Overcoming sensitivity-bandwidth trade-off in mid-infrared spectroscopy by a microresonator-anchored swept laser

Abstract

Optical frequency combs have revolutionized high-precision spectroscopy, yet an intrinsic trade-off between spectroscopic signal-to-noise ratio (sSNR) and measurement bandwidth ($B$) fundamentally constrains sensitive, broadband measurements. While broadband swept lasers offer a potential solution, generating broadband, ultrafast and linearly sweeping lasers with a narrow linewidth remains a significant challenge, particularly in the fingerprint mid-infrared (mid-IR) band. Here we overcome this limitation by using a microresonator-anchored ultrafast sweeping Fourier domain mode-locked (FDML) laser for mid-IR spectroscopy. We introduce a dual-microresonator-anchor approach: a microcomb provides frequency calibration and a high-Q microresonator resolves the instantaneous FDML lasing lineshape. The strategy enables accurate correction of the FDML laser's sweep nonlinearity and broad linewidth in the near-IR, allowing the FDML laser to function as a high-fidelity mid-IR light via difference frequency generation. The system achieves a record sSNR$\times$$B$ of 1.3$\times$10$^5$ THz$\cdot \sqrt{\rm Hz}$ and methane sensing precision of 9 ppb$\cdot$m$\cdot$$\sqrt{\rm s}$, while retaining GHz resolution to distinguish methane isotope. We further demonstrate broadband, coherent swept laser phase spectroscopy in the mid-IR, tolerating losses up to 78 dB. This work leverages advances in integrated photonics to overcome the fundamental limitations of precision spectroscopy, paving the way for next-generation, broadband, and ultra-sensitive mid-IR spectroscopic sensing systems.

Figures (4)

• Figure 1: Concept and performance of microresonator-anchored mid-IR FDML spectroscopy. a, Microresonator-anchored DFG FDML laser can be used for precise mid-IR spectroscopy, breaking the universal trade-off between sSNR and $B$ for comb spectroscopy. With dual-microresonator-calibration, the FDML laser features high power, broad bandwidth and fast scan rate, as well as high effective high sweep linearity and narrow linewidth, making it invaluable for many applications. b, Experimental setup for microresonator-anchored mid-IR FDML spectroscopy. An integrated Si$_3$N$_4$ soliton microcomb calibrates the FDML lasing frequency, while an AlN microresonator with a drop-port measures the lasing lineshape. FBS, flip beam splitter; FM, flip mirror; BPD, balanced photodetector. c, Mid-IR FDML laser spectra generated via DFG. Numerous gas samples are covered within the available lasing bandwidth. d, Mid-IR FDML spectroscopy performance compared with DCS (including near-IR DCS) reports. Open star means measurement stitching three 1 $\mu$m pump wavelengths. e, Mid-IR FDML laser performance compared with other mid-IR swept lasers, showing an excellent combination of bandwidth and scan rate.
• Figure 2: FDML laser lineshape measurement and mid-IR spectroscopy. a, AlN chip and transmission measured at its drop-port. The microresonator has a linewidth of 0.36 GHz based on a Lorentzian fit. b, Measured FDML laser lineshape and the corresponding linewidth. The inset shows zoom in of the measured lineshape. The bottom panel shows the linewidth of the FDML laser, which generally follows the chirp rate of the FDML laser (dashed curve). c, Measured absorption spectrum of a gas mixture of methane and ethane spanning 9.5 THz by stitching three 1.06 $\mu$m CW pump wavelengths. The gray and blue curves show the reference spectra without and with accounting for the FDML laser lineshape. The bottom panel shows the residual error between the measured result and the reference spectra with or without lineshape correction. d, Zoom in of the dashed box in panel c. e, sSNR under different FDML lasing bandwidths, all scaling as $\sqrt{t}$ ($t$ is the measurement time) and showing similar sSNR. The dashed lines are sSNR for reported DCS systems. f, Measurement precision of methane and ethane concentration evaluated by Allan deviation using different rovibrational branches annotated in panels c, d. The inset shows the measurement precision can be enhanced by using multi-branches for fit. The normalized precision improves as 1/$\sqrt{N}$ ($N$ is the number of used branches).
• Figure 3: Mathane isotope spectroscopy.a, Absorption spectrum measured by the DFG FDML laser and reference spectra with and without considering laser lineshape correction. The residual error greatly reduces by including lineshape correction (bottom panel). b, c, Zoom in of the boxed absorption branches in panel a. Absorption branches from $^{12}$CH$_4$ and $^{13}$CH$_4$ overlap in this measured band. The measured absorption spectra only agree with the reference spectra when accounting for the FDML laser lineshape. d, Gas concentrations of $^{12}$CH$_4$ and $^{13}$CH$_4$ derived by using the absorption spectra in b. The right panels show the distribution of the gas concentration.
• Figure 4: Coherent phase spectroscopy.a, Experimental setup for coherent phase spectroscopy using the microresonator-anchored FDML laser. Neutral density filters (NDFs) with different attenuation levels were inserted to evaluate the system under different received signal powers. PD, photodetector; BS, beam splitter. b, c, Measured signals using coherent or incoherent detection scheme. The labeled signal powers are total signal power on both detectors. d, Measured phase spectroscopy over 9 THz range by stitching three 1.06 $\mu$m pump wavelengths; the bottom panel shows the residual error with and without including the lineshape correction. e, Phase spectra measured by a single shot and averaged over 20 ms, with a signal arm power of 7 nW and a pump wavelength of 1048 nm. f, sSNR of the absorption phase spectra measured (averaged in 2 or 20 ms) under different NDF attenuation factors. The inset shows the tolerable loss versus averaging time when taking sSNR=1 as the detection limit.