The ABINIT Users Mailing List ( CLOSED )

[you peng <pengyoucool@yahoo.com.cn>]

Dear Fabien,
        Thank you for your attention.But I'm sorry to unapprehend your reply.I have calculated the band of my compound with full potential in LAPW method.It is a semiconductor,not metallic. I think there should not be
any difference between full potential method and the pseudo potential method about the special k points in the gap.So I work with GW correction in the gap directly. If I'wrong ,any comments will be appreciated.
        What'more,I don't know how the two parameters "occopt" and "tsmear" work for their purposes.I didn't find out any informations about them in the input file.
        Appreciate your help in advance.
 
                                                                                                    Peng
Fabien Bruneval <fabien.bruneval@polytechnique.fr> 写道：

Dear Peng,

There might be a bunch of errors in your input file. According to your
input file, you want to calculate PbTe, but with some huge displacements
with respect to the usual rock salt structure.
Before starting ANY GW run, you should first know what is the
DFT-Kohn-Sham band structure. First of all, is your compound metallic or
semiconducting within DFT?
I would guess that your system is metallic within DFT. Unfortunately,
your input file does not include any smearing on the electronic occupancies.
So please consult the input variables:
occopt
tsmear

and then try to figure out what is the DFT band structure. After and
only after, you might consider to run GW calculations.


Fabien


PS: please do a convergence study on your cut-off...




you peng wrote:
> Dear Matteo,
> First,thank you for your help.I have known the meaning of "istwfk"
> from you.
> But there is also error in the abinit process.I don't know what's
> wrong with nband or nkptgw. The case I payed attention to is indirect
> band_gap. I want to calculate the gap corrections in the GW
> approximation.So I put the two special points in kptgw3.The error
> written in the log file is the same as what is in log file that I have
> put in my last letter. Here,I append my input file.I don't know how to
> tackle with the abinit process.It can't go along.Often,it stops after
> seconds,even at once.
> Thank you for giving me so much help.
>
> yours,
>
> Friend
>
> */Matteo Giantomassi /* 写道：
>
> Dear peng,
>
> On Tue, 2007-03-13 at 21:47 +0800, you peng wrote:
> > Dear abinit users,
> > When I used the input file number1 in GW calculation in the
> > tutorial, I encountered some problems. I focus on the orthorhombic
> > cell ,and I have modified the input parameters in tgw_1.in. But after
> > I run abinis,it paused with error. The log file says as follows,
> >
> > DATASET 1 : space group Pn m a (# 62); Bravais oP
> > (primitive ortho.)
> > getkgrid : length of smallest supercell vector (bohr)= 3.390926E
> > +01
> > Simple Lattice Grid
> > symkpt : found identity, with number 1
> >
> > inkpts : istwfk preprocessed, gives following first values (max. 6):
> > 1 1 1 1 1 1
> >
> > chkneu : ERROR -
> > Initialization of occ, with occopt= 1
> > There are not enough bands to get charge balance right
> > Action : modify input file ...
> > (check the pseudopotential charges, the variable charge,
> > and the declared number of bands, nband)
> >
> > But when I increased the nband1,there is also error with
> > "bdgw3".
> > In my case,there are 10 bands occupied.So I set "bdgw3=10 11",and
> > "nband1=21". I don't know what's wrong with my input file.
>
> What is the error message you are receving?
> I cannot be sure because you didn't send the input file, but I think
> that the problem with bdgw is related to the fact that you have to
> specify the starting and ending band index for each k point where you
> are calculating GW corrections.
>
> For example:
>
> nkptgw 2
> kptgw 0.0 0.0 0.0
> 0.0 0.0 0.5
>
> bdgw 10 11
> 10 11
>
>
> > Besides,I am not clear with "istwfk1",although I have looked through
> > the tutorial and input variables. I don't know how to adjust
> "istwfk1"
> > according to specific cases.
>
> istwfk is used for GS calculations to reduce the number of G vectors
> for particular k-points using time-reversal symmetry (like for example
> the Gamma point).
> Unfortunately you cannot take advantage of this symmetry during the
> generation of the KSS file and, as a consequence, you have to force
> abinit to use all the G vectors for each irreducible k-point.
>
> In a nutshell:
>
> istwfk = 1* (number of irred k-points in the KSS file)
>
> If you don't know the number of irreducible k-points associated to the
> k-grid used to generate the KSS file, just run abinit, kill the job
> after few seconds and look for the value of nkpt reported in the output
> file (only the value of nkpt in the DATASET where the KSS file is
> generated)
>
> Once you know the value of nkpt, you can easily modify the value of
> istwfk in the input file
>
> Hope this helps
> Matteo
>
>
> > Look forward to your help. Thank you very much.
> >
> >
> > yours,
> >
> >
> > Friend
> >
> >
> >
> >
> >
> ______________________________________________________________________
> > 雅虎免费邮箱-3.5G容量，20M附件
>
>
> ------------------------------------------------------------------------
> Mp3疯狂搜-新歌热歌高速下
>

---

抢注雅虎免费邮箱-3.5G容量，20M附件！

---

抢注雅虎免费邮箱-3.5G容量，20M附件！