Hydrodechlorination of Alachlor Using a Molecular Electrocatalyst

Reductive hydrodechlorination is an effective approach to enhance the degradation rate of chlorinated herbicides such as alachlor, which are frequently detected in ground and surface water. In this study, a cobalt porphyrin complex with eight triazole units and alkyl chains, CoPor8T, was synthesized to catalyze the reductive hydrodechlorination of alachlor. Mechanistic study was performed using a combination of voltametric, spectroscopic, and electrospectroscopic techniques. A conversion yield of 84 % at −1.8 V vs. Fc/Fc+ and chloride ion concentration of 96 % was obtained after electrocatalysis. This work provides a new avenue of using molecular catalysts for electrocatalytic chlorinated herbicide remediation.


Introduction
Alachlor is a herbicide from the chloroacetanilide herbicide family and is widely used to control annual grasses and broadleaf weeds in crop cultivation. [1] However, it is characterized as a Class B2 human carcinogen by the US Environmental Protection Agency.
It persists in the soil and its residue and degradation metabolites have been frequently detected in ground and surface water. [2] Extensive and uncontrolled use of these herbicides is an environmental hazard, which has been a topic of concern in recent years.
Various abiotic and biotic strategies have been employed for alachlor degradation to date. [3] The most common strategy is to use ozone to oxidize alachlor in drinking water. [4] Qiang et al. have investigated the degradation of alachlor by advanced oxidation processes such as ozonation and H 2 O 2 pretreatment. [5] Fenton systems have also been used for alachlor removal owing to their high efficiency and environmental benignity. [6] However, the accumulation of ferric ions during the process slows down the reaction owing to ion blockage. The use of protocatechuic acid (PCA) improved the alachlor degradation process, which was attributed to the formation of a stable complex of PCA and the accumulated ferric ions, thus preventing the blockage. [7] Jeong et al. coupled gamma irradiation with hydrogen peroxide and enhanced the efficiency of alachlor degradation. [8] Zero-valent iron particles have also been employed for the reduction of alachlor in an aqueous medium. [3] Although oxidative methods are widely used for the degradation of organic compounds, hydrodechlorination reactions present some disadvantages. Hazardous byproducts such as inorganic and organic chloro species have been observed during electrooxidation processes. [9] Notably, various microbial strains induce alachlor degradation in the environment, and based on this, few studies have employed microorganisms for alachlor degradation. [3] Electrochemical strategies have been used as an efficient and green tool for the reductive dehalogenation reactions. This method can be utilized in the environmental remediation of halogenated substrates and in the production of useful substances such as chemical feedstocks. Moreover, electro reductive dechlorination can also be combined with biodegradation and oxidation processes for the complete mineralization of otherwise unreactive chlorinated compounds. [1,10] Rondinini et al. [11] have elucidated the mechanism of the electroreduction of organic halides. Ni I salen complexes generated at glassy carbon electrodes in acetonitrile (MeCN) or N,N'-dimethylformamide (DMF) have shown excellent catalytic activity in the reduction of halogenated organic compounds. The vast majority of dehalogenation catalysts are either single metal atoms or metal electrodes (carbon, [12] copper, [13] palladium, [14] silver, [15] and iron [10b,16] ) and metal oxides (SnO 2 , PbO 2 , TiO 2 , IrO 2 , etc.). [17] These electrodes demonstrate high efficiency and reactivity but lack structural tunability or product selectivity. [17a] Porphyrins have been studied in various electrocatalytic reactions. [18] In addition, catalytic metal centers are required for substrate binding and electron transfer. Porphyrin ligands paired with redox-innocent metal centers are used in redox processes. [19] Porphyrins have gained significant attention as molecular catalysts because of their facile synthesis, modular design for meso-and β-substitutions, and formation of strong metal complexes. [20] We previously reported a copper di-triazole electrocatalyst for the hydrodechlorination of dichloromethane (CH 2 Cl 2 ) to produce C1 and C2 hydrocarbons in a heterogeneous aqueous system. [19b] We also recently reported the electrocatalytic activity of a free-base triazole-porphyrin in the hydrodechlorination of CH 2 Cl 2 in MeCN. [19a] In this study, we introduced a cobalt porphyrin with triazole units in the second coordination sphere, CoPor8T, for the electro-hydrodechlorina-tion of alachlor ( Figure 1). A control compound with eight 1,2,3triazole units and a hydrophilic poly(ethylene glycol) (PEG) chain on each of the eight triazole units, 2HPEG8T, was employed for comparison. Electronic spectroscopy and cyclic voltammetry were used to investigate the catalytic efficiency and elucidate the mechanism of the hydrodechlorination reaction.
It was previously reported that the incorporation of triazole units as a second coordination sphere promotes the formation of a hydro shell and consequently enhances the activity of catalytic CO 2 reduction and hydrodechlorination of CH 2 Cl 2 . [19a] In this study, we adopted the same multifunctional porphyrin ligand but incorporated a more redox-active metal center, cobalt. Catalyst CoPor8T with eight 1,2,3-triazole units and a hydrophobic alkyl chain on each of the eight triazole units was synthesized in 61 % yield using the free-base triazole porphyrin 2HPor8T [21] as the starting material ( Figure 1). A previously reported free base catalyst, 2HPEG8T, without any metal center was used as the control. [18a] The PEG units are incorporated in the control compound to enhance its solubility in MeCN to allow for homogeneous studies. UV-visible spectroscopy was used to confirm metalation of CoPor8T. The difference in the Qband patterns ( Figure S1) indicates successful metalation. Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) was used to characterize the catalyst CoPor8T ( Figure S2). All porphyrin compounds were purified and characterized prior to the electrochemical and electrocatalytic studies.

Results and Discussion
The catalytic properties of the homogeneous catalyst CoPor8T for the hydrodechlorination of alachlor (Scheme 1) were first studied by cyclic voltammetry (CV) using 0.5 mM of the catalyst in MeCN with 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF 6 ) as the supporting electrolyte. Glassy carbon was used as the working electrode with ferrocene for internal potential calibration. A Pt wire was used as the counter electrode. For CoPor8T, two successive one-electron peaks were observed at À 0.8, and À 2.1 V vs. Fc/Fc + (Figure 2A). The first peak was reversible and could be attributed to the reduction of Co II to Co I . The second reversible redox couple correspond to the ligand-centered redox process. [21] CV scans at different rates (100-500 mV/s) were performed under the same condition ( Figure S3). The linear relationships between the peak currents at À 2.1 V vs. Fc/Fc + and the square roots of the scan rates indicates a diffusion-controlled process according to the Randles-Sevcik equation (Diffusion constant, D = 3.45x10 À 6 cm 2 / s). [18a] For catalytic performance studies, the substrate, alachlor (10 mM), and proton source, water (2 M), were successively added to determine the activity of each component during CV measurements (Figure 2A, blue and purple traces, respectively). An increase in the current was observed upon addition of 10 mM alachlor. After the addition of 2 M H 2 O, a catalytic current increase was observed at À 1.8 V, which could be attributed to either proton reduction or hydrodechlorination of alachlor. The electrochemical properties of 2HPEG8T were studied under the same conditions as those for CoPor8T. The increase in current for 2HPEG8T was lesser ( Figure 2B, purple trace) than that for CoPor8T at À 1.8 V, indicating that the catalytic activity of the Co catalyst is higher than that of the metal-free catalyst.
To gain insights on kinetics, CV studies of CoPor8T were carried out at different scan rates under the same conditions ( Figure S4). Although no S-shaped curve was observed during the study, indicating the absence of a pure kinetic condition, the catalytic activity of CoPor8T was found to be independent of the scan rate in the range of 300 À 500 mV/s, suggesting a mass-diffusion limit, presumably due to the low concentration of the alachlor substrate, which is limited by its solubility in aqueous medium. Controlled-potential electrolysis (CPE) was conducted in an H-cell separated by a Nafion membrane using MeCN with 0.1 M TBAPF 6 and 2 M H 2 O to evaluate the catalytic activity of CoPor8T. Carbon paper was used as the working electrode and carbon rods as counter electrodes. Proton nuclear magnetic resonance ( 1 H NMR) spectroscopy, gas chromatography-mass spectrometry (GC-MS), and ion chromatography (IC) were employed to detect the products in solution. GC was used to detect gaseous products formed during electrolysis. In Figure 2A, E cat /2 for CoPor8T was observed at À 1.8 V vs. Fc/Fc + (Figure 2A); hence the two other potentials for CPE were chosen at lower values than E cat /2 at À 1.7 V and À 1.6 V vs. Fc/Fc + .
Thus, the electrolysis was performed at three different potentials, À 1.6, À 1.7, and À 1.8 V vs. Fc/Fc + for 3 h each ( Figure 3). After each electrolysis, hydrodechlorinated alachlor was observed via GC-MS ( Figure S5). A minor amount of byproduct (5 % by GC) was also observed in the electro-lyte solution ( Figure S5), which may have been an electrochemical hydroxy alachlor product based on its mass pattern in the GC-MS spectra. To confirm that the byproducts were produced electrochemically, a separate test was performed in which 10 mM of the substrate alachlor was added to the catalyst solution and left for 3 h with no applied potential. No product peak was observed in the GC-MS spectra, which suggests that the byproduct should have been produced electrochemically, presumably via a nucleophilic substitution mechanism. [22] Column chromatography was performed to purify the reaction mixture, and the pure dechlorinated product was confirmed by 1 H NMR and highresolution electrospray ionization mass spectrometry (HR-ESI-MS) ( Figures S6, S7, and S8). GC-MS analysis indicated that a maximum product yield of 84 % was obtained (turnover number (TON) = 19.3 and turnover frequency (TOF) = 0.107 s À 1 ) after the 3 h electrolysis at À 1.8 V vs. Fc/ Fc + ( Figure S5). In addition, a maximum current of À 4.8 mA/ cm 2 was observed. The product yields at À 1.7 V and À 1.6 V vs. Fc/Fc + observed during the 3 h electrolysis was 81 % and 66 % with current densities of À 4.5 and À 3.8 mA/cm 2 , respectively ( Figure 3). Furthermore, IC was used to detect Cl À ions produced from the hydrodechlorination of alachlor ( Figure S10). A dechlorination yield of 96.2 % was obtained at -1.8 V vs. Fc/Fc + , indicating the high catalytic hydrodechlorination efficiency of CoPor8T.
To evaluate the role of the metal center in dechlorination catalysis, control experiments were performed under the same conditions using the metal-free catalyst 2HPEG8T ( Figure S9A). A maximum product yield of 30 % was obtained from GC-MS analysis at a potential of À 1.8 V vs. Fc/Fc + , and a current density of À 3 mA/cm 2 was observed. The current density of the free-base porphyrin catalyst was comparatively low. The solutions of 0.1 M TBAPF 6 in MeCN and 2 M H 2 O were also studied under the same conditions as a control experiment to determine whether any dechlorinated products were formed in the absence of a catalyst. At  À 1.8 V vs. Fc/Fc + , the product yield was 22 % at a low current density of À 1.71 mA/cm 2 ( Figure S9B).Thus, CoPor8T has been validated as an active catalyst for the hydrodechlorination of alachlor.
The main reaction pathways for the electroreductive cleavage of the carbon-halogen bond leads to a total or partial dehalogenated products with the formation of CÀ H or single, double and triple CÀ C bonds. [10b,11,17a,19a,23] Based on previous reports, we propose a plausible two-electronone-proton mechanism for the reductive dechlorination of alachlor ( Figure 4A). [24] Initially, the metal center Co II undergoes reduction to Co I followed by ligand-centered reduction. [25] The reactive Co I species then eliminates the chloride anions and subsequently undergoes oxidative addition to the alkyl chain of alachlor. [25,26] Finally, the Co IIalkyl complex undergoes homolytic cleavage to form alkyl radicals. The intermediate then accepts one proton from H 2 O to afford the corresponding dechlorinated product. A spectro electrochemical study was conducted with and without the alachlor substrate to elucidate the catalytic mechanism of CoPor8T. UV-visible spectra of the electrolytic solution, MeCN with 0.1 M TBAPF 6 , and proton source, 2 M H 2 O, ( Figure 4B and 4 C) were collected at the third redox event (À 1.8 V vs. Fc/Fc + ) under an inert atmosphere. In the absence of alachlor ( Figure 4B), upon electroreduction, there was a continuous decrease in the intensity of the Soret (450 nm) and Q bands (560 nm), which is attributed to both ligand and metal center reductions. However, in the presence of the substrate, as shown in Figure 4C, the intensity of the Soret and Q bands gradually decreased, shifted hypsochromically at a certain time, and then remained nearly constant. The change in the spectra was associated with the isosbestic point at 545 nm ( Figure 4C). This can be explained by the reduction of the precatalyst Co II to Co I followed by electron transfer from the catalyst to the substrate alachlor. Thus, the spectro electrochemical study validated our proposed catalytic cycle shown in Figure 4A.

Conclusion
In this study, a cobalt porphyrin complex with eight triazole units and alkyl chains, CoPor8T, has been synthesized and characterized. Its electrocatalytic activity in the reductive dechlorination of alachlor was investigated using water as the proton source. At À 1.8 V vs. Fc/Fc + , a current density of À 4.8 mA/cm 2 was observed and a dechlorination yield of 84 % was obtained. For comparison, we conducted the same experiment with the previously reported metal-free catalyst, 2HPEG8T. Following the electrocatalysis, the chloride ion concentration obtained from the dechlorination was determined to be 96.2 % via ion chromatography. Mechanistic studies were conducted by time-dependent UV-visible spectroscopy in the presence and absence of alachlor. Based on these studies, we proposed a mechanism for the catalytic hydrodechlorination of alachlor. This study provides an efficient electrochemical method for alachlor remediation under ambient conditions.

Experimental Section
Logic VSP potentiostat. The UV-visible absorption spectra were obtained using an Agilent Technologies Cary 8454 UV-visible spectrometer. Proton nuclear magnetic resonance spectroscopy (1H NMR) was recorded using a Bruker AV 400 MHz NMR. Matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) was performed on a Bruker Biflex III MALDI-MS instrument using 1,4-bis(5-phenyl-2-oxazolyl)benzene as the matrix. High-resolution electrospray ionization mass spectrometry (HR-ESI-MS) analysis was performed on an Orbitrap Fusion Lumos mass spectrometer from Thermo Scientific. All liquid products from the electrolysis solutions were detected via gas chromatography mass spectrometry (GC-MS) using an Agilent 5977A/7890B GC/MS System. The chloride ion (Cl À ) concentration from the post-electrolysis electrolyte were detected by ion chromatography (IC) using a Metrohm Eco IC instrument. Alachlor was purchased from Sigma Aldrich (� 98 % purity) and Caymen Chemicals (� 95 % purity). It should be noted that an unidentified minor impurity (~9.5 % from GC analysis) was detected from both samples. Compounds 2HPEG8T was synthesized based on our previous work. [19a] Solutions for all the CV studies were prepared using 0.5 mM catalyst in MeCN with 0.1 M TBAPF 6 as the electrolyte solution. CV studies were carried out using a glassy carbon working electrode, a platinum wire counter electrode, and Ag/AgO reference electrode. The reference electrode was calibrated using ferrocene (Fc) by measuring the Fc/Fc + redox couple every 10 min for 2 h to ensure potential stability prior to use in electrochemical measurements. Ferrocene was added to the post-catalysis solution for potential calibration. The glassy carbon electrode was polished using alumina particles (0.05 microns) before each measurement, and the platinum electrode was flame-cleaned.
CPE was performed in an H-cell configuration with a Nafion 117 cation-exchange membrane to separate the working and the counter cells to preserve any product collected during the electrolysis. For CPE, 6 mL of the catalyst (0.5 mM) solution in MeCN with 0.1 M TBAPF 6 and 2 M H 2 O with 10 mM of the substrate (alachlor) was used in the working cell, with the same volume of catalyst-free electrolyte in the counter cell. GDS 3250 carbon paper (0.5 cm × 1 cm) was used as the working electrode, and carbon rod as counter electrode.
The product formed by electrochemical hydrodechlorination of alachlor was analyzed by GC-MS. After each electrolysis, 100 μL of the post catalysis solution was added to a GC vial along with 100 μL of an internal standard, 3-chloropropylbenzene (10 mM in MeCN). The above solution was then diluted to 600 μL with MeCN before analysis.

Synthesis of CoPor8T
A sample of Co(OAc) 2 · 4H 2 O (0.180 g, 0.72 mmol) in methanol (5.0 mL) was added to a stirring solution of free-base porphyrin (2HPor8T [21] 0.082 g, 0.036 mmol) in chloroform (25 mL). The resulting solution was heated at 50°C and stirred for 30 h (Figure 1). The reaction mixture was then cooled to room temperature and diluted with CH 2 Cl 2 and washed with brine. The organic layer was evaporated and dried via rotatory evaporator. The residue was purified by column chromatography (10 % MeOH in CH 2 Cl 2 ) to afford the title compound