[DFTB-Plus-User] MD in liquid water
Christopher Mark Maupin
cmaupin at hec.utah.edu
Wed Oct 7 17:20:07 CEST 2009
Paulo,
I agree with Ben that having N terms in the file may be an issue. I
took you file and striped out all the carbons leaving just the water. I
then ran the code (with the H and O *.skf files) and the system did not
converge. I then noticed you had your Filling = Fermi {
Temperature[Kelvin] = 0}. I changed this this to be in agreement with
the system temp of 300.0 K. The system then converged in 28 cycles the
first step, and 10 - 12 cycles for the next few steps.
When I had question about this issue it was suggested that I use the
same electronic temperature as MD temp in addition to
AdaptFillingTemperature = Yes.
I sure hope this helps
Best wishes,
Mark
> Hello Paulo,
>
> the problem seems to be set of by the nitrogen atoms in the
> input. Your structure doesn't actually have any N atoms, but there are
> several places where there are settings for these, which when commented
> out/removed should cure the problem:
>
> TypeNames = {"O" "H" "C"}# "N"}
>
> # N-H = "N-H.skf"
> # H-N = "H-N.skf"
> # C-N = "C-N.skf"
> # N-C = "C-N.skf"
> # N-N = "N-N.skf"
> # O-N = "O-N.skf"
> # N-O = "O-N.skf"
>
> # N = "p"
>
> it looks like you have found a bug in the parsing of the input in
> this case.
>
> Regards
>
> Ben
>
> Paulo Cesar Piquini wrote:
>
>> Dear Mark,
>>
>> attached you will see the input and the output files of one of my tests.
>> I played with some variables as timestep and others but the result was
>> similar or worst.
>> As you will see, the SCC cycle does not achieve convergence. If one
>> increases the number of iterations, it goes well until some cycle and
>> then the problem appears again.
>> Thanks
>>
>> Yours
>> Paulo
>>
>> On Tue, Oct 6, 2009 at 12:19 PM, Christopher Mark Maupin
>> <cmaupin at hec.utah.edu <mailto:cmaupin at hec.utah.edu>> wrote:
>>
>> DEar Paulo Cesar Piquini,
>> I have conducted several liquid water simulations and have not seen
>> this issue. Can you please attach your input file and the output file.
>>
>> Best wishes,
>>
>> Mark
>>
>> Paulo Cesar Piquini wrote:
>> > Dear users of DFTB+
>> >
>> > I am a new user of DFTB+ and Iam having difficulties to make the SCC
>> > cycle to converge when I simulate a liquid water at 300 K.
>> > Have someone already performed similar calculations successfully?
>> >
>> > Thanks
>> > Paulo Piquini
>> >
>> > --
>> > Paulo Cesar Piquini
>> > Departamento de Física
>> > Universidade Federal de Santa Maria
>> > 97105-900
>> > Santa Maria, RS
>> > Brasil
>> > Fax: 55 55 3220-8032
>> > e-mail: ppiquini at gmail.com <mailto:ppiquini at gmail.com>
>> <mailto:ppiquini at gmail.com <mailto:ppiquini at gmail.com>>;
>> > ppiquini at smail.ufsm.br <mailto:ppiquini at smail.ufsm.br>
>> <mailto:ppiquini at smail.ufsm.br <mailto:ppiquini at smail.ufsm.br>>
>> > http://w3.ufsm.br/piquini
>> >
>> >
>> ------------------------------------------------------------------------
>> >
>> > _______________________________________________
>> > DFTB-Plus-User mailing list
>> > DFTB-Plus-User at dftb-plus.info <mailto:DFTB-Plus-User at dftb-plus.info>
>> > http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user
>> >
>>
>> _______________________________________________
>> DFTB-Plus-User mailing list
>> DFTB-Plus-User at dftb-plus.info <mailto:DFTB-Plus-User at dftb-plus.info>
>> http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user
>>
>>
>>
>>
>> --
>> Paulo Cesar Piquini
>> Departamento de Física
>> Universidade Federal de Santa Maria
>> 97105-900
>> Santa Maria, RS
>> Brasil
>> Fax: 55 55 3220-8032
>> e-mail: ppiquini at gmail.com <mailto:ppiquini at gmail.com>;
>> ppiquini at smail.ufsm.br <mailto:ppiquini at smail.ufsm.br>
>> http://w3.ufsm.br/piquini
>>
>>
>
>
More information about the DFTB-Plus-User
mailing list