Elsevier

Applied Surface Science

Volume 254, Issue 1, 31 October 2007, Pages 392-396
Applied Surface Science

Modeling of the deposition of Ni and Pd on Mo(1 1 0)

https://doi.org/10.1016/j.apsusc.2007.07.056Get rights and content

Abstract

Recent experimental work on the deposition of fcc metals on a bcc substrate motivates this atomistic modeling analysis of Ni and Pd deposition on Mo(1 1 0). A detailed atom-by-atom analysis of the early stages of growth, focusing on the formation of surface alloys and 3D islands is presented, identifying the interactions leading to each type of behavior. Further analysis describes the growth pattern as a function of coverage. Temperature effects are studied via Monte Carlo simulations using the Bozzolo–Ferrante–Smith (BFS) method for alloys for the energetics.

Introduction

Recent work combining different adsorbates and substrates has been performed in order to determine the growth process in each case. In particular, the study of fcc metal deposition on bcc substrates has been of interest not only because of the relatively less abundant literature than for deposition on fcc substrates, but also because of the interest on the formation of bidimensional superstructures.

Deposition of Pd on Mo(1 1 0) is one particular case of a bcc substrate with relevance in the study of catalysis, due to the bimetallic nature of the surface, and in microelectronics, in relation to interdiffusion in the semiconducting substrate. Recent experimental evidence [1] indicates that a surface alloy can form at low temperatures even for a refractory substrate. The work of Park et al. [1] shows that there is pseudomorphic growth for 1 ML coverage at room temperature, with no evidence of Mo diffusion in the Pd surface when annealed at 750 K. Instead, a Pd–Mo alloy covered by 1 ML Pd forms upon annealing at 770 K for coverages between 1 and 2 ML. For even higher coverage (∼12 ML) Mo atoms diffuse forming an alloy upon annealing at 770 K.

For Ni deposition on the same substrate, the LEED, AES, and TDS studies by He et al. [2] show the prevalence of a Frank–van der Merwe (FM) growth mode for T  115 K, but switches to the formation of 3D Ni islands for T  330 K. Tikhov and Bauer [3], using LEED and AES, established FM growth for low temperatures, Stranski-Krastanov growth for intermediate temperatures and surface alloy formation for high temperatures. The RHEED study by Tsunematsu and Gotoh [4] shows that for submonolayer Ni coverage (0.1 ML < θ < 0.5 ML) at room temperature, Ni atoms are adsorbed randomly in the Mo(1 1 0) substrate, whereas for higher coverage (0.5 ML < θ < 0.9 ML) a superstructure is observed. The formation of 2D or 3D islands is apparent for θ > 1 ML and temperatures in the range 923–1123 K.

Given the uncertainties in the experimental results and the relevance of solid films on bcc substrates, this work reports the results of an atomistic modeling of the deposition of Ni or Pd on Mo(1 1 0), using the BFS method for alloys for the energetics.

Section snippets

The BFS method for alloys

The BFS method [5] is a simple quantum approximate method based on the concept that the energy of formation ΔH of a given atomic configuration is the superposition of the individual atomic contributions, ΔH = Σɛi. Each contribution by atom i, ɛi, is the sum of two terms: a strain energy, εiS, computed in the actual lattice as if every neighbor of the atom i was of the same atomic species i, and a chemical energy, εiC, computed as if every neighbor of the atom i was in an equilibrium lattice site

Results and discussion

Monte Carlo–Metropolis simulations provide information on the ground state of the system which is not always reached under experimental conditions. Therefore, the BANN algorithm [8], a variant of the standard scheme, was used in this work due to its proven record in providing simulation results that yield a better match with experimental results. BANN only allows for exchanges between atoms in nearest-neighbor sites, which are accepted or rejected depending on the available thermal energy [8].

Conclusion

Atomistic modeling indicates that small differences in the energetics of the interaction between adatoms (Pd or Ni) with substrate (Mo) atoms can result in drastically different behavior with increasing temperature and coverage. The trend for Pd atoms forming bonds with Mo atoms can eventually overcome the unfavorable energy cost of Mo atoms migrating to the overlayer, resulting in a disordered subsurface alloy. For the Ni case, however, it is the trend for Mo atoms to bond with Ni atoms what

Acknowledgements

Fruitful discussions with N. Bozzolo are gratefully acknowledged. This work was partially supported by the NASA Fundamental Aeronautics program.

References (8)

  • Ch. Park et al.

    Surf. Sci.

    (1985)
  • M. Tikhov et al.

    Surf. Sci.

    (1990)
  • P. Blaha et al.

    Comput. Phys. Commun.

    (1990)
  • G. Bozzolo et al.

    Comp. Mater. Sci.

    (2002)
There are more references available in the full text version of this article.
View full text