This thread is locked.Only browsing is available.
Top Page > Browsing
too short distance between atoms warning
Date: 2022/01/18 21:18
Name: Jack

Hi,
I'm trying to run openmx with some models that I built using VESTA, according to a known point group and lattice parameters.
When I try to run some calculations in openMx I get the following message:

"warning: too short distance between 93 89 (Rn=445): r= 0.000000000000"
This is an example for two atoms, but I get this warning for each atom.

What does this message mean and does it affect the calculation?

Thank you in advance
メンテ
Page: [1]

Re: too short distance between atoms warning ( No.1 )
Date: 2022/01/19 18:26
Name: Naoya Yamaguchi

Hi,

It means that a distance of atoms 93 and 89 is too short and 0 Bohr. You need to check if the input atomic coordinates are correct.

Regards,
Naoya Yamaguchi
メンテ
Re: too short distance between atoms warning ( No.2 )
Date: 2022/01/19 22:06
Name: Jack

Hi.
Thank you for the reply. Those are standard input coordinates of the Diamond structure, with all the known parameters.
Could there be another parameter in the input file that affects the distance between atoms?

Thank you
メンテ
Re: too short distance between atoms warning ( No.3 )
Date: 2022/01/20 01:56
Name: Naoya Yamaguchi

Dear Jack,

>Could there be another parameter in the input file that affects the distance between atoms?
There are cell vectors. Distances between atoms are determined from the atomic coordinates and cell vectors.
A case of duplicate atoms is a common problem.

Regards,
Naoya Yamaguchi
メンテ
Re: too short distance between atoms warning ( No.4 )
Date: 2022/01/20 03:37
Name: Jack

Hi,

I am using the following atom coordnates:
1 C 0 0 0 2 2
2 C 0 0 3.567 2 2
3 C 0 3.567 0 2 2
4 C 0 3.567 3.567 2 2
5 C 3.567 0 0 2 2
6 C 3.567 0 3.567 2 2
7 C 3.567 3.567 0 2 2
8 C 3.567 3.567 3.567 2 2
9 C 0 1.7835 1.7835 2 2
10 C 3.567 1.7835 1.7835 2 2
11 C 1.7835 1.7835 0 2 2
12 C 1.7835 1.7835 3.567 2 2
13 C 1.7835 0 1.7835 2 2
14 C 1.7835 3.567 1.7835 2 2
15 C 2.67525 0.89175 2.67525 2 2
16 C 0.89175 0.89175 0.89175 2 2
17 C 0.89175 2.67525 2.67525 2 2
18 C 2.67525 2.67525 0.89175 2 2

and the following cell vectors:
<Atoms.UnitVectors
3.567 0 0
0 3.567 0
0 0 3.567
Atoms.UnitVectors>

are these correct?

Thank you very much
メンテ
Re: too short distance between atoms warning ( No.5 )
Date: 2022/01/20 16:23
Name: Naoya Yamaguchi

Dear Jack,

For example, if the units are ang., atoms 1-8 are duplicate.

Regards,
Naoya Yamaguchi
メンテ
Re: too short distance between atoms warning ( No.6 )
Date: 2022/01/20 18:47
Name: Jack

Hello,

Thank you. I ran another calculation without atoms 1 to 8, and the warning did not appear. When I tried to run unit cell optimization (using RFC5) on the first MD step the warning did not appear, but on the second MD step (and forward) it appeared again.
Other than that, on the 10th MD step, the dUele is 0 in every SCF step, and the calculation gets terminated due to the illegal SCF calculation.
What could be the issue?

Thank you very much
メンテ
Re: too short distance between atoms warning ( No.7 )
Date: 2022/01/20 19:34
Name: Naoya Yamaguchi

Dear Jack,

There remain duplicate ones such as atoms 11 and 12, for example.
In the simple cubic call with the lattice constant of 3.567 ang., there should be 8 atoms for the pristine diamond case.
You need to check your structure carefully.

Regards,
Naoya Yamaguchi
メンテ

Page: [1]