[gpaw-users] some question about GPAW

[Jess Wellendorff Pedersen]

Hi Eskil,

in a way you're right:
Starting from scratch, vdW-DF type functionals are in general written 
E_xc = E_x_GGA + E_c_LDA + E_c_nl.
The last term, E_c_nl, is the principal term separating vdW-DF from 
other functionals. It accounts for non-local correlations by a 6D 
integral over densities interacting with all other densities in space 
through a kernel. Guess you already know about that.
Anyway, the E_c_nl kernel, and thereby the whole E_c_nl integral, is 
invariant under spin-polarization, i.e. it doesn't care about 
spin-polarization.
However, the two other terms, E_x_GGA and E_c_LDA, do most certainly 
care about spin-polarization. So when specifying the GPAW vdW-DF 
implementation to be spin-polarized you really just ask it to do 
spin-polarized calculations for the E_x_GGA and E_c_LDA terms....

Hope that answers your question.
Cheers,
Jess W.
CAMD, DTU, DK

Eskil Varenius wrote:
> Hi!
> I have a follow up question about spin-polarised vdW-DF. As far as I 
> knew the functionals vdW-DF 1 and 2 were only derived as spin-paired 
> expressions. I didn't know it was possible to do a spin-polarised 
> calculation with these. If it really is possible it is a great thing, 
> but is there any article or other source describing the 
> "spin-polarisation" of vdW-DF?
>
> Best regards,
> Eskil Varenius, Master student at Chalmers, Sweden
>
>
> 2011/3/31 Jess Wellendorff Pedersen <jesswe at fysik.dtu.dk 
> <mailto:jesswe at fysik.dtu.dk>>
>
>     Dear Wei-Bing,
>
>     It appears that those particular GPAW wiki pages that you refer to
>     have
>     not been updated for some time. We apologize for this.
>     The answers to your questions are simple:
>
>     Wei-Bing Zhang wrote:
>     >
>     > Dear all
>     >
>     >
>     >
>     > I am trying to do some calculations using GPAW. But I found
>     something
>     > which I can’t understand fully.
>     >
>     > 1) On the webpage
>     >
>     https://wiki.fysik.dtu.dk/gpaw/tutorials/bandstructures/bandstructures.html
>     > there is a sentence “One should note that as GPAW only works with
>     > orthorhombic cells …..”.
>     >
>     > Does this means one has to use orthorhombic cells in the
>     > calculations??  I use a non-orthorhombic cell for surface
>     calculation
>     > with GPAW, but the code didn’t complaint.  I want to know
>     whether the
>     > results including energy and geometry obtained using
>     non-orthorhombic
>     > cell in GPAW is valid or not???
>     >
>     GPAW natturally handles non-orthorhombic unit cells, and total
>     energy as
>     well as forces are completely valid.
>     >
>     > 2) On the webpage
>     >
>     https://wiki.fysik.dtu.dk/gpaw/documentation/xc/vdw.html#doing-a-van-der-waals-calculation
>     >
>     >
>     > I found the “vdw = FFTVDWFunctional(nspins=1, verbose=True)”. It
>     seems
>     > that there is a keywords “nspin”, which should control spin
>     polarized
>     > or spin paired calculations. However, in the newer version GPAW, I
>     > can’t find a keyword relate to spin polarized calculations. I
>     want to
>     > know whether the new version off vdW xc can treat the spin polarized
>     > calculations or not??
>     >
>     The GPAW exchange-correlation library has been rearranged such
>     that the
>     vdW-DF type functionals (vdW-DF and vdW-DF2) are used on the same
>     footing as e.g. LDA and GGA xc-functionals. This means that you simple
>     do " xc = 'vdW-DF' " or " xc = 'vdW-DF2' " in order to use these to
>     functionals. Spinpolarization is handled automatically "behind the
>     scenes" or, if the ASE Atoms object has no magnetic moments but
>     you for
>     whatever reason still wants GPAW to do a spinpolarized
>     calculation, you
>     enforce this with "spinpol=True" in the GPAW object.
>     Example:
>     sys = Atoms(.......)
>     calc = GPAW(xc='vdW-DF', txt='out.txt',spinpol=True)
>     sys.set_calculator(calc)
>
>     cheers,
>     Jess Wellendorff
>     >
>     >
>     >
>     > Thanks and Best Wishes
>     >
>     >
>     >
>     > Wei-Bing Zhang
>     >
>     >
>     ------------------------------------------------------------------------
>     >
>     > _______________________________________________
>     > gpaw-users mailing list
>     > gpaw-users at listserv.fysik.dtu.dk
>     <mailto:gpaw-users at listserv.fysik.dtu.dk>
>     > https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>
>     --
>     Jess Wellendorff Pedersen
>
>     ph.d student
>     Center for Atomic-scale Materials Design (CAMD)
>     Department of Physics
>     Technical University of Denmark (DTU)
>
>     _______________________________________________
>     gpaw-users mailing list
>     gpaw-users at listserv.fysik.dtu.dk
>     <mailto:gpaw-users at listserv.fysik.dtu.dk>
>     https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> gpaw-users mailing list
> gpaw-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users

-- 
Jess Wellendorff Pedersen

ph.d student
Center for Atomic-scale Materials Design (CAMD)
Department of Physics
Technical University of Denmark (DTU)