A QMC-deep learning method for diffusivity estimation in random domains

TL;DR

This paper introduces a novel quasi-Monte Carlo deep learning approach to accurately and efficiently estimate exciton diffusion length in complex, high-dimensional random heterojunctions, aiding the design of advanced opto-electronic devices.

Contribution

It combines quasi-Monte Carlo sampling with deep neural networks to model exciton diffusion in high-dimensional random domains, surpassing traditional simulation capabilities.

Findings

High-accuracy estimation of exciton diffusion length.

Unprecedented efficiency in parameter space exploration.

Provides insights into interfacial effects on diffusion.

Abstract

Exciton diffusion plays a vital role in the function of many organic semiconducting opto-electronic devices, where an accurate description requires precise control of heterojunctions. This poses a challenging problem because the parameterization of heterojunctions in high-dimensional random space is far beyond the capability of classical simulation tools. Here, we develop a novel method based on quasi-Monte Carlo sampling to generate the training data set and deep neural network to extract a function for exciton diffusion length on surface roughness with high accuracy and unprecedented efficiency, yielding an abundance of information over the entire parameter space. Our method provides a new strategy to analyze the impact of interfacial ordering on exciton diffusion and is expected to assist experimental design with tailored opto-electronic functionalities.