Open Access
Add to BibSonomy
Abstract
A single-cavity dual-wavelength all-fiber femtosecond laser is designed to generate 1030 nm wavelength for high resolution multiphoton imaging and 1700 nm wavelength for long penetration depth imaging. Considering two-photon and three-photon microscopy (2PM and 3PM), the proposed laser provides the single-photon wavelength equivalent to 343 nm, 515 nm, 566 nm and 850 nm, that can be employed to excite a wide variety of intrinsic fluorophores, dyes, and fluorescent proteins. Generating two excitation wavelengths from a single laser reduces the footprint and cost significantly compared to having two separate lasers. Furthermore, an all-reflective microscope is designed to eliminate the chromatic aberration while employing two excitation wavelengths. The compact all-fiber alignment-free laser design makes the overall size of the microscope appropriate for clinical applications.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Multiphoton microscopy (MPM) has been recognized as a powerful technique for imaging of biological samples at submicron optical resolution. Compared to fluorescence confocal microscopy, MPM has the following advantages: 1) since multiphoton absorption happens only at the focal point of the objective lens, MPM provides optical sectioning capability and avoids out of focus photo-bleaching; 2) employing near infrared (NIR) wavelength results in less scattering and allows imaging deeper into the tissues; 3) MPM provides various nonlinear imaging modalities such as second harmonic generation (SHG), third harmonic generation (THG), two and three-photon excited fluorescence (2PEF/3PEF) which can be used to image different structures in a specimen [1–5].
In most studies, a Ti:sapphire or a fiber-based femtosecond laser with a single wavelength has been employed as the excitation source for MPMs [6–9]. Therefore, only a few fluorophores could be imaged with a given MPM design which significantly limits its applications. To simultaneously image various fluorophores with optimum sensitivity and selectivity, a multi-wavelength femtosecond source is required. Such a multiphoton microscope design will expand its applications in genetics (DNA sequencing) [10], analyzing neural network architecture [11], examining various immune cells behavior [12,13], and tracking tumor cell [14].
In most cases, the optical system in an MPM is optimized for a specific wavelength, therefore, the femtosecond laser cannot be simply replaced with another one without correcting for chromatic aberrations introduced to the system especially when the wavelength difference between two sources is significant. Amirsolaymani et al. proposed an all-reflective optical design to remove the adverse effects of chromatic aberration when using two separate lasers with 1040 nm and 1550 nm wavelengths [15]. Alternatively in [16], chromatic aberration compensation is applied for label-free imaging using a mode-locked Ti:sapphire pulsed laser (700 to 900 nm) and an optical parametric oscillator (1000 to 1550 nm). Although microscope designs mentioned above carefully correct for chromatic aberration, they use two separate laser sources to generate two wavelengths which increases the size and cost of the microscope.
Wang et al. have demonstrated a fiber-based three-color femtosecond source using a single cavity for simultaneous excitation of three fluorescent proteins [17]. In this work, soliton self-frequency shift along with frequency doubling are employed to generate three excitation wavelengths at 775 nm, 864 nm, and 950 nm chosen to excite red, cyan, and yellow fluorescent proteins through two-photon absorption. Even though the three wavelengths are generated from a single cavity which reduces the cost, the laser source is not an all-fiber design which makes it vulnerable to alignment and vibration. Similar approach has been recently presented in [18,19].
In this paper, a single-cavity dual-wavelength all-fiber femtosecond laser is proposed which can be employed to do simultaneous imaging of various intrinsic and extrinsic fluorophores through 2PM and 3PM modalities. The proposed laser generates two femtosecond pulse trains at 1030 nm and 1700 nm. The excitation wavelength 1030 nm can be used for high resolution three-photon imaging of aminocoumarin, Hoechst, DAPI, etc. In two-photon excitation, this wavelength can be employed to image Cy3, CyTRAK Orange, YPet, TurboYFP, etc. On the other hand, it has been shown in numerous studies that 1700 nm excitation wavelength is of particular interest for long penetration depth MPM imaging due to less scattering in biological tissues and lower water absorption in this spectral window [20,21]. Furthermore, in 3PM imaging, this wavelength is equivalent to 565 nm which is an ideal wavelength for exciting a wide selection of fluorophores such as G-Dye200, Cy3B, Rhodamine B, LDS 751, RFP, etc. Therefore, the clever selection of the two wavelengths provides imaging flexibility and enables us to image a variety of samples with two-and three-photon processes. However, it has to be noted that the excitation efficiency of various imaging modalities can be significantly different depending on the sample structure. For instance, collagen in tissue shows a very strong SHG signal but almost no THG signal.
The excitation wavelength 1030 nm is obtained by a supercontinuum pumped by an Er-doped mode-locked laser followed by an Yb-doped fiber amplifier [22]. The other output of the laser at 1700 nm is generated by propagating the Er-doped mode-locked laser output through 11 meters of a standard single mode (SM) fiber by means of soliton self-frequency shift (SSFS) [23–26]. Finally, a microscope with no chromatic aberration is designed to perform multimodal multiphoton imaging using these two excitation wavelengths.
2. Laser design and characterization
As shown in Fig. 1, an Er-doped fiber laser cavity is passively mode-locked using a fiber-pigtailed saturable absorber mirror (SAM-1550-40-2ps, BATOP GmbH, Germany). The cavity dispersion is optimized to generate $\approx$ 400 fs transform-limited pulses at $\approx$ 37 MHz repetition rate. The oscillator output power is $\approx$ 2 mW which is split into two equal parts where each part is separately amplified using a single stage Er-doped amplifier to about 80 mW.
The Er-doped gain fiber has a normal dispersion which prevents pulse-breaking while the spectral broadening occurs in the amplifier. The chirped pulses at the output of the amplifiers are compressed to 200 fs using a standard single mode fiber (SMF28, Corning, USA). The upper amplifier in Fig. 1 is used to generate 1030 nm output as described in the followings. A short piece (4 cm) of highly nonlinear fiber (HNLF-ST, OFS, USA) is spliced to the SMF28 fiber. The splice is optimized to reduce the loss to less than 20%. The supercontinuum generated at the output of the HNLF contains $\approx$ 1 mW power near 1030 nm wavelength which is amplified to 100 mW by the reverse-pumped Yb-doped amplifier. Note that the supercontinuum spectral distribution could be fine-tuned by the Er-doped amplifier pump power and the length of compression fiber.
The output spectrum of the Yb-doped amplifier is shown in Fig. 2(a). The output pulses from Yb-doped amplifier were highly chirped due to normal dispersion of the gain fiber and self-phase modulation effect. However, the compression down to FWHM $\approx$ 270 fs was accomplished by 2.9 m of a hollow core photonics crystal fiber (HC-1060, NKT Photonics, Denmark) using a series of cutbacks. The splice between SM fiber and hollow core fiber has been optimized using the procedure explained in [27] and a minimum splice loss of 1.5 dB was achieved. A fiber isolator is used to mitigate the back reflection from the air/silica interface in this splice. The compression fiber did not introduce any nonlinearity and the 1030 nm spectral window remained unchanged. This was expected due to the low Kerr effect of the hollow core fiber. As shown in the autocorrelation trace (Fig. 2(b)), the pulse is not fully compressed, which can be explained by the contribution of the higher-order dispersion of the compression fiber [28]. However, the pulse is very stable in time. Note that any instability in the pulse energy will result in a dramatic change in the multiphoton signal due to quadratic or cubic relationship of the signal with the peak power [29,30]. A shorter pulse can be achieved by using a compression fiber with lower higher-order dispersion at 1030 nm spectral window. The residual power outside the 1030 nm spectral window is filtered out using a bandpass filter (FF01-1055/70, Semrock, USA).
Fig. 2. Dual-wavelength laser output spectra and autocorrelation traces: (a, b) 1030 nm output. (c, d) 1700 nm output.
The soliton self-frequency shift is a well-known method to red-shift an ultrashort pulsed laser wavelength through intra-pulse stimulated Raman scattering which was first experimentally observed by Mitschke [31] and further explored in more details later by Lee [32]. In our design, the lower Er-doped amplifier in Fig. 1 is used to generate 1700 nm output through SSFS in a piece (11 m) of a standard single mode fiber (SMF28, Corning, USA). The pump power of the corresponding Er amplifier as well as the length of the SMF fiber are optimized to obtain conversion efficiency of around 50%. The output power at 1700 nm is 40 mW. The output spectrum is shown in Fig. 2(c). The residual power at 1550 nm is filtered out using a long pass filter (edge at 1650 nm). The full width half maximum bandwidth (FWHM) at 1700 nm is 30 nm and the pulse is nearly transform-limited (FWHM$\approx$120 fs). The autocorrelation trace of the pulse is shown in Fig. 2(d). To scale up the output power, one can employ a polarization maintaining (PM) double-stage Er-amplifier followed by a novel pulse compression method along with a PM large mode area (LMA) fiber for SSFS. Note that the linewidths of both lasers are narrow enough to prevent bleed-through when imaging multiple fluorophores. The entire laser could be fitted in a 20 cm$\times$20 cm box which makes it suitable for most clinical applications.
3. Microscope design and performance analysis
A compact mirror-based microscope is designed to eliminate chromatic aberration so that it can operate with two wavelengths without further adjustment (Fig. 3). Two $90^o$ off-axis parabolic mirrors (RC08APC-P01, Thorlabs, USA) are employed to collimate both lasers. The dichroic mirror DM1 (87-050, Edmund Optics, USA) is used to combine two wavelengths in a single path. The collimated beams are then focused on the sample using a reflecting microscope objective (5004-000, Beck, England). This 36X objective has a numerical aperture (NA) of 0.5 and a long working distance (8.6 mm) which makes it convenient to use for imaging various samples. The emitted light is reflected by DM2 (Di02-R980, Semrock, USA) and then separated to three spectral windows:$\lambda$<495 nm, 495 nm<$\lambda$<555 nm, and $\lambda$>555 nm using DM3 and DM4 (FF555-Di03, FF495-Di03, Semrock, USA). Appropriate filters are used to pass the desired portion of each spectral window toward the corresponding photo-multiplier tube (PMT). To maximize the detected signal to noise ratio (SNR), PMTs (H7422-40, Hamamatsu, Japan) are chosen so that they are highly sensitive in the desired spectral window.
Fig. 3. Mirror-based multiphoton microscope schematic diagram. ML: mode-locked, PM: parabolic mirror, DM: dichroic mirror, F: filter, PMT: photo-multiplier tube, RMO: reflecting microscope objective.
Instead of using a conventional galvo mirror system to scan the beam, we employed a resonant XY piezo stage (P-611.20, Physik Instrumente, Germany) to raster scan the sample at the focal plane of the microscope objective. Compared to a galvo mirror scanning microscope, our approach has the following advantages: 1) since the objective is being used only for on-axis beam, it does not introduce any field-dependent aberration (such as coma, astigmatism, field curvature, etc.) to the system, therefore, the image resolution does not degrade as we go toward the edge of the field of view (FOV); 2) the relay lenses that were required in conventional scanning microscope to image the galvo mirror to the back aperture of the objective can be eliminated; 3) the FOV of the system is as large as the travel range of the piezo cube which can be more than 1.8 mm. Note that the travel range of the resonant piezo stage used in the present design is 120 $\mu$m.
To characterize the microscope performance, the 1700 nm output is used to acquire the THG image of a silicon nano-waveguide attached to a ring resonator (Fig. 4(a)). Since the width of the waveguide is $\approx$ 200 nm, it is used to quantitatively measure the resolution of the microscope. The intensity profile of a cross section of the ring resonator is plotted in Fig. 4(b). The FWHM of the Gaussian fit is 2.2 $\mu$m which is in agreement with the theoretical diffraction-limited spot size of the proposed microscope $(1.22\lambda /(\sqrt {3}$ NA$))$. Based on Rayleigh criteria, the minimum resolvable feature in THG mode is about $\approx$ 1.1 $\mu$m which can be improved by using a microscope objective with higher NA. Note that all images acquired with the proposed microscope are $512\times 512$ pixels and the acquisition time for each image is 10 sec which is limited by the piezo scanning cube.
Fig. 4. (a) THG image of silicon nano-waveguide using $\lambda _2$=1700 nm laser output. (b) Intensity profile of the waveguide. The FWHM of the Gaussian fit is 2.2 $\mu$m. (c) SHG image of 2 $\mu$m polystyrene beads using $\lambda _1$=1030 nm laser output. The scale bar is 25$\mu$m. (d) The corresponding SHG/THG spectra of the images shown in (a) and (c).
Furthermore, in Fig. 4(c), we used the 1030 nm output of the laser to acquire SHG image of 2 $\mu$m polystyrene beads (Spherotech Inc., USA). The corresponding SHG/THG spectra of the silicon waveguide and polystyrene beads measured by replacing the PMTs with a spectrometer (USB2000+, Ocean Optics, USA) are shown in Fig. 4(d).
As discussed before having a dual-wavelength femtosecond laser and a multiphoton microscope with no chromatic dispersion will help to expand the variety of samples that can be imaged through different MPM modalities. For instance, we could not acquire a multiphoton image from euphorbia cactus leaf using 1700 nm output. However, by employing 1030 nm laser, we could image this leaf through intrinsic two photon excitation fluorescence as shown in Fig. 5(a). The opposite scenario happened in imaging a stained dog olfactory membrane. The 1030 nm output did not generate any multiphoton signal while 1700 nm laser was used to acquire the 3PEF image in Fig. 5(b). Note that switching between two outputs of the laser could be done in the software by turning on the corresponding amplifier pump laser diode with no further change or adjustment in the microscope.
Fig. 5. (a) Intrinsic 2PEF image of an euphorbia cactus leaf acquired using 1030 nm output. (b) 3PEF image of a stained section of a dog olfactory membrane using 1700 nm output. (c) 2PEF (red) using 1030 nm and THG (green) using 1700 nm from 5 $\mu$m polystyrene beads drop casted on a GaAs chip. The scale bar is 25$\mu$m.
In Fig. 5(c), we employed two wavelengths to image 5 $\mu$m polystyrene beads drop casted on GaAs chip. Using 1030 nm, we observed 2PEF from the surface of GaAs (shown in red) and using 1700 nm wavelength, we obtained THG from polystyrene beads.
4. Conclusion
A single-cavity dual-wavelength all-fiber femtosecond laser along with a dispersion-less microscope are designed to facilitate imaging various intrinsic and extrinsic fluorophores through 2PM and 3PM modalities. The proposed laser generates two femtosecond pulse trains at 1030 nm and 1700 nm. The truly all-fiber design of the laser makes it compact, robust and alignment-free. Finally, a mirror-based microscope is designed to prevent any chromatic dispersion. As a result, multimodal multiphoton imaging using two excitation wavelengths could be performed with no further adjustment.
Funding
NSF ERC CIAN; State of Arizona TRIF Photonics.
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. W. Denk, J. H. Strickler, and W. W. Webb, “Two-photon laser scanning fluorescence microscopy,” Science 248(4951), 73–76 (1990). [CrossRef]
2. W. R. Zipfel, R. M. Williams, and W. W. Webb, “Nonlinear magic: multiphoton microscopy in the biosciences,” Nat. Biotechnol. 21(11), 1369–1377 (2003). [CrossRef]
3. K. König, “Multiphoton microscopy in life sciences,” J. Microsc. 200(2), 83–104 (2000). [CrossRef]
4. J. N. Kerr and W. Denk, “Imaging in vivo: watching the brain in action,” Nat. Rev. Neurosci. 9(3), 195–205 (2008). [CrossRef]
5. R. A. Aviles-Espinosa, S. I. Santos, A. Brodschelm, W. G. Kaenders, C. Alonso-Ortega, D. Artigas-García, and P. Loza-Alvarez, “Third-harmonic generation for the study of caenorhabditis elegans embryogenesis,” J. Biomed. Opt. 15(4), 046020 (2010). [CrossRef]
6. C. Xu, J. Guild, W. W. Webb, and W. Denk, “Determination of absolute two-photon excitation cross sections by in situ second-order autocorrelation,” Opt. Lett. 20(23), 2372–2374 (1995). [CrossRef]
7. L. Fu, X. Gan, and M. Gu, “Nonlinear optical microscopy based on double-clad photonic crystal fibers,” Opt. Express 13(14), 5528–5534 (2005). [CrossRef]
8. S. Tang, J. Liu, T. B. Krasieva, Z. Chen, and B. J. Tromberg, “Developing compact multiphoton systems using femtosecond fiber lasers,” J. Biomed. Opt. 14(3), 030508 (2009). [CrossRef]
9. K. Kieu, S. Mehravar, R. Gowda, R. A. Norwood, and N. Peyghambarian, “Label-free multi-photon imaging using a compact femtosecond fiber laser mode-locked by carbon nanotube saturable absorber,” Biomed. Opt. Express 4(10), 2187–2195 (2013). [CrossRef]
10. K. M. Berland, “Detection of specific dna sequences using dual-color two-photon fluorescence correlation spectroscopy,” J. Biotechnol. 108(2), 127–136 (2004). [CrossRef]
11. J. Livet, T. A. Weissman, H. Kang, R. W. Draft, J. Lu, R. A. Bennis, J. R. Sanes, and J. W. Lichtman, “Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system,” Nature 450(7166), 56–62 (2007). [CrossRef]
12. G. D. Victora, T. A. Schwickert, D. R. Fooksman, A. O. Kamphorst, M. Meyer-Hermann, M. L. Dustin, and M. C. Nussenzweig, “Germinal center dynamics revealed by multiphoton microscopy with a photoactivatable fluorescent reporter,” Cell 143(4), 592–605 (2010). [CrossRef]
13. Z. Fan, J. A. Spencer, Y. Lu, C. M. Pitsillides, G. Singh, P. Kim, S. H. Yun, V. Toxavidis, T. B. Strom, C. P. Lin, and M. Koulmanda, “In vivo tracking of’color-coded’effector, natural and induced regulatory T cells in the allograft response,” Nat. Med. 16(6), 718–722 (2010). [CrossRef]
14. E. B. Voura, J. K. Jaiswal, H. Mattoussi, and S. M. Simon, “Tracking metastatic tumor cell extravasation with quantum dot nanocrystals and fluorescence emission-scanning microscopy,” Nat. Med. 10(9), 993–998 (2004). [CrossRef]
15. B. Amirsolaimani, B. Cromey, N. Peyghambarian, and K. Kieu, “All-reflective multiphoton microscope,” Opt. Express 25(19), 23399–23407 (2017). [CrossRef]
16. A. Filippi, E. Dal Sasso, L. Iop, A. Armani, M. Gintoli, M. Sandri, G. Gerosa, F. Romanato, and G. Borile, “Multimodal label-free ex vivo imaging using a dual-wavelength microscope with axial chromatic aberration compensation,” J. Biomed. Opt. 23(9), 091403 (2018). [CrossRef]
17. K. Wang, T.-M. Liu, J. Wu, N. G. Horton, C. P. Lin, and C. Xu, “Three-color femtosecond source for simultaneous excitation of three fluorescent proteins in two-photon fluorescence microscopy,” Biomed. Opt. Express 3(9), 1972–1977 (2012). [CrossRef]
18. A. Zach, M. Mohseni, C. Polzer, J. Nicholson, and T. Hellerer, “All-fiber widely tunable ultrafast laser source for multimodal imaging in nonlinear microscopy,” Opt. Lett. 44(21), 5218–5221 (2019). [CrossRef]
19. L. Huang, X. Zhou, Q. Liu, C. E. MacAulay, and S. Tang, “Miniaturized multimodal multiphoton microscope for simultaneous two-photon and three-photon imaging with a dual-wavelength er-doped fiber laser,” Biomed. Opt. Express 11(2), 624–635 (2020). [CrossRef]
20. N. G. Horton, K. Wang, D. Kobat, C. G. Clark, F. W. Wise, C. B. Schaffer, and C. Xu, “In vivo three-photon microscopy of subcortical structures within an intact mouse brain,” Nat. Photonics 7(3), 205–209 (2013). [CrossRef]
21. F. Akhoundi, Y. Qin, N. Peyghambarian, J. K. Barton, and K. Kieu, “Compact fiber-based multi-photon endoscope working at 1700 nm,” Biomed. Opt. Express 9(5), 2326–2335 (2018). [CrossRef]
22. K. Kieu, R. J. Jones, and N. Peyghambarian, “High power femtosecond source near 1 micron based on an all-fiber er-doped mode-locked laser,” Opt. Express 18(20), 21350–21355 (2010). [CrossRef]
23. K. Wang and C. Xu, “Tunable high-energy soliton pulse generation from a large-mode-area fiber and its application to third harmonic generation microscopy,” Appl. Phys. Lett. 99(7), 071112 (2011). [CrossRef]
24. P. Cadroas, L. Kotov, L. Abdeladim, M. Likhachev, D. Lipatov, A. Hideur, W. Supatto, J. Livet, E. Beaurepaire, and S. Février, “Three-photon microscopy with a monolithic all-fiber format laser emitting at 1650 nm,” in Specialty Optical Fibers, (Optical Society of America, 2016), pp. SoM4F–4.
25. T. N. Nguyen, K. Kieu, D. Churin, T. Ota, M. Miyawaki, and N. Peyghambarian, “High power soliton self-frequency shift with improved flatness ranging from 1.6 to 1.78 µm,” IEEE Photonics Technol. Lett. 25(19), 1893–1896 (2013). [CrossRef]
26. C. Xu and F. Wise, “Recent advances in fibre lasers for nonlinear microscopy,” Nat. Photonics 7(11), 875–882 (2013). [CrossRef]
27. J. T. Kristensen, A. Houmann, X. Liu, and D. Turchinovich, “Low-loss polarization-maintaining fusion splicing of single-mode fibers and hollow-core photonic crystal fibers, relevant for monolithic fiber laser pulse compression,” Opt. Express 16(13), 9986–9995 (2008). [CrossRef]
28. D. Turchinovich, X. Liu, and J. Lægsgaard, “Monolithic all-pm femtosecond yb-fiber laser stabilized with a narrow-band fiber bragg grating and pulse-compressed in a hollow-core photonic crystal fiber,” Opt. Express 16(18), 14004–14014 (2008). [CrossRef]
29. A. Fernández, L. Grüner-Nielsen, M. Andreana, M. Stadler, S. Kirchberger, C. Sturtzel, M. Distel, L. Zhu, W. Kautek, R. Leitgeb, A. Baltuska, K. Jesperson, and A. Verhoef, “Optimizing pulse compressibility in completely all-fibered ytterbium chirped pulse amplifiers for in vivo two photon laser scanning microscopy,” Biomed. Opt. Express 8(8), 3526–3537 (2017). [CrossRef]
30. N. Davoudzadeh, G. Ducourthial, and B. Q. Spring, “Custom fabrication and mode-locked operation of a femtosecond fiber laser for multiphoton microscopy,” Sci. Rep. 9(1), 4233 (2019). [CrossRef]
31. F. M. Mitschke and L. F. Mollenauer, “Discovery of the soliton self-frequency shift,” Opt. Lett. 11(10), 659–661 (1986). [CrossRef]
32. J. H. Lee, J. van Howe, C. Xu, and X. Liu, “Soliton self-frequency shift: experimental demonstrations and applications,” IEEE J. Sel. Top. Quantum Electron. 14(3), 713–723 (2008). [CrossRef]
