Transition metal induced ring-cluster structures on Si(111)

Abstract

We characterize a ‘‘ring-cluster’’ (RC) structure produced by high temperature annealing of transition metals on Si(111), studied by scanning tunneling microscope and low-energy electron microscopy. This structure consists of a single metal atom on a substitutional site in the top layer of Si(111) plus an overlying ring of six Si adatoms with an unusual bridge bonded topology. We find that the same structure occurs for all near noble metals specifically Fe, Co, Ni, Pd, Pt and Ir. It does not occur for refractory metals (Ti, W) or for noble metals (Cu, Ag). Clear phase separation of clean Si(111)-7×7 and large triangular domains containing a lattice gas of RCs occurs only for quenching from above the 7×7 to 1×1 phase transition at T=840 °C. The density of RCs within a domain increases with average metal coverage. The domain orientation is defined by its boundary with the faulted half of 7×7 unit cells. The large scale morphology for RC induced structures and its dependence on quench rate suggests a precipitation reaction. Dissolution rate of nickel RCs into bulk Si was found to be t=20 s at 800 °C assuming exponential decay.

Comparative STM study of SPE growth of FeSi2 nanodots on Si(111)7×7 and -R30°-B surfaces

“…1(b)). Our STM observations have revealed that the structure of the disordered regions is basically similar to that of the so-called ''1 · 1''-RC (ring-cluster) phase [16][17][18][19][20]. This phase is known to be induced by many silicide-forming metals (e.g., Co, Ni, Pd, Pt, Ir) and it comprises a disordered array of the ring clusters.…”

Formation of two-dimensional silicide on Ni(100) surface

“…19 -R23.4°, and (1 × 1)-RC phases on a Si(111) surface, [7][8][9][10][11][12][13][14][15][16][17][18][19][20] and ´n 2 and 2 2-R45°p hases on a Si(100) surface 11,[21][22][23] have no analogous bulk silicide. Although the lattice mismatch between Si and bulk NiSi 2 is negligible (−0.44%), no epitaxial 2-D silicides corresponding to the low-index face of NiSi 2 or other bulk silicides are formed on Si substrates.…”

“…In contrast to Si on Ni, Ni-induced reconstructions on Si substrates do not have a uniform 2-D silicide covering the entire surface with a single phase. [10][11][12][13][14]16,17) This fact is due to the large diffusion of Ni into the bulk Si which prevents a flat 2-D silicide forming on Si substrates, 54) and before a uniform 2-D silicide is completed, deposited Ni atoms diffuse into the bulk Si and precipitate more stable 3-D silicides, which induce inhomogeneity of the surface layer.…”

“…19 -R23.4°, and (1 × 1)-ring cluster (RC) phases. [7][8][9][10][11][12][13][14][15][16][17][18][19][20] These surface silicides are thought to be intermediate states before the formation of thick epitaxial silicides. It has been reported that ´n 2 and 2 2-R45°phases form on a Si(100) substrate depending on the Ni coverages and substrate temperatures.…”

Interface Effects on Epilayer Surface Density of States by Scanning Tunneling Spectroscopy and Density Functional Theory

“…STM micrographs at low (Figure a) and high (Figure d) magnification show that the terrace‐step bunch structure of the vicinal Si(111) substrate is covered by a thin quasi‐continuous flat and reconstructed film. Inside the holes in the film, a secondary phase of ring cluster (RC) patches—a well‐known sub‐monolayer (ML) surface structure of transition metal silicides—is clearly seen . The characteristic Si(111)‐(7 × 7) reconstruction could not be observed.…”