neb : question about the number of scf optimizations for each band step for each replica

[afloris]

Dear cp2k users,
 
I am running for the first time some NEB calculations...and I am using cp2k 
since few days only.

>From the output of a particular replica, e.g.  xxx-BAND2.out,  
(xxx being the name of the project, 2 is the replica), I see that for each 
band step, several scf optimizations are performed for each replica.
I have some questions: 

1. why  more than one scf optimization is performed? (every optimization 
being achieved with ~40 scf steps) 
2. Are these purely electronic optimizations, or there is a geometrical 
minimization involved?
(this question is motivated because different forces are found after each 
scf optimization...although no "new" atomic positions are printed...) 

Can someone clarify this to me?

Thank you very much 
Kind Regards
Andrea


-- 
 Dr Andrea Floris
 Research Associate
 King's College London
 Strand, London WC2R 2LS
 United Kingdom
 Phone: +44 (0) 207 848 2064
 Fax    : +44 (0) 207 848 2420
 Location:  Strand Building, 4th floor, Room 4.02 
 Emails: andrea... at kcl.ac.uk, an.f... at gmail.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20130517/3260ea56/attachment.htm>