extract isovalue of the transmission eigenstates

[Anders Blom]

This looks more like a problem that perhaps we and VESTA don't agree on the exact specification of the CUBE file format, because they are not even close to each other... I used to have VESTA installed but I don't anymore (I found it terribly difficult to use ) so I can't really test it myself.

Is there any particular plotting you can't do in VNL that you need VESTA for? I don't know if you noticed, but 13.8 is now available in beta-release, and the 3D viewer has some new options for controlling the plotting.

[Anders Blom]

Saaaaaaaaaay. I never noticed that extra column in the CUBE format... The coordinates are given in the example, at the link I sent, by 5 columns. The 2nd one is always zero, not sure what it's for - but it's missing in our export. Try changing line 54 in the script to

Code: python

            stream.write("%5i%13.6e%13.6e%13.6e%13.6e\n" % (e.atomicNumber(),0,c[0],c[1],c[2]))

and see if that improves things.

[baizq]

Ok. I have checked the coordinates. If converting the unit from Bohr to Angstrom, they appear to be correct. Then the problem should lie in the importing of the cube file to VESTA...

Thank you for your help, Anders.

[baizq]

Ok...Ok...That might be where the problem lies...it is five column..."the first is the atom number, second (?), the last three are the x,y,z coordinates of the atom center".

Let me try to change the code first...

But... why do they make a space for such an extra meaningless column....

[Anders Blom]

And above all: who is responsible for this file format, officially? Gaussian? Why is the top Google hit for the format specification a page from 2003 where the 2nd column indeed is unknown?

The column might be for charge, or some other property. Just weird nobody knows  

[baizq]

Anders, you are right. Inserting a zero column in between the atomic number and coordinates solves the problem(see attach file). VESTA may by default treat the 3rd column as x, 4th as y, and 5th as z, following the format of the weird Gaussian cube file. Since we missed a column at first, it cannot find the fifth column and just sets the z coordinates as pure zero. This explains why in my first plot all atoms are accumulated at the two ends of the supercell.

The reason I insist to use VESTA is simply that.... I have plotted all the other figures in the manuscript with VESTA...just to make sure of style consistency...Anyway I promise next time I will use VNL for plotting...

Thanks for your help. Have a nice weekend.

baizq