QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: markmick on August 9, 2016, 06:25

Title: Gibbs Free Energy and VASP
Post by: markmick on August 9, 2016, 06:25

I've successfully performed binding energy calculations with VASP, but I'm stumped with Gibbs free energy.  Based on an article I'm studying, Gibbs Energy for an adsorbed hydrogen is calculated by:

ΔG_H = ΔE_H + ΔE_ZPE - TΔS

Some of these variables I understand, others I don't:

* T is temperature in Kelvin (Setting this in INCAR I believe)
* ΔS is entropy, which I believe is pulled from OUTCAR
* ΔE_H  is the binding enegy, and I know how to do this one
* ΔE_ZPE (Zero-point-energy).  This one has me stumped. 

ΔE_ZPE  = ΔE^nH - ΔE^(n-1)H - 1/2 E^H2

ΔE_ZPE has something to do with meV and THz values pulled from OUTCAR.  My INCAR is set with:

NELMIN = 4
ISMEAR = 0
IBRION = 7
ISIF = 0
LEPSILON = .TRUE.
NSW = 1
NWRITE = 3

and this produces an OUTCAR with final values like:

63 f  =    0.665149 THz     4.179253 2PiTHz   22.186976 cm-1     2.750835 meV

How to translate this OUTCAR data into the values for ΔE_ZPE elude me.  I assume just use the meV value?   Thanks for any assistance.  I'm accustomed to Gibbs from Thermodynamic calculations, but using VASP for this is a new frontier.

Sincerely,
Mark




 

Title: Re: Gibbs Free Energy and VASP
Post by: Anders Blom on August 9, 2016, 08:51

You will probably have more luck getting this question properly answered at the VASP forum than here.

Title: Re: Gibbs Free Energy and VASP
Post by: zh on August 21, 2016, 10:30

For an A2 diatomic molecule, the zero point energy is 0.5* h *v, where h is Planck's constant, and v is the vibration frequency of this molecule.