[Pw_forum] bands calculation

[Gabriele Sclauzero]

udayagiri sai babu wrote:
> Dear all i am calculating electronic band structure of LaoFeP 
> superconductor. I have two questions regarding this.
> As I have understood from the example01, I need to perform an "scf" 
> calculation followed by "bands" calculation.

That's right.

> My question is
> 1) Why cant we do bands calculation alone, as all we need is energy 
> eigen values at different k_points for all available bands. This can be 
> done by solving Kohn-Sham equations at those k points for all bands.

Because DFT is a self-consistent field (SCF) method! I hope this is clear to you, with all 
that this implies.

Essentially, to get the self-consistent KS Hamiltonian you need the self-consistent charge 
density, which is defined as an integral in the 3D BZ. This integral is approximated with 
a sum over an appropriate set of k-points in the BZ (PWscf uses uniformly spaced regular 
grids, possibly shifted from the origin). Since you have to perform this kind of summation 
many times in a self consistent cycle (until converge is reached), you choose the minimal 
set of k-points which gives you the desired accuracy.

Once you have the SCF Hamiltonian (only AFTER you got the SCF charge density), you can 
diagonalize it at any desired k-point to get the band structure along a path in reciprocal 
space. This path is not chosen in order to give a good sampling of the BZ (as instead the 
regular grid of above), but usually follows some high symmetry lines of the crystal.


> 2) what data from scf calculation is used by bands calculation.

The self-consistent charge density is the key quantity. See also the replies from other 
people.

> 
> 3)when I did scf calculation for the following input file fermienergy is 
> not written in the output file. can somebody tell me if something is wrong?

Nothing wrong. The Fermi energy is not well defined for an insulator.

HTH

GS

> 
> &control
> calculation ='scf'
> restart_mode='from_scratch',
> outdir='/home/sai/laofep/',
> prefix='laofep'
> pseudo_dir='/home/sai/softwares/espresso-4.1.2/pseudo'
> /
> &system
> ibrav=0,
> celldm(1)=7.487,
> celldm(3)=16.083,
> nat=8,
> ntyp=4,
> ecutwfc=50,
> ecutrho=500,
> occupations='fixed',
> nbnd=38,
> /
> &electrons
> conv_thr=1.D-8,
> diagonalization='david',
> mixing_mode='plain',
> mixing_beta= 0.7
> /
> ATOMIC_SPECIES
> La 138.905  La.pw91-nsp-van.UPF
> O 15.999 O.pw91-van_ak.UPF
> P 30.973 P.pw91-van_ak.UPF
> Fe 55.845 Fe.pw91-sp-van_ak.UPF
> ATOMIC_POSITIONS (bohr)
> La 0.000234206 3.687761102 10.487650296
> La 3.687732709 0.000277249 5.739475052
> O 0.000215849 0.000260237 8.115083977
> O 3.687739522 3.687788174 8.115072590
> Fe 0.000245844 0.000231960 0.091985046
> Fe 3.687778452 3.687741841 0.091985975
> P 3.687720269 0.000225890 14.089592313
> P 0.000266430 3.687752524 2.133535171
> K_POINTS {automatic}
> 10 10 5 0 0 0
> CELL_PARAMETERS
> 0.985045304 0.000000280 -0.000002934
> 0.000000280 0.985045750 0.000000396
> -0.000004795 -0.000000455 2.142143981
> 
> -- 
> U.Saibabu
> PhD student,
> Deformation mechanisms modeling group,
> Materials engineering department,
> IISc Bangalore,
> India.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

-- 


o ------------------------------------------------ o
| Gabriele Sclauzero, PhD Student                  |
| c/o:   SISSA & CNR-INFM Democritos,              |
|        via Beirut 2-4, 34014 Trieste (Italy)     |
| email: sclauzer at sissa.it                         |
| phone: +39 040 3787 511                          |
| skype: gurlonotturno                             |
o ------------------------------------------------ o