There is a variable in the &ions namelist called first_last_opt that if set .true. makes the minimization of the first and last images along with the NEB optimization... I think that usually it is more convenient to optimize these two images separately and then keep them fixed because the NEB optimization itself is not very efficient, but if you already have a rather good estimate of them it may be convenient.
stefano Nay Lin wrote: > Dear all, > > I am using NEB to calculate the adatom diffusion barrier on metallic > surface. > > Do I need to relax the first and last image in separate calculations before > writing the input file for NEB calculation with intermediate images? Or can > be done automatically in single NEB file? What is the criteria for > convergence? (i.e. is it the third column of output file (image > energy(eV) *error(eV/A)* fronze) ? > > regards > Richard Lin > > > ------------------------------------------------------------------------ > > _______________________________________________ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum >
