TCE and MP2

[Javier Hernández Rojas]

Hi all,

I am calculating the energy minimum for H3- (singlet) using the MP2 method and the TCE module with MBPT2, and I obtain different energies (?). In the case of MP2 calculation, it gives -1.68312126 (a.u.) and for TCE module the energy is -1.68332105 (a.u.). I am employing, in both cases, the aug-cc-pVTZ basis set, a "threshold" of 1.e-8 and an "eprec"=1.e-7. Thanks in advance.

Regards,

Javier

[Edoardo Aprà]

Could you post input and output files?

[jeff.science]

I see no problems with H3- in a linear geometry.  Attached log has all the details.

SCF      -1.595027922321
Direct:
MP2      -1.665835021695
Semidirect:
MP2      -1.665835021695
CCSD     -1.680596118285250
CCSD(T)  -1.682221142932596

TCE:
MBPT(2)  -1.665835021695284
CCSD     -1.680596109666521
CCSD[T]  -1.682240638301308
CCSD(T)  -1.682221132775974

Jeff

[Javier Hernández Rojas]

Yes, here are.  I have renamed the input files. Thanks.

Javier

[Edoardo Aprà]

Since you are using spherical basis in the MP2 input file and the default cartesian basis in the TCE input file, you  end up using 69 basis functions in MP2 and 75 functions in TCE. This results in discrepancies in all the energies, starting from the Hartree-Fock reference (smaller difference than MP2/MBPT, of course)

$ grep basis h3-MP2.nw h3-tce.nw
h3-MP2.nw:basis spherical
h3-tce.nw:basis

$ grep functions out_MP2 |head -1
  functions       =    69

$ grep functions out_tce |head -1
  functions       =    75

[Javier Hernández Rojas]

Thanks Edoardo. That is the origin of the small difference.

Javier

--
You received this message because you are subscribed to the Google Groups "NWChem Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nwchem-forum...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/nwchem-forum/ca95db51-520a-48f5-92f7-7c8ffd88d4adn%40googlegroups.com.

--

Prof. Javier Hernández-Rojas

Director del Instituto Universitario de

Estudios Avanzados en Física Atómica,

Molecular y Fotónica (IUdEA).

Departamento de Física.

Facultad de Ciencias.

Universidad de La Laguna.

38205, La Laguna, Tenerife

SPAIN, Tlf + 34 922 318255

*************************************************************************

Este mensaje puede contener  información confidencial o legalmente protegida. El acceso a su contenido por personas distintas a las destinatarias no está autorizado. Por otro lado, el tratamiento desautorizado de datos de carácter personal puede suponer una infracción de la normativa vigente en Protección de Datos. 

Si usted ha recibido este mensaje por error, le informamos que la utilización, divulgación o reproducción de su contenido sin autorización está prohibida en virtud de la legislación vigente. En tal caso deberá comunicarlo al remitente y proceder de manera inmediata a su destrucción, así como la de cualquier documento adjunto al mismo.

This message may contain confidential or legally protected information. Access to its content by persons other than the recipients is not authorized. On the other hand, the unauthorized treatment of personal data may imply an infraction of the current regulations on Data Protection.

If you have received this message in error, we inform you that the use, disclosure or reproduction of its content without authorization is prohibited under current legislation. In this case, you must notify the sender and proceed immediately to its destruction, as well as any document attached to it.

[Javier Hernández Rojas]

Hi all,

I am trying to calculate the interaction energy in the

global minimum  (optimization) between He and H3+

using the TCE module and carrying out

CCSD(T) calculations, but I get the error:

"scf: negative no. of electrons ?". Any help?

Thanks in advance. I attach the input and output files.

Regards,

Javier

[Edoardo Aprà]

Your failure shows that the BSSE module does not seem be compatible with numerical gradients (since TCE does use numerical gradients).

A fix could be to use just the BSSE energy correction after having optimized the molecule without BSSE.

[Edoardo Aprà]

Opened a github issue about this bug https://github.com/nwchemgit/nwchem/issues/617

Not likely that a fix is going to surface any time soon.

[Javier Hernández Rojas]

Many thanks for your quick response.

Javier

--
You received this message because you are subscribed to the Google Groups "NWChem Forum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nwchem-forum...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/nwchem-forum/14230c40-bbf9-406a-ab8c-b15b99d4d38cn%40googlegroups.com.