Fig. 1. Principle of the agile CRDS with moderate optical feedback to a QCL. HWP: half-wave plate; AOM: acousto-optic modulator; PBS: polarizing beamsplitter; PM: plane mirror; P: power meter; ISO: optical isolator; M1, M2: cavity mirrors; PZT: piezoelectric actuator; PD: photodetector; DAQ: data acquisition card.

Fig. 2. Characterization of QCL linewidth. (a) Experimental setup. The absorption line of CO sealed in a gas cell (15 Torr) is used as a frequency discriminator. (b) Typical transmission signal of the QCL through the CO reference cell. (c) Variation of QCL linewidth with the feedback rate. The dashed line indicates the linewidth of the free-running QCL.

Fig. 3. Comparison of the cavity-transmitted signals with different levels of optical feedback. Four ringdown events obtained by scanning the cavity length with a triangle waveform (5 Hz) are shown in each figure. The QCL is operating either at the free-running state or with optical feedback (from −49 dB to −14.8 dB).

Fig. 4. Cavity ringdown measurement with a feedback rate of −18.2 dB. The cavity is in a vacuum. (a) Characterization of a single cavity ringdown event with an incident optical power of 0.4 mW. The residual between the measurement and exponential fit is plotted in the bottom panel, and the SNR of the ringdown signal is 162. (b) Allan deviation analysis of the ringdown time measurement. (c) Continuous measurement of more than 300 ringdown events.

Fig. 5. High-resolution and high-sensitivity molecular spectroscopy of trace amount of N₂O. (a) Measured absorption spectrum of 10 ppb N₂O. The residual of the spectral fitting by the Lorentzian function is plotted in the bottom panel. (b) Comparison of the CRDS measurements with the nominal concentrations of N₂O (1–10 ppb). The horizontal error bar represents the uncertainty in the nominal gas concentration, while the vertical error bar (too small to be visible) represents the 1−σ standard deviation of 300 CRDS measurements.