Environ. Sci. Technol. 2008, 42, 6942–6948
Correlation of 2-Chlorobiphenyl Dechlorination by Fe/Pd with Iron Corrosion at Different pH Y U A N X I A N G F A N G †,‡ A N D S O U H A I L R . A L - A B E D * ,† National Risk Management Research Laboratory, U.S. Environmental Protection Agency, 26 W. Martin Luther King Dr. Cincinnati, Ohio 45268
Received March 20, 2008. Revised manuscript received July 9, 2008. Accepted July 9, 2008.
The rate of 2-chlorobiphenyl dechlorination by palladized iron (Fe/ Pd) decreased with increasing pH until pH > 12.5. Iron corrosion potential (Ec) and current (jc), obtained from polarization curves of a rotating disk electrode of iron, followed the Tafel equation at pH e 5.5 and pH g 9.5. The pH dependence of the dechlorination rate constant (k1) suggests four pH regimes. In the low pH regime (3-5.5), |Ec| and jc decreased with increasing pH and k1 was linearly correlated to |Ec| and jc0.5. The correlation between k1 and jc0.5 indicates direct involvement of active hydrogen species (on the Pd surface) in PCB dechlorination. In the mid pH regime (5.5-9.5), no significant effect of pH was evident on the values of k1, jc, and Ec, a combined result of limiting anodic oxidation of iron to an intermediate product (iron hydroxide) and a proton-independent overall reaction. Both |Ec| and jc increased significantly as pH increased from 9.5 to 14. A clear trough of the k1 values in solutions of pH between 12 and 13 and the mismatch between the kinetic and corrosion data suggest two pH regimes (9.5-12.5 and 12.5-14) of different corrosion mechanisms.
1. Introduction Palladized iron (Fe/Pd) is an effective catalyst for the dechlorination of polychlorinated biphenyls (PCBs), a group of persistent organic pollutants (POPs) (1, 2). For this group of highly hydrophobic and recalcitrant contaminants, recent studies indicate the great potential of Fe/Pd technology for both in situ and ex situ remediation, which has been a very tough challenge (3-8). PCB dechlorination using Fe/Pd involves the hydrogen produced at the palladium surface as a product of iron corrosion with water (eq 1) (1, 2): H2 + RX f RH + H+ + X
(1)
Fe + 2H2O f Fe + H2 + 2OH
(2a)
0
+
2+
Fe +2 H f Fe +H2
* Corresponding author phone: 513-569-7849; fax: 513-569-7879; e-mail:
[email protected]. † U.S. Environmental Protection Agency. ‡ Current address: Sriya Innovations, Inc., 1831 West Oak Parkway, Suite B, Marietta, Georgia 30062,
[email protected]; phone: 770419-9809; Fax: 770-419-8907. 9
2. Electrochemical Analysis The iron corrosion in aqueous acid (eq 2b) consists of two half-reactions: Fe0)Fe2+ + 2e-
(2b)
Palladium catalyzes hydrodehalogenation of organics is a well-known phenomenon in organic synthesis (9). The Pd on the Fe surface acts as a collector of hydrogen species that
6942
is produced by iron corrosion (eqs 2a and 2b). The PCBs that are adsorbed on the Pd or the Pd/Fe surface are dechlorinated by the active hydrogen species dissolved in the Pd lattice. The mechanism is also manifested in the electrocatalytic dechlorination of chlorinated aromatic compounds at Pd modified carbon cloth and graphite electrodes (10, 11) and in the hydrodechlorination of such compounds over palladium on alumina (12). This dechlorination process can be highly pH sensitive because protons are involved both directly and indirectly (13). For dechlorination following eq 1, protons may directly and significantly affect the reaction extent. The rate of reduction of chlorinated organic compounds and inorganic compounds such as nitrate using zerovalent iron generally decreases with increasing pH. The reduction rates were commonly correlated to the pH or H+ concentration (14-16). In addition, H+ is indirectly involved in the hydrodechlorination, in which H+ is directly involved in the generation of hydrogen that is subsequently directly involved in the dechlorination (eq 1). The rate of hydrogen generation is related to iron corrosion potential and current. Nevertheless, no studies have been published on relationships between dechlorination rates and the corrosion parameters. Therefore, the aim of this study is (i) to evaluate the pH effect on PCB dechlorination by Fe/Pd using 2-chlorobiphenyl (2-ClBP) as a model compound, and (ii) to investigate relationships between the rates of dechlorination and the iron corrosion parameters. The dechlorination rates were obtained by conducting short-term experiments of 2-ClBP dechlorination using Fe/ Pd in different pH solutions. Short-term experiments that last for a very short period of time can minimize the longterm effects of catalyst deactivation, active surface loss to catalyst aging (17), and solution pH change on the dechlorination kinetics. Although short-term results may differ in general from the actual long-term performance, they allow for a careful kinetic study of Fe/Pd dechlorination under various initial conditions, preventing effects of complicated processes and undefined varying conditions, such as corrosion and passivation (4). When the reactants are involved in competitive adsorption by catalysts such as Fe/Pd, shortterm experiments provide data to calculate the actual rate of dechlorination. The corrosion potential and current were obtained from the potential-current sweeping curve of a disk electrode of iron in different pH solutions. Attempts were made to correlate the dechlorination rate values with the corrosion parameters in different pH regimes.
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 18, 2008
+
(3)
-
2 H (aq) + 2e )H2(g)
(4)
The current-potential curves for the above two electrochemical reactions can be expressed as
[ [
] ) ]
jM ) jM 0 exp
-RMF(E - EeM) βMF(E - EeM) - exp RT RT
(5)
jH ) jH 0 exp
-RHF(E - EeH) βHF(E - EeH - exp RT RT
(6)
10.1021/es800805y CCC: $40.75
2008 American Chemical Society
Published on Web 08/19/2008
where R and β ) 1 - R are, respectively, the cathodic and anodic transfer coefficients of the half-cell reaction; Ee is the equilibrium half-cell potential; j0 is the exchange current density at equilibrium; and superscripts M and H denote half-cell reactions 3 and 4, respectively. The net current j for reaction 2b is jH + jM. A state of zero net current represents a condition in which the anodic and cathodic components of the current are equal, jH ) |jM|. Equations 5 and 6 can be solved to obtain the corrosion potential Ec and corrosion current density jc (18), and j c ) jH ) jH 0 exp
|| |
-RHF(Ec - EeH) RT
) jM
) - jM 0 exp
βMF(Ec - EeM) RT
|
(7)
The net current density-potential curve can be expressed in terms of Ec and jc as in eq 8 in the form of the Butler-Volmer equation (19). At a potential sufficiently more negative than EeH (and Ec), the cathodic curve can be expressed in eq 9 in the form of the Tafel equation; at a potential sufficiently above EeM (and Ec), the anodic curve can be expressed in eq 10.
[
j ) jc exp
-RHF(E - Ec) βMF(E - Ec) - exp RT RT
j ) jH ) jcexp
]
(8)
-RHF(E - Ec) RT
(9)
βMF(E - Ec) RT
(10)
j ) jM ) -jcexp
Typical values of RH and βM for iron in aqueous acids are 0.4 and 1.5 (18), respectively. Note RH * 1 - βM and a relative smaller value of RH (compared to βM) does not necessarily indicate the half-cell reaction 4 is the rate-limiting process. Fitting eqs 8 and 9 to the proper sections of the currentpotential curves on the two sides of Ec can result in two jc values, which nevertheless can differ.
3. Materials and Methods 3.1. Chemicals. Iron particles ranging in size from 210 to 290 µm were obtained from Fisher Scientific. Palladium acetate Pd(CH3CO2)2 (purity >98%) was obtained from GFS Chemicals. Biphenyl (BP) and 2-ClBP (purity >99%), with an internal standard (3, 3′, 4-trichlorobiphenyl)sand standard BP and 2-ClBP solutions were obtained from Ultra Scientific. A cosolvent of acetone and hexane (30:70) was prepared for extraction of organic compounds from aqueous solutions and Fe/Pd particles. Hexane (Optima grade) and acetone (GC grade) were obtained from Fisher Scientific. The HCl (6 N) used was from Aldrich. The water used in the experiments was oxygen-free, obtained by bubbling nitrogen gas overnight through milli-Q water (resistivity g18 MΩ cm). 3.2. Preparation of Fe/Pd Particles. The palladium was deposited on Fe particles in an ethanol solution of palladium acetate using a method modified from one used previously (4, 20). Prior to deposition, a given amount of Fe particles was washed with 6 N HCl for a few seconds and was immediately washed five times with oxygen-free milli-Q water. The acid-washed iron particles were moved to a plastic centrifuge tube that was fully filled with an ethanol solution of palladium acetate (no headspace in the tube); the concentration of palladium acetate was calculated to obtain a target Pd loading of 0.585% based on the amount of Fe
before acid-washing. The tube was placed on a tumbler mixer at 30 rpm for 2 h for complete deposition of Pd. At the end of deposition, the ethanol solution was sampled and analyzed for Pd concentration in the final solution; the remaining solution was discarded. The Pd-deposited iron particles (Fe/ Pd) were washed four times with (oxygen-free) water, each time using the same volume as the ethanol solution (i.e., no headspace in the tube). After water-washing, the Fe/Pd particles were ready for use in the dechlorination experiments. All batches of Fe/Pd particles used in this study contained the same Pd target content of 0.585%. This Pd content was chosen so that the resulting rates of dechlorination varied responsively to changes of reaction conditions in the experiments that completed in a few hours. 3.3. Dechlorination of 2-ClBP in Solutions of Different pH. Each dechlorination experiment consisted of several sets of 40 mL vials. Vials in different sets contained solutions of different pH. Solutions with a pH different than seven were prepared by adding hydrochloric acid or adjusted by adding NaOH in water. The five vials in each set contained solutions of the same pH, but the five solutions in the each set contained 2-ClBP of different concentrations that ranged between 2 and 16 µm. All vials were fully filled with solutions to prevent any head space (and presence of air); the volumes of solutions (43.2 mL ( 0.1 mL) were calculated from the weights of the solutions in the vials. About 0.1 g of Fe/Pd (from a freshly prepared batch) was delivered to each vial in a set at a time, and the set of vials was immediately placed on a tumbler mixer (at 30 rpm) as the dechlorination experiment started. After 10 min of dechlorination, 1.5 mL of solution was taken from each vial in a set into a centrifuge bottle, to which a volume of 1.5 mL cosolvent was added (for extraction). The remaining solution in the vials was discarded using a magnetic bar to attract and keep Fe/Pd particles at the bottom of the vial; a volume of 4.5 mL cosolvent was added to the vials for the extraction of compounds adsorbed on Fe/Pd particles and on the glass wall. The same amount of Fe/Pd particles was consistently delivered to each vial in pieces of plastic tubing (of the same length) filled with the Fe/Pd particles. A 3.18 mm ID and 6.35 mm OD polytetrafluoroethylene (PTFE) tubing was first marked (with a shallow cut) for as many pieces of units that were needed; each piece was 0.4 cm in length. The tubing was then filled with freshly prepared Fe/Pd and cut in pieces as marked; the amount of Fe/Pd in one piece was delivered to one vial. All pieces of empty tubing (after air-dried) were weighed, and the actual amount of Fe/Pd in a piece (delivered to one vial) was calculated based on the weight of the piece, the total weight of all the pieces, and the total amount of Fe particles (before acid wash). 3.4. Analytical Methods. All extractions were conducted at 35 °C by placing these vials and bottles in a shaker for 18 h. The top extracts (0.5 mL) were transferred to 2 µL GC vials from extraction bottles and vials after the bottles were centrifuged at 1350 rpm for 10 min, and the particles in the vials were allowed to settle for 2 h. After being injected with a 10 µL internal standard, the extracts were analyzed using the HP 5980 GC/MS based on the EPA SW-846 method 8270C for semivolatile organic compounds such as PCBs (http:// www.epa.gov/epaoswer/ hazwaste/test/8_series.htm). The GC/MS was equipped with a Supelco SPB-5 30 m × 0.32 µm × 0.25 µm column. Calibration curves were developed based on six-point data with coefficients of determination (r (2)) of greater than 0.998; the method’s detection limit was 50 µg L-1 for BP and 2-ClBP. The efficiency of extraction of BP and 2-ClBP on Fe/Pd and in the solution was 85-95%. The palladium samples were analyzed using an IRIS Intrepid ICP spectrometer (Thermo Elemental). The detection limit was 0.5 mg L-1. VOL. 42, NO. 18, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6943
TABLE 1. Rate constant of 2-ClBP Dechlorination by Fe/Pd at Different pH in Three Experimentsa pH
k1 (L g-1 h-1)b
pH
k1 (L g-1 h-1)c
pH
k1 (L g-1 h-1)d
3 3 3.5 4 5 6 7 8 9 10 11 11 12 13 14
0.78 ( 0.08 0.74 ( 0.13 0.66 ( 0.08 0.51 ( 0.10 0.41 ( 0.05 0.41 ( 0.07 0.40 ( 0.08 0.37 ( 0.03 0.36 ( 0.05 0.33 ( 0.04 0.21 ( 0.05 0.22 ( 0.03 0.053 ( 0.011 0.077 ( 0.014 0.19 ( 0.07
3 3.3 4.5 4.75 5.5 5.5 6.5 7 7.5 8.5 9.5 10.5 11.5 12.5 13.5
0.729 ( 0.098 0.635 ( 0.065 0.307 ( 0.042 0.236 ( 0.039 0.218 ( 0.028 0.199 ( 0.034 0.193 ( 0.027 0.207 ( 0.017 0.183 ( 0.023 0.148 ( 0.023 0.145 ( 0.029 0.150 ( 0.025 0.056 ( 0.016 0.033 ( 0.013 0.083 ( 0.025
11 11.5 12 12.5 13 13.5 14
0.033 ( 0.008 0.029 ( 0.010 0.030 ( 0.011 0.040 ( 0.014 0.050 ( 0.012 0.065 ( 0.026 0.095 ( 0.060
a All experiments were conducted at 23.3 °C for 10 min for the dechlorination of 2-ClBP using Fe/Pd from three different batches. The target Pd content was 0.585% based on Fe. The averaged values of the rate constant k1 and their 95% confidence intervals were calculated methods developed in ref 4. The number of data points (N) for each k1 value is 5; the initial concentrations of 2-ClBP were between 2 and 16 µm. The solution pH was adjusted by adding NaOH or HCl. b Detailed results were summarized in Supporting Information Table SI 1. c Detailed results were summarized in Supporting Information Table SI 2 d Detailed results were summarized in Supporting Information Table SI 3.
3.5. Polarization Curves of Iron in Various pH Solutions. The polarization curves of iron were obtained using a rotating disk electrode (RDE) in various pH solutions in a standard 125 mL glass cell. The RDE was an iron disk (model AFD050P040Fe, Pine Instrument Company, PA) with a 5 mm diameter with a geometric surface area of 0.1963 cm2. It was inserted inside an AFE3 M shaft (Pine Instrument), which was mounted on a MSR analytical rotator with a MSRX speed control (model AFMSRX, PINE Instrument). A coiled platinum wire 0.5 mm in diameter (BAS MW-1033, Bioanalytical Systems, Inc. IN) was used as the counter electrode. The reference electrode was an Ag/AgCl electrode (model MF2052, Bioanalytical Systems). Solutions with a pH less than 7 were prepared using HCl, whereas solutions with a pH greater than 7 were prepared using NaOH. Milli-Q water was used for the pH 7 solution. All solutions were oxygen-free through nitrogen-bubbling for 8 h. Before each experiment, the iron disk was polished with a BAS polishing alumina solution (CF-1050, Bioanalytical Systems). The speed of the RDE was 1000 rpm. Electrode potentials were controlled and recorded with an electrochemical workstation BAS Epsilon-EC (Bioanalytical Systems). Sweeping analyses were performed in the solutions at potentials ranging from -2200 to +1200 mV at a speed of 5 mV s-1. The potential and current were recorded at a rate of 10 Hz. All potentials are reported relative to the Ag/AgCl electrode (3 M NaCl, 0.209 V vs NHE), and cathodic currents were reported as positive.
4. Results 4.1. Dechlorination of 2-ClBP by Fe/Pd in Various pH Solutions. Three experiments were conducted for the dechlorination of 2-ClBP in various pH solutions using Fe/ Pd from three batches. For the preparation of each Fe/Pd batch, the palladium in the ethanol solution was completely deposited on the iron surface in 2 h, as supported by the evidence that the Pd concentration in the final solutions was always below the detection limit. The actual Pd loadings could vary slightly from batch to batch as a small amount of fine Fe particles (during acid washing) and Fe/Pd particles (following Pd deposition and during water washing) were lost (but not quantified). The amount of Fe/Pd delivered to each vial in the experiments was very consistent, averaging about 0.100 g of Fe with an average standard deviation of less than 0.005 (based on the amounts of Fe/Pd delivered to 6944
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 18, 2008
FIGURE 1. The rate constant k1 for 2-ClBP dechlorination in different pH solutions using two sets (∆ O) of Fe/Pd particles. Error bars are 95% confidence intervals. The dependence of k1 on pH indicates four distinct regimes: the low pH regime (pH < 5.5), the mid pH regime (5.5 < pH < 9.5), the mid high pH regime (9.5 < pH < 12.5), and the high pH regime (12.5 < pH). all the vials in each experiment). The solution pH was not buffered in all the experiments to prevent any effects of buffers on iron corrosion kinetics or even mechanisms (14, 15). The solutions were randomly selected for pH measurement at the end of dechlorination experiments, and the final pH values (presented in Supporting Information Tables SI 1 and SI 2) remained constant in high and low pH solutions, varying by less than 0.5 in solutions of between pH 6 and 8. The results of the 10 min-dechlorination of 2-ClBP in various pH solutions in three experiments using Fe/Pd particles from three different batches are summarized in Supporting Information Tables SI 1, 2, and 3. The amounts of 2-ClBP converted to BP, between 2.3 and 29% of 2-ClBP (based on initial amounts of 2-ClBP), varied according to the pH of the solutions. The conversion values were calculated by dividing the total amount of BP at the end by the initial amount of 2-ClBP. Although the conversion values at similar pH differ in these experiments, the trends shown from the results of the three experiments are consistent. (These experiments were conducted by different technicians at different times several months apart. The differences could be a combined result of Pd loadings, Fe particles sizes, acid washing, and personal skills.) In each experiment, the
TABLE 2. Corrosion Potential and Exchange Current in Different Ph Solutionsa pH 3 3 3.3 3.5 3.75 4 4.5 4.75 5 5.5 5.5 6 6.5 7 7 7 7.5 8 8.5 8.5 9 9.5 10 10.5 11 11.5 12 12.5 13 13.5 14
Ec (V)
b
-0.53 -0.54 -0.44 -0.44 -0.31 -0.25 -0.23 -0.20 -0.22 -0.20 -0.20 -0.18 -0.21 -0.19 -0.18 -0.14 -0.20 -0.18 -0.20 -0.18 -0.23 -0.24 -0.31 -0.34 -0.40 -0.50 -0.62 -0.65 -0.76 -0.87 -1.08
jc (A m-2) 5.40 ( 0.05 6.51 ( 0.03 3.88 ( 0.01 2.48 ( 0.01 0.85 ( 0.01 1.19 ( 0.00 0.33 ( 0.00 0.12 ( 0.00 0.14 ( 0.00 0.04 ( 0.00 0.05 ( 0.00 0.05 ( 0.00 0.04 ( 0.00 0.03 ( 0.00 0.06 ( 0.00 0.02 ( 0.00 0.03 ( 0.00 0.03 ( 0.00 0.03 ( 0.00 0.03 ( 0.00 0.03 ( 0.00 0.03 ( 0.00 0.10 ( 0.00 0.19 ( 0.00 0.52 ( 0.02 0.90 ( 0.01 1.93 ( 0.03 3.54 ( 0.05 8.39 ( 0.31 15.2 ( 1.1 66.0 ( 9.0
β (V-1)
c
0.969 ( 0.017 0.899 ( 0.010 0.776 ( 0.009 0.796 ( 0.006 1.197 ( 0.022 0.559 ( 0.007 0.650 ( 0.005 1.500 ( 0.080 0.910 ( 0.014 0.910 ( 0.140 0.910 ( 0.000 0.900 ( 0.029 0.600 ( 0.004 0.700 ( 0.041 1.100 ( 0.101 0.800 ( 0.232 0.700 ( 0.087 1.995 ( 0.033 0.800 ( 0.004 1.580 ( 0.004 2.163 ( 0.009 2.350 ( 0.011 7.000 ( 0.042 4.220 ( 0.008 7.000 ( 0.287 5.780 ( 0.040 4.650 ( 0.075 3.660 ( 0.021 3.900 ( 0.039 4.651 ( 0.111 4.500 ( 0.276
b (A m-2)
c
3.236 ( 0.034 4.005 ( 0.019 2.755 ( 0.017 1.746 ( 0.012 1.234 ( 0.017 1.029 ( 0.013 0.373 ( 0.028 0.146 ( 0.018 0.162 ( 0.005 0.053 ( 0.014 0.061 ( 0.010 0.057 ( 0.014 0.043 ( 0.014 0.035 ( 0.014 0.069 ( 0.007 0.025 ( 0.015 0.033 ( 0.015 0.037 ( 0.010 0.036 ( 0.017 0.132 ( 0.008 0.238 ( 0.007 0.042 ( 0.013 0.359 ( 0.007 0.212 ( 0.004 1.185 ( 0.013 0.786 ( 0.006 1.586 ( 0.007 2.303 ( 0.006 3.316 ( 0.016 3.521 ( 0.020 2.382 ( 0.054
limiting currentd C C C C C C A A A A A A A A A A A A A A A A A A A A A A A A A
a All experiments were conducted using a rotating disk electrode of iron at 1000 rpm. The applied potential changed from negative to positive at a rate of 5 mV s-1. Solutions with a pH of less than seven were adjusted using HCl; solutions with a pH of greater than seven were adjusted using NaOH. Pure milli-Q water was used for the solution with a pH of 7. b Ec is the corrosion potential; the uncertainty of the values is less than 5 mV. jc is the exchange current with its 95% confidence intervals, which were calculated from eq 14. c β is the slope of the Tafel line of limiting current with its 95% confidence interval; b is the intercept of the Tafel line of the limiting current at Ec ) 0. d A and C, respectively, denote that the corrosion current is anodicly and cathodicly limited.
conversion values decreased in general, with increasing pH in solutions with a pH lower than 12.5. This trend is consistent with the percentages of total biphenyls adsorbed on Fe/Pd and with the 2-ClBP percentages among the adsorbed biphenyls. The conversion decrease in solutions of pH between 5.5 and 9.5 was not statistically significant, but the decrease was dramatic in solutions with a pH higher than 11, which was accompanied by significantly increased amounts of 2-ClBP adsorbed on Fe/Pd, indicating a dramatic decrease in the activity of the catalyst in these solutions. In solutions with a pH higher than 12.5, the conversion values exhibited a significant increase, which was accompanied by less 2-ClBP being adsorbed on Fe/Pd. Such an increase was observed in all three dechlorination experiments, which consistently indicates that the catalyst exhibited enhanced activity in solutions with pH higher than 12.5. As confirmed in our previous study (4), the dechlorination of 2-ClBP by Fe/Pd followed first-order kinetics; Table 1 lists values of the first-order rate constant k1, defined in ref 4, for 2-ClBP dechlorination in different pH solutions in the three experiments. Each k1 value was calculated, according the method presented in ref 4, from the results of dechlorination in the five solutions of a given pH in a set of five vials. Figure 1 presents the k1 values (against pH) along with their 95% confidence intervals in the first and second experiments. Figure 1 indicates that four regimes likely existed in which the solution’s pH level played different roles in the catalytic dechlorination of 2-ClBP by Fe/Pd. There was clear dependence of the k1 values on pH in solutions of pH < 5.5 and of pH > 9.5. The k1 values in solutions of pH between 5.5 and
9.5 were not affected by pH. In addition, the k1 values exhibited a significant increase in solutions of pH higher than 12.5, which is consistent with the conversion data. The results in solutions of pH below 10 are consistent with published observations of dechlorination using zerovalent iron (21-24). 4.2. Corrosion Potential and Current at Different pH. The corrosion potential (Ec) is the potential applied at the iron RDE when the polarization curve (the net current) is zero. For all the polarization curves, the negative-to-positive sweeping was adopted so that the hydrogen evolution took place (when the iron surface was untainted) before the ZVI disk electrode was oxidized. (The difference in Ec values obtained from the forward and backward sweepings was less than 5 mV.) The sweeping rate of 5 mV/s was used because (a) a slower rate of 2 mV/s resulted in an Ec value that differed by only 5 mV and (b) the time to complete the entire sweeping was about 11 min, which is comparable with the time (10 min) for the dechlorination experiments. Table 2 summarizes the important results abstracted from the polarization curves. Figure 2a plots the Ec values against the solution pH. The Ec values decreased as the pH increased from 3 to 5.5, then remained relatively unchanged until pH 9.5, and finally increased as the solution pH increased to pH 14. The corrosion current (jc) was obtained using the Tafel lines of a polarization curve. In general, two Tafel lines were obtained from a polarization curve, one approaching from each side of the corrosion potential. From the negative side, the Tafel line (eqs 9 and 11) of the (cathodic) hydrogen evolution curve is used, with a negative value of slope. From VOL. 42, NO. 18, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6945
of pH, between 5.5 and 9.5, both Ec and jc seemed stabilized as the solution pH changed. In the mid high and high pH regimes, both Ec and jc increased significantly as pH increased from 9.5 to 14.
5. Discussion The values of jc and Ec in the low pH regime and in the mid high and high pH regime follow Tafel eqs 15 and 16, respectively. jc ) 0.012exp(-12.23Ec), R2 ) 0.883(N ) 11), 3 < pH < 5.5 (15) jc ) 0.009exp(-8.717Ec), R2 ) 0.969(N ) 10), 9.5 < pH < 14 (16)
FIGURE 2. (a) Corrosion potential Ec and (b) corrosion current jc of iron in solutions of different pH. Clearly, three regimes of dependence of jc on pH emerge: the low pH regime (pH e 5.5), the mid pH regime (5.5 < pH < 9.5), and the mid high and high pH regimes (pH g 9.5). Error bars indicate the 95% confidence intervals of jc. the positive side, the Tafel line (eqs 10 and 12) of the (anodic) iron oxidation curve is used, with a positive value of slope. The intercepts of the two Tafel lines at Ec are usually not the same, as otherwise indicated theoretically in eqs 9 and 10. The corrosion current jc (A m-2) is the lesser of the two intercept values at Ec. When jcathodic (Ec) < | janodic (Ec) |, the corrosion current is cathodicly controlled (limited), and when jcathodic (Ec) > | janodic (Ec) |, it is called anodicly controlled (limited). jcathodic(E) ) bcexp(-βcE)
(11)
janodic(E) ) -baexp(βaE)
(12)
jc ) min[jcathodic(Ec), |janodic(Ec)|]
(13)
where E (V) is the applied potential and β and b are the slope and the intercept at Ec ) 0 of the limiting anodic or cathodic current at the corrosion potential. The 95% confidence interval of jc was calculated using eq 14. (jc ) ((b)exp(-βEc) + exp(-β(∆Ec)) + |exp(- (β + ((β))Ec) - exp(-βEc)| (14) where (β and (b are the 95% confidence intervals of β and b values of the limiting current and ∆Ec is the confidence interval of Ec, which is 5 mV. The values of the limiting β and b as well as the limiting process are listed in Table 2. Figure 2b presents the values jc at different pH levels. Corresponding to the Ec values, the jc values varied according to pH. Figure 2 indicates three distinct regimes for the complicated process of corrosion of iron in different pH solutions. In the low pH regime, the values of Ec and jc decreased as pH increased from 3 to 5.5. In the mid regime 6946
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 18, 2008
The iron corrosion under each (Ec, jc) state was at electrochemical equilibrium, in which the rate of iron oxidation was electrochemically balanced with the rate of hydrogen evolution. The potential sweeping results indicate that the corrosion equilibrium was controlled by the cathodic reaction of hydrogen evolution in solutions of pH below 4 and by the anodic oxidation of iron in solutions of pH above 4.5. The k1 values fall in the same range as those obtained in both short-term and long-term experiments in our previous studies (3, 4). They are of the same magnitude as those for the dechlorination of monochlorophenols and p-dichlorobenzene (25, 26) and for the dechlorination of PCBs (2, 27). Furthermore, attributing to the short-term experiment method, the k1 values obtained in this study carry small variations, thereby differentiating the impact of pH on the rates of dechlorination in different pH regimes. 5.1. k1 versus Ec and jc in Low pH Regime. The k1 values at pH between 3 and 5.5 present a linear and logarithmic relationship with the Ec and jc values, respectively, which is indeed consistent with the Tafel relationship between jc and Ec (eq 15). Figure 3a plots k1 versus Ec for the two experiments of dechlorination in solutions of pH e 5.5. The remarkable linear (rather than exponential) relationship suggests that the dechlorination of 2-ClBP by Fe/Pd was not a direct reduction, which is consistent with our previous studies (1, 3, 4). However, it is very surprising that a linear relationship between k1 and jc was not observed. It was expected that the rate of PCB hydrodechlorination by Fe/Pd would be related to the rate of hydrogen generation (which is proportional to jc) and that k1 would be related linearly to jc. Careful investigation indicates that the hydrodechlorination is proportionally related to the hydrogen activity (or concentration) at the Fe/Pd surface (rather than to the rate of hydrogen production). It is well-known that the solubility (or concentration) of hydrogen in palladium is directly proportional to the square root of the hydrogen pressure (concentration) in the bulk phase (28). Therefore, an attempt was made to correlate k1 with the square root of jc, as shown in Figure 3b. The remarkable linear relationship is well established between k1 and jc0.5. The rate of PCB dechlorination is related to the hydrogen activity at the Pd surface, which is proportional to the square root of the hydrogen concentration in the bulk phase, and the hydrogen concentration is proportional to hydrogen production rate, which is directly proportional to jc. 5.2. k1 versus Ec and jc in mid pH Regime. In solutions of pH between 5.5 and 9.5, there is not statistically significant correlation between k1 and Ec or jc. The Ec values averaged -0.187, with a standard deviation of 0.028, and jc values averaged 0.036, with a standard deviation of 0.012 (based on 13 data points). The values of Ec and jc follow approximately eq 16, which was established based on the values of Ec and
FIGURE 3. The k1 values versus the values of Ec (a) and jc0.5 (b) in solutions of pH between 3 and 5.5. Marks (∆ O) denote data obtained in two dechlorination experiments using different batches of Fe/Pd. Error bars indicate the 95% confidence intervals of the k1 values. jc in solutions of pH between 9.5 and 14. The k1 values remained very close to each other. For the k1 values of the two experiments presented in Figure 1, the averages of k1 (with the standard deviation) were 0.968 ( 0.045 (N ) 4) and 0.447 ( 0.068 (N ) 7), where N is the number of k1 values in the mid pH regime. The pH impact on both dechlorination and corrosion was not observed in this pH regime. The result indicates that the limiting anodic oxidation (see previous section) was not eqs 2a and 2b. Instead, the limiting process was the oxidation of iron to an intermediate product (iron hydroxide), Fe + 2 H2O ) Fe(OH)2 + H+ + 2 e-. The overall reaction was Fe + 2 H2O ) Fe(OH)2 + H2. Protons were directly involved in both cathodic and anodic half-reactions, but the effects on the half-reactions were canceled out in the overall reaction. As a result, the rate of dechlorination and the values of Ec and jc do not reveal significant correlation to pH. 5.3. k1 versus Ec and jc in Mid High and High pH Regimes. Figure 2 shows that the jc values increased exponentially as the Ec values became more negative in solutions of pH between 9.5 and 14. However, Figure 1 shows that the k1 values exhibit a clear trough in solutions of pH between 12 and 13. The mismatch between the kinetics data and the corrosion potential is understandable because the obvious difference between the rotating disk electrode of iron used in corrosion potential measurements and the Pd/ Fe particles used in the PCB dechlorination reaction. Nevertheless, that trough divides the pH ranges into the mid high pH (between 9.5 and 12.5) and high pH (above 12.5) regimes. The presence of such a trough was indicative of different corrosion mechanisms of iron in these two regimes. In the mid high pH regime, iron corrosion resulted in formation of solid Fe(OH)2 layer on the iron surface and thus was limited by the surface passivation. In the high pH regime, iron corrosion resulted in soluble HFeO2 ions (http:// corrosion-doctors.org/Corrosion-Thermodynamics/ PotentialpH-diagram-iron.htm). Therefore, the rate of corrosion process, which was not limited by surface passivation, is positively related to jc and |Ec|, their values increased simultaneously with pH. Accordingly, the k1 values increased as the pH increased beyond 12.5, which were observed in all the three experiments (Table 1). 5.4. Significance in Iron-Related Remediation. This study has made a significant step toward understanding the complicated process of PCB dechlorination by Fe/Pd. The correlations between the dechlorination rate with corrosion potential and current have suggested four possible (controlling) mechanisms for the dechlorination of 2-ClBP by Fe/Pd in four different pH regimes. The rate-determining factors may include the electric potential at the Fe/Pd surface (rather than the Eh of the solutions), the limiting electrochemical
(electron transfer) process, and the starting species (iron or magnetite) and resulting products (ferrous or iron oxides and hydroxides) of the limiting process in addition to palladium coverage (sites for hydrogen evolution and hydrodechlorination). The understanding obtained from this study may lead future research in this field to examine each pH regime for further detailed investigation of the complicated processes involved in the dehalogenation of halogenated compounds by zerovalent iron and iron bimetals.
Acknowledgments This research was funded and conducted by the National Risk Management Research Laboratory of the U.S. Environmental Protection Agency, Cincinnati, OH. This paper has not been subjected to internal policy review of the U.S. Environmental Protection Agency. Therefore, the opinions presented herein do not, necessarily, reflect the views of the Agency or its policy. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. We thank Mr. Eric Graybill of Pegasus Technical Services (PTS), Inc., and Mr. Yufu Liang formerly with PTS for conducting part of the experiments.
Supporting Information Available Results of 2-ClBP dechlorination by Fe/Pd in solutions of different pH values. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Grittini, C.; Malcomson, M.; Fernando, Q.; Korte, N. Rapid dechlorination of polychlorinated biphenyls on the surface of a Pd/Fe bimetallic system. Environ. Sci. Technol. 1995, 29 (11), 2898–2900. (2) West, O. R.; Liang, L.; Korte, N. E.; Fernando, Q.; Clausen, J. L. Degradation of Polychlorinated Biphenyls (PCBs) Using Palladized Iron, ORNL/TM-13217; Oak Ridge National Laboratory: Oak Ridge, TN, 1996. (3) Fang, Y.; Al-Abed, S. R. Partitioning, Desorption, and Dechlorination of a PCB Congener in Sediment Slurry Supernatants. Environ. Sci. Technol. 2007, 41, 6253–6258. (4) Fang, Y.; Al-Abed, S. R. Dechlorination kinetics of monochlorobiphenyls by Fe/Pd: Effects of solvent, temperature, and PCB concentration. Appl. Catal., B 2008, 78, 371–380. (5) Lowry, G. V.; Johnson, K. M. Congener-specific dechlorination of dissolved PCBs by microscale and nanoscale zerovalent iron in a water/methanol solution. Environ. Sci. Technol. 2004, 38 (19), 5208–5216. (6) Wang, C.-B.; Zhang, W.-X. Sythesizing nanoscale iron particles for rapid and complete dechlorination of TCE and PCBs. Environ. Sci. Technol. 1997, 31 (7), 2154–2156. (7) Zhang, W.-X. Nanoscale iron particles for environmental remediation: An overview. J. Nanopart. Res. 2003, 5 (3-4), 323– 332. VOL. 42, NO. 18, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6947
(8) Kim, Y.-H.; Shin, W. S.; Ko, S.-O. Reduction dechlorination of chlorinated biphenyls by palladized zero-valent metals. J. Environ. Sci. Health., Part A 2004, 39 (5), 1177–1188. (9) Wise, E. M. Palladium: Recovery, Properties, and Users; Academic Press: New York, 1968. (10) Cheng, I. F.; Fernando, Q.; Korte, N. Electrochemical dechlorination of 4-chlorophenol to phenol. Environ. Sci. Technol. 1997, 31 (4), 1074–1078. (11) Fang, Y.; Al-Abed, S. R. Palladium-faciliatated electrolytic dechlorination of 2-chlorobiphenyl using a granular-graphite electrode. Chemosphere 2006, 66, 226–233. (12) Schuth, C.; Reinhard, M. Hydrodechlorination and hydrogenation of aromatic compounds over palladium on alumina in hydrogen-saturated water. Appl. Catal., B 1998, 18, 215–221. (13) Matheson, L. J.; Tratnyek, P. G. Reductive dehalogenation of chlorinated methanes by iron metal. Environ. Sci. Technol. 1994, 28 (12), 2045–2053. (14) Zawaideh, L. L.; Zhang, T. C. The effects of pH and addition of an organic buffer (HEPES) on nitrate transformation in Fe0water systems. Water Sci. Technol. 1998, 38 (7), 107. (15) Choe, S.; Liljestrand, H. M.; Khim, J. Nitrate reduction by zerovalent iron under different pH regimes. Appl. Geochem. 2004, 19 (3), 335–342. (16) Chen, J.-L.; Al-Abed, S. R.; Ryan, J. A.; Li, Z. Effects of pH on dechlorination of trichloroethylene by zero-valent iron. J. Hazard. Mater. 2001, 83 (3), 243–254. (17) Zhu, B.-W.; Lim, T.-T. Catalytic reduction of chlorobenzenes with Pd/Fe nanoparticles: Reactive sites, catalyst stability, particle aging, and regeneration. Environ. Sci. Technol. 2007, 41, 7523–7529. (18) Rieger, P. H. Electrochemistry. 2nd ed.; Chapman & Hall, Inc.: New York, 1994.
6948
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 18, 2008
(19) Bard, C. V.; Faulkner, L. R. Electrochemical Methods; John Wiley and Sons: New York, 1980. (20) Zhang, W.-X.; Wang, C.-B.; Lien, H.-L. Treatment of chlorinated organic contaminants with nanoscale bimetallic particles. Catal. Today 1998, 40, 387–395. (21) Li, T.; Farrell, J. Reductive dechlorination of trichloroethene and carbon tetrachloride using iron and palladized-iron cathodes. Environ. Sci. Technol. 2000, 34, 173–179. (22) Liu, Y.; Lowry, G. V. Effect of particle age (Fe0 content) and solution pH on NZVI reactivity: H2 evolution and TCE dechlorination. Environ. Sci. Technol. 2006, 40, 6085. (23) Liu, Y.; Phenrat, T.; Lowry, G. V. Effect of TCE concentration and dissolved groundwater solutes on NZVI-promoted TCE dechlorination and H2 evolution. Environ. Sci. Technol. 2007, 41, 7881–7887. (24) Song, H.; Carraway, E. R. Reduction of chlorinated ethanes by nanosized zero-valent iron: Kinetics, pathways, and effects of reaction conditions. Environ. Sci. Technol. 2005, 39, 6237–6245. (25) Liu, Y.; Yang, F.; Yue, P. L.; Chen, G. Catalytic dechlorination of chlorophenols in water by palladium/iron. Water Res. 2001, 35 (8), 1887–1890. (26) Xu, X.; Zhou, H.; He, P.; Wang, D. Catalytic dechlorination kinetics of p-dichlorobenzene over Pd/Fe catalysts. Chemosphere 2005, 58 (8), 1135–1140. (27) Korte, N. E.; West, O. R.; Liang, L.; Gu, B.; Zutman, J. L.; Fernando, Q. The effect of solvent concentration on the use of palladizediron for the step-wise dechlorination of polychlorinated biphenyls in soil extracts. Waste Manage. 2002, 22 (3), 343–349. (28) Lacher, J. A theoretical formula for the solubility of hydrogen in palladium. Proc. R. Soc. London, Ser. A 1937, 161 (907), 525–545.
ES800805Y
