Dear All

I am simulating CuZr Metallic glass and to simulate i am using the command lines mentioned below,

fix 1 all ave/time 100 1 100 c_myRDF file Cu46Zr54_melt.rdf mode vector

fix 2 all npt temp 300.0 2500.0 0.2 iso 0.0 0.0 0.2
run 200000
fix 2 all npt temp 2500.0 2500.0 0.2 iso 0.0 0.0 0.2
run 100000
fix 2 all npt temp 2500.0 300 0.2 iso 0.0 0.0 0.2
run 200000

To check the melting i am using volume-temperature curve, and got MP around 1060 K, while reported is 1215 K from MD, 1207K from experiment and from CALPHAD it is 1198K all for B2 structure, what possibly the reason ?

I saw one of the old post in which steve mentioned the possible reason due to system size effect,PBC and nucleation for Mg-Al system, however i have tried it with bigger system size and the result is not appreciably changing, PBC and nucleation i think is ok because for one different system i.e.Cu62Zr38 results seems to be convincing.

Thanks in advance

Anik shrivastava
Senior Research Fellow
Naval Materials Research Lab,DRDO

I think you are asking why a simulation of the melting point
could be 10-15% different than expt. There are any number

of reasons. Potentials, the difficulty of getting an exact

melting point with MD (there is a big literature on that), expt error,