Mechanism of Adsorptive Removal of tert ... - ACS Publications

University, Chikusa-ku, Nagoya 464-8603, Japan, and Department of Materials and Life Science, Faculty of Science and Technology, Seikei University...
0 downloads 0 Views 160KB Size
3480

J. Phys. Chem. C 2007, 111, 3480-3485

Mechanism of Adsorptive Removal of tert-Butanethiol under Ambient Conditions with Silver Nitrate Supported on Silica and Silica-Alumina Ken-ichi Shimizu,*,† Shin-ichi Komai,‡ Toshinori Kojima,§ Shigeo Satokawa,§ and Atsushi Satsuma† Department of Applied Chemistry, Graduate School of Engineering, Nagoya UniVersity, Chikusa-ku, Nagoya 464-8603, Japan, Department of Applied Chemistry, School of Engineering, Nagoya UniVersity, Chikusa-ku, Nagoya 464-8603, Japan, and Department of Materials and Life Science, Faculty of Science and Technology, Seikei UniVersity, 3-3-1 Kichijoji-kitamachi, Musashino-shi, Tokyo 180-8633, Japan ReceiVed: December 6, 2006; In Final Form: January 7, 2007

Vapor-phase adsorptive removal of tert-butanethiol (TBT) over silver nitrate supported on silica (AgSi) and silica-alumina (AgSiAl) is tested under ambient conditions. Saturation uptakes on AgSi and AgSiAl are close to that on Ag-exchanged Y zeolite, which was shown to have high sulfur capacity in the literature. The structural analyses by XRD, TEM, Ag K-edge XANES/ EXAFS, and in situ UV-vis show that the reaction of Ag+ species on these samples with TBT yields AgSH species, Ag2S monomer, Ag4S2 cluster, and Ag2S particles. Dynamic changes in adsorbed intermediates and the silver sulfides were followed by in situ FTIR and in situ UV-vis, and the following reaction mechanism is presented: (1) reaction of Ag+ with TBT to produce butenes and AgSH species; (2) reaction of the two AgSH to produce Ag2S monomer and H2S; (3) aggregation of Ag2S monomer to (Ag2S)n clusters and Ag2S particle.

Introduction Pipeline natural gas is one of the most useful fuels for stationary applications of polymer electrolyte fuel cells (PEFC) because of the existence of its supply infrastructure. Generally, a few parts per million of sulfur-containing odorants, such as dimethylsulfide (DMS) and tert-butanethiol (TBT), are generally added to natural gas fuel in order to give people warning of gas leakage. Since these sulfur compounds are a severe poison for steam reforming catalysts,1 pipeline natural gas requires deep desulfurization before introduction to reforming process for PEFC applications. Current hydrodesulfurization (HDS) process, involving catalytic H2 treatment to remove organosulfur compounds as H2S, combined with subsequent adsorption of H2S on zinc oxide, are operated at elevated temperatures and pressures.2 It is difficult to apply HDS process for residential PEFC system, which requires quick and easy start-up, simple operation and small reactor size. Recently, Satokawa et al. reported that Ag-exchanged Y zeolite (Ag-Y) is favorable adsorbent to remove sulfur compounds from pipeline natural gas fuel at ambient conditions even in the presence of water vapor, which can be a new desulfurization process for PEFC system.3,4 Several reports also demonstrated that Ag-exchanged zeolites are effective for removing sulfur compounds such as DMS,5 tetrahydrothiophene6 and thiophenes.7-11 For a practical application, development of new adsorbents based on amorphous oxides with low cost, such as silica, is desired. Although most practical adsorptive separations exploit van der Waals interaction between adsorbates and surfaces, selective adsorption of organosulfur compound can be caused by chemical interaction. Iglesia et al. showed that H-zeolite showed high * Corresponding author. Fax +81-52-789-3193, e-mail: kshimizu@ apchem.nagoya-u.ac.jp † Graduate School of Engineering, Nagoya University. ‡ School of Engineering, Nagoya University. § Seikei University.

selectivity for thiophene adsorption in thiophene/toluene gas mixtures, which is caused by ring-opening and oligomerization of thiophene-derived adsorbed species.12-14 Yang7-9 demonstrated a high thiophene capacity of silver exchanged zeolite due to a π-complexation between these cations and thiophenic aromatic rings. For understanding a chemical interaction between organosulfur compound and adsorption sites, spectroscopic characterization of the adsorption complex is indispensable. Although much attempts have been focused on spectroscopic characterization of silver sulfides cluster formed by the reaction of Ag+-containing solids with H2S,15-18 very few fundamental studies focused on the reaction with organosulfur compounds. In our previous study, we reported that the reaction of Ag+ species in Ag-Y zeolite with TBT results in the formation of AgSH species and Ag2S clusters during TBT adsorption, and we proposed detailed reaction mechanism of TBT-adsorptive removal by Ag-Y.19 In this study, we show that silver nitrate supported on silica (AgSi) and silica-alumina (AgSiAl) are highly effective for TBT adsorptive removal under ambient condition. In-depth characterization of TBT-saturated adsorbents show the structure of TBT-saturated samples. The reaction mechanism was discussed on the basis of dynamic spectroscopic data using in situ FTIR and in situ UV-vis spectroscopies. Experimental Section Preparation of Adsorbents. Amorphous silica (JRC-SIO8, a reference catalyst of the Catalysis Society of Japan, SBET ) 303 m2 g-1) and silica-alumina (JRC-SAL-2, SiO2/Al2O3 ) 5.6, SBET ) 560 m2 g-1) were supplied from the Catalysis Society of Japan. Silver nitrate supported on silica (hereafter named as AgSi) or silica-alumina (hereafter named as AgSiAl) was prepared by impregnating silica or silica-alumina with an aqueous solution of silver nitrate at 298 K for 6 h, followed by evaporating, and subsequently drying in air at 323 K for 4 h.

10.1021/jp068408r CCC: $37.00 © 2007 American Chemical Society Published on Web 02/06/2007

Adsorptive Removal of tert-Butanethiol

J. Phys. Chem. C, Vol. 111, No. 8, 2007 3481 TABLE 1: Summary of TBT Adsorption Experiments samples e

AgSi AgSiAle,f Ag-Ye,f Na-Yf H-Yf

Nta, mmol g-1

Nrb

Nirrc

S/Agd

1.73 1.95 1.80 0.49 1.00

0.25 0.41 0.02 0.13 0.04

1.48 1.54 1.78 0.36 0.96

1.06 1.11 1.28 -

a Total amount of adsorbed TBT. bAmount of reversibly adsorbed TBT. cAmount of irreversibly adsorbed TBT. dThe ratio of the amount of irreversibly adsorbed TBT to silver content. eAg content is 15 wt %. fData from ref 19.

Figure 1. Breakthrough of TBT in a fixed-bed adsorber with AgSi and AgSiAl adsorbents for TBT(500 ppm)/He feed at 298 K.

Silver content of these samples is 15 wt %. Each adsorbent was crashed and sieved to obtain its particle size between 0.2 mm and 0.4 mm. Adsorption Experiment. The adsorption runs were carried out in a fixed bed flow tubular reactor (inner diameter: 3 mm). TBT (500 ppm by volume)/He mixture was fed to adsorbents (0.2 g) at a flow rate of 100 cm3 min-1 at ambient temperature (298 K) The adsorbents were used without any heat treatment. TBT concentrations in the reactor effluent were measured by mass spectrometry (Anelva): ionic current of m/e ) 90 was analyzed. To prevent changes in the operating pressure of reactor, the excess flow from the cell is sent to vent. Ionic currents of m/e ) 34 and 56 in the effluent gas were also monitored, and changes in the ionic currents due to butenes (m/e ) 56) and H2S (m/e ) 34) were estimated by subtracting ionic currents of m/e ) 56 and 34 attributed to the fragmentation of TBT. Characterization of Adsorbent. Powder X-ray diffraction patterns were taken by a Rigaku RINT 1200 X-ray diffraction meter with Cu KR radiation. Transmission electron microscopy (TEM) was carried out on a JEOL JEM-2010 operating at 200 kV. Ag K-edge X-ray absorption spectroscopy (XAS) were carried out at BL-10B of Photon Factory in High-energy Accelerator Research Organization in Tsukuba (Japan), with a ring energy of 2.5 GeV and stored current of 250-350 mA. AgSi and AgSiAl samples after adsorption experiments in Figure 1, named as AgS-S and AgSiAl-S, were sealed in cells made of polypropylene. The spectra were recorded in a transmission mode at room temperature with a Si(311) channel cut monochromator. The energy was defined by assigning the first inflection point of the Cu foil spectrum to 8980.3 eV. The REX version 2.5 program was used for the analysis of X-ray absorption fine structure (XANES), and extended X-ray absorption fine structure (EXAFS). For the curve fitting analysis of Ag-O and Ag-S shells in the Ag K-edge EXAFS, parameters extracted from Ag2SO420 and R-Ag2S21 were used, respectively. In situ FTIR spectra were recorded on a JASCO FT/IR-620 equipped with the IR cell connected to a conventional flow

reaction system. The sample was pressed into a 0.02 g of selfsupporting wafer and mounted into a quartz IR cell with CaF2 windows. Spectra were measured with a resolution of 4 cm-1 at 298 K. A reference spectrum of the catalyst wafer in a flow of He was subtracted from each spectrum. Prior to each experiment, the catalyst was heated in He at 373 K for 30 min, then cooled to 298 K and purged for 30 min with He, and then TBT/He gas mixture was fed at a flow rate of 100 cm3 min-1. In situ diffuse reflectance UV-vis spectra were recorded with UV-vis spectrometer (JASCO V-550) equipped with an in situ flow cell with quartz window used in our previous study.22 A diffuse reflectance sample cell is connected with a gas flow system. The light source is lead to the center of an integrating sphere by optical fiber. Reflectance was converted to pseudoabsorbance using Kubelka-Munk function. BaSO4 was used to collect a background spectrum. TBT(500 ppm)/He gas mixture was fed to the sample (50 mg) at a flow rate of 100 cm3 min-1, and in situ UV-vis spectra were recorded at 298 K. Results and Discussion Adsorption of TBT. A TBT(500 ppm)/He gas mixture was passed through a fixed-bed column and outlet concentration of TBT was monitored as a function of time. Figure 1 shows breakthrough curves for AgSi and AgSiAl samples. Note that adsorbents were tested without any pretreatment and hence physically adsorbed water are present on the adsorbents before adsorption experiments. For both adsorbents, TBT was initially depleted from inlet streams for ca. 100 min, after which its concentration increased and ultimately reached inlet level as the adsorbent reached saturation coverage. Saturation adsorbed capacities can be estimated from a time integral of the differences between inlet and outlet concentrations and are listed in Table 1. Saturation adsorbed capacities of AgSi and AgSiAl were 1.73 and 1.95 mmol/g, respectively. When these samples were purged with He after saturation, weakly adsorbed TBT was desorbed. The amount of irreversibly adsorbed TBT on AgSi and AgSiAl were estimated to be 1.48 and 1.54 mmol/g, respectively. In our previous study, TBT adsorption experiments under the same condition as in Figure 1 were tested for zeolitebased adsorbents.19 The amount of irreversibly adsorbed TBT on AgSi and AgSiAl are close to that of Ag-exchanged zeolite Y with the same silver loading (15 wt %) and are higher than those of Na-Y and H-Y. The S/Ag value, defined as the ratio of the amount of irreversibly adsorbed TBT to silver content, was 1.06 and 1.11 for AgSi and AgSiAl, suggesting that sulfur species are immobilized in a form of silver-sulfur 1:1 complex. Note that no formation of butenes (m/e ) 56) and H2S (m/e ) 34) was confirmed during adsorption experiments. XAS Studies on Silver Sulfides in Ag-Y. To clarify the structure of the Ag species in TBT-saturated samples, AgSi-S and AgSiAl-S, the sample after TBT adsorption experiments in Figure 1 was characterized by XRD, TEM and XAS as shown below. The XRD pattern of the fresh AgSi sample shows

3482 J. Phys. Chem. C, Vol. 111, No. 8, 2007

Shimizu et al.

Figure 2. TEM images of (a,b) AgSi-S and (c) AgSiAl-S.

Figure 4. Fourier transforms of k3-weighted Ag K-edge EXAFS of fresh and TBT-saturated adsorbents.

TABLE 2: Curve-Fitting Analysis of Ag K-Edge EXAFS samples

shell

CNa

R/Åb

σ2/Å2 c

Rf/%d

AgSi-S AgSiAl-S AgSi AgSiAl Ag2S

S S O O S

3.0 3.3 3.2 2.5 (4)e

2.49 2.48 2.68 2.62 (2.52)e

0.060 0.096 0.098 0.109

0.4 1.5 5 8

a Coordination number. bBond distance. cDebye-Waller factor. Residual factor. eCrystallographic values used for the curve-fitting analysis.21

d

Figure 3. Ag K-edge XANES spectra of fresh and TBT-saturated adsorbents and reference compounds.

diffraction lines due to silver nitrate. TEM image of AgSi showed the presence of silver nitrate particles below 10 nm (result not shown). The XRD pattern of the TBT-saturated sample, AgSi-S, shows very weak diffraction lines at d ) 0.2858 and 0.2652 nm due to R-Ag2S (JCPDS file, No. 1472). The TEM pictures of AgSi-S show that Ag2S clusters with size below 8 nm are present on most part of the silica surface as illustrated in Figure 2a. However, a small number of large Ag2S particles with size in a range 10-60 nm were also observed on AgSiAl-S (Figure 2b). The XRD pattern of the fresh AgSiAl sample showed no diffraction lines due to silver nitrate. TEM image of AgSiAl showed the presence of silver nitrate clusters below 2 nm. These results indicate that X-ray undetectable silver nitrate clusters are highly dispersed on silica-alumina. The XRD pattern of AgSiAl did not essentially change after TBT saturation; diffraction lines due to silver nitrate as well as Ag2S were hardly observed for AgSiAl-S. A TEM picture of AgSiAl-S are shown in Figure 2c. Over most part of the silica-alumina surface, Ag2S clusters with size below 8 nm were observed. From these results, it is suggested that Ag2S clusters are mainly formed on the TBT-saturated adsorbents, which is supported by the following characterization results. The structure of the X-ray amorphous species can be investigated with XAS (XANES and EXAFS), because it potentially provides an average structural information of all the Ag species in the sample. As shown in Figure 3, the XANES feature of fresh AgSi and AgSiAl samples was close to that of bulk silver nitrate, in which Ag+ ions are surrounded by oxygen atoms. This indicates the presence of silver nitrate species in the fresh samples. XANES features of AgSi-S and AgSiAl-S are close to each other and are similar to that of crystalline

R-Ag2S, suggesting that silver sulfides are the dominant silver species in TBT-saturated samples. The Fourier transforms (FTs) of Ag K-edge EXAFS of fresh and TBT-saturated samples are shown in Figure 4. The FTs of EXAFS significantly changed after TBT-saturation; an intense peak due to the first neighboring atom appeared. EXAFS feature of AgSi-S and AgSiAl-S are close to that of R-Ag2S. Curve-fitting analyses were performed for the first shell shown in Figure 4. The curve fitting results listed in Table 2 confirm the formation of the Ag-S bond after the desulfurization reaction of TBT over AgSi and AgSiAl. Ag-S coordination numbers of the first Ag-S shell for AgSi-S and AgSiAl-S were smaller than that for crystalline R-Ag2S (CN ) 4), suggesting the smaller particle size of silver sulfide species in AgSi-S and AgSiAl-S. In crystalline R-Ag2S, the shortest Ag-S distances are in the range of 2.40-2.42 Å and the next range of 2.57-2.60 Å.21 The Ag-S distances of AgSi-S and AgSiAl-S samples (R ) 2.48-2.49 Å) are shorter than the average Ag-S distance of crystalline R-Ag2S, suggesting that the structure of silver sulfides in these samples are not the same as bulk Ag2S. In Situ FTIR Studies. Infrared spectra of TBT-derived adsorbed species on AgSi and AgSiAl were measured at room temperature under a flow of TBT(500 ppm)/He. Figure 5A shows changes in the IR spectra of adsorbed species on AgSi. The reaction pathway derived from the following characterizations is shown in Scheme 1. When TBT was introduced to AgSi sample for 2 min, a strong ν(CdC) band assignable to olefins,23-25 and broad ν(SH) band at 2710 cm-1 and a shoulder around 2600 cm-1 were observed. A band at 2290 cm-1 was also observed. Previously, we reported IR spectra of adspecies formed by the reaction of TBT with Ag-Y zeolite and assigned broad bands at 2500 and 2300 cm-1 to ν(SH) modes of H2S molecule on Ag+ site and AgSH species, respectively, and a

Adsorptive Removal of tert-Butanethiol

J. Phys. Chem. C, Vol. 111, No. 8, 2007 3483

Figure 5. Changes in the IR spectra of adsorbed species on (A) AgSi and (B) AgSiAl as a function of time of TBT(500 ppm)/He flowing at 298 K.

SCHEME 1

band at 1640 cm-1 to ν(CdC) of iso-butene adsorbed on the zeolite.19 Garcia and Lercher reported IR study of H2S adsorption on Na-ZSM-5 zeolite and assigned the IR band at 2580 cm-1 to H2S adsorbed coordinatevely on Na+ cation via the sulfur atom.26 In general, higher frequency of ν(SH) mode is expected for weakly adsorbed H2S species with higher S-H bond strength. From these facts, we assigned the bands observed during TBT adsorption on AgSi as follows: a ν(SH) band at 2710 cm-1 with a shoulder at 2600 cm-1 to adsorbed H2S (2 in Scheme 1), a ν(SH) band at 2290 cm-1 to AgSH species (1 in Scheme 1), and a band at 1670 cm-1 to ν(CdC) of adsorbed iso-butene. It should be noted that a broad and very strong band centered around 3150 cm-1 appeared, which should indicate the formation of acidic OH species such as HNO3. In the case of AgSiAl (Figure 5B), a band around 2540 cm-1 assignable to ν(SH) modes of adsorbed H2S was observed, and its intensity increased with time. A band at 2290 cm-1 due to AgSH species was not clearly observed. This may result from a broad nature of several ν(SH) bands or instability of AgSH on this sample. There are at least three different bands at 1660, 1606, and 1540 cm-1 in ν(CdC) region, indicating the presence of several kinds of alkenes. The band at 1660 cm-1 could be assigned to ν(Cd C) of iso-butene adsorbed on silica-alumina. From the above results, it is suggested that TBT undergoes C-S bond rupture to form adsorbed H2S (2) and alkenes possibly via AgSH intermediate (1). In Situ UV-Vis. Diffuse reflectance UV-vis spectroscopy has been conventionally used for characterizing silver sulfide clusters formed by treating Ag-exchanged zeolite with H2S. A group of Calzaferri reported detailed UV-vis and luminescent studies on the formation of silver sulfide clusters, such as Ag2S monomer and Ag4S2, in zeolite A.16-18 On the basis of quantum chemical calculation, the authors showed that the electronic absorption spectrum of Ag2S monomer is composed of four transitions (390, 354, 325, 300 nm with oscillator strength of 0.26, 0.62, 0.73, 0.30, respectively), while the spectrum of AgSH composed of a single transition at 276 nm (with oscillator

Figure 6. (A) Charges in the in situ UV-vis spectra of AgSi as a function of time of TBT(500 ppm)/He flowing at 298 K. Each spectrum was subtracted by the spectrum of fresh sample. (B) Increase in the height of the bands due to Ag2S monomers (360 nm), Ag4S2 clusters (440 nm), and Ag2S particles (600 nm).

strength of 0.73).17 They also showed that a red shift of the electronic absorption bands occurs upon interaction of Ag2S with another Ag2S monomer, forming a Ag4S2 cluster,16 and the spectrum of Ag4S2 cluster exhibits a broad band around 440 nm.18 Figure 6A shows UV-vis spectra of AgSi as a function of exposure time to a TBT/He flow. Note that the spectrum of fresh AgSi sample was subtracted from each spectrum. Time course of the band heights at 360, 440, and 600 nm are plotted in Figure 6B. Adopting the band assignment by Calzaferri et al.,16-18 dynamic changes in the state of silver sulfide species are described as follows. As TBT was introduced on AgSi for 3 min, a band at 290 nm possibly due to AgSH molecule (1 in Scheme 1) appeared. The formation of AgSH molecule (1) on AgSi is consistent with the IR result (Figure 5A). After an induction period for 3 min, intensity of the band at 360 nm assignable to Ag2S monomer (3) increased with time. Intensity of the shoulder band at 300 nm is likely to accompany with that of the band at 360 nm, and we assume that the band at 300 nm is also assigned to Ag2S monomer (3) according to the previous assignment.17 Further increase in the reaction time resulted in a red shift of the absorption edge, and the band height at 440 nm increased with time, indicating the increase in the amount of larger Ag2S clusters, such as Ag4S2 cluster (4). After 40 min, a band at 600 nm assignable to larger Ag2S species, that is Ag2S particles, appeared, and its intensity increased with time. These results indicate the formation of Ag2S monomers (3) via AgSH (1) and its subsequent agglomeration to larger (Ag2S)n clusters (4) and Ag2S particle (5). In situ UV-vis result for the reaction of AgSiAl with TBT is shown in Figure 7. Mechanistic information derived from Figure 7 is basically the same as that from Figure 6 for AgSi. This suggests that the reaction mechanism on AgSiAl is similar to that on AgSi. A relatively high intensity at 600 nm for TBT-treated AgSiAl (after 120 min) may be explained as follows. Although the amount of Ag2S particles mainly contributes to the band height at 600

3484 J. Phys. Chem. C, Vol. 111, No. 8, 2007

Shimizu et al. as Ag2S-like species, the stoichiometry of the overall reaction can be described as follows:

2AgNO3 + 2C4H9SH f Ag2S + H2S + 2C4H8 + 2HNO3 (5) This stoichiometry is consistent with the S/Ag values for AgSi-S and AgSiAl-S (Table 1). The reaction of AgNO3 on AgSi or AgSiAl with TBT yields iso-butene (1660-1670 cm-1) and AgSH species (1, λ) 290 nm, νSH) 2290 cm-1), which then converted to the adsorbed H2S (2, νSH) 2710, 2600 cm-1 for AgSi and νSH) 2540 cm-1 for AgSiAl), Ag2S monomer (3, λ) 360 nm) and HNO3. The formation of HNO3 can be evidenced by the strong IR band around 3150 cm-1 assignable to hydrogen-bonded OH species. As the concentration of Ag2S monomer on the support is increased, the reaction of Ag2S monomers results in agglomeration of them to form Ag4S2 cluster (4) (λ) 440 nm) and Ag2S particles (5, λ) 600 nm). When all the available Ag+ ions are changed to silver sulfides, breakthrough occurs. Conclusions

Figure 7. (A) Charges in the in situ UV-vis spectra of AgSiAl as a function of time of TBT(500 ppm)/He flowing at 298 K. Each spectrum was subtracted by the spectrum of fresh sample. (B) Increase in the height of the bands due to Ag2S monomers (360 nm), Ag4S2 clusters (440 nm), and Ag2S particles (600 nm).

nm, contribution of carbon deposits derived from acid-catalyzed reaction of butenes can be included in the band height. Note that AgSi and AgSiAl, having similar level of the saturation capacities of TBT (Table 1), show basically the same reaction mechanism, which suggests that surface acidity of silicaalumina support does not significantly contribute to the TBT conversion to silver sulfides. The structural model of AgSi-S and AgSiAl-S derived from in situ UV-vis is basically consistent with that from XRD, TEM, XANES, and EXAFS. TEM result evidenced the presence of Ag2S particles (5) with different size distributions: below 60 nm on AiSi-S and below 8 nm on AgSiAl-S. The difference in size may be attributed to the higher acidity of silica-alumina than silica, which could lead to the higher degree of interaction between Ag2S species and the support. Ag2S particles (5) should be a minor silver species in AgSi-S and AgSiAl-S, because intense diffraction lines due to Ag2S are not observed by XRD. EXAFS data also indicates that small Ag2S clusters are the major species in AgSi-S and AgSiAl-S; the Ag-S distance of AgSi-S and AgSiAl-S (R ) 2.48-2.49 Å) is longer than that of AgSH (R ) 2.32 Å) estimated by quantum chemical calculation17 and is close to the Ag-S distance for Ag2S monomer (R) 2.42 Å).17 Reaction Mechanism. Summarizing the above characterization results, the mechanism of silver sulfides formation (Scheme 1) is proposed as follows:

AgNO3 + C4H9SH f AgSH + HNO3 + C4H8

(1)

AgSH + AgSH f Ag2S + H2S

(2)

Ag2S + Ag2S f Ag4S2

(3)

Taking into account the result of XANES (Figure 3) and EXAFS (Table 2), indicating that most of the silver species are present

Silver nitrate supported on silica and silica-alumina are found to be effective adsorbents for vapor-phase adsorptive removal of TBT under ambient conditions. Ag+ species on these adsorbents are sulfided by TBT to AgSH, Ag2S clusters and Ag2S particles. The reaction mechanism is presented as follows: (1) reaction of Ag+ with TBT to yield iso-butene and AgSH species; (2) reaction of the AgSH to yield Ag2S monomer and H2S; (3) aggregation of Ag2S monomer to Ag2S clusters and particles. Acknowledgment. The X-ray absorption experiments were performed under the approval of the Photon Factory Program Advisory Committee (Proposal No. 2003G-274). References and Notes (1) Rostrup-Nielsen, J. R. in Anderson, J. N.; Boudan, M. (Eds.), Catalysis Science and Technology, Springer-Verlag, Berlin, 1984, Vol. 5, p. 95. (2) Topsøe, H.; Clausen, B. S.; Massoth, F. E. in Anderson, J. N.; Boudart, M. (Eds.), Catalysis Science and Technology, Springer-Verlag, Berlin, 1996, Vol. 11. (3) Satokawa, S.; Kobayashi, Y.; Fujiki, H. Stud. Surf. Sci. Catal. 2003, 145, 399. (4) Satokawa, S.; Kobayashi, Y.; Fujiki, H. Appl. Catal. B 2005, 56, 51. (5) Kasaoka, S.; Sasaoka, E.; Funahara, M. Nihon Kagaku Kaishi 1981, 13, 1945. (6) Bezverkhyy, I.; Bouguessa, K.; Geantet, C.; Vrinat, M. Appl. Catal. B 2006, 62, 299. (7) Hernandez-Maldonado, A. J.; Yang, R. T. J. Am. Chem. Soc. 2004, 126, 992. (8) Hernandez-Maldonado, A. J.; Yang, R. T. Catal. ReV. 2004, 46, 111. (9) Yang, R. T.; Hernandez-Maldonado, A. J.; Yang, F. H. Science 2003, 301, 79. (10) McKinley, S. G.; Angelici, R. J. Chem. Commun. 2003, 2620. (11) Hayashi, A.; Saimen, H.; Watanabe, N.; Kimura, H.; Kobayashi, A.; Nakayama, H.; Tsuhako, M. Langmuir. 2005, 21, 7238. (12) Chica, A.; Strohmaier, K.; Iglesia, E. Langmuir. 2004, 20, 10982. (13) Chica, A.; Strohmaier, K.; Iglesia, E. Appl. Catal. B 2005, 60, 223. (14) Yu, S. Y.; Garcia-Martinez, J.; Li, W. Meitzner, G. D.; Iglesia, E. Phys. Chem. Chem. Phys. 2002, 4, 1241. (15) Takahashi, A;, Yang, R. T.; Munson, C. L.; Chinn, D. Ind. Eng. Chem. Res. 2001, 40, 3979. (16) Bru¨hwiler, D.; Seifert, R.; Calzaferri, G. J. Phys. Chem. B 1999, 103, 6397. (17) Bru¨hwiler, D.; Leiggener, C.; Glaus, S.; Calzaferri, G. J. Phys. Chem. B 2002, 106, 3770.

Adsorptive Removal of tert-Butanethiol (18) Leiggener, C.; Bru¨hwiler, D.; Calzaferri, G. J. Mater. Chem. 2003, 13, 1969. (19) Shimizu, K.; Kobayashi, N.; Satsuma, A.; Kojima, T.; Satokawa, S. J. Phys. Chem. B 2006, 110, 22570. (20) Brese, N. E.; O’keeffe, M.; Ramakrishna, B. L.; Von Dreele, R. B. J. Solid State Chem. 1990, 89, 184. (21) Bagatur’yants, A. A.; Safonov, A. A.; Stoll, H.; Werner, H. J. Chem. Phys. 1998, 109, 3096.

J. Phys. Chem. C, Vol. 111, No. 8, 2007 3485 (22) Satsuma, A.; Shibata, J.; Wada, A.; Shinozaki, Y.; Hattori, T. Stud. Surf. Sci. Catal. 2002, 145, 235. (23) Ivanov, P.; Papp, H. Langmuir. 2000, 16, 7769. (24) Kondo, J. N.; Yoda, E.; Ishikawa, H.; Wakabayashi, F.; Domen, K. J. Catal. 2000, 191, 275. (25) Trombetta, M.; Busca, G.; Rossini, S.; Piccoli, V.; Cornaro, U. J. Catal. 1997, 168, 349. (26) Garcia, C. L.; Lercher, J. A. J. Phys. Chem. 1992, 96, 2231.