[Wien] Radial wave function at Fermi level

[Shin-ya Ayukawa]

Dear Prof. Blaha,

Thank you for your kind reply.

I'm afraid to say it seems like my explanation causes misunderstanding.
"several calculations" doesn't mean "in the middle of a calculation",  
but
"several sessions".  SCF cycles run in each session.

I have tried to optimize the RMT parameters in the hydride.
Then, I set different RMT parameters in several sessions with same other
parameters, for example, space group, lattice parameter, mixing  
factor....

Next, I have set the energy parameter in case.in1 to EF for converged  
sessions
and rerun lapw1, lapw2.

It looks like below

                  RMT     RMT     RMT                    after x lapw1
Session 1 : Sr 2.4   Pd 1.7   H 1.4  --->  O.K.
Session 2 : Sr 2.4   Pd 1.7   H 1.5  --->  O.K.
Session 3 : Sr 2.4   Pd 1.7   H 1.6  --->  'SELECT' - no energy limits  
found for L= 0
Session 4 : Sr 2.4   Pd 1.7   H 1.7  --->  O.K.
Session 5 : Sr 2.4   Pd 1.7   H 1.8  --->  O.K.
Session 6 : Sr 2.4   Pd 1.8   H 1.4  --->  'SELECT' - no energy limits  
found for L= 0
                            .
                            .
                            .

So, do I have to run again SCF cycles using EF for Session 3 and 6 ?
Thanking for your help in advance.

Regards,
Shin-ya Ayukawa

_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/
Shin-ya Ayukawa
Department of Applied Physics
Tohoku University
e-mail : ayu at teion.apph.tohoku.ac.jp
Tel :   +81-22-795-7977
FAX : +81-22-795-7975
_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/

On 2010/03/12, at 22:25, Peter Blaha wrote:

>
>> I have set the energy parameter in case.in1 to EF and rerun lapw1,  
>> lapw2
>> for several calculations using different RMT parameters (other  
>> parameters
>> are same).
>
> Of course you cannot change RMT "in the middle of a calculation".
> (You have to use either clminter or reinitialize with init_lapw).
>
>
>> The results are fine in some calculations.  But errors (no energy  
>> limits found
>> for... and so on) are shown in lapw1 in other calculations.  Then,  
>> I have changed
>> 'STOP' to 'CONT' in case.in1 and rerun lapw1, but gone wrong.
>
> When you want to fix an energy parameter at a certain energy, you  
> MUSTNOT "search",
> i.e. the corresponding line must contain 0.000 for the search like:
> 0   .73062     0.000 CONT 1
>
>> Do I have to run again SCF cycles with using reduced mixing factor  
>> and
>> EF decided by first SCF cycles?
>
> No, of course not. You are NOT interested in the new  
> eigenvalues,.... only
> in the radial wave function at EF.
> You would just change in1, rerun (for one k-point only)  x lapw1 and  
> x lapw2
> and take radwf.
> -- 
>
>                                      P.Blaha
> --------------------------------------------------------------------------
> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
> Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
> Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
> --------------------------------------------------------------------------
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien