reading wrong number of atoms in surface calculation

[Fernan Saiz]

Dear all,
I am trying to simulate a semi-infinite slab of amorphous boron nitride with a vacuum gap placed along the z axis. Hence, the geometry input that I need to use is a 'surface' type, as you might read in my attached input script. The problem that I face is that atom 293 is missing in the output generated by nwchem. As a consequence, I obtained a charged system with an odd number of electrons. I obtained the same error when using different versions of nwchem: 6.6, 7.0.2, or the one readily available when installing it on ubuntu distributions.

It looks to me that nwchem has here a bug when reading the coordinates. The initial system has 428 atoms vs. 427 particles read by nwchem. Therefore, I would be grateful to hear from your suggestions. I might be missing an obvious input keyword after all.

Best regards,
- Fernan Saiz, PhD
Research Scientist
Physical Science and Engineering Division (PSE)
King Abdullah University of Science and Technology (KAUST)
Thuwal 23955-6900, Saudi Arabia

[Edoardo Aprà]

Are you planning to use NWChem for a calculation with periodic boundary conditions (you do that the cell parameters in your input file)?

If this is the case, the LCAO/DFT code is not ready for periodic boundary conditions.

[Edoardo Aprà]

If you do eliminate the 8 lines corresponding to the system_surface input section, the code will recognized 428 atoms.

However, I am not sure this is what you plan to do. One of the 428 atoms (# 246) seems   far away from the rest of the molecule.

[Fernan Saiz]

Hello,

Thanks for the prompt response. I would like to run a simulation only periodic in the x and y directions to avoid the propagation of the surface dipole error in the electrostatic potential along the z axis. I am not sure if nwchem has the surface dipole correction implemented in the pseudopotential plane wave module.

--
You received this message because you are subscribed to a topic in the Google Groups "NWChem Forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/nwchem-forum/w8gxukV3PjU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to nwchem-forum...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/nwchem-forum/2043cd01-74e6-4eb2-9d70-fa31267803adn%40googlegroups.com.

[Edoardo Aprà]

I believe the slab input is not fully compatible with the pseudopotential plane wave module.

https://nwchemgit.github.io/SYSTEM----Lattice-parameters-for-periodic-systems.html

surface – system with 2-d translational symmetry (not currently available with NWPW module).

[Fernan Saiz]

hi,

If I run the PSPW with the crystal keyword as shown in the attached input script, nwchem reads now 428 atoms. However, I am afraid that surface dipole correction is not available in NWchem. Conversely, if I use the surface keyword, nwchem reads 427 atoms again.

- Fernan

To view this discussion on the web visit https://groups.google.com/d/msgid/nwchem-forum/98c8521d-2303-4bf8-b9d0-d97a4ff58f34n%40googlegroups.com.