Mark T. Paffett

The interactions of O2, CO and CO2 with Ag(110)

The interactions and reactions of O2, CO and CO2 with Ag(110) have been studied with AES, XPS, TDS and LEED. Three states of adsorbed oxygen, molecularly adsorbed, atomically adsorbed, and an unreactive (probably subsurface) form, are characterized by O(1s)-XPS peaks at 529.3, 528.1 and 528.5 eV BE, respectively. By dosing at 50 Torr O2 and 485 K and transferring rapidly (∼ 17 s) back into UHV, a coverage θO=0.67 of atomically adsorbed oxygen could be achieved. This gave a new c(6 × 2)-O overlayer structure in LEED and a new, lower-temperature thermal desorption peak of O2 at 565 K. The reaction of CO gas with this state proceeded with a reaction probability of about 0.05 at room temperature to produce CO2gas, until the coverage reached θO=0.5. Below this coverage, the well-known p(2 × 1)-O LEED pattern appeared and the reaction probability for CO dropped suddenly by a factor of about five. Gaseous CO2 reacted in 1-to-1 stoichiometry with oxygen adatoms to produce surface carbonate, CO3,a, which was characterized by a C(1s) peak at 287.7 eV and a broadened O(1s) peak at 529.9 eV BE. The XPS data for O2,a and CO3,a are in agreement with structures for these species which were suggested by earlier vibrational analysis.

Model studies of ethylene epoxidation catalyzed by the Ag(110) surface

Abstract The selective oxidation of ethylene to ethylene epoxide (C 2 H 4 + 1 2 O 2 → C 2 H 4 O) over Ag is the simplest prototype for the entire class of kinetically-controlled, selective catalytic reactions. We have studied the mechanism of this reaction on a well-characterized Ag(110) surface by combining high-pressure kinetic measurements with ultrahigh vacuum surface analysis in a single apparatus. In a typical experiment, the surface cleanliness and order are established by AES and LEED; the sample is transferred into a microreactor where a steady-state reaction rate is established; and the sample is rapidly (17–45 s) transferred back into UHV for surface characterization (AES, LEED, XPS, TDS). In this way we ware able to develop a method for quantitatively measuring the coverage of atomically adsorbed oxygen (θO) under steady-state reaction conditions. The oxygen adatoms are shown by LEED to exist in p(2 × 1) islands of local coverage θO = 0.5. The effects of temperature and reactant pressures upon the rate and selectivity over Ag(110) agree perfectly with results on high-surface-area supported Ag catalysts, although the specific activity (per Ag surface atom) is some 100-fold higher on Ag(110). At constant temperature and ethylene pressure, the rate of ethylene epoxidation varies linearly with θO while the rate of the side reaction leading to CO2 shows a sharp break in slope near θO = 0.02. This and our other data are interpreted by a mechanism involving molecularly adsorbed oxygen and adsorbed ethylene in a common rate-determining step for both ethylene epoxide and CO2 formation. Another pathway for CO2 production via atomically adsorbed oxygen and carbon predominates at very low or very high θO. Oxygen adatoms are necessary in ethylene epoxide formation only for creating special ethylene adsorption sites (Agδ+). We were unable to produce observable amounts of ethylene epoxide by reacting ethylene gas with oxygen adatoms alone.

An electrochemical, ellipsometric, and surface science investigation of the PtRu bulk alloy surface

We describe results of in situ and ex situ characterizations of the surface of a bulk PtRu alloy in sulfuric acid solutions at room temperature. Ellipsometric measurements taken during potential cycling show that a chemisorbed oxygen species forms on the PtRu alloy at potentials as low as +0.25 V vs. RHE. This is found also for Ru electrodes in the same solution. However, the surface oxygen species seems to form more reversibly on the alloy surface. The surface composition of the PtRu alloy appears to be stable up to +1.0 V vs. RHE. Both the in situ and ex situ techniques employed showed partial Ru loss from the surface of the alloy for potential excursions above +1.0 V. Implications of these results for the use of such PtRu alloys as carbon monoxide tolerant anode electrocatalysts in acid fuel cells are discussed.

Surface modification of Pt(111) by Sn adatoms: Evidence for the formation of ordered overlayers and surface alloys

Abstract Vapor deposition of Sn onto Pt(111) in ultrahigh vacuum (UHV) has been examined by Auger electron spectroscopy (AES), X-ray photoelectron spectroscopy (XPS), low energy electron diffraction (LEED) and low energy ion (Ne + ) scattering spectroscopy (LEISS). The AES uptake plots for depositions at substrate temperatures between 140 and 400 K indicate a uniform layer growth up to a Sn exposure of 9.1 × 10 14 atoms cm -2 . LEED data taken simultaneously indicate the successive appearance of the following patterns at submonolayer coverages: (√3 × √3) R 30° followed by c(4 × 2). Simple overlayer real space structures are suggested for these ordered Sn overlayers. For Sn exposures greater than 9.1 × 10 14 atoms cm -2 at these substrate temperatures, nonlinear AES uptake plots and an increasingly diffuse LEED pattern are observed. For Sn depositions above 450 K, Pt-Sn alloy formation is strongly suggested from AES uptake plots and LEISS data. Annealing the Sn-Pt interface to 1000 K. produces ordered (as judged by LEED) alloy surfaces. Evidence for alloy formation is also inferred from variable incidence angle LEISS data. Real space models consistent with the data are proposed.

Cu adsorption on Pt(111) and its effects on chemisorption: A comparison with electrochemistry

Abstract The growth modes and interaction of vapor-deposited Cu on a clean Pt(111) surface have been monitored by Auger electron spectroscopy (AES), low energy electron diffraction (LEED), and work function measurements. The LEED data indicate that below 475 K Cu grows in p(1 × 1) islands in the first monolayer with the interatomic Cu spacing the same as the Pt(111) substrate. The second monolayer of Cu grows in epitaxial, rotationally commensurate Cu(111) planes with the CuCu distance the same as bulk Cu. For substrate temperatures below ∼ 475 K, the variation of work function and “cross-over beam voltage” with Cu coverage show characteristic features at one monolayer that are quite useful for calibration of θ Cu . Above 525 K, Cu-Pt alloy formation was observed in AES and LEED data. Thermal desorption spectroscopy of H 2 and CO has demonstrated that simple site blocking of the Pt(111) surface by vapor-deposited Cu occurs linearly with chemisorption being essentially eliminated at θ Cu = 1.0–1.15. Conclusions drawn from this work correlate very favorably with the well-known effects of under potentially deposited copper on the electrochemistry of the H 2 2H + couple at platinum electrodes.

Borazine adsorption and decomposition at Pt(111) and Ru(001) surfaces

Abstract The adsorption and decomposition of borazine (B3N3H6) was examined over the substrate temperature range 140–1200 K on Pt(111) and Ru(0011) surfaces. The surface reactivity was examined by conventional ultra high vacuum (UHV) methods that included Auger electron spectroscopy (AES), thermal desorption mass spectroscopy (TDMS), X-ray photoelectron spectroscopy (XPS), electron energy loss spectroscopy (EELS), low energy electron diffraction (LEED), and low energy ion surface scattering (LEISS). For room temperature adsorption, borazine adsorbed irreversibly with effective BN coverages of 0.36 monolayers (ML) on Pt(111) and 0.15 ML on Ru(001) (determined from attenuation of substrate AES signals). The drastic difference in room temperature borazine saturation coverage between the two substrates is attributed to a significant degree of dissociation on Pt(111) compared to the Ru surface. For room temperature adsorption, the TDMS data indicated only hydrogen (H2) desorption from both surfaces over a very broad temperature range (e.g., 350–900 K). Borazine exposures of > 5.0 × 10−5 Torr. s at either surface at 1000 K produced ordered close-packed hexagonal BN overlayers with saturation BN surface coverages of 1.22 ML on Pt(111) and 1.16 ML on Ru(001). LEED from the BN M ( M=Pt (111) or Ru (001)) surfaces prepared in this manner gave hexagonal spot splittings of approximately 10th and 12th order, respectively. This is consistent with the formation of a coincidence lattice h-BN overlayer on both substrates. Producing thicker BN overlayerrs by thermal exposures in the high vacuum regime was not possible. The evidence for h-BN overlayers is derived from EELS and LEED data from the annealed BN M surfaces. Multilayer adsorption/desorption of borazine was observed to occur at ∼ 150 K for initial substrate adsorption temperatures of 140 K.

Chemisorption of ethylene on ordered Sn/Pt(111) surface alloys

Abstract The adsorption of ethylene on ordered Sn/Pt surface alloys has been investigated using temperature programmed desorption (TPD), ultraviolet photoelectron spectroscopy (UPS), and high resolution electron energy loss spectroscopy (HREELS). The ordered surface alloys are prepared by annealing (to 1000 K) monolayer amounts of Sn vapor deposited onto Pt(111). This produces Sn/Pt(111) surfaces with p(2 × 2) or (√3 × √3)R30 ° LEED patterns, depending on the conditions used. These surfaces are postulated [M.T. Paffett and R.G. Windham, Surface Sci. 208 (1989) 34] to be the (111) face of Pt3Sn and a substitutional surface alloy of composition Pt2Sn, respectively. Ethylene is reversibly adsorbed at temperatures below 150 K on both ordered Sn/Pt(111) surface alloys: no ethylene decomposition occurs upon heating either surface to above 600 K. As the Sn concentration in the surface alloy is increased, there is a marked decrease in the ethylene desorption temperature from 285 K on Pt(111) to 240 K on the p(2 × 2) alloy and to 184 K on the (√3 × √3)R30 ° alloy. This indicates a significant electronic effect induced by Sn on the Pt-ethylene chemisorption bond strength. However, UPS and HREELS data are consistent with retention of the di-σ-bonding interaction of ethylene on both surface alloys as on Pt(111). The adsorption/desorption behavior and decomposition of ethylene on these surface alloys is consistent with the ensemble size requirement of 4 Pt atoms for ethylene adsorption and 6 Pt atoms for ethylene decomposition on Pt(111) estimated by us previously [14], from studies of C2H4/Bi/ Pt(111). However, electronic effects of alloying Sn with Pt may complicate this simple model and are discussed in describing ethylene chemistry on these surface alloys.

Adsorption and growth modes of Bi on Pt(111)

The vapor deposition of Bi on Pt(111) at 110 and ∼600 K have been characterized by Auger electron spectroscopy (AES), thermal desorption mass spectroscopy (TDMS), low‐energy electron diffraction (LEED), and changes in the work function (Δφ). At 110 K Bi growth follows a layer‐by‐layer mechanism. At ∼600 K Bi fills the first monolayer (θBi≂0.56) relatively uniformly, followed by 3D island growth. Bi desorption is characterized by a large, coverage‐dependent desorption energy, Edes =(81−34.2 θBi ) kcal mol−1, in the first monolayer, and zero‐order kinetics with constant activation energy (Edes =53–56 kcal mol−1) for the multilayer. Many LEED patterns are observed within the first monolayer for both cold and hot substrates. Structural models for these are proposed which are consistent with coverages obtained by AES. Annealed structures show continuous compression of hexagonal Bi overlayers with increasing coverage, subject to mild substrate constraints. At 110 K and θBi >0.33, uniaxial compression is instead seen, due to an unsurmounted energy barrier. Weakly repulsive lateral Bi–Bi interactions (due to dipole repulsions) dominate submonolayer growth. These results for the semimetal Bi are intermediate in behavior between alkali and transition metal overlayers on Pt(111). This is consistent with the relative strengths of the surface dipole of these adsorbed metals.The vapor deposition of Bi on Pt(111) at 110 and ∼600 K have been characterized by Auger electron spectroscopy (AES), thermal desorption mass spectroscopy (TDMS), low‐energy electron diffraction (LEED), and changes in the work function (Δφ). At 110 K Bi growth follows a layer‐by‐layer mechanism. At ∼600 K Bi fills the first monolayer (θBi≂0.56) relatively uniformly, followed by 3D island growth. Bi desorption is characterized by a large, coverage‐dependent desorption energy, Edes =(81−34.2 θBi ) kcal mol−1, in the first monolayer, and zero‐order kinetics with constant activation energy (Edes =53–56 kcal mol−1) for the multilayer. Many LEED patterns are observed within the first monolayer for both cold and hot substrates. Structural models for these are proposed which are consistent with coverages obtained by AES. Annealed structures show continuous compression of hexagonal Bi overlayers with increasing coverage, subject to mild substrate constraints. At 110 K and θBi >0.33, uniaxial compression is instead...

The role of chlorine promoters in catalytic ethylene epoxidation over the Ag(110) surface

Abstract We have recently shown that the kinetics and selectivity of ethylene epoxidation over clean Ag(110) mimic identically the results over unpromoted, supported Ag catalysts, except for a 100-fold enhancement of the specific activity (per surface Ag atom) on Ag(110) [1,2]. In this report, we will discuss results in which we have modeled the role of chlorine promoters in this reaction by combining ultrahigh vacuum surface analysis (XPS, AES, LEED, TDS) before and after high-pressure (≈ 100 torr) kinetic measurements. In this way, we were able to correlate the reaction rate and selectivity not only with temperature and reactant pressures, but also with the coverages of atomically adsorbed oxygen and chlorine. In industrial catalysis, trace amounts of chlorinated organics are added to the feed stream to promote selectivity with, however, some decrease in the reaction rate. We have obtained these same results on Ag(110) by predosing atomically adsorbed chlorine. The major effect occurs between the p(2 × 1) ( θ Cl = 0.5) and the c(4 × 2) ( θ Cl = 0.75) structures, suggesting an ensemble rather than electronic effect. Chlorine coverages above 0.3 completely suppress the rate of dissociative O 2 adsorption and consequently the steady-state coverage of atomic oxygen under reaction conditions. Both ethylene oxide and CO 2 production utilize molecularly absorbed oxygen as the oxidizing agent. Atomically absorbed oxygen and chlorine play a similar role in the reaction by creating Ag δ+ sites for ethylene adsorption.

Electrochemical and surface science investigations of PtCr alloy electrodes

Abstract Electrodes of supported Pt, modified with Cr, have shown an increase in electrochemical activity for oxygen reduction in phosphoric acid fuel cells over supported Pt only electrodes. To clarify the role of chromium and its chemical nature at the electrode surface, we have characterized a series of Pt x Cr( 1- x ) bulk alloys ( x = 0.9, 0.65, 0.5, 0.2) by electrochemical and ex-situ surface science methods. In this paper we report the surface characterization of native and post-electrochemical electrodes by XPS, cyclic voltammetry in 0.05 M H 2 SO 4 and 85% H 3 PO 4 , and analysis of 0.05 M H 2 SO 4 electrolyte following electrochemical treatment. The surface Cr(1 to 2 nm) was oxidized to Cr 3+ oxide for surfaces at open circuit and those exposed to potentials M H 2 SO 4 and 2 PO 4 . In 0.05 M H 2 SO 4 the Cr component was electrooxidized to solube Cr 6+ species at potentials > +1.3 V with the extent of Cr dissolution dependent on initial alloy stoichiometry. Alloys with Cr content ⪖ 0.5 are capable of producing (dependent on time spent at potentials above +1.3 V in 0.05 M H 2 SO 4 ) very porous Pt-rich surfaces. Loss of Cr was also observed in 85% H 3 PO 4 for the alloys with Cr content ≧ 0.5, although at the more positive potential limit of +1.55 V. For the Pt 0.2 Cr 0.8 , treatment in 85% H 3 PO 4 at +1.4 V and above led to the appearance of Pt 4+ and Cr 6+ species, apparently stabilized in a porous phosphate overlayer up to 5 nm thick (dependent on time spent at potentials above this limit). The enhancement reported for supported Pt+Cr oxygen cathodes is discussed in the light of these results.