Computer simulation of defect clusters in UO2 and their dependence on composition

Nicholas A. Brincat, M. Molinari, Stephen C. Parker, G. C. Allen, Mark T. Storr

Research output: Contribution to journalArticlepeer-review

34 Citations (Scopus)


It is recognised that point defects play a key role in the behaviour and properties of many technologically significant oxides. What is less well understood is how these defects cluster together and, crucially, the extent to which the clusters change with composition. We chose to investigate this phenomenon by considering UO2, a nuclear fuel material for which there is contradictory data in the literature concerning defect clustering as a function of oxygen content. Early studies of fluorite UO2+x proposed a model based on 2:2:2 Willis clusters whilst more recent research suggests cuboctahedral or split quad-interstitial defect clustering. Here we use the PBE + U functional to simulate defective UO2+x and find for 0.125 < x < 0.25, chains of edge-sharing 2:2:2 Willis clusters to be most stable. Below x = 0.125 these chains destabilise, transforming in to split di-interstitial clusters, demonstrating that the type of oxygen cluster present is dependent on local environment and stoichiometry.

Original languageEnglish
Pages (from-to)329-333
Number of pages5
JournalJournal of Nuclear Materials
Early online date13 Oct 2014
Publication statusPublished - 1 Jan 2015
Externally publishedYes


Dive into the research topics of 'Computer simulation of defect clusters in UO2 and their dependence on composition'. Together they form a unique fingerprint.

Cite this