Impurity modes and effect of clustering in diluted semiconductor alloys

TL;DR

This paper investigates how impurity vibrational modes in mixed zincblende semiconductors are affected by local clustering and alloy concentration, using a unified approach and first-principles calculations to understand spectral variations.

Contribution

It introduces a first-principles framework to analyze impurity modes and clustering effects in diluted semiconductor alloys, linking vibrational spectra to local structural variations.

Findings

Impurity phonon modes evolve predictably with clustering.

Vibrational frequency splitting depends on local bond length variations.

First-principles calculations can reproduce experimental spectral features.

Abstract

The variation of TO zone-center vibration spectra with concentration in mixed zincblende-type semiconductors can be understood within a paradigm of unified "one bond - two modes" approach, which has been recently outlined as a rather general concept, and emerges from a number of previous experimental and theoretical studies. The crucial issue is that the vibration frequency, associated with a certain cation-anion bond, depends on the length of the latter, and the bond length, in its turn, depends not only on the average alloy concentration, but on local variations of it. In an (A,B)C substitutional alloy, the A-C bond length differ in A-rich and A-poor regions, yielding a splitting of the A-C vibration frequency. Such splittings can be measured and reproduced in first-principles calculations. An analysis of vibration spectra helps to get an insight into the structural short-range (clustering) and long-range (formation of extended chains of certain cation-anion pairs and other structural motives at the mesoscopic scale) tendencies. For this however, one needs first-principles benchmark calculations for representative model systems. The simplest yet important result from first-principles calculations is a prediction of how the impurity phonon mode evolves as isolated (distant) impurities get clustered.