[Pw_forum] combining vdW-DF and hybrids?
Jose C. Conesa
jcconesa at icp.csic.es
Tue Feb 15 16:57:59 CET 2011
Thanks for the suggestion, but it does not help me. I'm not concerned
only with molecules, but with oxide layer compounds and surfaces as
well; and for these the standard Grimme's parameters are not adequate
(see J Phys Chem C _114_(2010)22718). For this I should like to try
hse06 and vdW-DF together, but it's not clear if Q-E allows it.
El 15/02/2011 6:06, Giuseppe Mattioli escribió:
> I would only add a perhaps useless suggestion.
> If you are dealing with molecules and your main interest is in using the HSE06
> functional, then try to use the london=.true. keyword. The semiempirical
> london correction behaves well with molecules and could fit to your system.
> On Saturday 12 February 2011 13:46:26 CONESA CEGARRA, JOSE CARLOS wrote:
>> I am doing some tests with the cvs version of Q-E. I wonder, could one
>> combine the HSE06 functional (that refers only to exchange) with the vdw-DF
>> correlation? I tried a line such as input_dft='nox+pw+hse+vdw'
>> but got a message of incompatibility of functionals.
>> Also, some question to the developers:
>> a) is there work for allowing soon the use of hybrid functionals with USPP
>> or PAW potentials?
>> b) will the vdW-DF2 functional be available (with both ions and cell
>> relaxation, if possible) in the 4.3 version?
>> All the best,
José C. Conesa
Instituto de Catálisis y Petroleoquímica, CSIC
Marie Curie 2, Cantoblanco
28049 Madrid, Spain
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the users