IBRION=3 bulk optimization: Difference between revisions

From Wiki
Jump to navigation Jump to search
Fdattila (talk | contribs)
No edit summary
Fdattila (talk | contribs)
No edit summary
Line 21: Line 21:
   -31.491683333336404      -29.378686935390292      -14.997850470930652      -9.6057143126896563      -5.4292953596452289      -1.6881155935524064      -1.3100285123688000      -1.1825456664410310      -0.59115181564548613  
   -31.491683333336404      -29.378686935390292      -14.997850470930652      -9.6057143126896563      -5.4292953596452289      -1.6881155935524064      -1.3100285123688000      -1.1825456664410310      -0.59115181564548613  


3.
3. Taken the highest eigenvalue, the proper POTIM is given, in this case, by
 
1/31.491683333336404*0.7=0.0222280909
 
where 0.7 is set as a ''control parameter''.

Revision as of 11:10, 23 October 2018

To choose a proper value for the POTIM in case of a IBRION=3 bulk optimization

1. You first need to perform a frequency calculation, setting in the INCAR file

frequencies:
 IBRION =   5     #     
 EDIFF  =  1E-7   # 
 POTIM  =   0.010 #                  
 NFREE  =   2     # 

PLEASE NOTE: just the degrees of freedom of the investigated bonds must be relaxed.

2. When this calculation converge, based on the POSCAR and OUTCAR of the frequencies calculation, you need to execute

getdimer POSCAR OUTCAR POSCAR

which will generate a eigenvectors.dat file, presenting the following format

Info:            0
Eigenvalues
 -31.491683333336404       -29.378686935390292       -14.997850470930652       -9.6057143126896563       -5.4292953596452289       -1.6881155935524064       -1.3100285123688000       -1.1825456664410310      -0.59115181564548613 

3. Taken the highest eigenvalue, the proper POTIM is given, in this case, by

1/31.491683333336404*0.7=0.0222280909

where 0.7 is set as a control parameter.