# Deep Learning Framework for Atomic-Level Design and Presynthesis Prediction of Coinage-Metal Nanoclusters

## TL;DR

This paper introduces a deep learning framework called CoLiM that predicts the compatibility of nanocluster components before synthesis, enabling precise atomic-level design of coinage-metal nanoclusters.

## Contribution

The novel contribution is a deep learning model for presynthesis prediction of inorganic core-ligand compatibility in nanoclusters, enabling inverse design.

## Key findings

- CoLiM achieves an AUC exceeding 0.83 on a test set of nanocluster structures.
- The model successfully guided the synthesis of a modified nanocluster through single-atom editing.
- The framework demonstrates practical utility in inverse synthesis and atomic-level tailoring of nanoclusters.

## Abstract

The atomically precise
nature of coinage-metal nanoclusters (CMNs)
enables systematic exploration of structure–property relationships
and motivates application oriented inverse design. However, the synthesis
of CMNs typically relies on trial-and-error methods, with atomic-level
structures only revealed through crystallography (postsynthesis),
posing a major challenge to the deterministic synthesis of predesigned
cluster structures, which is known as inverse synthesis. Here, we introduce CoLiM, a deep neural network framework that
predicts the chemical compatibility between the unexplored inorganic
core and ligands before synthesis. CoLiM employs
a dual-encoder architecture and is trained on a newly constructed
dataset comprising 1,989 reported CMN structures, supplemented by
an additional gas-phase cluster dataset. The optimal CoLiM model achieves
an area under the curve (AUC) exceeding 0.83 on a held-out test set,
outperforming all of the baseline methods. To demonstrate its practical
utility, CoLiM is applied to address the long-standing challenge of
achieving atomically precise structural tailoring. Starting from [Cu20Cl­(PET)12(PPh3)4­(MeCOO)6]+, we successfully performed single-atom editing
on its inorganic core to synthesize [Cu19Cl­(PET)12(PPh3)3­(HCOO)6] guided by
the prediction of CoLiM, validating the model’s generalizability
under real experimental conditions. Our framework facilitates the
inverse synthesis and precise atomic-level modification of nanoclusters,
underscoring its substantial potential to accelerate rational nanocluster
discovery.