Open Access
Add to BibSonomy
Abstract
We demonstrate the generation of high-energy femtosecond pulses by a spectrally combined four-diode-pumped Ti:sapphire laser with semi-conductor saturable absorber mirror mode locking. To achieve energy scaling, the laser cavity was extended using a design based on the Herriott multipass cell. The laser operates at a 17.5 MHz repetition rate and generates pulses with energies as high as 16.3 nJ and 80 fs in duration. The signal-to-noise ratio at the fundamental frequency showed an extinction ratio of >50 dB relative to the carrier. This compact single laser was applied to video-rate forward and backward coherent anti-Stokes Raman scattering spectro-microscopy with a pixel dwell time of 122 ns, which is the lowest dwell time ever achieved, to the best of our knowledge.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Ti:sapphire (TiS) crystals with broad gain bandwidths [1] are the most attractive materials for femtosecond pulse generation in the near-infrared wavelength range. TiS femtosecond lasers have been successfully deployed in industry and basic science. In particular, TiS femtosecond lasers are widely utilized in bioimaging applications, such as two-photon microscopy, coherent Raman scattering microscopy, and optical coherence tomography [2–5]. Despite such extensive use, there are limitations to this approach because TiS femtosecond lasers are typically pumped with multiwatt frequency-doubled green lasers such as diode-pumped neodymium-doped yttrium aluminum garnet (Nd:YAG) and neodymium-doped yttrium lithium fluoride (Nd:YLF) lasers. Such pump sources make the laser system highly complex, bulky, and expensive.
High-power commercial laser diodes in the blue-green wavelength region have paved the way for an alternative that overcomes these drawbacks. Since the demonstration of the first TiS femtosecond laser using a direct blue diode [6], mode-locked laser powers have increased. Roth et al. reported a TiS laser with 101 mW output power and 111 fs pulses at 127 MHz based on two-way pumping [7]. Gurel et al. utilized two laser diodes and obtained 450 mW output power and 58 fs pulses at 418 MHz from a TiS laser using Kerr-lens mode-locking (KLM), as well as 350 mW output power and 39 fs pulses at 414 MHz using SESAM mode-locking; the corresponding pulse energies were 1 and 0.8 nJ [8]. Backus et al. reported 145 mW of output power and 13 fs pulses at 78 MHz from a TiS laser, corresponding to a pulse energy of 1.85 nJ [9]. Rohrbacher et al. also used two laser diodes emitting at 450 nm and obtained 460 mW of output power and 82 fs pulses at 92 MHz, which corresponds to a pulse energy of 5 nJ [10]. However, these pulse energies are still lower than ideal for most applications. In 2021, we reported 1 W output power and 55 fs pulses from a TiS laser at 68.8 MHz with spectrally combined three-diode-pumping and a shared cylindrical beam expander [11]. The corresponding pulse energy was 14.5 nJ in the standard configuration. At higher output powers, TiS lasers experience strong instabilities caused by enhanced nonlinear effects owing to the increased peak power [12,13]. In addition, femtosecond lasers with high pulse energy and low repetition rates are required to avoid thermal damage in various types of nonlinear bio-imaging, such as two-photon microscopy and coherent Raman scattering microscopy [14].
Here, we present the pulse energy scaling of a TiS laser by means of spectrally combined four-diode pumping and cavity length extension. The TiS laser with a SESAM for self-starting mode-locking generates 16.3 nJ, 80 fs pulses at 17.5 MHz, corresponding to 286 mW output power. A laser configuration based on a Herriott multipass cell was used to maintain the q-parameter. The overall dimensions of this laser are only 60 $\times$ 38 cm$^2$, including a 18 $\times$ 8 cm$^2$ footprint for four pump diodes with three dichroic filters and a 40 $\times$ 8 cm$^2$ footprint for a multipass cell setup. This high-energy femtosecond laser was applied to video-rate [7.5 frames per second (fps), 1024$\times$1024 pixels per frame] forward and backward multiplex coherent anti-Stokes Raman scattering (CARS) spectro-microscopy. The remainder of this paper is organized as follows. Section 2 is principally concerned with introducing the development of a four-diode-pumped femtosecond TiS laser. In section 3, the application of this laser to CARS spectro-microscopy is presented. Finally, the development of the TiS laser and the CARS spectro-microscopy results are summarized.
2. TiS laser
The TiS laser setup is shown in Fig. 1. The laser cavity had a standard and extended X-folded configuration. The standard cavity (up to the dashed line, which represents the position of the output coupler for the extended cavity alignment) consists of a SESAM, a concave mirror (CM1) with radius of curvature 30 cm for the SESAM, a flat chirped mirror (M1) that generates negative group delay dispersion (GDD), concave dichroic pump mirrors (CM2 and CM3) with radius of curvature $R$ of 10 cm, and a TiS gain medium. The Brewster-cut TiS gain medium (GT Advanced Technologies) with a 6 mm length, absorption coefficient of 2.71 cm$^{-1}$ at 532 nm, and figure-of-merit (FOM) of >150, was water-cooled on a copper mount. The standard cavity has a long arm of 515 mm (CM2 to SESAM) and a short arm of 253 mm (CM3 to dashed line). Thus, the length of the standard cavity was 875 mm, which corresponds to a repetition rate of 171.4 MHz. In addition, the Herriott multipass cell (dashed line to OC) consisted of three flat chirped mirrors (M2, M3, and M4) that generated negative GDD, one curved mirror (HM1, $R$ = 4 m), one flat mirror (HM2), and an output coupler with 5$\%$ transmission. The Herriott multipass mirrors (HM1 and HM2) had a notch for ease of beam entry and exit [15]. Owing to the zero effective length of the Herriott multipass cell configuration, the mode-locking condition of the standard cavity was preserved, but the repetition rate of this compact laser was reduced [15–17]. An intra-cavity net GDD of approximately -800 fs2 was intentionally maintained to obtain laser pulses with a suitable spectral bandwidth for CARS spectro-microscopy. A SESAM (RefleKron) for self-starting mode-locking was placed before CM3. In addition, it has a modulation depth of 1.5$\%$, low intensity reflectivity of 98$\%$, and nonsaturable loss of <0.5$\%$.
As shown in Fig. 1(a) and (b), laser beams from four diodes (Nichia Corporation), generating wavelengths of 450, 470, 492, and 526 nm, respectively, were combined and utilized as the pump source. The pump source consists of four diodes and four aspheric lenses ($f$ = 4 mm, NA = 0.6) to collimate each diode beam along the fast axis, and an 8$\times$ cylindrical telescope ($f$ = -10 mm and 80 mm) to collimate the combined beam along the slow axis. The pump source was spectrally and coaxially combined using dichroic filters (DF1, DF2, and DF3, incident angle: 45$^\circ$) and then transmitted through the cylindrical telescope, and then to the TiS gain medium. The overall transmission of the pump source (four aspheric lenses, three dichroic filters, and one cylindrical telescope) was 90$\%$. The TiS gain medium was held at a temperature of 16 $^\circ$C on a water-cooled aluminum mount. The 450, 470, 492, and 526 nm diodes emitted output powers of 4.0, 5.5, 2.0, and 1.5 W (upper power limits), respectively, as measured in front of the focusing lens (L). The total output power of the combined beam was approximately 13 W. The single-pass power absorption of the TiS gain medium was measured to be 80$\%$ when the pump power was 13 W. Compared with our previous study [11], the pump absorption for the TiS gain medium was 5.3$\%$ lower because of the lower absorption of the 450 nm LD. No power degradation or spectral shift of the combined beam was observed over 24 h.
Fig. 1. (a) Schematic of TiS laser setup. LD, laser diode; AL, aspheric lens; DF, dichroic filter; VM, visible flat mirror; L, lens; CM1, curved mirror; CM2, CM3, dichroic curved mirror; TiS, Ti:sapphire gain medium; M, flat mirror; HM, Herriott multipass mirror; OC, output coupler. (b) Spectrum of the combined pump beam. (c) Schematic of Herriott multipass mirrors with a notch and photograph of HM1 showing nine bounces. (d) Beam waist on each optical element in the cavity, from the SESAM to the OC. The dotted line, the position of the standard cavity output coupler
For the extended cavity, including the HMM mirrors, as shown in Fig. 1(a) and (c), a standard cavity was first built. The HMM mirrors were aligned using a beam from the standard cavity. In the extended cavity design with ten round trips, curved (HM1, R = 4 m) and flat (HM2) mirrors with a notch were used. Although the notch meant that one round trip was missed, as shown in the photo in Fig. 1(c) (nine bounces on HM1 are apparent), the q-parameter was preserved, as described in detail in [17]; the distance between the two HMM mirrors was 38.2 cm and the angular advance between each successive beam on each mirror was 2$\pi$/10 = 36$^\circ$. The output coupler was positioned 45 mm from the HM2 mirror. Thus, the total length of the laser cavity was 8.56 m, corresponding to a repetition rate of 17.5 MHz. This repetition rate is reduced by a factor of $\sim$10 compared with the length of the standard cavity (171.4 MHz). Using ABCD matrix calculations, the beam waist on each component in the laser cavity, from the SESAM to the OC, was obtained, as shown in Fig. 1(d). The beam waists on the TiS gain medium and the SESAM were designed to be 66 $\mu$m and 238 $\mu$m, respectively. The fluence on the SESAM was 174 $\mu$J/cm$^2$, which is $\sim$ three times higher than the saturation fluence of 60 $\mu$J/cm$^2$. After self-starting SESAM mode-locking had been initiated, the mode-locked (ML) output power was stable. Figure 2(a) shows the variation of ML output power as a function of LD pump power. The diodes were powered up in sequence as follows: 470 nm, 450 nm, 492 nm, then 525 nm. Self-mode-locking was achieved as soon as the gain medium was pumped with both 5.5 W of the 470 nm LD and either the 450 nm, 492 nm, or 525 nm LD. Thereafter, the ML output power versus the LD pump power was measured. At 5.5 W of pump power from the 470 nm LD, lasing occurred. The ML output power increased to 73 mW with the addition of the 450 nm laser, and the slope efficiency was 1.0$\%$. Sequentially, as a function of the addition of the 492 nm LD and 525 nm LD pumps, the ML output power increased to 190 and 286 mW, and the slope efficiencies were 5.4$\%$ and 5.0$\%$, respectively. The overall slope efficiency was estimated to be 3.7$\%$. Using 13 W, i.e., full power for all four pump diodes, the ML output power was obtained up to 286 mW, which corresponded to a pulse energy of 16.3 nJ, which is the highest pulse energy ever reported for a multiple-diode-pumped femtosecond TiS laser. As shown in Fig. 2(b), the root-mean-square stability of the ML output power over 24 h was 1.34$\%$. No degradation of the ML output power was observed.
Fig. 2. (a) Mode-locked (ML) output power variation as the pump power supplied by four diodes of different wavelengths was increased sequentially in the order 470 nm, 450 nm, 492 nm, then 525 nm. (b) ML output power stability over 24 h.
Finally, to verify that the laser generated stable femtosecond pulses, the ML output spectrum, pulse duration, optical pulse train, and RF spectrum were measured. As shown in Fig. 3, the ML output spectrum had a bandwidth [full-width at half maximum (FWHM)] of 11.6 nm centered at the wavelength of 808 nm. Spectral shifting was not observed over 48 h. The corresponding pulse duration was characterized using a commercial autocorrelator (MiniPD, APE GmbH), which revealed a pulse autocorrelation width of 123 fs (FWHM), corresponding to an 80 fs (FWHM) pulse duration assuming a sech$^2$ intensity profile. The oscilloscope (Teledyne Lecroy, WaveRunner 610Zi, 1GHz bandwidth) showed an ML pulse train with a period of 57.1 ns, corresponding to a repetition rate of 17.5 MHz. The RF spectrum (Teledyne Lecroy, T3SA3200 spectrum analyzer) of the ML pulses was recorded at the fundamental frequency with a resolution bandwidth of 1 kHz over a span of 1 MHz. An extinction ratio of >52 dB was measured. The inset shows a RF spectrum measured over a span of 160 MHz with a resolution bandwidth of 10 kHz. The high signal-to-noise ratio and the wide-span RF spectrum verify the existence of stable ML pulses without Q switching instability or multiple pulsing owing to the Kerr nonlinearity of the gain medium and increased intra-cavity peak power. These findings verify the suitability of the laser for multiplex CARS spectro-microscopy, which is reported in the next section.
Fig. 3. (a) Measured output spectrum over 48 h. (b) Corresponding intensity autocorrelation, (c) pulse train, and (d) RF spectrum at the fundamental frequency with a resolution bandwidth (RBW) of 1 kHz (inset: RF spectrum spanning 160 MHz with a resolution bandwidth of 10 kHz).
3. Coherent anti-Stokes Raman scattering spectro-microscopy
CARS spectro-microscopy is a powerful label-free imaging technique that can produce vibrational images with high sensitivity and selectivity compared with spontaneous Raman microscopy. Since the introduction of CARS microscopy in 1982 [18], the CARS technique, which does not require staining and molecular tagging, has become widely used in the life sciences. Multiplex CARS spectro-microscopy using two synchronized beams (pump and Stokes) [3] is more straightforward than monochromatic CARS imaging. In this study, the above-described TiS laser generating 16.3 nJ, 80 fs pulses was utilized in a home-built multiplex CARS spectro-microscopy setup, as shown in Fig. 4. As soon as the output beam was emitted from the TiS laser, the beam encountered a negative-GDD mirror pair (ChM), generating -1000 fs$^2$ per bounce to pre-compensate the GDD generated by the other optical elements (HW, OL, PCF, etc.). The total GDD generated by the ChM mirror pair was -10,000 fs$^2$. The output beam from the TiS laser was split using a polarized beam splitter. The split ratio was controlled at 150 mW (pump): 130 mW (Stokes) using a half-wave plate (HW). The pump beam was transmitted via a delay stage, Keplerian telescope ($f$ = 50 mm $\times$ $f$ = 35 mm for mode matching with the Stokes beam), bandpass filter (3 nm FWHM corresponding to a spectral resolution of $\sim$46 cm$^{-1}$), and half-wave plate. The Stokes beam was spectrally broadened by a photonic crystal fiber (NKT Photonics, FemtoWhite CARS) using two objective lenses (Olympus, 20$\times$) and polarization-controlled using half-wave plates. We achieved a coupling efficiency of 26.5$\%$. After being transmitted through the longpass filter (LPF, RG830 4 mm thick), the Stokes spectrum up to 1200 nm, corresponding to a Raman shift of 4043 cm$^{-1}$, was finally obtained, as shown in the inset of Fig. 4. The two beams were spatially superimposed on the combining mirror (CM, Semrock, LP02-808RU) and temporally synchronized using a motorized delay stage. The combined beam was sent to a sample via a scanner, a flip mirror (FM) for bright-field imaging, and a dichroic mirror (DM) for backward imaging. The scanner consists of a resonant and a galvanometric mirror: rapid scanning in the x-direction (fast axis) was performed using a resonant silver-coated mirror (Cambridge Technology, CRS, 8kHz), while a galvanometric silver-coated mirror (Thorlabs, GVS001) was scanned in the y-direction (slow axis). Finally, 1024 $\times$ 1024 pixel images were acquired at 7.5 fps. This corresponds to the short pixel dwell time of 122 ns, which is the lowest dwell time to be achieved thus far, to the best of our knowledge. The combined beam (pump and Stokes beams) was focused on the sample by a 40$\times$ objective lens (Olympus, NA 0.75), re-collimated using a 10$\times$ objective lens (Olympus, NA 0.30), and the signal was transmitted through a short pass filter (Semrock, 720SP). The forward and backward signals were simultaneously detected using photomultiplier tubes (PMT, Hamamatsu R10699) and sent to a computer for image processing using a program written in Labview (National Instruments). The filter wheel (F) was designed to hold six filters, that is, six filters were selected to allow the PMT to detect the Raman shift or two-photon absorption signals. In addition, a bright-field image was obtained using a CMOS camera, and a CARS spectrum using a spectrograph (Andor Technology, Monochromator SR-303i-A).
Fig. 4. CARS setup. M, dielectric mirror; ChM, chirped mirror; HW, halfwave plate; PBS, polarized beam splitter; TL, telescope; BPF, bandpass filter; CM, combining mirror; OL, objective lens; PCF, photonic crystal fiber; SM, silver mirror; LPF, longpass filter; SPF, shortpass filter; FM, flip mirror; F, filter-wheel.
Finally, as a demonstration of CARS spectro-microscopy using this setup, a mixture of polystyrene (PS) beads [Sigma-Aldrich, 20 $\mu$m beads and blue fluorescent ($\lambda _{emission}$ = 420 nm) 2 $\mu$m beads for two-photon (TP) absorption] deposited on a slide as samples, were imaged, as shown in Fig. 5. Figure 5(a) and (b) shows a bright-field image of the samples and a CARS spectrum of the 20 $\mu$m bead in the center of Fig. 5(c). We established that the CARS spectro-microscope was very effective for strong-resonance imaging in the CH vibration region (2800-3100 cm$^{-1}$). The prominent peaks at 2878 cm$^{-1}$ (CH$_{2}$ symmetric stretching) and 3037 cm$^{-1}$ (CH stretching) were detected [19,20]. The spectral resolution was $\sim$46 cm$^{-1}$, being determined by the spectral bandwidth (3 nm FWHM) of the pump beam. The average laser powers as measured just before the scanner were 17 mW (pump) and 3 mW (Stokes). As shown in Fig. 5(c)–(f), we acquired forward (c) and backward (e) CARS images of the mixed PS bead sample using the intensity of the Raman shift in the range of 2760–3100 cm$^{-1}$, with a 13 nm FWHM bandpass filter (Semrock, FF01-650/13) and averaging over 10 frames to increase the contrast. In addition, a backward TP absorption image of the 2 $\mu$m beads, acquired using a 40 nm FWHM bandpass filter (Semrock, FF01-400/40), is shown in Fig. 5(d), and the images in Fig. 5(c) and (d) were merged to form the image shown in Fig. 5(f). The field of view (FOV) was 218.8 $\mu$m $\times$ 218.8 $\mu$m in Fig. 5(c)–(f). In Fig. 5(f), the two different types of beads can be identified. The contrast was higher in the forward CARS images (Fig. 5(c)) than in the backwards CARS image (Fig. 5(e)). Video-rate merged forward CARS and backward TP images (7.5 fps without frame averaging, 2.5 fps with frame averaging over 3 frames, and 1.5 fps with frame averaging over 5 frames) were successfully recorded, as detailed in the supplemental files (Visualization 1, Visualization 2 and Visualization 3).
Fig. 5. (a) Bright-field image of the sample (a mixture of 20 $\mu$m polystyrene beads and 2 $\mu$m blue fluorescent polystyrene beads). (b) CARS spectrum of the 20 $\mu$m polystyrene bead. (c) Forward CARS image showing the intensity in the Raman shift range of 2760–3100 cm$^{-1}$. (d) Backward two-photon absorption image of 2 $\mu$m blue fluorescent beads. (e) Backward CARS image showing the intensity in the Raman shift range of 2760–3100 cm$^{-1}$. (f) Merged image created from the images shown in (c) and (d).
4. Conclusion
We first present a femtosecond self-starting mode-locked TiS laser based on an extended cavity incorporating a Herriott multipass cell design that generated pulse energies of up to 16.3 nJ using four pump diodes. The diodes were spectrally combined in a system consisting of an aspheric lens for collimation along the fast axis and a shared cylindrical telescope for collimation along the slow axis. The laser generated 286 mW ML average power and 80 fs pulses at 17.5 MHz. The signal-to-noise ratio at the fundamental frequency showed an extinction ratio of >50 dB relative to the carrier. The laser design has only a 60 $\times$ 38 cm$^2$ footprint. In addition, we demonstrate that this single laser can be successfully utilized for video-rate multiplex forward and backward CARS spectro-microscopy with the shortest pixel dwell time ever reported (122 ns). This compact CARS system with a TiS laser has advantages over other diode-pumped solid-state femtosecond laser systems in terms of simplicity, robustness, and cost. In addition, femtosecond TiS lasers are very attractive for emerging ultrafast applications in photodisruption, nonlinear bio-imaging, and spectroscopy [21–23]. The proposed laser will be further developed in future studies with the aim of increasing the output pulse energy using bidirectional diode pumping.
Funding
Electronics and Telecommunications Research Institute (22YR1410); Korea Creative Content Agency (KOCCA) (SR202106003).
Acknowledgments
The authors thank Incheon Song and Byung-Seon Chun (Nanoscope Systems) for assistance with the microscope setup. The authors thank Eero Koivusalo (RefleKron) for supplying SESAMs.
Disclosures
The authors declare that they have no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available but may be obtained from the corresponding author upon reasonable request.
References
1. P. F. Moulton, “Spectroscopic and laser characteristics of ti:al2o3,” J. Opt. Soc. Am. B 3(1), 125–133 (1986). [CrossRef]
2. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning fluorescence microscopy,” Science 248(4951), 73–76 (1990). [CrossRef]
3. J. Cheng, A. Volkmer, L. D. Book, and X. S. Xie, “Multiplex coherent anti-stokes raman scattering microspectroscopy and study of lipid vesicles,” J. Phys. Chem. B 106(34), 8493–8498 (2002). [CrossRef]
4. X. Zhang, J. Hu, R. W. Knighton, X.-R. Huang, C. A. Puliafito, and S. Jiao, “Dual-band spectral-domain optical coherence tomography for in vivo imaging the spectral contrasts of the retinal nerve fiber layer,” Opt. Express 19(20), 19653–19659 (2011). [CrossRef]
5. M. Hofer, N. K. Balla, and S. Brasselet, “High-speed polarization-resolved coherent raman scattering imaging,” Optica 4(7), 795 (2017). [CrossRef]
6. P. W. Roth, A. J. Maclean, D. Burns, and A. J. Kemp, “Direct diode-laser pumping of a mode-locked ti:sapphire laser,” Opt. Lett. 36(2), 304–306 (2011). [CrossRef]
7. P. W. Roth, D. Burns, and A. J. Kemp, “Power scaling of a directly diode-laser-pumped ti:sapphire laser,” Opt. Express 20(18), 20629–20634 (2012). [CrossRef]
8. K. Gürel, V. J. Wittwer, M. Hoffmann, C. J. Saraceno, S. Hakobyan, B. Resan, A. Rohrbacher, K. Weingarten, S. Schilt, and T. Südmeyer, “Green-diode-pumped femtosecond ti:sapphire laser with up to 450 mw average power,” Opt. Express 23(23), 30043–30048 (2015). [CrossRef]
9. S. Backus, M. Kirchner, C. Durfee, M. Murnane, and H. Kapteyn, “Direct diode pumped ti:sapphire ultrafast regenerative amplifier system,” Opt. Express 25(11), 12469–12477 (2017). [CrossRef]
10. A. Rohrbacher, O. E. Olarte, V. Villamaina, P. Loza-alvarez, and B. Resan, “Multiphoton imaging with blue-diode-pumped sesam-modelockedj ti:sapphire oscillator generating 5 nj 82 fs pulses,” Opt. Express 25(9), 10677–10684 (2017). [CrossRef]
11. D. H. Song and H.-S. Seo, “Spectrally combined three-diode-pumped compact femtosecond ti:sapphire laser exceeding 1 w mode-locked power,” Opt. Express 29(20), 32649–32657 (2021). [CrossRef]
12. D. H. Song, S. I. Hwang, and D.-K. Ko, “Dynamics of sub-microjoule femtosecond ppulse formation in a negative dispersion regime,” J. Korean Phys. Soc. 61(5), 730–734 (2012). [CrossRef]
13. V. L. Kalashnikov, E. Podivilov, A. Chernykh, S. Naumov, A. Fernandez, R. Graf, and A. Apolonski, “Approaching the microjoule frontier with femtosecond laser oscillators: theory and comparison with experiment,” New J. Phys. 7(1), 217 (2005). [CrossRef]
14. B. Talone, M. Bazzarelli, A. Schirato, F. D. Vicario, D. Viola, E. Jacchetti, M. Bregonzio, M. T. Raimondi, G. Cerullo, and D. Polli, “Phototoxicity induced in living hela cells by focused femtosecond laser pulses: a data-driven approach,” Biomed. Opt. Express 12(12), 7886–7905 (2021). [CrossRef]
15. D. H. Song, W. B. Cho, H. W. Lee, D. H. Shin, D.-K. Ko, and M. Y. Jung, “A compact kerr-lens mode-locked ti:sapphire oscillator with 330 nj soliton-like pulses,” Laser Phys. Lett. 10(6), 065003 (2013). [CrossRef]
16. S. H. Cho, B. E. Bouma, E. Ippen, and J. G. Fujimoto, “Low-repetition-rate high-peak-power kerr-lens mode-locked ti:al2o3 laser with a multiple-pass cavity,” Opt. Lett. 24(6), 417–419 (1999). [CrossRef]
17. D. H. Song, S. I. Hwang, and D.-K. Ko, “Compensation of incomplete round trip in an herriott multipass-based kerr-lens mode-locked ti:sapphire oscillator via an output coupler position,” Jpn. J. Appl. Phys. 50, 032705 (2011). [CrossRef]
18. M. D. Duncan, J. Reintjes, and T. J. Manuccia, “Scanning coherent anti-stokes raman microscope,” Opt. Lett. 7(8), 350–352 (1982). [CrossRef]
19. J. Kiefer, S. Wagenfeld, and D. Kerle, “Chain length effects on the vibrational structure and molecular interactions in the liquid normal alkyl alco,” Spectrochim. Acta, Part A 189, 57–65 (2018). [CrossRef]
20. A. Krylov, S. Krylova, S. Kopyl, A. Krylov, F. Salehli, P. Zelenovskiy, A. Vtyurin, and A. Kholkin, “Raman spectra of diphenylalanine microtubes: Polarisation and temperature effects,” Crystals 10(3), 224 (2020). [CrossRef]
21. W. Watanabe, N. Arakawa, S. Matsunaga, T. Higashi, K. Fukui, K. Isobe, and K. Itoh, “Femtosecond laser disruption of subcellular organelles in a living cell,” Opt. Express 12(18), 4203–4213 (2004). [CrossRef]
22. H. Timmers, A. Kowligy, A. Lind, F. C. Cruz, N. Nader, M. Silfies, G. Ycas, T. K. Allison, P. G. Schunemann, S. B. Papp, and S. A. Diddams, “Molecular fingerprinting with bright, broadband infrared frequency combs,” Optica 5(6), 727–732 (2018). [CrossRef]
23. R. He, Y. Xu, L. Zhang, S. Ma, X. Wang, D. Ye, and M. Ji, “Dual-phase stimulated raman scattering microscopy for real-time two-color imaging,” Optica 4(1), 44–47 (2017). [CrossRef]
