All-fiber optical parametric oscillator (OPO), offering advantages like robustness, compactness and low lost, has attracted intense interest in coherent anti-Stokes Raman scattering spectroscopy. In typical fiber-based OPO configurations, detrimental nonlinear effects due to intense pump field in fiber coupling devices would inevitably degrade the spectral purity and conversion efficiency, especially when the OPO operated at low repetition rates. Here we demonstrated a new OPO design by placing the main amplifier inside the cavity, where the amplified pump pulses were directly coupled into the nonlinear fiber. Consequently, lower threshold, higher output power and narrower spectrum were obtained. In particular, effective suppression of spectral noise was experimentally observed, resulting in threshold reductions of 37.5%, 17.2%, and 5.2% with a comparison to a conventional OPO operating at repetition rates of 1, 2 and 3 MHz, respectively. Furthermore, the generated synchronized two-color laser sources at a low repetition rate were then employed to detect CH vibrational bands in an ethanol sample. This spectral tailored cavity design is expected to greatly promote the spread of compact all-fiber laser source to nonlinear biomedical imaging.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Coherent anti-Stokes Raman Scattering (CARS) spectroscopy has been widely used to perform label-free, chemical selective analysis in biomedical researches and life sciences [1–5]. Since the performance of a CARS system is largely determined by characteristics of its light source, increasing efforts are dedicated to the development of temporal synchronized, spatial overlapped, and wavelength-tunable ultrafast lasers [6–10]. Compared to bulky laser sources used at early stage, fiber laser sources possess advantages of low cost, compact size, alignment-free operation and environmental stability [11–14]. Especially, the use of photonic crystal fiber (PCF) in laser systems could help to realize tunable single-band CARS detections . For example, a two-color fiber-delivered picosecond laser source was implemented to perform CARS imaging of mouse skin . Spectral compression of a high-power femtosecond fiber laser has been reported to produce transform-limited picosecond pulses, which could be used in CARS and other imaging modalities .
Depending on applications, the laser source in CARS systems is required to operate at different optimized repetition rates. Specifically, ultrafast fiber lasers with high repetition rates (20-80 MHz) are more suitable for CARS detection of less dense biological samples [18–21]. On the other hand, for dense samples such as myelin sheath, commonly used to study the transmission of action potential and nerve impulses, laser pulses with low repetition rates of a few MHz would be preferable due to the reduced photodamage [22–24]. Recently, a 1-MHz fiber master oscillator power amplifier (MOPA) system was built to pump a PCF in single-pass configuration, which generated dual-wavelength light at 780 nm and 1030 nm via four-wave mixing (FWM) for probing the aromatic CH-stretch vibrational resonance . However, the spectral bandwidth of the newly generated sideband in spontaneous FWM was relatively broad and its central wavelength was hard to be tuned, thereby limiting the spectral resolution and detection range. These limitations can be overcome by building an optical parametric oscillator (OPO), where the generated signal or idler pulses are temporally broadened before coupling back into the parametric gain medium. The so-called dispersion filtering effect could lead to the generation of tunable and narrow-band new frequency components . The aforementioned technique has been adopted to improve the spectral resolution and increase the wavelength tuning speed for CARS applications . However, the OPO cavity in these works was constructed with free-space elements in spite of using a nonlinear fiber as the gain medium. In 2017, an all-fiber OPO was developed in a linear cavity arrangement, generating tunable signal pulses between 764 and 960 nm at a repetition rate of 9.5 MHz. Owing to the ultra-low noise performance and automated tunable operation, this all-fiber laser source facilitated a reliable tool for stimulated Raman Scattering .
Although all-fiber integration can be realized by simply replacing the free-space coupling element with a fiber wavelength division multiplex (WDM), yet the strong pump field on the fiber coupler could inevitably induce detrimental nonlinear effects, which would introduce spectral noise and decrease conversion efficiency during the parametric process. With further decrease of the repetition rate, the pulse energy as well as the peak power would increase accordingly. Consequently, the performance of the all-fiber OPO would deteriorate even more significantly. This degradation caused by the fiber coupler in the traditional cavity design hinders the development of the all-fiber OPO with repetition rate of a few MHz. Here, we proposed and implemented a novel cavity design for the all-fiber OPO, where the main amplifier was placed inside the cavity, thus making it possible to directly splice the fibers for pump amplification and nonlinear conversion. Consequently, the OPO could operate at a repetition rate as low as 1 MHz, and deliver two-color pulse trains with temporal synchronization and spatial overlap, thus facilitate the CARS-based spectroscopic analysis in dense samples. Moreover, some important figures of merit like output spectrum, average power and spectral bandwidth are experimentally and theoretically investigated at repetition rates of 1, 2, and 3 MHz with a comparison to the conventional OPO configuration. We found that lower threshold, higher output power and narrower spectrum were obtained by using the new OPO scheme. Thanks to the all-fiber configuration of the pump laser and OPO cavity, the whole system including electronics was assembled in an aluminum box with dimensions of 39 cm × 28 cm × 13 cm. Such compact and portable laser source was then employed to conduct CARS spectroscopic analysis of CH vibrational bands in ethanol. It is worth noting that the reported work here is a continuation of our previous studies on generation and manipulation of low-repetition-rate modulation instabilities in an OPO, but the bulky components such as coupling lens and three-dimensional stages limited compact implementation and alignment-free operation .
This paper was organized as follows. In Section 2, we first presented simulation investigation and experimental measurement of nonlinear effects caused by the fiber coupling device between the gain and parametric media. Then in Section 3, output characteristics of the fiber OPO constructed in different configurations were comparatively analyzed at various repetition rates, which showed effective suppression on spectral noise in our scheme. Finally, observation of CARS spectrum from ethanol sample was given in Section 4, which verified the competence of the implemented laser source in CARS spectroscopy.
2. Investigation of nonlinearity for transmission fiber between gain and PCF
Numerical simulations were made by the commercially available software Fiberdesk to understand the evolution of spectral noise caused by the transmission fiber of the fiber coupling device. The simulation model included an input pump pulse, a piece of transmission fiber with different lengths, and a 40-cm PCF. The temporal duration of the input pump pulse was set to be 30 ps with a pulse energy of 0.15 μJ at 1030 nm, corresponding to an average power of 0.15 W at a repetition rate of 1 MHz. The type of the transmission fiber was HI1060 single mode fiber (SMF) with mode field diameter (MFD) of 6.2 μm, which was commonly used in producing fiber devices operating at 1030 or 1064 nm. Nonlinear effects including dispersion, Raman, self-phase modulation (SPM), self-steepening and temporal gain saturation were all activated during the simulation. We used four different lengths of the SMF, which were 0, 20, 50, and 100 cm, respectively. Length of 0 cm represented our proposed scheme for OPO, where the gain fiber and PCF were spliced directly. Length of 20 cm was almost the shortest length for splicing two polarization maintaining (PM) fibers while 50 and 100 cm were the typical pigtail lengths of fiber elements. The PCF used in our simulation and experiment consisted of a fused silica core of 5 μm diameter surrounded by air holes with hexagonal structure (NKT Photonics, LMA-PM-5). The MFD of PCF was 4.2 μm, resulting in an effective cross-sectional area Aeff of 13.9 μm2. The nonlinear coefficient γ was 10 W−1km−1, which can be deduced by the Aeff. All the parameters were consistent with those used in the experiment.
The calculated output spectra after the transmission fiber and PCF for various fiber lengths are shown as the black lines in Fig. 1. Figures 1(a)-(d) depict the spectra of the output pump pulses after propagating through different lengths of transmission fibers. Under the same condition of input pump pulse, the nonlinear effects especially the Raman-induced spectral noises at 1079 nm take up more energy from the main peak at 1030 nm with the increased fiber length. This Raman-induced spectral noise would be detrimental to the FWM in the subsequent parametric conversion process, as shown in Figs. 1(e)-(h). The simulation predictions were then verified by experimentally measuring the corresponding output spectra with an optical spectrum analyzer (Yokogawa AQ6370C). The experiment results are shown as the red lines in Fig. 1, which agree well with the simulations. Both the simulated and experimental results show that the accumulated nonlinear effects caused by the transmission fiber would give rise to additional spectral noises. In particular, when the length of the transmission fiber is longer than 50 cm, a supercontinuum-like spectrum is generated and leading the domination over FWM signal.
Therefore, it is favorable to design an OPO without the transmission fiber between the gain and parametric fibers, thus eliminating the detrimental nonlinear effects to obtain better spectral purity. In the following section, a novel scheme in contrast to the conventional one will be presented, which facilitates the implementation of an all-fiber OPO operating at low repetition rate.
3. Development of a low-repetition-rate OPO
3.1 Basic configuration of two kinds of OPO
The experimental setup of the all-fiber integrated OPO is shown in Fig. 2. It consisted of two parts: the pump source and the feedback cavity. The pump source contained a fiber oscillator, a pre-amplifier and a pulse picker. The fiber oscillator was in a linear-cavity structure. A semi-conductor saturable absorber mirror (SESAM) acted as a cavity end mirror. The other end mirror was a fiber Bragg grating (FBG) with a bandwidth of 0.2 nm centered at 1030 nm. The FBG had a reflectivity of 80% and it was also used as the output coupler. The gain fiber was 0.6 m with a pump absorption of 250 dB/m. The output power of fiber oscillator was 0.5 mW at the pump power of 80 mW. The pulse duration and spectral bandwidth were 30 ps and 0.09 nm, respectively. The pre-amplifier was composed of 1.1-m gain fiber and used to boost the output power of the oscillator from 0.8 to 38 mW. After the pre-amplifier, the spectral bandwidth was broadened to 0.18 nm. The pulse picker was based on a fiber-coupled acoustic optical modulator (FCAOM, Gooch & Housego). The repetition rate of the pump laser was reduced from 20 MHz to 1, 2 or 3 MHz. After the pulse picker, the average powers were 1.5, 3.1 and 4.5 mW at repetition rates of 1, 2, and 3 MHz, respectively.
Two kinds of feedback cavity schemes were experimentally demonstrated depending on the relative position between the feedback loop and the main amplifier. One was termed as amplifier outside the cavity (AOC) and the other was termed as amplifier inside the cavity (AIC). In the AOC configuration, the low-power pump pulses were firstly coupled into the main amplifier for power enhancement. Then the amplified pulses passed through the WDM and were injected into the PCF for parametric conversion. The fiber length of the WDM was set to be 0.5 m. In this case, the WDM inserted between the gain of the main amplifier and the PCF would induce inevitable nonlinear effects. As for the AIC configuration, the pump pulses passed through the WDM firstly. These pulses were then coupled into the main amplifier and the cascade PCF. In this case, the gain of the main amplifier was directly spliced with the PCF without any transmission fiber. After that, one part of the parametric pulses were coupled out, and the other part of the generated parametric pulses were coupled to the PCF again by the SMF.
Both of the feedback cavities were composed of the same elements including a main amplifier, a piece of PCF, a fiber delay line, an output coupler and a WDM. The main amplifier was pumped by a 976-nm single mode diode laser with a maximum output power of 400 mW. The gain fiber for the main amplifier was 0.3 m with a pump absorption of 1200 dB/m. The length of the PCF used in the OPO was 0.4 m. The fiber delay line was composed of 67, 100 or 200 m SMF and a variable fiber delay stage (General Photonics, MDL-002-D). The fiber delay line and WDM were used to build the feedback loop for the generated parametric signal. The output coupler was composed of a half wave plate and a polarization beam splitter. By rotating the half wave plate, most of the pump laser was extracted from the cavity. The small residual part of pump laser in the cavity would be blocked by the WDM due to its transmission bandwidth. The whole system including electronics was assembled in an aluminum box with dimensions of 39 cm × 28 cm × 13 cm, as the photos shown in Fig. 2. Three pump LDs were already assembled in the box, which were used to pump the fiber oscillator, the pre-amplifier and the main amplifier, respectively. All the elements in the presented laser source were fiber integrated devices, which ensured long-term stability and alignment-free operation.
3.2 Performance of two kinds of OPO
Now, we will turn to characterize the main output characteristics of the implemented OPO. The performance of a CARS imaging system was greatly affected by the output characteristics of its light source. For example, the average power on the sample would affect the intensity of the anti-Stokes signal and the spectral width of the exciting laser would determine the resolution of the resolved vibrational bands. The output characteristics of the OPO varied with different peak powers when changing the repetition rate. Therefore, we measured the output spectra, average powers and spectral widths at three different repetition rates to get a clear insight into the nonlinear effects in the FWM process.
Firstly, we set the repetition rate of the pump pulses to be 1 MHz and changed the length of the SMF to be 200 m. The maximum output power of the pump pulses after the main amplifier was 152 mW. By optimizing the position of the delay line, the corresponding repetition rate of the OPO cavity can be the same as that of the pump laser. In this case, the resonant FWM signal could be observed. The minimum pump power of the 1030-nm pulses for observing the resonance of the FWM signal was designated as the threshold (Pth) of the OPO. The threshold for the AIC and AOC configuration operating at 1 MHz were 55 and 88 mW, respectively, corresponding to a threshold reduction of 37.5%. With the increase of the pump power, the FWM signal was amplified notably, resulting in a maximum output power of 6.6 mW (pump power: 123 mW) and 5.5 mW (pump power: 152 mW) for AIC and AOC, respectively. Further increase of the pump power would give rise to other nonlinear effects and cause a degradation of signal power and spectral purity. The corresponding spectra are shown in Figs. 3(a) and 3(b) for the AIC and AOC, respectively. It can be seen that the noisy spectral peaks caused by the 1st Raman (1079 nm), 2nd Raman (1142 nm), 3rd Raman (1187 nm) and corresponding Raman induced FWM (985, 938, 910 nm) are severe in the AOC configuration. These spectral peaks were originated from the fiber of the WDM inserted between the gain and PCF. The evolution of the signal power and spectral width in AIC and AOC are shown in Figs. 3(c) and 3(d), respectively. Both the average power and the spectral width of generated FWM signal pulses around 800 nm increase almost linearly with the pump power. Although signal pulses with 1 MHz repetition rate can be obtained both in AIC and AOC configurations under the current experimental condition, the pump power threshold, signal power and spectral purity in AIC scheme were improved compared to those in AOC.
Secondly, the repetition rate of the pump pulses was set to be 2 MHz and the length of the SMF was changed to 100 m. The maximum output power of the pump pulses after the main amplifier was about 170 mW. The thresholds were 120 and 145 mW while the maximum output power were 10.0 and 8.7 mW for the AIC and AOC configuration, respectively. The threshold for AIC scheme was reduced by 17.2% compared to AOC. The corresponding spectra are shown in Figs. 4(a) and 4(b). The noisy spectral pair at 985 and 1079 nm still exist in the AOC configuration. The signal power and spectral width with respect to the pump power in AIC and AOC are shown in Figs. 4(c) and 4(d), respectively. They all increase almost linearly with the pump power.
Finally, the repetition rate of the pump pulses was set to be 3 MHz and the length of the SMF was changed to 67 m. The maximum output power of the pump pulse was 178 mW. The thresholds were 164 and 173 mW while the maximum output power were 5.4 and 2.9 mW for AIC and AOC, respectively. The threshold for AIC scheme was reduced by 5.2% compared to AOC. The corresponding spectra are shown in Figs. 5(a) and 5(b). Two small noisy spectral peaks are still observed in the AOC configuration. Figures 5(c) and 5(d) illustrate the evolution of the signal power and spectral width in AIC and AOC, respectively. Their linear relationship with the pump power still preserve. Due to the limited pump power and increased threshold, the output power of 3-MHz OPO is smaller than that of either 1-MHz or 2-MHz OPO. By utilizing high power diode laser or pump combining technique, the output power of OPO could be improved.
From the comparison, we found that our presented AIC scheme had three advantages than the AOC scheme: First, the spectral noise, especially the noisy peaks caused by Raman-induced FWM was eliminated due to the direct splicing of the gain and PCF, which would increase the spectral purity. Second, the threshold of OPO was reduced and the output signal power was increased because the majority of the pump pulses were concentrated at 1030 nm and then converted to FWM sidebands. The reduction of the threshold would benefit for improving the signal power under the condition of limited pump power. Third, the spectral width of the generated signal laser was narrowed. This could be ascribed to the reduced SPM effects for the pump pulses caused by shorter interaction fiber length. As for the drawbacks, inserting a main amplifier inside the fiber OPO cavity might induce additional loss for the feedback pulses. For example, the spectral range of the feedback pulses might lie in the absorption spectral band of the gain medium. In this case, one should carefully select the wavelengths or use enough pump to saturate the gain medium. Additionally, there might be residual recirculating pump light in the cavity. To solve this, optical filters or WDMs could be used to block the unwanted pump light.
The generated signal wavelength is determined by the phase-matching condition defined by involving interaction wavelengths and properties of nonlinear parametric medium. Previous studies have shown that using 1030-nm pump wavelength and PCF with type of LMA-PM-5 would result in a signal wavelength around 800 nm, corresponding to wavenumber difference matching the vibrational bands of CH chemical bonds. What’s more, the residual pump pulses and generated signal pulses are temporally synchronized and spatially overlapped owing to the nonlinear conversion process of FMW and endlessly single-mode property of the PCF, thus avoiding the complicate operation for temporal adjustment and spatial alignment. By changing the delay line in the OPO cavity, the signal wavelength could be tuned from 788 to 797 nm due to the dispersion filtering . This tuning range would be enough to detect the CH chemical bonds of biomedical samples around 2900 cm−1. As for other vibrational bands such as those located in the fingerprint region, using special designed fiber , changing the pump wavelength  or manipulating the input pump polarization  would be a practical way to extend the wavelength tunability. The pulse width of the signal was measured to be 14 ps at a repetition rate of 1 MHz. The measured average power of 6 mW thus indicated a peak power of 430 W, which is sufficient for CARS microscopy. These achieved characteristics of the signal pulses not only favor high-fidelity transmission in a short length of guiding fiber due to the dispersion effect, but also enable a high spectral resolution allowed by the Fourier transformation limit. In our experiment, the spectral bandwidth of the signal was 1.1 nm, corresponding to a spectral resolution of 18 cm−1 in CARS spectroscopy. Considering that the Raman bandwidth of ethanol is about 30 cm−1, the obtained spectral resolution can meet the requirement of spectroscopic identification. Further reducing the spectral bandwidth could be achieved by using pump pulses with narrower spectral bandwidth and longer dispersive cavity . In addition, the average power of the signal pulses was recorded by a power meter (Thorlabs, PM100D) with a sampling rate of 1Hz, showing a RMS fluctuation of 2.6% within one hour.
4. Application to CARS spectroscopy
We measured the CARS spectra of ethanol sample with FOPO in conventional AOC configuration as well as our proposed AIC scheme. The repetition rates of laser pulses were set to be 1 MHz. The experimental setup was shown in Fig. 6(a). The output pulses were filtered by a 750-nm long pass filter to exclude spontaneous FWM in the PCF. Two identical microscope objectives (Daheng Optics 20X NA0.40) were used to focus the pump (788-797 nm) and Stokes (1030 nm) pulses on the sample, and collimate the CARS signal pulses in forward direction. Then the CARS signal passed through a short pass filter (700 nm) and a lens with 50 mm focal length, and was focused into a multimode fiber connected to a fiber optical spectrometer (Ocean Optics, HR4000).
The average powers of pump (788-797 nm) and Stokes (1030 nm) pulses used in our experiment were 5.0 and 3.0 mW for both FOPOs, respectively. Due to the loss of the microscopy objectives, the total excitation power on the sample was about 6.0 mW. The spectral bandwidths of pump and Stokes pulses for both FOPO schemes were shown in Figs. 6(b) and 6(c), respectively. The Stokes spectra show significant SPM structures due to the high peak power. Especially, the spectral bandwidth of Stokes pulse for AOC scheme is almost twice as that for AIC scheme due to the nonlinear effect caused by the fiber coupler. By carefully tuning the focal position and laser wavelength, anti-Stokes radiation of ethanol sample around 646 nm was detected. The AIC cavity design eliminated the spectral noise caused by the fiber coupler, thus outputted laser pulses with narrower spectral bandwidth. As a result, the intensity and the spectral bandwidth of CARS signal at 646 nm were improved in AIC scheme compared to AOC scheme. The signal intensity was increased by two times while the signal bandwidth was decreased by 30%, as shown in Fig. 6(d). After post processing the detected CARS signal, the spectrum of the ethanol sample excited by two FOPO schemes from 2830 cm−1 to 2980 cm−1 were obtained, as shown in Fig. 6(e). The strong peaks around 2900 cm−1 correspond to the vibrational resonances of CH2 and CH3 stretches, which validates the application of our laser source in CARS detection.
In conclusion, we have proposed and implemented an all-fiber OPO operating at a low repetition rate of 1 MHz, which could provide a compact light source suitable for spectroscopic analysis of dense samples by using single-band CARS technique. Investigation on output parameters such as output spectra, signal powers and spectral bandwidths were made at repetition rates of 1, 2 and 3 MHz, indicating effective suppression of nonlinear noise caused by the fiber coupling device. Comparing with previous works, the novel design of OPO configuration enables lower threshold, higher output power and narrower spectral bandwidth. Moreover, the generated synchronized output pulses from our all-fiber OPO were used to detect the CARS spectrum of ethanol sample, which verified the practical utility of this approach. Additionally, the implemented all-fiber laser source was assembled into a portable aluminum box, which is expected to spread the CARS spectroscopy to dense samples with low photodamage and compact setup.
National Natural Science Foundation of China (11504235, 11727812, 11434005, &11404211).
References and links
1. A. Zumbusch, G. R. Holtom, and X. S. Xie, “Three-Dimensional Vibrational Imaging by Coherent Anti-Stokes Raman Scattering,” Phys. Rev. Lett. 82(20), 4142–4145 (1999). [CrossRef]
2. C. H. Camp Jr, Y. J. Lee, J. M. Heddleston, C. M. Hartshorn, A. R. Hight Walker, J. N. Rich, J. D. Lathia, and M. T. Cicerone, “High-Speed Coherent Raman Fingerprint Imaging of Biological Tissues,” Nat. Photonics 8(8), 627–634 (2014). [CrossRef] [PubMed]
3. L. Gao, H. Zhou, M. J. Thrall, F. Li, Y. Yang, Z. Wang, P. Luo, K. K. Wong, G. S. Palapattu, and S. T. Wong, “Label-free high-resolution imaging of prostate glands and cavernous nerves using coherent anti-Stokes Raman scattering microscopy,” Biomed. Opt. Express 2(4), 915–926 (2011). [CrossRef] [PubMed]
4. X. N. He, J. Allen, P. N. Black, T. Baldacchini, X. Huang, H. Huang, L. Jiang, and Y. F. Lu, “Coherent anti-Stokes Raman scattering and spontaneous Raman spectroscopy and microscopy of microalgae with nitrogen depletion,” Biomed. Opt. Express 3(11), 2896–2906 (2012). [CrossRef] [PubMed]
5. C. Liao, P. Wang, C. Y. Huang, P. Lin, G. Eakins, R. T. Bentley, R. Liang, and J. Cheng, “In Vivo and in Situ Spectroscopic Imaging by a Handheld Stimulated Raman Scattering Microscope,” ACS Photonics 5(3), 947–954 (2017). [CrossRef]
6. T. Ideguchi, S. Holzner, B. Bernhardt, G. Guelachvili, N. Picqué, and T. W. Hänsch, “Coherent Raman spectro-imaging with laser frequency combs,” Nature 502(7471), 355–358 (2013). [CrossRef] [PubMed]
7. Y. Zhang, C. S. Liao, W. Hong, K. C. Huang, H. Yang, G. Jin, and J. X. Cheng, “Coherent anti-Stokes Raman scattering imaging under ambient light,” Opt. Lett. 41(16), 3880–3883 (2016). [CrossRef] [PubMed]
8. K. Chen, T. Wu, H. Wei, T. Zhou, and Y. Li, “Quantitative chemical imaging with background-free multiplex coherent anti-Stokes Raman scattering by dual-soliton Stokes pulses,” Biomed. Opt. Express 7(10), 3927–3939 (2016). [CrossRef] [PubMed]
9. Y. Yang, F. Li, L. Gao, Z. Wang, M. J. Thrall, S. S. Shen, K. K. Wong, and S. T. Wong, “Differential diagnosis of breast cancer using quantitative, label-free and molecular vibrational imaging,” Biomed. Opt. Express 2(8), 2160–2174 (2011). [CrossRef] [PubMed]
11. C. W. Freudiger, W. Yang, G. R. Holtom, N. Peyghambarian, X. S. Xie, and K. Q. Kieu, “Stimulated Raman scattering microscopy with a robust fibre laser source,” Nat. Photonics 8(2), 153–159 (2014). [CrossRef] [PubMed]
12. H. Mikami, M. Shiozawa, M. Shirai, and K. Watanabe, “Compact light source for ultrabroadband coherent anti-Stoke Raman scattering (CARS) microscopy,” Opt. Express 23(3), 2872–2878 (2015). [CrossRef] [PubMed]
13. S. Bégin, B. Burgoyne, V. Mercier, A. Villeneuve, R. Vallée, and D. Côté, “Coherent anti-Stokes Raman scattering hyperspectral tissue imaging with a wavelength-swept system,” Biomed. Opt. Express 2(5), 1296–1306 (2011). [CrossRef] [PubMed]
14. X. Chen, X. Xu, D. T. McCormick, K. Wong, and S. T. C. Wong, “Multimodal nonlinear endo-microscopy probe design for high resolution, label-free intraoperative imaging,” Biomed. Opt. Express 6(7), 2283–2293 (2015). [CrossRef] [PubMed]
15. E. S. Lamb, S. Lefrancois, M. Ji, W. J. Wadsworth, X. S. Xie, and F. W. Wise, “Fiber optical parametric oscillator for coherent anti-Stokes Raman scattering microscopy,” Opt. Lett. 38(20), 4154–4157 (2013). [CrossRef] [PubMed]
18. A. Zeytunyan, K. T. Crampton, R. Zadoyan, and V. A. Apkarian, “Supercontinuum-based three-color three-pulse time-resolved coherent anti-Stokes Raman scattering,” Opt. Express 23(18), 24019–24028 (2015). [CrossRef] [PubMed]
19. R. Xie, J. Su, E. C. Rentchler, Z. Zhang, C. K. Johnson, H. Shi, and R. Hui, “Multi-modal label-free imaging based on a femtosecond fiber laser,” Biomed. Opt. Express 5(7), 2390–2396 (2014). [CrossRef] [PubMed]
20. E. Capitaine, N. O. Moussa, C. Louot, S. M. Bardet, H. Kano, L. Duponchel, P. Lévêque, V. Couderc, and P. Leproux, “Fast epi-detected broadband multiplex CARS and SHG imaging of mouse skull cells,” Biomed. Opt. Express 9(1), 245–253 (2018). [CrossRef] [PubMed]
21. Y. H. Zhai, C. Goulart, J. E. Sharping, H. Wei, S. Chen, W. Tong, M. N. Slipchenko, D. Zhang, and J. X. Cheng, “Multimodal coherent anti-Stokes Raman spectroscopic imaging with a fiber optical parametric oscillator,” Appl. Phys. Lett. 98(19), 191106 (2011). [CrossRef] [PubMed]
23. C. Krafft, M. Schmitt, I. W. Schie, D. Cialla-May, C. Matthäus, T. Bocklitz, and J. Popp, “Label-Free Molecular Imaging of Biological Cells and Tissues by Linear and Nonlinear Raman Spectroscopic Approaches,” Angew. Chem. Int. Ed. Engl. 56(16), 4392–4430 (2017). [CrossRef] [PubMed]
24. K. Yang, J. Jiang, Z. Guo, Q. Hao, and H. Zeng, “Tunable Femtosecond Laser From 965 to 1025 nm in Fiber Optical Parametric Oscillator,” IEEE Photonics Technol. Lett. 30(7), 607–610 (2018). [CrossRef]
25. M. Baumgartl, M. Chemnitz, C. Jauregui, T. Meyer, B. Dietzek, J. Popp, J. Limpert, and A. Tünnermann, “All-fiber laser source for CARS microscopy based on fiber optical parametric frequency conversion,” Opt. Express 20(4), 4484–4493 (2012). [CrossRef] [PubMed]
26. T. Gottschall, T. Meyer, M. Baumgartl, B. Dietzek, J. Popp, J. Limpert, and A. Tünnermann, “Fiber-based optical parametric oscillator for high resolution coherent anti-Stokes Raman scattering (CARS) microscopy,” Opt. Express 22(18), 21921–21928 (2014). [CrossRef] [PubMed]
27. M. Brinkmann, S. Janfrüchte, T. Hellwig, S. Dobner, and C. Fallnich, “Electronically and rapidly tunable fiber-integrable optical parametric oscillator for nonlinear microscopy,” Opt. Lett. 41(10), 2193–2196 (2016). [CrossRef] [PubMed]
28. T. Gottschall, T. Meyer, C. Jauregui, F. Just, T. Eidam, M. Schmitt, J. Popp, J. Limpert, and A. Tünnermann, “All-fiber optical parametric oscillator for bio-medical imaging applications,” Proc. SPIE 10083, 100831E (2017).
29. K. Yang, P. Ye, S. Zheng, J. Jiang, K. Huang, Q. Hao, and H. Zeng, “Polarization switch of four-wave mixing in a lawtunable fiber optical parametric oscillator,” Opt. Express 26(3), 2995–3003 (2018). [CrossRef] [PubMed]
30. L. Velázquez-Ibarra, A. Díez, E. Silvestre, and M. V. Andrés, “Wideband tuning of four-wave mixing in solid-core liquid-filled photonic crystal fibers,” Opt. Lett. 41(11), 2600–2603 (2016). [CrossRef] [PubMed]