[Pw_forum] Phonon of single layer MoS2

[Ari P Seitsonen]

Dear "Shopping",

   Adding to Paolo's and Nicola's comments, I would rather use 'ibrav = 4' 
and 'a' and 'c' to define the lattice, making sure that the hexagonal 
symmetry group is indeed found, so that the k points are reduced into the 
irreducible wedge of the Brillouin zone - I am not sure whether it is a 
good idea of shifting the k points away from Gamma, as there is an issue 
with the Monkhorst-Pack (or, uniform) grid of k points (the Appendix of 
my PhD thesis from billions of years ago contains some examples, and a 
short version is, this shift was indeed introduced so that the sampling 
in hexagonal cell _could_ include the Gamma point).

   In order to make the coordinates look more symmetric, I would put the 
coordinate of Mo along the high-symmetry axis and the corresponding 
coordinates of S to +-0.04907565.

   Just to make the message clearer - also in the phonons there is no need 
to sample along the direction corresponding to the direction of the vacuum 
- 'nq3 = 1'.

   If you want accurate frequencies, I would consider reducing 'tr2_ph' for 
example to 10^-14, but this you can test.

     Greetings from Paris,

        apsi

-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
   Ari Paavo Seitsonen / Ari.P.Seitsonen at iki.fi / http://www.iki.fi/~apsi/
   Ecole Normale Supérieure (ENS), Département de Chimie, Paris
   Mobile (F) : +33 789 37 24 25    (CH) : +41 79 71 90 935


On Mon, 13 Jul 2015, Shopping wrote:

> Dear all:
>
> I am calculating phonons of singel-layer MoS2 (SLMoS2). I have two confusions here.
>
> First is usually it takes me two days to complete the phonon calculation for just 3 atom unit cell of SLMoS2. I am wondering if any settings in my input file is too high like some threshold, but no idea why it takes so long. I have attached  the scf.in and phonon.in file for scf calculation below.
>
> The second is do I have to include smearing in the scf calculation? Does it have large influence on the results? Since it takes too long to calculate I have not got the chance to try, anyone has similar experience willing to give me some advice here? I will be very appreciated, thanks.
>
> scf.in:
> &control
>    calculation='scf',
>    restart_mode='from_scratch',
>    nstep=50
>    tstress = .true.
>    tprnfor = .true.
>    prefix='SLMoS2',
>    pseudo_dir='/home/jpeng/Downloads/espresso-5.2.0/mypotential/'
>    outdir='/home/jpeng/Downloads/espresso-5.2.0/tmpdir/',
> /
> &system
>    ibrav= 0,
>    A=3.192,
>    nat=  3, ntyp= 2,
> /
> &electrons
>    conv_thr=1e-10
>    mixing_beta = 0.7
> /
> CELL_PARAMETERS
> 1.0 0.0 0.0
> 0.5 0.866 0.0
> 0.0 0.0 10
> ATOMIC_SPECIES
> Mo  95.94  Mo.pbe-spn-kjpaw_psl.0.2.UPF
> S  32.065  S.pbe-n-kjpaw_psl.0.1.UPF
> ATOMIC_POSITIONS (crystal)
> Mo 2/3 2/3 0.04907565
> S 1/3 1/3 0
> S 1/3 1/3 0.0981513
> K_POINTS automatic
> 6 6 6 1 1 1
>
> phonon.in:
> phonons of SLMoS2
> &inputph
>  tr2_ph=1.0d-12,
>  prefix='SLMoS2',
>  ldisp=.true
>  nq1=4
>  nq2=4
>  nq3=4
>  amass(1)=95.94,
>  amass(2)=32.065,
>  fildyn='SLMoS2.dyn',
>  outdir='/home/jpeng/Downloads/espresso-5.2.0/tmpdir/'
> /
>
> Best
> Jie Peng
> PhD student
> University of Maryland
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>