Metrology-grade sub-Doppler spectroscopy of CHF3 at 8.6 μm

Year: 2017

Authors: Gambetta A., Vicentini E., Wang Y., Coluccelli N., Fernandez T.T., Fasci E., Castrillo A., Gianfrani L., Santamaria L., Di Sarno V.D., Maddaloni P., Laporta P., Galzerano G.

Autors Affiliation: Dipartimento di Fisica – Politecnico di Milano, Piazza Leonardo da Vinci 32, Milano, 20133, Italy; Istituto di Fotonica e Nanotecnologie – CNR, Piazza Leonardo da Vinci 32, Milano, 20133, Italy; Dipartimento di Matematica e Fisica – Seconda Università di Napoli, Viale Lincoln 5, Caserta, 81100, Italy; CNR-INO, Istituto Nazionale di Ottica, Via Campi Flegrei 34, Pozzuoli (NA), 80078, Italy

Abstract: We report on metrological-grade saturated absorption spectroscopy of CHF3 at 8.63 μm based on CW distributed-feedback QCL laser and a mid-IR self-frequency referenced optical comb. The experimental setup for the saturation spectroscopy of CHF3 is shown in Fig. 1. A CW DFB-QCL with a maximum output power of ~40 mW tunable in the wavelength range from 8.55 to 8.65 µm, in coincidence with the υ5 (asymmetric FCF bending) vibrational band of CHF3, is used to probe the gas sample. The QCL output beam is focused at the input of a 30-dB optical isolator to a diameter of 1.6~mm, by using antireflection-coated aspheric lens mounted onto the laser housing. Then the laser beam passing through a 20-cm plano-convex lens and a 50% beam splitter is coupled to a 25-cm-long stainless steel cell containing the CHF3 gas (98% purity). The transmitted beam first passes through a second 50% beam splitter to detect the single-pass CHF3 absorptions and then is back-reflected into the gas cell to implement the double pass configuration (collinear pump and probe saturated absorption technique). The probe beam is detected exploiting the reflection on the first beam splitter by a four-stage electric cooled MCT detector (10~MHz bandwidth). By modulating the QCL driving current at a frequency fm=100 kHz also a wavelength modulation method is implemented to retrieved the first derivative signal by means of a first-harmonic coherent demodulation. To absolute calibrate the QCL emission frequency a difference frequency generation optical frequency comb, covering the 8–14 µm spectral region, is used [1]. The mid-IR comb is properly tuned to a central wavelength of 8.6 µm (average output power of 4~mW in an optical bandwidth of 0.8 µm) and its repetition frequency is stabilized against a RF synthesizer locked to a GPS-disciplined Rb frequency standard (Allan deviation σ(τ)=10−11τ−1/2 and fractional frequency accuracy of 10−13, respectively, where τ is the integration time). The mid-IR optical frequency comb and QCL beams are then superimposed using a 50% ZnSe beam splitter, filtered with a 0.01-µm monochromator, and then focused onto a low-noise 200-MHz bandwidth MCT detector. Two different and complementary absolute frequency measurement strategies were implemented using the beat note signal: the first one consists in counting the beat note frequency when both the QCL and mid-IR comb sources are frequency stabilized against the sub-Doppler CHF3 transition, using the first derivative signal (see Fig. 1 b), and the Rb-frequency standard, respectively; the second method relies on a tight phase-lock of the QCL to the closest mid-IR comb tooth (closed-loop bandwidth of 500 kHz) whereas the comb repetition frequency is finely tuned to record the molecular absorption profile [2]. In both experimental approaches, preliminary fractional frequency accuracies in the determination of line centre frequencies down to 10−11, limited only by the reproducibility of the QCL frequency locking, have been obtained.

Conference title:

KeyWords: Line shape; Spectrometers; Shift coefficients