Highly sensitive optically heterodyned, Raman-induced Kerr-effect spectrometer using pulsed lasers

Abstract

A highly sensitive, optically heterodyned Raman-induced Kerr-effect (OHD-RIKE) spectrometer designed for the spectroscopy of radicals and molecular ions in discharges is presented. A linearly polarized, pulsed dye laser beam (probe) is crossed in the sample with the circularly polarized second harmonic of a pulsed Nd:YAG laser beam (pump). The nonlinear interaction between these beams and the sample induces a birefringence that creates a polarization component of the probe laser orthogonal to its original polarization. The magnitude of the birefringence peaks at Raman resonances as the probe laser is tuned. The combination of detection of the birefringence by the use of polarizers that have an extinction coefficient of the order of 108 with a pulsed probe laser with sufficient power to maximize the signal relative to the shot noise in the heterodyne field provides the basis for the high sensitivity. By careful adjustment of the polarization of the pump laser, we can decrease nonresonant contributions to the birefringence and further increase the signal-to-noise ratio. By subtracting a fraction of the magnitude of a reference signal from the OHD-RIKE signal, we further enhance the sensitivity by reducing pulse-to-pulse fluctuations of the pulsed probe laser to near (10 times) the shot-noise limit of the local oscillator intensity. The spectrometer is tested on the CO2 molecule in the Fermi-resonance region. Using a high-power (750-mJ) single-mode YAG laser as both the Raman pump laser and the pump for the dye laser that is used as the probe laser allows us to operate near Raman gain saturation, as observed in other coherent Raman spectroscopies. Preliminary studies of saturation for the OHD-RIKE in CO2 are presented. We present both broadband and narrow-band spectra of CO2. For broadband (∼10- GHz) studies we achieve sensitivities of 6×1013 molecules cm-3 for a signal-to-noise ratio of ∼1. In narrow-band spectra (∼400 MHz) we observe increased backgrounds owing to problems that arise because of the high intensities and the extremely high polarization extinction coefficients. These higher backgrounds prevent attainment of the improved signal-to-noise ratio expected from the larger Raman gains that are presumed to result from narrow-band operation. Possible solutions to the background problems are discussed.