If it helps, and to compelement Nicolas' e-mail, I recall seeing a paper by Wei
Cai and Giulia Galli on the subject of finite magnetic fields and DFT with
periodic boundary conditions:
http://prl.aps.org/abstract/PRL/v92/i18/e186402
Ab Initio Calculations in a Uniform Magnetic Field Using Per
Dear list,
I am running the most simple calculation to obtain the lattice constant for
Pt. I am using the LDA pseudo from ABINIT, my calculation is spin unpolarized
for the time being.
I am getting the error shown as the title of this e-mail. The basis is the
standard DZP. I looked into the
Hi Janakiraman,
In transiesta the z-direction is that of charge transport. In other words it
is non-periodic. Therefore, when transiesta runs it will override the user's
choice of k-points in this direction. If you pick k-points along the x and y
directions transiesta should keep them th
Hi Janakiraman ,
The problem you have has many faces to it, in a literal sense.
The unit cell is the most symmetric way to place the atoms in the FCC cell, but
you could imagine yourself slicing your bulk in such a way that you get more
than one atom. You can slice it in whatever angle you wa
Hi Magda,
For a given k-point, the bands file will have spin up and spin down entries if
the run was spin-polarized. To see where the spin down entries start, look for
a sudden jump in the eigenvalues (ie, from large positive values back to
negative ones) right at the middle of the dataset for
Hi Andrea. I thought from your error message you are using the intel fotran
compiler. Can you indicate which version/compiler you are using?
I think you need to get into you arch.make file and place the (similar)
debugging options in there; all compilers have an option to dump the sequence
of
Hi,
The segmentation fault occurs whenever you are exceeding the limits of
fixed-size arrays. Predefined sizes of arrays might have changed in between
versions of the code. Can you recompile your code with the debug options:
-g -O0 -traceback
in order to determine the offendig line in your c
iceVectors (or pi/a as you wish)
%block WaveFuncKPoints
0.000 0.000 0.000 from 1 to 2 # Gamma point wavefunctions 1 to 2
% endblock WaveFuncKPoints
And these are enough.
I am using siesta-3.0-rc2，and go well with the settting above.
Best.
Guangping Zhang
发件人 : "
Hi Carlo,
so what did you do?
Was it just an error in your input file?
I am facing a similar problem; I beleive my input file is correct. Can you
share the info on how you solved the problem?
Thanks,
-Salvador.
- Original Message -
From: "Carlo Motta"
To: siesta-l@uam.es
Sent: T
Hi Nguyen,
I would say if you want to get something out of your PDOS file you have to
mine the data yourself. Below I am sharing a script I generated in matlab; it's
got a number of lines hardcoded (so it could be better coded for a general
purpose) but it does what I need. You would need t
GMT -05:00 US/Canada Eastern
Subject: Re: [SIESTA-L] Error in FoX
You can add " writeXML F" to avoild this error
On Wed, Oct 13, 2010 at 4:21 AM, Barraza-lopez, Salvador < s...@mail. gatech .
edu > wrote:
Hi,
I am compiling SIESTA 3.0.b on ember, a new computer at NCSA . I a
Hi,
I am compiling SIESTA 3.0.b on ember, a new computer at NCSA. I am getting the
following error at FoX:
[s...@ember Efield1]$ more r78bilayer1.e772
set_SCR: using existing PBS job directory /scratch/batch/772
ERROR(FoX)
xml_AddCharacters: Invalid character in chars
MPI: MPI_COMM_WORLD rank
Hi,
If you have siesta 3.0, you can use DENCHAR to plot them. It is found in the
Util directory. Denchar will produce a number of files for you: Norm and phase
of wavefunctions, real and imaginary parts, and so on... it works for any
k-point.
Once you run denchar, you can plot them using any
Hi Diego,
The error comes from the undirected input/output; this is why you get the
following line:
reinit: Reading from standard input
In other words, the program is *not* reading your input file and attempts to
read an empty INPUT_DEBUG file.
Try running your job the following way:
mpir
Hi Alexander,
Thank you for your guidance. I am looking at some emails from 2007, presumably
on siesta 2.0. The three emails were between Marcos Verisimo Alves, Yurko
Natanzon and Heribert Reis. I don't see a solution on that thread. That's all
that's said apparently regarding this issue.
I
Dear users and developers,
I want to comment on a sudden memory increase we observe in slab calculations
with electric field. We are running calculations on a slab with 800 atoms, on a
~30x30x20 Ang^3 unit cell, so it is decidedly large. The energy cutoff is 300
Ry.
The memory used is about
La verdad no se, supongo que es simple interpolacion como tu dices. En lo que
alguien con mas conocimiento te contesta, si tienes un archivo que vibra
produce con extension VIB (por decir algo) entonces puedes buscar la rutina en
fortran que te genera dicho archivo de la siguiente manera:
grep
Dear Dr. Shalabi,
I agree with the previous message from Dr. Alan. We are sorry you are facing
difficulties. But we, as users, don't need to know about the troubles you are
facing while trying to get the package.
Now, here's a faithful translation of the message you received:
"Your message
Zhen (Alex) Huang
Ph.D. Student
Nanoscale Transport Research Group
Laboratory for Computational Methods in Emerging Technologies
Cooling Technologies Research Center
School of Mechanical Engineering
Purdue University
Tel: 765 237 9733
On Wed, Mar 3, 2010 at 10:17 AM, Barraza-lopez, Sa
pseudos are often tried and tested, even though it is imperative to
test every pseudo in simple cases before going for the real thing.
Just my 10 cents on the subject.
Cheers,
Marcos
On Wed, Mar 3, 2010 at 3:21 PM, Barraza-lopez, Salvador
wrote:
>
> Dear Zhen,
> I would recommend tha
logies Research Center
School of Mechanical Engineering
Purdue University
Tel: 765 237 9733
On Mon, Mar 1, 2010 at 10:14 AM, Barraza-lopez, Salvador < s...@mail.gatech.edu
> wrote:
Hi all,
I've been following this post and want to mention something else that has not
c
Hi all,
I've been following this post and want to mention something else that has not
come out yet. Ti has in the vale nce two d and two s electrons. d-electrons are
responsible for mag net ic behavior. So, I'd like to ask: Have you taken your
Ti pseudo (is it LDA, GGA pseudo?) and ran a
test
So you mean the 6p orbit is not appropriate to be used in the
pseudopotential generation?
On Thu, 25 Feb 2010 11:33:32 -0500 (EST), "Barraza-lopez, Salvador"
wrote:
> Well, the promotion of electrons has as it underlying motivation the
change
> in occupations in real materia
Well, the promotion of electrons has as it underlying motivation the change in
occupations in real materials. I think some of the occupation changes in your
list, such as 6s1 5d5 6p5, would be unphysical... would not be surprised if in
those cases you face an error. Remember that in the pseudo
0.4224046.664435
Requested rc inside node ! ** rzero, rextr, rc_input: 0.260 0.432
0.000
rc changed to 0.415
error in pseud2 - nonconvergence in finding
starting delta for angular momentum 3
stop parameter =824
On Wed, 24 Feb 2010 17:36:46 -0500 (EST), "Barraza-lopez, Salvador&qu
Hi Ma Xiao,
Mind the spacing. You can take a pseudo file from the examples directory and
replace the numbers and labels you want. That way you ensure having a *format*
the ATOM program will understand: ATOM does not have the FDF features that
SIESTA has, so you have to not add a single more n
Hi Akram,
If you are doing a Gamma-point calculation (isolated molecule) [I think you
are], and you know the energy range you are interested in, you can just ask the
program to dump the eigenfunction info. In the standard output it will dump the
amplitude on each basis channel. You can u
Dear Feng,
As Heiko indicates, your in-plane parameters are too large; more than 16
Angstrom. your lattice parameters, as Heiko says, are *multiplied* by the
lattice constant, a choice like
1.0 1.0 4.0 90 90 90
may be more sensible one, that will allow for atomic bonding on the x-y plane.
I
Thank you Teng. This is not what I meant. In principle dk could take any value
I choose, well beyond 0.0 and 0.5. In particular I'd like my choice to be such
that I explicitly include the K point of graphene (notice the uppercase here).
Right now I am obtaining eigenvalues at the K point from a
Dear users,
I am performing a calculation on a graphitic system and would like to include
the K point explicitly into the k-point grid.
Wondering if dk is given in units of pi/a. Your help on this will be greatly
appreciated.
--
Salvador Barraza-Lopez
Postdoctoral Fellow
School of Physics
Th
Hi Javad,
Your error message says that you are trying to deallocate a variable that has
not been dynamically allocated.
I would recommend compiling in debug mode to see the *specific line and
routine* where the deallocate command is called.
Then you should be able to get into the offending
Marcos is right. It seems like at the beginning something is being configured
on the FoX directory, but as long as you first copy your own arch.make that
used to work in previous versions you are fine (are you compiling SIESTA for
the very first time?) .
I compiled 3.0b with my arch.make
Dear Jeffei,
I a normal calculation, you would want the Hartree to *match* between the
leads and the scattering region. Since the potential is defined up to a
constant, this matching is done *by hand*, by using some appropriate flags
(don't know them by heart). So that at a given location (a m
contributions
of each atom in the given layer and plot them?
Cheers,
Marcos
On Sat, Oct 24, 2009 at 5:45 PM, Barraza-lopez, Salvador
wrote:
>
> Hello,
>
> I would like to color a band structure plot according to the atomic species
> contributing the most to a given band: I
Hello,
I would like to color a band structure plot according to the atomic species
contributing the most to a given band: I would like to decompose the band
structure according to the atomic species in the system. The system is a
layered structure, so it makes sense to think of a given layer
35 matches
Mail list logo
