[QE-users] Physical constants conversion inconsistency

[Giovanni Cantele]

Dear S. Kumar,

maybe some developer will give you a more precise answer. However, just to make a remark, what I would say is that what one should also check
is to what extent a change in the conversion and/or in the alat would modify your results. 
For example, as far as the the number
of decimal places is concerned, let us suppose that we are making a calculation of a cubic crystal with a = 3.40613 A and in place of it
we specify a = 3.4061 A. That would correspond to an underestimation of the lattice constant of less that 0.0009 %, much less than typical accuracy that
can be reached by DFT in estimating the lattice constants. With such a small error, if you compute for example the band structure with
the two values of the lattice parameter, you would appreciate no difference between the two (I would guess with changes in the eigenvalues
much less that 1 meV).

Just to make an exercise, I tried with the following input, that concerns bulk silicon:
&CONTROL
    calculation  = 'scf'
    title        = 'Silicon'
    restart_mode = 'from_scratch'
    outdir       = './tmp/'
    prefix       = 'Si'
    tprnfor      = .true.
    pseudo_dir   = './'
/
&SYSTEM
    ibrav        = 2
    a              = 5.43
    nat          = 2
    ntyp         = 1
    ecutwfc    = 44
    ecutrho    = 200
    nbnd       = 8
/
 &ELECTRONS
/
ATOMIC_SPECIES
        Si   28.09     Si.pbe-n-kjpaw_psl.1.0.0.UPF
ATOMIC_POSITIONS { alat }
Si       0.00  0.00  0.00
Si       0.25  0.25  0.25
K_POINTS { automatic }
4 4 4    0  0  0


With the shown value of a you get a total energy of -93.43935682 Ry and eigenvalues at Gamma point equal to
-5.7252   6.2495   6.2495   6.2495   8.7945   8.7945   8.7945   9.6160

If I change a from 5.43 to 5.43005 (on the 5th digit, which corresponds to a change of 0.0009%) the previous quantities become
-93.43935815 Ry and
-5.7253   6.2492   6.2492   6.2492   8.7941   8.7942   8.7942   9.6154.
The total energy changes by -0.00000133 Ry (-0.018 meV), the gap at Gamma point from 2.545 to 2.5449 (-0.1 meV).


Giovanni


> On 23 Jan 2020, at 06:00, Sonu Kumar <1009ukumar at gmail.com> wrote:
> 
> Dear QE developers,
> 
> I found that physical constant conversion factors are incorrect in the q-e version 6.2 and 6.4.1. For example, Bohr to ang conversion factor is around 0.5291772085900001, while nist codata value (https://physics.nist.gov/cgi-bin/cuu/Value?bohrrada0 <https://physics.nist.gov/cgi-bin/cuu/Value?bohrrada0>) is 5.291 772 109 03 e-11 m.  This error is introduced in data-file-schema.xml files as lattice parameters are written in Bohr (provided input file contain lattice parameters in angstrom units). This error might lead to other erroneous physical quantities involving angstrom units. 
> 
> Also, output file contains 'lattice parameter (alat)  =' values upto 4 decimal places; It will be better (in my opinion) to add more decimal places so as to check conversions to angstrom units or so.
> 
> Looking into the constants.f90 files in Modules in the src directory, I found the same. This is true for other constants mentioned in this file. 
> 
> With kind regards, sonu
> 
> 
> With kind regards,
> S Kumar |  PhD.
> ICReDD
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users

-- 

Giovanni Cantele, PhD

CNR-SPIN
c/o Dipartimento di Fisica
Universita' di Napoli "Federico II"
Complesso Universitario M. S. Angelo - Ed. 6
Via Cintia, I-80126, Napoli, Italy

e-mail: giovanni.cantele at spin.cnr.it
            gcantele at gmail.com
Phone: +39 081 676910
Skype contact: giocan74
Web page: https://sites.google.com/view/giovanni-cantele

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200123/6b6e7114/attachment.html>