Open Access
Add to BibSonomy
Abstract
A sub-picosecond mid-infrared laser based on hollow-core silica fiber is demonstrated for the first time. By using deuterium-filled revolver fiber as an active medium, we realized efficient two-cascade Raman conversion 1.03 → 1.49 → 2.68 µm pumped by chirped pulses of a femtosecond ytterbium laser. The gas fiber Raman laser generates ∼920 fs pulses at 2.68 µm with output pulse energy as high as 10 µJ. It is shown that SRS can dominate other nonlinearities even in highly transient regime implemented in the mid-IR. The approach used may be applied to develop mid-IR laser sources of various types, such as frequency combs, supercontinuum and few-cycle pulse sources.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Mid-infrared (mid-IR) femtosecond and picosecond lasers are required for a wide range of applications spanning from material processing to metrology and biomedicine. Various rapidly advancing techniques, such as dual-comb spectroscopy, supercontinuum and high harmonic generation, need new developments of ultrafast lasers. The lasers of particular interest are those operating at wavelengths λ > 2.4 µm, which are not accessible by mature silica-core fiber lasers.
Various approaches are now used to generate ultrashort pulses (USP) in the mid-IR [1]. For example, remarkable progress was demonstrated in direct lasing based on Cr2+ and Fe2+ ions hosted in ZnSe or ZnS crystals [2–8]. Another widely used approach relies on frequency conversion of near-IR USP by using nonlinear crystals [9,10]. Both approaches can offer wide wavelength tunability, few-cycle pulse duration, and mJ-level pulse energy. However this is achieved at the expense of complex design of the laser system.
To reduce the complexity of mid-IR ultrafast lasers, fiber-based solutions are desirable. Significant progress was made in rare-earth-doped fluoride fiber lasers, in which mode-locked operation at 2.8 µm [11,12] and tuning range from 2.8 to 3.6 µm [13] was reported. Output pulse energy as high as 37 nJ and peak power of about 200 kW was achieved [13]. However scaling of the pulse energy is a challenge because of detrimental nonlinear effects in the soft-glass core of a fiber.
An alternative approach to develop ultrafast lasers in the mid-IR may rely on the concept of gas fiber lasers (GFL) [14,15]. An active medium of such lasers is a gas-filled hollow-core fiber (HCF), which not only supports high-energy pulse propagation, but also can serve as a nonlinear pulse compressor to produce few-cycle optical pulses.
Particularly promising candidates are GFLs based on stimulated Raman scattering (SRS) [15]. Without the need for phase matching, the Raman GFLs could potentially downshift the frequency of well-developed near-IR frequency combs by a precisely fixed amount that equals to the Stokes shift ΩR of a chosen gas. Keeping in mind the large vibrational Stokes shifts of the lightest molecular gases (ΩR = 4155, 2987, and 2917 cm−1 for H2, D2, and CH4, respectively), the mid-IR frequency combs may be realized by very few SRS cascades.
Up to now, the Raman GFLs at λ > 2.4 µm were only reported in nano- and picosecond pulsed regimes [16–22]. Transform-limited 12-ps-long pump pulses were the shortest one used so far [22]. Moving to femtosecond pump pulses results in a highly transient SRS regime that requires high intensity to reach the Raman threshold. As a result, competing nonlinear effects such as self-phase (SPM) and cross-phase (XPM) modulation come into play, thus impeding the SRS.
To suppress the unwanted nonlinear effects the pump intensity can be reduced by chirping the frequency of pump pulses. The Stokes pulses generated by SRS replicate the chirp of the pump and can be compressed to transform-limited duration, if needed [23]. This technique was demonstrated in bulk gas cells and used to generate fs-pulses in the visible and near-IR spectral range [24–29]. Very recently, chirp-assisted vibrational SRS has been tested in gas-filled HCFs [30–33], where pulse duration as short as 590 and 39 fs was achieved at the wavelength of 1.46 and 1.8 µm, respectively. However fs-pulsed Raman GFLs have never been realized in the mid-IR.
Since Raman gain is usually reduced with wavelength and Kerr nonlinearity becomes increasingly pronounced with pulse shortening, the extension of the Raman GFLs operation into mid-infrared spectral domain and, simultaneously, into femtosecond time domain is a nontrivial task that requires further investigation.
In this work, we report on the first demonstration of a sub-picosecond mid-infrared laser based on hollow-core silica fiber. By using deuterium-filled revolver-type HCF as an active medium, we realized efficient two-cascade Raman conversion 1.03 → 1.49 → 2.68 µm pumped by chirped pulses of a commercially available femtosecond ytterbium laser. The gas fiber Raman laser generates ∼920 fs pulses at 2.68 µm with output pulse energy as high as 10 µJ, which is at least two orders of magnitude higher compared with other mid-IR sub-picosecond fiber lasers demonstrated so far.
2. Experimental setup
The Raman GFL was built in a single-pass scheme (Fig. 1(a)). As a pump source we used an ytterbium laser (TETA-6, Avesta) that generates linearly polarized transform-limited 250-fs-long pulses at 1032 nm with pulse energy of up to 400 µJ. A pulse stretcher, which is factory-installed at the output of the pump source, was tuned to introduce positive frequency chirp to the optical pulses, thus producing 10-ps-long pulses at the output of the pump source.
Fig. 1. (a) Experimental setup. (b) SEM image of the revolver-type HCF cross-section. c) Simulated loss spectrum for the fundamental mode of the HCF. Spectral positions of the pump, first Stokes (S1), second Stokes (S2), and first anti-Stokes (AS1) are indicated.
A lens L1 couples the pump pulses into a 2.9-m-long piece of a revolver-type HCF [17] with core diameter of 75 µm (Fig. 1(b)). The wall-thickness of the cladding capillaries equals to 1.15 µm and, thus, defines the shape of the optical loss spectrum of the fiber (Fig. 1(c)), which has low-loss bands near the expected spectral position of the first and second Stokes waves (1490 and 2680 nm, respectively), as well as at the pump wavelength.
The theoretical losses of the fiber were calculated by the finite element method in the СOMSOL package. The calculations used the cross section of a real fiber (Fig. 1(b)) measured with an electron microscope. The loss values at the wavelengths of 1.03, 1.49 and 2.68 µm were calculated to be 0.6, 2.2 and 225 dB/km, respectively. The calculated values correspond to the losses in a fundamental mode of the fiber provided that the fiber is straight and uniform along its length. In the experiment, however, the real fiber was coiled with a 40 cm diameter, thus giving rise to some additional bent-induced attenuation, especially for the short wavelength part of the spectrum.
The HCF was filled with D2 at a pressure of 5 bar. Both ends of the fiber were hermetically sealed into small gas cells, which had 1-mm-thick fused silica (input) and sapphire (output) windows to couple/decouple the radiation. The output radiation was collimated by a ZnSe lens and, then, analyzed by powermeter (Ophir), scanning autocorrelator (IRA-MIR, Avesta), optical spectrum analyzer (AQ6317B, Ando) and Fourier-spectrometer (OSA207, Thorlabs). When needed, the radiation at pump (1032 nm) and first Stokes (1490 nm) wavelengths were filtered out by 2-mm-thick Si plate and Ge filter (WG91050-C9, Thorlabs), respectively. Additionally, a bandpass filter (FB2750-500, Thorlabs) was used to pick out the radiation at around the second Stokes wavelength only.
3. Results and discussion
Optical spectra measured at the HCF output revealed a new component at 1490 nm, when pump pulse energy coupled to the HCF exceeded EP = 23 µJ (Fig. 2(a)). This spectral peak corresponds to the Stokes wave generated by SRS of the pump on vibrations of D2 molecules (ΩR = 2987 cm−1). Increasing EP above 50 µJ made the wave at 1490 nm strong enough to generate its own vibrational Stokes at 2680 nm (Fig. 2(b)).
Fig. 2. Output spectra of the Raman GFL measured (a) with 1 nm resolution in the near-IR at EP = 60 µJ and (b) with 0.5 cm−1 in the mid-IR at EP = 80 µJ. For the mid-IR spectrum the pump wave was filtered out by 2-mm-thick Si plate to avoid equipment damage.
Note, we have not observed rotational SRS peaks (ΩR = 414 cm−1), since this nonlinear process was suppressed by maintaining linear polarization of the pump along the HCF length [17]. Moreover, the output spectra are free from vibrational anti-Stokes component (788 nm), as it was suppressed by phase mismatch associated with close vicinity of the high-loss spectral band of the HCF, where dispersion changes very strongly (Fig. 1(c)). At the same time, any significant amount of unwanted nonlinear spectral broadening is also avoided by using positively chirped 10-ps-long pump pulses. Thus, pure vibrational Raman generation at 2.68 µm has been realized.
The temporal properties of the 2.68 µm output pulses were investigated by scanning autocorrelator. Two modes of the Raman GFL operation can be distinguished. One operation mode is observed when pump pulse energy exceeds ∼120 µJ. In this mode the autocorrelation function (ACF) shows that the main peak at zero delay is accompanied by almost structureless background, which grows with the pump energy and is distributed over delay values as large as 20 ps (Fig. 3, blue). Such ACF shape may indicate pulse breakup caused by competing nonlinear effects. A reliable measurement of the pulse duration can hardly be done in this regime.
Fig. 3. Intensity autocorrelation of the second Stokes pulses at 2.7 µm measured at coupled pump pulse energy of EP = 90 µJ (black) and EP = 220 µJ (blue). Fourier transform of the second Stokes spectrum shown in Fig. 4 is also plotted for comparison (red).
However, at pump pulse energy EP < 120 µJ there is a second operation mode, in which the background is strongly suppressed (Fig. 3, black). In this case the main peak is clearly distinguished on the ACF and corresponds to pulse duration of 920 fs, if Gaussian pulse shape is assumed. In addition, the ACF has sidelobes, the strongest of which appear at delay values of about τ ≈ 3, 7.5 and 11 ps and are highly reproducible in position (Fig. 3, black).
To gain insight into the origin of the ACF sidelobes, the spectrum of the second Stokes wave at 2.68 µm can be considered in more details. Measured with 0.5 cm−1 resolution, the shape of the spectrum is strongly indented (Fig. 4, red). Importantly, most of the dips on the spectrum correspond to absorption peaks of water vapor (Fig. 4, blue), which has strong absorption band near 2.7 µm. The source of H2O molecules is a 3-m-long free-space path that was passed by the laser beam before the beam reached the Fourier-spectrometer and autocorrelator. Carbon dioxide is another ambient air molecule that has absorption band near 2.7 µm, however water vapor absorption is more pronounced and, in general, explains the shape of the spectrum observed (Fig. 4). Overall, the Raman GFL enabled the detection of about 15 absorption lines of H2O molecules.
Fig. 4. Measured spectrum (red) of the second Stokes pulses when 80 µJ of pump pulse energy was coupled to a 2.9-m-long HCF filled by D2 at a pressure of 5 bar. The free-space path from HCF output to the spectrometer was about 3 m. Based on HITRAN database, transmittance of water vapor is also shown (blue) for the ambient conditions during the measurement.
The complex spectral shape of the pulses must affect the pulse temporal properties. By taking inverse Fourier transform of the measured spectrum (Fig. 4, red) a quantity known as electric field autocorrelation Γ(2)(τ) = ∫E(t)·E*(t-τ)dt was calculated (Fig. 3, red) [34]. The field ACF contains exactly the same information as the spectrum does, so it misses spectral phase data and could not be a measure of the pulse duration. Nevertheless, the calculated field ACF reproduces nicely the temporal positions of all the experimentally observed sidelobes on intensity ACF A(2)(τ) ∼ ∫I(t)·I(t-τ)dt that was measured by scanning autocorrelator (Fig. 3, black).
This can be understood by taking into account that each pair of peaks observed in the spectrum (Fig. 4, red) and separated in wavenumber by Δν should produce temporal beating with a period of τ = 1/(c·Δν), where c is speed of light. Although complex in structure, the spectrum is formed by transmission through water vapor and, thus, should have many peaks with almost equal separations Δν that are governed by combinations of rotational frequencies of H2O molecules. As a result, the corresponding beating periods τ should be detected as a sidelobes in both field and intensity ACFs. For example, the ACF sidelobe at delay value of τ ≈ 3 ps (Fig. 3, red and black) correlates well with the beating period between the two groups of spectral peaks located at 3729.3 and 3739.8 cm−1 in Fig. 4 (red). These groups of peaks are even better illustrated in Fig. 6 (red curve, the most intense components at δν ≈ 17.3 and 27.7 cm−1), where the spectrum measured with lower resolution is shown.
To understand the influence of spectral phase a numerical modeling was done. The spectral amplitude |E(ν)| of the laser pulse was modeled as a Gaussian function multiplied by amplitude transmission of water vapor. The width of the Gaussian function was chosen to provide spectral intensity bandwidth of 20 cm−1 (full width at half maximum). As a result, the intensity spectrum S(ν) = |E(ν)|2 of the modeled pulse (Fig. 5(a), black) fits nicely the shape of measured pulse spectrum (Fig. 5(a), red). Now, by using modeled spectral amplitude |E(ν)| and applying various assumptions about the spectral phase, the temporal properties of the pulses can be calculated.
Fig. 5. (a) The simulated spectral shape of the output pulses at 2680 nm (black line). Reproduced from Fig. 4, the experimental spectrum is also shown for comparison (red). (b) The intensity ACF calculated for the modeled pulses assuming that the pulses are transform-limited (black solid line) and linearly chirped to have 2-ps-long duration (blue dashed line). Reproduced from Fig. 3, the experimental ACF is also shown for comparison.
Assuming that the spectral phase is constant across the spectrum (i.e. the pulse is transform-limited), an intensity ACF A(2)(τ) was calculated (Fig. 5(b), black). As can be seen, the sidelobes appear immediately on the ACF and correlate well with the experimental data in temporal position (Fig. 5(b), red). Since pure Gaussian pulse spectrum does not produce any sidelobes on ACF, the sidelobes can be explained only by the influence of H2O absorption peaks, which modulate the spectral and temporal properties of the pulses after the output end of the Raman laser. Compared with modeled ACF sidelobes, the measured sidelobes are more intense. This is probably due to the fact that experimental data contains sharper spectral features compared with the modeled spectrum.
If the spectral phase is assumed to be quadratic function of wavenumber (i.e. the pulse is linearly chirped), the sidelobes on the intensity ACF become lower in magnitude and disappear as the value of chirp increases. An example of such behavior is illustrated in Fig. 5(b) (blue dashed line), where intensity ACF modeled for a 2-ps-long linearly chirped pulse is shown. The sidelobe-free ACF of a chirped pulse may be explained as follows: produced by different pairs of equally separated spectral peaks, the temporal pulsations will have the same period τ but different initial phases, thus cancelling each other. In this way, sharp ACF sidelobes should be transformed into smooth background as the pulse chirp increases. In fact, this explanation should be applicable to any nonlinear spectral dependence of the pulse spectral phase. Moreover, in view of this discussion, the presence of strong sidelobes on the measured ACF (Fig. 5(b), red) indirectly indicates that pulses generated at 2680 nm are almost transform-limited.
Additional data about cascaded Raman GFL operation can be extracted from the spectral shape of the pulses (Fig. 6). Low frequency components, which propagate at the leading edge of the positively chirped pump pulse (Fig. 6, dashed black), remain unconverted and are detected as a residual pump wave at the GFL output (Fig. 6, blue). The spectral shape of the first Stokes pulses (Fig. 6, green) also indicates some amount of positive chirp, as conversion to the second Stokes is observed only for high frequency part of the spectrum, which should be at the trailing edge of the 1490 nm pulses. Finally, the second Stokes pulses with ∼20 cm−1 bandwidth are generated (Fig. 6, red). Although pulse duration of 920 fs was measured in the experiment, the spectral bandwidth indicates that ∼750-fs-long transform-limited pulses can be produced at 2680 nm.
Fig. 6. Spectral shape of 1.03 µm (blue), 1.49 µm (green) and 2.7 µm (red) pulses measured at the output of the Raman GFL. Pump pulse energy was 84 µJ. The shape of input pump spectrum is also shown (dashed black). On the X-axis δν represents the wavenumber ν shifted by νP, νP - ΩR, and νP - 2ΩR for the pump, first and second Stokes, respectively, where νP is the central frequency of the pump spectrum, and ΩR = 2987 cm−1 is the vibrational Stokes shift of D2 molecules. The Y-axis is normalized so that the area under each curve is proportional to the number of photons in the corresponding pulse.
Note, the pulse duration may be reduced even further, since gas-filled hollow-core fibers provide an efficient way not only for SRS, but also for nonlinear pulse compression. Both functionalities may be combined in a single piece of a HCF, as have been recently shown in [32,33], where 39-fs pulses was generated at λ = 1.8 µm. Thus, Raman gas fiber lasers have great potential for generation of few-cycle pulses in the mid-infrared range.
Quantum efficiency of the two-cascade Raman GFL is presented in Fig. 7(a), where normalized photon numbers are shown as a function of the pump pulse energy coupled into the HCF. As high as 28% of the input photons were converted to the second Stokes wave (Fig. 5(a), red). The pulse energy at 2.68 µm grows monotonically without saturation and reaches a 30-µJ-level (Fig. 7(b)). Since background-free ACF was observed only at pump energy below 120 µJ (Fig. 7(b), shaded region), the energy of at least 10 µJ was generated in 920-fs-long pulses at 2.68 µm. We believe that conversion efficiency and energy of the pulses with background-free ACF may be increased by using pump pulses with stronger chirp. Such approach could help to suppress unwanted nonlinear effects, but may require post-compression stage to achieve sub-picosecond output pulses. An optimization of the fiber length and the gas pressure may also increase the output pulse energy.
Fig. 7. (a) The output number of photons at pump (blue), first (green) and second Stokes (red) wavelengths as a function of coupled pump pulse energy. The curves are normalized to the number of coupled pump photons. The total photon number is also shown (dashed black). (b) The output pulse energy at 2.7 µm as a function of coupled pump pulse energy. Shaded region corresponds to operation mode where background-free ACF was observed.
The fraction of residual pump and first Stokes photons was measured to be 13 and 23%, respectively (Fig. 7(a), blue and green). These photons correspond to that part of the leading edge of the pulses where the instantaneous peak power is below the SRS threshold (Fig. 6, blue and green). The total number of photons detected at the HCF output is 64% of the input pump photons (Fig. 7(a), dashed black), which means that 36% of photons are lost.
Such a large amount of lost photons could not be explained by the values of optical losses that were calculated for the fundamental mode of the fiber (Fig. 1(c)). However, at least two reasons may be responsible for additional photon loss. First, small non-uniformities of the real fiber cross section along the fiber length can give rise to higher optical attenuation of the fundamental mode compared to the case of ideally uniform fiber. This is especially important at the second Stokes wavelength, where optical loss has the highest value (0.22 dB/m at 2.68 µm) even if the fiber is uniform. The second possible reason includes excitation of the higher order modes (HOM) that could have much higher attenuation compared with the fundamental mode. The influence of HOM should be more pronounced at the pump wavelength, since modal content of Stokes waves is usually purified from HOM because of low spatial overlap of the HOM and the fundamental mode. Detailed study of the modal content and optimization of the conversion efficiency is a matter of future work.
4. Conclusion
To conclude, we have experimentally demonstrated that gas fiber Raman lasers can be efficient sources of ultrashort high-energy pulses in the mid-infrared spectral range. A simple setup, which consist of a commercially available femtosecond laser at 1.03 µm and D2-filled revolver-type HCF, enabled us to generate almost transform-limited sub-picosecond pulses with duration as short as ∼920 fs and energy as high as ∼10-µJ at the wavelength of 2.68 µm. We showed that SRS can dominate other nonlinearities even in highly transient regime implemented in the mid-IR. The approach used in this work may be applied to develop mid-IR laser sources of various types, such as frequency combs, supercontinuum and few-cycle pulse sources, each of which are of high importance for numerous applications in science, biomedicine and technology.
Funding
Russian Science Foundation (19-12-00361).
Disclosures
The authors declare no conflict of interest.
References
1. K. L. Vodopyanov, Laser-based Mid-infrared Sources and Applications (John Wiley & Sons, Inc.2020).
2. S. Vasilyev, I. Moskalev, M. Mirov, S. Mirov, and V. Gapontsev, “Multi-watt mid-IR femtosecond polycrystalline Cr2+:ZnS and Cr2+:ZnSe laser amplifiers with the spectrum spanning 2.0–2.6 µm,” Opt. Express 24(2), 1616–1623 (2016). [CrossRef]
3. E. Slobodchikov, L. R. Chieffo, and K. F. Wall, “High peak power ultrafast Cr:ZnSe oscillator and power amplifier,” Proc. SPIE 9726, 972603 (2016). [CrossRef]
4. N. Tolstik, E. Sorokin, and I. T. Sorokina, “Graphene mode-locked Cr:ZnS laser with 41 fs pulse duration,” Opt. Express 22(5), 5564–5571 (2014). [CrossRef]
5. N. Nagl, S. Gröbmeyer, V. Pervak, F. Krausz, O. Pronin, and K. F. Mak, “Directly diode-pumped, Kerr-lens mode-locked, few-cycle Cr:ZnSe oscillator,” Opt. Express 27(17), 24445–24454 (2019). [CrossRef]
6. S. Vasilyev, I. Moskalev, M. Mirov, V. Smolski, S. Mirov, and V. Gapontsev, “Ultrafast middle-IR lasers and amplifiers based on polycrystalline Cr:ZnS and Cr:ZnSe,” Opt. Mater. Express 7(7), 2636–2650 (2017). [CrossRef]
7. S. Vasilyev, V. Smolski, J. Peppers, I. Moskalev, M. Mirov, Y. Barnakov, S. Mirov, and V. Gapontsev, “Middle-IR frequency comb based on Cr:ZnS laser,” Opt. Express 27(24), 35079–35087 (2019). [CrossRef]
8. A. V. Pushkin, E. A. Migal, S. Tokita, Y. V. Korostelin, and F. V. Potemkin, “Femtosecond graphene mode-locked Fe:ZnSe laser at 4.4 µm,” Opt. Lett. 45(3), 738–741 (2020). [CrossRef]
9. G. Andriukaitis, T. Balčiūnas, S. Ališauskas, A. Pugžlys, A. Baltuška, T. Popmintchev, M.-C. Chen, M. M. Murnane, and H. C. Kapteyn, “90 GW peak power few-cycle mid-infrared pulses from an optical parametric amplifier,” Opt. Lett. 36(15), 2755–2757 (2011). [CrossRef]
10. S. Wandel, M.-W. Lin, Y. Yin, G. Xu, and I. Jovanovic, “Parametric generation and characterization of femtosecond mid-infrared pulses in ZnGeP2,” Opt. Express 24(5), 5287–5299 (2016). [CrossRef]
11. S. Duval, M. Bernier, V. Fortin, J. Genest, M. Piché, and R. Vallée, “Femtosecond fiber lasers reach the mid-infrared,” Optica 2(7), 623–626 (2015). [CrossRef]
12. M. R. Majewski, R. I. Woodward, and S. D. Jackson, “Ultrafast mid-infrared fiber laser mode-locked using frequency-shifted feedback,” Opt. Lett. 44(7), 1698–1701 (2019). [CrossRef]
13. S. Duval, J.-C. Gauthier, L.-R. Robichaud, P. Paradis, M. Olivier, V. Fortin, M. Bernier, M. Piché, and R. Vallée, “Watt-level fiber-based femtosecond laser source tunable from 2.8 to 3.6 µm,” Opt. Lett. 41(22), 5294–5297 (2016). [CrossRef]
14. A. V. V. Nampoothiri, A. M. Jones, C. Fourcade-Dutin, C. Mao, N. Dadashzadeh, B. Baumgart, Y. Y. Wang, M. Alharbi, T. Bradley, N. Campbell, F. Benabid, B. R. Washburn, K. L. Corwin, and W. Rudolph, “Hollow-core Optical Fiber Gas Lasers (HOFGLAS): a review,” Opt. Mater. Express 2(7), 948–961 (2012). [CrossRef]
15. A. V. Gladyshev, A. F. Kosolapov, A. N. Kolyadin, M. S. Astapovich, A. D. Pryamikov, M. E. Likhachev, and I. A. Bufetov, “Mid-IR hollow-core silica fibre Raman lasers,” Quantum Electron. 47(12), 1078–1082 (2017). [CrossRef]
16. A. V. Gladyshev, A. F. Kosolapov, M. M. Khudyakov, Y. P. Yatsenko, A. N. Kolyadin, A. A. Krylov, A. D. Pryamikov, A. S. Biriukov, M. E. Likhachev, I. A. Bufetov, and E. M. Dianov, “2.9, 3.3, and 3.5 µm Raman Lasers Based on Revolver Hollow-Core Silica Fiber Filled by 1H2/D2 Gas Mixture,” IEEE J. Sel. Top. Quantum Electron. 24(3), 1–8 (2018). [CrossRef]
17. M. S. Astapovich, A. V. Gladyshev, M. M. Khudyakov, A. F. Kosolapov, M. E. Likhachev, and I. A. Bufetov, “Watt-Level Nanosecond 4.42-µm Raman Laser Based on Silica Fiber,” IEEE Photonics Technol. Lett. 31(1), 78–81 (2019). [CrossRef]
18. W. Huang, Y. Cui, Z. Li, Z. Zhou, Y. Chen, and Z. Wang, “Diode-pumped single-pass tunable mid-infrared gas Raman source by methane-filled hollow-core fiber,” Laser Phys. Lett. 16(8), 085107 (2019). [CrossRef]
19. Z. Li, W. Huang, Y. Cui, and Z. Wang, “Efficient mid-infrared cascade Raman source in methane-filled hollow-core fibers operating at 2.8 µm,” Opt. Lett. 43(19), 4671–4674 (2018). [CrossRef]
20. Y. Wang, M. K. Dasa, A. I. Adamu, J. E. Antonio-Lopez, M. S. Habib, R. Amezcua-Correa, O. Bang, and C. Markos, “High pulse energy and quantum efficiency mid-infrared gas Raman fiber laser targeting CO2 absorption at 4.2 µm,” Opt. Lett. 45(7), 1938–1941 (2020). [CrossRef]
21. A. Benoît, E. Ilinova, B. Beaudou, B. Debord, F. Gérôme, and F. Benabid, “Spectral–temporal dynamics of high power Raman picosecond pulse using H2-filled Kagome HC-PCF,” Opt. Lett. 42(19), 3896–3899 (2017). [CrossRef]
22. L. Cao, S. Gao, Z. Peng, X. Wang, Y. Wang, and P. Wang, “High peak power 2.8 µm Raman laser in a methane-filled negative-curvature fiber,” Opt. Express 26(5), 5609–5615 (2018). [CrossRef]
23. C. Jordan, K. A. Stankov, G. Marowsky, and E. J. Canto-Said, “Efficient compression of femtosecond pulses by stimulated Raman scattering,” Appl. Phys. B 59(4), 471–473 (1994). [CrossRef]
24. A. V. Konyashchenko, L. L. Losev, and S. Y. Tenyakov, “Raman frequency shifter for laser pulses shorter than 100 fs,” Opt. Express 15(19), 11855–11859 (2007). [CrossRef]
25. N. V. Didenko, A. V. Konyashchenko, P. V. Kostryukov, L. L. Losev, V. S. Pazyuk, S. Y. Tenyakov, V. Y. Molchanov, S. I. Chizhikov, and K. B. Yushkov, “40-fs hydrogen Raman laser,” Quantum Electron. 45(12), 1101–1104 (2015). [CrossRef]
26. C. Vicario, M. Shalaby, A. Konyashchenko, L. Losev, and C. P. Hauri, “High-power femtosecond Raman frequency shifter,” Opt. Lett. 41(20), 4719–4722 (2016). [CrossRef]
27. A. V. Konyashchenko, P. V. Kostryukov, L. L. Losev, and V. S. Pazyuk, “Minimum possible laser pulse duration for SRS,” Quantum Electron. 47(7), 593–596 (2017). [CrossRef]
28. A. V. Konyashchenko, L. L. Losev, and V. S. Pazyuk, “Femtosecond Raman frequency shifter–pulse compressor,” Opt. Lett. 44(7), 1646–1649 (2019). [CrossRef]
29. N. V. Didenko, A. V. Konyashchenko, and L. L. Losev, “Decreasing the amplitude of macroscopic quantum fluctuations in the case of transient SRS,” Quantum Electron. 50(9), 834–837 (2020). [CrossRef]
30. A. V. Gladyshev, M. S. Astapovich, Y. P. Yatsenko, A. F. Kosolapov, A. G. Okhrimchuk, and I. A. Bufetov, “SRS generation of femtosecond pulses in a methane-filled revolver hollow-core optical fibre,” Quantum Electron. 49(12), 1089–1092 (2019). [CrossRef]
31. D. Kergoustin, F. Amrani, B. Debord, F. Gérôme, and F. Benabid, “Infrared Ultra-Short Pulses Generation using Stimulated Raman Scattering in Gas-Filled HC-PCF,” 2019 Conference on Lasers and Electro-Optics Europe & European Quantum Electronics Conference (CLEO/Europe-EQEC), Munich, Germany, 2019, p. cd_3_5.
32. S. Loranger, D. Novoa, and P. S. J. Russell, “Efficient Chirp-Assisted SRS in H2-Filled Hollow-Core PCF for Generation of Ultrashort LP01 Pulses at 1.8 µm,” in Conference on Lasers and Electro-Optics Technical Digest 2020 (Optical Society of America, 2020), p. SM1P.1.
33. S. Loranger, P. St. J. Russell, and D. Novoa, “Generation of sub-40 fs pulses at 1.8 µm by chirp-assisted Raman scattering in hydrogen-filled hollow-core fibre,” arXiv:2006.16782v1 (2020).
34. R. Trebino, Frequency-Resolved Optical Gating: The Measurement of Ultrashort Laser Pulses (Kluwer Academic Publishers2000).
