[CPMD-list] RE : Geometry Optimization of NaAlH4

[Belabbas Imad]

  Dear Phani,
 
 As far as know in the crystallographic databases, it only provided the positions of the generating atoms ( in fractional coordinates) in your case, they are three atoms Al, Na and H. Specifying the space group helps to determine the rest of the atoms attached to one node of the crystal. To do you have to apply all the symmetry elements of the space group (in your case: I41/a Z) to the generating atoms. 
 
 I have checked (but you should to double check) that the number of symmetry elements of the your space group is 32 (Tetragonal : I41/a Z) (these symmetry elements can be found in databases). So you have to apply all these symmetry elements to your 3 atoms and you will end up, at most, with 3x32 atoms per each lattice node. However, You have to check the positions of the atoms you will generate by the previous procedure, as it may be that some atoms will be repeated.
 
 One you generate all the atoms of your system you will be able to start your calculations.
 
 It will be also, important for you to fix the space group during the relaxation.
 
 One final remark with respect to your geometry convergence criterion (2e-07), I think it's too much strict !!!. 5e-04 or 1e-03 will be sufficient !! 
 
 hope this will help !!
 
 Cheers,
 Imad
  
             
---------------------------------
 Ne gardez plus qu'une seule adresse mail ! Copiez vos mails vers Yahoo! Mail 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://cpmd.org/pipermail/cpmd-list/attachments/20080110/2be83f3b/attachment.html