[Pw_forum] Transition state calculation error

Paolo Giannozzi p.giannozzi at gmail.com
Wed May 4 10:21:39 CEST 2016


For sure your structure looks much better if you set
CELL_PARAMETER angstrom
...
ATOMIC_POSITIONS angstrom
...
You should also set K_POINTS gamma: it is faster and takes less memory

Paolo


On Wed, May 4, 2016 at 10:04 AM, Paolo Giannozzi <p.giannozzi at gmail.com>
wrote:

> This is unfortunately a nasty error that is not easy to solve. It is
> either due to pseudopotentials or to really bad structures. If the problem
> arises in the first NEB step, verify that your starting positions do not
> have atoms too close
>
> Paolo
>
> On Tue, May 3, 2016 at 2:48 PM, Cecil Humphrey Botchway <
> humphrey2g4 at gmail.com> wrote:
>
>> Hi There,
>> I am trying to run a transition state calculation using neb.x with
>> methanol C-OH bond breaking in the zeolite micro pore (the system is
>> composed of 120 atoms). I have tried relaxing the whole structure with the
>> adsorbate and freezing only the zeolite leaving the adsorbate. I have even
>> tried increasing and decreasing the numbers of processor cores from 8-16. I
>> encounter the same error (which is below) regardless .
>>
>>
>>
>>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>      task #         0
>>      from rdiaghg : error #       818
>>      S matrix not positive definite
>>
>>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>>
>>
>> What could be the possible cause?
>> This is how my input looks like
>>
>> BEGIN
>> BEGIN_PATH_INPUT
>> &PATH
>>   restart_mode      = 'from_scratch'
>>   string_method     = 'neb',
>>   nstep_path        = 20,
>>   ds                = 2.D0,
>>   opt_scheme        = "broyden",
>>   num_of_images     = 7,
>>   k_max             = 0.3D0,
>>   k_min             = 0.2D0,
>>   CI_scheme         = "auto",
>>   path_thr          = 0.1D0,
>> /
>> END_PATH_INPUT
>> BEGIN_ENGINE_INPUT
>> &CONTROL
>>   prefix         = "TS1"
>>   outdir         = "./tmp/",
>>   pseudo_dir     = "/home/cbotchway/Cecil",
>> /
>> &SYSTEM
>>   ibrav                  = 0,
>>   nat                    = 120,
>>   ntyp                   = 5,
>>   ecutwfc                = 40.0D0,
>>   ecutrho                = 240.0D0,
>>   nspin                  = 2,
>>   starting_magnetization = 0.5D0,
>>   occupations            = "smearing",
>>   degauss                = 0.03D0,
>> /
>> &ELECTRONS
>>   conv_thr    = 1.D-6,
>>   mixing_beta = 0.2D0,
>>   electron_maxstep =1000000,
>> /
>> &IONS
>> /
>> ATOMIC_SPECIES
>> O    15.99940    O.pbe-van_ak.UPF
>> Si   28.0855    Si.pbe-n-van.UPF
>> Al   26.98154   Al.pbe-n-van.UPF
>> H     1.00794    H.pbe-van_ak.UPF
>> C    12.0107     C.pbe-van_ak.UPF
>> BEGIN_POSITIONS
>> FIRST_IMAGE
>>
>> Attached to this post is a copy of my input file
>> Any help or advice would be appreciated
>> Thank you in advance
>> --
>> Cecil Huymphrey Botchway
>> Mphil. Material Science and Computational Chemistry
>> Department of Chemistry
>> Kwame Nkrumah University of Science and Technology
>> +23329002372
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>
>
>
>
> --
> Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
> Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
>
>


-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20160504/07544016/attachment.html>


More information about the users mailing list