This thread is locked.Only browsing is available.
Top Page > Browsing
about charged molecule
Date: 2006/11/16 16:12
Name: Xinyuan Zhang   <>

Dear Prof.
I want to ask a question about the calculation of Openmx on the charged semiconductor molecule. As widely known, the DFT would underestimate the band gap. Hence, if an electron is doped into a neutral molecule, it will occupy the lowest conduction band. If DFT underestimate the band gap, does it mean it will also underestimate the total energy of the doped molecule? How to justify it?
Thank you in advance.

Xinyuan Zhang
Page: [1]

Re: about charged molecule ( No.1 )
Date: 2006/11/21 22:05
Name: Vasilii Artyukhov

The DFT problem is not with the band gap, but it is the unoccupied states in the conduction band that are treated incorrectly. Since your doping level is going to be occupied, all the well-known "band gap" errors will concern states that are above it. On the other hand, as your total energy is the integral of energy-weighted DOS over all occupied states, you are going to get the correct value of total energy - as usual in DFT calculations.
Re: about charged molecule ( No.2 )
Date: 2006/11/22 01:13
Name: Xinyuan Zhang  <>

Thank you very much

Page: [1]