[Pw_forum] transition state using NEB

[Ravi Kiran]

Dear Dr. Lorenzo,

I have tried all possible paths from octa-octa but still I get the same
value. The full input script I am using to calculate activation energy is
as follows .
BEGIN
BEGIN_PATH_INPUT
&PATH
  string_method     = 'neb',
  restart_mode      = 'restart',
  nstep_path        = 60,
  ds                = 2.D0,
  opt_scheme        = "broyden",
  num_of_images     = 7,
first_last_opt      = .TRUE.,
  minimum_image     = .TRUE.,
   k_max             = 0.3D0,
  k_min             = 0.2D0,
  path_thr          = 0.01D0,
/
END_PATH_INPUT
BEGIN_ENGINE_INPUT
&control
    calculation='relax',
    prefix='NiHeN',
    tprnfor =.t.,
    pseudo_dir = '/home/ras256/nebH',
    outdir='/home/ras256/nebH/tmp',
/
 &SYSTEM
    nat = 5,
    ntyp = 2,
    degauss = 0.018,
    nbnd = 48,
    nspin = 2,
    starting_magnetization(1) = 0.0,
    starting_magnetization(2) = 0.5,
    ecutwfc = 45,
    ibrav = 0,
    occupations = 'smearing',
    smearing = 'gauss',
    ecutrho = 700,
 /
 &ELECTRONS
    conv_thr = 1e-08,
    mixing_beta = 0.7,
    mixing_mode = 'local-TF',
 /
&IONS
 /
&CELL
/
ATOMIC_SPECIES
    H   1.00   H.pbe-vbc.UPF
    Ni 58.6934 Ni.pbe-nd-rrkjus.UPF
K_POINTS automatic
   6 6 6   0 0 0
BEGIN_POSITIONS
FIRST_IMAGE
ATOMIC_POSITIONS crystal
    Ni 0 0 0
    Ni 0 0.5 0.5
    Ni 0.5 0 0.5
    Ni 0.5 0.5 0
    H  0.5 0.5 0.5
LAST_IMAGE
ATOMIC_POSITIONS crystal
    Ni 0 0 0
    Ni 0 0.5 0.5
    Ni 0.5 0 0.5
    Ni 0.5 0.5 0
    H  0.5 0.0 0.0
END_POSITIONS
CELL_PARAMETERS angstrom
    3.522900 0.000000 0.000000
    0.000000 3.522900 0.000000
    0.000000 0.000000 3.522900
END_ENGINE_INPUT
END
> Hi,
>
> I am using neb to calculate activation energy of hydrogen in unit cell
> of nickel. I have placed hydrogen in octahedral position and I am
> moving it to another octahedral position as shown below. My question
> is whether neb can calculates transition state which is tetrahedral
> site?. I have used a path_thr of 0.01eV with 7 images. I have
> calculated the activation energy which I am getting is 1.05 eV as
> shown in output, whereas the experiments indicate 0.42.
> Are you sure that initial and final positions are first-neighbours?
> The NEB code should produce an axsf file that you can examine with
> xcrysden to check the path.
>If it is ok, there are innumerable other possible sources of error, most
of which we cannot guess from the very limited input/output you provide
-- 
Thanks
Ravi
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150201/07f8635c/attachment.html>