Dear Sanjay,
You cannot completely 'switch off' the Frohlich interaction, but if lpolar=.false. then the long-range part of the matrix element is not treated correctly, and essentially you cut much of it off your calculations.
By the way, in case you haven't come across this paper, it might be interesting for you:
https://pubs.acs.org/doi/10.1021/acs.nanolett.8b04799Best,
Carla