fe2+ can be chelated by some organic ligands (e.g., EDTA) to avoid the rapid conversion of Fe2+ to Fe3+ and to provide a Fe2+Fe3+ redox couple to facilitate persulfate activation [21–23]. Additionally, it is presumed that the EDTA could inhibit the precipitation of Fe2+/
Fig. 4. a) Generation of soluble iron ions from nZVI corrosion in the absence and presence of EDTA; b) Reduction of Cr(VI) by nZVI in the absence and presence of EDTA (nZVI: EDTA = 1:1 (molar ratio), Cr(VI) = 50 mg/L, nZVI = 500 mg/L).
416
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
Fe3+ onto the surface of nZVI, which could alleviate the surface passivation of nZVI. Accordingly, the effect of EDTA on TCE degradation was examined in the systems of nZVI and nZVI-activated persulfate in the following sections.
3.4. Influence of EDTA on TCE Degradation by nZVI/nZVI-Activated Persulfate In order to elucidate the effect of EDTA on the degradation of TCE, the kinetics of TCE degradation by nZVI, nZVI/EDTA, nZVI/persulfate, and nZVI/persulfate/EDTA were compared and the results are shown in Fig. 3a. The results show that the presence of EDTA increased the degradation of TCE by nZVI but drastically decreased the degradation of TCE by nZVI/persulfate in a 360 min reaction period. The nZVI/EDTA system removed TCE more efficiently than either nZVI or nZVI/persulfate/EDTA system. Nearly 10.6 lM TCE was removed by nZVI/EDTA within 360 min, while only 3.6 lM and 7.5 lM of TCE was removed by nZVI/persulfate/EDTA and nZVI, respectively. However, the increasing ratio of EDTA/nZVI from 0.5 to 2 (via increasing the concentration of EDTA) did not lead to any obvious effect on TCE degradation, revealing that the
effect of EDTA is limited. Given that TCE removal by nZVI still exhibited a downward trend when the experiment was finished at 360 min (Fig. 3a), further experiments were carried out in a prolonged reaction time of 6 days (Fig. 3b). It was interesting to find that TCE was removed most by nZVI alone in the 6-day experiment, and about 31.2 lM of TCE was degraded. The degradation of TCE by nZVI/EDTA and nZVI/persulfate/EDTA system seemed to be ceased after 1 day of reaction. In the previous section, it has been discussed that EDTA could possibly inhibit the precipitation of Fe2+/Fe3+ onto the surface of nZVI, reducing the surface passivation of nZVI and thus generating more reactive sites for contaminant removal. This could be the reason for the enhancement in TCE degradation by nZVI in the presence of EDTA in the short time reaction. As demonstrated in Fig. 4a, the concentrations of dissolved iron ions from nZVI corrosion in the absence and presence of EDTA within 60 min were examined and compared. It was found that the dissolved iron concentration significantly increased with time in the presence of EDTA. The soluble iron ions concentration reached 76 mg/L after 60 min in nZVI/EDTA system, while the concentration was only 4.75 mg/L in the single nZVI system. This proved that EDTA could
Fig. 5. Determination of EDTA degradation (a: TOC concentration, b: EDTA concentration) in the nZVI/nZVI + persulfate system. nZVI = 2 g/L. nZVI:EDTA = 1:1 (molar ratio); nZVI:EDTA:PS = 1:1:1 (molar ratio). Error bars indicate a relative error of ±5% for concentration measurements.
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
inhibit the surface passivation of nZVI and enhance the corrosion of nZVI in water. In regard to the adverse impact of surface passivation of nZVI on the contaminant reduction, it has been widely reported in the reduction of Cr(VI) by nZVI, in which the reactivity of nZVI drops overtime during the reaction [29–32]. Therefore, taking Cr(VI) as an target contaminant, the role of EDTA in alleviating the surface passivation of nZVI was indirectly identified (Fig. 4b). The experiments on Cr(VI) reduction by nZVI in the absence and presence of EDTA demonstrated that the Cr(VI) reduction was significantly facilitated by EDTA. 98% Cr(VI) was reduced by nZVI/ EDTA (Fe: EDTA = 1: 1), while only 69% Cr(VI) was reduced by nZVI alone in 60 min. This further provided evidence for the effect of EDTA in alleviating the surface passivation of nZVI and facilitating the process of nZVI corrosion from Fe0 to Fe2+. However, given that the generated Fe2+ ions have no ability to degrade TCE, it was presumed that the termination of TCE degradation after 1 day in the nZVI/EDTA system should be ascribed the fast dissolution and consumption of nZVI (i.e., conversion of Fe0 to Fe2+) caused by EDTA. In addition, Fig. 3 also shows that the presence of EDTA remarkably decreased the TCE degradation by nZVI/persulfate. It has been reported [22] that the ability of free Fe2+ ions to activate persulfate for the oxidation of TCE is usually limited by the scavenging of generated sulfate radicals with excess Fe2+ and a rapid conversion of Fe2+ to Fe3+. Accordingly, the resulting rapid dissolution of nZVI caused by EDTA led to the generation of a large amount of Fe2+ ions in short time, which could possibly result in the rapid consumption of sulfate radicals originally for the degradation of TCE (as demonstrated in Eq. (7)). Besides, some studies reported incomplete mineralization of EDTA by oxidation [23,33]. For instance, Christina et al. investigated the generation of reactive oxygen species (e.g., HO) and its ability to degrade EDTA in ZVI/air/water system [33]. They reported that the degradation of Fe2+-EDTA or Fe3+-EDTA by HO yields low-molecular carboxylic acids, e.g., propionic, oxalic and iminodiacetic acids. Liang et al. [23] also reported the partial degradation of EDTA in the EDTA/Fe3+ activated persulfate oxidation system. In this study, the TOC concentration was examined for the measurement of EDTA degradation (Fig. 5a). The results showed that the TOC concentration was nearly unchanged. EDTA concentration was further examined and it was found that the EDTA concentration decreased in the nZVI/EDTA/PS system. The unchanged TOC and the decreased EDTA suggest that EDTA was partially degraded in the oxidation system. Thus, it was presumed that the partial degradation of EDTA should partly contribute to the inhibition of TCE degradation in the nZVI/persulfate/EDTA system. Therefore, it was proposed that the quick conversion of Fe0 to Fe2+ ions, the consequent rapid consumption of sulfate radicals and the partial degradation of EDTA led to the significant decrease in TCE degradation in the presence of EDTA.
4. Conclusions and environmental implications Recently, the activation of persulfate by ZVI or nZVI particles has been intensively studied and used for the remediation of contaminated soils, groundwater and sediments. This study compared the kinetics of nZVI and nZVI activated persulfate for the degradation of TCE and analyzed the final reaction products in both systems. The degradation of TCE in nZVI/persulfate system can be described by pseudo-first-order kinetics, and the rate constant increased with the increase of the molar ratio of persulfate and nZVI. In both the two systems, it was found that the main reaction products were ethene, and small amount of ethane and acetylene. The effects of EDTA addition on TCE degradation by nZVI and nZVI/ persulfate were also examined. It was found that EDTA had a positive effect on the degradation of TCE within 360 min, whereas the TCE degradation was nearly stopped after 1 day in a 6-day experi-
417
ment, resulting in a lower TCE degradation than that without EDTA. Besides, in the nZVI/persulfate system, the presence of EDTA severely inhibited TCE degradation. The presence of EDTA could possibly inhibit the precipitation of Fe2+/Fe3+ ions onto the surface of nZVI and thus reduce the surface passivation of nZVI, which might contribute to the improved TCE degradation by nZVI. However, EDTA could contribute to the rapid corrosion of nZVI and thus resulted in a short life time of nZVI, which would limit its efficiency in the long-term reaction with contaminants. The generation of a large amount of Fe2+ ions in short time as caused by EDTA might also lead to the scavenging of sulfate radicals with excess Fe2+ and a quick conversion of Fe2+ to Fe3+ in the nZVI/persulfate system, which limited the ability of sulfate radicals for the oxidation of TCE. Besides, the partial oxidation of EDTA by sulfate radicals could also be partially responsible for the decreased TCE degradation. Accordingly, even though EDTA has been widely applied in the Fe2+/Fe3+ activated persulfate system for the advanced oxidation of persistent organic pollutants, the appropriateness of EDTA application in the nZVI or nZVI/persulfate system should be further evaluated under various operational conditions (e.g., different pH and molar ratios of EDTA/nZVI/persulfate). The roles of EDTA in the nZVI or nZVI/persulfate system should be more complicate than that in the Fe2+/Fe3+ activated persulfate system, which entails further investigation in future research work.
Acknowledgments This research was supported by the National Natural Science Foundation of China (51409100, 51521006, 51378190), the Fundamental Research Funds for the Central Universities and the Program for Changjiang Scholars and Innovative Research Team in University (IRT-13R17).
References [1] M.A. Dahmani, K.H. Huang, G.E. Hoag, Sodium persulfate oxidation for the remediation of chlorinated solvents (USEPA superfund innovative technology evaluation program), Water, Air, Soil Pollut. 6 (1–2) (2006) 127–141. [2] R. Brown, D. Robinson, D. Sethi, P. Block, Second generation persulfate ISCO, in: Second International Conference on Oxidation and Reduction Technologies for In-Situ Treatment of Soil and Groundwater. Toronto, Ontario, Canada 2002, 17–21. [3] G.D. Fang, J. Gao, D.D. Dionysiou, C. Liu, D.M. Zhou, Activation of persulfate by quinones: free radical reactions and implication for the degradation of PCBs, Environ. Sci. Technol. 47 (9) (2013) 4605–4611. [4] G.P. Anipsitakis, D.D. Dionysiou, Degradation of contaminants in water with sulfate radicals generated by the conjunction of peroxymonosulfate with cobalt, Environ. Sci. Technol. 37 (2003) 4790–4797. [5] Y.H. Guan, J. Ma, X.C. Li, J.Y. Fang, L.W. Chen, Influence of pH on the formation of sulfate and hydroxyl radicals in the UV/peroxymonosulfate system, Environ. Sci. Technol. 45 (2011) 9308–9314. [6] R.H. Waldemer, P.G. Tratnyek, R.L. Johnson, J.T. Nurmi, Oxidation of chlorinated ethenes by heat-activated persulfate: kinetics and products, Environ. Sci. Technol. 41 (2007) 1010–1015. [7] O.S. Furman, A.L. Teel, R.J. Watts, Mechanism of base activation of persulfate, Environ. Sci. Technol. 44 (2010) 6423–6428. [8] C. Liang, C.J. Bruell, M.C. Marley, K.L. Sperry, Persulfate oxidation for in situ remediation of TCE. I. Activated by ferrous ion with and without a persulfatethiosulfate redox couple, Chemosphere 55 (2004) 1213–1223. [9] T. Zhang, Y. Chen, Y. Wang, J.L. Roux, Y. Yang, J. Croué, An efficient peroxydisulfate activation process not relying on sulfate radical generation for water pollutant degradation, Environ. Sci. Technol. 48 (2014) 5868–5875. [10] C.J. Liang, M.C. Lai, Trichloroethylene degradation by zero valent iron activated persulfate oxidation, Environ. Eng. Sci. 25 (7) (2008) 1071–1078. [11] G.D. Fang, D.D. Dionysiou, S.R. Al-Abed, D.M. Zhou, Superoxide radical driving the activation of persulfate by magnetite nanoparticles: Implications for the degradation of PCBs, Appl. Catal. B 129 (2013) 325–332. [12] Mohammed A. Al-Shamsi, Neil R. Thomson, Treatment of organic compounds by activated persulfate using nanoscale zerovalent iron, Ind. Eng. Chem. Res. 52 (38) (2013) 13564–13571. [13] S.Y. Oh, H.W. Kim, J.M. Park, H.S. Park, C. Yoon, Oxidation of polyvinyl alcohol by persulfate activated with heat, Fe2+, and zero-valent iron, J. Hazard. Mater. 168 (1) (2009) 346–351.
418
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
[14] S. Yang, P. Wang, X. Yang, G. Wei, W. Zhang, L. Shan, A novel advanced oxidation process to degrade organic pollutants in wastewater: microwaveactivated persulfate oxidation, J. Environ. Sci. (China) 21 (2009) 1175–1180. [15] C. Liang, Y.-T. Lin, W.-H. Shih, Treatment of trichloroethylene by adsorption and persulfate oxidation in batch studies, Ind. Eng. Chem. Res. 48 (2009) 8373–8380. [16] C.J. Liang, H.W. Su, Identification of sulfate and hydroxyl radicals in thermally activated persulfate, Ind. Eng. Chem. Res. 48 (2009) 5558–5562. [17] Yoko Furukawa, Kim Jin-wook, Janet Watkins, Richard T. Wilkin, Formation of ferrihydrite and associated iron corrosion products in permeable reactive barriers of zero-valent iron, Environ. Sci. Technol. 36 (24) (2002) 5469–5475. [18] H.R. Dong, Y.K. Xie, G.M. Zeng, L. Tang, J. Liang, Q. He, F. Zhao, Y.L. Zeng, Y.N. Wu, The dual effects of carboxymethyl cellulose on the colloidal stability and toxicity of nanoscale zero-valent iron, Chemosphere 144 (2016) 1682–1689. 2+ 2+ [19] G.V. Buxton, T.N. Malone, G.A. Salmon, Reaction of SO and 4 with Fe , Mn Cu2+ in aqueous solution, J. Chem. Soc., Faraday Trans. 93 (1997) 2893. [20] O.A. Travina, Y.N. Kozlov, A.P. Purmal, I.Y. Rod’ko, Synergism of the action of the sulfite oxidation initiators, iron and peroxydisulfate ions, Russ. J. Phys. Chem. 73 (1999) 1215. [21] C.J. Liang, C.J. Bruell, M.C. Marley, K.L. Sperry, Persulfate oxidation for in situ remediation of TCE. II. Activated by chelated ferrous ion, Chemosphere 55 (2004) 1225–1233. [22] M. Ahmad, A.L. Teel, O.S. Furman, J.I. Reed, R.J. Watts, Oxidative and reductive pathways in iron-ethylenediaminetetraacetic acid-activated persulfate systems, J. Envion. Eng. ASCE 138 (2012) 411–418. [23] C.J. Liang, C.P. Liang, C.C. Chen, PH dependence of persulfate activation by EDTA/Fe(III) for degradation of trichloroethylene, J. Contam. Hydrol. 106 (2009) 173–182. [24] Y. Sun, J.J. Pignatello, Chemical treatment of pesticide wastes. Evaluation of iron(III) chelates for catalytic hydrogen peroxide oxidation of 2,4-D at circumneutral pH, J. Agric. Food Chem. 40 (2) (1992) 322–327.
[25] X. Zhang, B. Deng, J. Guo, Y. Wang, Y. Lan, Ligand-assisted degradation of carbon tetrachloride by microscale zero-valent iron, J. Environ. Manage. 92 (4) (2011) 1328–1333. [26] N.R. Hartley, D.C. Tsang, W.E. Olds, P.A. Weber, Soil washing enhanced by humic substances and biodegradable chelating agents, Soil Sedim. Contam. 23 (2014) 599–613. [27] Z. Zou, R. Qiu, W. Zhang, H. Dong, Z. Zhao, T. Zhang, X. Wei, X. Cai, The study of operating variables in soil washing with EDTA, Environ. Pollut. 157 (2009) 229–236. [28] H.R. Dong, Q. He, G.M. Zeng, L. Tang, C. Zhang, Y.K. Xie, Y.L. Zeng, F. Zhao, Y.N. Wu, Chromate removal by surface-modified nanoscale zero-valent iron: Effect of different surface coatings and water chemistry, J. Colloid Interf. Sci. 471 (2016) 7–13. [29] H.R. Dong, Y.L. Zeng, G.M. Zeng, D.L. Huang, J. Liang, F. Zhao, Q. He, Y.K. Xie, Y.N. Wu, EDDS-assisted reduction of Cr(VI) by nanoscale zero-valent iron, Sep. Purif. Technol. 165 (2016) 86–91. [30] X.H. Guan, Y.K. Sun, H.J. Qin, J.X. Li, I.M.C. Lo, D. He, H.R. Dong, The limitations of applying zero valent iron technology in contaminants sequestration and the corresponding countermeasures: the development in zero valent iron technology in the last two decades (1994–2014), Water Res. 75 (2015) 224– 248. [31] Y.K. Xie, H.R. Dong, G.M. Zeng, L. Tang, Z. Jiang, C. Zhang, J.M. Deng, L.H. Zhang, Y. Zhang, The interactions between nanoscale zero-valent iron and microbes in the subsurface environment: a review, J. Hazard. Mater. 321 (2017) 390–407. [32] H.R. Dong, F. Zhao, G.M. Zeng, L. Tang, C.Z. Fan, L.H. Zhang, Y.L. Zeng, Q. He, Y.K. Xie, Y.N. Wu, Aging study on carboxymethyl cellulose-coated zero-valent iron nanoparticles in water: chemical transformation and structural evolution, J. Hazard. Mater. 312 (2016) 234–242. [33] Christina E. Noradoun, I. Francis Cheng, EDTA degradation induced by oxygen activation in a zerovalent iron/air/water system, Environ. Sci. Technol. 39 (18) (2005) 7158–7163.
Contents lists available at ScienceDirect
Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej
Degradation of trichloroethene by nanoscale zero-valent iron (nZVI) and nZVI activated persulfate in the absence and presence of EDTA Haoran Dong ⇑, Qi He, Guangming Zeng, Lin Tang, Lihua Zhang, Yankai Xie, Yalan Zeng, Feng Zhao College of Environmental Science and Engineering, Hunan University, Changsha, Hunan 410082, China Key Laboratory of Environmental Biology and Pollution Control (Hunan University), Ministry of Education, Changsha, Hunan 410082, China
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
The kinetics of TCE degradation by
nZVI/nZVI activated persulfate were compared. The final products of TCE degradation in both the two systems were analyzed. The effects of EDTA on TCE degradation by nZVI and nZVI/ persulfate were examined. Either positive or negative effect of EDTA was observed in different systems.
a r t i c l e
i n f o
Article history: Received 11 November 2016 Received in revised form 18 January 2017 Accepted 30 January 2017 Available online 1 February 2017 Keywords: Persulfate Nanoscale zero-valent iron EDTA ISCO Trichloroethene
a b s t r a c t This study investigated the degradation of trichloroethene (TCE) by nanoscale zero-valent iron (nZVI) w/o persulfate (PS) in the absence and presence of ethylenediaminetetraacetic acid (EDTA). In the absence of EDTA, the degradation of TCE obeyed the pseudo-first-order kinetics, and the rate constants increased exponentially with increasing molar ratio of PS and nZVI. Ethene was found to be the main reaction products in either the single nZVI system or nZVI/PS system. The presence of EDTA increased TCE degradation by nZVI in a short reaction period (within 360 min), whereas the TCE degradation was almost ceased after 1 day in a long-term (6 days) experiment, resulting in a lower TCE degradation than that in the absence of EDTA. However, EDTA drastically decreased TCE degradation in nZVI/PS system. The positive effect of EDTA in the nZVI system should be attributed to its possible inhibition of the precipitation of Fe2+/ Fe3+, which could reduce the surface passivation of nZVI. On the other hand, the adverse effect should be due to the rapid corrosion of nZVI by EDTA and the generation of a large amount of Fe2+ ions in short time, which could possibly lead to a rapid consumption of sulfate radicals originally for the degradation of TCE. Ó 2017 Elsevier B.V. All rights reserved.
1. Introduction In recent years, the persulfate-based in-situ chemical oxidation (ISCO) technologies have been widely applied in the remediation of contaminated soils and groundwater [1–4]. Activation of persulfate ⇑ Corresponding author at: College of Environmental Science and Engineering, Hunan University, Changsha, Hunan 410082, China. E-mail address: [email protected] (H. Dong). http://dx.doi.org/10.1016/j.cej.2017.01.118 1385-8947/Ó 2017 Elsevier B.V. All rights reserved.
(PS) can be achieved by a variety of reactions, including heat, base, or transition metals (Eqs. 1–3) [5–8]. Sulfate radicals (SO 4 ) are generated from the activation of persulfate, and they are very reactive with a wide range of contaminants, such as trichloroethene (TCE) and many other halogenated organic contaminants (HOCs) [9]. heat
S2 O2 8 ! 2SO4
ð1Þ
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418 OH
2 þ 2S2 O2 8 þ 2H 2 O ! SO4 þ 3SO4 þ O2 þ 4H
ð2Þ
2 nþ nþ1 S2 O2 ! SO 8 þM 4 þ SO4 þ M
ð3Þ
Recently, zero-valent iron (ZVI) and nanoscale zero-valent iron (nZVI) have been reported to be able to activate persulfate for the degradation of contaminants [10–17]. Firstly, Fe0 corrosion occurs via reaction with persulfate (Eq. (4)) [10] and can also be initiated by waters under anaerobic conditions (Eq. (5)) [18–20], resulting in the release of Fe2+:
411
tion. Pristine nZVI particles (Nanofer 25, produced from nanosized ferrihydrite) were purchased from the NANOIRONÒ Company (Czech Republic, EU). The mixed gas standards (8.10% acetylene, 12.3% ethylene, 3.11% ethane; argon was used as the balance gas) were prepared using gasses (the purities of these gases were >99.99%) purchased from Airichem Specialty Gases & Chemicals Co., Ltd. (Dalian, China). The chemical stock solutions were prepared by dissolving chemicals into ultrapure water. Ultrapure water was purged with argon for 30 min prior to usage.
2þ Fe0 þ S2 O2 þ 2SO2 8 ! Fe 4
ð4Þ
2.2. Degradation of TCE
Fe0 þ 2H2 O ! Fe2þ þ 2OH þ H2
ð5Þ
The TCE degradation was tested in a series of duplicated batch experiments. Batch experiments were conducted in 43 mL serum vials capped by Teflon Mininert valves. Each vial contained 20 mL of 2 g/L nZVI w/o EDTA (at molar ratio of 1:1) suspension (with 23 mL of headspace). All the samples were prepared in the anaerobic glovebox so the vial headspace was initially filled with argon. TCE degradation was initiated by spiking 0.1 mL of a TCE stock solution (48 g/L TCE in methanol) into the nZVI or nZVI-EDTA suspension, resulting in an initial TCE concentration of 240 mg/L at pH 7.3. After complete mixing, 0.3 mL of 23.5 M persulfate solution was quickly added into the vials. Then, the vials were sealed and kept shaking at 250 rpm on a reciprocating shaker at room temperature (22 ± 1 °C). Control experiments were also carried out with only persulfate or EDTA added into the TCE solutions. During the reaction, 0.1 mL of the aqueous sample was withdrawn using a 250 lL gastight syringe at selected time intervals. The sample was then transferred into a 2 mL GC (gas chromatograph) vial containing 1 mL of hexane for the extraction of TCE. After phase separation, the TCE in the extract was analyzed by using a Shimadzu QP2010 ultra GC equipped with a mass spectrometer (MS) and a DB-5 capillary column. The gas samples (100 lL) in the headspace were withdrawn using a tight gas syringe for the analysis of the main gas products (using Shimadzu QP2010 ultra GC–MS). The scan range of MS was from 2 to 100 m/z, where the sum of these ions is referred to as the total ion count (TIC). Quantification of the gas products was performed by integrating the TIC and comparing the peak areas with the calibration curves obtained with the mixed gas standards (8.10% acetylene, 12.3% ethylene, 3.11% ethane; argon was used as the balance gas).
Then, the Fe2+ ions released from Fe0 corrosion can activate persulfate to generate sulfate radicals (Eq. (6)) [10,13,19,20]: 2 2þ 3þ S2 O2 ! SO 8 þ Fe 4 þ SO4 þ Fe
ð6Þ
However, Fe0 corrosion usually occurs rapidly and can produce a large amount of Fe2+ ions in a short time. The high Fe2+ concentration in the system of persulfate activation could possibly induce the scavenging of sulfate radicals (SO 4 ), and thus result in the decrease in the removal of contaminants (Eq. (7)) [10,13,19]: 3þ Fe2þ þ SO þ SO2 4 ! Fe 4
ð7Þ
To prevent the sulfate radical scavenging due to the presence of excess Fe2+, Fe2+ can be chelated by some organic ligands to form a stable metal chelate, which is effective in activating persulfate [21–24]. Among a variety of chelating agents, ethylenediaminetetraacetic acid (EDTA) is most widely used in industrial processes such as metal finishing, textile manufacturing, pulp and paper production [25–27]. EDTA is a well-known chelating agent with six potential sites (four carboxyl and two amino groups) available for binding with metal cations. It has been successfully used as a chelating agent to avoid the rapid conversion of Fe2+ to Fe3+ and to provide a Fe2+Fe3+ redox couple to facilitate persulfate activation [23]. Besides, it was reported that the organic ligands (e.g., EDTA) could prevent the formation of passivation layer of Fe3+ (hydro)oxides on the external surface of microscale ZVI through chelating of EDTA with Fe3+, which maintained the exposure of active sites on the ZVI surface [25]. Accordingly, it is expected that EDTA could influence the performance of both nZVI and nZVI/persulfate systems. However, no studies have yet systematically investigated the influence of EDTA on the degradation of organic contaminants by nZVI or nZVI activated persulfate. In this study, we used nZVI to activate persulfate and examined its potential environmental application for the degradation of trichloroethene (TCE). EDTA as a chelating agent was added into the system to explore its influence on the removal of TCE. The specific research objectives were to (i) quantify the reaction kinetics and main products in the system of nZVI w/o persulfate; (ii) investigate the influence of EDTA on the degradation of TCE by single nZVI or nZVI activated persulfate; and (iii) gain insights into the reaction rates, pathways and underlying mechanisms of TCE degradation in different systems. 2. Experimental section 2.1. Chemicals and Materials All chemical reagents used in this study, including sodium persulfate, TCE, cis-1,2-dichloroethene (c-DCE), trans-1,2dichloroethene (t-DCE), 1,1-dichloroethene (1,1-DCE), vinyl chloride (VC), methanol, hexane, K2Cr2O7, HCl, NaOH and EDTA were analytical reagent grade and used without further purifica-
2.3. Removal of Cr(VI) To compare the removal efficiency of Cr(VI) by nZVI in the absence and presence of EDTA, the kinetics experiments were conducted using a 500 mg/L nZVI suspension and 50 mg/L Cr(VI) in deionized water in the absence and presence of EDTA at pH 7.3. The final solutions (200 mL) were placed on a rotary shaker at 22 ± 1 °C and 250 rpm. At selected time intervals, the mixed suspensions were sampled and filtered immediately (through 0.45 lm membranes) for the measurement of Cr(VI) and soluble Fe concentration. The Cr(VI) concentrations were measured following the 1,5diphenylcarbazide method with a UV–visible spectrophotometer (UV-2550) at wavelength of 540 nm [28]. The concentration of soluble Fe ions in the samples was measured following a phenanthroline spectrophotometry method with UV–visible spectrophotometer (UV-2550) at wavelength of 510 nm [29]. The EDTA concentration was measured following the Standard Test Method for Tetrasodium Salt of EDTA in Water (ASTM D3113-80). Besides, the EDTA degradation was also determined by the measurement of the total organic carbon (TOC) with TOC-VCPH/CPN (Shimadzu Corporation).
412
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
3. Results and discussion 3.1. Kinetics of TCE Degradation by nZVI/ nZVI-Activated Persulfate Fig. 1a depicts the kinetics of TCE degradation by nZVI and nZVI with different mass of persulfate (PS) in batch tests under anaerobic conditions. Only 21% of TCE was removed by nZVI alone in a 360 min period, while the TCE removal significantly increased with the increasing concentration of persulfate. 72% and 84% of TCE were removed at nZVI/PS molar ratio of 1:0.5 and 1:1, respectively. Approximately 97% of TCE was removed at nZVI/PS molar ratio of 1:2. The control experiment shows that the single addition of persulfate could not degrade TCE. Thus, the remarkable degradation of TCE in the nZVI/PS system should be due to the formation of sulfate radicals [10–17]. The reaction kinetics for the reductive degrada-
tion of TCE by nZVI was able to be described by the following pseudo-first-order kinetic model.
dC TCE ¼ kobsTCE C TCE dt
ð8Þ
where C TCE is the concentration of TCE at sampling time and kobsTCE is the observed pseudo-first-order rate constant. The kinetic rate constants kobsTCE were determined by the pseudo-first-order kinetic model using the data collected within 360 min. The kobsTCE of TCE degradation by nZVI without persulfate was just 0.036 h1. Much faster degradation kinetics for the dechlorination of TCE by nZVI was observed with the presence of persulfate. The kobsTCE of nZVI with persulfate was 0.234 h1, 0.324 h1, 0.702 h1, respectively, at PS/nZVI ratio of 0.5, 1, and 2. The rate constant increased exponentially with the increasing
Fig. 1. (a) Effect of the molar ratio of persulfate and nZVI on the degradation of TCE. The solid lines show the pseudo-first-order fit curve for the reaction. Error bars indicate a relative error of ±5% for concentration measurements. (b) Changes in the kinetic rate constants for the degradation of TCE with respect to the PS/nZVI molar ratio. TCE = 36.5 lM, nZVI = 2 g/L.
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
molar ratio of persulfate and nZVI (Fig. 1b). Liang et al. [23] also reported the pseudo-first-order kinetic kobs of TCE in the PS/Fe3+/ EDTA oxidation system at different solution pH, which were 0.096 h1, 0.142 h1 and 0.502 h1 at pH 3, 7 and 10, respectively. Their study showed that the TCE degradation by persulfate activated by Fe3+/EDTA was pH dependent and the higher pH increased the activation strength in the Fe3+/EDTA activated persulfate system. In this study, all the experiments were carried out at the same initial pH (7.3), however, the solution pH were not controlled and varied during the reactions. The final pHs in the PS/nZVI systems were recorded and found to be around pH 3.5. In comparison, it can be observed that the kobsTCE of nZVI/PS were generally higher than that in Liang’s study [23] at the similar pH levels (pH 3 7). However, it should be noted that higher solution pH might con-
413
tribute to the better TCE degradation by nZVI/PS system based on the findings of Liang et al. [23], which needs to be further investigated in the future study. 3.2. Analysis of Reaction Products The kinetic studies indicate that nZVI-activated persulfate could be a promising approach to remove TCE compared to the single nZVI reduction method. To further verify the effectiveness of TCE declorination by nZVI-activated persulfate, the final reaction products were analyzed and compared with that in the single nZVI reduction system. TCE dechlorination and the major reaction products are shown in Fig. 2 (a-b). It was found that ethene was the main product in all experiments, and a small amount of ethane
Fig. 2. Degradation of TCE and formation of main products by (a) nZVI and (b) nZVI activated persulfate, Fe:PS = 1:1 (molar ratio). TCE = 36.5 lM, nZVI = 2 g/L. Error bars indicate a relative error of ±5% for concentration measurements.
414
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
and acetylene was also yielded. Besides, the trivial amounts of higher coupling products (propane, propene, butanes, butenes, pentanes, etc.) and less-chlorinated intermediates (cis-DCE, VC) were also detected. However, these reaction intermediates (i.e., the higher coupling products, cis-DCE and VC) were not quantified partly due to the deficits in the molar balance, which might result from losses during sampling caused by the overpressure generated from the reduction of water to H2 by nZVI. Besides, it was suspected that these reaction intermediates were rather reactive and quickly converted to the final products (i.e., ethane, ethane and acetylene). Hence, no apparent accumulation of the intermediates occurred during the reactions. Under the reaction conditions of
nZVI alone, it took 6 days to dechlorinate 31.2 lM TCE, and the final products were ethane (4.3 lM), ethene (16.1 lM), and acetylene (1.5 lM). In the system of nZVI-activated persulfate (PS/ Fe = 1), 29.8 lM TCE were transformed within reaction period of 360 min, primarily to two carbon unsaturated compounds, including acetylene (2.1 lM) and ethene (15.0 lM), and ethane (2.5 lM). 3.3. Possible Factors influencing TCE Degradation The above results reveal that the presence of persulfate radically accelerated the degradation of TCE. However, it should be noted that the degradation of TCE by nZVI-activated persulfate
Fig. 3. (a) Degradation of TCE by nZVI/nZVI + persulfate in the absence and presence of EDTA in a reaction period of 360 min; (b) Degradation of TCE by nZVI/nZVI + persulfate in the absence and presence of EDTA in a reaction period of 6 days. TCE = 36.5 lM, nZVI = 2 g/L. nZVI:EDTA = 1:1 (molar ratio); nZVI:EDTA:PS = 1:1:1 (molar ratio). Error bars indicate a relative error of ±5% for concentration measurements.
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
was rapid in the first 2–3 h and became slower afterwards (as demonstrated in Fig. 1a and Fig. 2b). The change in TCE reaction rate in nZVI-activated persulfate system is suspected to be caused by consumption of persulfate, depletion of nZVI particles, or passivation of reaction sites on the nZVI particles. The increasing concentration of persulfate contributed to the higher TCE degradation, which indicates that the concentration of persulfate is one factor that determined the reaction rate and the nZVI particles should not be depleted. Besides, Al-Shamsi and Thomson [12] found that the nZVI particles were passivated quickly following exposure to persulfate, causing the decrease of reaction rate with time. As mentioned previously, nZVI can activate persulfate indirectly by releasing Fe2+ into the aqueous systems (Eqs. (4) and (5)), and then the released Fe2+ can activate persulfate directly to
415
generate free radicals (Eq. (6)). The final iron species in this system is Fe3+. Theoretically, all Fe0 should be converted into Fe3+ as an indicator of the complete depletion of nZVI [30]. However, it was observed that some precipitates were formed during the reaction, which is likely a result of the following reaction pathway (Eq. (9)):
Fe3þ þ 3H2 O ! FeðOHÞ3 # þ3Hþ
ð9Þ
Thus, it is presumed that the passivation of nZVI surface could also possibly be one reason for the decreasing degradation rate of TCE with reaction time. Precious studies have reported that Fe2+ can be chelated by some organic ligands (e.g., EDTA) to avoid the rapid conversion of Fe2+ to Fe3+ and to provide a Fe2+Fe3+ redox couple to facilitate persulfate activation [21–23]. Additionally, it is presumed that the EDTA could inhibit the precipitation of Fe2+/
Fig. 4. a) Generation of soluble iron ions from nZVI corrosion in the absence and presence of EDTA; b) Reduction of Cr(VI) by nZVI in the absence and presence of EDTA (nZVI: EDTA = 1:1 (molar ratio), Cr(VI) = 50 mg/L, nZVI = 500 mg/L).
416
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
Fe3+ onto the surface of nZVI, which could alleviate the surface passivation of nZVI. Accordingly, the effect of EDTA on TCE degradation was examined in the systems of nZVI and nZVI-activated persulfate in the following sections.
3.4. Influence of EDTA on TCE Degradation by nZVI/nZVI-Activated Persulfate In order to elucidate the effect of EDTA on the degradation of TCE, the kinetics of TCE degradation by nZVI, nZVI/EDTA, nZVI/persulfate, and nZVI/persulfate/EDTA were compared and the results are shown in Fig. 3a. The results show that the presence of EDTA increased the degradation of TCE by nZVI but drastically decreased the degradation of TCE by nZVI/persulfate in a 360 min reaction period. The nZVI/EDTA system removed TCE more efficiently than either nZVI or nZVI/persulfate/EDTA system. Nearly 10.6 lM TCE was removed by nZVI/EDTA within 360 min, while only 3.6 lM and 7.5 lM of TCE was removed by nZVI/persulfate/EDTA and nZVI, respectively. However, the increasing ratio of EDTA/nZVI from 0.5 to 2 (via increasing the concentration of EDTA) did not lead to any obvious effect on TCE degradation, revealing that the
effect of EDTA is limited. Given that TCE removal by nZVI still exhibited a downward trend when the experiment was finished at 360 min (Fig. 3a), further experiments were carried out in a prolonged reaction time of 6 days (Fig. 3b). It was interesting to find that TCE was removed most by nZVI alone in the 6-day experiment, and about 31.2 lM of TCE was degraded. The degradation of TCE by nZVI/EDTA and nZVI/persulfate/EDTA system seemed to be ceased after 1 day of reaction. In the previous section, it has been discussed that EDTA could possibly inhibit the precipitation of Fe2+/Fe3+ onto the surface of nZVI, reducing the surface passivation of nZVI and thus generating more reactive sites for contaminant removal. This could be the reason for the enhancement in TCE degradation by nZVI in the presence of EDTA in the short time reaction. As demonstrated in Fig. 4a, the concentrations of dissolved iron ions from nZVI corrosion in the absence and presence of EDTA within 60 min were examined and compared. It was found that the dissolved iron concentration significantly increased with time in the presence of EDTA. The soluble iron ions concentration reached 76 mg/L after 60 min in nZVI/EDTA system, while the concentration was only 4.75 mg/L in the single nZVI system. This proved that EDTA could
Fig. 5. Determination of EDTA degradation (a: TOC concentration, b: EDTA concentration) in the nZVI/nZVI + persulfate system. nZVI = 2 g/L. nZVI:EDTA = 1:1 (molar ratio); nZVI:EDTA:PS = 1:1:1 (molar ratio). Error bars indicate a relative error of ±5% for concentration measurements.
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
inhibit the surface passivation of nZVI and enhance the corrosion of nZVI in water. In regard to the adverse impact of surface passivation of nZVI on the contaminant reduction, it has been widely reported in the reduction of Cr(VI) by nZVI, in which the reactivity of nZVI drops overtime during the reaction [29–32]. Therefore, taking Cr(VI) as an target contaminant, the role of EDTA in alleviating the surface passivation of nZVI was indirectly identified (Fig. 4b). The experiments on Cr(VI) reduction by nZVI in the absence and presence of EDTA demonstrated that the Cr(VI) reduction was significantly facilitated by EDTA. 98% Cr(VI) was reduced by nZVI/ EDTA (Fe: EDTA = 1: 1), while only 69% Cr(VI) was reduced by nZVI alone in 60 min. This further provided evidence for the effect of EDTA in alleviating the surface passivation of nZVI and facilitating the process of nZVI corrosion from Fe0 to Fe2+. However, given that the generated Fe2+ ions have no ability to degrade TCE, it was presumed that the termination of TCE degradation after 1 day in the nZVI/EDTA system should be ascribed the fast dissolution and consumption of nZVI (i.e., conversion of Fe0 to Fe2+) caused by EDTA. In addition, Fig. 3 also shows that the presence of EDTA remarkably decreased the TCE degradation by nZVI/persulfate. It has been reported [22] that the ability of free Fe2+ ions to activate persulfate for the oxidation of TCE is usually limited by the scavenging of generated sulfate radicals with excess Fe2+ and a rapid conversion of Fe2+ to Fe3+. Accordingly, the resulting rapid dissolution of nZVI caused by EDTA led to the generation of a large amount of Fe2+ ions in short time, which could possibly result in the rapid consumption of sulfate radicals originally for the degradation of TCE (as demonstrated in Eq. (7)). Besides, some studies reported incomplete mineralization of EDTA by oxidation [23,33]. For instance, Christina et al. investigated the generation of reactive oxygen species (e.g., HO) and its ability to degrade EDTA in ZVI/air/water system [33]. They reported that the degradation of Fe2+-EDTA or Fe3+-EDTA by HO yields low-molecular carboxylic acids, e.g., propionic, oxalic and iminodiacetic acids. Liang et al. [23] also reported the partial degradation of EDTA in the EDTA/Fe3+ activated persulfate oxidation system. In this study, the TOC concentration was examined for the measurement of EDTA degradation (Fig. 5a). The results showed that the TOC concentration was nearly unchanged. EDTA concentration was further examined and it was found that the EDTA concentration decreased in the nZVI/EDTA/PS system. The unchanged TOC and the decreased EDTA suggest that EDTA was partially degraded in the oxidation system. Thus, it was presumed that the partial degradation of EDTA should partly contribute to the inhibition of TCE degradation in the nZVI/persulfate/EDTA system. Therefore, it was proposed that the quick conversion of Fe0 to Fe2+ ions, the consequent rapid consumption of sulfate radicals and the partial degradation of EDTA led to the significant decrease in TCE degradation in the presence of EDTA.
4. Conclusions and environmental implications Recently, the activation of persulfate by ZVI or nZVI particles has been intensively studied and used for the remediation of contaminated soils, groundwater and sediments. This study compared the kinetics of nZVI and nZVI activated persulfate for the degradation of TCE and analyzed the final reaction products in both systems. The degradation of TCE in nZVI/persulfate system can be described by pseudo-first-order kinetics, and the rate constant increased with the increase of the molar ratio of persulfate and nZVI. In both the two systems, it was found that the main reaction products were ethene, and small amount of ethane and acetylene. The effects of EDTA addition on TCE degradation by nZVI and nZVI/ persulfate were also examined. It was found that EDTA had a positive effect on the degradation of TCE within 360 min, whereas the TCE degradation was nearly stopped after 1 day in a 6-day experi-
417
ment, resulting in a lower TCE degradation than that without EDTA. Besides, in the nZVI/persulfate system, the presence of EDTA severely inhibited TCE degradation. The presence of EDTA could possibly inhibit the precipitation of Fe2+/Fe3+ ions onto the surface of nZVI and thus reduce the surface passivation of nZVI, which might contribute to the improved TCE degradation by nZVI. However, EDTA could contribute to the rapid corrosion of nZVI and thus resulted in a short life time of nZVI, which would limit its efficiency in the long-term reaction with contaminants. The generation of a large amount of Fe2+ ions in short time as caused by EDTA might also lead to the scavenging of sulfate radicals with excess Fe2+ and a quick conversion of Fe2+ to Fe3+ in the nZVI/persulfate system, which limited the ability of sulfate radicals for the oxidation of TCE. Besides, the partial oxidation of EDTA by sulfate radicals could also be partially responsible for the decreased TCE degradation. Accordingly, even though EDTA has been widely applied in the Fe2+/Fe3+ activated persulfate system for the advanced oxidation of persistent organic pollutants, the appropriateness of EDTA application in the nZVI or nZVI/persulfate system should be further evaluated under various operational conditions (e.g., different pH and molar ratios of EDTA/nZVI/persulfate). The roles of EDTA in the nZVI or nZVI/persulfate system should be more complicate than that in the Fe2+/Fe3+ activated persulfate system, which entails further investigation in future research work.
Acknowledgments This research was supported by the National Natural Science Foundation of China (51409100, 51521006, 51378190), the Fundamental Research Funds for the Central Universities and the Program for Changjiang Scholars and Innovative Research Team in University (IRT-13R17).
References [1] M.A. Dahmani, K.H. Huang, G.E. Hoag, Sodium persulfate oxidation for the remediation of chlorinated solvents (USEPA superfund innovative technology evaluation program), Water, Air, Soil Pollut. 6 (1–2) (2006) 127–141. [2] R. Brown, D. Robinson, D. Sethi, P. Block, Second generation persulfate ISCO, in: Second International Conference on Oxidation and Reduction Technologies for In-Situ Treatment of Soil and Groundwater. Toronto, Ontario, Canada 2002, 17–21. [3] G.D. Fang, J. Gao, D.D. Dionysiou, C. Liu, D.M. Zhou, Activation of persulfate by quinones: free radical reactions and implication for the degradation of PCBs, Environ. Sci. Technol. 47 (9) (2013) 4605–4611. [4] G.P. Anipsitakis, D.D. Dionysiou, Degradation of contaminants in water with sulfate radicals generated by the conjunction of peroxymonosulfate with cobalt, Environ. Sci. Technol. 37 (2003) 4790–4797. [5] Y.H. Guan, J. Ma, X.C. Li, J.Y. Fang, L.W. Chen, Influence of pH on the formation of sulfate and hydroxyl radicals in the UV/peroxymonosulfate system, Environ. Sci. Technol. 45 (2011) 9308–9314. [6] R.H. Waldemer, P.G. Tratnyek, R.L. Johnson, J.T. Nurmi, Oxidation of chlorinated ethenes by heat-activated persulfate: kinetics and products, Environ. Sci. Technol. 41 (2007) 1010–1015. [7] O.S. Furman, A.L. Teel, R.J. Watts, Mechanism of base activation of persulfate, Environ. Sci. Technol. 44 (2010) 6423–6428. [8] C. Liang, C.J. Bruell, M.C. Marley, K.L. Sperry, Persulfate oxidation for in situ remediation of TCE. I. Activated by ferrous ion with and without a persulfatethiosulfate redox couple, Chemosphere 55 (2004) 1213–1223. [9] T. Zhang, Y. Chen, Y. Wang, J.L. Roux, Y. Yang, J. Croué, An efficient peroxydisulfate activation process not relying on sulfate radical generation for water pollutant degradation, Environ. Sci. Technol. 48 (2014) 5868–5875. [10] C.J. Liang, M.C. Lai, Trichloroethylene degradation by zero valent iron activated persulfate oxidation, Environ. Eng. Sci. 25 (7) (2008) 1071–1078. [11] G.D. Fang, D.D. Dionysiou, S.R. Al-Abed, D.M. Zhou, Superoxide radical driving the activation of persulfate by magnetite nanoparticles: Implications for the degradation of PCBs, Appl. Catal. B 129 (2013) 325–332. [12] Mohammed A. Al-Shamsi, Neil R. Thomson, Treatment of organic compounds by activated persulfate using nanoscale zerovalent iron, Ind. Eng. Chem. Res. 52 (38) (2013) 13564–13571. [13] S.Y. Oh, H.W. Kim, J.M. Park, H.S. Park, C. Yoon, Oxidation of polyvinyl alcohol by persulfate activated with heat, Fe2+, and zero-valent iron, J. Hazard. Mater. 168 (1) (2009) 346–351.
418
H. Dong et al. / Chemical Engineering Journal 316 (2017) 410–418
[14] S. Yang, P. Wang, X. Yang, G. Wei, W. Zhang, L. Shan, A novel advanced oxidation process to degrade organic pollutants in wastewater: microwaveactivated persulfate oxidation, J. Environ. Sci. (China) 21 (2009) 1175–1180. [15] C. Liang, Y.-T. Lin, W.-H. Shih, Treatment of trichloroethylene by adsorption and persulfate oxidation in batch studies, Ind. Eng. Chem. Res. 48 (2009) 8373–8380. [16] C.J. Liang, H.W. Su, Identification of sulfate and hydroxyl radicals in thermally activated persulfate, Ind. Eng. Chem. Res. 48 (2009) 5558–5562. [17] Yoko Furukawa, Kim Jin-wook, Janet Watkins, Richard T. Wilkin, Formation of ferrihydrite and associated iron corrosion products in permeable reactive barriers of zero-valent iron, Environ. Sci. Technol. 36 (24) (2002) 5469–5475. [18] H.R. Dong, Y.K. Xie, G.M. Zeng, L. Tang, J. Liang, Q. He, F. Zhao, Y.L. Zeng, Y.N. Wu, The dual effects of carboxymethyl cellulose on the colloidal stability and toxicity of nanoscale zero-valent iron, Chemosphere 144 (2016) 1682–1689. 2+ 2+ [19] G.V. Buxton, T.N. Malone, G.A. Salmon, Reaction of SO and 4 with Fe , Mn Cu2+ in aqueous solution, J. Chem. Soc., Faraday Trans. 93 (1997) 2893. [20] O.A. Travina, Y.N. Kozlov, A.P. Purmal, I.Y. Rod’ko, Synergism of the action of the sulfite oxidation initiators, iron and peroxydisulfate ions, Russ. J. Phys. Chem. 73 (1999) 1215. [21] C.J. Liang, C.J. Bruell, M.C. Marley, K.L. Sperry, Persulfate oxidation for in situ remediation of TCE. II. Activated by chelated ferrous ion, Chemosphere 55 (2004) 1225–1233. [22] M. Ahmad, A.L. Teel, O.S. Furman, J.I. Reed, R.J. Watts, Oxidative and reductive pathways in iron-ethylenediaminetetraacetic acid-activated persulfate systems, J. Envion. Eng. ASCE 138 (2012) 411–418. [23] C.J. Liang, C.P. Liang, C.C. Chen, PH dependence of persulfate activation by EDTA/Fe(III) for degradation of trichloroethylene, J. Contam. Hydrol. 106 (2009) 173–182. [24] Y. Sun, J.J. Pignatello, Chemical treatment of pesticide wastes. Evaluation of iron(III) chelates for catalytic hydrogen peroxide oxidation of 2,4-D at circumneutral pH, J. Agric. Food Chem. 40 (2) (1992) 322–327.
[25] X. Zhang, B. Deng, J. Guo, Y. Wang, Y. Lan, Ligand-assisted degradation of carbon tetrachloride by microscale zero-valent iron, J. Environ. Manage. 92 (4) (2011) 1328–1333. [26] N.R. Hartley, D.C. Tsang, W.E. Olds, P.A. Weber, Soil washing enhanced by humic substances and biodegradable chelating agents, Soil Sedim. Contam. 23 (2014) 599–613. [27] Z. Zou, R. Qiu, W. Zhang, H. Dong, Z. Zhao, T. Zhang, X. Wei, X. Cai, The study of operating variables in soil washing with EDTA, Environ. Pollut. 157 (2009) 229–236. [28] H.R. Dong, Q. He, G.M. Zeng, L. Tang, C. Zhang, Y.K. Xie, Y.L. Zeng, F. Zhao, Y.N. Wu, Chromate removal by surface-modified nanoscale zero-valent iron: Effect of different surface coatings and water chemistry, J. Colloid Interf. Sci. 471 (2016) 7–13. [29] H.R. Dong, Y.L. Zeng, G.M. Zeng, D.L. Huang, J. Liang, F. Zhao, Q. He, Y.K. Xie, Y.N. Wu, EDDS-assisted reduction of Cr(VI) by nanoscale zero-valent iron, Sep. Purif. Technol. 165 (2016) 86–91. [30] X.H. Guan, Y.K. Sun, H.J. Qin, J.X. Li, I.M.C. Lo, D. He, H.R. Dong, The limitations of applying zero valent iron technology in contaminants sequestration and the corresponding countermeasures: the development in zero valent iron technology in the last two decades (1994–2014), Water Res. 75 (2015) 224– 248. [31] Y.K. Xie, H.R. Dong, G.M. Zeng, L. Tang, Z. Jiang, C. Zhang, J.M. Deng, L.H. Zhang, Y. Zhang, The interactions between nanoscale zero-valent iron and microbes in the subsurface environment: a review, J. Hazard. Mater. 321 (2017) 390–407. [32] H.R. Dong, F. Zhao, G.M. Zeng, L. Tang, C.Z. Fan, L.H. Zhang, Y.L. Zeng, Q. He, Y.K. Xie, Y.N. Wu, Aging study on carboxymethyl cellulose-coated zero-valent iron nanoparticles in water: chemical transformation and structural evolution, J. Hazard. Mater. 312 (2016) 234–242. [33] Christina E. Noradoun, I. Francis Cheng, EDTA degradation induced by oxygen activation in a zerovalent iron/air/water system, Environ. Sci. Technol. 39 (18) (2005) 7158–7163.