Author Topic: basis sets and pseudopotentials for Pr  (Read 903 times)

0 Members and 1 Guest are viewing this topic.

Offline merlys.borges

  • New QuantumATK user
  • *
  • Posts: 1
  • Country: cl
  • Reputation: 0
    • View Profile
basis sets and pseudopotentials for Pr
« on: February 23, 2024, 18:07 »
I'm attempting to calculate a crystal containing praseodymium and have a few inquiries. Is it possible to compute this compound accurately? The program documentation regarding basis sets and pseudopotentials does not specify anything for lanthanides or actinides, and previous forum posts suggested that calculations for Pr and other elements were not supported. Are these elements now supported? With version 2023, I can perform calculations using GGA with LCAO without encountering any errors, and the input parameters seem good. However, attempting to use PAW shows me that no pseudopotentials are available for Pr, and LDA yields an error. I'm uncertain whether the parameters for describing Pr atoms in the GGA calculation are correct.

Additionally, I aim to incorporate spin-orbit coupling (SOC) and make some corrections to the bandgap in the calculation. The DFT-1/2 correction is not defined for Pr. Does this mean the parameter is not automatically defined, and I can assign it a value, or does it indicate that this correction is not applicable for Pr? Can those corrections be applied to systems with Pr?

Thank you for your assistance.

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5411
  • Country: dk
  • Reputation: 89
    • View Profile
    • QuantumATK at Synopsys
Re: basis sets and pseudopotentials for Pr
« Reply #1 on: March 4, 2024, 21:48 »
PAW-GGA-LCAO should work fine, although there are some other details in PAW that might not work. You could also use the planewave calculator with PAW.

Yes, there is no predefined +1/2 parameters for Pr (or in fact a lot of exotic elements). I would probably recommend GGA+U for Pr, but either way you will need to fit the parameter.