[Pw_forum] Application of LDA+U in phonon calculations.

saswata halder sanky.h at gmail.com
Thu Sep 24 20:58:56 CEST 2015


Dear All,

At present I am trying to calculate the phonon modes of double
perovskite Ba2SmTaO6 having a cubic Fm3m structure. The detailed input
of the vc-relax file is provided below (and also attached alongwith).

 &CONTROL
    calculation='vc-relax',
    restart_mode='from_scratch',
    prefix='BST'
    pseudo_dir = '/home/saswata/espresso/potentials/',
    outdir='./'
    tstress=.true.
    tprnfor=.true.
    verbosity='high'
    wf_collect=.true.
    nstep=200
    etot_conv_thr=1.0D-7
    forc_conv_thr=1.0D-4
    disk_io='default'
 /
 &SYSTEM
    ibrav = 2, celldm(1) =16.0208, nat = 10, ntyp= 4,
    ecutwfc = 80.0, ecutrho=800, occupations='smearing',
    smearing='gaussian', degauss=0.02, nspin=2,
    starting_magnetization(2)=0.4,
    lda_plus_u=.true.,
    Hubbard_U(2)=7.0,

 /
 &ELECTRONS
    diagonalization='david',
    mixing_mode='plain',
    mixing_beta = 0.7D0
    conv_thr =  1.0D-9
    electron_maxstep=200
 /
&IONS
    ion_dynamics='bfgs'
    upscale=10D0
    trust_radius_max=1.0D0
    trust_radius_min=1.0D-5
    trust_radius_ini=0.5D0
    w_1=0.01D0
    w_2=0.5D0
/
&CELL
    cell_dynamics='bfgs',
    press=0.0,
    cell_factor=2.5D0,
    press_conv_thr=0.2,
/
ATOMIC_SPECIES
 Ba  137.327    Ba.pbe-mt_fhi.UPF
 Sm  150.360    Sm.pbe-mt_fhi.UPF
 Ta  180.94788  Ta.pbe-mt_fhi.UPF
 O   15.9994     O.pbe-mt_fhi.UPF
ATOMIC_POSITIONS crystal
 Ba 0.25  0.25  0.25
 Ba 0.75  0.75  0.75
 Sm 0.00  0.00  0.00
 Ta 0.50  0.50  0.50
 O  0.264 0.00  0.00
 O  0.736 0.00  0.00
 O  0.00  0.264 0.00
 O  0.00  0.736 0.00
 O  0.00  0.00  0.264
 O  0.00  0.00  0.736
K_POINTS automatic
    4 4 4  1 1 1

  I specifically have two questions regarding the output of this calculation

1. The structural relaxation occurs but with a pressure value of
P=-13.71. What is the significance of this negative pressure?

2. The phonon calculation terminates with an error message 'LDA+U is
not implemented in phonon'. Is there any other way to go about with
this calculation in the existing QE versions?

Thanking you in advance and awaiting a reply eagerly,

Saswata Halder.

Senior Research Fellow

Bose Institute.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bst.relaxcell.in
Type: application/octet-stream
Size: 1504 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150925/0003e4a7/attachment.obj>


More information about the users mailing list