Error in fildvscf with Hubbard U correction.

Post here questions linked with issue while running the EPW code

Moderators: hlee, hlee

Post Reply
choi90
Posts: 1
Joined: Sat Jul 23, 2022 3:40 am
Affiliation: Advanced Light Source

Error in fildvscf with Hubbard U correction.

Post by choi90 »

I need to calculate the system with Hubbard U correction.
So, I put the Hubbard code in scf file as below.

lda_plus_u = .TRUE.
lda_plus_u_kind = 0
Hubbard_U(1) = 3
Hubbard_U(2) = 0

Then I run ph.x with ph.in file with "fildvscf = 'dvscf'", because of later EPW calculation, like "dvscf_dir = './save'" in epw.in.
But ph.x calculation ends up with an error like below.

Error in routine diropn (34):
can't open a connected unit

I searched that error and found that someone suggest just remove the "fildvscf = 'dvscf'" in the ph.x in the case of Hubbard U correction.
I attach the link of that suggestion.

https://lists.quantum-espresso.org/pipe ... 46445.html

So, I tested calculation without "fildvscf = 'dvscf'", and it is running without an error.
But I need to input the both Hubbard U correction in scf and "fildvscf = 'dvscf'" in ph to calculate the EPW.
How can I solve this problem?
Is there any way to epw calculation without "fildvscf = 'dvscf'"?

hlee
Posts: 378
Joined: Thu Aug 03, 2017 12:24 pm
Affiliation: The University of Texas at Austin

Re: Error in fildvscf with Hubbard U correction.

Post by hlee »

Dear choi90:
Is there any way to epw calculation without "fildvscf = 'dvscf'"?
No. EPW needs the perturbing potential from ph.x.

More importantly, currently, EPW doesn't support the calculations with Hubbard U correction.

Sincerely,

H. Lee

Post Reply