BaTiO3 Rhombohedral structure - Structure Optimization

Running Abinit Ground state
1

Hello,
Please I really need help. I was trying to figure out the issue but failed miserably.

After carrying out ENCUT (36 Hartree) and KPOINTS grid (12x12x12) convergence tests (<1mHa), I aim at performing structure optimization for BaTiO3 perovskite. However, when running BaTiO3.abi, I got an error:

--- !ERROR
src_file: m_kg.F90
src_line: 181
mpi_rank: 3
message: |
    Choice of acell, ngfft, and ecut
    ===> basis sphere extends BEYOND fft box !
    Recall that boxcut=Gcut(box)/Gcut(sphere)  must be > 1.
    Action: try larger ngfft or smaller ecut.
    Note that ecut=effcut/boxcut**2 and effcut=    6.531650
    This situation might happen when optimizing the cell parameters.
    Your starting geometry might be crazy.
    See https://wiki.abinit.org/doku.php?id=howto:troubleshooting#incorrect_initial_geometry .

I increased ngfft value from 24 24 24 to 60 60 60, but such change did not work.

Could anyone please assist me in this problem?

I appended my ABINIT input file.

BaTiO3.abi (2.2 KB)
error.txt (528 Bytes)

2

can you post or send me the cif file you used to get your starting structure? rhombohedral cases can be tricky.

3

Hi,

One issue that I noted is the input lattice vectors which are 4 Bohr. I guess you intended to specify Angstrom. In ABINIT the default units are atomic units, so in this case,it is necessary to mention the use of Angstrom after acell declaration.

Good luck!
I hope this solves your problem
Bogdan

4

Indeed; after mentioning the use of Angstrom, the calculations went good .
Thank you