Pd Coverage Research Articles

Characterization of PdCl2/SnCl2 metallization catalysts on a polyetherimide surface by XPS and RBS

AbstractA metallization catalyst (based on the PdCl2/SnCl2 system) adsorbed on polyetherimide surfaces was characterized by x‐ray photo‐electron spectroscopy (XPS), Rutherford back scattering (RBS) and transmission electron microscopy (TEM). The overall Pd/Sn ratio increased from 0.2 to 3.8 when the initial sensitized surface was treated with the accelerator solution. XPS results indicated that before acceleration, only Sn(IV) (stannic hydroxide) is obserable, and after acceleration both metallic Sn and Sn(IV) are present; palladium is present primarily in a metallic form both before and after the accelerator step. Total Pd and Sn coverages were also obtained by RBS, and the total Pd/Sn ratios agree with the XPS surface measurements for specimens just prior to metallization. TEM micrographs showed the adsorbed catalyst consisted of particles 10–50 Å in size covering only a fraction of the surface.

Morphology and structural phase transitions of Pd monolayers on Ta(110).

The morphology of Pd overlayers on Ta(110) was studied for various Pd coverages before and after heat treatment. As was reported for Pd on Nb(110), a structural phase transition from a commensurate epitaxial overlayer to an incommensurate overlayer was found at approximately one monolayer (ML) coverage (1 ML=1.89 A\r{}). Valence-band photoemission and Auger-electron energy distribution curves show modifications of the Pd-derived valence-band states attributable to Pd\char22{}Ta bonding and reduced Pd nearest-neighbor coordination. Little change is seen in the Pd density of states during the commensurate-incommensurate transition. Thicker Pd layers (thetag1 ML) evolve toward the structure and photoemission properties characteristics of Pd(111). Thermal treatment of the thick Pd overlayer triggered the interdiffusion of Ta and Pd and produced an extended Pd-Ta surface compound. Further heat treatments caused additional Pd and Ta interdiffusion and left a surface-retained Pd ``monolayer'' with the same structure as the Ta(110) substrate.

Photoemission study of the reactive Pd/InP(110) interface

We present a soft x-ray photoemission spectroscopy (SXPS) study of the Pd/InP(110) interface. SXPS has been used to study chemical reactions, overlayer morphology, and Fermi level pinning behavior during the early stages of interface formation. Strong reactions involving Pd and both species of the substrate have been documented and discussed. We took advantage of the tunability of the synchrotron source to study p–d valence band hybridization near to, and far from the Pd 4d cross section Cooper minimum. This was combined with a detailed study of the line shape of the P LVV Auger transition. Because of the strong reactions we had to unfold the core level spectra into their different components to follow the Fermi level pinning. Such an analysis shows that Fermi level pinning is established at the same energy, 0.45 eV below the conduction band minimum, for n and p-type crystals with Pd coverages of about 1 ML. This energy agrees with barrier heights established from the current–voltage characteristics measured on thick diodes grown under identical conditions. On the basis of the data, correlations proposed to describe barrier heights and chemical reactions at metal/InP interfaces are evaluated.

Thickness dependence of the reactions at the interface of Pd and Si〈111〉

The reactions at the interface of Si〈111〉 and Pd are studied as a function of the thickness of Pd coverage from 5 to 50 Å. Surface conditions are monitored with Auger spectroscopy and the structural interactions are probed by Raman spectroscopy. The results indicate that at 5 Å coverage, strong Si–Pd bonding is evident from the Auger, but the Raman spectra show no evidence of crystalline silicide formation. At 10 Å coverage a sharp peak is observed in the Raman spectrum which is attributed to the formation of crystalline silicide. The results indicate that the initial interactions can be modified after reactions proceed to greater thicknesses. The effect is attributed to critical cluster sizes necessary to form stable nuclei.

Determination of the frequency-dependent resistivity of ultrathin metallic films on Si(111).

Inelastic electron scattering is used to determine the frequency-dependent resistivity of ultrathin metallic films on Si(111). The experimental data are analyzed in the single-scattering regime using dipole scattering theory. An unusual frequency dependence of the resistivity is found for low coverages of Pd on Si(111) and analyzed using the Bruggeman effective-medium theory. This analysis together with hydrogen-titration studies indicates the presence of metallic clusters embedded in the surface. We also show that electron tunneling via surface states gives an important contribution to the dc conductivity of these ultrathin granular metal films on Si(111).

Gold at the Si(111)-Pd interface; modification of atomic interdiffusion

We present the results of an angle integrated photoemission ( hv = 21.2eV) and Auger investigation of the effect of Au interlayers on the growth of the Si(111)-Pd interface. Gold does not remain totally blocked at the interface with increasing amounts of deposited Pd. As a consequence the same amount of Au (around 6 monolayers) acts as a promotor of Si-Pd mixing at low Pd coverages (below about 8 monolayers) and as an inhibitor at higher Pd coverages. The effect is dependent on the amount of Au so that Au at 10 monolayers is an inhibitor independent of the Pd coverage. The implications on the interface growth mechanism are discussed.

Catalysis of hydrogen evolution via deposition of palladium onto electrodes modified with an N,N'-dialkyl-4,4'-bipyridinium-based polymer: dependence of rate of palladium coverage

The H/sub 2/ evolution activity of indium tin oxide (ITO) (PQ/sup 2+/+/)/sub n//Pd surfaces has been characterized in pH 2.8/1 M KCl aqueous solution where (PQ/sup 2+/+/)/sub n/ is a redox polymer derived from an N,N'-dialkyl-4,4'-bipyridinium monomer and Pd is deposited onto the outer surface of the polymer by reduction of PdCl/sub 4//sup 2 -/. The coverage of (PQ/sup 2+/+)/sub n/ used, (4-5) x 10/sup -8/ mol/cm/sup 2/, can carry a maximum current density, i/sub E/, of 11-15 mA/cm/sup 2/. When the Pd coverage is approx. 5 x 10/sup -8/ mol/cm/sup 2/ the maximum current from kinetic limitations only, i/sub k/, for H/sub 2/ evolution is approx. 20 mA/cm/sup 2/. However, the observed current, i/sub obsd/, is limited by both I/sub E/ and i/sub k/. We predict that a polymer coverage of approx. 1 x 10/sup -8/ mol/cm/sup 2/ with a Pd coverage of 5 x 10/sup -8/ mol/cm/sup 2/ is sufficient to give an observed H/sub 2/ evolution current density of 15-20 mA/cm/sup 2/. These considerations apply at pH 2.8 and an electrode potential of -0.78 V vs. SCE. The Pd deposited onto the polymer is found to be significantly less active than Pd deposited directly onto ITO, butmore » structured ITO/(PQ/sup 2+/+/)/sub n//Pd interfaces have been demonstrated to have good integrity and can provide significant current density, greater than or equal to 10 mA/cm/sup 2/, at modest, < 10/sup -7/ mol/cm/sup 2/, Pd coverages. 12 references, 3 figures, 1 table.« less

The formation of the reacted interface for the Si/Cu/Pd system: An ultraviolet photoemission study

Various coverage of Pd were deposited on substrates formed by few layers of Cu on Si (111) face at room temperature. Photoemission results show that the formation of Pd 2 Si-like compound is strongly inhibited by the presence of the Cu/Si interlayer and suggest that for the studied system the heat of formation of the compound is not the only parameter that controls the thickness of the intermixed region.

Observation of a first-order structural phase transition in a monolayer of Pd on Nb(110) and its relationship to electronic structure

It is shown that a commensurate 1\ifmmode\times\else\texttimes\fi{}1 monolayer of Pd on a Nb(110) surface undergoes a first-order structural phase transition to an incommensurate monolayer Pd(111) structure when \ensuremath{\approx} 18% more Pd atoms are added to the surface. We find, however, that this phase transition has no substantial effect on the surface electronic structure. A steady shift toward the Fermi level of the local density of states at the surface is found as the Pd coverage is increased beyond about one monolayer. This shift correlates quite well with a previously observed increase in the hydrogen uptake rate.

Electronic structure of transition-metal-transition-metal interfaces: Pd on Nb(110)

With the use of a self-consistent tight-binding scheme, local electronic densities of states (LDOS) are calculated for commensurate (C) Pd* (110) and incommensurate (IC) Pd(111) layers on Nb(110). At low coverages ($\ensuremath{\Theta}&lt;1$) Pd induces a Pd-Nb resonance near the bottom of the Nb $d$ band. In agreement with the experimental results this resonance shifts to lower binding energy as the coverage is increased. Comparing detailed features of the LDOS with the photoemission data we predict a C-IC structural transformation for approximately one monolayer of Pd. This we also conclude from a simple interface energy calculation. The $d$-electron charge transfer across the interface and the LDOS of Pd at the Fermi energy (${E}_{F}$) are small. The latter, however, increases with the Pd coverage. The LDOS of the top Pd layer is nearly the same for Pd(111) on Pd* (110)-Nb(110) or Pd(111)-Nb(110) interfaces and in all cases there is a significant decrease of Nb $d$ states near the ${E}_{F}$. This shows that the LDOS is mainly perturbed at the interface. Implications of these results for hydrogen uptake and work-function changes are discussed.

Auger and electron energy-loss study of the Pd/SiC interface and its dependence on oxidation

Auger and electron energy-loss spectroscopies have been used to observe the formation and oxidation of the Pd/SiC interface. The interface is reactive at room temperature, and intermixing produces a surface phase consisting of Pd silicide with some dissolved C. The dependence of the Si(L 2,3VV) lineshape on Pd coverage and on annealing suggests that the local environment of Si in the silicide is that of a mixture of Pd 3Si and Pd 4Si rather than that of Pd 2Si, which is formed when Pd is deposited on elemental Si. There is no indication of Pd carbide formation even after annealing at ∼800°C. Oxidation-induced changes in the spectra indicate that (for room-temperature exposures up to 10 6 L (=1 Torr s)) adsorption occurs preferentially on Si sites.

Chemisorption properties of thin Pd overlayers on Nb(110) and Ag(111)

A simple model of hydrogen adsorption on a Pd covered Nb(110) substrate is presented. In agreement with recent experiments, it is found that for the commensurate bcc (110) phase, dissociative H 2 adsorption is not favoured in contrast to the incommensurate fcc (111) phase. The major reason for the dramatic reduction in surface reactivity for the bcc (110) phase, is due to the increase in Pd-Pd distances which results in a narrowing of the d band and also a lowering of the antibonding resonance made up of the H 1s orbital and the substrate d states, to an energy position where it is now occupied. A similar behaviour is expected for H adsorbed on a Pd/Ag(111) composite substrate, where at low Pd coverages (less than one monolayer) the expanded Pd layer is expected to act against the H 2 dissociation. Calculations of the one-electron energies due to the metal d states for H on Pd and Nb surfaces, indicate a larger chemisorption energy for Nb than Pd. The H-surface bond is thus more easily broken for a Pd covered Nb surface, than for a clean Nb surface, thereby resulting in a faster H penetration into the bulk of the composite substrate.

Chemisorption properties of thin Pd overlayers on Nb(110) and Ag(111)

Abstract A simple model of hydrogen adsorption on a Pd covered Nb(110) substrate is presented. In agreement with recent experiments, it is found that for the commensurate bcc (110) phase, dissociative H 2 adsorption is not favoured in contrast to the incommensurate fcc (111) phase. The major reason for the dramatic reduction in surface reactivity for the bcc (110) phase, is due to the increase in Pd-Pd distances which results in a narrowing of the d band and also a lowering of the antibonding resonance made up of the H 1s orbital and the substrate d states, to an energy position where it is now occupied. A similar behaviour is expected for H adsorbed on a Pd/Ag(111) composite substrate, where at low Pd coverages (less than one monolayer) the expanded Pd layer is expected to act against the H 2 dissociation. Calculations of the one-electron energies due to the metal d states for H on Pd and Nb surfaces, indicate a larger chemisorption energy for Nb than Pd. The H-surface bond is thus more easily broken for a Pd covered Nb surface, than for a clean Nb surface, thereby resulting in a faster H penetration into the bulk of the composite substrate.

Enhanced oxygen adsorption on palladium silicide at room temperature

Following Pd deposition on Si(111), Pd2Si formation was observed at room temperature up to a Pd coverage of about 10 A. Exposure to oxygen revealed an estimated four times larger oxygen uptake on the thin Pd2Si layers with respect to pure Si. Thicker Pd2Si layers formed upon heating of unreacted Pd layers showed the same enhanced oxygen adsorption. The Si(LVV) Auger line shape does not indicate SiO2 formation but rather maintains the form characteristic for Pd2Si.

Low energy electron loss spectroscopic study of Pd-Si(111) system

Effect of Pd deposition on a clean Si(111) surface was studied by ELS and AES methods for submonolayer [1 ML = 7.8 × 10 14 atoms cm -2 for Si(111)] to several tens of monolayers. ELS spectra showed that the electronic nature of Pd-Si bonding for ≲ 1 ML of Pd coverage is different from Pd 2Si formed for ≳ 3 ML. Namely, it was shown that some critical thickness for Pd on Si(111) exist for inducing interfacial intermixing reaction at room temperature.

Exploiting energy-dependent photoemission in Si d-metal interfaces: The Si(111)–Pd case

We present synchrotron radiation photoemission results from Si(111) with increasing Pd coverages ϑ (in monolayer units from 0.5 to 23 monolayers). We have measured photoemission spectra in the Cooper minimum for the Pd 4d electrons (hν = 130 eV) and outside this region (hν = 80 eV) where the d emission in strong. By comparing the two cases, one obtains information on the Si (sp) and Pd (d) states. Making use of this, we discuss in detail the coverage dependence of the spectra in connection with the nature of the chemical bond between Si and Pd and the structure of the Si–silicide interface.

Electronic states and atomic structure at the Pd2Si–Si interface

Electronic states and atomic structure at the Pd2Si–Si interface have been investigated using AES and UPS in conjunction with transmission electron microscopy (TEM). By following the core and valence spectra from submonolayer to several atomic layers of Pd coverages, the changes in the partial state density of the Pd and Si atoms during interface formation have been observed. Stoichiometry calibration based on relative variation of Auger intensities reveals extra Si states existing within a few Å of the interface. Analysis of the spectral shapes and peak shifts shows that these Si states are derived from a Si-rich environment at the interface and are located mostly near the Si bandgap region. TEM lattice images reveal a structurally sharp Pd2Si–Si (111) interface with misfit dislocations and atomic steps present within several Å of the interface. Combining the spectroscopy and TEM results, we infer that the interface states originate from chemical bonds associated with the atomic defect structures produced by silicide formation.

Chemical and structural properties of the Pd/Si interface during the initial stages of silicide formation

The initial stage of silicide formation at the Pd/Si interface has been investigated during Pd deposition on the Si(111) surface using a combination of AES and TEM techniques. The formation of silicide was found to change the valence state of Si significantly as observed from the shape change of the Si LVV Auger spectra. The ratios of the normalized peak intensities were used to calibrate the composition and the thickness of the reacted layer. The calibration results coupled with TEM phase identification showed uniform formation of the Pd2Si compound even at room temperature up to about 10 Å Pd coverage. Using thin Pd coverages of about 10 Å, epitaxial growth of Pd2Si was not observed on sputter-cleaned Si surfaces. Crystalline order was clearly observed at the Pd2Si/Si interface in annealed samples with less than 4 Å Pd2Si, indicating a relatively sharp interface with no significant amorphous-like structural disorder.

Sodium without electrolysis

Oxygen reduction reaction (ORR) was studied on polycrystalline gold, Au(poly), modified by nanosized palladium islands in 0.1 M HClO4 solution using rotating disc electrode technique. Paladium was spontaneously deposited from 1 mM PdSO4·2H2O + 0.5 M H2SO4 solution for various deposition times at a submonolayer coverage. Topography of obtained Pd/Au(poly) surfaces was observed by tapping mode atomic force microscopy (AFM), while chemical surface composition changes were detected by simultaneously performed phase AFM. Deposited Pd islands were nonuniform in size and randomly distributed over the gold substrate. Size and number of Pd islands increase with the increase of the deposition time, and consequently surface coverage increases too. Deposited Pd was also identified by the changes in cyclic voltammetry (CV) profiles from which an active surface area was estimated. Obtained Pd/Au(poly) surfaces have shown a significant catalytic activity towards oxygen reduction reaction which increases with the increase of Pd islands coverage. The initial potential of ORR was shifted positively, while reaction pathway changes from 2e-reduction on pure gold to 4e-reduction on Pd modified gold. Additionally, obtained Pd/Au(poly) surfaces have shown a significant activity towards hydrogen peroxide reduction (HPRR), which appear as an intermediate during ORR. This supports the assumption that ORR on Pd/Au(poly) occurs partly through 4e-series reduction pathway, the fraction of which increases with the increase of Pd coverage.