Page 2 of 2

Re: HSE06 with yambo

Posted: Sat Jun 19, 2021 2:35 pm
by haseebphysics1
Dear Dr. Daniele,

The scf in and out files are attached.

Here I just changed the functional type to HSE and did not introduce any hybrid parameters (to tackle divergence problem) and vacuum thing.

The Scalar PPs were taken from PseudoDojo.


Thanks,

Re: HSE06 with yambo

Posted: Sat Jun 19, 2021 5:30 pm
by Daniele Varsano
Dear Haseeb,
we will take a look.
Just for your info: when using HSE06 in QE you need to specify the divergence treatment as "gygi-baldereschi" while the vacuum treatment can be handled just adding enough vacuum.

Pseudo dojo pseudo potential are not optimal for hybrid functional calculation as you can see for the QE output:
BEWARE: nonlinear core correction is not consistent with hybrid XC

for the same reason they are not consistent with Many Body calculation, so I suggest you to use sg15 pseudpopontentials (http://www.quantum-simulation.org/potentials/sg15_oncv/).

In any case, the problem you are facing seems not related with the functional and pseudo.

Best,
Daniele

Re: HSE06 with yambo

Posted: Tue Jun 22, 2021 4:30 pm
by haseebphysics1
Dear Dr. Daniele,

Sorry for disturbing you again with the same query!

I just want to know, what should I do to circumvent the above-mentioned issue (Optics with HSE06 using Yambo)?

I checked the older version (4.5), which also yielded the same error! So, is it something in re-compiling p2y interface?

Thanks again,

Re: HSE06 with yambo

Posted: Wed Jun 23, 2021 8:46 am
by Daniele Varsano
Dear Haseeb,
unfortunately I still did not find the time to reproduce your error.
I hope to find it soon and I will be back to you asap.

Best,
Daniele

Re: HSE06 with yambo

Posted: Wed Jun 23, 2021 12:02 pm
by haseebphysics1
Alright, dear Dr. Daniele. Thanks for your time and help.

Re: HSE06 with yambo

Posted: Tue Jul 27, 2021 4:29 pm
by Daniele Varsano
Dear Haseeb,

sorry for the very late reply.
I had a look to your problem.
You can overcome it by:

1) in QE use input_dft="HSE" instead of HSE06

2) next in Yambo you need to change the following lines in:
/interfaces/p2y/mod_p2y.F

subroutine get_xc

Code: Select all

- &                            XC_FACTOR,XC_HYB_GGA_XC_GAUPBE,XC_HYB_GGA_XC_PBEH,XC_HYB_GGA_XC_B3LYP      
+&                            XC_FACTOR,XC_HYB_GGA_XC_GAUPBE,XC_HYB_GGA_XC_PBEH,XC_HYB_GGA_XC_B3LYP,XC_HYB_GGA_XC_HSE06 
around line 911:

Code: Select all

-         &  GS_xc_FUNCTIONAL==XC_HYB_GGA_XC_B3LYP*XC_FACTOR.or.GS_xc_FUNCTIONAL==0) then
+         &  GS_xc_FUNCTIONAL==XC_HYB_GGA_XC_B3LYP*XC_FACTOR.or.& 
+         &  GS_xc_FUNCTIONAL==XC_HYB_GGA_XC_HSE06*XC_FACTOR.or.GS_xc_FUNCTIONAL==0) then
where "-" is remove and "+" is add lines.

essentially you need to define XC_HYB_GGA_XC_HSE06 and include the GS_xc_FUNCTIONAL==XC_HYB_GGA_XC_HSE06*XC_FACTOR condition in the if statement.

Then recompile "make p2y" and it should work correctly.

If you have problem please post it again.

Best,
Daniele

Re: HSE06 with yambo

Posted: Tue Jul 27, 2021 6:35 pm
by haseebphysics1
Dear Daniele,
sorry for the very late reply.
No, it's okay!! Thanks for the help, in fact.


In terms of the procedure, I tried this and p2y worked perfectly fine! So, thanks again!