Siesta

NEB trajectory request

Asked by rezabma

Dear Nick,
Could you please guide me to have following items by flos NEB. I am not familiar with LUA language.

===================================================================================
1- I need "movie file" for each NEB step. Its name should be "NEB.nnn.xyz" which "nnn" is its neb step. Its structure should be:

N
Energy: A1
SYMB X Y Z
.
.
.
N
Energy: A2
SYMB X Y Z
.
.
.

which N is total number of atoms, A1 is the energy of first image, A2 is energy of the second image, ..., SYMB is chemical sysmble of atoms with X Y Z the coordination of that atom.

This file could be readily read by jmol. This is urgently needed.
====================================================================

2- a summary file named as NEB.MAXF which has following information:

#iteration number maximum force barrier energy
1 M1 B1
2 M2 B2
3 M3 B3
.
.
.
.

which M1 is the maximum force for neb step 1, M2 is for neb step 2, M3 is for neb step3 ,...
B1 is barrier for step 1, B2 is barrier for step 2, ...

====================================================================

3- How could we quench the force and coordination details in neb flos?We need not all details about force components and DR.

=====================================================================

4- How could we have neb input data in fdf file? For example, spring constant, delta t, total time, criteria for climbing, on/off climbing, ....

==================================================================

Thank you
Reza

Question information

Language:
English Edit question
Status:
Solved
For:
Siesta Edit question
Assignee:
No assignee Edit question
Solved by:
rezabma
Solved:
Last query:
Last reply:
Revision history for this message
Nick Papior (nickpapior) said : 		#1

Dear Reza,

I thank you for your suggestions (and also on github).

1) Yes, this would be really nice! However, this requires some restructuring for now.
It would probably be easiest to write a script which manipulates the *.R and NEB.results file. This should be easily manageable using your favourite scripting language.

2) This is basically a post-processing of the NEB.results file. Again, this can be calculated using a simple scripting language.

In terms of 1 and 2) I developed the NEB with help of a colleague which was used to the VASP output from TST. Many of the things were inspired by the outputs of those (as far as I know) and hence it became like that. This also means that I was not the expert in the NEB stuff and implemented stuff with guidance, he then post-processed the results to calculate the required values.

3) If you don't need it, simply delete the files after or while running. You can easily comment out the lines in flos/special/neb.lua that opens the files you don't need, see e.g. line 393 and 516 for where to comment out stuff you don't need (simply put -- in front of the line)

4) The algorithms are handled in Lua, the parameters are handled in Lua, and if one start by adding flags to be read through Lua I think the additional flags would be too immense. In its current state the fdf variables are not hooked that far in the Lua environment, but they could. This is not something I have thought too much about because currently I believe such customizations inherent to the NEB should belong to the environment it was developed in, in this case Lua.

Thanks for all your inputs, they are truly valuable.

Revision history for this message
Nick Papior (nickpapior) said : 		#2

If you wish to help, that would be awesome!
If so, please open your 1) as a ticket on github, then I'll direct you towards the code.

Revision history for this message
rezabma (rezabma) said : 		#3

Dear Nick,
OK, Thank you.
Reza