[Pw_forum] Spin-Polarized NEB

Vic Bermudez c.h.bermudez at cox.net
Mon Feb 8 12:26:29 CET 2016


Hello,

	I'm doing a NEB calculation for a reaction where I suspect that the
transition state might be paramagnetic. In the "engine_input" section I set
'nspin=2', and I set a non-zero starting magnetization for one of the atom
types. In the first SCF cycle of the first NEB iteration the absolute
magnetization is initially non-zero and converges to a zero absolute
magnetization. The total magnetization remains zero at all times. This
applies to all 15 of the NEB images.
	However, in successive NEB iterations the SCF starts right from the
beginning with zero absolute and total magnetizations and remains at zero
for both quantities right up to SCF convergence.
	Here's my question. Given all this, am I really doing a spin-polarized
calculation, or is my calculation actually spin-restricted. If the latter,
what should I be doing to get a spin-unrestricted NEB run ?

Thank you in advance for your help.

This is the relevant part of the input. Here I'm restarting a calculation
that was spin-restricted and appeared to be getting somewhat close to NEB
convergence

BEGIN
BEGIN_PATH_INPUT
&PATH
restart_mode='restart',
string_method='neb',
ds=1.D0,
opt_scheme='broyden',
nstep_path=4000,
num_of_images=15,
k_max=0.6D0,
k_min=0.1D0,
first_last_opt=.TRUE.,
use_freezing=.TRUE.,
CI_scheme='auto'
/
END_PATH_INPUT

BEGIN_ENGINE_INPUT
&CONTROL
title='NEB Run',
! restart_mode='from_scratch',
pseudo_dir='/lustre/cmf/scratch/b/bermudez/QE_PP/',
! outdir='/lustre/cmf/scratch/b/bermudez/NEB-SO2_2/',
etot_conv_thr=1.0D-4,
forc_conv_thr=1.0D-3,
verbosity='default'
/

&SYSTEM
ibrav=8,
a=14.5898, b=14.7815, c=35.00,
nat=138,
ntyp=4,
ecutwfc=40.0,
ecutrho=400.0,
nspin=2,
starting_magnetization(1)=0.0,
starting_magnetization(2)=0.0,
starting_magnetization(3)=0.0,
starting_magnetization(4)=0.3333,
london=.true.,
occupations='smearing',
smearing='gaussian',
degauss=0.002,
nosym=.TRUE.
/

&ELECTRONS
electron_maxstep=200,
conv_thr=3.0D-9,
mixing_beta=0.7D0,
mixing_ndim=8,
mixing_mode='local-TF'
/

ATOMIC_SPECIES
Zr 91.22   Zr.pbe-spn-kjpaw_psl.0.2.3.UPF
O  15.9994  O.pbe-n-kjpaw_psl.0.1.UPF
H   1.0079  H.pbe-kjpaw_psl.0.1.UPF
S  32.06    S.pbe-n-kjpaw_psl.0.1.UPF


Victor M. Bermudez
E-mail: bermudez at alum.mit.edu





More information about the users mailing list