Saturated polarizalson spectroscopy with a picosecond laser for quantitative concentration measurements s

The collisional dependence of saturated polarization spectroscopy with a picosecond laser is investigated by probing hydroxyl in a flow cell. 01999 Optical Society of America OCIS Codes: (120.1740) Combustion diagnostic (300.6420) Spectroscopy,nonlinear While nanosecond lasers have been used often for nonlinear dia=mostic measurements of flame composition, picosecond lasers provide a potentially superior source for such techniques. Compared to a nanosecond laser, picosecond lasers produce significantly greater peak power for the same pulse energy, and this could improve the signal strength of multi-photon techniques such as degenerate four-wave mixing (DFWM) and polarization spectroscopy (I%). The spectral bandwidth of a transform-limited, 50-150 ps pulsewidth laser [1] is sufficiently narrow to couple effectively with the molecules, and thus such lasers have been successfully used for laser-induced fluorescence (L~ concentration measurements [2,3]. Such pulse widths are on the order of the characteristic collisional time in an atmospheric combustion environment. It has been suggested that the signal produced by such lasers would be less dependent on the collisional environment because the behavior of the molecular system probed by short-pulse lasers is governed more by the spectral width of the laser and the Doppler effect .[4]. To investigate the collisional dependence of the polmization spectroscopy signal generated with a picosecond laser, we probe the A222+X211(0,0) band of OH in a flow cell. In this well-controlled environment, we monitor the change in signal stren=@ as we vary the buffer gas pressure by a factor of 50. In PS two beams, a high-power pump and a low-power probe, are crossed at the measurement location. The probe beam is passed through two orthogonally oriented linear polarizers positioned on opposite sides of the measurement location. The pump beam is either circularly polarized or lineady polarized 45° with respect to the pokirization of the probe beam. When the pump beam is tuned to a resonance of the interaction medium, it induces a change in both the real and imaginary parts of the refractive index. This causes the polarization of the probe beam to rotate as it passes through the medium, and the PS signal is observed as leakage of the probe beam through the second polarizer. The experimental setup is pictured in Fig. 1, The laser source is the tlequency-doubled output of a distributed feedback dye laser (DFDL) with motorized tuning and active wavelength stabilization. The energy of the doubled output is controlled by a polarizer preceded by a quarter-wave plate (to compensate for the slightly elliptical polarization of the beam) and half-wave plate (to rotate the polarization of the beam). The beam is down-collimated and directed toward an uncoated UV 3° wedge. The front surface reflection ti-om the wedge forms the probe beam while the back surface reflection is sent to two photodiodes (Thorlabs DET 110 and DET 210) to monitor the beam energy. The portion of the beam that passes through the wedge forms the pump beam. The pump and probe beams are sent through separate delay lines. Before passing through the vertically-oriented polarizer, the probe beam is directed through a half-wave plate rotated to maintain the probe beam energy at 0.5% of the pump beam energy. A quarter wave plate is placed in the pump beam path to change the vertically polarized pump beam to a circular polarization. Hydroxyl (OH) is created by photolysis of hydrogen peroxide (HzOJ using the softly focussed 266-rim output of a Continuum PL9020 Nd:YAG laser. Hydrogen peroxide is obtained by flowing 8 standard cm3 s“’argon through a bubbler containing 95 wt. ?ZOaqueous Hz02.. Further details of the photolysis system are