Davi Febba | Searcharxiv

arXiv2026

Autonomous Reliability Qualification of Ga$_2$O$_3$-based diode sensors via Safe Active Learning

Ultra-wide bandgap (UWBG) Ga$_2$O$_3$ is a promising semiconductor for high-power and high-temperature electronics. Reliable qualification of these devices under extreme operating conditions is essential, yet conventional reliability testing is inherently time-consuming. Autonomous experimentation offers a new paradigm by enabling measurement planning and model refinement to evolve in parallel in real time. We present a Safe Active Learning (SAL) framework for autonomous reliability characterization of Ga$_2$O$_3$-based diode sensors under coupled thermal and hydrogen stress. We first evaluate SAL in simulation, where it safely expands the explored region while learning the evolving rectification surface. Second, we demonstrate SAL experimentally on an automated high-temperature probe-station platform using a Pt/Cr$_2$O$_3$:Mg/$\beta$-Ga$_2$O$_3$ diode sensor of H$_2$ and temperature, spanning 0-800 ppm H$_2$ and 350-550 {\deg}C. Finally, we use the SAL-generated dataset for offline long-horizon forecasting of the diode current at a target voltage with a structured Gaussian-process model. Its condition-dependent Kohlrausch--Williams--Watts mean and residual covariance kernel were engineered with artificial-intelligence assistance using the SAL data and an auxiliary validation dataset spanning 1,000 hours at 400 {\deg}C across multiple H$_2$ concentrations. This dataset guided kernel design and validation, and the resulting model captures its long-time, saturating degradation trends. Although demonstrated here for a rectifying Ga$_2$O$_3$-based diode, SAL is applicable to other device classes whenever a suitable safety observable can be measured in situ.

physics.app-ph↗

arXiv2024

Rapid screening of molecular beam epitaxy conditions for monoclinic (InxGa1-x)2O3 alloys

Molecular beam epitaxy is one of the highest quality growth methods, capable of achieving theoretical material property limits and unprecedented device performance. However, such ultimate quality usually comes at the cost of painstaking optimization of synthesis conditions and slow experimental iteration rates. Here we report on high-throughput molecular beam epitaxy with rapid screening of synthesis conditions using a novel cyclical growth and in-situ etch method. This novel approach leverages sub-oxide desorption present during molecular beam epitaxy and as such should be broadly applicable to other material systems. As a proof of concept, this method is applied to rapidly investigate the growth space for the ternary alloy (InxGa1-x)2O3 on (010) oriented beta-Ga2O3 substrates using in-situ reflection high energy electron diffraction measurements. Two distinct growth regimes are identified and analyzed using machine learning image recognition algorithms, the first stabilizing a streaky 2x surface reconstruction typical of In-catalyzed beta-Ga2O3 growth, and the second exhibiting a spotty/faceted pattern typical of phase separation. Targeted growth of (InxGa1-x)2O3 is performed under conditions near the boundary of the two regimes resulting in a 980 nm thick epitaxial layer with In mole fraction up to 5.6%. The cyclical growth/etch method retains the ~1 nm surface roughness of the single crystal substrate, increases experimental throughput approximately 6x, and improves single crystal substrate utilization by >40x. The high-throughput MBE method enables rapid discovery of growth regimes for ultra-wide bandgap oxide alloys for power conversion devices operating with high efficiency at high voltages and temperatures, as well as optical devices such as ultraviolet photodetectors.

cond-mat.mtrl-sci↗