Environ. Sci. Technol. 2001, 35, 3566-3570
Competitive Photocatalytic Oxidation of Cu(II)-EDTA and Cd(II)-EDTA with Illuminated TiO2 JAE-KYU YANG† AND ALLEN P. DAVIS* Department of Civil and Environmental Engineering, University of Maryland, College Park, Maryland 20742
Competitive photocatalytic oxidation (PCO) of mixtures of Cu(II)-EDTA and Cd(II)-EDTA was studied with variation of molar ratio of these two complexes (1 × 10-4:0, 8 × 10-5: 2 × 10-5, 5 × 10-5:5 × 10-5, 2 × 10-5:8 × 10-5, 0:1 × 10-4 M) and in the pH range of 4-8. PCO rates for each compound can be described using a combined aqueous + adsorbed pathway: -dC/dt ) k1Caq/(1+ k2Caq) + kadsCads. This expression is valid under both noncompetitive and competitive conditions. Differences in rates under competition result from differences in the partitioning of the two species between the TiO2 surface and the aqueous phase. Total initial complex degradation rates (rTT), obtained by summation of the total destruction rates for Cu(II)-EDTA and Cd(II)-EDTA, were relatively constant at pH 4 and 5 for all ratios. At these pH values, contribution of adsorbed pathways to rTT was important, and rates were similar to those of the aqueous phase pathways. From pH 6 to 8, the degree of adsorption, and thus the adsorbed pathway rate, diminished. Through the adsorbed pathway, no difference in rate constants was found between Cu(II)-EDTA and Cd(II)-EDTA; Cd(II)-EDTA is somewhat more reactive through the aqueous phase pathway.
Introduction The fate of pollutant metal ions in the environment is influenced by solution pH and the presence of chelating agents such as EDTA (ethylenediaminetetraacetic acid) (14). In the presence of EDTA, the mobility of metal ions is generally increased at neutral pH due to the weaker adsorption of metal-EDTA complexes onto sorption media in subsurface or soil systems. Such contamination has been reported at several sites in the U.S. (1, 2, 5). Additionally, conventional metal treatment processes such as chemical precipitation are not effective in the treatment of metalEDTA wastes produced from several industrial sources due to the strong chelation by EDTA over wide pH ranges. Photocatalytic oxidation (PCO) with illuminated TiO2 has been ambitiously investigated as a water treatment process (6) and specifically has been evaluated as a promising technique for the treatment of metal-EDTA complexes (712). The EDTA becomes oxidized and at some point the metal is released from the complex. As desired, the metal can be adsorbed onto the TiO2 photocatalyst or left in solution by adjusting suspension pH. Hitherto, most PCO or other advanced oxidation processes have investigated the treatment of simulated wastewater containing single metal-EDTA or * Corresponding author phone: (301)405-1958; fax: (301)405-2585; e-mail:
[email protected]. † Present address: Civil Engineering, Auburn University, AL 36849. 3566
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 17, 2001
free EDTA and have reported reaction kinetics, products, and mechanisms in the destruction of these compounds (719). In most wastewaters, however, mixtures of metal-EDTA are expected depending on the relative concentrations of each metal species, the stability constant of each metalEDTA complex, and species adsorption characteristics. In a competitive PCO environment, multiple species will compete for limited adsorption sites and available oxidant. Considering the first competition, we have studied competitive adsorption between Cu(II)-EDTA and Cd(II)-EDTA onto TiO2 (20). Cd(II)-EDTA adsorption was only slightly affected by the presence of Cu(II)-EDTA; however, Cu(II)-EDTA adsorption was retarded by the presence of Cd(II)-EDTA, especially as the molar ratio of Cu(II)-EDTA/Cd(II)-EDTA decreased, indicating some competition for adsorption sites. Turchi and Ollis (21) studied competitive PCO of benzene and perchloroethylene (PCE) and reported negligible effect of PCE concentration on the initial degradation rate of benzene, while significant inhibition of the initial rate of PCE resulted due to benzene. The authors explained this trend through different distributions of PCE (present primarily in solution phase) and benzene (adsorbed onto TiO2 surface). Adsorbed benzene reacted with oxidants at the TiO2 surface, while PCE degraded primarily by reaction with solution-phase radicals. Therefore, surface-bound benzene would consume the surface radicals before they were able to reach the PCE. Al-Ekabi et al. (22) studied TiO2-mediated degradation of 4-chlorophenol (CP) alone and threecomponent mixtures of CP, 2,4-dichlorophenol (DCP), and 2,4,5-trichlorophenol (TCP). They reported that DCP and TCP compete with CP for the active sites on the TiO2 surface, causing a significant reduction of adsorbed CP. Nonetheless, the overall photocatalytic capacity from the degradation of CP alone (∼55% degradation) was the same as for the threecomponent mixtures (∼53% total degradation of all three compounds). To date, little information is available detailing PCO reaction kinetics in a multicomponent system containing metal chelates. Nonetheless, to appropriately employ PCO technologies for environmental benefits, an understanding of the behavior of these compounds within complex competitive environments is required. In this work, destruction rates of Cu(II)-EDTA and Cd(II)-EDTA complexes in single and binary systems were investigated with a purpose of quantifying the competitive kinetics of aqueous, adsorbed, and total (aqueous+adsorbed) destruction pathways. The competitive adsorption will produce different distributions of species with variation of the ratio of Cu(II)-EDTA/Cd(II)-EDTA and pH in the mixed system. Kinetic expressions will be developed to describe metal-EDTA degradation for comparison to noncompetitive situations. The focus of this work is on the fate of the metal, which is responsible for the toxicity in these mixtures. The metal fate, however, depends on the destruction of the EDTA and the subsequent decomplexation.
Experimental Section Experimental Procedure. All chemicals were analytical grade, and all solutions were prepared with deionized water (18 MΩ-cm) from a Hydro-Service reverse osmosis/ion exchange apparatus (Model LPRO-20). The PCO experiments were conducted using procedures described in previous papers (9-12) using P-25 TiO2 (Degussa Corp). The PCO reactor consisted of a Pyrex glass tube positioned directly adjacent to the UV source. A peristaltic pump was used to circulate 10.1021/es010563q CCC: $20.00
2001 American Chemical Society Published on Web 07/27/2001
the TiO2 suspension between the tube and a modified roundbottomed flask at a flow rate of 75 mL•min-1. The suspension was irradiated with a Spectronics UV lamp containing two integrally fitted UV tubes (15 W each, λmax ) 365 nm, model XX15-A). The photogeneration rate was determined as 1.3 × 10-4 einstein•min-1•L-1 using modified ferrioxalate actinometry measurements (7). Cu(II)-EDTA, Cd(II)-EDTA, and mixtures of Cu(II)-EDTA/Cd(II)-EDTA solutions were prepared by dissolving the metal-perchlorates and equimolar Na2H2EDTA•H2O in water. For investigation of competitive PCO between Cu(II)EDTA and Cd(II)-EDTA, the molar concentration ratios of the two species were 8 × 10-5:2 × 10-5, 5 × 10-5:5 × 10-5, and 2 × 10-5:8 × 10-5 with a total concentration fixed at 10-4 M in 670 mL total initial volume. All experiments were completed in 3 × 10-3 M NaClO4•H2O (Fisher Scientific) electrolyte and at room temperature (22-25 °C). Six or seven samples (∼20 mL each) were collected during a PCO run and immediately filtered through 0.2 µm membrane filters to separate the TiO2. The samples were immediately refrigerated before analyses. Analytical Methods. Cu(II)-EDTA and Cd(II)-EDTA concentrations were measured using a Dionex Model DX100 ion chromatograph (IC) with a conductivity detector. Activities of Cu2+ and Cd2+ were determined in situ using Orion specific Cu2+ and Cd2+ electrodes, respectively. Total aqueous Cu(II) and Cd(II) concentrations were measured using a Perkin-Elmer Model 5100ZL atomic absorption spectrophotometer (AAS). Dissolved organic carbon (DOC) was measured as total carbon using a Shimadzu total organic carbon analyzer (TOC, Model 5000). Aqueous carbon dioxide was analyzed employing a CO2 specific probe (Fisher Scientific) with correction to account for headspace and dissolved carbonate species. Details on analytical methods are described in previous works (9-12). System Species Analysis. To evaluate the competitive PCO reactions, the following metals and organic compound mass balances, as described previously (9-11), were used
Cu(II)T ) Cu(II) aq + Cu(II)ads
(1a)
Cd(II)T ) Cd(II) aq + Cd(II)ads
(1b)
TOC ) DOC + AdOC + CO2
(2)
Cu(II)aq ) Cu(II)-EDTAaq + Cuaq2+ + ΣCu-Yaq
(3a)
Cd(II)aq ) Cd(II)-EDTAaq + Cdaq2+ + ΣCd-Yaq
(3b)
DOC ) Cu(II)-EDTAaq + Cd(II)-EDTAaq + ΣCu-Yaq + ΣCd-Yaq + organic products (4) Cu(II)ads ) Cu(II)-EDTAads + Cuads2+ + ΣCu-Yads (5a) Cd(II)ads ) Cd(II)-EDTAads + Cdads2+ + ΣCd-Yads (5b) AdOC ) Cu(II)-EDTAads + Cd(II)-EDTAads + ΣCu-Yads + ΣCd-Yads + adsorbed organic products (6) where subscripts aq and ads represents aqueous and adsorbed species, respectively.
Results Competitive PCO results were obtained with variation of the Cu(II)-EDTA/Cd(II)-EDTA (CuE/CdE) ratio and pH (ranging from 4 to 6). Selected representative results are shown in Figures 1a-c. Initial adsorbed amounts of Cu(II)-EDTA and Cd(II)-EDTA agree well with previous competitive adsorption
FIGURE 1. Competitive PCO of Cu(II)-EDTA and Cd(II)-EDTA (2 g/L, 10-4 M total metal-EDTA): (a) pH 4, 8:2 ratio of Cu(II)-EDTA/ Cd(II)-EDTA; (b) pH 5, 5:5 ratio of Cu(II)-EDTA/Cd(II)-EDTA; and (c) pH 6, 2:8 ratio of Cu(II)-EDTA/Cd(II)-EDTA. results (20). The initial fraction of aqueous Cd(II)-EDTA was much lower than that of Cu(II)-EDTA under equimolar conditions due to competitive adsorption (Figure 1b). At all pH and ratios of CuE/CdE examined, with the exception of the 8:2 ratio of CuE/CdE at pH 6, both Cu(II)-EDTA and Cd(II)-EDTA were reduced to undetectable levels after 40 min of PCO. Cu(II)aq. At pH 4, the fate of Cu(II)aq differed with variation of the ratio of CuE/CdE. Cu(II)aq increased as the PCO proceeded at the 8:2 ratio (Figure 1a), indicating that the initial oxidation of Cu(II)-EDTA produced nonadsorbable products (12). At the 2:8 ratio, Cu(II)aq increased initially and subsequently decreased due to later adsorption of Cu2+aq and some intermediate complexed copper species, Cu-Yaq, at lower concentration. At constant ratios of CuE/CdE, Cu(II)aq systematically decreased during the PCO at higher pH due to the enhanced adsorption of various copper species. Liberation of Cu2+aq eventually occurred at all ratios, but this species is adsorbed at higher pH. At pH 4, Cu2+aq released increased as the molar ratio of CuE/CdE increased, matching the decreasing shift in the percentage adsorption for higher concentrations. Cd(II)aq. Somewhat different trends of Cd(II)aq were observed as compared to those of Cu(II)aq. Regardless of the ratio of CuE/CdE, Cd(II)aq increased significantly during PCO at pH 4. This is explained by the less favorable adsorption of Cd2+ as compared to Cu2+ (20) and by minimal formation of Cd-Yaq as compared to Cu-Yaq (9, 11). VOL. 35, NO. 17, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3567
The effect of pH on the Cd(II)aq trend was also different from that of Cu(II)aq. At the 2:8 ratio, Cd(II)aq increased continuously at pH 4, and was relatively constant, followed by an increase, at pH 5. At pH 6, Cd(II)aq decreased through 40 min and then increased (Figure 1c). This peculiar latter trend was explained through the formation of more adsorbable intermediates at initial PCO time followed by further oxidation of these adsorbed species, releasing Cd(II) into the aqueous phase (9). The concentrations of Cd(II)aq after 40 min at pH 6 were essentially equal to that of Cd2+aq, indicating negligible presence of any intermediate Cd-Yaq and result from the weaker adsorption of Cd(II) as compared to Cu(II). DOC and EDTA Mineralization. Initial DOC concentrations varied, depending on the initial adsorbed amounts of Cu(II)-EDTA and Cd(II)-EDTA. At pH 4, little decrease in DOC was observed for all ratios due to the formation of soluble metal-organic species from the destruction and release of adsorbed Cu(II)-EDTA and Cd(II)-EDTA, as observed during single species PCO (9-11). At pH 5 and 6, continuous DOC decrease occurred due to substrate mineralization. Comparing these results to the PCO of 10-4 M Cd(II)-EDTA (9) and Cu(II)-EDTA (11) at pH 6 (where adsorption was minimal), it was found that DOC destruction increased as the molar ratio of CuE/CdE decreased, indicating higher mineralization reactivity for Cd(II)-EDTA. In contrast, however, faster CO2 evolution was initially noted at the higher Cu(II) fractions at pH 4 (data not shown), but a nearly equal production was noted for all ratios after 180 min. These contrasting results are observed because of the complex suite of reactions occurring within these systems, to be discussed later. Kinetics of Competitive PCO. To gain a quantitative understanding of the PCO in the binary systems, the simple irreversible reactions previously described in the PCO of 10-4 M Cu(II)-EDTA were employed (11). Each of the metalEDTA species will partition between the aqueous phase and the TiO2 surface.
Cu-EDTAaq h Cu-EDTAads
(7a)
Cd-EDTAaq h Cd-EDTAads
(7b)
Each of these species can be oxidized through corresponding surface or aqueous-phase reactions.
Cu-EDTAaq + oxidantaq F products
(8a)
Cu-EDTAads + oxidantsurface F products
(8b)
Cd-EDTAaq + oxidantaq F products
(8c)
Cd-EDTAads + oxidantsurface F products
(8d)
Details and assumptions affiliated with these processes are given in previous work on competitive adsorption and photocatalysis of Cu-EDTA and Cd-EDTA (9, 11, 12, 20). In a competitive situation, a total initial degradation rate (rTT) can be expressed as the sum of the initial degradation rates (rT) of Cu(II)-EDTA and Cd(II)-EDTA:
rTT ) rT(Cu(II)-EDTA) + rT(Cd(II)-EDTA)
(9)
Assuming that both heterogeneous and homogeneous pathways act independently (11, 12), the total rates for each metal were divided into contributions of aqueous (raq) and adsorbed initial rates (rads, found assuming a constant metal-EDTA partitioning coefficient for each condition and through desorbing metal-EDTA from the surface). Both pathways are important at lower pH, while the homogeneous reaction predominates near neutral pH because of limited adsorption. 3568
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 17, 2001
TABLE 1. Observed Initial Aqueous (raq) and Adsorbed (rads) Degradation Rates of Single and Binary Metal-EDTA with Variation of pH (10-4 M Total Metal-EDTA)
pH 4
raq rads raq rads raq rads raq rads raq rads
5 6 7 8
Cu(II)-EDTA (mol L-1 min-1 × 106) Cu(II)-EDTA:Cd(II)-EDTA
Cd(II)-EDTA (mol L-1 min-1 × 106) Cu(II)-EDTA:Cd(II)-EDTA
10:0
8:2
5:5
2:8
0:10
2:8
5:5
8:2
3.0 2.4 4.0 1.5 3.2 0.3 2.8 0.1 2.0 0
2.7 1.6 3.2 0.8 2.6 0.4
1.8 1.1 2.5 0.6 2.3 0.2 2.1 0 1.7 0
0.7 0.5 0.8 0.2 0.9 0.1
2.5 2.4 3.4 2.3 4.3 0.7 4.3 0.1 3.4 0
1.8 2.0 2.6 1.4 3.2 0.3
0.8 1.9 1.4 1.4 2.0 0.1 1.6 0 1.5 0
0.3 1.1 0.7 0.6 0.8 0.2
2.0 0
1.1 0
2.8 0
0.6 0
In the case of Cu(II)-EDTA alone, raq was described well by a hyperbolic rate expression (11):
-
d[Cu(II)-EDTA]aq k1[Cu(II)-EDTA]aq ) raq ) dt 1+ k2[Cu(II)-EDTA]aq (10a)
The adsorbed-phase reaction was given by a first-order reaction:
-
d[Cu(II)-EDTA]ads ) rads ) kads[Cu(II)-EDTA]ads ) dt kads[S](Cu(II)-EDTA)ads (10b)
Here [ ] and () denote concentrations of substrate as mol•L-1 and mol•g-1, respectively. [S] is the concentration of TiO2 catalyst (adsorbent, here equal to 2 g•L-1). The resulting values for k1, k2, and kads were found to be 0.05 ( 0.01 min-1, 3840 ( 320 L•mol-1, and 0.06 ( 0.05 min-1, respectively for Cu(II)-EDTA over a concentration range from 10-6 to 10-3 M at pH 6 and at 10-4 M at pH 4-8 (11). Analogous expressions are applied to obtain rate constants and reactions orders for Cd(II)-EDTA. Effects of pH on the competitive destruction of Cu(II)-EDTA and Cd(II)-EDTA were evaluated from the pH 4-6 data (partially described in Figure 1a-c) and additional experiments at pH 7 and 8. Total Rates. Figure 2a,b show the total initial rates (rT) of Cu(II)-EDTA and Cd(II)-EDTA in mixtures of CuE/CdE and as single complexes. The maximum rate of Cu(II)-EDTA was found at pH 4 or 5. The rT gradually decreased as the fraction of Cu(II)-EDTA decreased, showing no pH dependence at 2:8. Slower destruction rates were observed above neutral pH where the Cu(II)-EDTA was mostly unadsorbed. In the case of Cd(II)-EDTA, rT also increased as the fraction of Cd(II)-EDTA increased. Compared to Cu(II)EDTA, rT is nearly constant and decreases just slightly with increasing pH, resulting in greater overall reaction for Cd(II)-EDTA than for Cu(II)-EDTA above pH 6. Aqueous Rates. Similar raq dependences on pH, demonstrating a maximum at pH 5, were observed for Cu(II)EDTA at all ratios except 2:8 (Table 1). This trend coincides with the PCO of aqueous Cu(II)-EDTA only. With Cd(II)EDTA (Table 1), raq for all ratios increased up to pH 6 and then slight decreases were observed. Figure 3a shows that aqueous initial rates of Cu(II)-EDTA under both single and competitive conditions for pH ranging from 4 to 6 are similar, within the scatter found in initial rates. The solid line was developed using eq 10a and values for k1 and k2 described above (11). The resultant model generally follows the experimental data under all conditions.
FIGURE 4. Comparison of initial adsorbed rates between single (filled) and competitive (open) metal-EDTA at several pH. Solid line represents a linear model prediction.
FIGURE 2. Observed initial total degradation rates of single and competitive metal-EDTA with variation of pH: a. Cu(II)-EDTA and b. Cd(II)-EDTA. FIGURE 5. Observed total initial degradation rates of Cu(II)-EDTA + Cd(II)-EDTA with variation of pH. Open and filled data represent competitive and single metal-EDTA, respectively.
FIGURE 3. Comparison of initial aqueous rates between single (filled) and competitive (open) metal-EDTA at several pH. Solid line represents a hyperbolic model prediction: a. Cu(II)-EDTA and b. Cd(II)-EDTA. Some points from the Cu(II)-EDTA-only study at higher concentrations used to calibrate the model fit are not shown in Figure 3a. These values contribute to the apparent underprediction of the modeled curve. Figure 3b shows the same information for Cd(II)-EDTA. As with Cu(II)-EDTA, the aqueous initial rates of Cd(II)EDTA were invariant with the competitive presence of Cu(II)-EDTA. The hyperbolic rate expression was also applied to the Cd(II)-EDTA data, producing the solid line. Values of k1′ and k2′ (for Cd(II)-EDTA) are found to be 0.05 ( 0.01 min-1 and 700 ( 800 L•mol-1, respectively. Thus, the aqueous metal-EDTA degradation in competitive systems can be described by eq 10a for both Cu(II)-EDTA and Cd(II)-EDTA,
with no difference between competitive and noncompetitive situations. Compared to Cu(II)-EDTA, k1′ is identical to k1; however, k2′ is less than k2, indicating that k2 is the controlling parameter, producing different initial aqueous degradation rates for the two complexes at the same concentration. The k1 value is considered to depend primarily on the type of catalyst and reaction conditions governing diffusion rates of oxidants from the TiO2 surface into solution (23); thus, identical k1 values were expected here. The difference in k2 and k2′ can be explained by differences in reactivity with OH• radicals by each complex (23), noting different electron densities of the EDTA donor groups for the Cu(II)-EDTA and Cd(II)-EDTA. The PCO reaction is initiated through a carboxyl group in Cu(II)-EDTA as a major reaction pathway for all pH values, while a pathway through an amine group is only important at neutral pH (12). Oxidation is more favorable where carboxyl and amine groups are less tightly bound to the metal ion. The stability constants (log K) for Cu(II)-EDTA and Cd(II)-EDTA are 20.5 and 18.2 (25 °C, I ) 0 M), respectively (24). Adsorbed Rates. Effects of pH and CuE/CdE on the destruction rates of adsorbed Cu(II)-EDTA and Cd(II)-EDTA are also shown in Table 1. For both complexes, increasing pH and decreasing metal-EDTA resulted in decreased rads due to decreased adsorption. The rads of Cd(II)-EDTA were somewhat greater, especially at lower pH. In our previous study (11), a first-order reaction with corresponding rate constant (kads ) 0.06 min-1) was obtained from a linear relationship between log(rads) and log[Cu(II)EDTA]ads. Figure 4 shows that competitive data for Cu(II)EDTA at all pH and concentrations studied also indicate a first-order reaction (1.01 ( 0.06). With the order now fixed at one, kads ) 0.045 ( 0.002 min-1 is found from a direct analysis of rads vis-a´-vis [Cu(II)-EDTA]ads. In the case of Cd(II)-EDTA, a linear relationship was also obtained between log (rads) and log[Cd(II)-EDTA]ads for pH ranging from 4 to 6, including all single and competitive data (Figure 4). A reaction order of one (0.97 ( 0.24) is apparent, and kads is VOL. 35, NO. 17, 2001 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3569
calculated as 0.048 ( 0.007 min-1 from the arithmetic linear relationship. The values of kads for both Cu(II)-EDTA and Cd(II)-EDTA are statistically indistinguishable. Thus, the adsorbed Cu(II)-EDTA and Cd(II)-EDTA species are similar enough that no difference is noted in rate constants for the adsorbed substrate pathway. By considering all aqueous and adsorbed initial rates of Cu(II)-EDTA and Cd(II)-EDTA, the resultant total complex initial rate (rTT ) rTCu(II)-EDTA + rTCd(II)-EDTA) can be expressed as
rTT )
0.05[Cu(II)-EDTA]aq
+ 0.045(Cu(II)1 + 3840[Cu(II)-EDTA]aq 0.05[Cd(II)-EDTA]aq EDTA)ads[S] + + 0.048(Cd(II)1 + 700[Cd(II)-EDTA]aq EDTA)ads[S] (11)
These rate expressions are the same under both single and competitive conditions for each metal-EDTA species and cover the pH range examined, from 4 to 8. Total complex initial rates (rTT) under competitive conditions, however, vary from noncompetitive because of the difference in the distribution of species. The ratio of (CuE)ads/ [CuE]aq is shifted lower as the ratio of CuE/CdE decreases, arising from competitive adsorption. This shift in species distribution affects the PCO rate through dependencies on concentration and pH in competitive situations. The total initial metal-EDTA PCO rates (rTT) as a function of pH are given in Figure 5. At pH 6 and above, the speciation and initial rates of Cu(II)-EDTA and Cd(II)-EDTA are dominated by the aqueous fractions and the cadmium rate is higher than that of the copper, which is suggested by the hyperbolic expressions in eq 11. At pH 4, both aqueous and surface reactions of Cu(II)-EDTA and Cd(II)-EDTA contribute to rTT at all ratios of CuE/CdE. Since the adsorbed pathway rates are identical, little difference is noted among rates at low pH. These trends are supported by the TOC andCO2 results described previously. In other competitive PCO studies, the competition between species was described through different tendencies among organic compounds to occupy active sites on the TiO2 photocatalyst (21, 22). Consequently, a general Langmuir-Hinshelwood rate form was used to describe the kinetics of both single and several binary reactants. Treatment of Mixtures of Metal-EDTA with Illuminated TiO2. Design of a PCO treatment process for mixtures of metal-EDTA should depend on the suspension pH. At a moderately acidic pH, separation of decomplexed metals may be possible, as was noted here at pH 5 since Cu(II) became adsorbed onto the TiO2, while Cd(II) remained dissolved (Figure 1b). Cu(II) and Cd(II) were liberated from EDTA but were not removed through adsorption at pH 4. To remove both metals through TiO2 surface accumulation, PCO should be operated at near-neutral or higher pH. Accumulation and recovery of metal via PCO has recently been
3570
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 35, NO. 17, 2001
demonstrated (25). Cu(II)-EDTA was effectively treated using a cyclic procedure of PCO, catalyst recovery, and Cu(II) recovery through acid stripping. The TiO2 remained photocatalytically active through several reaction cycles.
Acknowledgments This study was supported by the National Science Foundation through Grant BCS-9358209. We thank the Degussa Company for providing the TiO2 sample. We also thank Shalini Jayasundera and Mazyar Zeinali for performing TOC analyses.
Literature Cited (1) Riley, R. G.; Zachara, J. M.; Wobber, F. J. Chemical Contaminants on DOE Lands and Selection of Contaminant Mixtures for Subsurface Science Research; DOE/ER-0547T; April, 1992. (2) Means, J. L.; Crerar, D. A.; Duguid, J. O. Science 1978, 200, 14771481. (3) Jardine, P. M.; Taylor, D. L. Geoderma 1995, 67, 125-140. (4) Cleveland, J.; Rees, T. Science 1981, 212, 1506-1509. (5) Killey, R. W. D.; McHugh, J. O.; Champ, D. R.; Cooper, E. L.; Young, J. L. Environ. Sci. Technol. 1984, 18, 148-157. (6) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. Rev. 1995, 95, 69-96. (7) Madden, T.; Datye, A. K.; Fulton, M.; Prairie, M. R.; Majumdar, S. A.; Stange, B. M. Environ. Sci. Technol. 1997, 31, 3475-3481. (8) Kagaya, S.; Bitoh, Y.; Hasegawa, K. Chem. Lett. 1997, 155-156. (9) Davis, A. P.; Green, D. L. Environ. Sci. Technol. 1999, 33, 609617. (10) Vohra, M. S.; Davis, A. P. Water Res. 2000, 34, 952-964. (11) Yang, J.-K.; Davis, A. P. Environ. Sci. Technol. 2000a, 34, 37893795. (12) Yang, J.-K.; Davis, A. P. Environ. Sci. Technol. 2000b, 34, 37963801. (13) Gilbert, E.; Hoffmann-Glewe, S. Water Res. 1990, 24, 39-44. (14) Krapfenbauer, K.; Getoff, N. Radiat. Phys. Chem. 1999, 55, 385393. (15) Furlong, D. N.; Wells, D.; Sasse, W. H. F. A. Aust. J. Chem. 1986, 39, 757-769. (16) Low, G. K.-C., McEvoy, S. R.; Matthews, R. W. Environ. Sci. Technol. 1991, 25, 460-467. (17) Sorensen, M.; Zurell, S.; Frimmel, F. H. Acta Hydrochim. Hydrobiol. 1998, 26, 109-115. (18) Chung, H. H.; Rho, J. S. J. Ind. Eng. Chem. 1999, 5, 81-86. (19) Chung, H. H.; Rho, J. S. J. Ind. Eng. Chem. 1999, 5, 261-267. (20) Yang, J.-K.; Davis, A. P. J. Colloid Interface Sci. 1999, 216, 77-85. (21) Turchi, C. S.; Ollis, D. F. J. Catal. 1989, 119, 483-496. (22) Al-Ekabi, H.; Serpone N.; Pelizzetti, E.; Minero, C.; Fox, M. A.; Draper, R. B. Langmuir 1989, 5, 250-255. (23) Turchi, C. S.; Ollis, D. F. J. Catal. 1990, 122, 178-192. (24) Allison, J. D.; Brown, D. S.; Novo-Gradac, K. J. MINTEQA2/ PRODEFA2, A Geochemical Assessment Model for Environmental Systems: Version 3.0 User’s Manual; U.S. EPA: Athens, GA, 1991. (25) Rhoads, K.; Davis, A. P. submitted to J. Environ. Eng., ASCE 2001.
Received for review January 22, 2001. Revised manuscript received June 7, 2001. Accepted June 14, 2001. ES010563Q
