How to set up OPBE in cp2k

[jia... at vt.edu]

Dear Conrad,

I appreciate your quick reply. But I still have some uncertainty.

The paper [1] introduced the GGA_XC_OPBE_D, refitting procedure applied to 
PBE to optimize the (meta-)GGA functional. The prefix ‘o’ stands for 
optimized.
While the OPBE functional is Handy-Cohen’s 2001 OPTX correction for 
exchange plus Perdew-Burke-Ernzerhof’s 1996 correction for correlation. [2]
OPBE is already built into the Gaussian[3] and ADF[4], which shows some 
accuracy in capturing the features of metal complex. One paper [5] reports 
that employing OPBE calculate the Ru complex in CP2K, whereas the OPBE 
haven’t built in CP2K, I am really want to know the setting about OPBE is 
right or not. 


[1] Goerigk, Lars, and Stefan Grimme. “A General Database for Main Group 
Thermochemistry, Kinetics, and Noncovalent Interactions − Assessment of 
Common and Reparameterized (meta-)GGA Density Functionals.” Journal of 
Chemical Theory and Computation 6, no. 1 (January 12, 2010): 107–26. 
doi:10.1021/ct900489g.
[2] Liao, Meng-Sheng, John D. Watts, and Ming-Ju Huang. “Electronic 
Structure of Some Substituted Iron(II) Porphyrins. Are They Intermediate or 
High Spin?” The Journal of Physical Chemistry A 111, no. 26 (July 1, 2007): 
5927–35. doi:10.1021/jp070734z. 
[3] Andrikopoulos, Prokopis C., Carine Michel, Sandra Chouzier, and 
Philippe Sautet. “In Silico Screening of Iron-Oxo Catalysts for CH Bond 
Cleavage.” ACS Catalysis 5, no. 4 (April 3, 2015): 2490–99. 
doi:10.1021/cs500996k.
[4] Swart, Marcel, Andreas W. Ehlers, and * Koop Lammertsma. “Performance 
of the OPBE Exchange-Correlation Functional.” Molecular Physics 102, no. 
23–24 (December 10, 2004): 2467–74. doi:10.1080/0026897042000275017.
[5] Vallés-Pardo, José Luis, Marieke C. Guijt, Marcella Iannuzzi, Khurram 
S. Joya, Huub J. M. de Groot, and Francesco Buda. “Ab Initio Molecular 
Dynamics Study of Water Oxidation Reaction Pathways in Mono-Ru Catalysts.” 
ChemPhysChem 13, no. 1 (January 16, 2012): 140–46. 
doi:10.1002/cphc.201100546.


On Monday, July 18, 2016 at 1:35:50 PM UTC-4, Conrad wrote:
>
> Dear Jianmin,
>
>
> When you want to use an exotic XC functional, it is better to use the 
> LIBXC library.
> All available functionals are shown here: 
> http://www.tddft.org/programs/octopus/wiki/index.php/Libxc_3.0_functionals
> (Assuming you built CP2K with LIBXC version 3.0. Older versions are 
> elsewhere on the website.)
>
> In your input you include the functional like this:
>
>             &XC_FUNCTIONAL
>                 &LIBXC
>                     FUNCTIONAL GGA_XC_OPBE_D
>                 &END LIBXC
>             &END XC_FUNCTIONAL
>
> In this case, as it is a GGA functional there is nothing else to be done. 
> If it was a hybrid functional, for example, you would need also to make a 
> HF section. 
>
> For your pair potential section, set REFERENCE FUNCTIONAL to OPBE. 
>
> I'm not familiar with this functional, but this should be sufficient. Try 
> this and see if it works for you.
> It runs on my machine. Do some benchmarking before doing any production 
> runs, of course. 
>
> Best wishes,
> Conrad
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20160725/7d53db41/attachment.htm>