[Pw_forum] Ti4+ Pseudopotential

Manu Hegde mhegde at uwaterloo.ca
Tue Aug 30 23:25:30 CEST 2016 

Dear Thomas,
Thank you very much. I will have a look.
Thanks and Regards,
Manu

On Tue, Aug 30, 2016 at 5:49 AM, Thomas Brumme <thomas.brumme at mpsd.mpg.de>
wrote:

> Dear Manu,
>
> first of all, please note that I'm not an expert in pseudopotential
> creation! Anyway...
>
> If I use the input of the pslibrary 1.0 as a starting point, I have no
> problems in creating a pseudo...
> The logarithmic derivatives look fine, there are no ghosts and also the
> energies are ok... Yet, just ok...
> The transferability will be not so good as the difference between the
> all-electron calculation and the
> ps one for the normal, uncharged groundstate of Ti shows a difference of
> about 1 mRy for the states
> of the n=3 shell and also the 3s states for the +2 charged state is off by
> 1 mRy... Yet, these are
> semi-core states and maybe you're not interested in them...
>
> I think that it is always better to include all states of one shell
> (except maybe f-states) even if there are
> not occupied... In fact, even if the 3d shell is empty for the +4
> oxidation state, it is lower in energy than
> the 4s and 4p... Not including it in the pseudo makes not sense...
>
> Maybe you should try using not the +4 state to create the pseudo... Maybe
> the +2 or the normal state
> gives in the end better energies AND transferability... Or try changing
> the radii... Or wait until someone
> with more knowledge in the creation of pseudos answers :)
>
> Anyway, here the input I used:
>
>  &input
>    title='Ti d',
>    zed=22.0,
>    rel=1,
>    config='[Ar] 3d0.0 4s0 4p0',
>    iswitch=3,
>    dft='PBE'
>    nld=5,
>    rlderiv=2.0,
>    eminld=-14,
>    emaxld=2,
>    deld=0.012d0,
>  /
>  &inputp
>    lpaw=.true.,
>    pseudotype=3,
>    file_pseudopw='Ti.pbe-spn-kjpaw.UPF',
>    author='ADC modified by TB',
>    lloc=-1,
>    rcloc=1.6,
>    which_augfun='BESSEL',
>    rmatch_augfun_nc=.true.,
>    nlcc=.true.,
>    new_core_ps=.true.,
>    rcore=0.7,
>    tm=.true.
>    lgipaw_reconstruction=.true.
>    use_paw_as_gipaw=.true.
>  /
> 6
> 3S  1  0  2.00  0.00  0.85  1.30  0.0
> 4S  2  0  0.00  0.00  0.85  1.30  0.0
> 3P  2  1  6.00  0.00  0.85  1.60  0.0
> 4P  3  1  0.00  0.00  0.85  1.60  0.0
> 3D  3  2  0.00  0.00  0.85  1.50  0.0
> 3D  3  2  0.00  0.05  0.85  1.50  0.0
>
>
> and the input for the test:
>
>
>  &input
>    title='Ti d',
>    zed=22.0,
>    rel=1,
>    config='[Ar] 3d0.0 4s0 4p0',
>    iswitch=2,
>    dft='PBE'
>  /
> &test
> file_pseudo='Ti.pbe-spn-kjpaw.UPF',
> nconf=3
> configts(1)='3s2 3p6 3d0 4s0 4p0'
> configts(2)='3s2 3p6 3d2 4s2 4p0'
> configts(3)='3s2 3p6 3d0 4s2 4p0'
> ecutmin=50, ecutmax=200, decut=25
> rm=100
> /
>
> Regards
>
> Thomas
>
>
>
>
> On 08/29/2016 07:05 PM, Manu Hegde wrote:
>
> Hi Thomas,
> Here is my input file, I have modified the EPFL-THEOS one. I have attached
> both input and test file (out put). But I found a pw-91 GGA, one which is
> verified by Paolo. It has a valence configuration of [Ar] 3d2 4s0, which is
> close to Ti4+. I started using the same.
>
> One I have attached is PBE and my relax calculations did not converge.
>
> Regards,
> Manu
> (University of Waterloo)
>
> On Mon, Aug 29, 2016 at 7:21 AM, Paolo Giannozzi <p.giannozzi at gmail.com>
> wrote:
>
>> On Mon, Aug 29, 2016 at 1:17 PM, Thomas Brumme <thomas.brumme at mpsd.mpg.de
>> > wrote:
>>
>>
>>> I would say, that it is usually a bad sign if there are nodes which
>>> shouldn't be there, but I could be wrong...
>>>
>> ... but you aren't: pseudized atomic wavefunctions should be nodeless
>>
>> Paolo
>>
>> On 08/27/2016 09:06 PM, Manu Hegde wrote:
>>
>>> Hello All,
>>> Posting again there was a typo.
>>> I am trying to generate Ti pseudopotential with oxidation state *+4*. I
>>> have assumed valelnce electronic configurations as ''[Ar] 3d0.0 4s0 4p0''.
>>> When I did the PP test, i found this warning
>>>        Warning: n=1, l=0 expected 0 nodes, found 3
>>>        Setting wfc to zero for this iteration
>>>
>>> Is it okay to use this PP for further calculations?
>>>
>>> Thank You,
>>>
>>> Manu
>>>
>>> (University of Waterloo)
>>>
>>> On Sat, Aug 27, 2016 at 2:58 PM, Manu Hegde <mhegde at uwaterloo.ca> wrote:
>>>
>>>> Hello All,
>>>>
>>>> I am trying to generate Ti pseudopotential with oxidation state +5. I
>>>> have assumed valelnce electronic configurations as ''[Ar] 3d0.0 4s0 4p0''.
>>>> When I did the PP test, i found this warning
>>>>        Warning: n=1, l=0 expected 0 nodes, found 3
>>>>        Setting wfc to zero for this iteration
>>>>
>>>> Is it okay to use this PP for further calculations?
>>>>
>>>> Thank You,
>>>>
>>>> Manu
>>>>
>>>> (University of Waterloo)
>>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Pw_forum mailing listPw_forum at pwscf.orghttp://pwscf.org/mailman/listinfo/pw_forum
>>>
>>> --
>>> Dr. rer. nat. Thomas Brumme
>>> Max Planck Institute for the Structure and Dynamics of Matter
>>> Luruper Chaussee 149
>>> 22761 Hamburg
>>>
>>> Tel:  +49 (0)40 8998 6557
>>>
>>> email: Thomas.Brumme at mpsd.mpg.de
>>>
>>> _______________________________________________ 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
>>> <http://pwscf.org/mailman/listinfo/pw_forum>
>>
>> <http://pwscf.org/mailman/listinfo/pw_forum>
>> _______________________________________________ Pw_forum mailing list
>> Pw_forum at pwscf.org http://pwscf.org/mailman/listinfo/pw_forum
>
> _______________________________________________
> Pw_forum mailing listPw_forum at pwscf.orghttp://pwscf.org/mailman/listinfo/pw_forum
>
> --
> Dr. rer. nat. Thomas Brumme
> Max Planck Institute for the Structure and Dynamics of Matter
> Luruper Chaussee 149
> 22761 Hamburg
>
> Tel:  +49 (0)40 8998 6557
>
> email: Thomas.Brumme at mpsd.mpg.de
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20160830/a402a218/attachment.html>

More information about the users mailing list