Method for determining optimal supercell representation of interfaces

Abstract

The geometry and structure of an interface ultimately determines the behavior of devices at the nanoscale. We present a generic method to determine the possible lattice matches between two arbitrary surfaces and to calculate the strain of the corresponding matched interface. We apply this method to explore two relevant classes of interfaces for which accurate structural measurements of the interface are available: (i) the interface between pentacene crystals and the (1 1 1) surface of gold, and (ii) the interface between the semiconductor indium-arsenide and aluminum. For both systems, we demonstrate that the presented method predicts interface geometries in good agreement with those measured experimentally, which present nontrivial matching characteristics and would be difficult to guess without relying on automated structure-searching methods.

First principles study of hBN-AlN short-period superlattice heterostructures

“…1(c)). A similar picture holds for the larger period (L AlN ∼ 4 nm) (hBN ) 12 (AlN ) 16 SL. Corroborated with our findings for the ultrathin SL as well as the other aforementioned non-symmetric SL, this indicates that the critical AlN thickness beyond which the AlN phase changes from hexagonal to wurtzite lies between 1 and 2 nm.…”

“…The value of u for each of the SL (u SL ), normalized to the value of u for bulk wAlN (u wAlN = 0.58Å), defines an internal structure parameter s ≡ u SL /u wAlN that equals 0(1) for the pure hexagonal(wurtzite) phase. After optimization and near the middle of the AlN layer we find s = 0 for (hBN ) 6 (AlN ) 4 , (hBN ) 3 (AlN ) 4 , (hBN ) 3 (AlN ) 2 and (hBN ) 1 (AlN ) 4 , s = 0.78 for (hBN ) 6 (AlN ) 8 and s = 0.91 for (hBN ) 12 (AlN ) 16 and (hBN ) 6 (AlN ) 16 SL. This suggests that the structure of the AlN layer in the optimized SL depends mainly on the AlN layer thickness L AlN , with ultrashort/thicker AlN layers relaxing into the hexagonal/wurtzite phase.…”

“…The lattice constants a of hBN and wAlN show a significant mismatch along directions parallel to the SL interface (see Table I), posing a challenge to realistic ab initio arXiv:1812.07188v1 [cond-mat.mtrl-sci] 18 Dec 2018 simulations which require large supercells. We construct the supercell by replicating the surface unit cells for each layer until the hBN/AlN supercells can be matched with less than ∼ 1% tensile/compressive lateral strain [16,17].…”

“…Next we consider larger period SL, namely (hBN ) 6 (AlN ) 8 and (hBN ) 12 (AlN ) 16 . We simulate them using unrotated 5 × 5/4 × 4 configurations with 556 and 1112 atoms/supercell respectively.…”

Determination of low-strain interfaces via geometric matching

“…In a previous article [8], some of the authors have shown how to find good matches between specific surfaces of two crystals and a related, but more simplistic, method for such a two-dimensional (2D) match has also previously been * kwj@fysik.dtu.dk proposed [9]. With the present method, all possible crystal orientations and surfaces are investigated at the same time.…”