", " ", - " The descriptor also generalizes to double perovskites - compounds with substitutions at the , , or sites. Below you can explore the stability of compounds with 50/50 mixtures of ions on each or all of the sites - i.e., formulas.

", + " The descriptor also generalizes to double perovskites - compounds with substitutions at the , , or sites. Below you can explore the stability of compounds with 50/50 mixtures of ions on each or all of the sites - i.e., formulas.

", " ", " More details on the identification and application of are available within the manuscript and associated github repository.", "", @@ -1677,11 +1677,11 @@ "result": { "type": "BeakerDisplay", "innertype": "Html", - "object": "\n\n\n is a descriptor that takes as input the chemical composition and outputs a prediction of perovskite stability according to the formula:\n\n

\n \n where is the ionic radius of ion, and is the oxidation state of ion, , and indicates stability in the perovskite structure.

\n \n

This descriptor was identified by applying the SISSO algorithm developed by R. Ouyang, S. Curtarolo, E. Ahmetick, M. Scheffler, L. Ghiringhelli: Phys. Rev. Materials 2, 083802 (2018) [PDF] {Code} which efficiently identifies from a space of ~3,000,000,000 potential descriptors.

\n \n requires the same information as Goldschmidt's famous tolerance factor (noting that is an implict function of ):

\n \n While the functional forms are comparable, the accuracies are not. On a set of 576 compounds characterized experimentally at ambient conditions, achieves 92% accuracy in predicting whether the compound will or won't be stable as perovskite compared with 74% using .

\n \n is probabilistic, providing not only whether a given composition will crystallize as perovskite but also a probability on this prediction. \n \n Below you can input two cations - and - and one anion - - and the utility will automatically assign oxidation states and radii to each ion (more on this below) and provide the probability that the will form the perovskite structure. is also provided for context.

\n \n This result is visualized with respect to the cationic radii to show where in the space of stable and unstable perovskites the given composition sits.

\n \n The descriptor also generalizes to double perovskites - compounds with substitutions at the , , or sites. Below you can explore the stability of compounds with 50/50 mixtures of ions on each or all of the sites - i.e., formulas.

\n \n More details on the identification and application of are available within the manuscript and associated github repository.\n\n

\n \n where is the ionic radius of ion, and is the oxidation state of ion, , and indicates stability in the perovskite structure.

\n \n

This descriptor was identified by applying the SISSO algorithm developed by R. Ouyang, S. Curtarolo, E. Ahmetick, M. Scheffler, L. Ghiringhelli: Phys. Rev. Materials 2, 083802 (2018) [PDF] {Code} which efficiently identifies from a space of ~3,000,000,000 potential descriptors.

\n \n requires the same information as Goldschmidt's famous tolerance factor (noting that is an implict function of ):

\n \n While the functional forms are comparable, the accuracies are not. On a set of 576 compounds characterized experimentally at ambient conditions, achieves 92% accuracy in predicting whether the compound will or won't be stable as perovskite compared with 74% using .

\n \n is probabilistic, providing not only whether a given composition will crystallize as perovskite but also a probability on this prediction. \n \n Below you can input two cations - and - and one anion - - and the utility will automatically assign oxidation states and radii to each ion (more on this below) and provide the probability that the will form the perovskite structure. is also provided for context.

\n \n This result is visualized with respect to the cationic radii to show where in the space of stable and unstable perovskites the given composition sits.

\n \n The descriptor also generalizes to double perovskites - compounds with substitutions at the , , or sites. Below you can explore the stability of compounds with 50/50 mixtures of ions on each or all of the sites - i.e., formulas.

\n \n More details on the identification and application of are available within the manuscript and associated github repository.\n\n