Determination of bonding radii from small-molecule crystal structures

TL;DR

This paper introduces an unsupervised method to derive bonding radii from crystal structure data, improving the accuracy of identifying bonded atom pairs in small-molecule crystals.

Contribution

It presents a novel unsupervised workflow for determining bonding radii directly from crystal structures, enhancing bond identification accuracy.

Findings

Successfully derived bonding radii from the Crystallography Open Database

Improved bond detection compared to heuristic methods

Provides a data-driven approach for chemical bonding analysis

Abstract

X-ray crystallography rarely captures chemical bonding between atoms of a structure in question. Most of the time distance-based heuristics are applied to establish the pairs of bonded atoms. One class of such heuristics depends on a set of bonding radii that estimate the idealised size of each chemical element in a bond. This publication describes an unsupervised workflow for deriving a bonding radii set from crystal structure data in the Crystallography Open Database.