Hi, Bin,

This subroutine is called by other subroutines for basis generation,
and there is not an independent executable file. I guess you may mean
the compiled obj file, but the compilation is compiler- or
system-dependent. If you can see the output message of LDAUprojgen in
the example of Fe, your compilation of ldauproj_gen.f90 should be
fine.

I haven't tested ldau parameters for Ni. You may already have this
reference: S L Dudarev et al, Electron-energy-loss spectra and the
structural stability of nickel oxide: an LSDU+U study, PRB 57, 1505
(1998).

Cheers,
Haibo

On Tue, Sep 14, 2010 at 5:40 PM, Bin Shao <[email protected]> wrote:
> Hi, Haibo,
>
> Could you give a list of the files about ldauproj_gen in your Src fold? or
> is there a execute file naming ldauproj_gen? I am afraid some wrong happens
> in my compiling process.
>
> On Tue, Sep 14, 2010 at 2:37 PM, Haibo Guo <[email protected]> wrote:
>>
>> Hi, Bin,
>>
>> It seems that the subroutine ldauproj_gen (in Src/ldauproj_gen.f90)
>> did not run well. If the subroutine runs to end, the output should
>> have some messages like the following.
>>
>> ...
>> LDAUprojgen begin
>>
>> LDAUprojs with angular momentum L= 2
>> LDAUproj generation method   2
>>
>> LDAUproj corresponding to state 3d
>>  radial_log schro: updating the rc to:   19.9234748429000
>> LDAUproj is an extended PAO orbital cut off with a
>> Fermi function 1/[1+exp(r-rc)/w] with
>>  rc=    2.300000
>>  w =    0.150000
>> LDAUproj cutoff radious     2.979846
>>
>> LDAUprojgen end
>> ...
>>
>> You may want to do some debug to see what's going wrong with the
>> execution of this subroutine if you have time. I am not sure whether
>> it helps resolve the problem if you specify the cutoff radii
>> explicitly (not using 0.0 to let the program choose default ones), but
>> here are my inputs for Fe. Hope this working example helps.
>>
>> LDAU.ProjectorGenerationMethod   2
>> LDAU.CutoffNorm                  0.9
>> %block LDAU.proj
>> Fe 1
>>  n=3 2 E 50.0 2.5
>>  5.5  1.0
>>  2.3  0.15
>>  1.0
>> %endblock LDAU.proj
>> LDAU.FirstIteration  false
>> LDAU.ThresholdTol    1.0E-2
>> LDAU.PopTol          1.0E-3
>> LDAU.PotentialShift  false
>>
>>
>> Cheers,
>> Haibo
>>
>>
>> On Tue, Sep 14, 2010 at 1:10 PM, Bin Shao <[email protected]> wrote:
>> > Hi, Haibo,
>> >
>> > Thank you very much for your reply!
>> >
>> > The attachment is my input and output files with the effective U = 1.95
>> > eV.
>> > And you'll find the Eldau is zero.
>> >
>> > On Tue, Sep 14, 2010 at 10:44 AM, Haibo Guo <[email protected]> wrote:
>> >>
>> >> Hi, Bin,
>> >>
>> >> The input file seems fine to me. You have switched on both LDAU and
>> >> spin polarization. My calculations with LDAU have a non-zero Eldau.
>> >> Can you show us your output file?
>> >>
>> >> cheers,
>> >> Haibo
>> >>
>> >>
>> >> On Mon, Sep 13, 2010 at 6:49 PM, Bin Shao <[email protected]>
>> >> wrote:
>> >> > Dear, Haibo,
>> >> >
>> >> > Thank you for your reply.
>> >> >
>> >> > Actually, there is no error or warning messages in my computation.
>> >> > But I
>> >> > note that the job can be finished in serial mode, maybe the Fe system
>> >> > is
>> >> > too
>> >> > small, some error happens in the process of allocating the memory.
>> >> >
>> >> > Indeed, I want to using the LDA+U in NiO systems, the attachment is
>> >> > my
>> >> > input
>> >> > file. The job can run successfully, but the result shows no
>> >> > difference
>> >> > between the system plus U and without U. Moreover, the Eldau is 0 in
>> >> > the
>> >> > output file.
>> >> >
>> >> > Could you give me some suggestion about my computation setting,
>> >> > especially,
>> >> > about the LDAU. Thank you in advance!
>> >> >
>> >> > Best,
>> >> >
>> >> > On Mon, Sep 13, 2010 at 1:03 PM, Haibo Guo <[email protected]>
>> >> > wrote:
>> >> >>
>> >> >> Hi, Bin Shao,
>> >> >>
>> >> >> There seem to be no error or warning messages shown in your email.
>> >> >> After a quick scan to your input file, I found the settings of basis
>> >> >> vectors may need your attention.
>> >> >>
>> >> >> > %block LatticeVectors
>> >> >> >  0.50000   0.500000  0.500000
>> >> >> >  0.50000  -0.500000  0.500000
>> >> >> >  0.50000   0.500000 -0.500000
>> >> >> > %endblock LatticeVectors
>> >> >>
>> >> >> But this may not be relevant to the unexpected stop of SIESTA.
>> >> >>
>> >> >> Cheers,
>> >> >> Haibo
>> >> >
>> >> >
>> >> >
>> >> > --
>> >> > Bin Shao, Ph.D. Candidate
>> >> > College of Information Technical Science, Nankai University
>> >> > 94 Weijin Rd. Nankai Dist. Tianjin 300071, China
>> >> > Email: [email protected]
>> >> >
>> >
>> >
>> >
>> > --
>> > Bin Shao, Ph.D. Candidate
>> > College of Information Technical Science, Nankai University
>> > 94 Weijin Rd. Nankai Dist. Tianjin 300071, China
>> > Email: [email protected]
>> >
>
>
>
> --
> Bin Shao, Ph.D. Candidate
> College of Information Technical Science, Nankai University
> 94 Weijin Rd. Nankai Dist. Tianjin 300071, China
> Email: [email protected]
>

Responder a