Deciphering Cryptic Behavior in Bimetallic Transition-Metal Complexes with Machine Learning

Abstract

We demonstrate an alternative, data-driven approach to uncovering structure–property relationships for the rational design of heterobimetallic transition-metal complexes that exhibit metal–metal bonding. We tailor graph-based representations of the metal-local environment for these complexes for use in multiple linear regression and kernel ridge regression (KRR) models. We curate a set of 28 experimentally characterized complexes to develop a multiple linear regression model for oxidation potentials. We achieve good accuracy (mean absolute error of 0.25 V) and preserve transferability to unseen experimental data with a new ligand structure. We also train a KRR model on a subset of 330 structurally characterized heterobimetallics to predict the degree of metal–metal bonding. This KRR model predicts relative metal–metal bond lengths in the test set to within 5%, and analysis of key features reveals the fundamental atomic contributions (e.g., the valence electron configuration) that most strongly influence the behavior of these complexes. Our work provides guidance for rational bimetallic design, suggesting that properties, including the formal shortness ratio, should be transferable from one period to another.

Identifying Underexplored and Untapped Regions in the Chemical Space of Transition Metal Complexes

“…The Cambridge Structural Database 27 (CSD) is a centralized repository of structural data that contains TMCs, enabling quantitative analysis of crystal structures over a broad chemical space. Recent studies have leveraged experimental TMC structures in the CSD to understand spin crossover 28 , redox behavior in bimetallic complexes 29 , oxidation state 30 and charge 2 assignment, and electronic properties. 3 Previously curated subsets of TMCs from the CSD analyzed either a subset of metals (e.g.…”

Stability Trends in disubstituted Cobaltocenium Based on the Analysis of the Machine Learning Models

“…5 Nevertheless, the ML has already made significant contributions to the TM chemistry. 2 For instance, ML has played a crucial role in predicting the HOMO−LUMO gap for TM complexes, uncovering structure−property relationships to aid in the rational design of heterobimetallic TM complexes, 6 predicting relative-energy and total-energy values for organic and TM-containing molecules, 7 and even predicting the line shape of first-row transition metal K-edge X-ray absorption nearedge structure (XANES) spectra 8 such as multiple linear regression, kernel ridge regression, and deep learning techniques. This work is motivated by the experimental interest in the TM-containing cations as components of anion exchange membranes (AEM) in fuel cells, 9−11 because these cations exhibit excellent thermal and alkaline stability under operating conditions, while allowing for high anion mobility, and because the properties of the cobaltocenium−anion complexes can be chemically tuned through the substituent groups on the cyclopentadienyl rings (Cp) of the cation, CoCp 2 + .…”

“…Nevertheless, the ML has already made significant contributions to the TM chemistry . For instance, ML has played a crucial role in predicting the HOMO–LUMO gap for TM complexes, uncovering structure–property relationships to aid in the rational design of heterobimetallic TM complexes, predicting relative-energy and total-energy values for organic and TM-containing molecules, and even predicting the line shape of first-row transition metal K-edge X-ray absorption near-edge structure (XANES) spectra using advanced ML methods, such as multiple linear regression, kernel ridge regression, and deep learning techniques.…”

Identifying Underexplored and Untapped Regions in the Chemical Space of Transition Metal Complexes

“…The Cambridge Structural Database (CSD) is a centralized repository of structural data that contains TMCs, enabling quantitative analysis of crystal structures over a broad chemical space. Recent studies have leveraged experimental TMC structures in the CSD to understand spin crossover, redox behavior in bimetallic complexes, oxidation state and charge assignment, and electronic properties . Previously curated subsets of TMCs from the CSD analyzed either a subset of metals (e.g., cell2mol), complexes that are closed shell in nature (e.g., tmQM), or specific TMCs for spin-crossover applications .…”