Update (Dec 19, 2011): The recommended procedure now is:

Please, ask in comments for more details!

For general information about Bader analysis see this presentation.

For Bader analysis with VASP or Gaussian check Graeme Henkelman’s page.

Introduction

Analysis of charge on atoms can give a lot of interesting information in terms of understanding the nature of chemical bonding. There are different ways to do it:

• Mulliken analysis is implemented directly in SIESTA - it shows the direct population of the orbitals (basis) used in calculations.
• Natural bond orbital analysis. Similar to Mulliken in that it shows the population of the orbitals, but the interesting point is that it recalculates the population of orbitals of free atoms, i.e. in the way how we usually interpret the simplest bonds. Check the NBO program homepage for more details.
• Löwdin analysis. I don’t know too much about it. Check the original paper: On the non-orthogonality problem connected with the use of atomic wave functions in the theory of molecules and crystals. J. Chem. Phys., 18(3), p.365, 1950.
• Bader analysis. Based on the analysis of the final total charge density and division of it into basins associated with each atom. Sometime it can result in a basin non-associated with any atom which has a special physical meaning. Check this presentation for introduction.

(more…)