Phonon calculations for gas molecules (O2, NH3) using finite displacement: is a large box sufficient without supercell?

[yuanchen_gao]

Hello everyone,

I recently started performing DFT calculations on gas-phase molecules. At the moment, I need to compute phonons for gas molecules (O₂ and NH₃), and I plan to use the finite displacement method. The POSCAR for the gas molecule is constructed following the VASP example (with a larger cell parameter), for instance:

Code: Select all

O2 atom in a box
1.0
15.0   0.0    0.0
0.0    15.0   0.0
0.0    0.0    15.0
O
2
cart
0 0 0
0 0 1.21

I wonder if it is ok to perform phonon calculation based on this large unit cell, without creating any more supercell?

Thank you very much.

[alexey.tal]

Dear yuanchen_gao,

Since you are simulating a non-periodic system, you don't need to create a supercell, but the box should be large enough to avoid the interactions across the periodic boundary conditions. Similar logic applies if you do a regular SCF calculation for a molecule in a box, you don't need to include the k-points.

Best regards,
Alexey

[yuanchen_gao]

Thank you very much, Alexey!