[DFTB-Plus-User] DFTB+ about the process for calculating relaxed density

이인성 (화학과) islee at unist.ac.kr
Wed May 31 09:16:10 CEST 2017

Thank you for your advice. (DEBUG level >= 2)

I check the source code for dipole moment using Hellman-Feymann theorem.

And I find the analytic form of dipole moment integral.

d_mu_nu = e * r * S_mu_nu.

If this is wrong, then please let me know about the form.


In Seong Lee.

보낸 사람: Ben Hourahine <benjamin.hourahine at strath.ac.uk> 대신 DFTB-Plus-User <dftb-plus-user-bounces at mailman.zfn.uni-bremen.de>
보낸 날짜: 2017년 5월 30일 화요일 오후 7:26:19
받는 사람: dftb-plus-user at mailman.zfn.uni-bremen.de
제목: Re: [DFTB-Plus-User] DFTB+ about the process for calculating relaxed density

Dear Seong Lee,

The ground state dipole is, as you mention, the derivative of energy with respect to the field. This can be calculated using the Hellmann-Feynman theorem through the derivative of the hamiltonian with respect to the external field. This gives the same value as obtained using the Mulliken charges and locations of atoms (this is due to the use of Mulliken partitioning in constructing the contribution from an external electric fields to the total energy).

When DFTB+ is compiled at DEBUG >= 2, it will evaluate dipole moments for molecules using both of these methods.

The density matrix is written in the basis of the atomic orbitals used for DFTB and it is these functions that are explicitly dependent on r, not the density matrix itself. DFTB avoids evaluation of integrals during the calculation by re-casting as much as possible in terms of the overlap elements between atomic basis functions. So for example, the contribution from an external field is

H_ij = 0.5 S_ij ( V_i + V_j)

where V_i is the external potential at the atom containing orbital i. There is more detail in several of the articles listed at http://www.dftb.org/about-dftb/references/

Are you instead asking about the oscillator strength for something else, such as electronic or vibrational excitations?



On 30/05/17 10:33, 이인성 (화학과) wrote:

Hello, DFTB+ developers

I am a new user for dftb+ and I want to know about relaxed density matrix.

To obtain the information about the oscillator strength, we have to know relaxed density matrix.

And to obtain the relaxed density matrix, we have to know the dipole moment integrals.

Differentiation of the total energy with respect to the field F gives the dipole moment.

d_mu_nu = e * < chi_mu | r | chi_nu > = \partial(h_mu_nu) / \partial(field)

where d_mu_nu is dipole moment integrals, chi is AO, e is electron charge, and r is the electron coordinate vector.

I am sorry for low quality of upper equations.

Thus, I want to know 1. the analytic formula for d_mu_nu and 2. the process how dftb+ calculate this dipole moment for oscillator strength.

Thank you.

In Seong Lee

DFTB-Plus-User mailing list
DFTB-Plus-User at mailman.zfn.uni-bremen.de<mailto:DFTB-Plus-User at mailman.zfn.uni-bremen.de>

      Dr. B. Hourahine, SUPA, Department of Physics,
    University of Strathclyde, John Anderson Building,
            107 Rottenrow, Glasgow G4 0NG, UK.
    +44 141 548 2325, benjamin.hourahine at strath.ac.uk<mailto:benjamin.hourahine at strath.ac.uk>

2013/14 THE Awards Entrepreneurial University of the Year
      2012/13 THE Awards UK University of the Year

   The University of Strathclyde is a charitable body,
        registered in Scotland, number SC015263
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.zfn.uni-bremen.de/pipermail/dftb-plus-user/attachments/20170531/0b76ef0a/attachment.html>

More information about the DFTB-Plus-User mailing list