Efficient Heterogeneous Catalytic Reduction of Perchlorate in Water

Feb 15, 2007 - Clearly, nitrate reduction is a much faster reaction with the Pd−Cu catalyst than is perchlorate reduction with the ReO−Pd catalyst...
6 downloads 18 Views 127KB Size
Download PDF
Environ. Sci. Technol. 2007, 41, 2044-2049

Efficient Heterogeneous Catalytic Reduction of Perchlorate in Water KEITH D. HURLEY AND JOHN R. SHAPLEY* Center of Advanced Materials for the Purification of Water with Systems and Department of Chemistry, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801

A new heterogeneous catalyst that promotes the reduction by hydrogen of perchlorate ion in water under mild conditions has been developed. The catalyst is prepared by adsorption of a rhenium(VII) precursor (either ammonium perrhenate or methylrhenium trioxide) onto carbon powder containing 5% palladium by weight. Under standard batch conditions of room temperature, 1 bar of hydrogen, and 200 ppm perchlorate (as HClO4), reduction proceeded to less than 1 ppm in as little as 5 h. Extended reaction times led to residual perchlorate at low parts per billion levels. Chloride was the only observed product, with good material balance. Catalytic materials ranging from 3% to 13% Re showed (pseudo) first-order rates linearly dependent on the Re content. Representative normalized rate constants for catalysts with 5-9% Re were in the range 0.1-0.3 L h-1 (g of cat.)-1. Inhibition by chloride was not significant, with little change in perchlorate reduction rate in the presence of excess chloride to 1000 ppm. However, optimal activity occurred in acidic solutions (pH ca. 3), and both the rate and extent of reaction decreased at higher values of pH. In its current form the catalyst might be best applied to destroy perchlorate in the acidic regeneration stream from selective ion exchange columns.

Introduction The perchlorate ion (ClO4-) has been reported as a contaminant in surface water, drinking water wells, and soil in up to 35 states and Puerto Rico (1). Studies have also detected perchlorate in commercial samples of lettuce and milk (2) as well as in human breast milk (3). High local concentrations of perchlorate have been associated with the manufacture, handling, or use of ammonium perchlorate as an oxidant in rocket fuel, munitions, or blasting materials (4). Since perchlorate salts are characterized by high water solubility, low adsorptive capacity, and kinetic inertness, perchlorate can easily be spread widely and be quite persistent in surface water and groundwater systems. A recent analytical focus has been on collecting information to distinguish anthropogenic sources from possible natural sources of perchlorate in the environment (5), and the relative impact of various sources on the food chain has been evaluated (6). High levels of perchlorate have been shown to interfere with iodide uptake into the thyroid gland; however, quantifying the effects of continuing low levels of perchlorate exposure is difficult, especially for the most sensitive populations (7, 8). For adults, thyroid hormones regulate metabolic activity, but for infants and fetuses, the thyroid * Corresponding author phone: (217) 333-0297; fax: (217) 2443186; e-mail: [email protected]. 2044

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 6, 2007

has a major role also in normal central nervous system development and skeletal growth (4, 8). Since the most likely way for pregnant women and mothers to ingest perchlorate is through contaminated drinking water, the U.S. EPA has set an official reference dose (RfD) of 0.0007 (mg of perchlorate/kg of body weight)/day, which translates to a drinking water equivalent level (DWEL) of 24.5 ppb (4, 8). However, since this is not an enforceable federal maximum contaminant limit (MCL), various states are considering their own specific standards. Massachusetts recently became the first state to set a legal MCL, in this case at 2 ppb (9), and California has a proposed MCL of 6 ppb (10). A variety of perchlorate treatment technologies are available and have been used in either pilot-scale or fullscale remediation cases (1, 4, 11). These technologies fall into two general categories: (i) physical/chemical, such as ion exchange, granular activated carbon, and electrodialysis, and (ii) biological, such as bioreactors, composting, in situ bioremediation, and phytotechnology. The physical/chemical approaches are ex situ methods that provide for more or less efficient separation of perchlorate from the treated water; complete destruction of the perchlorate, however, requires a subsequent, generally high-temperature, procedure. Phytoremediation is also fundamentally a separation method, since the perchlorate-loaded biomass requires a separate disposal stage. Bioreaction systems, in contrast, take advantage of naturally occurring microorganisms to enzymatically catalyze the overall reduction of perchlorate to chloride and oxygen (12). The microbes are induced to grow on sand, carbon, plastic, or glass supports in various ex situ bioreactor configurations or enhanced in the soil for in situ approaches. A reducing agent (e.g., hydrogen, ethanol, acetate, or lactate) must be added continuously as a food source for the microorganisms (13). Advantages cited for bioreactor systems over ion exchange systems, especially for remediating highlevel contamination sites, include lower overall cost and the possibility of in situ application (4, 11). On the other hand, the biological systems can be especially sensitive to the incoming water chemistry (redox potential, oxygen concentration), require closer supervision, and cannot be operated intermittently. Furthermore, biological purification for drinking water necessarily requires posttreatment disinfection steps and, despite preliminary regulatory approval in California, might face debilitating public opposition from perceived fears of pathogens (11). Ion exchange is the only entry in the list of best available technologies (BATs) for achieving compliance with the new Massachusetts MCL (9), an indication of its widespread regulatory acceptance and history of successful application (1). However, ion exchange systems are also sensitive to influent water quality; high total suspended solids can clog the resin bed, and high total dissolved solids can provide competitive uptake by other anions (4). Furthermore, oncethrough resins require frequent replacement and disposal, adding to the complexity and cost. New anion exchange resins that are highly selective for perchlorate have been developed (14, 15), but these resins in turn require development of new chemistry for effective regeneration. Development of a chemical (i. e., nonbiological) catalyst for room-temperature reaction of perchlorate with a simple reducing agent compatible with drinking water would fill a major void in the current roster of perchlorate remediation strategies. There are several transition-metal ions, such as V(III), Ti(III), and Ru(II), that can slowly reduce perchlorate stoichiometrically in aqueous solution at room temperature 10.1021/es0624218 CCC: $37.00

 2007 American Chemical Society Published on Web 02/15/2007

TABLE 1. Catalytic Perchlorate Reduction Experiments expt no.a

[Re] (wt %)

[Pd] (wt %)

C(cat.) (mg/50 mL)

k(obsd) (h-1)

k(Re) (h-1 (mg of Re)-1)

k(cat.) (L h-1 (g of cat.-1)

1 2 3 4 5 6 7 8 9 10 11 12 13b 14c 15 16 17 18d 19e 20 21f 22f 23g 24 25h

0.69 0.69 0.69 2.91 2.91 2.91 5.30 5.30 5.30 5.39 5.39 5.68 5.68 5.68 5.86 5.86 5.86 5.86 7.19 9.41 9.41 9.41 9.41 13.05 13.05

5.83 5.83 5.83 5.60 5.60 5.60 5.34 5.23 5.34 5.44 5.44 5.28 5.28 5.28 5.26 5.26 5.26 5.26 4.46 5.59 5.59 5.59 5.59 4.71 4.71

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 59.4 100 50 25 200 100 100

0.0041 0.0029 0.0034 0.052 0.051 0.068 0.21 0.25 0.26 0.26 0.24 0.33 0.26 0.049 0.35 0.32 0.34 0.037 0.24 0.60 0.22 0.086 1.40 0.88 0.15i

0.0059 0.0042 0.0049 0.018 0.019 0.023 0.040 0.047 0.049 0.048 0.045 0.058 0.046 0.0086 0.060 0.055 0.058 0.0063 0.056 0.064 0.048 0.033 0.074 0.067 0.012

0.0021 0.0015 0.0017 0.026 0.026 0.034 0.11 0.13 0.13 0.13 0.12 0.17 0.13 0.025 0.18 0.16 0.17 0.019 0.20 0.30 0.22 0.17 0.35 0.44 0.075

a All experiments conducted under standard conditions: 1 bar of H , 25 °C, 2 mM HClO , 50 mL total volume, 100 mg of catalyst formed from 2 4 [NH4][ReO4] unless indicated otherwise. b A 1000 ppm concentration of chloride (as NaCl) added. c pH adjusted to 3.70 using NaOH. d pH adjusted e to 3.72 using NaOH. Catalyst formed from ReO3CH3 (MTO); reduction run in 100 mL of water with ∼120 mg of catalyst. f Reduced quantity of catalyst in reactor. g Increased quantity of catalyst in reactor. h pH adjusted to 3.72 using NaOH. i k(obsd) calculated for first half-life only.

(16, 17), and it has recently been shown that Fe(II) at high temperatures (140-170 °C) is an effective reducing agent (18). There is evidence that active metals, particularly zerovalent iron, can react with perchlorate, but the reactions under normal conditions are very slow (16, 19-23). Similarly, the catalytic or electrocatalytic reduction of perchloric acid with dihydrogen observed with noble metal powders or solid electrodes is extremely slow and inefficient under mild conditions (16, 22, 23). However, in 1995 Abu-Omar and Espenson (24) showed that a rhenium(V) complex in aqueous solution, formulated as methylrhenium dioxide (MDO), can react relatively rapidly with perchlorate by an oxygen atom transfer (OAT) reaction to form a rhenium(VII) complex (methylrhenium trioxide, MTO) and chlorate. A cycle could be closed by reducing the Re(VII) complex back to the Re(V) complex with hypophosphorous acid (H3PO2) (25). Subsequent work by Abu-Omar and co-workers (26-28) led to a Re(V) complex that was a more efficient perchlorate reduction catalyst in a polar organic solvent, but the oxygen acceptor was an organic sulfide. Such a homogeneous catalyst with a soluble phosphorus or sulfur reducing agent is not readily compatible with water purification systems. However, we have found that simple oxorhenium(VII) compounds, when dispersed on a carbon support along with Pd metal particles, form a new heterogeneous catalyst that efficiently promotes the complete transformation of aqueous perchlorate to chloride and water by using hydrogen as the reducting agent. In this paper we report the preparation of the catalyst together with aspects of its operation.

Experimental Section Chemicals. All experiments were conducted by using 18.2 MΩ cm-1 Milli-Q water generated in a Synergy 185 Millipore with a Simpak2 purifying system (Millipore, Billerica, MA). Tanks of hydrogen (H2, 99.995%) were supplied by Praxair (Danbury, CT). Sodium perchlorate, ammonium perrhenate ([NH4][ReO4]), and MTO (ReO3CH3) were purchased from

Sigma-Aldrich (St. Louis, MO). Sodium chloride, sodium hydroxide, and perchloric acid (70%, w/w) were purchased from Fischer Scientific. Palladium was purchased in the form of 5% Pd (w/w) on activated carbon (wet, Degussa type, E101) from Sigma-Aldrich. All chemicals were used as received without further purification unless otherwise noted. Catalyst Preparation. The Pd/C material as received was first dried at 110 °C for 1 h in air and then treated at 250 °C for 1 h under an atmosphere of flowing hydrogen, allowed to cool to room temperature, and stored in air. The catalyst was prepared by combining the treated Pd/C powder with the desired amount of either ammonium perrhenate or MTO and placing the mixture and a Teflon-coated stir bar in a 250 mL thick-walled glass bottle with 100 mL of 2 mM HClO4 in nanopure water. The reaction vessel was sealed with a pressure valve assembly, flushed with hydrogen, and then pressurized with hydrogen to 3.5 bar (1 bar ) 100 kPa) for 5 h at room temperature. Then the pressure bottle was vented, the contents were filtered, and the recovered black solid was first washed with nanopure water and then dried in air for 1 h at 110 °C. A sample of each catalyst prepared was selected for analysis to determine the Re and Pd contents; the results are given in Table 1. All catalyst materials were stored and handled with exposure to air. Reactor System. The perchlorate reduction reactions were carried out in a batch reactor at room temperature (determined periodically to be 23 ( 1 °C). The reactor was a threeneck, 250 mL round-bottom flask to which the perchlorate solution, the desired amount of powdered catalyst, and a Teflon-coated stir bar were added. Hydrogen was introduced to the system via the submerged tip of a glass tube seated in a rubber-O-ring-sealed adapter in one neck of the flask. A second neck was sealed with a rubber septum fitted with a 25 gauge needle to allow for gas outflow and flow rate monitoring. The final neck was sealed with a clamped glass stopper. Most of the experiments were conducted under a set of standard conditions, which consisted of 100 mg of VOL. 41, NO. 6, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2045

catalyst in 50 mL of a 2 mM HClO4 solution in nanopure water under 1 bar of flowing hydrogen. Aliquots (0.7 mL) of the reaction solution were removed by syringe at various times for subsequent analysis. Analytical Methods. Samples from reaction solutions were analyzed for perchlorate and chloride by ion chromatography, in most cases by using a Metrohm Basic 792 ion chromatograph with a Cetac AN1-SC column, equipped with a suppressed anion conductivity detector (1.9 mmol of NaHCO3/Na2HCO3 eluent, 0.8 mL/min flow rate, 20 µL sample loop); the detection limit for perchlorate was ca. 1 ppm. To determine chloride and perchlorate at low concentrations, a Dionex ICS-2000 ion chromatograph was used (Dionex IonPac AS16 column, 36 mM KOH eluent, 1 mL/min flow rate, 1000 µL sample injection loop), also with a suppressed anion conductivity detector; the detection limit with this instrument was 5 ppb. UV-vis spectra were obtained with an HP8452 diode-array spectrophotometer. Measurements of pH were conducted with a ThermoOrion 420 meter and standard pH electrode. The metals content of the solid catalysts was determined by ICP-MS in the Microanalysis Laboratory of the School of Chemical Sciences at the University of Illinois at Urbana-Champaign (UIUC). XPS spectra were obtained on a Kratos Axis ULTRA imaging X-ray photoelectron spectrometer by using monochromatized Al KR radiation with a hemispherical mirror analyzer. The dried and powdered solid materials were mounted on Cu-coated carbon tape, and the spectra were referenced to the position of the C 1s peak at 284.5 eV (29). The spectrometer is part of the Center for the Microanalysis of Materials at the Frederick Seitz Materials Research Laboratory at UIUC. Evaluation of Reaction Kinetics. Catalyst activity was evaluated by fitting the perchlorate concentration vs time data to the exponential form of a pseudo-first-order rate expression (eq 1). The “best” value of k(obsd) was determined

C(t) ) C(0) exp(-k(obsd)t)

(1)

by minimizing the rms deviations between the observed and calculated values of C(t). In most cases the final perchlorate concentration observed was 1 ppm or less. To normalize the observed rates for different amounts of catalytic material, a parameter k(cat.) was calculated by dividing k(obsd) by the concentration of the catalyst (g/L) actually employed in a given experiment. No term for hydrogen was included in the rate expression, since we assumed complete hydrogen equilibration with the palladium active sites under these conditions.

Results and Discussion Synthesis and Characterization of the Catalytic Materials. By monitoring the UV absorption bands of MTO in water (25), we found that suspended Pd/C powder under a hydrogen atmosphere caused rapid disappearance of MTO from the solution. Similar but slower behavior was observed for ammonium perrhenate. In either case, isolation of the black powder by filtration followed by drying it at 110 °C gave materials that proved to be active catalysts for perchlorate reduction with hydrogen. Qualitatively similar catalytic behavior was observed for materials prepared from either Re(VII) precursor; however, since MTO is ca. 10 times more expensive than the perrhenate salt, the latter precursor was adopted for most of the experiments shown in Table 1 (see experiment 19 for the MTO precursor). Elemental analyses of the prepared catalysts showed that all of the samples have a Pd content near the 5 wt % nominal value of the purchased Pd/C material as well as a varying Re content that depends on the amount of Re precursor used in the preparation and the amount of time allowed for adsorption. 2046

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 6, 2007

FIGURE 1. Typical perchlorate reduction and chloride formation profiles for ReO/Pd/C catalyst. Further physical characterization of the ReO/Pd/C catalytic materials is limited at this time. The palladium is expected to be present as nanocrystallites dispersed over the carbon surface, which will also be covered with varying amounts of an oxorhenium species deposited from solution during the process of catalyst synthesis (30). XPS data obtained on the catalyst materials (handled in air) are consistent with this picture. The 5 wt % Pd/C material exhibits a binding energy (BE) for the Pd 3d5/2 peak of 335.3 eV, consistent with handbook values for Pd(0) (29). For the catalysts (both from MTO and from ReO4-) the observed shift for the Pd 3d5/2 peak is +0.6 eV, suggesting a slightly more oxidized Pd/PdO surface, which could be due to oxoRe species present in close proximity to the Pd particles (30). A high-resolution scan of the Re 4f7/2 region revealed BE values of 45.5 and 45.4 eV for the MTO and perrhenate catalysts, respectively, consistent with the presence of Re in the +7 oxidation state (29). Operation of the Catalyst. In the presence of hydrogen at atmospheric pressure the new ReO/Pd/C materials efficiently catalyze the reduction of 200 ppm perchlorate (as perchloric acid) to chloride in a matter of hours. Figure 1 shows a typical profile for perchlorate loss, which usually was followed to a detection limit of ca. 1 ppm. Concomitant formation of chloride as the product is seen, and there is good material balance. This is consistent with the overall reaction given in eq 2. By using an ion chromatograph with

ClO4- + 4H2 a Cl- + 4H2O

(2)

a larger sample loop and a lower limit of detection, we determined that solutions exposed to the reaction conditions for longer periods of time can have perchlorate levels near or below the lower limit of detection of ca. 5 ppb. Used catalysts that were exposed to fresh batches of perchlorate showed qualitatively similar reduction behavior. All of the reduction experiments analyzed quantitatively are listed in Table 1. The perchlorate reduction profiles are well fit by a (pseudo) first-order dependence on the perchlorate concentration, proceeding in most cases to >99% reduction. Sample fits for catalytic materials with different Re loadings are shown in Figure 2. The good first-order fits to relatively high conversion imply that chloride does not inhibit the coordination and reduction of perchlorate at the catalyst surface. In fact, there is only a very modest decrease in reduction rate when the reaction is conducted in the presence of 1000 ppm added chloride, which is 5 times greater

FIGURE 4. Observed pseudo-first-order reduction rate constants as a function of catalyst concentration in the reactor. The specific data shown are for the 9.4% Re catalyst.

FIGURE 2. Reduction profiles and model fits at various Re loadings on the catalyst. In each case the catalyst concentration was 2 g/L in the reactor.

FIGURE 3. Observed pseudo-first-order rate constant for perchlorate reduction as a function of the weight percent of Re in the catalyst (concentration of the catalyst in the reactor, 2 g/L). The dashed line shows the linear correlation, excluding the data for the 0.7% Re catalyst. than the initial concentration of perchlorate (compare experiments 13 and 12 in Table 1 and further discussion below). The rhenium centers are clearly the key determinants of the perchlorate reduction process. As shown in Figure 3, the observed reduction rates are linearly dependent on the Re loading in the catalyst between ca. 3% and 13% by weight; the latter figure is determined by the maximum amount of ammonium perrhenate adsorbed under our preparative conditions. The slight lag in onset of the Re dependence may indicate a small amount of inactive rhenium species on the carbon surface. The observed reaction rate constants also may be normalized by the amount of Re on the catalyst (see Table 1), which gives relatively similar values for k(Re) of 0.02-0.06 h-1 (mg of Re)-1 under the standard conditions of 100 mg of catalyst in 50 mL of solution ()2 g/L). It is important to note that the catalyst prepared from MTO as the Re(VII) precursor (Table 1, experiment 19) has a value of k(obsd) that also fits this correlation; this specific value is included in the relation depicted in Figure 3. Qualitatively, we observed no difference in reduction rates between reactions conducted at 1 bar and at 3.5 bar of hydrogen pressure, which supports the expectation that there is a steady state (saturation) between the hydrogen atmosphere and adsorbed hydrogen on the catalyst surface. The lack of mass-transfer limitations on delivery of hydrogen to the catalyst is indicated by the linear dependence of k(obsd) on the amount of catalyst present in the reactor as shown in Figure 4. Values of the rate constants k(cat.), which are normalized for the actual concentration of the catalyst, are given in Table 1. The standard conditions we adopted for most reactions involves 200 ppm (2 mM) perchloric acid, which leads to pH 2.7 for the solution, since perchloric acid is a very strong acid

FIGURE 5. Effect on perchlorate reduction profiles due to a change in pH vs addition of excess NaCl (5.68% Re catalyst). Lines are drawn to aid the eye; for quantitative fit results see Table 1. and fully dissociated at this concentration. Under these conditions there is no difference in the measured pH before and after complete reduction. This is consistent with the stoichiometry given in eq 2, which does not involve any term in [H+]. However, when we raised the pH of the reaction solution by adding NaOH, the initial perchlorate reduction rate decreased. As indicated by data in Table 1, for catalysts with three different [Re] (%) values, a change in pH by 1 unit to 3.7 caused the corresponding observed rate constant to drop by a factor of 6-9 (cf. experiments 14 and 12, 18 and 17, and 25 and 24). This indicates a kinetic (catalytic) role for a proton in the perchlorate reduction mechanism. However, at even higher values of pH the overall reaction was incomplete; after an initial perchlorate reduction of ca. 50%, with corresponding chloride formation, further reduction was very slow or nonexistent. This observation suggests a structural role for protons in maintaining the integrity of the catalytic material. Figure 5 compares the reduction profile effects of increased pH vs increased chloride for the specific case of the 5.68% Re catalyst. Catalytic Mechanism. We suggest that the metal centers in this catalyst have two different modes of operation as shown in Scheme 1. The Pd component serves to bind and “activate” H2, forming adsorbed surface H, which can then “spill over” (31) to the Re-O species dispersed on the support. The reaction of ReVIIdO with 2H will reduce it to ReV(OH2), and then dissociation of the water ligand provides an open site on the rhenium center that can complex perchlorate. Oxygen atom transfer, as discussed thoroughly for the soluble cases (25-28), regenerates the original ReVIIdO center and releases chlorate. Subsequent reactions with chlorate, chlorite, and hypochlorite will complete the stepwise reduction VOL. 41, NO. 6, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2047

TABLE 2. Comparison of Metal-Based Perchlorate-Reducing Systems Operating at Room Temperature system ZVI + microbes ZVI + microbes nanoiron 5.9% Re(O) + 5.2% Pd/C

SCHEME 1. Reduction

reducing agent Fe(0), H2 Fe(0), H2 Fe(0) H2

pH 8.5 7 8 2.7

[Co(ClO4)] (mg/L) 0.5 65 200 200

Mechanistic Scheme for Catalytic Perchlorate

to chloride and water. In separate experiments we have shown that chlorate is reduced to chloride much faster than is perchlorate under our standard conditions (in solution the rate increase is 10-1000 times (24)), consistent with chlorate as an unobserved but probable intermediate. The pH dependence of the reduction rate suggests that a proton assists in the complexation and activation of perchlorate. Since perrhenic acid (ReO3OH; pKa ≈ -1.25) is a weaker acid than perchloric acid (ClO3OH; pKa ≈ -7) (32), an extra proton incorporated into a rhenium-perchlorate surface complex should be attached primarily to a more basic rhenium-bound oxygen atom. However, a hydrogen-bonding interaction with one of the ancillary perchlorate oxygen atoms would serve to stabilize the intermediate by providing a specialized (enzyme-like) binding site. (Note that bidentate bridging coordination of perchlorate to metal and nonmetal centers is known (33).) This idea is consistent also with the observed lack of inhibition by chloride: although chloride is generally much better than perchlorate as a monodentate ligand with metals, chloride would be incapable of forming the secondary hydrogen-bonding interaction with the catalytic reaction center. One implication of this model is that direct incorporation of proton-donating capability into or onto the catalyst support (i.e., a solid acid) could provide the needed rate enhancement without requiring the solution as a whole to be acidic. This possibility is being examined in ongoing work. Comparative Activity of the Catalyst. Table 2 lists representative data from three recent studies (21, 34, 35) involving the use of zerovalent iron (ZVI), in some cases with added microbes, for perchlorate reduction. ZVI is a stoichiometric reagent when used directly; however, it may also provide reducing equivalents for the microbial catalysts, since rates are comparable with or without added hydrogen. However, on a weight-adjusted basis, the best rate coefficients k(cat.) shown by these systems (ca. 6 × 10-4 L h-1 (g of catalyst)-1) are several hundred-fold less than that found for 2048

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 6, 2007

half-life (h) 12 96 432 2

k(obsd) (h-1) 0.06 0.007 0.0016 0.34

C(cat.) (g/L) 95 16 20 2

k(cat.) (L h-1 (g of cat.)-1) 10-4

6× 5 × 10-4 8 × 10-5 1.7 × 10-1

ref 34 35 21 this work

a midrange version of the new ReO/Pd/C catalyst (0.17 L h-1 (g of catalyst)-1). This advantage should translate into much shorter treatment times for a given quantity of contaminated water or a much smaller overall “footprint” for the reactor needed. On the other hand, it is informative to compare this new catalyst with other palladium-containing catalysts that are being examined for specific reductive remediation reactions. For instance, a recent study of heterogeneous catalytic nitrate reduction using a 5% Pd/1.5% Cu/Al2O3 catalyst (36) showed rate coefficients k(cat.) of up to 15 L h-1 (g of cat.)-1 in pure water. Clearly, nitrate reduction is a much faster reaction with the Pd-Cu catalyst than is perchlorate reduction with the ReO-Pd catalyst. Furthermore, studies of a 1% Pd/Al2O3 catalyst for chlorocarbon hydrodehalogenation (37) also showed generally much faster rates, with a k(cat.) of up to 20 L h-1 (g of cat.)-1 for TCE reacting with hydrogen at 1 bar of pressure. Technological Application. This study has shown that a new bimetallic heterogeneous catalyst is effective for perchlorate reduction with hydrogen in controlled aqueous environments. In its current form it appears possible to develop this technology for application in a two-stage system to the purification of the perchlorate-contaminated brine evolved during regeneration of selective ion-exchange columns. In particular, the regeneration procedure with tetrachloroferrate developed at Oak Ridge National Laboratory (14) involves an acidic HCl solution that should be compatible with this ReO/Pd/C catalyst; treatment in a batch mode would lead to complete destruction of the perchlorate, and the resulting brine could likely be recycled. However, direct application of this technology for treating contaminated drinking water will require significant improvements in catalyst activity and stability, especially at near-neutral pH, as well as an understanding of possible interference by natural water constituents. These developments are the focus of ongoing research.

Acknowledgments This work was supported by WaterCAMPWS, a Science and Technology Center of Advanced Materials for the Purification of Water with Systems, under National Science Foundation Agreement Number CTS-0120978. For help in obtaining the XPS data, we thank Rick Haasch of the UIUC Center for Microanalysis of Materials, which is partially supported by the U.S. Department of Energy under Grant DEFG 02-91ER45439.

Literature Cited (1) U.S. EPA. Perchlorate Treatment Technology Update, Federal Facilities Forum; Washington, DC, May 2005; available on the Internet at http://www.clu-in.org/download/remed/542-r-05015.pdf; erratum (September 2005) available at http://www.epa.gov/fedfac/pdf/ffrro_perchlrt_errata.pdf. (2) U.S. FDA. Exploratory Data on Perchlorate in Food; Rockville, MD, 2004; available on the Internet at http://www.cfsan.fda.gov/ ∼dms/clo4data.html. (3) Kirk, A. B.; Martinelango, P. K.; Tian, K.; Dutta, A.; Smith, E. E.; Dasgupta, P. K. Perchlorate and Iodide in Dairy and Breast Milk. Environ. Sci. Technol. 2005, 39, 2011-2017. (4) ITRC (Interstate Technology & Regulatory Council). Perchlorate: Overview of Issues, Status, and Remedial Options; PER-

(5) (6) (7)

(8) (9)

(10)

(11)

(12) (13) (14)

(15)

(16) (17)

(18)

(19) (20) (21) (22)

CHLORATE-1; Perchlorate Team, ITRC: Washington, DC, 2005; available on the Internet at http://www.itrcweb.org. Bo¨hlke, J. K.; Sturchio, N. C.; Gu, B.; Horita, J.; Brown, G. M.; Jackson, W. A.; Batista, J.; Hatzinger, P. B. Perchlorate Isotope Forensics. Anal. Chem. 2005, 77, 7838-7842. Dasgupta, P. K.; Dyke, J. V.; Kirk, A. B.; Jackson, W. A. Perchlorate in the United States. Analysis of Relative Source Contributions to the Food Chain. Environ. Sci. Technol. 2006, 40, 6608-6614. Buffler, P. A.; Kelsh, M. A.; Lau, E. C.; Edinboro, C. H.; Barnard, J. C.; Rutherford, G. W.; Daaboul, J. J; Palmer, L.; Lorey, F. W. Thyroid Function and Perchlorate in Drinking Water: An Evaluation among California Newborns, 1998. Environ. Health Perspect. 2006, 114, 798-804. U.S. EPA. Perchlorate and Perchlorate Salts; Washington, DC; available on the Internet at: http://www.epa.gov/iris/subst/ 1007.htm. Massachusetts Department of Environmental Protection. Perchlorate Fact Sheet for Public Water Suppliers; Boston, 2006; available on the Internet at http://mass.gov/dep/water/drinking/perchfs.htm. California Department of Health Services. Perchlorate in Drinking Water; R-16-04; Sacramento, 2006; available at http:// www.dhs.ca.gov/ps/ddwem/publications/regulations/proposedregulations.htm. Zerovnik, G.; Longville, S. L. Perchlorate in Groundwater: Time to Put New Technology to Work. Report from seminar at California State University, San Bernardino, April 21, 2006; available at http://ottc.csusb.edu/documents/PerchlorateWhitePaper.pdf. Min, B.; Evans, P. J.; Chiu, A. K.; Logan, B. E. Perchlorate Removal in Sand and Plastic Media Bioreactors. Water Res. 2004, 38, 47-60. Shrout, J. D.; Parkin, G. F. Influence of Electron Donor, Oxygen, and Redox Potential on Bacterial Perchlorate Degradation. Water Res. 2006, 40, 1191-1199. Gu, B.; Brown, G. M.; Maya, L.; Lance, M. J.; Moyer, B. A. Regeneration of Perchlorate (ClO4-)-loaded Anion Exchange Resins by a Novel Tetrachloroferrate (FeCl4-) Displacement Technique. Environ. Sci. Technol. 2001, 35, 3363 - 3368. Xiong, Z.; Dimick, P.; Zhao, D.; Kney, A.; Tavakoli, J. Removal of Perchlorate from Contaminated Water Using a Regenerable Polymeric Ligand Exchanger. Sep. Sci. Technol. 2006, 41, 24442574. Lang, G. G.; Horanyi, G. Some Interesting Aspects of the Catalytic and Electrocatalytic Reduction of Perchlorate Ions. J. Electroanal. Chem. 2003, 552, 197-211. Cai, Y.; Espenson, J. H. Catalysis by Methyltrioxorhenium(VII): Reduction of Hydronium Ions by Europium(II) and Reduction of Perchlorate Ions by Europium(II) and Chromium(II). Inorg. Chem. 2005, 44, 489-495. Gu, B.; Dong, W.; Brown, G. M.; Cole, D. R. (2003) Complete Degradation of Perchlorate in Ferric Chloride and Hydrochloric Acid under Controlled Temperature and Pressure. Environ. Sci. Technol. 2003, 37, 2291-2295. Ujvari, M.; Lang, G.; Horanyi, G. Stability of Perchlorate Ions in Acid Medium: Interaction with Iron. J. Appl. Electrochem. 2002, 32, 1403-1406. Moore, A. M.; de Leon, C. H.; Young, T. M. Rate and Extent of Aqueous Perchlorate Removal by Iron Surfaces. Environ. Sci. Technol. 2003, 37, 3189-3198. Cao, J.; Elliott, D.; Zhang, W.-X. Perchlorate Reduction by Nanoscale Iron Particles. J. Nanopart. Res. 2005, 7, 499-506. Lang, G.; Inzelt, G.; Vrabecz, A.; Horanyi, G. Electrochemical Aspects of some Specific Features Connected with the Behavior

(23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

(35)

(36)

(37)

of Iron Group Metals in Aqueous Perchloric Acid/Perchlorate Media. J. Electroanal. Chem. 2005, 582, 249-257. Rusanova, M. Y.; Polaskova, P.; Musika, M.; Fawcett, W. R. Electrochemical Reduction of Perchlorate Ions on PlatinumActivated Nickel. Electrochim. Acta 2006, 51, 3097-3101. Abu-Omar, M. M.; Espenson, J. H. Facile Abstraction of Successive Oxygen Atoms from Perchlorate Ions by Methylrhenium Dioxide. Inorg. Chem. 1995, 34, 6239-6240. Abu-Omar, M. M.; Appelman, E. H.; Espenson, J. H. OxygenTransfer Reactions of Methylrhenium Oxides. Inorg. Chem. 1996, 35, 7751-7757. Abu-Omar, M. M.; McPherson, L. D.; Arias, J.; Bereau, V. M. Clean and Efficient Catalytic Reduction of Perchlorate. Angew. Chem., Int. Ed. 2000, 39, 4310-4313. Arias, J.; Newlands, C. R.; Abu-Omar, M. M. Kinetics and Mechanisms of Catalytic Oxygen Atom Transfer with Oxorhenium(V) Oxazoline Complexes. Inorg. Chem. 2001, 40, 21852192. McPherson, L. D.; Drees, M.; Khan, S. I.; Strassner, T.; AbuOmar, M. M. Multielectron Atom Transfer Reactions of Perchlorate and Other Substrates Catalyzed by Rhenium Oxazoline and Thiazoline Complexes: Reaction Kinetics, Mechanisms, and Density Functional Theory Calculations. Inorg. Chem. 2004, 43, 4036-4060. Crist, B. V. Handbook of Monochromatic XPS Spectra. Vol. 1: The Elements and Native Oxides; XPS International, Inc.: Kawasaki, Japan, 1999. Neri, G.; Rizzo, G.; Milone, C.; Galvagno, S.; Musolino, M. G.; Capannelli, G. Microstructural Characterization of Doped Pd/C Catalysts for the Selective Hydrogenation of 2,4-Dinitrotoluene to Arylhydroxylamines. Appl. Catal., A 2003, 249, 303-311. Li, Y.; Yang, R. T. Hydrogen Storage in Metal-Organic Frameworks by Bridged Hydrogen Spillover. J. Am. Chem. Soc. 2006, 128, 8136-8137. Bailey, N.; Carrington, A.; Lott, K. A. K.; Symons, M. C. R. Structure and Reactivity of the Oxyanions of Transition Metals. Part VIII. Acidities and Spectra of Protonated Oxyanions. J. Chem. Soc. 1960, 290. Pascal, J.-L.; Favier, F. Inorganic Perchlorato Complexes. Coord. Chem. Rev. 1998, 178-180, 865-902; erratum 2000, 209, 453457. Yu, X.; Amrhein, C.; Deshusses, M. A.; Matsumoto, M. R. Perchlorate Reduction by Autotrophic Bacteria in the Presence of Zero-Valent Iron. Environ. Sci. Technol. 2006, 40, 1328-1334. Son, A.; Lee, J.; Chiu, P. C.; Kim, B. J.; Cha, D. K. Microbial Reduction of Perchlorate with Zero-Valent Iron. Water Res. 2006, 40, 2027-2032. Chaplin, B. P.; Roundy, E.; Guy, K. A.; Shapley, J. R.; Werth, C. J. Effects of Natural Water Ions and Humic Acid on Catalytic Nitrate Reduction Kinetics Using an Alumina Supported Pd-Cu Catalyst. Environ. Sci. Technol. 2006, 40, 3075-3081. Lowry, G. V.; Reinhard, M. Hydrodehalogenation of 1- to 3-Carbon Halogenated Organic Compounds in Water Using a Palladium Catalyst and Hydrogen Gas. Environ. Sci. Technol. 1999, 33, 1905-1910.

Received for review October 9, 2006. Revised manuscript received December 19, 2006. Accepted January 16, 2007. ES0624218

VOL. 41, NO. 6, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2049