High Dynamic Range Laser Dispersion Spectroscopy of Saturated Absorption Lines Kale J. Franz 1 , Damien Weidmann 2 , and Gerard Wysocki* 1 1 Department of Electrical Engineering, Princeton University, Princeton, NJ 08544 USA 2 Space Science and Technology Department, STFC Rutherford Appleton Laboratory, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0QX, UK *e-mail: gwysocki@princeton.edu, web: www.princeton.edu/~gwysocki/ Abstract: A spectroscopic detection of molecular dispersion based on frequency-chirped laser is presented. Unlike non-linear direct absorption methods yielding line saturation, this method provides linear signal response and accuracy over a wide range of sample concentrations. ©2010 Optical Society of America OCIS codes: (300.6390) Spectroscopy molecular; (300.6310) Spectroscopy heterodyne; (260.2030) Dispersion. 1. Background Direct optical absorption is the predominant physical mechanism in the majority of today’s molecular laser sensing systems. Whilst direct laser absorption spectroscopy (LAS) can achieve detection limits down to pptv (parts-per-trillion by volume) levels, several practical weaknesses degrade performance and limit LAS capabilities. Absorption signals result from transmission of light through a predetermined sample space and are described by the Beer-Lambert law. When detecting low concentrations, absorption signals consists of small changes in the total laser power arriving at the detector. Therefore the signal due to molecular absorption is small and superimposed on a several order of magnitude greater background which is subject to fluctuations. Hence, laser amplitude noise, interference effects, and other intensity fluctuations are the main source of measurement error. When detecting high concentrations yielding ≥ 10% absorption, the Beer-Lambert law becomes highly non-linear . Above 10% absorption saturation starts and close to 100% absorption even large changes in concentration can no longer be measured. This fundamentally limits the achievable accuracy for high sample concentration. Here is presented results from an alternative molecular detection technique that exploits the anomalous dispersion occurring around a molecular absorption line. The method provides several advantages over traditional direct LAS techniques. For example, the measurement of anomalous dispersion is background-free in nature. Therefore, the signal is baseline-free and immune to fluctuations of the laser power incident on the detector. Also, unlike the measurement of direct absorption that saturates at high concentrations, the measurement of dispersion is a completely linear process. Thus, large dynamic range of concentration measurements is achievable, and the accuracy is maintained over this range. Dispersion signals are small, and measuring them is challenging, as previously confronted by techniques making use of optical dispersion [1]. Fig. 1(a) and 1(b) shows calculated absorption and corresponding dispersion for the R(10.5e) and R(10.5f) doublet of nitric oxide (NO) centered at 1912.076 (1% concentration at 5 Torr and a 15 cm path length). While the magnitude of the direct absorption is ~25%, the corresponding refractive index change is ~1.5×10 -5 . The method presented here relies on high laser chirp rates to boost molecular dispersion signals. A two color laser beam is produced using an acoustic optical modulator (AOM) and the dispersion signal is derived from the frequency demodulated heterodyne beat note of the two colors. The experimental setup is shown in Fig. 1(c). The demodulated frequency signal containing the refractive index change information scales with the laser chirp rate, which can be further exploited to 1912.10 1912.09 1912.08 1912.07 1912.06 1912.05 0 5 10 15 20 25 (a) Absorption (%) Wavenumbers (cm -1 ) 1912.10 1912.09 1912.08 1912.07 1912.06 1912.05 -8 -6 -4 -2 0 2 4 6 8 (b) Dispersion ( × 10 -6 ) Wavenumbers (cm -1 ) Fig. 1. Calculation of Voigt absorption profile (a) and corresponding dispersion (b) for the nitric oxide (NO) R(10.5e) and R(10.5f) transitions. Path length is 15 cm, NO concentration is 1%, and total pressure is 5 Torr. A schematic of the experimental set-up is shown in (c). Laser AOM 0 th order 1 th order Frequency shifted Molecular dispersion signal Laser frequencies Two-color Laser beam Sample Detector Mirror Mirror Beamsplitter RF analyzer (c)