Environ. Sci. Technol. 2003, 37, 171-176
Characterization of γ-Alumina-Supported Manganese Oxide as an Incineration Catalyst for Trichloroethylene TING-KE TSENG, HSIN CHU,* AND HAN-HSUAN HSU Department of Environmental Engineering, National Cheng Kung University, 1 University Road, Tainan 701, Taiwan
Trichloroethylene (TCE) decomposition over a MnOx/ γ-Al2O3 catalyst in a fixed-bed reactor was conducted in this study. The MnOx/γ-Al2O3 powders were prepared by the incipient wetness impregnation method with aqueous solution of manganese nitrate. The catalysts were characterized by DTA-TGA, XRD, porosity analysis, SEM, EDX, and XPS. The results show that the main distinct weight loss is found at the temperature around 373 and 873 K, the MnO peaks (2θ ) 34.9° and 40.5°) are only observed crystal phase on the fresh catalyst, the SEM image of the MnOx-impregnated γ-Al2O3 support is much different from the calcined catalyst, and the Mn element quantity on the catalyst surface is higher than that of the impregnated support. The products and reactants distributions from the oxidation of TCE over MnOx/γ-Al2O3 were analyzed by GC. The results show that the TCE conversion starts from 5% at 443 K and rises to very high values in the 673873 K ranges and that the CO2 yield also pushes to 99% at the same temperature ranges. HCl and Cl2 are the other main products with little halogenated VOC intermediates.
Introduction Halogenated volatile organic compounds (VOCs) emissions are associated to a wide range of industrial processes; for instance, trichloroethylene (TCE) is mainly used in metal degreasing processes and is known to be hazardous to the environment and public health. The increasing stringent environmental regulations limiting emissions of halogenated hydrocarbons have increased enormously the demand for technology to remove efficiently such halogenated compounds from waste streams. The current technology for the complete oxidation of hydrocarbons is thermal incineration, which requires extremely high temperatures of about 1000 °C (1). While it is a simple and often effective method of control, the high temperatures required culminate in a relatively fuel-intensive technique with little control over the ultimate products. The latter is particularly problematic and can result in incomplete oxidation of the waste stream and the formation of toxic byproducts such as dioxins, dibenzofurans, and oxides of nitrogen if conditions are not carefully controlled. Some people believe that the best technique to eliminate these toxic materials in the waste stream is catalytic oxidation (2). Therefore, heterogeneous catalytic incineration has been * Corresponding author telephone: (886)-6-208 0108; fax: (886)6-275 2790l e-mail:
[email protected]. 10.1021/es0255960 CCC: $25.00 Published on Web 11/26/2002
2003 American Chemical Society
paid the most attention lately because it is a final disposal and energy-saving process (3, 4). However, chlorinated VOCs may deactivate the catalyst and reduce the advantage of catalytic incineration. Catalysts reported for destructive oxidation of chlorinated VOC mostly consist of base metal oxides and noble metals on acidic supports (5-7). Most research has been focused on noble meal catalysts, and very little has been reported on combustion of chlorinated VOC over supported base metal oxides. The supported noble metal systems show high activity for the oxidation of many VOCs, with high selectivity to carbon oxide products. However, these tend to be relatively expensive and can be rapidly deactivated by the presence of chlorinated compounds, sulfur, or other metals in the waste stream. The second class of catalysts is metal oxides, and some of the most active ones are based on copper (8), cobalt (9), chromium (5), and manganese (10). The present study focuses on the characterization of supported manganese oxide catalysts on the activity and product yield in the vapor-phase oxidation of TCE under excess of air between 448 and 773 K, with the aim of enhancing our understanding of the reaction pathway.
Experimental Section Catalysts Preparation. The commercial γ-alumina spherical support, which was supplied by Macherey Nagel, was used in this study. From the surface and pore size analysis, the following properties were obtained: BET surface area, 128 m2 g-1; pore volume, 0.24 cm3 g-1; and average pore radius, 7.50 nm. The active phase (Mn2O3) was obtained by adsorption from aqueous solution of the saltssMn(NO3)2‚4H2O. To do so, 0.61 cm3 of the solution was sprayed on each gram of γ-alumina. The nominal compositions of the prepared catalysts of Mn was 5 wt %. After being dried for 1 h at room temperature followed by drying for 24 h at 393 K in an oven, the supported catalysts were activated by calcining at 873 K in a furnace with an airstream for 8 h and then reduced at 873 K with a N2/H2 ) 10/1 stream for an additional 8 h. Catalyst Characterization (BET, DTA-TGA, XRD, XPS, and TPR). The catalysts’ surface areas and average pore diameters were measured through N2 adsorption at liquidnitrogen temperature by a surface area analyzer (Micromeritics ASAP 2400). The DTA-TGA (differential thermal analysis-thermogravimetric analysis) analysis were carried out in air (100 mL/min) on a thermogravimetric analyzer (model SDT 2960 and Thermo analysis 2000, TA instruments); the temperature cycle was programmed from 313 to 1073 K at a rate of 50 K/min. An X-ray diffractometer (Rigaku D/max III V XRD) was used to analyze the catalysts’ structures. The radiation source was Cu KR. The applied current and voltage were 30 mA and 40 kV, respectively. During the analysis, the sample was scanned from 20 to 80° at a speed of 0.4°/min. The XPS chemical analyzing instrument is a VG Micro Lab MKIII XPS analyzing instrument (where Mg KR was the radiation source). The sample was initially tapped on a sample supporting plate. The plate was then placed in a pretreatment chamber attached to the instrument. The chamber pressure was used then decreased from 100 to 1.33 × 10-10 kPa via a turbo pump. The time needed to reach the final pressure was approximately 4 h. A gate valve between the pretreatment chamber and a vacuum chamber was then opened after the pretreatment. The sample was moved to the vacuum chamber for analysis. The pressure in the vacuum chamber was VOL. 37, NO. 1, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
171
maintained at 1.33 × 10-10 kPa via an ion pump. The following analyzing conditions were utilized: resolution ) 0.1 eV; number of scans ) 100. The binding energy spectra were obtained under the above-mentioned conditions and a predetermined scanning range. The apparatus used for the temperature-programmed reduction (TPR) was described by Jones and McNicol (11). A gas stream of 10% H2 in argon passed through the catalyst sample (0.5 g) in a quartz reactor heated at 10 °C/min to 760 °C with a temperature-programmed furnace. The water produced by reduction was trapped into a column of silica gel. The amount of H2 consumption was detected with a thermal conductivity detector (TCD). The reduction temperature was monitored by a K-type thermocouple. Catalytic Activity Measurement. The catalytic incineration of this study was conducted in a bench-scale fixed-bed reactor under atmospheric pressure. It consisted of a 1.6 cm i.d., 2.0 cm o.d., and 45 cm length stainless steel tube located inside an electrical furnace. A 200-mesh SS316 sieve was set in the reactor, 24 cm below the top of the tube, to support the catalyst. The weight of catalyst packing was 1.7 g (thickness 0.75 cm). Two K-type thermocouples were inserted into the reactor to the positions on the top and bottom of the catalyst packing, respectively, to control (up position) and measure (down position) the inlet and outlet temperature. To start a run, after the catalyst bed temperature had decreased to the wanted reaction temperature (445-863 K), a gas mixture containing 100 ppm TCE balanced by dry air was passed through the catalyst bed. The gas hourly space velocity (GHSV) was calculation at 80 000 h-1. The conversion of TCE and C2Cl4 concentration were determined by analyzing inlet and outlet gases by a GC unit (Shimadzu, GC-14B) with an FID detector, and the concentration of CO and CO2 in the effluent gas was measured with a CO/CO2/O2 analyzer (model 300 type). Carbonaceous material changes of the catalyst samples were determined by an elemental analyst (EA, Elementar Analysensysteme GmbH); metal composition and surface changes of the catalysts were analyzed by a SEM/EDS analyzer (Hitachi S-2500). Analysis of both Cl2 and HCl was performed by bubbling the effluent stream through a NaOH solution. Then, the Cl2 concentration was determined by titration with ferrous ammonium sulfate using N,N-diethyl-p-phenylenediamine as the indicator. The concentration of chlorine ions in the bubbled solution was determined by using a Dionex DX-100 ion chromatograph to obtain the HCl concentration. The conversion (X) of TCE and the yield (YC and YCl) of carbon and chlorine products are defined as follows:
X)
YC(CO,CO2,C2Cl4) )
Cin - Cout × 100% Cin
CCO × 100% 2Cin
CCO2 2Cin
× 100% 2CC2Cl4 2Cin
YCl(Cl2,HCl,C2Cl4) )
2CCl2 3Cin
× 100%
(1)
CHCl × 100% 3Cin 4CC2Cl4 3Cin
× 100% (2)
× 100% (3)
Results and Discussion Porosity of Catalysts at Various Status. Table 1 shows the physical properties of the catalysts at various status, such as γ-Al2O3 (CA-Al), impregnated catalyst (CA-I), calcined catalyst 172
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 1, 2003
TABLE 1. Basic Properties of MnOx/γ-Al2O3 Catalyst catalyst
BET surface area (m2/g)
pore vol (cm3/g)
avg pore diameter (nm)
CA-Al CA-I CA-C CA-R CA-U
128 120 105 106 101
0.24 0.21 0.22 0.22 0.22
7.5 7.0 8.3 8.4 8.6
FIGURE 1. XRD spectra of (A) the catalysts by various calcining temperatures and (B) the catalysts at various states: (a) 573 K, (b) 673 K, (c) 773 K, (d) 873 K, (e) 973 K, (f) CA-Al, (g) CA-I, (h) CA-C, (i) CA-R, (1) γ-Al2O3, (2) Mn2O3, (3) MnO, and (4) Mn3N2. (CA-C), reduced catalyst (CA-R), and 638 K incinerated catalyst (CA-U). As shown in Table 1, the surface area decreases and the average pore diameter increases slightly in the sequence of the impregnation, calcinations, reduction, and incineration treatment. These may be due to the fact that the micropore was sintered during the high-temperature treatment (CA-C and CA-R). The observation of CA-U catalyst reveals that some chlorine species cover the manganese sites and that its surface area is less than that of the CA-R catalyst. Since the small pore was blocked, the average pore diameter of CA-U is larger than that of the CA-R catalyst. Surface Structure (XRD, TGA/DTA, XPS, TPR, SEM/EDS). Figure 1A shows the X-ray diffraction spectra of the different calcination temperature for the calcined catalysts. The principal peaks at 2θ ) 32.88° and 55.04° in the spectrum of
FIGURE 2. DTA-TGA curve of the impregnated catalyst. 873 K calcination temperature do not appear in the spectrum of calcination temperature 573-773 K, and the peak intensity of the 973 K calcination temperature is lower than for the 873 K calcination temperature. According to the Joint Committee on Powder Diffraction System (JCPDS) file, the principal peaks represent Mn2O3 [78-0390]. The result suggests that the best calcination temperature of the impregnated catalyst for obtaining supported Mn2O3 catalyst is 873 K. The X-ray diffraction spectra of the catalysts at various state are shown in Figure 1B. The principal peaks at 2θ ) 34.96° and 40.59° in the spectrum of reduced catalyst do not appear for calcined catalyst. According to the Joint Committee on Powder Diffraction System file, the principal peaks represent MnO [78-0424]. The result suggests that the calcined catalyst reduced by H2 in the reducing processes let the manganese oxide on γ-Al2O3 support become amorphous. In addition, there are small peaks representing the manganese oxide crystal phase appearing in the spectra of the impregnated catalyst. According to the Joint Committee on Powder Diffraction System file, the small peaks represent Mn3N2 [81-0299]. From this observation, we infer that the major part of Mn3N2 of the impregnated catalyst is converted to Mn2O3 after the 873 K calcination process and leads to the MnO by the reducing process. The result of DTA-TGA analysis at a heating rate of 10 °C/min for the impregnated catalyst is shown in Figure 2. From this figure, the TGA curve indicates that the weight of impregnated catalyst declines from 98.5% to 90% and that two principal peaks at 373 and 873 K appear in the DTA curve. This results suggests that the weight loss phenomenon might be due to the loss of moisture adsorbed on impregnated catalyst and the loss of the manganese oxide crystal changes at 373 and 873K, respectively. According to Figure 1A, the result proved that the crystal forming temperature of Mn2O3 is 873 K. The XPS spectra for the impregnated, calcined, reduced, and incinerated catalysts were recorded. The binding energy (BE) values for the Mn2P3/2 peak in impregnated, calcined, and reduced catalysts are drawn in Figure 3. For the impregnated sample, value at 641.2 eV is consistent with the expected BE for Mn2+ species. This suggests that the impregnated precursor during the drying for 24 h at 393 K in an oven; the manganese is oxidized to Mn2+ or Mn>2+ species. For the calcined catalyst, exposed to an air pulse, the Mn2P3/2 profile changes to a value at 641.7 eV that is consistent with the expected BE for Mn3+ species. This finding suggests that the manganese is oxidized to Mn2O3 during the 873 K calcination process. These results were also observed earlier on the XRD spectra. For the reduced catalyst, the Mn2P3/2 profile may be resolved into three peaks. According
FIGURE 3. Mn2P3/2 XPS spectra of the catalysts at various states.
FIGURE 4. TPR profiles for the fresh and incinerated catalyst. to other studies, the manganese species corresponding to the BE at 642.6 eV would be Mn4+ species (12, 13). For the incinerated catalyst, in Figure 3, the Mn2P3/2 profile changes for a BE value at 641.7 eV. This finding suggests that the manganese is reduced to Mn2O3 during the incineration process. Figure 4 displays the TPR patterns of the fresh and incinerated catalysts. The reduction peaks of the fresh catalyst at 593, 799, 788, and 873 K may represent the different oxidation states. One reduction peak (833 K) appears in the pattern of the incinerated catalyst. The H2 standard (0.5 mL) was injected into the TPR apparatus, and the peak area was measured to calibrate the hydrogen consumption of the reduction peaks for the fresh and incinerated catalysts. The amount of hydrogen consumption for the fresh and incinerated catalyst were 342 and 244 mL, respectively. This reveals that the average oxidation state of the fresh and incinerated catalyst was 4.2 and 3, respectively. It shows that the oxidation VOL. 37, NO. 1, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
173
TABLE 2. EDX Analysis for the Catalysts at Various Statuses element mass percentage (%) catalysts
Al
Mn
Fe
Cl
CA-Al CA-I CA-C CA-R CA-U
86.5 5.5 8.9 18.4 11.8
0.0 85.8 83.8 73.0 80.5
13.5 8.6 7.3 8.6 7.4
0.0 0.0 0.0 0.0 0.3
FIGURE 6. Product distributions of the TCE decomposed on the 5 wt % MnOx/γ-Al2O3 catalyst.
FIGURE 5. Light-off curve of TCE decomposed on the 5 wt % MnOx/ γ-Al2O3 catalyst. state of the fresh catalyst changes to a stable state after decomposition of TCE. According to the XPS results, the stable crystal form may be Mn2O3. According to the results of XRD and XPS spectra, such a phase change (Mn3N2fMn2O3fMnOxfMn2O3) has been found in other previous studies (12, 13). Figure 4 displays the TPR patterns of the fresh and incinerated catalysts. The reduction peaks of the fresh catalyst at 593, 799, 788, and 873 K may represent the different oxidation states. One reduction peak (833 K) appears in the pattern of the incinerated catalyst. The stander H2 pulse (0.5 mL) was injected into the TPR apparatus, and the peak area was measured to calculate the hydrogen amount of the reduction peaks for the fresh and incinerated catalysts. The amount of hydrogen for the fresh and incinerated catalyst was 342 and 244 mL, respectively. This reveals that the average oxidation state of the fresh and incinerated catalyst was 4.2 and 3, respectively. This phenomenon performed that the oxidation state of the fresh catalyst changed to be a stable state after decomposition of TCE. According to the XPS results, the stable crystal phase may be Mn2O3. The SEM spectra for the impregnated and calcined catalysts were recorded. For the impregnated catalyst, some spherical particles are formed. After calcinations at 873 K, small needle-shape particles are formed. Although it is impossible to carry out a quantitative EDS analysis of the catalyst, the semiquantitative information can be obtained. Table 2 shows the results corresponding to the catalysts at various state. For example, the Al and Mn contents of the calcined catalyst are respectively higher and lower than for the impregnated catalyst. The light-off curve of TCE decomposition was performed and is shown in Figure 5. As shown in Figure 5, the performance of the catalyst declines gradually for 3 h and then remains at a stable condition for 7 h. This phenomenon may be due to the possibility that certain activated sites of 174
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 1, 2003
the catalyst would form irreversible sites with chlorine species that would need time to accomplish the irreversible reaction. The remaining sites of the catalyst could be reversible sited. This phenomenon is consistent with the TPR test because some activated sites may be deactivated when the H2 consumption of the catalyst drops after incineration. Although the process is kinetic for the first couple hours, we take the data when the conversion reaches a stable value. Therefore, the process is in a sense of stoichiometry. Product Distributions of TCE Decomposition. A series of blank tests by replacing catalyst packing with glass fiber or the alumina support were performed. The results are shown in Figure 6. It suggests that the homogeneous reaction occurs above 623 K. This is similar to the finding of Ballikaya et al. (14), in which they showed that the homogeneous reaction of methylene chloride started above 673 K. The activity of the alumina support is between the homogeneous and the catalytic reactions. This relationship can be attributed to the surface acidity of the alumina, which plays an important role in the adsorption of chlorocarbon occurring at acid sites via hydrogen bonding between chorine and surface proton (15-18). Figure 6 also shows the distributions the products and reactants for the oxidation of TCE over MnOx/γ-Al2O3 with correct index. The conversion of TCE was extremely low below 573 K. As the temperature exceeds 598 K, the conversion rapidly increases and CO2, HCl, and Cl2 are the main products, with only some trace amounts of incomplete combustion products (e.g., C2Cl4 and CO). According to this figure, C2Cl4 is formed from the beginning, presenting a peak at 668 K, 34.9 ppm. The concentration of CO is also peaked at 698 K, 34.2 ppm. From the results shown in Figure 6, to correlate concentrations of different components in the figure, the following reaction scheme can be proposed for the destruction of TCE:
C2HCl3 + 2O2 f 2CO2 + HCl + Cl2
(4)
C2HCl3 + O2 f 2CO + HCl + Cl2
(5)
C2HCl3 + Cl2 f C2Cl4 + HCl
(6)
C2Cl4 + O2 f 2CO + 2Cl2
(7)
1 CO + O2 f CO2 2
(8)
as well as the Deacon reaction (19) since water and chlorine
FIGURE 7. Relationship between the yield of CO, CO2, and C2Cl4 and the conversion of TCE.
FIGURE 8. Relationship between the yield of Cl2, HCl, and C2Cl4 and the conversion of TCE.
are present in the reaction environment. TCE decomposition primarily undergoes reactions 4-6 at low temperature. As the temperature increases to 725 K, high conversion of TCE (above 95%) is obtained with some amounts of C2Cl4 (which is destroyed by reaction 7 in the range 698-750 K) and a yield to Cl2 around 60% and to HCl around 18 % at 750 K is observed. The side reaction associated with the chlorination of the feed (TCE) to tetrachloroethylene was also observed by Wang et al. (20) over 1.5% platinum on γ-alumina with 95.6% conversion at 723 K but low yield to HCl in the range of 20-23% at temperatures between 623 and 723 K. These authors (21) also studied the TCE oxidation over PdO on γ-Al2O3 that reported 95.6% conversion at 823 K with yield to HCl of 31% and selectivity to C2Cl4 of 1.43%. Thus, the catalysts studied in this work produces higher yield to the desirable Cl2 and HCl than those reported elsewhere (20, 21). Formation of CO from incomplete oxidation of TCE, reaction 5, can be observed over MnOx/γ-Al2O3 catalysts at lower temperatures around 551 K but not at high temperature due to reaction 8. The peak around 698 K obtained with the catalyst corresponds to the oxidation of C2Cl4 (reaction 7) since the formation of CO coincides with the decrease of C2Cl4 concentration. Figures 7 and 8 show the TCE conversion and product yield against the change of reaction over the MnOx/γ-Al2O3 catalyst. To thoroughly understand the reaction pathway of the decomposition of TCE, the mass balance on C and Cl atoms was performed. The TCE simulation conversion (Xi-simul) and the yields (Yi) of products are defined as
the chlorine atoms (Cl) measured at outlet and inlet of the reactor are not balanced at high temperatures. It is noticeable that in all cases chlorine recovery was within 76-100%. A change in color of the catalyst from black to light green was observed at the end of the activity tests. These colors fit quite well with those reported in the literature for metal chloride complexes (22). Therefore, the interaction of chlorine with the metal in the presence of halocarbons or adsorption of chlorine on the γ-Al2O3 support could explain the unfitted chlorine balance. All of our work was performed under dry inlet conditions in order that water vapor effects were not examined. The effects of water vapor are beyond the scope of this study, deserving further attention in the future, although it is wellknown that, on one hand, water can sinter the precious metal inhibiting the catalyst activity and, on the other hand, it is a source of hydrogen in the TCE oxidation promoting the selectivity to HCl (20, 21).
Xi(C,Cl)-simul )
∑Y
i(C,Cl)
(9)
in which Yi is the experiment data (Figure 6) of products yield obtained from eqs 2 and 3; Xi-simul can be calculated by summation of the yields (Yi). As shown in Figure 7, the TCE simulation conversions are attached with the raw TCE conversion data. It reveals that carbon atoms (C) measured at outlet and inlet of the reactor remains balanced at all temperatures. In fact, the carbon atoms recovery in all case was within 93-103%. This finding is consistent with eqs 4-8, which show that the reactant is TCE and that the products containing carbon atoms are CO, CO2, and C2Cl4. The dominant product is CO2. Figure 8 shows that the TCE simulation conversions do not fit well with the raw TCE conversion data. It reveals that
Literature Cited (1) Artizzu-Duart, P.; Brulle´, Y.; Gaillard, F.; Garbowski, E.; Guilhaume, N.; Primet, M. Catal. Today 1999, 54, 181. (2) Toledo, J. M.; Corella, J.; Sanz, A. Environ. Prog. 2001, 20, 167. (3) Van der Vaart, D. R.; Vatavuk, W. M.; Wehe, A. H. J. Air Waste Manage. Assoc. 1991, 41 (1), 92. (4) Van der Vaart, D. R.; Vatavuk, W. M.; Wehe, A. H. J. Air Waste Manage. Assoc. 1991, 41 (4), 497. (5) Agarwal, S. K.; Spivey, J. J.; Butt, J. B. Appl. Catal. A 1992, 82, 259. (6) Bickle, G. M.; Suzuki, T.; Mitarai, Y. Appl. Catal. B 1994, 4, 141. (7) Rossin, J. A.; Farris, M. M. Ind. Eng. Chem. Res. 1993, 32, 1024. (8) Kang, Y. M.; Wan, B. Z. Appl. Catal. A 1994, 114, 35. (9) Drago, R. S.; Jurczyk, K.; Singh, D. L.; Young, V. Appl. Catal. B 1995, 6, 155. (10) Watanabe, N.; Yamashita, H.; Miyadera, H.; Tominaga, S. Appl. Catal. B 1996, 8, 405. (11) Jones, A.; McNicol, B. Temperature-Programmed Reduction for Solid Materials Characterization; Marcel Dekker: New York, 1986. (12) Wangner, C. D.; Riggs, W. M.; Moulder, J. F. Handbook of X-ray Photoelectron Spectroscopy; Perkin-Elmer Corp.: Minneapolis, 1979. (13) Zaki, M. I.; Hasan, M. A.; Pasupulety, L.; Kumari, K. New J. Chem. 1998, 22, 875. (14) Ballikaya, M.; Atalay, S.; Alpay, H. E.; Atalay, F. S. Combust. Sci. Technol. 1996, 120, 169. (15) Chintawar, P. S.; Greene, H. L. Appl. Catal. B 1997, 13, 37. (16) Windawi, H.; Zhang, Z. C. Catal. Today 1996, 30, 99. (17) Jiang, X. Z.; Zhang, L. Q.; Wu, X. H.; Zheng, L. Appl. Catal. B 1996, 9, 229. (18) Greene, H. L.; Prakash, D. S.; Athota, K. V. Appl. Catal. B 1996, 7, 213. VOL. 37, NO. 1, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
175
(19) Weldon, J.; Senkan, S. M. Combust. Sci. Technol. 1986, 47, 229. (20) Wang, Y.; Shaw, H.; Farrauto, R. J. Catalytic Control of Air Pollution; Mobile and Stationary Sources, ACS Symposium Series 495; American Chemical Society: Washington, DC, 1992. (21) Yu, T. C.; Shaw, H.; Farrauto, R. J. Catalytic Control of Air Pollution, Mobile and Stationary Sources, ACS Symposium Series 495; American Chemical Society: Washington, DC, 1992.
176
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 1, 2003
(22) Gonza´lez-Velasco, J. R.; Aranzabal, A.; Gutie´rrez, J. I.; Lo´pezFonseca, R.; Gutie´rrez-Ortiz, M. A. Appl. Catal. B 1998, 19, 189.
Received for review February 21, 2002. Revised manuscript received September 13, 2002. Accepted October 28, 2002. ES0255960
