Author Topic: Global Scale Factor/ S6 value  (Read 4840 times)

0 Members and 1 Guest are viewing this topic.

Offline Dipankar Saha

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 516
  • Country: in
  • Reputation: 5
    • View Profile
Global Scale Factor/ S6 value
« on: June 10, 2014, 13:45 »

What will be the "Global Scale Factor/ S6" value in the case of the LDA functionals?? Is it taken as the same as 0.75 (GGA)??

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5446
  • Country: dk
  • Reputation: 89
    • View Profile
    • QuantumATK at Synopsys
Re: Global Scale Factor/ S6 value
« Reply #1 on: June 10, 2014, 15:14 »
 ???

Offline Dipankar Saha

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 516
  • Country: in
  • Reputation: 5
    • View Profile
Re: Global Scale Factor/ S6 value
« Reply #2 on: June 10, 2014, 16:11 »
Instead of GGA~PBE-D2+CP...if I use LDA~PW...then do BSSE CP correction and then do Dispersion correction using DFT-D2...
What should be the value of ..... S6  in "extendedCalculatorWithDFTD2(bsse_calculator, global_scale_factor= ? )" ???
« Last Edit: June 10, 2014, 16:14 by Dipankar Saha »

Offline Umberto Martinez

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 479
  • Country: dk
  • Reputation: 26
    • View Profile
Re: Global Scale Factor/ S6 value
« Reply #3 on: June 10, 2014, 17:12 »
You can find a list to functionals for which the s6 scale factors have been determined in  S. Grimme, J. Comp. Chem. 27, 1787, 2006.
namely, 0.75 (PBE), 1.2 (BLYP), 1.05 (B-P86), 1.0 (TPSS), and 1.05 (B3LYP).

to my knowledge, this global scale factor has not been determined for LDA.

Offline Dipankar Saha

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 516
  • Country: in
  • Reputation: 5
    • View Profile
Re: Global Scale Factor/ S6 value
« Reply #4 on: June 10, 2014, 17:30 »
Thank you Umberto.... :)

 But I would like ask you another question....

If we do interfacing between two layers...maintain the distance as per the expmntl. data.... and use LDA-PW, then may be the binding energy value will be improper..../ But how the that will affect the nature of the band diagram, dispersion relation, stress etc. ??
To be more specific.... to get more accurate dispersion relation, BG, stress etc. .... which one is better (say, the inter-layer distance and, Eb are not the ONLY concerns)__
1) GGA~PBE-D2+CP  or,
2) LDA-PW with out DFT-D2 correction ......??
« Last Edit: June 10, 2014, 19:35 by Dipankar Saha »

Offline Umberto Martinez

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 479
  • Country: dk
  • Reputation: 26
    • View Profile
Re: Global Scale Factor/ S6 value
« Reply #5 on: June 13, 2014, 09:56 »
These are corrections to the energy by adding extra terms.
They do not affect the electronic structure (besides giving you a different geometry) which is calculated by the DFT part (GGA or LDA in your case).

Offline Dipankar Saha

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 516
  • Country: in
  • Reputation: 5
    • View Profile
Re: Global Scale Factor/ S6 value
« Reply #6 on: June 13, 2014, 10:34 »

"besides giving you a different geometry" ....means??

Offline Umberto Martinez

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 479
  • Country: dk
  • Reputation: 26
    • View Profile
Re: Global Scale Factor/ S6 value
« Reply #7 on: June 13, 2014, 10:47 »
means the electronic structure depends also on the geometry of your system.
grimme dispersion correction and CP correction may give you a different geometry (interlayer distance in your case) with respect to GGA/LDA only.

Offline Dipankar Saha

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 516
  • Country: in
  • Reputation: 5
    • View Profile
Re: Global Scale Factor/ S6 value
« Reply #8 on: June 20, 2014, 19:57 »
Now suppose...

if  there are no stacked layers....thus no concept of vdW force...., then for finding the electronic band structure, as well as phonon dispersion....which one should we prefer...LDA-PW or GGA-PBE? Any comparative adv. or disadv. of GGA_PBE ??
(N.B. :- Again, here is no question of under estimating or overestimating the inter layer distance or, binding energy....as it's a mono layer)

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5446
  • Country: dk
  • Reputation: 89
    • View Profile
    • QuantumATK at Synopsys
Re: Global Scale Factor/ S6 value
« Reply #9 on: June 27, 2014, 05:37 »
Why not just do both and compare, then you will know if there is a difference? That will be an interesting result to include in your paper. Do note however, that you may need a larger mesh cut-off for GGA for good accuracy.

Offline Dipankar Saha

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 516
  • Country: in
  • Reputation: 5
    • View Profile
Re: Global Scale Factor/ S6 value
« Reply #10 on: June 27, 2014, 10:30 »
Dr. A. Blom,

Thank u for mentioning the point regarding the mesh cut-off for GGA....
...let me check... :) 

Regards__
Dipankar