Local structural evolution in the anionic solid solution ZnSexS1−x

Abstract

The century-old Vegard's law has been remarkably accurate in describing the evolution of the lattice parameters of almost all solid solutions. Contractions or expansions of lattice parameters of such systems depend on the size of the guest atom being smaller or larger than the host atom it replaces to form the solid solution. This has given rise to the concept of "chemical pressure" in analogy to the physical pressure. We have investigated using EXAFS the evolution of the local structure in terms of atom-pair distances extending up to the third-nearest neighbors in the family of compounds, ZnSe x S 1−x as an example of an anionic solid solution, in contrast to all previous studies focusing on cationic solid solutions. Our results establish several common features between these two types of solid solutions, while strongly suggesting that the concept of a chemical pressure is inaccurate and misleading. Most interestingly, we also find a qualitative difference between the cationic solid solutions, reported earlier, and the anionic solid solution.

Structural and electronic properties of the random alloy ZnSe$_x$S$_{1-x}$

“…1. As expected by Vegard's law, we can observe a linear increase when going from ZnS to ZnSe, which is also in line with what was reported by both experiment and theory [17]. Naturally, for the cubic phase we have only one lattice parameter, whose value is intermediate to the values of a and c found in the wurtzite structure.…”

“…This reflects what we pointed out in our recent work on the cubic phase, i.e. the behavior of the lattice parameter, being determined by a long-range averaging over many unit cells, does not depend crucially on the precise description of the local environments in ZnSe x S 1−x [17]. The comparison with recently reported [17,39] experimental data (black dots, up and down triangles) is also favorable.…”

“…Understanding these differences is of paramount importance to shed light on the connection between microscopic changes at the atomic scale and macroscopic changes at the crystal scale, following Vegard's law. In this sense, a limitation of the previous two studies [16,17] is that the EXAFS analysis yields only an average value for each bond distance. This is a gross approximation of the real world, where one expects a wide distribution of bond distances between two atoms, based on their local environment.…”

“…Several studies in the last three decades have demonstrated that the nearest neighbor bond lengths do not change drastically with composition, but remain close to their values in the pure systems [10][11][12][13][14][15][16]. In two of the most recent studies [16,17], we combined experimental measurements by EXAFS with theoretical data via density functional theory (DFT) [18,19] to identify unambiguously to what extent Vegard's law applies at the atomic scale in both cationic and anionic solid solutions, namely Zn x Cd 1−x S and ZnSe x S 1−x . The usage of large supercells constructed via special quasi-random structures (SQS) [20] allowed us to avoid the limitations of the single-site effective medium theories, discussed above, providing a meaningful comparison between theory and experiment.…”

“…While cationic and anionic solid solutions show similar features, some distinctive differences are also present. The most significant difference manifests primarily in the variations of the bond angles as a function of composition, which leads to identifying different kinds of elementary building blocks for the two types of solid solutions [16,17]. Understanding these differences is of paramount importance to shed light on the connection between microscopic changes at the atomic scale and macroscopic changes at the crystal scale, following Vegard's law.…”