Quantum engineering of GaAs nanoribbons for advanced thermoelectric energy conversion

Gallium arsenide (GaAs) nanoribbons are promising candidates for advanced thermoelectric applications due to their unique quantum confinement effects, low-dimensional electronic properties, and tunable vibrational characteristics. In this study, we employed a robust computational approach combining the Non-Equilibrium Green's Function (NEGF) formalism with density functional-based tight binding (DFTB) to investigate the electronic, phononic, and thermoelectric transport properties of GaAs nanoribbons. A distinct ≈40 meV acoustic-optical phonon gap is identified, which suppresses acoustic-optical scattering but is counterbalanced by dominant size effects including strong boundary/edge scattering, confinement-induced reductions in group velocity, and acoustic-acoustic umklapp scattering at elevated temperatures, collectively reducing lattice thermal conductivity. Quantum confinement-induced modifications in electron transport mechanisms yield pronounced peaks and sign reversals in Seebeck and Peltier coefficients near the Fermi level, with calculated ZT values exceeding 0.6 at 300 K. To extend these findings across temperature ranges, we developed a physics-informed neural network (PINN) machine learning model that predicts temperature-dependent thermoelectric behavior from 100 K to 600 K. The machine learning (ML) analysis shows that the best operating temperature is 450 K, where ZT reaches 0.85, which is 37% better than performance at room temperature. The model identifies distinct operational regimes: cryogenic temperatures (100-200 K) maximize Seebeck coefficients for cooling applications, while elevated temperatures (400-550 K) optimize power generation efficiency through the optimal trade-off between preserved quantum confinement effects and thermally enhanced electrical conductivity. This integrated first-principles and machine learning framework provides comprehensive insights into temperature-dependent transport phenomena and establishes GaAs nanoribbons as promising candidates for next-generation thermoelectric devices across a broad operational spectrum.