Abstract
Pourbaix diagrams are a widespread tool in electrochemistry, enabling a direct understanding of the main changes in phase stability with potential and pH. While originally founded in the combination of theoretical equations and experimental data, in recent years there has been a growing interest in the prediction of this kind of diagrams by means of computational data, fueled by the success of massive databases such as the Materials Project. Herein, we propose a scheme to estimate the uncertainty associated to Pourbaix diagrams, by repurposing the energy recalculation schemes used to build the diagram to determine the population of all competing electrochemical phases. In this way, we may characterize the regions of the diagram (pH and potential ranges) in which the major phase would be more likely to be mispredicted. Furthermore, we propose additional analyses on the computed populations to improve the quantification of these competitive phases and to improve the comparison of different electrochemical systems, modifying not only the target element combination but also the initial composition of mixtures of increasing complexity.

This publication has 30 references indexed in Scilit: