Dear Axel and other ATAT users,
I am a new ATAT user. I am facing the following issues in my calculations:
-
I have been trying to calculate the cluster expansion of ZnO-ZnSe alloy in wurtzite structure for quite some time now. 164 structures have been generated so far and 4 ground states have been identified (including pure ZnO and pure ZnSe) but the CV score is not reducing below 0.04. Is the cluster expansion with above mentioned CV score acceptable or should I wait until it is less than 0.025 (as mentioned in the manual).
-
I am using quantum espresso for the relaxation calculations. But I am not able to use the ‘checkrelax’ command. An empty output file is generated. Since, ZnO-ZnSe is a mismatched system therefore I am not expecting the output of ‘checkrelax’ to be empty. Do I have to modify the script for ‘checkrelax’? If so how should I do it? Also how is ‘str_relax.out’ generated?
-
When I run MC simulations on ground state of pure ZnO and ZnSe using a precision for concentration of 0.001 i.e. -dx=1e-3, the simulation gets stalled and no output is generated for any value of temperature and chemical potential. The problem persists when I reduce the precision to 0.01. I have also tried using -eq and -n for different number of MC simulations. On increasing -eq and -n, the average energy does not converge. I have varied -eq from 50 to 1000 and -n from 100 to 2000. Please let me know how to determine the convergence of -eq and -n.
Please find the output files below. Any suggestion and advice would be extremely helpful.
Thanking you.
Regards
Arini Kar
QE parameters:
KPPRA=1300
&control
calculation = ‘vc-relax’ ,
prefix = ‘pwscf’ ,
outdir = ‘./out’ ,
pseudo_dir = 'PseudoDIR/' ,
disk_io = 'none' ,
restart_mode = 'from_scratch',
verbosity = 'high' ,
nstep = 50 ,
/
&system
ibrav = 0 ,
celldm(1) = 1.89 ,
ecutwfc = {CUT_WFC} ,
ecutrho = ${CUT_CHG} ,
occupations = ‘tetrahedra’ ,
/
&electrons
electron_maxstep = 200 ,
conv_thr = 1.0d-7 ,
mixing_mode = ‘plain’ ,
mixing_beta = 0.7 ,
diagonalization = ‘david’ ,
/
Map.log
Maps version 3.36
The internal database of structures extends at least up to 20 atoms/unit cell, see predstr.out
Among structures of known energy, true and predicted ground states agree
No other ground states of 24 atoms/unit cell or less exist.
Concentration range used for ground state checking: [0,1].
Crossvalidation score: 0.0468895