all 0.0 forces from nebbarrier.pl
Posted: Thu Feb 18, 2016 7:31 pm
Hello,
I am running VASP 5.3.5 implemented with the VTST tools on NERSC and when I run the current version of nebbarrier.pl (downloaded yesterday from http://theory.cm.utexas.edu/vtsttools/scripts.html) I get all zero forces for the intermediate images in the neb.dat file:
0 0.000000 0.000000 -0.033023 0
1 1.885349 0.565354 0.000000 1
2 3.784811 0.306657 0.000000 2
3 5.576834 1.287476 0.000000 3
4 7.285078 0.084631 0.000000 4
5 8.857768 -0.286671 0.025546 5
I have traced this back to a difference in the OUTCAR file from NEBs I used to run with VASP 4.6.35 that had NEB sections looking like this:
--------------------------------------------------------------------------------------------------------
NEB: the previous image is higher in energy: F
NEB: the next image is higher in energy : T
NEB: only next energy greater
NEB: Tangent
----------------------------------------------
... (values for each atom in geometry) ...
NEB: forces: par spring, perp REAL, dneb 0.142174 0.267327 0.000000
NEB: distance to prev, next image, angle between 0.373571 0.402006 152.397069
NEB: projections on to tangent (spring, REAL) 0.142174 -0.376604
FORCES: max atom, RMS 0.100471 0.026954
FORCE total and by dimension 0.201706 0.098389
to the NEB section of the current OUTCARs that look like this:
--------------------------------------------------------------------------------------------------------
NEB: the previous image is higher in energy: F
NEB: the next image is higher in energy : F
NEB: image is at an extrema
NEB: diff energy (min, max): 2.013657 2.235832
NEB: Tangent
----------------------------------------------
... (values for each atom in geometry) ...
NEB: forces: par spring, perp REAL, dneb 1.373534 2.107970 0.000000
NEB: distance to prev, next image, angle between 1.776348 1.501642 112.783846
-1.373534 -1.133137
stress matrix after NEB project (eV)
0.00000 0.00000 0.00000
0.00000 0.00000 0.00000
0.00000 0.00000 0.00000
FORCES: max atom, RMS 1.600415 0.281295
FORCE total and by dimension 2.515975 1.374603
Stress total and by dimension 0.000000 0.000000
I see that the nebbarrier.pl script greps "NEB: projections" and reports the 2nd number as the force for each image, but the current OUTCAR does not have the NEB: projections line, therefore the forces in neb.dat are 0.000000. I am trying to figure out if this is (a) an issue with the incorporation of the VTST tools in VASP 5.3.5 on NERSC or (b) an intentional change in the VTST code that was not updated in the nebbarrier.pl script. By drawing a parallel between the two OUTCARs, the 2nd number in the line after the "NEB: distance ..." line looks promising for the image force, but I am hoping someone can speak definitively to this issue?
Thank you!
I am running VASP 5.3.5 implemented with the VTST tools on NERSC and when I run the current version of nebbarrier.pl (downloaded yesterday from http://theory.cm.utexas.edu/vtsttools/scripts.html) I get all zero forces for the intermediate images in the neb.dat file:
0 0.000000 0.000000 -0.033023 0
1 1.885349 0.565354 0.000000 1
2 3.784811 0.306657 0.000000 2
3 5.576834 1.287476 0.000000 3
4 7.285078 0.084631 0.000000 4
5 8.857768 -0.286671 0.025546 5
I have traced this back to a difference in the OUTCAR file from NEBs I used to run with VASP 4.6.35 that had NEB sections looking like this:
--------------------------------------------------------------------------------------------------------
NEB: the previous image is higher in energy: F
NEB: the next image is higher in energy : T
NEB: only next energy greater
NEB: Tangent
----------------------------------------------
... (values for each atom in geometry) ...
NEB: forces: par spring, perp REAL, dneb 0.142174 0.267327 0.000000
NEB: distance to prev, next image, angle between 0.373571 0.402006 152.397069
NEB: projections on to tangent (spring, REAL) 0.142174 -0.376604
FORCES: max atom, RMS 0.100471 0.026954
FORCE total and by dimension 0.201706 0.098389
to the NEB section of the current OUTCARs that look like this:
--------------------------------------------------------------------------------------------------------
NEB: the previous image is higher in energy: F
NEB: the next image is higher in energy : F
NEB: image is at an extrema
NEB: diff energy (min, max): 2.013657 2.235832
NEB: Tangent
----------------------------------------------
... (values for each atom in geometry) ...
NEB: forces: par spring, perp REAL, dneb 1.373534 2.107970 0.000000
NEB: distance to prev, next image, angle between 1.776348 1.501642 112.783846
-1.373534 -1.133137
stress matrix after NEB project (eV)
0.00000 0.00000 0.00000
0.00000 0.00000 0.00000
0.00000 0.00000 0.00000
FORCES: max atom, RMS 1.600415 0.281295
FORCE total and by dimension 2.515975 1.374603
Stress total and by dimension 0.000000 0.000000
I see that the nebbarrier.pl script greps "NEB: projections" and reports the 2nd number as the force for each image, but the current OUTCAR does not have the NEB: projections line, therefore the forces in neb.dat are 0.000000. I am trying to figure out if this is (a) an issue with the incorporation of the VTST tools in VASP 5.3.5 on NERSC or (b) an intentional change in the VTST code that was not updated in the nebbarrier.pl script. By drawing a parallel between the two OUTCARs, the 2nd number in the line after the "NEB: distance ..." line looks promising for the image force, but I am hoping someone can speak definitively to this issue?
Thank you!