Chiral molecules tend to absorb one circular polarization of light more than the other. Is this ever used to isolate a particular enantiomer?
Yes. For some chiral compounds, each enantiomer in a racemic mixture absorbs certain frequencies of light differently depending on the circular polarization type of the light.
That means that if you can pump in sufficiently intense and energetic (e.g. ultraviolet) circular polarized light pulses into the racemic mix of such compounds, you may in some cases increase the rate of breakdown of the enantiomer that absorbs that particular circular polarization (left or right), resulting in an increase in the relative concentration of the other enantiomer.
You can view the first page of a paper on asymmetric photolysis using circularly polarized ultraviolet light here.
The optical rotation is certainly used to follow the separation of optical isomers (enantiomers if they are mirror images, or diastereomers if you have more than one stero-center and the two isomers are not mirror images). You can see how the optical rotation changes as you use the normal separation techniques (chromatography, distillation, crystallization, etc.). If you know the true value for the substance, you can determine the purity. But I don't know how polarized light would be used to do the actual separation.