Abstract

Photoelastic modulation (PEM) makes possible (with phase-sensitive detection) high precision measurement of small polarization changes in light that has interacted with matter. In an ideal modulator an oscillating birefringence (piezoelectrically induced in a silica crystal) leads to phase variations of the form cos(a₁), sin(a₁), where a₁ = A sin(2πft); f is the oscillation frequency and A is the maximum retardation. This standard model cannot account for the results of recent PEM-based experiments to measure optical activity by means of light reflection.¹ A more accurate theory, supported by experiment, is given that predicts phase variations of the form cos(b), sin(b)/ 2b, where $2b={\left[a_1^2+a_2^2+2a_1{a}_2\cos \left(2g\right)\right]}^{1/2}$; a₂ is the retardation produced by a distributed static birefringence with principal axis inclined at angle g to the modulation axis. Theoretical analysis of the light flux for selected experimental configurations of spectroscopic and polarimetric interest leads to signals at f and 2f that differ markedly from the predictions of the standard model and a two-plate theory with discrete static and dynamic birefringence. Experimental procedures are suggested for circumventing the birefringence-related anomalies in PEM-based reflection spectroscopy of chiral materials.

© 1989 Optical Society of America