own-VASP dimer-method vs VTST dimer-method (test)

[perez2013]

hello

I have run a test which consits in trying to reproduce the results of the example system (H impurity diffusion in Ir) provided at the Henkelman (Univ Texas) VTST website/code specifically for the dimer-method transition state search, but instead of using VTST in the calculation, I use own-VASP dimer-method. I mean IBRION=44 as specified in the vasp manual, following and adapting my system to the vasp manual provided example for IDM (ammonia inversion)

Please note that I do not claim my using high accuracy in this test or anything of the sort, I am simply trying to get the test running to the very end. As clearly explained in the manual, I did vibrational analyses, before and after the improved dimer method calculation, in order to get the hessian matrix first, and to check for imaginary frequencies in the second case

Apparently everything looks "alright", what I mean is that at the very end, after the TST calculation is converged, the vibrational analysis shows clearly imaginary frequencies (3 of them). I think this is a transition state, especially when you consider that in the initial state vibrational analysis (optimised), I had not a single imaginary frequency

The reason I am writing is that the VTST example has inputs and also output files, which means you can compare the provided VTST calculation result (without running it yourself) with my own own-VASP dimer result.

Please find enclosed the two CONTCAR files (see below), ie the geometries of the transition states calculated by own-vasp and by vtst. I would like to ask you to have a look, the difference is very clear, unless I am confused. If you concentrate on the H atom, which is what should diffuse mainly, the coordinates are really different (the last line of coordinates)

Something else, the energy barrier I get is more or less 0.02 eV, and the barrier from the OUTCAR in the vtst example is about 0.01 eV

Would you say I have found a different transition state?

regards

Manuel

PS I have included the contcar files as text, sorry I could not find how to attach files

-----------------------------own-VASP CONTCAR------------------------------------------------
Ir H
3.84347853250000
1.0000000000000000 0.0000000000000000 0.0000000000000000
0.0000000000000000 1.0000000000000000 0.0000000000000000
0.0000000000000000 0.0000000000000000 1.0000000000000000
Ir H
4 1
Selective dynamics
Direct
0.1250000000000000 0.1250000000000000 0.1250000000000000 F F F
0.1237896947429418 0.6270542442143225 0.6270567424333305 T T T
0.6270514331274946 0.1237860505624337 0.6270607822618688 T T T
0.6270528579012905 0.6270591185148645 0.1237885910683551 T T T
0.6109785802023180 0.6109355199195470 0.6109365564276165 T T T

0.00000000E+00 0.00000000E+00 0.00000000E+00
-0.57536085E-01 0.39122599E-01 0.39126289E-01
0.39178420E-01 -0.57590607E-01 0.39171445E-01
0.39180770E-01 0.39169172E-01 -0.57587042E-01
-0.57187633E+00 -0.57175819E+00 -0.57175741E+00
-----------------------------------------------------------------------------------------------------
--------------------------vtst CONTCAR provided-----------------------------------------------
Ir H
3.843478532500000
1.0000000000000000 0.0000000000000000 0.0000000000000000
0.0000000000000000 1.0000000000000000 0.0000000000000000
0.0000000000000000 0.0000000000000000 1.0000000000000000
4 1
Selective dynamics
Direct
0.1249999999999998 0.1249999999999998 0.1249999999999998 F F F
0.1103244031242872 0.6324325493749443 0.6324337785600408 T T T
0.6324290628192405 0.1103233518612962 0.6324312532324496 T T T
0.6324273785156075 0.6324303626390698 0.1103230182224636 T T T
0.4631022385830526 0.4631031148914402 0.4631011306934644 T T T

NaN NaN NaN
NaN NaN NaN
NaN NaN NaN
NaN NaN NaN
NaN NaN NaN
-------------------------------------------------------------------------------------------------

[support_vasp]

Hi,

We're sorry that we didn’t answer your question. This does not live up to the quality of support that we aim to provide. The team has since expanded. If we can still help with your problem, please ask again in a new post, linking to this one, and we will answer as quickly as possible.

Best wishes,

VASP