[Pw_forum] Atomic rho
In reciprocal space the sum of atomic densities is
sum_rho_at(G) = \sum_ityp S_ityp(G) rho_atomic_ityp(G)
where rho_atomic_ityp(G) is the Fourier trasform of the atomic charge
density for atomic type ityp and S_ityp(G) is the strcuture factor for
that atomic type
S_ityp(G) = \sum_{na such that ityp_of_na=ityp) esp(-i G tau_na) / Omega
If you want a single atom charge density you simply need to use the
espression above with
S_ityp(G) = esp(-i G tau_na_you_want) / Omega
this is actually a periodic charge density corresponding to atom
na_you_want plus its
periodic images ... so the cell must be large enough that these images
do not overlap.
Hope this helps,
stefano
Quoting Leonardo Matheus :
> Thanks for your reply Axel.
>
>> well, for practical reasons it is more efficient to loop over
>> atom types and atoms per atom type. per atom-type parameters
>> only need to be computed once this way.
>
> I understand why the calculation is like this, but still couldn't find
> this "atoms per atom type" loop.
>
>> i suggest you first have a look into textbook material
>> explaining the implementation of electronic structure
>> calculation with plane wave pseudopotentials (e.g. the
>> books by jorge kohnoff or richard martin). and then
>> read the marx / hutter review which also gives an overview
>> on how the implementation can be done (on the example of
>> the CPMD code, which is not that different as far as the
>> basic principles are concerned). the CPMD code actually
>> does have an implementation of hirshfeld charges determination,
>> but only for norm-conserving pseudopotentials, IIRC, so you
>> can look up how the implementation principles.
>
> I've already read the book by Kohanoff, and now I'm waiting for the
> licence to download the CPMD code, to see how Hirshfeld charges are
> implemented there, thanks for the hint. ABINIT code also has Hirshfeld
> charges implemented, I'm looking there too.
>
>> for implementing it into QE you'll have to figure out
>> how to handle ultra-soft psps contribution to the atomic
>> density on top of everything else.
>
> I'm also using the SaX code, which accepts just NC PPs, so I haven't
> thought deeply on this, but I would first try to adapt the "addusdens"
> subroutine to the individual atomic densities (at least, this is what
> I understood the subroutine does).
>
> Actually, I've based most of my code on the "voronoy" program, so what
> I really need now is just to obtain the individual atomic rho, the
> rest (I think) is mostly done.
>
>> good luck,
>> axel.
>
> Thanks
>
> Leonardo Matheus
> Doctorate student - Physics Institute, University of S?o Paulo, Brazil
> Nanomol group - http://nanomol.if.usp.br
> ___
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
SISSA Webmail https://webmail.sissa.it/
Powered by Horde http://www.horde.org/
[Pw_forum] Getting DOS at Fermi Energy
Paul M. Grant wrote:
> BTW, have you had a chance to look
> at the k-points file for aluminum I sent?
it looks like a uniform unshifted 12x12x12 grid in a simple
cubic lattice. You should get it (or an equivalent one)
directly from pw.x using options nosym=.true., noinv=.true.
(the latter is undocumented: removes q=>-q symmetry) and
K_POINT {automatic}
12 12 12 0 0 0
or else, from kpoints.x, answering 'T' to the question
write all k? [f] >>
I am not aware of any limitation to the simple cubic grid,
but this should be asked to the authors of epsilon.x
Paolo
--
Paolo Giannozzi, Democritos and University of Udine, Italy
[Pw_forum] Atomic rho
Thanks for your reply Axel. > well, for practical reasons it is more efficient to loop over > atom types and atoms per atom type. per atom-type parameters > only need to be computed once this way. I understand why the calculation is like this, but still couldn't find this "atoms per atom type" loop. > i suggest you first have a look into textbook material > explaining the implementation of electronic structure > calculation with plane wave pseudopotentials (e.g. the > books by jorge kohnoff or richard martin). and then > read the marx / hutter review which also gives an overview > on how the implementation can be done (on the example of > the CPMD code, which is not that different as far as the > basic principles are concerned). the CPMD code actually > does have an implementation of hirshfeld charges determination, > but only for norm-conserving pseudopotentials, IIRC, so you > can look up how the implementation principles. I've already read the book by Kohanoff, and now I'm waiting for the licence to download the CPMD code, to see how Hirshfeld charges are implemented there, thanks for the hint. ABINIT code also has Hirshfeld charges implemented, I'm looking there too. > for implementing it into QE you'll have to figure out > how to handle ultra-soft psps contribution to the atomic > density on top of everything else. I'm also using the SaX code, which accepts just NC PPs, so I haven't thought deeply on this, but I would first try to adapt the "addusdens" subroutine to the individual atomic densities (at least, this is what I understood the subroutine does). Actually, I've based most of my code on the "voronoy" program, so what I really need now is just to obtain the individual atomic rho, the rest (I think) is mostly done. > good luck, > axel. Thanks Leonardo Matheus Doctorate student - Physics Institute, University of S?o Paulo, Brazil Nanomol group - http://nanomol.if.usp.br
[Pw_forum] scf-convergence
Dear Asa, A few comments on your input file. 1. UsXCrysDen have shown that, presumably, what you study is not a monowire, it looks quite strange. Be sure that you have correct structure. 2. You use UltraSoft pseudopotentials, so ecutwfc should be around 10-12 times of ecutrho. 3. Again, you mean you have body centered orthorhombic structure, but looking at the structure (by XCrysDen) there is no cell center. Bests, Eyvaz. --- Prof. Eyvaz Isaev, Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com --- On Mon, 10/13/08, asa aravindh wrote: > From: asa aravindh > Subject: [Pw_forum] scf-convergence > To: pw_forum at pwscf.org > Date: Monday, October 13, 2008, 3:28 PM > Dear all.. > I am trying to do scf calculation of an fe/co nanowire > along the bcc (110) direction. > However the calculation is not getting converged..after > even 100 iterations.. the scf accuarcy is not reached and > suddenly the program stops ... > can anybody pls tell me if anything is wrong with my input > file... > the input file is attached > any reply would be greatly appreciated > regards, > asa > > > Add more friends to your messenger and enjoy! Go to > http://messenger.yahoo.com/invite/___ > Pw_forum mailing list > Pw_forum at pwscf.org > http://www.democritos.it/mailman/listinfo/pw_forum
[Pw_forum] possible bug in scatter_forw.f90
Dear Alexader,
Thank you very much for your reply. I was out of town and could not see
the message till yesterday. I have some questions on it, it would be great
if you could answer them.
On Fri, 10 Oct 2008, Alexander wrote:
> Dear Manoj
>
> On Friday 10 October 2008 05:20, Manoj Srivastava wrote:
> > Any follow ups on the message below? Alexander?
> >
> > Regards,
> > Manoj
> >
> > On Wed, 8 Oct 2008, Manoj Srivastava wrote:
> > > Dear All,
> > > Is any body familiar with the scatter_forw.f90 subroutine of PWCOND? I
> > > think there is a bug in this subroutine at the place where you calculate
> > > intw2, which is z integration of nonlocal wavefunction with beta
> > > function. I have looked up the necessary formulae in the Choi & Ihm's
> > > paper (PRB 59, 2267, Jan 1999). In the paper, nonlocal wavefunction has 3
> > > terms, one of which contains beta function, say W. The other two parts
> > > dont have any W in them, rather they are just plane wave solution.
> > > (Please have a look at equation 24 and 26 of the paper ). So, when we are
> > > doing z integration of nonlocal wavefunction with beta function W, we
> > > should have three terms, one of which should contain two W, but rest
> > > should just have one W. On the other hand in the code all three terms
> > > contain two beta functions!! (please have a look at line 228 of
> > > scatter_forw.f90). I am wondering if my understanding is right?
>
> Your understanding is wrong. On the line 228 you add to the integral ONLY
> contribution with two beta functions (from the 1st term in the nonlocal
> function, see Eq. 24) but on the line 393 the contribution from the 2nd term
> is added too. There is no 3rd term since every step you rotate your local
> solutions \psi_n in such a way that b_{\lambda \alpha lm} vanish (see the
> paragraph after Eq. 37).
I believe that the equations 40, 41 and 42 implemented in the code are
in the momentum space. In the code, intw2 has 3 terms. We can tell this by
looking at the subroutine integrals.f90, where int2d is defined. Now, by
looking at the choi& Ihm 's paper, the very first term of psi{alpha,l,m}
has f{lambda,alpha,l,m}(equation26).And, if we substitute this in equation
40, we should just get one term. I am just confuesd on the fact that how
come this one term in paper gets transformed into 3 terms of the code!
Would you mind explaining? Also, the integrals that are defined in the
code are only in the slabs (nz1), so how are you taking the integration in
the region (0,d), which we need according to equations 40, 41 and 42 ?
>
> > > I have one more question in the later part of the same subroutine. What
> > > does the lapack subroutine
> > > ZGESV(2*n2d,2*n2d+norb*npol,amat,2*n2d,ipiv,xmat,2*n2d,info)
> > > do?
> > > I tried to look it up and i got the idea that it is trying to solve
> > > amat*x=xmat, with amat and xmat known and x unknown, and at the end of
> > > calculation it stores x in xmat.
> > > so, basically it does-- x=[(amat)^(-1)]*xmat. Am i right? So, it changes
> > > the structure of xmat from what is defined in 'constructs matrices'
> > > part. Is it correct?
>
> Here, you are right.
>
> > > Also, afterwards in this code where it 'rotates integrals' is not very
> > > clear to me.
> > > Could somebody please tell me in little detail, what is going on
> > > here?
> As I explained before rotation means that every step (at each slab)
> you make a linear combination of local solutions \psi_n with the
> transformation matrix h_{nn'} (see the last paragraph on the p. 2270).
> so that the new solutions have the property:
> b_{\lambda n} = \delta_{\lambda n} and the new nonlocal solutions have
> b_{\lambda \alpha lm}=0. Doing this transformation of local and nonlocal
> solutions you should also perform the corresponding transformations
> of the integrals.
>
> > > Also, is this subroutine written just on the basis of Choi and Ihm
> > > paper, or are there more reference to it? If yes, would someone mind
> > > mentioning them?
> Unfortunately, the paper of Choi and Ihm is very detail, and in the
> PWCOND code we followed it very closely, so no more references
> are needed. In our papers we just extended those ideas to ultrasoft
> pseudopotentials, magnetism, spin-orbit coupling and so on.
>
> Hope this helped you,
> regards, Alexander
>
> > >
> > > Regards,
> > > Manoj Srivastava
> > > Physics Graduate Student
> > > University of Florida,
> > > Gainesville,FL, USA
> >
> > ___
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://www.democritos.it/mailman/listinfo/pw_forum
> ___
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
Once again thank you very much for your help. I appreciate it.
Regards,
Manoj Srivastava
[Pw_forum] question about output of SCF
I have done a scf calculation of ground state Pbnm CaCrO3 crystal , the total band nbnd=101 there are 152 irreducible k-points ,but the output file in X.save directory only contians 76 folders : K1-K00076, and the charge-desity.xml only contains to , what is the relation between charge desity grid, and the number of K000XX folders. I also do not know why we define Main grid and wavefunction grid respectively ? Since the grid number are determined by the same ecutwfc , could someone explain more clearly about the all the grids system (include FFT grid, and charge density grid in real or G-space) in DFT ? Many thanks -- Sincerely Y. M. Qian Lab.of Condensed Matter Theory and Materials Computation Institute of Physics Chinese Academy of Sciences Tel: + 8610 8264 9147 E-Mail:yuminqian at gmail.com <E-Mail%3Ayuminqian at gmail.com> P.O.Box 603 Beijing 100190 China -- next part -- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20081014/5ba1febb/attachment.htm
[Pw_forum] Bug in axsf generation?
On Fri, 2008-10-10 at 08:23 -0600, Jan Beck wrote: > I have an axsf file that I have generated with quantum espresso's cppp. I > have checked with the specification of the axsf file format, and the > output from quantum espresso is consistent with that. > WARNING: Atom 2 and atom 6 overlap !!! > Atom 6 deleted !!! > WARNING: Atom 2 and atom 10 overlap !!! > Atom 10 deleted !!! I have looked at your file. It is nothing wrong with its syntax, but the structure is a complete mass: there are many overlapping atoms (look into the file). Therefore you receive the atom-deleted warnings! Either your original cp.x file already contains massy structure, or the cppp utility has a bug (i never used the latter). Regards, Tone -- Tone Kokalj J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia (tel: +386-1-477-3523 // fax:+386-1-477-3822)
[Pw_forum] k-point grids...and all that
>Finally, on the point of mesh visualization. I use XCrySDen a lot, >especially for BZ and charge-potential visualizations. But I don't know how >to configure or generate a .xsf file for a 3D mesh of discrete points. W.r.t. visualization of k-points, if the coordinates of k-points and the boundary of the BZ are known, they can be visualized in VMD with the .xyz file format. Below is an example of the 10 fcc special k-points (blue spheres) in the irreducible wedge (white spheres). http://picasaweb.google.com/deyu.lu/Misc#5257056499972518002 Best Deyu Lu *** Deyu Lu (Ph.D) 190 Chemistry Building University of California, Davis One Shields Avenue Davis, CA 95616 Office phone: (530) 754-9663 Group Webpage: http://angstrom.ucdavis.edu/ ***
[Pw_forum] Getting DOS at Fermi Energy
Paul M. Grant wrote: > Look in the PH directory and not PW or PP? How obvious! it is the only part of the entire distribution where the DOS at E_F is presently used. There is a reason for everything in Q-E. Sometimes it is not a good reason, or it is an obsolete reason, or nobody knows any longer what the reason was, but there is always a reason... > However, thanks for pointing to the dos_ef function in elphon.f90 > (which also needs the external routine w0gauss.f90) it is in PW/w0gauss.f90 and is used in many other places > I think my main challenge will be to figure out how to extract > the arrays et and wk from the nefarious *.save folders. > Maybe you can point me to some routines where this task > is handled...and for spin polarized states as well? if you start from scratch, it will take a lot of time. I think the minimum energy path here is one of the following: - add a call to dos_ef in the main pw.x code, whenever the Fermi energy is calculated (PW/weights.f90) ; - modify any of the codes in PP/ that read the nefarious .save folders. They typically rebuild the structure of variables as used in pw.x ; - build upon the "qexml" library, as suggested by Axel, but beware: I am not sure it is updated Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy
[Pw_forum] question about output of SCF
yumin qian wrote: > I have done a scf calculation of ground state Pbnm CaCrO3 crystal , the > total band nbnd=101 > there are 152 irreducible k-points ,but the output file in X.save > directory only contians 76 folders : K1-K00076 so there are 76 irreducible k-points, and since you are using LSDA, the code doubles the number of k-points during the calculation, so that you have 76 k-points for spin-up states and 76 for spin-down. At the end, when results are saved, spin-up and spin-down states are collected into a single folder. > I also do not know why we define Main grid and wavefunction grid > respectively ? Since the grid number are determined by the same > ecutwfc not for USPP or PAW > , could someone explain more clearly about the all the grids system > (include FFT grid, and charge density grid in real or G-space) in DFT ? the best explanations I managed to produce are reported in the 3rd paper listed here: http://www.quantum-espresso.org/wiki/index.php/Ref:_First-principles_Molecular_Dynamics You can get a copy here: http://www.fisica.uniud.it/~giannozz/Papers/JCP_57.pdf Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy
[Pw_forum] Getting DOS at Fermi Energy
..."there is always a reason"...sounds like some sort of creationist mantra to me! Thanks for the insight into pw.x...maybe in a few light-years I'll be able to contribute to the cause. I'll have a look at method you suggest, but last night I think I've found a straightforward (relatively) way to hack epsilon.f90 with a dos_ef like option. Let you know how it turns out. BTW, have you had a chance to look at the k-points file for aluminum I sent? Paul M. Grant, PhD Principal, W2AGZ Technologies Visiting Scholar, Applied Physics, Stanford (2005-2008) EPRI Science Fellow (Retired) IBM Research Staff Member Emeritus w2agz at pacbell.net http://www.w2agz.com ? ? -Original Message- From: pw_forum-bounces at pwscf.org [mailto:pw_forum-boun...@pwscf.org] On Behalf Of Paolo Giannozzi Sent: Tuesday, October 14, 2008 12:25 AM To: PWSCF Forum Subject: Re: [Pw_forum] Getting DOS at Fermi Energy Paul M. Grant wrote: > Look in the PH directory and not PW or PP? How obvious! it is the only part of the entire distribution where the DOS at E_F is presently used. There is a reason for everything in Q-E. Sometimes it is not a good reason, or it is an obsolete reason, or nobody knows any longer what the reason was, but there is always a reason... > However, thanks for pointing to the dos_ef function in elphon.f90 > (which also needs the external routine w0gauss.f90) it is in PW/w0gauss.f90 and is used in many other places > I think my main challenge will be to figure out how to extract > the arrays et and wk from the nefarious *.save folders. > Maybe you can point me to some routines where this task > is handled...and for spin polarized states as well? if you start from scratch, it will take a lot of time. I think the minimum energy path here is one of the following: - add a call to dos_ef in the main pw.x code, whenever the Fermi energy is calculated (PW/weights.f90) ; - modify any of the codes in PP/ that read the nefarious .save folders. They typically rebuild the structure of variables as used in pw.x ; - build upon the "qexml" library, as suggested by Axel, but beware: I am not sure it is updated Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy ___ Pw_forum mailing list Pw_forum at pwscf.org http://www.democritos.it/mailman/listinfo/pw_forum
[Pw_forum] Bug in axsf generation?
Jan Beck wrote: > this is my third attempt at posting this reply, I cannot paste the > .axsf file or attach it, because its too large. The message bounces > back to me with a "moderator approval pending" message. I think > approval is still pending apparently you didn't notice the "Important notice" here: http://www.quantum-espresso.org/wiki/index.php/Introduction#Contacts Executive summary: --- Messages "waiting for moderator approval" are automatically deleted with no further processing (sorry, too much spam) --- > so I apologize for making you download the > file from a 3rd party site: > > http://rapidshare.com/files/153722255/data.axsf.html you shouldn't apologize: this is the correct way to send large files, in my opinion. Paolo -- Paolo Giannozzi, Democritos and University of Udine, Italy
