High-Throughput Combinatorial Screening of Refractory Amorphous Alloys for Fusion Capillary Systems via a Calibrated First-Principles Continuum-Thermodynamic Model
Year: 2026
Type: preprint
Abstract: Liquid metal divertors utilizing Capillary Porous Systems (CPS) represent a critical armor technology for handling extreme heat fluxes (> 10–20 MW/m2) in magnetic confinement fusion. However, conventional crystalline refractory matrices suffer from liquid-metal-induced corrosion along grain boundaries, thermal embrittlement, and grain coarsening. Amorphous metallic alloys eliminate grain boundaries, but identifying high-temperature glass formers remains computationally prohibitive via density functional theory or molecular dynamics. Here, we present a closed-form analytical framework that couples first-principles screened nuclear ion affinities (Ai), Rose Universal Binding Energy Relations (UBER), Voigt–Reuss–Hill (VRH) elastic homogenization, Egami topological size polydispersity (δ), and Miedema macroscopic mixing enthalpy (∆Hmix) with the Johnson–Samwer Cooperative Shear Model (CSM) to predict elastic moduli and glass transition temperatures (Tg) a priori directly from alloy stoichiometry. Nonlinear least-squares calibration against an experimental bulk metallic glass database demonstrates that atomic size strain distortion is intrinsically captured by the elastic shear work (GVRH · Vm, yielding α = 0.000), while core valence charge disparity creates local electrostatic screening perturbations that enhance the shear arrest barrier (β = 0.384), reducing residual root-mean-square error by 18.5% to 41.9 K. Applying this calibrated framework to a combinatorial sweep of the quaternary Zr–Ta–Ni–Ti system evaluates 839 candidate compositions meeting CPS engineering thresholds (Tg ≥ 700 K, GVRH ≥ 45 GPa) in < 1 second. We identify Zr5Ta55Ni35Ti5 as an optimal refractory candidate exhibiting a predicted Tg = 1019.1 K, shear modulus GVRH = 67.0 GPa, bulk modulus KVRH = 178.3 GPa, Poisson’s ratio ν = 0.333, deep exothermic stabilization ∆Hmix = −30.41 kJ/mol, and robust topological frustration δ = 8.23%. This framework establishes a zero-cost computational pipeline for accelerated discovery of extreme-environment amorphous materials.
Source: ChemRxiv
Author: Matthew B. Minus
Institution: Prairie View A&M University
Language: English
Subfield: Mechanical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7223177503