# Artificial Intelligence for Perovskite Additive Engineering: From Molecular Screening to Autonomous Discovery

## TL;DR

This paper explores how artificial intelligence can revolutionize the discovery of additives for perovskite solar cells, speeding up the development process.

## Contribution

The paper introduces AI-driven methods for additive discovery, integrating machine learning and autonomous laboratories.

## Key findings

- AI-driven approaches can efficiently identify suitable additives for perovskite solar cells.
- Active learning algorithms reduce the need for extensive experimental iterations in precursor formulation.
- Integration of large language models with autonomous labs enables closed-loop discovery.

## Abstract

Additive engineering plays a crucial role in enhancing the performance of perovskite solar cells (PSCs), yet identifying suitable additives within the vast chemical space remains a significant challenge. This paper describes a paradigm shift in additive discovery from trial-and-error methods to AI-driven approaches. We first establish the physicochemical foundations of additive engineering and the descriptors commonly employed in machine learning algorithms. Next, we discuss intelligent process optimization, highlighting how active learning algorithms effectively tune complex precursor formulations with minimal experimental iterations. Additionally, we explore the role of AI in mechanism elucidation and the potential prospects of generative models in the field of additives. Finally, we emphasize the emerging trend of integrating large language models with autonomous laboratories for closed-loop autonomous discovery, offering a promising pathway to accelerate the commercialization of PSCs.