plotting eigenstate and determining corresponding energy of the state

[sathyajit]

Hi,

According to the documentation I can calculate the eigenstate of a quantum number with Eigenstate method.
1. Is it possible to do it for a device configuration, if so, what method must I use ?

projection_list
Specifies the orbitals within the configuration to include in the calculation.

Type: ProjectionList.

Default: All atoms included : ProjectionList(All)

2.In the above does ProjectionList imply atoms in the device configuration or the orbitals within a single atom ? :
This is contrary to what is said here:
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.projectionlist.html
3. How do I know the energy value of the eigenstate ?
4. How do I extract the 3D matrix for the eigenstate so that I can plot it ? (Is this option already there in VNL ?)

Thanks
Sathyajit

 

[Anders Blom]

1. Yes, you just use the same Eigenstate class.
2. You can project on many thing, both atoms and orbitals, it just depends on which argument(s) you use, "atoms", "orbitals", etc. The ProjectionList is general, but when you use the Eigenstate method for a device, typically what you want is to project on the central molecule.
3. Project on the same atoms, and compute the MolecularEnergySpectrum http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.molecularenergyspectrum.html
4. Save the Eigenstate in an NC file, then you can just open it in the Viewer in VNL to plot it in 3D

[sathyajit]

Thanks Dr.Blom.
The description of ProjectionList in
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.eigenstate.html

must read as atoms instead of orbitals. Am I missing something here ?

[Anders Blom]

You are right, it's not written very well in the docstring. Actually, you can project both on atoms and orbitals (and both), but it should be made clearer in the docstring.

[sathyajit]

Dr. Blom,

I am unable to get the 3s and 3p orbital strucutre from the script below. Can you please let me know what I am missing ?

The error I am getting is

Traceback (most recent call last):
  File "/tmp/3336602494981770.py", line 45, in <module>
    eigenstate=Eigenstate(molecule_configuration,3,proj_list)
  File "./zipdir/NL/Analysis/Eigenstate.py", line 62, in __init__
  File "./zipdir/NL/Analysis/Eigenstate.py", line 391, in calculateEigenstate
  File "./zipdir/NL/Analysis/AnalysisUtilities.py", line 133, in setAndCheckQuantumNumber
NL.ComputerScienceUtilities.Exceptions.NLValueError:  The parameter, quantum_number, must lie in the range [0,number of orbitals - 1], which is [0,1] in this case.

Regards
Sathyajit

[Anders Blom]

Your basis set is not large enough to contain 3s and 3p. A DoubleZetaPolarized set for Hydrogen contains two 1s + 2p as you can see in the Script Generator if you plot the basis set. There is also, actually, no 2s orbitals.

[sathyajit]

Dr.Blom,

I am facing the same error after I have modified the script looking into the documentation. Can you pls take a look at the modified script and the log file ?

Also when we define our own basis sets to include principal quantum numbers for 3d orbitals what is the expected computation increase ? I want to do the orbital calculation for a small 2 terminal device.

Regards
Sathyajit

[Anders Blom]

This is quite far outside the scope of regular support, so you're a bit on your own here. But generally speaking, it makes no sense to create a basis set with 4s only, you need to start from 1s anyway, because that's your ground state. The code works technically for 3d since there are no 1d and 2d states, but if you add 4s it will still find the 1s ground state, it's just that you now have a rather odd basis function which gives a bad energy for the 1s state.

If you want to add 3d or higher orbitals, you should just consider one of the larger predefined basis sets in ATK, like Tier 8 for the HGH pseudopotentials. Of course the computational increase will be noticeable.