Cavity ring-down spectroscopy with a pulsed distributed feedback quantum cascade laser

Abstract

A pulsed distributed feedback quantum cascade laser operating near 970 cm -1 (10.3 µm) was coupled with the technique of cavity ring-down spectroscopy, as described here for the first time. The newly constructed set-up was tested by recording three relatively weak rotational lines of the 10 0 0→00 0 1 vibrational band of CO 2 in the range from 966.75 cm -1 to 971.5 cm -1 . The CO 2 lines were recorded by measuring the decay time of a CO 2 -N 2 mixture flowing through an open sample tube placed between the cavity ring-down mirrors. The quantum cascade laser frequency was tuned at a rate of ~ 0.071 cm -1 /K by changing the heat sink temperature in the range between -20 and 50 °C. The first results demonstrated the applicability and high sensitivity of the cavity ring-down spectroscopy -pulsed quantum cascade laser combination and encouraged us to extend our research to the study and detection of ammonia. We demonstrated that a detection limit of ammonia of ~ 25 ppbv can be attained with the current set-up. Basic instrument performance and optimization of the experimental parameters for sensitivity improvement are discussed.

Trace gas measurements using optically resonant cavities and quantum cascade lasers operating at room temperature

“…The k values from the ICOS spectrum are below 6 ϫ 10 −5 cm −1 , clearly larger than that obtained from the CRDS analysis but almost a factor of 10 less than expected from a calculation using the HITRAN database ͑2004 edition͒. 61 The two CH 4 Similar observations were made by Sukhorukov et al 49 They concluded that the dual effects of line broadening and line position shifts might be minimized to some extent by reducing the pulse widths used to operate the laser. In our case it is 60 ns and it was found that reducing it lead to an insufficient signal for detection after the cavity.…”

“…Possible explanations for the effects observed here can be found in the work of Sukhorukov et al 49 They discussed the deviations in their measurement series ͑i.e., absorption lines with an increased linewidth and shifted line positions with increasing pulse width͒, which were not as pronounced as those presented here, in terms of the well-known frequency-down chirp of the QCL. The seed current for the laser generates heat during the pulse which, in turn, makes the frequency decrease throughout the duration of the laser pulse.…”

“…11 If now the chirped QCL pulse is taken into account the picture has to be modified. Although this situation has been considered earlier by Silva et al 8 and Sukhorukov et al, 49 the effect is usually underestimated. Silva et al determined an upper limit for the effective laser linewidth of their pulsed QCL, from a fit to NO absorption lines, to be 0.024 cm −1 .…”

“…Clearly, this could be overcome by calibration 48,49 but in doing so, one of the major advantages of CRDS, namely, the calibration free measurement approach, is lost. Conversely this raises the question as to whether the combination of an optical cavity with a pulsed QCL in order to benefit from reduced sample volumes compared to multiple pass cells of the same effective length is desirable as a calibration free method.…”

“…Other studies combining TE cooled pulsed QCLs with the CRDS approach recognized the influence of the frequency chirp of QCLs leading to disturbed line profiles followed by reduced sensitivity and selectivity. 48,49 Further development was forthcoming from the availability of DFBQCLs with increasingly good performance such as output power and near ambient temperature operation, in the beginning limited to the 5 -10 m range. Several applications of QCLs combined with either the CRDS or the CEAS/ICOS techniques for trace gas detection [50][51][52][53][54][55][56][57] followed, and sensitivities down to 2 ϫ 10 −9 cm −1 Hz −1/2 were reported.…”