Beyond diamond: Interpretable machine learning reveals design principles for quantum defect host materials

Solid-state spin defects in wide-band-gap semiconductors are leading candidates for quantum information processing, but systematic identification of suitable host materials remains limited by the cost of first-principles screening across vast chemical spaces. We address this with a composition-only machine learning framework built on heterogeneous Rashomon set ensembles: by contrasting the feature attributions of seven diverse classifiers, we extract consensus design rules that no single model identifies alone---filled valence $s$-, $d$-, and $f$ shells, low chemical heterogeneity, and enrichment in C, S, Si, and O favor quantum compatibility. Screening $\ensuremath{\approx}45\phantom{\rule{0.28em}{0ex}}000$ thermodynamically stable compounds, we identify 122 high-confidence candidates (confidence $>0.95$), recovering most experimentally verified hosts (C, $\mathrm{SiC}, \mathrm{ZnO}$, and $\mathrm{ZnS}$) and predicting unexplored materials including ${\mathrm{TiO}}_{2}, {\mathrm{PbWO}}_{4}$, and layered chalcogenides (${\mathrm{HfS}}_{2}, {\mathrm{ZrS}}_{2}$). Density-functional perturbation theory calculations on 12 representative materials validate dielectric screening as a coherence proxy (${R}^{2}=0.89$ against experimental ${T}_{2}$), and vacancy calculations for ${\mathrm{TiO}}_{2}$ reveal deep, isolated midgap states favorable for spin-defect hosting. The framework provides transferable, physically grounded design principles for rational quantum materials discovery beyond traditional carbide and nitride hosts.