[Wien] Impurity sites in supercell

[Stefaan Cottenier]

OK, I see better now what your problem is. It can be done in an easier way:

1) x supercell
2) in the newly made supercell that is created, split *only* the block 
with N atoms that contains the position at which you want to put the 
impurity into a block of N-1 atoms and a block with the impurity. Of 
course, change the MULT values for these two blocks, and increase the 
number of inequivalent positions at the top of case.struct by 1. 
(perhaps at this stage you might want to remove all labels, but it might 
not be necessary)
3) x sgroup

Stefaan

>I think this is a general issue related to multiplicity. Assuming atom
>A has N multiplicity with coordinate (x,y,z) in the original cell, there
>will be L A atoms in the LxLxL supercell and each has N multiplicity with
>coordinates (x+n1*a,y+n2*b,z+n3*c), where a,b and c are the original
>lattice parameters and n1,n2 and n3 are integers less than L. In the
>StructGen template, only the first one among the N positions can be edited
>and all others are changed automatically be symmetry operations. So a
>localized defect cluster is created in stead of a single defect. This will
>happen as long as atom A has a multiplicity greater than 1.
>What I am doing now is first to build a P type cell and put a label on
>every position so that multiplicity is 1. Then make change on the position
>as I want. Finally, remove all the labels on other positions and run
>init_lapw to let Wien2k to find out the remaining symmetry. My case
>contains 28 positions, it is a kind of pain to build the cell. For more
>complex cases, this way is not very practical.
>  
>


Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm