The Growth Mechanism, Thermal Stability, and Reactivity of Palladium Mono- and Multilayers on Cu(110)

Abstract

The room-temperature growth of palladium (Pd) on Cu(110) has been studied by X-ray photoelectron spectroscopy (XPS), scanning tunneling microscopy (STM), low-energy electron diffraction (LEED), and temperature-programmed desorption (TPD). XPS signal versus deposition time plots rule out a simple layer-by-layer growth mechanism. STM/LEED indicates formation of regions of (2 × 1) overlayer at low Pd coverages (θPd < 1 ML), with considerable disorder in the form of monolayer deep pits and islands. Higher Pd coverages lead to the formation of a granular film consisting of densely packed, flat-topped Pd clusters of average size 75 × 150 Å and with largely a rectangular shape. The favored growth mechanism is of multilayered Pd islands above a mixed (2 × 1) CuPd interface of two to three atomic layers thickness. The thermal stability of the Pd/Cu(110) system was investigated with XPS peak intensity versus annealing temperature plots that indicate that bulk intermixing takes place rapidly between 500 and 600 K. The Pd 3d5/2 XPS peak widths narrow, suggesting the formation of a largely homogeneous CuPd surface alloy. STM indicates that heating to 500 K leaves the Pd clusters in a largely unaltered morphology with no sign of Ostwald ripening, whereas annealing to 600 K leads to considerable changes in topography. The granular structure of the Pd film is disrupted, leading to a surface with irregularly shaped flat domains separated by mono-atomic steps. High temperature (720 K) annealing leads to further flattening and appearance of regular parallel lines in STM images. The spacing of these lines varies with Pd loading, and they are assigned to strain due to lattice mismatch between the “capping” copper monolayer and the underlying mixed CuPd alloy. The reactivity of the Pd/Cu(110) surface has been probed by dosing formic acid and monitoring formate decomposition. High Pd coverages lead to a substantial destabilization of the formate relative to clean Cu(110), which is assigned to formate adsorption on mixed CuPd sites.

Pd Deposition onto the Thermally Roughened Cu(110) Surface

“…The thermally roughened Cu(110) surface was prepared from the clean surface by annealing at~1000 K. For both deposition experiments a Pd coverage of~2 ML was deposited with the specimen held at room temperature and the coverage was determined using XPS to measure the ratio of the Pd 3d 5/2 to the Cu 2p 3/2 core level photoelectron lines. The coverage was calculated using the equation [4] I Pd…”

“…The terms l Pd and l Cu are the inelastic mean free paths of the Pd 3d 5/2 and the Cu 2p 3/2 core levels [4].…”

“…CuPd alloys have catalytic applications and there have been a number of studies of the surfaces of bulk CuPd alloys [1][2][3] and the growth of Pd on Cu surfaces [4][5][6][7]. In the latter case, most studies have concentrated on the deposition of Pd onto the Cu(100) surface.…”

“…In the latter case, most studies have concentrated on the deposition of Pd onto the Cu(100) surface. Recently, studies of Pd deposition onto the Cu(110) surface have been reported [4][5][6] with a complex growth mechanism found for Pd coverages up to 1 ML resulting in regions of the surface consisting of (2 Â 1) reconstructed CuPd surface alloy and regions of disorder in the form of monolayer deep pits and monolayer high islands. The outermost surface structure is somewhat simplified for Pd coverages greater than 1 ML with the formation of a granular film of densely packed Pd islands above the complex (2 Â 1) interface [4].…”

The structure formed by the deposition of a sub-monolayer quantity of platinum onto Cu(110) investigated using medium energy ion scattering

“…This system has been studied primarily using scanning tunnelling microscopy (STM) [12,13,14,15,16] in combination with Auger electron microscopy (AES) [15], LEED [15,16], X-ray photoelectron spectroscopy [16] and reflection anisotropy spectroscopy [13]. There is also a report of an investigation using LEIS [17] which found that for deposition of 1 ML equivalent at room temperature only 15% of the palladium was visible in the surface layer, with copper atoms remaining visible.…”