Page 1 of 1

Charge transfer between two different orbitals in one atom

Posted: Tue Dec 09, 2014 3:32 pm
by s.zarrini
Dear all,

As I know by the bader analysis, the transferred charge between two atoms in the unit cell can be seen, but , I was wondering how I can see the probable transferred charge between two different orbitals in the one atom of the cell? For example, I want to know how much charge are going from 4s orbital to 3d orbital of Ni atom in a specific metallic crystal.
By the way, the VASP is the program I am using.

Best regards,

Salman

Re: Charge transfer between two different orbitals in one at

Posted: Wed Dec 10, 2014 4:21 am
by graeme
This is not really part of the Bader approach. I suggest looking at the site-projected density of states, that you can do in vasp. Also, if you use DFT+U, you can get even more detailed information about the occupancy matrix around each metal center.

Re: Charge transfer between two different orbitals in one at

Posted: Wed Dec 10, 2014 4:08 pm
by s.zarrini
Dear Graeme,

Thank you for your answer.
I knew about DFT+U, but I do not know how to interpret the result, I even wrote in VASP forume about it, but no answer I could get. For example below are the "occupancies and eigenvectors" of "Ni" and "B" atoms in one calculation, I would be appreciated if you could explain how I can found out the intera atomic charge or electron transferring in the d orbital of Ni. I know every column is related to different d suborbital in Ni (dxy, dxz,.....) , and (pxy,pxz,pxz) for p orbital of B, but not more.

Best regards,
Salman Zarrini

+++++++++++++++++++++++++++++++++++++++++
onsite density matrix for Ni

spin component 1

0.8418 -0.0082 0.0000 0.0000 0.0000
-0.0082 0.8405 0.0000 0.0000 0.0000
0.0000 0.0000 0.8413 -0.0080 0.0141
0.0000 0.0000 -0.0080 0.8283 0.0000
0.0000 0.0000 0.0141 0.0000 0.8623

spin component 2

0.8418 -0.0082 0.0000 0.0000 0.0000
-0.0082 0.8405 0.0000 0.0000 0.0000
0.0000 0.0000 0.8413 -0.0080 0.0141
0.0000 0.0000 -0.0080 0.8283 0.0000
0.0000 0.0000 0.0141 0.0000 0.8623

occupancies and eigenvectors

o = 0.8233 v = 0.0000 0.0000 -0.5183 -0.8343 0.1881 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.8233 v = 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 -0.5183 -0.8343 0.1881
o = 0.8329 v = 0.0000 0.0000 0.0000 0.0000 0.0000 0.6798 0.7334 0.0000 0.0000 0.0000
o = 0.8329 v = 0.6798 0.7334 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.8389 v = 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 -0.7188 0.5441 0.4328
o = 0.8389 v = 0.0000 0.0000 -0.7188 0.5441 0.4328 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.8493 v = -0.7334 0.6798 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.8493 v = 0.0000 0.0000 0.0000 0.0000 0.0000 -0.7334 0.6798 0.0000 0.0000 0.0000
o = 0.8698 v = 0.0000 0.0000 0.4634 -0.0891 0.8816 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.8698 v = 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.4634 -0.0891 0.8816

++++++++++++++++++++++++++++++

++++++++++++++++++++++++++++++
onsite density matrix for B

spin component 1

0.1845 0.0000 0.0000
0.0000 0.1823 0.0053
0.0000 0.0053 0.2022

spin component 2

0.1845 0.0000 0.0000
0.0000 0.1823 0.0053
0.0000 0.0053 0.2022

occupancies and eigenvectors

o = 0.1810 v = 0.0000 0.9702 -0.2425 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.1810 v = 0.0000 0.0000 0.0000 0.0000 0.9702 -0.2425 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.1845 v = 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.1845 v = 0.0000 0.0000 0.0000 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.2035 v = 0.0000 0.0000 0.0000 0.0000 0.2425 0.9702 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
o = 0.2035 v = 0.0000 0.2425 0.9702 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000
+++++++++++++++++++++++++++++++++++++++

Re: Charge transfer between two different orbitals in one at

Posted: Fri Dec 12, 2014 12:52 am
by xph
I think "charge transfer" here means charge distribution differences between Ni atom in the compounds and Ni atom in vacuum. If you already know the orbital occupations of isolated Ni atom, you just need to calculate the occupation numbers in the compound.

For the density matrix, the diagonal numbers are the number of electrons in the five d-orbitals, but there is no value for the s orbital that has no U applied.
As Graeme also suggested, the site projected DOS will give all the information you need. This can be done by setting LORBIT = 11.