why the "surface isovalue" is different for different ATK version ?

[fangyongxinxi]

Dear,

I build a device model, and calculate the "voltage drop" (0V vs 2V)
I want to see the voltage drop along z direction  through "surface isovalue"

Problems are :
1.
with 11.8 version, I can see the "surface isovalue" change from 0V to 2V,
with 13.8 version, the the "surface isovalue " is quite small (about 0.07), I think it is because of the unit. right?

2.
I know how to get the average voltage drop
http://quantumwise.com/images/stories/mini-tutorials/iv/voltage_drop_analyzer.py
, but Is there any simple script to export the 3-d data ?
I hope to extract the data, and draw picture in Matlab.

3. I set the  "   electrode_voltages=( 1.0*Volt, -1.0*Volt)"
the result is like the attachment.

However, I hope to get the voltage drop from left to right
why the average value is not like that.


Thanks.

[Anders Blom]

1. There might be some issue with the isosurfaces in 13.8, which we want to fix for 13.8.1. To help us, could you provide the input data for your calculation?

2. You can export the voltage drop 3D grid data as a CUBE file from VNL.

3. The picture is correct, if you want it the other way you would need the opposite sign of the bias.

[fangyongxinxi]

1. There might be some issue with the isosurfaces in 13.8, which we want to fix for 13.8.1. To help us, could you provide the input data for your calculation?

2. You can export the voltage drop 3D grid data as a CUBE file from VNL.

3. The picture is correct, if you want it the other way you would need the opposite sign of the bias.

Dear Sir,
Thanks for your reply

1. the attachment "0v.py" & "1v.py" are the input files, to calculate the voltage drop from 1V~-1V (left is 1V).

after calculation(11.8 version), I use this kind of line to get the voltage_drop:
voltage_drop=potential[1v]-potential[0v]
and get  the average voltage_drop, which is like yesterday's attachment (yesterday's picture is the voltage drop 2V-0V).

2. I still have problem on the "export the voltage drop 3D grid data as a CUBE file "

I am trying to use VESTA to get the 3d data, (not familiar with VESTA or Gaussian), so what kind of "file.*" I should export ? I export the data as ".txt" format, but it can not be import by VESTA.

3. I set the  "electrode_voltages=( 1.0*Volt, -1.0*Volt)" in "1v.py", I hope to get the voltage drop from left to right, but the average voltage drop seem like drop from right (1V)to left(-1V) (like yesterday's attachment), I don understand it.

you said "The picture is correct, if you want it the other way you would need the opposite sign of the bias.". could you explain it more detail ?  Thanks.

4. I have a problem about the voltage_drop surface isovalue, as the attachment shows(calculated and analyised by ATK11-8).

I set "electrode_voltages=( 1.0*Volt, -1.0*Volt)"

4.1,
The average voltage drop shows in the left side is "-1V", not "1V", why ?
4.2,
the surface isovalue is "0.9", as the blue color shows, this "0.9" is 0.9V(like the voltage drop that I hope to get), or something else?
4.3,
the unit of the color-bar is deltaVH(eV), and the value is from (-0.9-0.9), is there some connection between this 0.9(in the color bar) with that 0.9 (the surface isovalue)?


One more thing, could set a larger size abot the attachment ? ,400KB seems a little small.


Thanks
 

[Anders Blom]

1. Sorry, I don't understand the problem here.

2. You open 13.8, you select the grid on the LabFloor, you click "Export CUBE file". How to open in VESTA I don't know, I never used it, but maybe you can also indicate why you need to? Is there something you can't do in VNL?

3. If you want the bias the other way, use electrode_voltages=( -1.0*Volt, 1.0*Volt)

4.1 See 3

4.2 What you are plotting is the effective potential, which is -e*V where V is the voltage (electrons have negative charge). Hence the unit is eV, and hence your confusion about the direction of the potential vs the voltage. The unit is written on the colorbar. So it's not 0.9 V, but 0.9 eV.

4.3 The colorbar is rather useless for isosurfaces since you are plotting only points with a given value (the isovalue), so you only get one color.

400 kB is plenty for PNG files - don't use BMP (uncompressed format).

[fangyongxinxi]

Thanks for your reply.
it helps me alot.

[fangyongxinxi]

Dear,

thanks for your help, I still have one question about you second reply:
"2. You open 13.8, you select the grid on the LabFloor, you click "Export CUBE file". How to open in VESTA I don't know, I never used it, but maybe you can also indicate why you need to? Is there something you can't do in VNL?"

After voltage drop calculation (with 11.8 or 13.8 version), drop the "*.nc" file to "Viewer",
I hope to get voltage_drop in the form of "cut plane " (of course we can also choose the "isosurface" form, but it can not show several "isosurface value" in one picture), and want the exact value of the voltage drop (not in the form of color change)
So I need the data file.

Problem is: when I export the data in "*.cube" , I can't open this cube file in Gaussian-view or VESTA, it shows as the attachment.
I don't know is something wrong with my operation, or some thing wrong in the exported cube file.


Thanks.

1. Sorry, I don't understand the problem here.

2. You open 13.8, you select the grid on the LabFloor, you click "Export CUBE file". How to open in VESTA I don't know, I never used it, but maybe you can also indicate why you need to? Is there something you can't do in VNL?

3. If you want the bias the other way, use electrode_voltages=( -1.0*Volt, 1.0*Volt)

4.1 See 3

4.2 What you are plotting is the effective potential, which is -e*V where V is the voltage (electrons have negative charge). Hence the unit is eV, and hence your confusion about the direction of the potential vs the voltage. The unit is written on the colorbar. So it's not 0.9 V, but 0.9 eV.

4.3 The colorbar is rather useless for isosurfaces since you are plotting only points with a given value (the isovalue), so you only get one color.

400 kB is plenty for PNG files - don't use BMP (uncompressed format).

[Anders Blom]

The Cube file you show contains a header - if you remove that (i.e. let the file start with the line CUBE FILE) it should be fine. Actually there may be a footer too - in case the last line of the file is something like +---------------+, remove that too.

HOWEVER, this Cube file is most likely not made with 13.8, because we removed this header.

I may have confused you with the option to "Export Cube file" - that's my own private plugin which also includes the geometry in the file I'll make it available! In 13.8 you should click "Export Data as Text", and in this case the file should directly open in VESTA.

[fangyongxinxi]

Yes, it works.
Thanks.

the reason it did not work before is I did not move the head part in the cube file.