Computational insights into aqueous speciation of metal-oxide nanoclusters: an in-depth study of the Keggin phosphomolybdate.

Herein, we present a new computational methodology that unlocks the prediction of the complex multi-species multi-equilibria processes involved in the formation of complex metal-oxo nanoclusters. Relying on our recently introduced method named POMSimulator, we extended its capabilities and challenged its accuracy with the well-known phosphomolybdate [PMoO] Keggin anion system. We show how the use of statistical techniques enabled the processing of a vast number of speciation models and their associated systems of non-linear equations efficiently and in a scalable manner. Subsequently, this approach is applied to generate statistically averaged speciation diagrams and their associated error bars. Then, we unveil the previously unreported speciation phase diagram under varying [Mo]/[P] ratios pH. Our findings align well with experimental data, indicating the prevalence of the Keggin {PMo} as the primary species at low pH, but the lacunary {PMo}and Strandberg {PMo} anions also emerge as major species at other concentration ratios. Finally, from 7 × 10 speciation models we inferred a plausible reaction network across the diverse nuclearities present within the system, which underlines the role of trimers as key intermediate building blocks.