[DFTB-Plus-User] (no subject)

[Jacek Jakowski]

Anjali,

Are you sure that you SK parameters belong all tot the same set? Mixing SK
parameters from different sets  can lead to this.

If everything  looks fine then you should  dump your overlap matrix and
diagonalize it.  There will be one or two negative eigenvalues.  You should
then look   corresponding eigenvectors and find out which atoms/basis  lead
to this  "non-positive  overlap".  Probably there  will be two or three
 atoms that contribute most.  Then you  can manually move a little but
those atoms to tilt them off from  "non-positive overlap" position, while
retaining  all  other atoms in original location.  You can then restart
your optmization. I bet the atoms you need to move are very close to each
other.

Jacek


On Thu, Oct 18, 2012 at 11:50 AM, Anjali Singh <anjalisinghromi at gmail.com>wrote:

> Dear Bikash
>         I can see geometry, but nothing is like that.
>
>
> On Thu, Oct 18, 2012 at 6:15 PM, bikash sankar kanungo <
> biku.kanungo at gmail.com> wrote:
>
>> Hi,
>>
>> Can you visualize the final geometry ? I suppose there must be some atoms
>> sitting on top of each other. If thats the case then the initial geometry
>> or boundary conditions seems to be faulty.
>>
>> Thank you,
>> Bikash
>>
>> On Thu, Oct 18, 2012 at 5:27 PM, Anjali Singh <anjalisinghromi at gmail.com>wrote:
>>
>>> Dear all,
>>>             I want help, I am getting following error when trying to
>>> relax cluster. Error is given below:
>>>
>>>   ERROR!
>>> -> Failure in diagonalisation routine dsygvd, non-positive definite
>>> overlap! Minor     89 responsible.
>>> -- INS
>>>  I am giving input file also. It is showing problem in convergence of
>>> forces also. Force component is not going below 1e-2 hartree/bohr .
>>>
>>> Geometry = GenFormat {
>>>     <<< "input.gen"
>>>     }
>>> Driver = ConjugateGradient {
>>>     MovedAtoms = 1:-1
>>>     MaxSteps = 500
>>>     ConvergentForcesOnly = No
>>>     #LatticeOpt= Yes
>>>     OutputPrefix = "geom.out"
>>>     AppendGeometries = Yes
>>>     }
>>> Hamiltonian = DFTB {
>>>     SCC = Yes
>>>     SCCTolerance = 1e-004
>>>     MaxSCCIterations = 1000
>>>     ReadInitialCharges = Yes
>>>     Mixer = Broyden{
>>>         MixingParameter = 0.1
>>>     }
>>>     SlaterKosterFiles = Type2FileNames {
>>>         Prefix = "./S_Kfiles/"
>>>         Separator = "-"
>>>         Suffix = ".skf"
>>>         LowerCaseTypeName = No
>>>     }
>>>     MaxAngularMomentum {
>>>         B  = "p"
>>>         N  = "p"
>>>     }
>>>     Filling = Fermi {
>>>         Temperature [Kelvin] = 10
>>>     }
>>>      }
>>>     Options {
>>>    # CalculateForces = Yes
>>>     WriteEigenVectors = Yes
>>>     WriteDetailedXML = Yes
>>>     WriteResultsTag = Yes
>>>     AtomResolvedEnergies = Yes
>>>     }
>>> ParserOptions {
>>>    ParserVersion = 4
>>> }
>>>
>>> Can somebody help me ?
>>>
>>> --
>>> With Thanks and Regards
>>> Anjali Singh
>>> Material Theory Group
>>> Theoretical Science Unit
>>> Jawaharlal Nehru Centre for Advanced Scientific Research
>>> Jakkur,Bangalore 560 064
>>> India
>>> email id : anjalisinghromi at gmail.com
>>> phone no:+91-89703-58107
>>>
>>>
>>> _______________________________________________
>>> DFTB-Plus-User mailing list
>>> DFTB-Plus-User at dftb-plus.info
>>> http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user
>>>
>>>
>>
>>
>> --
>> BIKASH SANKAR KANUNGO
>> Final Year Undergraduate student,
>> Mechanical Engineering Department,
>> INDIAN INSTITUTE OF TECHNOLOGY
>> KHARAGPUR.
>> +919749935409
>>
>>
>> _______________________________________________
>> DFTB-Plus-User mailing list
>> DFTB-Plus-User at dftb-plus.info
>> http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user
>>
>>
>
>
> --
> With Thanks and Regards
> Anjali Singh
> Material Theory Group
> Theoretical Science Unit
> Jawaharlal Nehru Centre for Advanced Scientific Research
> Jakkur,Bangalore 560 064
> India
> email id : anjalisinghromi at gmail.com
> phone no:+91-89703-58107
>
>
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at dftb-plus.info
> http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.dftb-plus.info/pipermail/dftb-plus-user/attachments/20121018/7aa91dce/attachment.html>