[Pw_forum] Epsilon calculation for non collinear spin system

[Kaustubh Kaluskar]

Dear All:

(Input files are produced below. I am using version 5.1)
I am trying to calculate dielectric properties of a system using epsilon
code. I am getting the following message and error:


  Fermi energy [eV] is: -3.36939
     The system is a metal...

     Performing eps calculation...

 
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%
     Error in routine grid_build (1):
     Non collinear spin  calculation not implemented
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%


In scf calculation preceding epsilon calculation, my Fermi energy came out
to be -2.6424 and dos showed it to be an insulator. My questions are:
(1) Why is epsilon code showing the system to be metal ? 
(2)   How to deal with the error given my system has non collinear spin
calculation in scf input file ? In this paper
(http://arxiv.org/abs/1305.6672) authors mention using a modified epsilon
code to account for full relativistic calculations. Does this limitation
still exists ? 

Regards,
Kaustubh
IISER Bhopal


!!!!!!!!!!!!!! SCF Input !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
&CONTROL
    calculation = 'scf',
    prefix = 'moSeSe',
    pseudo_dir='/home2/kkaluskar/PSEUDO'
    outdir='./tmp_mose2'
    tprnfor = .true.
    wf_collect=.true
/
&system
    ibrav =  4,
    celldm(1) = 6.2,
    celldm(3) = 15,
    nat = 3,
    ntyp = 2,
    nbnd=40,
nosym=.true.,
noinv=.true.,
    noncolin=.true.,
    lspinorb=.true.,
 starting_magnetization(1)=0,
    ecutwfc = 65,
    ecutrho = 510,
    occupations="smearing"
    smearing="fermi-dirac"
    degauss=0.005
 /
&electrons
    mixing_beta = 0.6
    electron_maxstep = 200
   ! diagonalization = 'cg'
 /
ATOMIC_SPECIES
Mo 95.94 Mo.rel-pbesol-spn-rrkjus_psl.0.3.0.UPF
Se 78.96 Se.pbe-van.UPF
ATOMIC_POSITIONS (alat)
Mo 0.0 0.576756641 1.000007469
Se 0.499487809 0.288378321 0.499997735
Se 0.499997735  0.289248195  1.499993206 K_POINTS crystal_b
 4
   0.0000000000     0.0000000000     0.0000000000     20
   0.3333333333    -0.6666666667     0.0000000000     20
   0.5000000000    -0.5000000000     0.0000000000     20
   0.0000000000     0.0000000000     0.0000000000     20
                                                           
!!!!!!!!!!!!!!!!!!! epsilon input !!!!!!!!!!!!!!!!!!!!!!!!!!!!!
&inputpp
outdir='./tmp_mose2'
prefix='moSeSe'
calculation='eps'
/
&energy_grid
smeartype='gauss'
intersmear=0.136d0,
intrasmear=0.0d0,
wmax=10.0d0,
wmin=0.0d0,
nw=500,
shift=0.0d0,
/