PDOS for different elements

[HUAN BI]

Dear CP2K users and developers, 

  How can CP2K output the density of states of all atoms? I can't define different atom labels in the xx.inp file, and I can't print out the density of states of different atoms.

 Can you tell me how to do.

  In addition, how do I convert the optimized file (.xyz or .restart) into a .cif file

 Thanks in advance

[Moon Yue]

Dear Bi,

For  output of  the density of states of all atoms, you can find the exercises titled ‘Projected density of states for WO3" .

As for converting the optimized file (.xyz or .restart) into a .cif file, there are many software that can do that, such as OVITO, ATOMS.

Best regards,

Yue Qiang

[HUAN BI]

Dear Qiang,

   Thank you for your answer. I have check the file you suggest. 

    But, I expect to output the DOS of each atom rather than a class of atoms. For example,
My input file contains 5 O atoms, and I expect to output a DOS map of each O atom instead of outputting a DOS map after combining 5 O atoms.
If I want to achieve the above goals, what keywords do I need to add? I tried many keywords, but none of them worked.

Thanks in advance

Huan BI

[Marcella Iannuzzi]

Dear Bi,

You should use the subsection LDOS of the section PDOS, where you specify in LIST the index of the atom you are interested in 

Regards

marcella

[HUAN BI]

Dear  marcella,

  Thanks for your answer.

  In fact, I want to output the DOS map of all individual atoms. Let's say I have 36 atoms in total, so I would write: LIST 1..36 Is this correct? Is the DOS of a single atom output in this way, or is the 36 atoms combined and then output into a DOS map?

 Looking forward to your reply, I think this is my last question.

 Huan BI

[Gianluca Rengo]

Dear Huan,

To get a PDOS file for each single atom, I typically include as many &LDOS sections as the atoms for which I want the DOS. 

If you include the COMPONENTS keyword you will get also get  the DOS projected on the different m numbers for each orbital. 

Here's an example for atoms from 1 to 3:

&PDOS
        NLUMO -1
        &LDOS
          COMPONENTS
          LIST 1
        &END LDOS
        &LDOS
          COMPONENTS
          LIST 2
        &END LDOS
        &LDOS
          COMPONENTS
          LIST 3
        &END LDOS
        &LDOS
[.....]

&END PDOS

If you use LIST 1..36 as you proposed, you should get a file with the sum of the PDOS of atoms from 1 to 36. So you loose the information about single atoms.

I hope it helps. 

Regards, 

Gianluca