About NEB calculations

[alok_shukla1]

Dear Users and Experts,
I am trying NEB calculations. I created 4 images in between initial and final relaxed structure and I am getting very high energy values.
00 -395.94910849 = 0
01 37.30329488 = 37.30329488+395.94910849 = 433.25240337
02 1018.48104960 = 1018.48104960+395.94910849 = 1414.43015809
03 1001.09925193 = 1001.09925193+395.94910849 = 1397.04836042
04 37.79655317 = 37.79655317+395.94910849 = 433.74566166
05 -395.95035434 = -395.95035434+395.94910849 = -0.0012458499999752348

Can anyone please suggest me what could be the possible reason of getting such high energies. Or, what can I do in this case. I have seen in the literature the energy values in few eVs, but in my case such a high energy values I am getting. Here I am attaching my structure files.


with warm regards,
Poonam S.

[henrique_miranda]

I would suggest that you visualize the intermediate structures with a program like VESTA (https://jp-minerals.org/vesta/en/) or xcrysden (http://www.xcrysden.org/) to see if they are like you would expect.
I think the problem in your case is that the atoms in the intermediate positions are overlapping other atoms in the structure.
You need to be careful when setting up these structures.

[alok_shukla1]

Dear Experts,
Yes, this is correct. How can I form these intermediate structures correctly. Please suggest about this.

[henrique_miranda]

I think the problem has to do with what atom you moved between the initial and final structure.
You need to choose another one. A visualization tool will help you to choose the atom correctly.
Without knowing exactly how you choose which atoms to move it is difficult for me to help you further.

[alok_shukla1]

Dear Expert,
In the attached Zip file, 00 and 06 are the initial and final relaxed structures where I did the Pd vacancy in the structure. I am willing to study the migration behaviour of these vacancies (how much energy barrier they cost) from one position to another. For this purpose, I created Pd vacancies at two nearby positions. I am also new to these calculations. Please let me know if I am not mentioning here some needed information.

[henrique_miranda]

I understand that 00/POSCAR and 05/POSCAR are the initial and final structures in your files.
The problem with these structures is that you are not 'moving' an atom between the initial and final structure, instead, you remove two different atoms which lead to the wrong trajectory where atoms are overlapping.

Try to generate your initial and final structures using the following procedure:
1. Remove one atom from the perfect supercell and copy its position (r_atom). This is your initial structure
2. Copy the initial structure to the final structure and change the position of the atom you want to migrate to the vacancy, for example, atom 6 replace its position by r_atom.

That should lead to the expected trajectory but you should visualize it before running the calculations.
I hope you understand that I cannot generate these files for you.

[alok_shukla1]

Dear Expert,

Thank you very much for your response.
Yes, I solved my problem with your suggestion.
Thanks once again.