Amending CALPHAD databases using a neural network for predicting mixing enthalpy of liquids Journal: Computational Materials Science

Clement Vincely, Amin Reiners-Sakic, Vedant Dave, Erwin Povoden-Karadeniz, Elmar Rueckert, Ronald Schnitzer, David Holec · 2025

Read Amending CALPHAD databases using a neural network for predicting mixing enthalpy of liquids Journal: Computational Materials Science by Clement Vincely, Amin Reiners-Sakic, Vedant Dave, Erwin Povoden-Karadeniz, Elmar Rueckert, Ronald Schnitzer, David Holec (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-18366.txt.

Description

Clement Vincely and co-authors teach a neural network to fill gaps in the thermodynamic maps used to design alloys in this 2025 paper. The CALPHAD method predicts a material's stability from stored data on simpler mixtures. Many binary liquid pairs have never been measured, and testing each one is slow.

The network learned the mixing enthalpy, the heat tied to blending two liquid metals, from a steel database of ninety-one binary systems plus fresh experimental reports. Tested on pairs it had never seen, it predicted well once some data on the pair's elements reached training. Its uncertainty stayed under one kilojoule per mole. The team then predicted every missing pair in the database and converted the answers into the standard parameters the files use. The authors offer the patch as a way to spare large experimental effort.

Catalog record

CREATORClement Vincely, Amin Reiners-Sakic, Vedant Dave, Erwin Povoden-Karadeniz, Elmar Rueckert, Ronald Schnitzer, David Holec
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE25 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-18366

Browse the full library · About the Tower of Babel