ABINIT Discussion Forums

Dear all,

I have finally got the phonon spectrum of CsH2PO4 crystal (Z=2, 16 atoms, P2_1/m) in Gamma point. Unfortunately, instead of three acoustic mode frequencies (stirctly zero) I've got -2, -2, +2 cm^-1. But this is other story. I would like to clarify this problem later. All other frequencies are quite resonable and correspond well to experimental ones. Now I am interested in the following question:

How may I get my 48 phonon frequencies classified according to irreducible representations (Ag, Bg, Au, Bu) of P2_1/m sp. group in BZ centre? Of course, using resources of ABINIT.

Sincerely,
Yaroslav

Dear Yaroslav,

To get the symmetry transformation of each mode and to fix the acoustic sum rule problem, you have to use the anaddb postprocessing program (http://www.abinit.org/documentation/helpfiles/for-v7.6/users/anaddb_help.html).
See the variable asr: http://www.abinit.org/documentation/helpfiles/for-v7.6/users/anaddb_help.html#asr for acoustic sum rule problem and see http://www.cecam.org/upload/files/file_1907.pdf for the mode symmetry assignment.

Let us know if you still have problems.
All the best,
Eric

Dear Eric,

thank you very much for your help. I need some time to be more familiar with ANADDB. After that, I guess, I'll let you know whether everything is OK.
Meantime, since I cannot operate well with ANADDB, I am trying to perform the block-diagonalization of dynamical matrix by myself using the symmetry coordinates and corresponding unitary transformation of dynamic. matrix.

Thank you once more,
All the best,
Yaroslav

Dear Abinit users,

it seems to me I faced with some intrinsic problem of Abinit. The problem concerns to the phonon mode classification according to irreducible representations. I have calculated the phonon spectrum of monoclinic crystal (P2_1/m, Z=2) with 16 atoms per unit cell. 48 normal modes are classified according to irreduc. representations in Gamma point as follows: 13Ag+8Bg+11Au+16Bu. All Ag, Bg modes are Raman active, whereas Au and Bu ones are IR active. ANADDB utility performes the symmetry classification in Gamma point without assuming the non-analyticity. However, if the phonon spectrum is calculated with non-analyticity in the
three different cartesian directions (1.00000 0.00000 0.00000; 0.00000 1.00000 0.00000; 0.00000 0.00000 1.00000) the frequencies of all Au and Bu symmetry modes are shifted due to the LO-TO splitting (sometimes rather significantly ~ 100 cm^-1). In other words, the initial order of phonon modes which is relevant for the analytic case (0 0 0 point) becomes violated for 24 IR active modes (Au and Bu) in the non-analytic case. For many modes I may detect the proper symmetry (Au or Bu) combining some indirect features, but not for all. However, it would be highly desirable to get directly with ANADDB the symmetry classification of modes calculated accounting the non-analyticity.

Best regards,
Yaroslav