July 25, 2024 Publication number: 232558 EP

This invention relates to a novel spectroscopic apparatus and method for measuring the spectral response of a sample, in particular of a biological sample to pulsed mid-infrared (MIR) radiation. The apparatus irradiates a sample with probe pulses and detects the sample response to the probe light in a temporally or spectrally resolved manner. The core of the invention is the use of one or more quantum cascade lasers (QCLs) to amplify either the probe-light pulses irradiating the sample, or to amplify the probe light and response after interaction with the sample. The QCL-based amplification allows to increase the measured signal, and thus, the detection sensitivity. While QCLs have typically limited bandwidth, the invention is also suitable for spectrally broadband probe pulses by employing a combination of several QCLs with spectrally shifted amplification windows, with the possibility to shape the amplified spectrum. Temporal gating of the amplification process further allows temporal shaping of the amplified probe pulses and/or sample response.

 Background

Infrared spectroscopy in the range 3-25 μm spectroscopy excites molecular vibrations and is sensitive to the chemical composition and 3-dimensional structure of the molecules in a sample. Using broadband mid-infrared (MIR) laser pulses enables new types of infrared spectroscopy with enhanced sensitivity. For example, it has recently been shown that laser-based field-resolved spectroscopy based on femtosecond mid-infrared laser pulses can achieve higher dynamic range, sensitivity and specificity for molecular detection when compared to current state-of-the-art Fourier transform infrared spectroscopy [2]. Despite its success and the prospect to reach multi-octave spectral coverage in the future, broadband laser-based MIR spectroscopy still faces several limitations that need to be overcome to fully harness the potential of the technique:

1. The intensity of the MIR driving pulses and the corresponding strength of the molecular response are limited by the low efficiency of the nonlinear MIR generation processes. For example, for intra-pulse difference frequency generation that provides a particularly phase-stable pulse, the efficiency is only 0.1-3% [3-5].
2. MIR generation generally relies on phase matching in nonlinear crystals, which produces spectra with non-uniform spectral density across the targeted wavelength range.
3. The current achievable optical signal/noise ratios start to reach the dynamic range limitations of detection electronics [2].  

The present invention overcomes the aforementioned restrictions by the amplification of MIR pulses based on quantum cascade lasers (QCLs). Such amplification can boost the MIR power from tens-of-milli-Watt range [2] to the multi-Watt level. The process can also be spectrally and temporally tailored to optimally capture the molecular fingerprints. 

Technology

The invention uses QCLs to amplify the broadband MIR pulses. The emission properties of QCLs can be designed by creating specific quantum well structures with semiconductor-multilayer sequences, providing a plethora of powerful and customizable on-chip lasers [6,7]. As the spectral bandwidth of a single QCL unit is currently limited to around 5 % of the center wavelength, the invention uses a combination of QCLs with shifted emission wavelengths, either in parallel or sequentially, to amplify the full spectrum of a broadband MIR pulse (Fig. 1). Such section-wise spectral amplification can be applied before or after interaction of the probe with the sample and further provides the possibility to tailor the output pulse spectrum, for example to flatten the pulse spectrum, to enhance the spectral wings, or to increase the power spectral density at frequencies of expected molecular resonances. 

Fig. 1: Segmental amplification of ultrabroad mid-IR spectrum in the quantum cascade lasers (QCLs).

In the time domain, a sample excited by an ultrashort broadband MIR-probe driving pulse emits a response signal in the wake of the excitation pulse (Fig. 2, bottom).  The invention allows time gating by using a radio frequency (RF) setup to generate RF pulses, synchronized with the MIR probe pulses that switch on/off the gain of the QCL amplifier. In particular, QCL amplification can be applied after the interaction of the probe light with the sample, and the time-gated amplification can be shifted such that preferentially the sample-response pulse is amplified (Fig. 2, top). This mode of operation selectively enhances the sample-specific spectroscopic signal with respect to the probe driving-pulse background, increasing the spectroscopic contrast and sensitivity.

Fig.2: Temporally gated QCL amplification. In this configuration, QCL amplification is synchronized to the MIR-probe driving pulse by an RF setup. After interaction with the sample, a sample-response pulse follows the probe driving pulse. By adjusting the time window of amplification, temporal sections of the probe or sample response can be selectively amplified.

The second application of time-gated QCL amplification acts on the full MIR waveform, either before or after sample interaction, with the on/off switching of amplification synchronized with the delay in a multi-pulse scanning experiment. In field-resolved detection with electro-optic sampling, exemplary shown in Fig. 3(a), the MIR electric-field waveform is mapped by the electro-optic effect, and the measurement is performed by scanning the delay between the MIR waveform and a gate pulse, and acquiring the EOS signal at each delay. As shown in Fig.3(a), the measured EOS waveform consists of data points that correspond to at least one laser shot each. Depending on the detection bandwidth and scan speed each data point may also be the time-averaged result of multiple laser shots. The time span between each data acquisition is thus at least equal or larger than the pulse-to-pulse temporal separation given by the repetition rate of the laser, typically longer than multiple nanoseconds, so that time-gated amplification can be switched on and off from one data acquisition point to the next. By switching on QCL amplification only at delays after the main pulse, it selectively enhances the signal measured for the molecular response in the wake of the pulse. In this multi-pulse application, the QCLs amplify the full MIR waveform (Fig. 2(b)), but the delay-dependence of the amplification leads to an increased sample response signal in the measured EOS waveform. This application of time gating is not limited by the time scale of the sample response, and is applicable also to picosecond or shorter MIR waveforms.

Fig.3 (a) EOS measurement controlled by the delay line. The black circle represents different measurements with different time delay between the probe and the MIR pulse. The gray bars on top show the temporal range where QCL is switched on/off. (b) RF pulse and EOS trace in the laboratory time.

Advantages of this invention

• Increased signal level and sensitivity in laser-based infrared spectroscopy.
• Selective spectral shaping: enhance spectroscopic features in specific frequency ranges.
• Selective temporal shaping: enhance spectroscopic features in specific time intervals.
• Increased spectroscopic contrast and dynamic range by time-gated amplification.

Applications

• Ultra-sensitive field-resolved infrared spectroscopy.
• Sensitivity and contrast enhancement in laser-based Fourier-transform infrared spectroscopy.
• Dual-comb spectroscopy.
• Infrared molecular fingerprinting.

References
[1] Lasch, P. & Kneipp, J. Biomedical Vibrational Spectroscopy (Wiley, 2010).
[2] Pupeza, Ioachim, et al. Field-resolved infrared spectroscopy of biological systems. Nature 577.7788, 52-59 (2020).
[3] Zhang, Jinwei, et al. Intra-pulse difference-frequency generation of mid-infrared (2.7–20 μm) by random quasi-phase-matching. Optics letters 44.12, 2986-2989 (2019).
[4] Wang, Qing, et al. Broadband mid-infrared coverage (2–17 μm) with few-cycle pulses via cascaded parametric processes. Optics letters 44.10 (2019): 2566-2569.
[5] Novák, Ondřej, et al. Femtosecond 8.5 μm source based on intrapulse difference-frequency generation of 2.1 μm pulses. Optics letters 43.6, 1335-1338 (2018).
[6] Williams, B. Terahertz quantum-cascade lasers. Nature Photon 1, 517–525 (2007).
[7] Rauter, Patrick, and Federico Capasso. Multi‐wavelength quantum cascade laser arrays. Laser & Photonics Reviews 9.5, 452-477(2015).