[CP2K-user] M06-2X DFT functional

[Frederick Stein]

Dear Tianhua,

Either use FRACTION 0.54 with SCALE_COULOMB 1.0 (default value for 
SCALE_COULOMB) or FRACTION 1.0 with SCALE_COULOMB 0.54. Both should work. 
The second way is more convenient when you mix interaction potentials. 
Basically you can use every combination of SCALE_COULOMB and FRACTION as 
long as the product of their input values is 0.54.

Best,
Frederick

Am Mittwoch, 16. Oktober 2019 03:44:37 UTC+2 schrieb Tianhua Wang:
>
> Dear Frederick,
> I also want to use the M06-2X functional in CP2K through the first option 
> you provided above, because of the version (5.1) I'm using, and 
> the &XC_FUNCTIONAL section in the input file is written as
> &XC_FUNCTIONAL
>   &LIBXC
>     FUNCTIONAL XC_MGGA_X_M06_2X XC_MGGA_C_M06_2X
>   &END LIBXC
> &END XC_FUNCTIONAL
>
> However, I'm hesitating at the keyword FRACTION in the &HF section. 
> The M06-2X functional has 54% HF, and I wonder if I should write "FRACTION 
> 0.54" in this section as
> &HF
>   FRACTION 0.54
>   &SCREENING
>     EPS_SCHWARZ 1.0E-6
>     SCREEN_ON_INITIAL_P TRUE
>   &END SCREENING
>   &INTERACTION_POTENTIAL
>     POTENTIAL_TYPE TRUNCATED
>     CUTOFF_RADIUS 6.0
>     T_C_G_DATA t_c_g.dat
>   &END INTERACTION_POTENTIAL
>   &MEMORY
>     MAX_MEMORY 5000
>     EPS_STORAGE_SCALING 0.1
>   &END MEMORY
> &END HF
>
> Thank you for your attention to my consultation and looking forward to 
> your reply.
> Sincerely,
> Tianhua Wang
>
>
> 在 2019年9月3日星期二 UTC+8下午8:15:53，Frederick Stein写道：
>>
>> Dear Subhasish,
>>
>> The DFT part of M06-2X should be XC_HYB_MGGA_XC_M06_2X. It is basically 
>> the combination of XC_MGGA_X_M06_2X and XC_MGGA_C_M06_2X. When you want to 
>> use M06-2X within CP2K, you have the following options:
>> 1) Within 6.1.0 use XC_MGGA_X_M06_2X and XC_MGGA_C_M06_2X.
>> 2) Within 6.1.0 adjust the code by either using the patches in 
>> https://groups.google.com/forum/#!searchin/cp2k/wb97x%7Csort:date/cp2k/ARya5uM-pA4/2wt5PgtkAQAJ by 
>> Stanislav or applying the patches from PRs 304 and 422 (
>> https://github.com/cp2k/cp2k/pull/304, 
>> https://github.com/cp2k/cp2k/pull/422) and use XC_HYB_MGGA_XC_M06_2X.
>> 3) Use the latest trunk from CP2K with XC_HYB_MGGA_XC_M06_2X. (The PRs 
>> from 2 have already been included.)
>> Please keep in mind that you need an additional HF section in all of 
>> these three cases.
>>
>> Best regards,
>> Frederick
>>
>> Am Dienstag, 3. September 2019 12:59:46 UTC+2 schrieb SUBHASISH MALLICK:
>>>
>>> I want to use M06-2X DFT functional in conjunction with 6-31G* basis 
>>> sets in cp2k version 6.1.0. In this purpose I used libxc library, where, 
>>> only XC_MGGA_C_M06_2X functional is supported in cp2k version 6.1.0 and 
>>> the  XC_MGGA_X_M06_2X and 
>>> XC_HYB_MGGA_XC_M06_2X are not supported in the current version of cp2k. 
>>> May I ask you among the above three M06-2X functionals mentioned in libxc 
>>> which is equivallent to the gaussian M06-2X and why other two functionals 
>>> are not supported in this version? 
>>> Is there any way to implement the HYB_MGGA functionals in cp2k version 
>>> 6.1.0.
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20191016/5c34282a/attachment.htm>