Unit cell of strained GeSi

Abstract

The local structure within the unit cell of strained-GeSi layers grown on Si͑001͒ has been examined by polarization-dependent extended x-ray-absorption fine structure. First-neighbor bond lengths are found to deviate only slightly from their unstrained values; however, the distortion of the cubic-unit cell by strain leads to measurable polarization-dependent changes in first-shell coordination and second-shell distances. A unifying picture of bond lengths and elasticity in strained-layer semiconductors is presented.

First-shell bond lengths inSixGe1−xcrystalline alloys

“…The experimental x-ray absorption fine structure ͑XAFS͒ results reported in the literature [21][22][23][24][25][26][27][28][29][30][31][32][33] differ qualitatively from the theoretical predictions. In contrast to the consistent theoretical picture of approximately 30% Vegard character, several XAFS studies on both amorphous and crystalline SiGe alloys have reported that the bonding in SiGe materials is essentially at the Pauling limit.…”

Phase-correct bond lengths in crystallineGexSi1−xalloys

“…This estimate coincides with a previous estimate based on a uniform differential model. 34 Consequently, it is not surprising that the bond lengths within the strained xϭ0.21 alloy may be as small if not smaller than the unstrained Ge-Ge, Ge-Si, and Si-Si bond lengths in the dilute xϭ0 limit.…”

Diffraction anomalous fine-structure study of strainedGa1−xInxAson GaAs(001)

“…Although the theory of elasticity 1 accurately describes the macroscopic distortions of the film, a consensus between experiment and theory concerning its microscopic distortions has only recently begun to form. 2,3 The ambiguity of previous experimental studies that have reported both strained 4-6 and unstrained [7][8][9][10][11][12] bond lengths in strained, quasibulk, thin-alloy films may be traced to the inherent rigidity of semiconductor bonds: strain effects often lie within the typical extended x-ray-absorption fine-structure ͑EXAFS͒ bond-length accuracy of ϳ0.02 Å. 7 Additionally, only a few theoretical studies 13,14 of the bond lengths within strained-layer semiconductors have existed with which to compare the varying experimental results, and these calculations have either been in conflict with experiment 2,4,14 or they have been performed for alloy/ substrate systems for which little or no experimental data exist.…”