Predictive Inorganic Synthesis based on Machine Learning using Small Data sets: a case study of size-controlled Cu Nanoparticles

TL;DR

This paper demonstrates that small, well-curated datasets combined with classical machine learning models can accurately predict the size of copper nanoparticles synthesized via microwave-assisted methods, highlighting the potential for efficient, resource-conscious materials research.

Contribution

The study shows that high-accuracy predictions of nanoparticle size are achievable using small datasets and ensemble regression models, with limited benefit from complex models like LLMs in data-scarce scenarios.

Findings

Ensemble regression models achieved R^2 = 0.74 in size prediction.

Classical statistical approaches had R^2 = 0.60, showing improvement with ML.

LLMs did not significantly outperform simpler models in small data conditions.

Abstract

Copper nanoparticles (Cu NPs) have a broad applicability, yet their synthesis is sensitive to subtle changes in reaction parameters. This sensitivity, combined with the time- and resource-intensive nature of experimental optimization, poses a major challenge in achieving reproducible and size-controlled synthesis. While Machine Learning (ML) shows promise in materials research, its application is often limited by scarcity of large high-quality experimental data sets. This study explores ML to predict the size of Cu NPs from microwave-assisted polyol synthesis using a small data set of 25 in-house performed syntheses. Latin Hypercube Sampling is used to efficiently cover the parameter space while creating the experimental data set. Ensemble regression models successfully predict particle sizes with high accuracy ($R^2 = 0.74$), outperforming classical statistical approaches ($R^2 = 0.60$). Additionally, classification models using both random forests and Large Language Models (LLMs) are evaluated to distinguish between large and small particles. While random forests show moderate performance, LLMs offer no significant advantages under data-scarce conditions. Overall, this study demonstrates that carefully curated small data sets, paired with robust classical ML, can effectively predict the synthesis of Cu NPs and highlights that for lab-scale studies, complex models like LLMs may offer limited benefit over simpler techniques.