B3LYP Calculations

Message

Mahlaga · #1 Post by **Mahlaga** » Mon Jul 20, 2009 9:33 am

Dear VASP Users ,

I need to do energy gap calculations using the B3LYP hybrid functional. A typical INCAR file is given in the manual only for the PBE0 and HSE hybrids, but no explanation of tags and flags is given for B3LYP functional. Can anyone who has performed calculations based on this functional please help.

Thank you in advance.

Mahlaga

panda · #2 Post by **panda** » Wed Jul 22, 2009 9:58 pm

Is there a specific reason that you need to use this functional? From the VASP manual, "The GGA-tag was added to perform GGA calculations with pseudopotentials generated with conventional reference configurations. The tag is named GGA. The options for the tag are:
GGA = PW Perdew-Wang
GGA = PB Perdew-Becke
GGA = LM Langreth-Mel-Hu
GGA = 91 Perdew-Wang 91
GGA = PE Perdew-Burke-Ernzerhof
GGA = RP revised Perdew-Burke-Ernzerhof

What system are you working with? PE is generally a good choice for GGA calculations as a first step. If you are not sure try all of the available functionals and see if there is any discrepancy.
B3LYP is available using the Gaussian software pacakge, I am not sure that it is available in VASP (in fact, I am almost certain it is not).

Mahlaga · #3 Post by **Mahlaga** » Tue Aug 04, 2009 8:10 am

B3LYP is in fact available in the recently released VASP5.2. For details of the implementation please see J. Paier et al. Journal of Chemical Physics 127, 024103 (2007). The tag GGA = B3 is used for B3LYP calculations.

panda · #4 Post by **panda** » Tue Aug 04, 2009 3:17 pm

I will read this, thanks. Is it available in 4.6 also? (that is the version that I was referring to as per reading the users manual)

VASP001 · #5 Post by **VASP001** » Sat Jan 09, 2010 6:12 pm

B3LYP only can be used in VASP5 or newer version.

VASP001 · #6 Post by **VASP001** » Sat Jan 09, 2010 6:16 pm

[quote=" B3 is used for B3LYP calculations. [/quote"]</span>

Mahlaga · #7 Post by **Mahlaga** » Fri Jan 29, 2010 9:42 am

Dear VASP001,

Yes I had to set LEXCH = B3 in the POTCAR file. The jobs (bandstructure calculations on various phases of ZnO) are running, but there is this persisting warning "Your generating k-point grid is not commensurate to the symmetry of the lattice. Ths can cause slow convergene with respect to k-points for HF type calculations"
I've used automatic k-point generation, and shifted the grid to Gamma as suggested but I still get the same warning. Do you have any idea of how to solve this problem?
Suggestions from anyone would be much appreciated.

VASP001 · #8 Post by **VASP001** » Wed Mar 31, 2010 3:30 am

The head-administor have given very clear answers ,you can look forward to it .

http://cms.mpi.univie.ac.at/vasp-forum/ ... ?4.6633.10

You can get answer from this link

Best wishes.
<span class='smallblacktext'>[ Edited Wed Mar 31 2010, 05:48AM ]</span>

Mahlaga · #9 Post by **Mahlaga** » Wed Mar 31, 2010 10:22 am

Got it. Thank you very much

emily.t · #10 Post by **emily.t** » Thu Jun 02, 2011 8:55 pm

I have been trying to run B3LYP calculations, and have been getting errors both of " RSPHER:running out of buffer" and "call to ZHEGV failed".