Year: 2025
Type: article
Abstract: Significant effort has been devoted to mitigating polarization fields in nitride LEDs, as these fields are traditionally viewed as detrimental to light emission, particularly for red emission. Contrary to this prevailing notion, we demonstrate that strong polarization fields can enhance the optical-transition strength of red-emitting AlInGaN quantum wells, which has been historically challenging to achieve. By leveraging machine-learning surrogate models trained on multiscale quantum-mechanical simulations, we globally explore the heterostructure design space and uncover that larger fields correlate with higher electron–hole overlap. This relation arises from the quantum-confined Stark effect, which enables thinner wells without requiring higher indium compositions, thus overcoming a key limitation in nitride epitaxy. Structural and compositional engineering of internal fields offers a unique dimension for designing polychromatic nitride LEDs, crucial for miniaturizing LED pixels to the micrometer scale for extended-reality and biomedical applications. Broadly, our work demonstrates how machine learning can uncover unexpected paradigms for semiconductor design.
Source: ACS Photonics
Authors: Nick Pant, Rob Armitage, Emmanouil Kioupakis
Institution: University of Michigan
Language: English
Subfield: Condensed Matter Physics
Field: Physics and Astronomy
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4410205238