Accelerated discovery of topological metals for nanoscale interconnects
Data files
Jan 10, 2026 version files 842.09 KB
-
Interconnect_Results_Summary.html
208.38 KB
-
NanowireData.pdf
538.70 KB
-
NanowireData.xlsx
93.65 KB
-
README.md
1.36 KB
Abstract
The sharp increase in resistivity of copper interconnects at ultra-scaled dimensions threatens the continued miniaturization of integrated circuits. Topological metals with gapless surface states (Fermi arcs) protected by bulk topological invariants offer robust, backscattering-immune conduction. We develop an efficient computational framework to quantify 0~K surface-state transmission in TSM nanowires derived from Wannier tight-binding models that faithfully reproduce relativistic density functional theory results. Utilizing the non-equilibrium Green's function formalism, we systematically screen materials across chemical potentials and transport directions, producing a dataset of 3000 surface transmission values. This dataset supports machine learning models for rapid interconnect compound identification.
Dataset DOI: 10.5061/dryad.12jm63zb7
Description of the data and file structure
Files and variables
File: NanowireData.xlsx
Description: Data table containing extract surface transmission as a function of geometry and doping. Computed surface energies are also included. Used to train regression model.
Variables
- Compound + Transmission Direction/Surface: Chemical formula and geometry of nanowire for computations
- Surface energy: Computed surface energy in eV/Å2
- Ef: Doping of Fermi energy in eV
- Normalized Surface Transmission: Extracted value of surface transmission in units of (e2/h)/H where H is the height of the nanowire as detailed in Fig. 1. of the accompanying manuscript.
File: Interconnect_Results_Summary.html
Description: Interactive scatter plot of data in NanowireData.csv points are colored by the metric detailed in Fig. 5 of the accompanying manuscript.
File: NanowireData.pdf
Description: PDF version of data in NanowireData.csv
File: SM_Results.pdf
Description: PDF containing crystal structure image, lattice constants and nanowire transmission scaling as a function of thickness for extraction of surface transmission for each material screened.