ICODE-ADC/lammps
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

removed a bug in fix_neb.cpp which prevented the freeend to work properly, plus added an example for the neb freeend

Browse Source
This commit is contained in:
Emile Maras
2017-05-29 15:49:04 +02:00
parent 32c87f3131
commit c904cfb8bc
2 changed files with 7 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
examples/neb/README
View File

@ -2,13 +2,19 @@ Run these examples as:
mpirun -np 4 lmp_g++ -partition 4x1 -in in.neb.hop1
mpirun -np 4 lmp_g++ -partition 4x1 -in in.neb.hop2
mpirun -np 4 lmp_g++ -partition 4x1 -in in.neb.hop1freeend
mpirun -np 3 lmp_g++ -partition 3x1 -in in.neb.sivac
mpirun -np 8 lmp_g++ -partition 4x2 -in in.neb.hop1
mpirun -np 8 lmp_g++ -partition 4x2 -in in.neb.hop2
mpirun -np 8 lmp_g++ -partition 4x2 -in in.neb.hop1freeend
mpirun -np 6 lmp_g++ -partition 3x2 -in in.neb.sivac
mpirun -np 9 lmp_g++ -partition 3x3 -in in.neb.sivac
Note that more than 4 replicas should be used for a precise estimate
of the activation energy corresponding to a transition.
If you uncomment the dump command lines in the input scripts, you can
create dump files to do visualization from via Python tools: (see
lammps/tools/README and lammps/tools/python/README for more info on

2
src/REPLICA/fix_neb.cpp
View File

@ -461,7 +461,7 @@ void FixNEB::min_post_force(int vflag)
if (ireplica < nreplica-1)
dotgrad = dotgrad /(gradlen*gradnextlen);
dot = 0.0;
if (FreeEndIni && ireplica == 0) {
if (tlen > 0.0) {
double dotall;