Machine learning-assisted design and explainable optimization of CdSnP2-based integrated solar-photodetector devices
Abstract
CdSnP2-based integrated solar cell-photodetector (SC-PD) devices employing CdS and CuGaSe2 (CGS) as the window and back surface field (BSF) layers, respectively are investigated using a hybrid machine learning (ML)-assisted SCAPS-1D framework. Device optimization is performed by varying the thickness, doping concentration, and defect density of individual layers. SCAPS-generated data are used to train six ML models and one deep learning model, with ensemble-based algorithms exhibiting the highest predictive accuracy. The ML-guided optimization identifies the n-CdS/p-CdSnP2 (CTP)/p+-CGS architecture as the optimum configuration among thirteen candidate structures. Incorporation of a 200 nm CGS BSF layer significantly enhances both photovoltaic and photodetection performance, increasing the efficiency from 20.67% to 32.69%, responsivity from 0.53 AW-1 to 0.72 AW-1, and detectivity from 2.51x1014 Jones to 1.78x1016 Jones. SHapley Additive exPlanations (SHAP) analysis reveales that band-offset engineering, particularly at the window/absorber and absorber/BSF interfaces, together with absorber properties, governs device performance. These findings demonstrate the potential of CdSnP2 and the proposed data-driven SCAPS-ML framework for the accelerated design of high-efficiency multifunctional optoelectronic devices.
Related
- Explainable AI-assisted optimization for Feynman integral reduction
- Resource optimization for RIS-UAV integrated air-ground cooperative semantic communication driven by deep reinforcement learning
- CFD and machine-learning-assisted multi-objective optimization of trailing-edge serrations and vortex generators for flow-separation control on a NACA 0012 Airfoil
- An Algorithm for computing Three- and Four-Body Density Matrices from CASCI Wavefunction and AI-Assisted Code Optimization
- Unveiling wing turbulence dynamics through explainable deep learning
BibTeX
@misc{pappu2026machine,
title = {{Machine learning-assisted design and explainable optimization of CdSnP2-based integrated solar-photodetector devices}},
author = {Md. Alamin Hossain Pappu and Kazi Abrar Shafin and Mainul Hossain and Jaker Hossain},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.03642},
eprint = {2609.03642},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.03642},
}
RIS
TY - UNPB
TI - Machine learning-assisted design and explainable optimization of CdSnP2-based integrated solar-photodetector devices
AU - Md. Alamin Hossain Pappu
AU - Kazi Abrar Shafin
AU - Mainul Hossain
AU - Jaker Hossain
PY - 2026
DA - 2026/09/03
PB - arXiv
DO - 10.48550/arxiv.2609.03642
UR - https://arxiv.org/abs/2609.03642
AB - CdSnP2-based integrated solar cell-photodetector (SC-PD) devices employing CdS and CuGaSe2 (CGS) as the window and back surface field (BSF) layers, respectively are investigated using a hybrid machine learning (ML)-assisted SCAPS-1D framework. Device optimization is performed by varying the thickness, doping concentration, and defect density of individual layers. SCAPS-generated data are used to train six ML models and one deep learning model, with ensemble-based algorithms exhibiting the highest predictive accuracy. The ML-guided optimization identifies the n-CdS/p-CdSnP2 (CTP)/p+-CGS architecture as the optimum configuration among thirteen candidate structures. Incorporation of a 200 nm CGS BSF layer significantly enhances both photovoltaic and photodetection performance, increasing the efficiency from 20.67% to 32.69%, responsivity from 0.53 AW-1 to 0.72 AW-1, and detectivity from 2.51x1014 Jones to 1.78x1016 Jones. SHapley Additive exPlanations (SHAP) analysis reveales that band-offset engineering, particularly at the window/absorber and absorber/BSF interfaces, together with absorber properties, governs device performance. These findings demonstrate the potential of CdSnP2 and the proposed data-driven SCAPS-ML framework for the accelerated design of high-efficiency multifunctional optoelectronic devices.
ER -