[Pw_forum] Cell geometry optimization issue

[Giovanni Cantele]

As explained here : http://www.quantum-espresso.org/faq/self-consistency/#6.2
before changing any calculation parameter such as mixing_beta it is advisable to
check if there is a problem with band input data, that means unit cell and/or atomic positions.

Just by visualising your input using XcrysDen you can easily find out that, for example,
atom 4
W   0.54140000000  0.0000000000  0.98510000000
and atom 22
W   0.45860000000  0.0000000000  0.98510000000

are just 0.5 A from each other : d ~ (0.54-0.46)*13*0.5291, a distance maybe too small to avoid self-consistency issues.
Maybe a similar, even though less critical issue is with atoms 7 and 13 (only 1.36 A).

So, before going on, I would check if your input structure is exactly the one you want to study, and if the input coordinates
are right in the right units.

Giovanni

> On 06 Nov 2015, at 14:59, Arijan Aleksić <aaleksic at fizika.unios.hr> wrote:
> 
> Dear all,
> 
> I am working on geometry optimization for WTe2 (Tungsten(IV) telluride).
> When running calculations, they failed to converge within limits in 100
> iterations.
> 
> First I have increased cell parameters for 10-15% to check if there will
> be any difference (also had big pressure). Next I went with reducing
> mixing_beta in steps of 0.1 to check for any changes which yielded no good
> results. And finally I used calculation = 'vc-relax' which gave even worse
> results and could not optimize the cell.
> 
> I will accept any advice on what should be my next action in order to
> succeed in getting optimization complete.
> 
> Here is the text from input file I used (version of Quantum Espresso is
> 5.2.1):
> 
> &control
>    calculation  = 'vc-relax',
>    restart_mode = 'from_scratch',
>    prefix       = 'wte2',
>    tprnfor      = .true.,
>    pseudo_dir   = '/home/aaleksic/Pseudo/',
>    outdir       = '/home/aaleksic/wte2/tmp/',
> /
> 
> &system
>    ibrav     = 8,
>    celldm(1) = 13,
>    celldm(2) = 0.53846,
>    celldm(3) = 2.15385,
>    nat       = 24,
>    ntyp      = 2,
>    nbnd      = 74,
>    ecutwfc   = 100.0,
> /
> 
> &electrons
>    mixing_beta = 0.1,
>    conv_thr    = 1.0d-12,
> /
> 
> &ions
> /
> 
> &cell
> /
> 
> ATOMIC_SPECIES
> W   74.0   W.pbe-mt_fhi.UPF
> Te  52.0   Te.pbe-mt_fhi.UPF
> 
> ATOMIC_POSITIONS (crystal)
> W   0.90050000000  0.5000000000  0.00000000000
> Te  0.29410000000  0.5000000000  0.09650000000
> Te  0.80020000000  0.0000000000  0.14000000000
> W   0.54140000000  0.0000000000  0.98510000000
> Te  0.35590000000  0.0000000000  0.34490000000
> Te  0.85170000000  0.5000000000  0.38930000000
> W   0.59950000000  0.5000000000  0.50000000000
> Te  0.20590000000  0.5000000000  0.59650000000
> Te  0.69980000000  0.0000000000  0.64000000000
> W   0.95860000000  0.0000000000  0.48510000000
> Te  0.14410000000  0.0000000000  0.84490000000
> Te  0.64830000000  0.5000000000  0.88930000000
> W   0.40050000000  0.5000000000  0.50000000000
> Te  0.79410000000  0.5000000000  0.59650000000
> Te  0.30020000000  0.0000000000  0.64000000000
> W   0.04140000000  0.0000000000  0.48510000000
> Te  0.85590000000  0.0000000000  0.84490000000
> Te  0.35170000000  0.5000000000  0.88930000000
> W   0.09950000000  0.5000000000  0.00000000000
> Te  0.70590000000  0.5000000000  0.09650000000
> Te  0.19980000000  0.0000000000  0.14000000000
> W   0.45860000000  0.0000000000  0.98510000000
> Te  0.64410000000  0.0000000000  0.34490000000
> Te  0.14830000000  0.5000000000  0.38930000000
> 
> 
> 
> K_POINTS (automatic)
> 4 4 4 0 0 0
> 
> Thank you.
> 
> Best regards,
> Arijan Aleksic,
> University J.J.Strossmayer, Croatia.
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
> 

-- 

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
Phone: +39 081 676910
Skype contact: giocan74

ResearcherID: http://www.researcherid.com/rid/A-1951-2009
Web page: http://people.na.infn.it/~cantele