Unveiling the thermodynamic driving forces for high entropy alloys formation through big data ab initio analysis

The fundamental thermodynamic driving forces beyond the existence of high entropy alloys (HEAs) are still not firmly understood. Here, using thermodynamic modeling combining ab initio computations with a regular solution model, we build a database of more than 100,000 BCC and FCC equimolar alloys formed using 27 common elements. We statistically study how enthalpic and entropic contributions evolve with the number of elements in a random solid solution. The commonly admitted rationalization of a stabilization of HEAs due to a growing importance of the entropy with the number of elements is somewhat contradicted. Entropic and enthalpic contributions favor mixing in average, but both driving forces weaken as the number of elements in the alloy increases. By adding binary intermetallics to our analysis, we conclude that the specific chemical compositions prone to form single phase HEAs need to combine an enthalpically favorable mixing of their elements on a given lattice with the absence of strongly competing intermetallics.