[Wien] Error for sphere overlapping

[Gavin Abo]

Yes, you can try decreasing the RMT, where it is recommended to use 
setrmt (terminal) or "set automatically RMT and continue editing" 
(w2web) [ 
http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg07996.html 
, http://www.wien2k.at/reg_user/faq/rmt.html ].

Or you might want to upgrade from WIEN2k 13.1 to 14.2 in order to use 
reduce_rmt_lapw that was released in WIEN2k 14.1.  As it says on the 
webpage at http://www.wien2k.at/reg_user/updates/ :

*reduce_rmt_lapw*: (should be called when a structure optimization stops 
due to overlapping spheres. It will save the current calc. as 
case_old_rmt_XX, reduce sphere sizes by a certain amount (3 % or -r XX) 
and extrapolate densities on the new radial mesh, so that afterwards you 
can easily continue with the minimization)

However, if any of the atomic positions have moved too close to each 
other, decreasing the RMT might not work and the atomic positions would 
have to be adjusted so that the atoms do not overlap.  Check the failing 
struct file with xcrysden and/or "x nn", these programs can help you see 
if the atomic positions are too close.

I assume the Li positions are constrained in the calculation as you 
mentioned before [ 
http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg12403.html ]. 
I haven't tried constraining positions during optimization.  So I don't 
know, but maybe constraining positions can also lead to such a problem.

Also, your RMT of Li looks like it might be too small.  From what I have 
seen in literature, the RMT of Li in LiFePO4 is around 1.97 a.u. [ J. 
Phys.: Condens. Matter vol. 22 p. 275501 (2010): 
http://dx.doi.org/10.1088/0953-8984/22/27/275501 ].

On 5/28/2015 4:39 AM, B Tankhilsaihan wrote:
> Hello Dear Wien2k users
> I am running calculation of minimization on Wien2k_13.01.
> I faced with error that caused by sphere overlapping due to atomic 
> movement.
> My structure is LiFePO4 and I chose Rmt as following
> Li-0.88
> Fe-2.0
> P-1.66
> O-1.21
> Then how can i solve this error?
> Can I decrease these values more?
>
>
> Thanks in advance
>
> Regards
> Tankhilaa
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20150528/0f888d18/attachment.html>