Machine Learning Accelerates Lead-Free Dual-Absorber Perovskite Solar Cell Design, Paving Way for 32.5% Efficiency

July 26, 2026

MDPI (Materials journal) Switzerland

Overview

A new study published in MDPI’s Materials journal introduces an automated optimization framework integrating SCAPS-1D simulation with machine learning to accelerate the design of lead-free dual-absorber perovskite solar cells. This approach demonstrates that dual-absorber devices can achieve over 30% power conversion efficiency by more effectively utilizing the solar spectrum. Notably, perovskite/silicon tandem cells reached 32.5% efficiency, marking a significant contribution to high-efficiency solar cell development.

In Depth

Key Findings

A recent study published in MDPI’s “Materials” journal introduces an automated optimization framework that integrates SCAPS-1D simulation with advanced machine learning techniques. This innovative approach dramatically accelerates the design of lead-free double-absorber perovskite solar cells, demonstrating the potential to achieve power conversion efficiencies of 32.5% in perovskite/silicon tandem configurations.

Technical / Clinical Details

The researchers employed various machine learning methods, including surrogate model-assisted reinforcement learning, multi-algorithm optimization comparison, and transfer learning, to optimize the structure and material composition of perovskite solar cells. The double-absorber architecture, in particular, enables efficient absorption across a broader range of the solar spectrum, allowing for power conversion efficiencies that surpass the limits of single-absorber devices. Simulations have shown that these double-absorber devices can achieve PCEs exceeding 30%.

This framework identifies optimal parameters such as material layer thicknesses, compositions, and doping concentrations with unprecedented speed and accuracy, outperforming traditional experimental trial-and-error or simple simulation methods. By specifically focusing on environmentally friendly lead-free perovskite materials, the research pioneers a path toward combining high performance with sustainability.

Background & Context

Perovskite solar cells are widely regarded as a pivotal technology for the future of photovoltaics due to their high efficiency potential and low manufacturing costs. Concurrently, there is a strong imperative to develop lead-free materials for environmental reasons, and tandem structures, which break through the efficiency limits of existing silicon solar cells, are at the forefront of research. However, the design space for these complex multi-layered devices is vast, making it challenging to find optimal designs using conventional methods. The integration of machine learning streamlines this design process, enabling much faster technological development.

Strategic Significance & Outlook

The machine learning-based optimization framework developed in this study is versatile and applicable not only to perovskite solar cell design but also to the development of other multi-layered optoelectronic devices and novel materials. The achievement of 32.5% efficiency represents a significant contribution towards setting new world records in photovoltaics, with the potential to profoundly impact the entire energy industry. Realizing high-efficiency, lead-free perovskite solar cells is a crucial step towards accelerating the commercialization of next-generation clean energy technologies and contributing to a sustainable society.

Source: https://www.mdpi.com/1996-1944/19/14/3091

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.