Article pubs.acs.org/est
Quantitative Assessment of the Formation of Polychlorinated Derivatives, PCDD/Fs, in the Electrochemical Oxidation of 2‑Chlorophenol As Function of the Electrolyte Type Marta Vallejo, M. Fresnedo San Román, and Inmaculada Ortiz* Dpto. Chemical Engineering and Inorganic Chemistry. ETSIIyT. Universidad de Cantabria, Avda. de los Castros, 39005, Santander, Spain ABSTRACT: The electrochemical degradation of 2-chlorophenol (2-CP) on boron-doped diamond (BDD) anodes was carried out using two electrolytes, NaCl and Na2SO4. Both electrolytes supported complete mineralization of 2CP, but faster rates of degradation were observed in NaCl. After 4 h of oxidation, the total organic carbon (TOC) balance neared 100% with Na2SO4 for identifiable compounds, whereas 4 mM of TOC remained unidentified with NaCl. Since chlorophenols are known to be precursors of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs), a rigorous assessment of intermediate products was carried out. When near complete mineralization was achieved, the use of NaCl resulted in the concentration of total PCDD/Fs increasing 2.68 × 104 times compared to the untreated sample, and to toxicity values several times higher than the maximum level established by U.S. Environmental Protection Agency for water ingestion. When Na2SO4 was used, the increase in total PCDD/Fs concentration was 134 times lower than with NaCl and there was no significant 2,3,7,8-PCDD/Fs formation. Thus, we emphasize the importance of electrolyte selection in electro-oxidation processes, especially when PCDD/Fs precursors are initially present or may be formed in the treated water samples.
■
INTRODUCTION
treatment to degrade many organic contaminants, including phenolic compounds.8−12 The electrochemical oxidation for the treatment of wastewater containing aromatic pollutants has attracted a great deal of attention recently, because of its ease of control, amenability to automation, high efficiency and environmental compatibility.13 The use of boron-doped diamond (BDD) anodes has been reported to yield higher organic oxidation rates and greater current efficiencies than other commonly used metal oxides.14 BDD anodes are known to have a greater oxidative power with high O2 overvoltage producing large amounts of OH· and leading to rapid rates of pollutants degradation.10 There are several studies that deal with the electrochemical oxidation of aqueous solutions containing chlorophenols that show a high mineralization of these organic contaminants.15−18 In spite of the high effectiveness of electrochemical oxidation in the oxidation of refractory compounds, it must be taken into account that the accumulation of (chlorinated)-aromatic intermediates, some of which may be more toxic than their parent compounds, can take place during the treatment process. In fact, according to Weber,19 one important criterion for the assessment of the treatment technologies for persistent organic pollutants (POPs) is the potential formation of new
Chlorophenols (CPs) comprise a group of organic compounds characterized by acute toxicity, emission of strong odor, bioaccumulation, and resistance to biodegradation; they are also suspected mutagens and carcinogens.1 Consequently CPs have been listed as priority pollutants by the U.S. Environmental Protection Agency’s (USEPA) Clean Water Act2 and by the European Decision 2455/2001/CE.3 CPs serve as intermediates in the synthesis of insecticides, herbicides, pharmaceuticals, and dyes, as wood preservatives and disinfectants. In addition, CPs may be formed during the incineration of waste, the bleaching of pulp and water disinfection.4 As a result of such wide use CPs have been identified in industrial wastewaters, surface and groundwater, soils and even in finished drinking waters.5,6 Various industrial effluents and municipal waste discharges typically contain CPs at concentration from 1 to 21 ppm.7 It is necessary to remove these hazardous compounds from wastewaters in order to reduce the negative impact on human health and environment. Biological processes are ineffective because the high toxicity of CPs inhibit the growth of microorganisms. Physical/chemical treatments such as flocculation, precipitation, adsorption or reverse osmosis require and additional post-treatment to remove the pollutants from the newly contaminated stream.6 Advanced oxidation processes (AOPs) that are based on the formation of very active hydroxyl radicals (OH·) have been applied successfully in wastewater © XXXX American Chemical Society
Received: July 22, 2013 Revised: September 26, 2013 Accepted: October 7, 2013
A
dx.doi.org/10.1021/es403246g | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
acetonitrile -80% H3PO4 0.01 M as mobile phase with a flow gradient up to 80% Acetonitrile −20% H3PO4 in 2.5 min at λ of 210, 254, and 280 nm. The analysis of organic acids, acetic, formic, malonic, maleic, fumaric, and oxalic acid, was carried out by means of ion chromatography with anionic suppression using the chromatograph Dionex ICS-1100 with a conductivity cell detector (ASRULTRA model). The column IonPac AS9-HC (4 mm) was used as stationary phase and a 9 mM solution of Na2CO3 flowing at 1 mL min−1 was the mobile phase. PCDD/Fs Analysis. Standard Method USEPA 1613 (1994) for PCDD/Fs analysis was used. Samples (0.5 L) were spiked with 10 μL of a 15 13C-labeled PCDD/Fs solution (EPA 1613 LCS) dissolved in acetone. PCDD/Fs were extracted with three portions of 60 mL aliquots of dichloromethane. The organic extract was concentrated in a rotary evaporator (Laborota 4000), transferred to n-hexane and treated with H2SO4. The organic phase was then dried with sodium sulfate and concentrated again in the rotary evaporator to approximately 1−2 mL. The extract was then filtered through a 0.45 μm PTFE filter and cleaned-up using an automated system (Power-Prep, Fluid Management Systems Inc.) through a multilayer silica column, then a basic alumina column, and finally a PX-21 active carbon column. The purified extract was concentrated by means of the rotary evaporator and transferred into a vial to be concentrated to dryness under nitrogen. Purified samples were analyzed by the Chromatography Service of the University of Cantabria (SERCROM). Before the chromatographic analysis, internal syringe standards (EPA 1613 ISS) were added to the sample. The analysis was carried out using high resolution gas chromatography coupled with high resolution mass spectrometry (HRGC-HRMS). A TRACE GC UltraTM gas chromatograph was used equipped with a split/ splitless injector (Thermo Electron S.p.A.) and a DB-5 MS fused silica capillary column (J&W Scientific). The initial temperature of the column, 120 °C, was kept constant for 2 min, then increased sequentially in 3 steps to 210, 230, and 310 °C at 15, 1, and 3 °C min−1 respectively. The column was connected through a heated transfer line kept at 270 °C to a DFS high-resolution magnetic sector mass spectrometer with a BE geometry (Thermo Fisher Scientific). Positive electron ionization (EI+) mode with ionization energy of 45 eV was used in the source and its temperature was set at 270 °C. The mass spectrometer was operated in SIM mode at 10 000 resolution power (10% valley definition). Detection limits were calculated as the concentration values that gave instrumental responses within a signal-to-noise ratio of 3. Quantitative determination was carried out by the isotopic dilution method. Relative response factors (RRFs), obtained from the calibration curve by analyzing CS-1 to CS-5 standard solution mixtures, were used to determine the target compounds’ concentration in the samples. The recoveries of labeled standards were calculated using a mixture of two labeled PCDD (ISS) that were added to the samples before the chromatographic analysis. Final results were expressed in pg L−1 and also in total toxic equivalent (TEQ) as pg I-TEQ L−1 for 2,3,7,8-PCDD/Fs obtained by adding up individual congener concentrations weighted by the international toxic factors (I-TEF). Quality Control. The reliability of the PCDD/Fs analytical methodology was assured by means of the analysis of ultrapure water samples spiked with native PCDD/Fs (PAR standard), blank samples and the use of labeled PCDD/Fs (LCS
POPs and other toxic byproducts, such as polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs). In fact, CPs are known to contain many chlorinated impurities, such as dioxins, and are potent precursors of PCDD/Fs. PCDD/Fs are a family of POPs comprised of 210 congeners; those with chlorine at 2,3,7,8 positions are of particular interest due to their toxicity and potential effects on human health, mainly due to their persistence and bioaccumulative behavior.20 Thus, a rigorous assessment of the occurrence and quantification of highly toxic substances such as PCDD/Fs is needed during the application of AOPs for the abatement of chlorinated organic compounds, especially if chlorine is present in the reaction medium. This study analyzed the distribution of intermediate products, paying special attention to the potential formation of PCDD/Fs, as a result of the electrochemical treatment of solutions containing 2-chlorophenol (2-CP) on BDD anodes. Furthermore, to our knowledge, this is the first time that an increase in PCDD/Fs concentration that is highly dependent on the operational conditions has been observed and quantified in electrochemical remediation processes. The role of the supporting electrolyte was determined using two different and commonly employed electrolytes, NaCl21,22 and Na2SO4.
■
MATERIALS AND METHODS Reagents and Materials. 2-CP, 2,4-dichorophenol, 2,5dichlorophenol, 2,3,6-trichlorophenol, 2-chlorobenzoquinone, 2,6-dichlorobenzoquinone, catechol, hydroquinone and pbenzoquinone were provided by Sigma-Aldrich. NaCl and Na2SO4, used as electrolytes, H3PO4 and Na2CO3 were purchased from Panreac. EPA 1613 standard solutions in nonane (CS-1 to CS-5, PAR, LCS and ISS, Wellington Laboratories, Ontario, Canada) were used for instrument calibration, recovery, quantification and quality control. Toluene, dichloromethane, hexane and acetone (for organic trace analysis), sulphuric acid and sodium sulfate were purchased from Merck. Silica, alumina and carbon columns were purchased from Technospec. 2.7 μm glass microfiber filter and 0.45 μm Millex syringe filters were purchased from Whatman and Millipore. Electrochemical Oxidation Experiments. Electro-oxidation experiments of 1L of 2-CP solutions (15.56 mM) were performed in batch mode in a laboratory DiaCell system (two circular electrodes: BDD on silicon anode and stainless steel cathode, with surface area of 0.007 m2 each and an electrode gap of 5 mm) working with a current density of 900 A m−2 and operational flow rate of 9 L min−1. Two supporting electrolytes were used, NaCl (34.2 mM) and Na2SO4 (21.1 mM) resulting in an initial conductivity of 7.5 mS cm−1. Two replicates of each experiment were carried out. TOC, COD, HPLC, and Ion Chromatography Analysis. Total organic carbon (TOC) analysis were performed using an TOC-V CPH (Shimadzu) and chemical oxygen demand (COD) was determined by the open reflux method following the analytical procedure 5220B from Standard Methods. 2-CP and aromatic reaction intermediates were measured in a high-pressure liquid chromatograph (HPLC) Agilent Series 1100, using a Supelco reversed-phase column LC-8 and a photo diode array (PDA) detector. 2-CP, catechol, hydroquinone and p-benzoquinone were measured using H2SO4 4 mM as mobile phase at λ = 210 and 245 nm. 2,4-dichorophenol, 2,5dichlorophenol, 2,3,6-trichlorophenol, 2-chlorobenzoquinone and 2,6-dichlorobenzoquinone were measured using 20% B
dx.doi.org/10.1021/es403246g | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
The limiting current density, which establishes the limit between the operating regime controlled by applied charge or by mass-transfer, was calculated according to the expression24 (eq 2):
standard). The average recoveries were in the range of 63− 126% for 13C12- labeled 2,3,7,8-CDD/Fs congeners indicating that the analysis performed satisfactorily as the recoveries were within the ranges established in the EPA 1613 method. Blanks covering the whole sample preparation methodology for PCDD/Fs analysis were included, demonstrating that congeners were either not detected or below the detection limits.
Jlim = 4FK mCOD
where F is Faraday’s constant (96500 C mol ), Km is the mass transport coefficient (1.46 × 10−5 m s−1, determined by means of the Sherwood (Sh), Reynolds (Re), and Schmidt (Sc) numbers), and COD is expressed as mol O2 m−3. Since the working current density, 900 A m−2, was higher than the limiting current density, 591 A m−2, the process should be mass-transfer-controlled. Therefore, the reaction between 2-CP and OH· is favored in the anode diffuse layer due to the high concentration of OH· electrochemically generated on the anode surface in comparison with the pollutant. Furthermore, due to the high concentration of OH·, the generic reaction of 2-CP oxidation may be described by a pseudo first-order kinetic equation according to what has been observed with experimental data. In addition to the aforementioned proposed electrooxidation mechanisms (a, b), evidence of mediated oxidation mechanisms by means of other electro-generated species in the bulk solution such as peroxodisulphates and active chlorine has been reported in the literature.25,26 Results of 2-CP electrooxidation (Figure 1a) indicate that considerably lower values of the specific charge were needed to decrease the concentration of 2-CP to negligible concentrations with NaCl than with Na2SO4. These results suggest a positive contribution of the indirect chlorine mediated oxidation of 2-CP in the bulk solution.27 The experimental data show that complete mineralization (decrease in TOC) required higher values of the specific charge than did the reduction of COD, independent of the electrolyte used, and higher also than those needed for the oxidation of 2CP. This indicates that the formation of oxidation intermediate products need longer times to be oxidized (Figure 1b). On the other hand, there is no clear evidence of the influence of the electrolyte type on the kinetic change for either TOC or COD, as the difference in results may be within experimental error. Hydroxylation and chlorination products with predominantly ortho/para substitutions are expected: the hydroxyl functional group in 2-CP is a highly active ortho/para-directing group, and OH· are strong electrophilic radicals.28 Furthermore, although the chlorine atom of the aromatic ring is a deactivating group it is also an ortho/para-directing group. When NaCl was used as the electrolyte, two dichlorosubstituted phenols, 2,4-diCP and 2,6-diCP, and a trichlorosubstituted phenol, 2,3,6-triCP, were formed during the first minutes of electro-oxidation treatment (Figure 2a). Maximum intermediate product concentrations were measured at 30 min, and at 2 h, intermediate concentrations were negligible. Higher concentrations of chlorinated intermediates were observed for 2,4-diCP, highlighting the preference of the chlorination of 2CP in the para position. Finally, 2,6-dichlorobenzoquinone was found only in the presence of NaCl, and at very low concentrations (
