LO and TO phonon branches

Alloy Theoretic Automated Toolkit (ATAT)
1

Hello,

I have calculated the phonon dispersion curves for a structure (BaBr2). The unit cell has has 12 atoms, consequently, other than the 3 acoustic modes, I have 3N-3 = 33 optical modes, of which 22 are transverse optical (TO) and 11 are longitudinal optical (LO) modes. I need to separate out the LO modes and I was wondering if the way the eigenfreq.out files are written, would allow me to do that.
As of now, I am finding it hard to distinguish between the LO and TO modes. The eigenfreq.out files has 36 columns corresponding to the acoustic and optical modes, of which the first three are acoustic. Of the remaining 33 columns, can I know which ones are LO and which ones are TO?

2

The option -apd (atom-projected DOS) could be useful for this.
It forces the eigenvectors of the normal modes to be output in the fitfc.log file after the line
"Phonon frequencies/modes/weights:"