Long-Range Machine Learning of Electron Density for Twisted Bilayer Moiré Materials
Zekun Lou, Alan M. Lewis, Mariana Rossi · 2026
Read Long-Range Machine Learning of Electron Density for Twisted Bilayer Moiré Materials by Zekun Lou, Alan M. Lewis, Mariana Rossi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-09938.txt.
Description
Twisted layers of two-dimensional materials host striking quantum effects, but modeling them from first principles is punishingly expensive. A 2026 paper led by Zekun Lou, with two co-authors, trains a Gaussian process model called SALTED on electron densities of small, shifted bilayers. The aim is to predict densities for the large supercells that small twist angles demand.
Long-range descriptors prove essential: without them, predicted densities yield unreliable band structures and electrostatic properties. The choice of descriptor also shapes where density errors land, which in turn moves downstream electronic results. Tested on twisted graphene, on hexagonal boron nitride, and on transition metal dichalcogenides, the model captures flat bands, narrowing bandwidths, domain-wall fields, and spin-orbit effects.
Catalog record
| CREATOR | Zekun Lou, Alan M. Lewis, Mariana Rossi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 120 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-09938