[Wien] STM

[Gavin Abo]

Not an expert on the STM, so I could be wrong about some things, but my 
comments below.

> Still on STM, I have a question regarding case.in3d, please what is 
> the unit of the start and end value in the case.in3d file? For 
> example, I have a test case with a vacuum of about 13 Angstrom.

Section 8.15.3 in the WIEN2k 18.2 usersguide has:

---------------- top of file: case.in3d --------------------
100 100 100 # number of 3D-gridpoints in a-, b- and c-direction
...
stm z 60 100 0.0001 # (optional) stmswitch, direction, start, end, density

------------------- bottom of file ------------------------

Looks to me that the start and end are unitless and are based on the 
gridpoints.

> What should be my start value (value above the surface atom) and end 
> value (value around middle of vacuum)?
Section 8.15.3 in the WIEN2k 18.2 usersguide has:

start specifies the first plane for the search of the given density 
(typically a
value above the surface atom).

end specifies the last plane for the search of the given density (typically
around middle of vacuum).

Sounds like something that can be found by trial and check.  If the 
start-end window is not able to capture the given density, move it or 
make it wider.  If the start-end search window is picking up an 
undesirable density, narrow the window to capture the one you want sort 
of thing.

> Also in the UG page 172, section 18.5 it's mentioned that "*Optionally 
> one can simulate a constant current STM image, where a height profile 
> (above a surface slab) corresponding to a specific electron density 
> value can be extracted from the case.xsf-file and written into case 
> stm.xsf-file*." Please the extraction of the electron density is it 
> automatic or it should be done manually, and which file case.xsf or 
> case_stm.xsf should be use for the plotting?

Section 8.15 3DDENS (3D electron density plots in whole cell) in the UG has:

This program generates the charge density (or the potential) in the 
whole conventional (or primitive)
unit cell on a three dimensional grid which can be used for plotting 
with an external program
that can read .xsf-files (e.g., XCrysden, VESTA).

As the above says, the charge density case.xsf file should be generated 
automatically.

Section 8.15.3 in the WIEN2k 18.2 usersguide has:

stmswitch case_stm.xsf file is written, containing height values for the 
given density
value.

As the above says, the stm image case_stm.xsf file should be generated 
automatically if the stmswitch parameter is set to "stm".

So if want the charge density plot, open case.xsf (in XCrySDen). If you 
want the STM image plot, open the case_stm.xsf.

If you use lapw5's STM option instead of 3ddens, I'm not sure, but I'm 
thinking it outputs a case.rho file for the stm image data.  The rho 
file can be plotted with different programs [ 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg07125.html 
].

You might have a look at the file SRC_lapw5/README_STM.docx. That 
mentions it will also output a file in SEMPER format that can be read by 
the semper7 code.  This is probably the semper-7.0beta image processing 
tool mentioned on Prof. Marks website [ 
http://www.numis.northwestern.edu/Software/Software.shtml ].


> On Friday, December 7, 2018, 3:20:56 PM GMT+8, Peter Blaha 
> <pblaha at theochem.tuwien.ac.at> wrote:
>
>
> STM simulations make only sense for a surface. You cannot make STM for a
> bulk ???
>
> 3ddens does not include tip vibrations, but is very fast and
> conveniently lets you choose the best hight above the surface to
> reproduce experimental images.
>
> Be careful with the STM mode of lapw5 in cases with angles differing
> from 90 degrees !
>
> STM simulations are always a bit of an art, because the quality of the
> experimental STM images changes a lot with group / time / ....
>
>
> Am 07.12.2018 um 03:32 schrieb Lawal Mohammed:
> > Dear Developers and Users,
> >
> > I need some advice on using the 3ddens program. My interest is to
> > simulate STM image of a 3D system to compare with our experimental 
> results.
> > My questions are:
> >
> > Whether it's possible for 3D system.
> > What determine the size of the grid points?
> > Do I have to edit any file in my case directory before x 3ddens -XX?
> >
> > I want to test run using TiC, what are the general tips? Any hint is
> > highly appreciated.
> >
> > Thanks a lot for your time.
> >
> > With kind regards.
> >
> > */Lawal
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20190105/19cd9118/attachment.html>