Ytterbium-doped fiber lasers (YDFLs) working in the near-infrared (NIR) spectral window and capable of high-power operation are popular in recent years. They have been broadly used in a variety of scientific and industrial research areas, including light bullet generation, optical frequency comb formation, materials fabrication, free-space laser communication, and biomedical diagnostics as well. The growing interest in YDFLs has also been cultivated for the generation of high-power femtosecond (fs) pulses. Unfortunately, the operating wavelengths of fs YDFLs have mostly been confined to two spectral bands, i.e., 970-980 nm through the three-level energy transition and 1030-1100 nm through the quasi three-level energy transition, leading to a spectral gap (990-1020 nm) in between, which is attributed to an intrinsically weak gain in this wavelength range. Here we demonstrate a high-power mode-locked fs YDFL operating at 1010 nm, which is accomplished in a compact and cost-effective package. It exhibits superior performance in terms of both short-term and long-term stability, i.e., <0.3% (peak intensity over 2.4 μs) and <4.0% (average power over 24 hours), respectively. To illustrate the practical applications, it is subsequently employed as a versatile fs laser for high-quality nonlinear imaging of biological samples, including two-photon excited fluorescence microscopy of mouse kidney and brain sections, as well as polarization-sensitive second-harmonic generation microscopy of potato starch granules and mouse tail muscle. It is anticipated that these efforts will largely extend the capability of fs YDFLs which is continuously tunable over 970-1100 nm wavelength range for wideband hyperspectral operations, serving as a promising complement to the gold-standard Ti:sapphire fs lasers.
© 2017 Optical Society of America
High-power ytterbium-doped fiber lasers (YDFLs) have gained popularity in the past decade for their high energy, wideband and reliable operations [1,2]. They also possess a broad range of attractive physical attributes, functionality and practicality that distinguish them from other bulky solid-state counterparts, including single longitudinal mode (also known as single frequency) [3,4], broad gain bandwidth , high power efficiency (>80%) [6,7], outstanding thermo-optical properties and fully alignment-free design. Excellent performance characteristics of continuous wave (CW), nanosecond (ns), picosecond (ps) and femtosecond (fs) YDFLs have been intensively demonstrated so far [8–10]. Therein, the working wavelength of YDFLs has long been limited to two typical spectral bands: 970-980 nm which exploits three-level transition and 1030-1100 nm which is dominated by quasi-three-level transition [11–14]. However, YDFLs working at 990-1020 nm have rarely been demonstrated, mainly due to the small gain coefficient and the gain competition from the three-level and quasi-three-level bands . Bridging this wavelength gap is crucial to achieve a wideband hyperspectral capability, which is essential in applications such as multicolor two-photon excited fluorescence (TPEF) microscopy  and hyperspectral coherent Raman scattering microscopy [16,17]. In addition, this special wavelength of 1010 nm by itself can potentially provide a higher TPEF efficiency for some widely-used biomolecules and dyes, e.g., Alexa Fluor 488 and emerald GFP, and thus reduce the illumination power requirement. Recently, several efforts have been devoted to shortening the operating wavelength of a quasi-three-level band to around 1020 nm by suppressing the three-level gain [18–20]. However, those prior works have mostly been demonstrated for narrow-linewidth CW lasers, although a few other attempts have been performed for pulsed lasers with long pulse widths (100’s ps to μs) [21,22]. Femtosecond pulse lasers operating at extremely high peak powers ranging from kW to MW , on the other hand, are of great use in a wide variety of applications such as light bullet generation , optical frequency comb formation , advanced 3D micro-/ nano-structure processing  and biological imaging . Comparing with visible light which has widely been used in single-photon fluorescence microscopy, two-photon fluorescence microscopy with near-infrared (NIR) excitation can penetrate deeper into highly scattering biological tissue, and is less likely to induce tissue damage [28–31]. Specifically, fs pulsed laser can significantly reduce phototoxicity and photobleaching effects [32,33]. The gold-standard solid-state Ti:sapphire lasers exhibit poor power efficiency beyond 1000 nm , while mode-locked YDFLs usually work at wavelengths beyond 1030 nm. As a result, it is technically interesting to exploit high-power fs YDFLs in the bridging wavelength range of 990-1020 nm, which remains an unmet challenge [34–38].
In this letter, we demonstrate a compact high-power ~300-fs YDFL covering a wavelength range of 1000-1020 nm centered at 1010 nm, which is particularly useful for multiphoton imaging applications. The fiber laser oscillator is built in a very simple and cost-effective configuration, where the transverse mode is confined to the fundamental Gaussian mode by using all single-mode fibers (SMFs). Femtosecond pulses at an average power of >500 mW are obtained with superior short-term and long-term stabilities, i.e., <0.3% peak-intensity and <4.0% average-power variation over 2.4 μs and 24 hours, respectively. The practical applications of this compact high-power fs YDFL are showcased as part of a custom-built multiphoton imaging system. Biological samples, e.g., mouse kidney, mouse brain, mouse tail and potato starch granules, are employed for two-photon excitation and second-harmonic generation (SHG) microscopy studies. Its capabilities in terms of deep tissue sectioning and multicolor fluorescence excitation are also investigated. The results confirm the robustness of this fs YDFL. We believe that these efforts will help to broaden the useful spectrum of high-power fs YDFLs on one hand, and largely facilitate applications requiring high peak-power hyperspectral operation by incorporating other conventional bands of YDFLs.
2. Experimental setup
The master fiber oscillator has a simple architecture, as shown in Fig. 1(a). The generation of the fs laser pulses is based on the mode-locking technology of nonlinear polarization rotation (NPR) [39,40], which can be self-started from initial noise pulses. It includes three key elements: gain, NPR and spectral filtering (SF). The gain fiber is a 40-cm long ytterbium-doped fiber (YDF, Thorlabs Yb 1200-4/125), which is highly doped and has a core pump absorption of 1200 dB/m at 976 nm. It is pumped by an optical integrated module (OIM) using a 976-nm laser diode (LD, II-VI LC96Z600), which has a SMF pigtail and provides a maximum power of up to 600 mW. The OIM integrates the functions of wavelength-division multiplexing (WDM), polarization-sensitive isolation (PS-ISO) and 50:50 signal beam splitting (BS). NPR is performed by incorporating an inline polarization controller (PC) and polarization-sensitive OIM. To force the oscillation at 1010 nm, SF is employed, which includes a 4-nm wide bandpass filter (BPF, Semrock LL01-1064-12.5) located between two fiber collimators (Col) on both sides. It should be noted that this free-space filter can also be replaced by a customized fiber-based bandpass filter, in which way an all-fiber configuration can be realized. A half-wave plate (λ/2) is utilized for flexible and fine control of the state of polarization. SF plays a crucial role in suppressing the oscillation in the conventional high-gain wavelength windows as mentioned before, which can prevent mode-locking and even lasing at 1010 nm. In addition, SF also benefits the self-started mode-locking particularly for an all-normal dispersion oscillator in this case, i.e., a wavelength of <1.3 μm . The mode-locked signal is extracted from the fiber oscillator through the tap port of the OIM, and subsequently launched into an external power booster. It should be pointed out that inside the OIM, both signal extraction and pump coupling work in reflection mode. The master fiber oscillator is designed for a pulse repetition rate of 80 MHz, which is useful for fast nonlinear imaging.
The external power booster, as shown in Fig. 1(b), is constructed with all SMFs. It includes two stages of amplifiers, i.e., Amp1 and Amp2. The slightly-chirped pulse output from the oscillator is further chirped to about 25 ps by using a 50-m SMF (Corning HI1060) to reduce excessive nonlinearity. In Amp1, the gain medium is another piece of highly-doped YDF (Thorlabs Yb 1200-4/125), 20 cm in length. It is backward pumped by a LD through the WDM in reflection mode. To isolate the backward-propagation light and to ensure the stability of the master fiber oscillator, a fiber isolator (ISO) specially designed for high efficiency at 1010 nm (>75%) is placed between the oscillator and external power booster. Without this fiber isolator, we found that mode-locking was susceptible to reflected light from the external power booster, e.g., the amplified spontaneous emission (ASE) noise and parasite lasing at around 1030 nm and 1060 nm. It is noted that the length of the gain fiber has experimentally been optimized from a number of YDF samples with different lengths. This is based on the consideration of the trade-off between pump absorption and signal gain at the targeted wavelength (i.e., 1010 nm in this case) . In Amp2, another 40-cm YDF (Thorlabs Yb 1200-4/125) serves as the gain medium, which is bidirectionally pumped through two sets of WDMs and LDs. After the power booster, the high-power laser pulses are launched into free space through another fiber collimator with a beam waist diameter of ~4.0 mm. They are subsequently de-chirped by a polarization-insensitive grating pair (GP, LightSmyth T-1000-1040-3212-94) with a single-pass power efficiency of >94% at 1010 nm. The de-chirped pulses are finally fed into a homemade laser scanning multiphoton microscope, as shown in Fig. 1(c).
In the multiphoton microscope, laser scanning is utilized by XY galvanometric scanning mirrors (GM, Cambridge Technology 6220H). The raster-scanned laser beam is relayed to the illumination objective lens (Obj, Olympus UPLSAPO60xW, NA 1.2, water immersion) through a two-lens telescope (Tel), whose magnification has been optimized to fully illuminate the back focal plane of the objective lens. The generated signal is collected in the forward direction by a condenser lens (Con, Olympus U-AAC, NA 1.4). The residual excitation beam is cleaned up by two shortpass filters (Semrock FF01-775/SP-25 and FF01-950/SP-25), while additional bandpass filters are used to pick up specific signals at different wavelength bands, i.e., two-photon excited fluorescence emitted by different dyes and SHG signal, to be introduced later. The image signal is detected by a large-area photomultiplier tube (PMT, Hamamatsu H10723-20) through a focusing lens, which limits the signal beam movement to the detector area in order to avoid scanning artifacts. The electrical signal from the PMT is subsequently digitized by an analogue-to-digital (A/D) converter card (National Instruments PCI-6110, 5 MS/sec). The control of the entire imaging system, including laser scanning, data collection and image reconstruction, is realized by a Matlab-based multifunctional program . It should be pointed out that the current multiphoton microscope operates in transmission mode with forward detection, while it can easily be modified to also implement epi-detection.
3. Results and discussion
3.1 Performance of master fiber oscillator
Self-started mode-locking at 1010 nm was obtained by increasing the pump power beyond 460 mW. At a pump power of 460-500 mW, single-pulse mode-locking was observed. It should be noted that an excessive pump power can easily produce multiple pulses that arise from the peak power clamping. The performance of the mode-locked pulses was analyzed in both time and frequency domains at the oscillator output. As shown in Fig. 2(a), the optical spectrum measured right after the oscillator by an optical spectrum analyzer (OSA, Agilent 86142B) is centered at 1010 nm and covers a wavelength range of 1000-1020 nm. A 3-dB bandwidth of ~8 nm corresponds to a transform-limited Gaussian pulse width of 188 fs. It is also noticed that the optical spectrum exhibits some fine features, which can be contributed to either intrinsic spectral features of the optical elements or excessive nonlinear effects inside the master oscillator. For the latter case, the induced nonlinear frequency chirp would affect the pulse compressibility. A 10-GHz fiber-coupled photodetector (PD, HP DET01-CFC) was used to receive the optical pulse train, which was subsequently digitized by a 20-GHz real-time oscilloscope (LeCroy SDA 820Zi-B), and shown in Fig. 2(b). As can be observed, the pulse train has a period of 12.5 ns, i.e., a repetition rate of 80.0 MHz. The peak intensity fluctuation of the pulse train was calculated to be about 0.28% (standard deviation, std) over 2.4 μs, which implies a reasonably good short-term stability. This can also be verified by its 80-dB signal-to-noise ratio (SNR) in the radio frequency (RF) spectrum shown in Fig. 2(c), which was measured by an electrical spectrum analyzer (Agilent, E4440A, 26 GHz). It is clear that the frequency peak is centered at 80.0 MHz, consistent with a pulse period of 12.5 ns shown in Fig. 2(b). The direct output pulse was frequency-chirped due to the dispersion from the optical fibers used to construct the master oscillator, yielding a pulse width of 2.4 ps at the output. As the frequency chirp produced by the second-order dispersion is nearly linear, it can be easily de-chirped by simply using a grating pair. The autocorrelation trace of the de-chirped pulse measured by an autocorrelator (Femtochrome Research FR-103 MN) exhibits a pulse width of 294 fs, as shown in Fig. 2(d). The discrepancy between the theoretical transform-limited pulse width (i.e., 188 fs) and compressed pulse width can be attributed to the imperfect spacing between the grating pair as well as the higher-order dispersion [38,39].
3.2 Performance of external power booster
The optical power obtained directly from the master oscillator was only 6.8 mW, which is not sufficient for practical applications. It was externally boosted up by using the two-stage amplification as shown in Fig. 1(b). Figures 3(a) and 3(b) illustrate the average output power as a function of the pump power in Amp1 and Amp2, respectively. In Amp1, a maximum average output power of 202 mW is obtained at a pump power of ~400 mW, corresponding to a slope efficiency of ~51%, which benefits from the high dopant level of the YDFs. It is also noted that no obvious power saturation is present in Fig. 3(a), meaning that a higher optical power can further be obtained by simply increasing the available pump power. Figure 3(c) shows the optical spectrum at the output of Amp1 when signal is amplified to 202 mW (red curve). Comparing with the spectrum measured at oscillator output (blue curve), Amp1 indeed introduces a red shift by about 1.5 nm on the left hand side, as indicated by the shaded circles. This can be attributed to the gain-window shifting as well as the nonlinear effects induced by the high peak power, which can be improved (if necessary) by placing a pre-chirped stage between the master oscillator and external power booster, e.g., a piece of SMF, which is usually performed for chirped-pulse amplification (CPA) . In Amp2, a maximum output power of 501 mW is obtained at a pump power of 850 mW, which is the total power of both the forward and backward pump laser diodes. Again, no obvious power saturation is observed, and a slope efficiency of ~59% is achieved at this amplification stage. Similar to Amp1, Amp2 also exhibits a red-shifted spectrum, as indicated in Fig. 3(c). It is noticed that, after Amp2, the side-mode suppression ratio (SMSR) of the optical spectrum on the right hand side becomes less than that on the left hand side, as indicated by the shaded rectangle at about 1020-1030 nm. It can be simply improved by using a bandpass filter if necessary. It should also be pointed out that the optical spectrum after Amp2 exhibits visible spectral modulation, which can be attributed to the interplay of nonlinearity. It can be improved by increasing the chirping rate of CPA. The far-field 2D-beam profile after Amp2 was measured by a CCD camera and is shown in the bottom-right inset of Fig. 3(b), while the 1D and its Gaussian fitting profiles are shown in the top-left inset. It is clear that a pure Gaussian beam (i.e., single mode) is obtained. To examine the long-term power stability, the output signal power from the external booster was monitored every ~300 ms over 24 hours, and is shown in Fig. 3(d). As can be observed, an optical power fluctuation of < 2% over one day is achieved for this room-temperature laser system. It is noted that output power remains the same each time the system is switched on, which is crucial for daily operations.
3.3 Two-photon excited fluorescence microscopy
To showcase the practical applications of this robust high-power fs YDFL, we first employed it to perform two-color two-photon-excited fluorescence imaging of a biological sample, i.e., a mouse kidney section slide (Thermo Fisher Scientific FluoCells® prepared slide #3). This sample is a 16-µm thick cryosection stained with Alexa Fluor 488 wheat germ agglutinin (W-11261), Alexa Fluor 568 phalloidin (A-12380) and DAPI (D-1306). It is ideal for demonstrating the optical sectioning properties of confocal microscopes. Since the wavelength of our fs YDFL is located between the two-photon excitation peak wavelengths of Alexa fluor 488 and Alexa fluor 568, i.e., 990 nm and 1156 nm, respectively, it can efficiently excite those two dyes at the same time, while the excitation of DAPI is relatively weak . Two-color two-photon images can thus be obtained by isolating the emission of these two dyes through bandpass filters centered at about 519 and 600 nm (Semrock, FF01-534/42-25 and FF01-600/52-25), respectively, as shown in Fig. 4(a). The image at a field of view (FOV) of 150 μm x 150 μm was collected at a line-scan rate of 500 Hz over an image size of 512 x 512 pixels, corresponding to a 2D frame rate of ~1 Hz. It is noted that the frame rate is mainly limited by the speed of the scanning mirrors, which can easily be improved by using resonant scanning mirrors. Figures 4(b) and 4(c) show the images of Alexa Fluor 568 and 488 dyes, respectively, while the composite image with false colors is shown in Fig. 4(d). As can be observed, the elements of the glomeruli and convoluted tubules are highlighted by the mostly membrane-staining wheat germ agglutinin Alexa Fluor 488, while filamentous actin prevalent in the glomeruli and brush border can be easily identified by Alexa 568. It is noted that all the imaging experiments demonstrated in this work were performed using only Amp1, since the required optical powers were all less than 100 mW.
Indeed, two-photon excited fluorescence imaging has been shown to be well-suited for deep tissue optical analysis involved in the studies of neuronal activity and anatomy, embryo development, as well as tissue morphology and pathology . For deep imaging of the brain, in particular, two-photon imaging at longer wavelengths (>900 nm) can largely alleviate scattering issues (i.e., the dominant attenuation factor in brain tissues) on one hand, and the generated fluorescence emission at correspondingly longer wavelengths has less absorption on the other hand. Consequently, it enables deeper penetration into brain tissue [28,44]. Here, we utilize our robust and cost-effective fs YDFL to perform two-photon imaging in mouse brain thick tissue. To achieve a wider field of view, we used a 40x objective lens (LEICA Achro 40/0.66). The sample was a 200-μm-thick mouse brain tissue sandwiched by two glass slides, where the mouse expresses the yellow fluorescent protein (YFP) in layer-V pyramidal neurons (Thy1-YFP H-line) . All experiments in these samples were approved and performed in accordance with institutional guidelines of The University of Hong Kong. As shown in Fig. 5(a), 3D two-photon imaging of the mouse brain tissue was performed at axial step sizes of 5 μm (i.e., coarse depth scanning). A typical sectioning image shown in Fig. 5(b) clearly illustrates the neuron distribution. This section image consists of 512 x 512 pixels, corresponding to a FOV of 225 x 225 μm. The line scan rate is 500 Hz and 2D frame rate is ~1 Hz. By modifying the microscope to epi-detection together with a long working distance objective lens (e.g., Olympus XL Plan N 25x, NA1.05, WD 2 mm), it should be possible to realize in vivo mouse brain imaging.
3.4 Second-harmonic generation imaging
In addition to two-photon fluorescence excitation, the SHG is another primary contrast mechanism for multiphoton imaging, which in fact has been one of the earliest contrast modes used for nonlinear biological microscopy . As SHG is polarization-sensitive, it can provide useful structural information about biological samples, e.g., the orientation and degree of structural organization of fibrous compounds. As a result, it has widely been used as the main contrast in a label-free manner to characterize the filamentous architecture of biological elements, particularly starch granules and collagen fibers . As shown in Fig. 6, a high-signal-to-noise-ratio SHG image of potato starch granules is obtained by using our fs YDFL at 1010 nm. The starch granule sample was prepared by scraping a slice from a fresh potato and sandwiching it between two 170-μm-thick cover glass slides. Figure 6(a) shows the SHG spectrum generated by the starch granules. Here, the objective lens used was also Olympus UPLSAPO60xW. Imaging FOV was 150 μm x 150 μm (512 x 512 pixels) and 2D frame rate was ~1 Hz. By further manipulating the illumination polarization of the laser beam, more information or even plant species specificity could be obtained .
To further study the polarization-sensitive characteristics of SHG imaging through our robust fs YDFL, a non-stained mouse tail slide was employed to perform SHG microscopy at varying polarization states of the illumination laser beam, as shown in Fig. 7. All the SHG images were taken under the same condition as Fig. 6. Figure 7(a) is a bright-field image of the investigated area, where the region between two red dotted lines is rich in collagen fiber. Illumination polarization was controlled as illustrated in Fig. 7(b). A linear polarizer (P), a half-wave plate (λ/2) and a quarter-wave plate (λ/4) were placed between the galvanometric scanning mirrors and the compression grating pair. Figures 7(c)-7(e) show typical SHG images of the mouse tail slide. For all three images, FOV was 150 μm x 150 μm (512 x 512 pixels) and 2D frame rate was ~1 Hz. Here, collagen in the tendon tissue exhibits very strong SHG signal, while the signal from the surrounding muscle tissue is weak. Optical spectrum of the signal was similar to Fig. 6(a), indicating that there was no two-photon excited fluorescence signal causing disturbance. We also studied the polarization dependent SHG signal intensity. Figure 7(c) and 7(d) have two orthogonal linear illumination polarization states, while Fig. 7(e) has a circular illumination polarization state. It is observed that the intensities of the second-harmonic signal at different locations change with the illumination polarization state, e.g., the areas indicated by the white dashed rectangle in Figs. 7(c) and 7(d). To show it more clearly, two line-scans in the same area with different illumination polarization states are plotted together in Fig. 7(f).
In summary, we have demonstrated a high-power fs ytterbium-fiber laser operating in the wavelength window of 990-1020 nm complementary to the conventional solid-state laser systems. Single pulse operation with a pulse width of about 300 fs and an average power of >500 mW is obtained. Pulse repetition rate is 80 MHz which is common for standard Ti:sapphire lasers. The short-term (2.4 μs) and long-term (24 hours) power fluctuations are less than 0.3% and 4.0%, respectively. Compared to its solid-state counterparts, it has a much more compact design (fits standard 19″ racks) and lower cost (see Table 1, Appendix). Moreover, it is a turnkey system and no adjustment is needed for daily operation. It is the first time that such high-power fs pulses at 1010 nm are obtained by using off-the-shelf ytterbium-doped fibers. This is an essential step in performing high-power hyperspectral operation with continuous tunability over an optimal wavelength range, i.e., from 970 nm to 1100 nm, by bridging the traditional wavelength windows of ytterbium fibers ― 970-980 nm and 1030-1100 nm. To showcase the validity of practical applications, this laser is employed to perform nonlinear imaging, including TPEF and SHG microscopy. Two-color fluorescence excitation and depth-scanning microscopy have been demonstrated for mouse kidney and brain tissue samples, while polarization-sensitive SHG microscopy has been demonstrated with potato and mouse tail samples. The results suggest that this high-power fs YDFL can be a versatile laser for various nonlinear imaging modalities, and we believe this is essential for wideband hyperspectral applications in the wavelength windows beyond the working range of traditional solid-state lasers.
The authors declare that there are no conflicts of interest related to this article.
This research is funded by Research Grants Council of the Hong Kong Special Administrative Region, China (Project Nos. HKU 17205215, HKU 17208414, and CityU T42-103/16-N); National Natural Science Foundation of China (N_HKU712/16); Innovation and Technology Fund (GHP/050/14GD); Germany/Hong Kong Joint Research Scheme sponsored by the Research Grants Council of Hong Kong and the Germany Academic Exchange Service of Germany (G-HKU708/14); University Development Fund of HKU; German Academic Exchange Service (DAAD) (57138104).
References and links
1. R. Paschotta, J. Nilsson, A. C. Tropper, and D. C. Hanna, “Ytterbium-doped fibre amplifiers,” IEEE J. Quantum Electron. 33(7), 1049–1056 (1997).
2. D. J. Richardson, J. Nilsson, and W. A. Clarkson, “High power fiber lasers: current status and future perspectives [Invited],” J. Opt. Soc. Am. B 27(11), B63–B92 (2010).
3. M. Leigh, W. Shi, J. Zong, J. Wang, S. Jiang, and N. Peyghambarian, “Compact, single-frequency all-fiber Q-switched laser at 1 µm,” Opt. Lett. 32(8), 897–899 (2007). [PubMed]
4. Y. Ma, X. Wang, J. Leng, H. Xiao, X. Dong, J. Zhu, W. Du, P. Zhou, X. Xu, L. Si, Z. Liu, and Y. Zhao, “Coherent beam combination of 1.08 kW fiber amplifier array using single frequency dithering technique,” Opt. Lett. 36(6), 951–953 (2011). [PubMed]
5. J. Limpert, T. Schreiber, T. Clausnitzer, K. Zöllner, H. Fuchs, E. Kley, H. Zellmer, and A. Tünnermann, “High-power femtosecond Yb-doped fiber amplifier,” Opt. Express 10(14), 628–638 (2002). [PubMed]
6. Y. Jeong, J. Nilsson, J. Sahu, D. Payne, R. Horley, L. Hickey, and P. Turner, “Power scaling of single-frequency ytterbium-doped fiber master-oscillator power-amplifier sources up to 500W,” IEEE J. Sel. Top. Quantum Electron. 13(3), 546–551 (2007).
8. C. Jauregui, J. Limpert, and A. Tünnermann, “High-power fibre lasers,” Nat. Photonics 7(11), 861–867 (2013).
9. M. E. Fermann and I. Hartl, “Ultrafast fibre lasers,” Nat. Photonics 7(11), 868–874 (2013).
10. C. Xu and F. W. Wise, “Recent advances in fibre lasers for nonlinear microscopy,” Nat. Photonics 7(11), 875–882 (2013). [PubMed]
11. J. R. Armitage, R. Wyatt, B. J. Ainsly, and S. P. Craig-Ryan, “Highly efficient 980 nm operation of an Yb-doped silica fiber laser,” Electron. Lett. 25(5), 298–299 (1989).
12. F. Röser, C. Jauregui, J. Limpert, and A. Tünnermann, “94 W 980 nm high brightness Yb-doped fiber laser,” Opt. Express 16(22), 17310–17318 (2008). [PubMed]
13. J. Boullet, Y. Zaouter, R. Desmarchelier, M. Cazaux, F. Salin, J. Saby, R. Bello-Doua, and E. Cormier, “High power ytterbium-doped rod-type three-level photonic crystal fiber laser,” Opt. Express 16(22), 17891–17902 (2008). [PubMed]
14. Á. Krolopp, A. Csákányi, D. Haluszka, D. Csáti, L. Vass, A. Kolonics, N. Wikonkál, and R. Szipőcs, “Handheld nonlinear microscope system comprising a 2 MHz repetition rate, mode-locked Yb-fiber laser for in vivo biomedical imaging,” Biomed. Opt. Express 7(9), 3531–3542 (2016). [PubMed]
15. P. Mahou, M. Zimmerley, K. Loulier, K. S. Matho, G. Labroille, X. Morin, W. Supatto, J. Livet, D. Débarre, and E. Beaurepaire, “Multicolor two-photon tissue imaging by wavelength mixing,” Nat. Methods 9(8), 815–818 (2012). [PubMed]
16. D. Fu, G. Holtom, C. Freudiger, X. Zhang, and X. S. Xie, “Hyperspectral imaging with stimulated Raman scattering by chirped femtosecond lasers,” J. Phys. Chem. B 117(16), 4634–4640 (2013). [PubMed]
17. S. Karpf, M. Eibl, W. Wieser, T. Klein, and R. Huber, “A time-encoded technique for fibre-based hyperspectral broadband stimulated Raman microscopy,” Nat. Commun. 6, 6784 (2015). [PubMed]
18. S. Mo, S. Xu, X. Huang, W. Zhang, Z. Feng, D. Chen, T. Yang, and Z. Yang, “A 1014 nm linearly polarized low noise narrow-linewidth single-frequency fiber laser,” Opt. Express 21(10), 12419–12423 (2013). [PubMed]
19. R. Steinborn, A. Koglbauer, P. Bachor, T. Diehl, D. Kolbe, M. Stappel, and J. Walz, “A continuous wave 10 W cryogenic fiber amplifier at 1015 nm and frequency quadrupling to 254 nm,” Opt. Express 21(19), 22693–22698 (2013). [PubMed]
20. H. Xiao, J. Leng, H. Zhang, L. Huang, J. Xu, and P. Zhou, “High-power 1018 nm ytterbium-doped fiber laser and its application in tandem pump,” Appl. Opt. 54(27), 8166–8169 (2015). [PubMed]
21. X. Qi, S. P. Chen, H. Y. Sun, B. K. Yang, and J. Hou, “1016nm all fiber picosecond MOPA laser with 50W output,” Opt. Express 24(15), 16874–16883 (2016). [PubMed]
22. M. Jiang, P. Zhou, H. Xiao, R. Tao, and X. Wang, “Pulsed Yb3+-doped fiber laser operating at 1011 nm by intra-cavity phase modulation,” Appl. Opt. 53(10), 1990–1993 (2014). [PubMed]
23. T. Brabec and F. Krausz, “Intense few-cycle laser fields: Frontiers of nonlinear optics,” Rev. Mod. Phys. 72(2), 545–591 (2000).
24. A. Chong, W. H. Renninger, D. N. Christodoulides, and F. W. Wise, “Airy-Bessel wave packets as versatile linear light bullets,” Nat. Photonics 4(2), 103–106 (2010).
25. S. T. Cundiff and J. Ye, “Femtosecond optical frequency combs,” Rev. Mod. Phys. 75(1), 325–342 (2003).
26. X. Liu, D. Du, and G. Mourou, “Laser ablation and micromachining with ultrashort laser pulses,” IEEE J. Quantum Electron. 33(10), 1706–1716 (1997).
27. K. König, “Multiphoton Microscopy in Life Sciences,” J. Microsc. 200(Pt 2), 83–104 (2000). [PubMed]
28. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning fluorescence microscopy,” Science 248(4951), 73–76 (1990). [PubMed]
29. W. R. Zipfel, R. M. Williams, and W. W. Webb, “Nonlinear magic: multiphoton microscopy in the biosciences,” Nat. Biotechnol. 21(11), 1369–1377 (2003). [PubMed]
30. M. Göppert-Maier, “Über Elementarakte mit zwei Quantensprüngen,” Ann. Phys. 9(3), 273–294 (1931).
31. W. Kaiser and C. G. B. Garrett, “Two-Photon Excitation in CaF2:Eu2+,” Phys. Rev. Lett. 7(6), 229–231 (1961).
32. E. P. Perillo, J. E. McCracken, D. C. Fernée, J. R. Goldak, F. A. Medina, D. R. Miller, H. C. Yeh, and A. K. Dunn, “Deep in vivo two-photon microscopy with a low cost custom built mode-locked 1060 nm fiber laser,” Biomed. Opt. Express 7(2), 324–334 (2016). [PubMed]
33. W. Liu, S. H. Chia, H. Y. Chung, R. Greinert, F. X. Kärtner, and G. Chang, “Energetic ultrafast fiber laser sources tunable in 1030-1215 nm for deep tissue multi-photon microscopy,” Opt. Express 25(6), 6822–6831 (2017). [PubMed]
34. H. Yang, “High energy femtosecond fiber laser at 1018 nm and high power Cherenkov radiation generation,” (Massachusetts Institute of Technology, 2014).
35. Y. Hua, W. Liu, M. Hemmer, L. E. Zapata, G. Zhou, D. N. Schimpf, T. Eidam, J. Limpert, A. Tünnermann, and F. X. Kaertner, “87-W, 1018-nm Yb-fiber ultrafast seeding source for cryogenic Yb: YLF amplifier,” in CLEO: Science and Innovations (Optical Society of America, 2016), pp. SM4Q. 5.
36. R. Royon, J. Lhermite, L. Sarger, and E. Cormier, “fs Mode-locked fiber laser continuously tunable from 976 nm to 1070 nm,” 2013Conference on Lasers & Electro-Optics Europe & International Quantum Electronics Conference CLEO EUROPE/IQEC, DOI: . [CrossRef]
37. O. G. Okhotnikov, L. Gomes, N. Xiang, T. Jouhti, and A. B. Grudinin, “Mode-locked ytterbium fiber laser tunable in the 980-1070-nm spectral range,” Opt. Lett. 28(17), 1522–1524 (2003). [PubMed]
38. A. Agnesi, L. Carrà, C. Di Marco, R. Piccoli, and G. Reali, “Fourier-Limited 19-ps Yb-Fiber Seeder Stabilized by Spectral Filtering and Tunable Between 1015 and 1085 nm,” IEEE Photonics Technol. Lett. 24(11), 927–929 (2012).
39. K. Tamura, E. P. Ippen, H. A. Haus, and L. E. Nelson, “77-fs pulse generation from a stretched-pulse mode-locked all-fiber ring laser,” Opt. Lett. 18(13), 1080–1082 (1993). [PubMed]
40. A. Chong, W. H. Renninger, and F. W. Wise, “All-normal-dispersion femtosecond fiber laser with pulse energy above 20 nJ,” Opt. Lett. 32(16), 2408–2410 (2007). [PubMed]
41. T. A. Pologruto, B. L. Sabatini, and K. Svoboda, ScanImage: Flexible software for operating laser scanning microscopes.
42. D. Strickland and G. Mourou, “Compression of amplified chirped optical pulses,” Opt. Commun. 56(3), 219–221 (1985).
44. D. Kobat, M. E. Durst, N. Nishimura, A. W. Wong, C. B. Schaffer, and C. Xu, “Deep tissue multiphoton microscopy using longer wavelength excitation,” Opt. Express 17(16), 13354–13364 (2009). [PubMed]
45. C. S. W. Lai, T. F. Franke, and W. B. Gan, “Opposite effects of fear conditioning and extinction on dendritic spine remodelling,” Nature 483(7387), 87–91 (2012). [PubMed]
46. R. Carriles, D. N. Schafer, K. E. Sheetz, J. J. Field, R. Cisek, V. Barzda, A. W. Sylvester, and J. A. Squier, “Invited review article: Imaging techniques for harmonic and multiphoton absorption fluorescence microscopy,” Rev. Sci. Instrum. 80(8), 081101 (2009). [PubMed]