Chemical Potential values for emc2

Alloy Theoretic Automated Toolkit (ATAT)
1

Dear Dr. Axel and ATAT users,

I need some help trying to understand the values of chemical potential that we input for the MC runs while using emc2. I ran the MAPS program for Ti-Al system and I am able to get the convex hull correctly for Ti-Al. Now I create a new folder where I copy my lat.in, gs_str.out, eci.out and clusters.out files. If I select a ground state, say 1, shouldn’t I be specifying the chemical potential range according to the convex hull plot I got? For example, mu0=-0.4 eV/atom and mu1=-0.5 eV/atom if my formation energy/atom value for this ground state (here no. 1) lies between those values.

In the manual, it says that we specify lower (e.g. 0.5) and upper limits (e.g. 1.5) such that the ground state here ‘1’ lies in between. I am having some trouble understanding the reason for this? Could you please provide some help?

Regards,
Aakash

2

The reason is that it is a bit difficult to calculate the chemical potential that stabilizes a given ground state, so the program gives you a way to specify it in a dimensionless fashion where you only need to know which ground state number you want.

I should emphasize that the integral values of the dimensionless chem pot correspond to 2-phase equilibria, so
-mu0=1 stabilizes a 2-phase equilibria between gs=0 and gs=1 at 0K.
-mu0=2 stabilizes a 2-phase equilibria between gs=1 and gs=2 at 0K.
Any value between 1 and 2 will stabilize gs=1 at 0K. So typically if you are interested in gs 1, you would pick -mu0=1.5 -mu1=2 to study one composition side of the phase region
and -mu0=1.5 -mu1=1 for the other.

If you want to by pass that convention and specify chem pot in eV/atom, then add the option -abs on the command line.