The carbon cost of materials discovery: Can machine learning really accelerate the discovery of new photovoltaics?

TL;DR

This paper evaluates the environmental and predictive trade-offs of replacing density functional theory with machine learning models in photovoltaic materials discovery, proposing hybrid strategies for more sustainable and efficient screening.

Contribution

It introduces a framework to quantify CO2 emissions of computational workflows and demonstrates hybrid ML/DFT strategies that balance accuracy and environmental impact in PV materials screening.

Findings

ML surrogates reduce computational emissions significantly.

Direct prediction of efficiency outperforms spectral intermediate steps.

ML trained on DFT data can surpass DFT in screening accuracy.

Abstract

Computational screening has become a powerful complement to experimental efforts in the discovery of high-performance photovoltaic (PV) materials. Most workflows rely on density functional theory (DFT) to estimate electronic and optical properties relevant to solar energy conversion. Although more efficient than laboratory-based methods, DFT calculations still entail substantial computational and environmental costs. Machine learning (ML) models have recently gained attention as surrogates for DFT, offering drastic reductions in resource use with competitive predictive performance. In this study, we reproduce a canonical DFT-based workflow to estimate the maximum efficiency limit and progressively replace its components with ML surrogates. By quantifying the CO$_2$ emissions associated with each computational strategy, we evaluate the trade-offs between predictive efficacy and environmental cost. Our results reveal multiple hybrid ML/DFT strategies that optimize different points along the accuracy--emissions front. We find that direct prediction of scalar quantities, such as maximum efficiency, is significantly more tractable than using predicted absorption spectra as an intermediate step. Interestingly, ML models trained on DFT data can outperform DFT workflows using alternative exchange--correlation functionals in screening applications, highlighting the consistency and utility of data-driven approaches. We also assess strategies to improve ML-driven screening through expanded datasets and improved model architectures tailored to PV-relevant features. This work provides a quantitative framework for building low-emission, high-throughput discovery pipelines.