double walled nanotube building

[Zexter]

Hi

I want to compute the interaction energy between a (5,5)@(10,10) nanotube. Thank you in advance. another one question, when I try to use your Nanotube.py script it shows the error

"No module named ATK.KohnSham"


I want to know whether the above mentioned one is built into vnl already or to be downloaded. how can i rectify it

the script mwcnt.py u have attached generates only the structure of the outer tube in this case tube1 and doesn't considers the tube into account.so when i execute it returns me only the tube 1 configuration.

[Anders Blom]

Do you have a specific idea about the chiralities of the tubes of interest? It would help to know a bit of Python, in the sense that we can rather easily show how to do it for a specific case, but you will want to play around and modify it a bit to fit your particular needs...

But if, for instance, you are only interested specifically in double-wall, and they are all say of the form (n1,0) and (n2,0), then it would be easy to make a simple Custom builder that you can use.

[Anders Blom]

Attached is a "simple" example on making a double-wall tube. I can be made much simpler if we turn it into a Custom builder, but this shows how you would do it in "raw" Python.

[Zexter]

I want to compute the interaction energy between a (5,5)@(10,10) nanotube. Thank you in advance. another one question, when I try to use your Nanotube.py script it shows the error

"No module named ATK.KohnSham"


I want to know whether the above mentioned one is built into vnl already or to be downloaded. how can i rectify it

the script mwcnt.py u have attached generates only the structure of the outer tube in this case tube1 and doesn't considers the tube2 into account.so when i execute it returns me only the tube 1 configuration.

next question is what does the error"shape mismatch: objects cannot be broadcast to single shape" mean and how can it be solved

[Anders Blom]

The script I attached my post is for ATK 10.8, while the Nanotube.py script you probably found in another post on the Forum is for 2008.10. It's important not to mix them, the two versions are not 100% compatible.

Yes, you can do the whole workflow in VNL 10.8. It's not hard but requires a short sequence of operations to center the XY coordinates of the respective two tubes. It's a bit difficult to describe in a few words; we'll make a small tutorial on it soon!

I too noticed that the script doesn't show you the double-wall tube if you drop it directly on the Builder etc. But if you run it from the command line ("atkpython mwcnt.py") it will create a NetCDF file containing the multiwall tube.

If I have time I'll make a better solution for (n1,n1)/(n2,n2) DWCNTs later this week.

[Zexter]

Dear Mr.Blom,
 
Is there any general procedure to build a double walled nanotube i.e a work flow kind of thing.what are the basic things that are to be followed to build a dwnt.

could you just help me infiguring out
1. how to center the XY coordinates of the two tubes
2. what does the term unit cell defined at the very beginning of the script mwcnt.py signifies
3. what does the variable period_length means
4. How to run the code in command line mode
5. general work flow (an algo kind of thing) to generate a Double walled nanotube
6. any tutorial for the present problem at present available

[Anders Blom]

The general way, provided the two tubes have the same Z-unit cell length, is defined by the script (and your points).

If you look in the script, you will see that it centers the coordinates and merges the two tubes. So, the script does precisely the work you need. To change the details in it, you just need to select the chiral indices on the first two lines, and also check the "xy" parameter for the X/Y unit cell size. That's all the changes that are needed, after that, open a terminal and run

atkpython mwcnt.py

and the DWCNT will be saved in mwcnt.nc which you can open in VNL and use the structure for further manipulations.

As I wrote, we'll make a tutorial on it later.

[Anders Blom]

I attached a commented version of the script

[Anders Blom]

If you drop the attached script on the "Custom" builder icon, I think you can have some fun!

[Zexter]

thank you for your Dwnt.py

could you please tell me what option has to be selected in the analysis part for getting the interwall interaction energy in the Dwnt.
2. whether molecular dynamics calculation can be performed in vnl
3. what should be the optimum kpoint sampling values to get a good result if the atom counts to 400-600

[zh]

The interaction energy cannot be obtained by directly choosing the options in VNL. To obtain the interaction energy of DCNT, three steps should be done:
i) calculate the total energies of interior and outer CNTs, respectively. That means you should separately build the respective configurations for the interior and outer CNTs.
ii) calculate the total energy of DCNT.
iii) the interaction energy can be obtained according to the formula listed here:  E_int = E_tot(DCNT) - E_tot(interior CNT) - E_tot(outer CNT).

The MD simulations cannot be performed in VNL, only the geometry optimization can be done.

The choice of k-point sampling depends on the length of DCNT, rather than the number of atoms.

[Anders Blom]

The interaction energy is derived from the total energy. See e.g. http://arxiv.com/abs/0812.0244
MD simulations will soon be possible in a future version of ATK, we have a prototype working.