[Pw_forum] Si bandgap using hybrid functionals

Gabriele Sclauzero sclauzer at sissa.it
Mon Jul 26 10:07:18 CEST 2010


On 07/20/2010 11:45 PM, Janam wrote:
> Hi,
>
> I'm trying to get the right band gap (~1.12 eV) for Silicon using hybrid
> functions. I'm using espresso 4.2. Unfortunately, the band gap doesn't
> seem to be affected by using the hybrid functionals.
>
> For example, when I use Si.pb-rrkj.UPF (no hybrid functional), I get a
> band gap of 0.633eV. When I use the same pseudopotential with B3LYP, I
> get a band gap of 0.583.
>    

I think you are not using the exact exchange method in the correct way. 
At the moment there are no pseudopotentials built upon exact exchange 
calculations, so that one uses the non-hybrid PP _corresponding_ to the 
hybrid functional.
For instance, you can use PBE PPs with PBE0 hybrid functional, but you 
CANNOT use PZ PPs with the PBE0 hybrid. For the same reason I guess that 
you need to use BLYP PPs with the B3LYP hybrid.
In your example below you are mixing a PBE PP with the B3LYP hybrid. I 
think this is wrong because it leads to an unconsistency.
Try with input_dft='pbe0', you should see the opening of the Si band gap.

At least, this is my (little) understanding so far...

HTH

GS

> I've posting my scf input file below. Any help/advice would be greatly
> appreciated
>
> &control
>      prefix='si-pbe-b3lyp',
>      pseudo_dir='/home/nanohub/janam/QuantumEspresso/Si/pseudo'
>      outdir ='/home/nanohub/janam/QuantumEspresso/Si/tmp',
>      tstress = .true.,
>      tprnfor = .true.,
>   /
>   &system
>      ibrav=  2, celldm(1) =10.35, nat=  2, ntyp= 1,
>      ecutwfc = 80.0, input_dft="b3lyp",
>   /
>   &electrons
>      conv_thr=1E-7
>   /
> ATOMIC_SPECIES
>   Si  28.086  Si.pbe-rrkj.UPF
> ATOMIC_POSITIONS
>   Si 0.00 0.00 0.00
>   Si 0.25 0.25 0.25
> K_POINTS automatic
>     8 8 8 0 0 0
>
> Sincerely
> ---
> Janam Jhaveri
> School of Electrical and Computer Engineering
> Purdue University
> C# (765)-409-1948
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>    


-- 

Gabriele Sclauzero, EPFL SB ITP CSEA
PH H2 462, Station 3, CH-1015 Lausanne



More information about the Pw_forum mailing list