We try to obtain stable supercontinuum (SC) generation with broad bandwidth under relative simple pump conditions. We use a 1.3-m-long highly nonlinear tellurite microstructured fiber and pump it by a 15 ps 1064 nm fiber laser. One segment of the fiber is tapered from a core diameter of 3.4 μm to 1.3 μm. For the first time five-order stimulated Raman scatterings (SRSs) are observed for soft glass fibers. SC covering 730-1700 nm is demonstrated with the pump-pulse-energy of several nJ. The mechanisms of SC broadening are mainly SRS, self-phase modulation (SPM) and cross phase modulation (XPM). The tapered segment has two advantages. Firstly it increases the nonlinearity of fiber by several times. Secondly, it acts as a compensation for the dispersion of the untapered segment, and mitigates the walk-off between pump pulse and SRS peaks.
©2011 Optical Society of America
Supercontinuum (SC) generation has had a revolutionary impact on the development of nonlinear optics . Applications of SC include optical coherence tomography (OCT) , multiwavelength optical sources , optical frequency metrology (OFM) , and pulse compression (PC) . Various highly nonlinear microstructured fibers benefit the development of SC light source greatly. A preferable SC light source is expected to have the following features: Firstly, it is compact and cost-effective. Though SC generation was demonstrated a long time ago, it found widely applications only after bench-top device with easy maintenance was developed. Secondly, the spectrum is broad enough for the practical applications. Generally an octave width is necessary for the applications such as OFM. Thirdly, it is expected to be stable and have an acceptable coherence. For the applications such as OCT, OFM and PC, a phase-stable SC is essential. Additionally, for some applications, spectral flatness, or power spectra density is important. Though various SC generations have been reported in recent years, the light sources which meet most of the requirements above are rare. In many demonstrations to obtain a broad spectrum, the pump laser is usually the femtosecond pulsed laser with a pulse width in the magnitude of hundreds of femtoseconds. The pump wavelength is in the anomalous dispersion region of the nonlinear fiber, and is close to the zero dispersion wavelength (ZDW). The SC broadening is mainly due to soliton propagation dynamics. The spectrum might be broad, but it is usually sensitive to input pulse fluctuations and pump laser shot noise. Consequently, these broad SCs are characterized by a complex temporal profile, pulse-to-pulse variations in intensity and phase as well as considerable fine structures over their bandwidth [6–8]. When using the pump pulse less than 50 fs, the soliton number can be less than 10. The low soliton number can ensure a high coherence . However, so far such femtosecond laser is still premature commercially. It is sensitive to ambient perturbation, and might be unstable temporally . The SC generation by pumping in the normal dispersion region has a much better coherence. The coherence is generally improved at the expense of spectral broadening given by the same pump pulse . Important advancements have been made in the latest experimental and numerical studies on SC generation by pumping in the normal dispersion region [12–14]. Very recently up to 1.5 octave spanning highly coherent and flat SCs were achieved in a specially designed all-normal dispersion microstructured fiber pumped with femtosecond pulses .
Generally, to construct a cost-effective and compact SC source, the requirements to pump laser should be simplified. If possible, sub-50-fs pulsed laser had better not be used. In particular, pump pulse with temporal width broader than 10 ps is intriguing because it can avoid the complications and inconveniences of using a complex femtosecond oscillator . In this case to construct a stable SC light source with good coherence, the only option might be pumping in the normal dispersion region . Additionally, to ensure a broad wavelength-bandwidth, for example, one octave, the dispersion of fiber should be tailored elaborately, and the nonlinearity of fiber should be as high as possible. Soft glass microstructured fibers, including lead silicate, chalcogenide and tellurite microstructured fibers, can have high nonlinear coefficient [16–19]. The high nonlinear coefficient owes to the high nonlinearity of the glass materials, which is higher than silica glass by one or several orders, and owes to their high linear refractive index which can provide a large index contrast between core and cladding for a compact modal field. However, the dispersions of these fibers are difficult to tailor because of the premature fabrication technology. So far most reported soft glass holey fibers just have a simple air-clad structure. Note that the preforms with complex microstructures might be prepared by advanced techniques, but it is still a challenge to draw them into fibers which can reproduce the microstructures.
In this work, we use a tapered highly nonlinear tellurite microstructured fiber to generate SC by a 15 ps fiber laser with the pulse energy of several nJ. The fiber has a longitudinally varying dispersion. The pump wavelength locates in the normal dispersion region. The SC generation is mainly due to stimulated Raman scattering (SRS), self-phase modulation (SPM) and cross phase modulation (XPM). The tapered segment not only increases the nonlinearity greatly, but also acts as a compensation for the dispersion of the untapered segment, and increases the interaction lengths of SRS and XPM.
The tellurite microstructured fiber is a homemade wagon-wheel-shaped fiber fabricated by the method of rod-in-tube. The detailed fiber-fabrication process can be found in reference . The cross-section of the fiber is shown in the inset of Fig. 1 . A 1.30-m-long fiber was used for tapering. Both tips of the fiber were sealed in advance to avoid the collapse of air-holes during the tapering process. The fiber was put inside the furnace of an elongation tower. The upper tip of the fiber was fixed on the upper movement stage and the bottom tip was fixed on the lower movement stage. The temperature was gradually increased to the softening temperature and then the fiber was drawn gradually. By elaborately controlling the temperature and tension, the tapered microstructure can have the same geometry as the original. The expected core diameter can be obtained by controlling the drawing speed.
The SC spectra were measured by using a commercially available 1064 nm picosecond pulsed fiber laser. The laser was connected to a single-mode fiber (SMF) by a connector. The beam from the SMF was collimated into parallel by a lens with a NA of 0.25. The parallel beam was focused and coupled into the tellurite microstructured fiber by a lens with a NA of 0.47. The output end of the microstructured fiber was mechanically spliced with a silica large-mode-area-fiber by using the butt-joint method. The other end of the large-mode-area-fiber was connected to an optical spectrum analyzer (OSA). The pulse width was 15 ps and the repetition rate was 80 MHz. The coupling efficiency, defined as the launched power divided by the power incident on the lens, was about 35%.
3. Results and discussions
Figure 1 shows the core diameters along the length of the tapered fiber. The original core diameter is 3.4 μm. The smallest core diameter is 1.3 μm. It is predicted from the outside diameter. Figure 2 shows the chromatic dispersion calculated by the fully vectorial finite difference method. The ZDW shifts to a shorter wavelength when the core diameter is reduced to 1.3 μm. ZDW is 1530 nm for the original fiber, while it is 1100 nm for the tapered segment with the core diameter of 1.3 μm. The nonlinear coefficient (γ) was calculated in the same way as reference . For the untapered segment with a core diameter of 3.4 μm, γ is 705 W−1km−1, while for the tapered segment with a core diameter of 1.3 μm, it is 3807 W−1km−1.
Figure 3 shows the measured SC spectra. With the highest pulse energy the whole SC spectrum covers 730-1700 nm. The 20 dB bandwidth covers 960-1610 nm. The increase in SC intensity around the shortest wavelength is ascribed to the noise of OSA (Agilent 86142B, USA). When the pulse energy is not so high, we can observe various orders of SRS peaks clearly. Due to the noise of OSA, the fifth order SRS peak is masked. We used another OSA (Yokogawa AQ6375, Japan) to record a spectrum, which is shown in Fig. 4 . In Fig. 4 we can see the fifth order SRS peak. The peak wavelengths of SRSs, from the first to the fifth, are 1150 nm, 1262 nm, 1386 nm, 1520 nm and 1660 nm, respectively. To our knowledge, it is the first time that five-order SRSs are observed by soft glass fibers.
When the pulse energy is no more than 4.0 nJ, the SC broadening on the short wavelength side of pump wavelength is due to anti-Stokes of SRSs and SPM. With the increasing pulse energy, the anti-Stokes peaks become obscure and the spectra become smooth. In reference  it was shown that the short wavelengths of SC in an all-normal dispersion microstructured fiber are generated through optical wave breaking (OWB) induced four wave mixing effects, which occur when different wavelength components contained in the same SC pulse overlap in time. OWB occurs in normal dispersion region . It results in the smoothing of the spectral profile when pulse energies are high. From Fig. 2 we know that on the short wavelength side the fiber is normal dispersion along the whole fiber length. We think under the pump pulses with high energies the mechanism of SC broadening in short wavelengths might be similar to that in reference .
In this demonstration the SC generation is stable, since the spectra can be repeated very well when we measure at different time by the same pump powers. However, it should be mentioned that this repeatability cannot prove the shot-to-shot stability of the spectra, which are not tested due to the limitation of experimental conditions.
To make a comparison we pumped an untapered 1.60-m-long tellurite microstructured fiber with the same pump laser and coupling conditions. The fiber is the same as the one which is used for tapering. The measured spectra are shown in Fig. 5 .
The walk-off length of pulses can be calculated by, where LW is the walk-off length, λ 1 and λ 2 are peak wavelengths of the pump pulse and SRSs, vg is the group velocity, and T 0 is the pulse width. We calculated the walk-off lengths of pump pulse and SRS peaks for the tapered and untapered fiber. The results are shown in Table 1 .
With the pulse energies around 6.0-7.0 nJ, the SC spectrum of the tapered fiber is much broader than the untapered fiber. The reasons are as follows: Firstly, the nonlinearity is increased greatly by tapering. Second, the dispersion along the tapered fiber is changed due to the variation of core diameter, which favors a broader SC generation. In the following paragraphs we discuss how the longitudinally varying dispersion of the tapered fiber contributes to the broadening of SC.
The width of the pump pulse is 15 ps. For optical pulse with such a width the intrapulse SRS can be neglected. The fiber is pumped at the wavelength away from ZDW, so the high order dispersions are insignificant. Since the pump wavelength locates in the normal
dispersion region, initially the SC is generated by SPM. Figure 6 shows clearly the characteristics of SPM under the high-resolution measurement. The strength of SPM increases linearly with the fiber length, while the strength of SRS increases exponentially with the fiber length [21,22]. Therefore, SRS gradually becomes the dominant mechanism of spectral broadening.
According to Table 1, for the untapered fiber, the walk-off length is 0.60 m between the pump pulse and the first order SRS, and is 0.34 m between the pump pulse and the second order SRS. The interaction length (Leff) of pump pulse and SRS is . Though the length of the untapered fiber is as long as 1.60 m, the pulse energy cannot be transferred from pump pulse to SRSs after the interaction lengths.
For the tapered fiber, after the pulse travels the initial 30-cm-long segment, it extends to a certain extent, but does not split yet, because the initial segment is shorter than the walk-off lengths between any two of the followings: the pump pulse, the first and second order SRS. The following segment which the pulse travels is the tapered section. From Table 1 it can be seen that the walk-off lengths are negative. In the following analyses we discuss how the SRS peaks move within the pulse. Note that what we discuss, is not the propagation velocity of the whole pulse, but the relative move of SRS peaks of intrapulse. The SRS peaks get away from each other in the initial 30-cm-long segment, and then in the tapered segment they move in the converse direction to the previous. Therefore, in the tapered segment firstly they get close to each other, and then get away from each other again when traveling in the final section of the tapered segment. In our case the total length of tapered segment is about 12 cm. It is less than all of the walk-off lengths. If the tapered segment is much longer than the walk-off lengths, the pulse would break into shorter pulses. The shorter pulses with peak wavelengths in the anomalous region would generate solitons which are probably detrimental to the coherence of SC. After the tapered segment, the SRS peaks move in the direction converse to that in the tapered segment again. Therefore, firstly the pulse experiences a process of compression, and then it extends again. Since the fiber length after the tapered segment is longer than the walk-off lengths, finally the pulse splits into shorter pulses. The fifth order SRS has the peak wavelength in the anomalous region. It can be converted into soliton. However, this can be avoided by using a fiber with a larger core diameter which has a longer ZDW, and by using a shorter fiber length after the tapered segment. In this work the fiber length after the tapered segment is too long. Such a length is used only for the safety of the tenuous tapered fiber. Nevertheless, the fifth order SRS is quite weak, so its influences must be insignificant. In Fig. 3 with the increasing pulse energies the SRS peaks merge into a united SC gradually. The troughs among peaks get shallow. It is due to not only SPM, but also XPM. Here XPM benefits from the varying dispersion along the fiber length the same as SRS.
SC by the mechanisms of SRS, SPM, and XPM does not necessarily has a unflattened spectrum which contains obvious SRS peaks and troughs, because the lower order SRS peaks can be consumed by the higher order SRSs, and the troughs can be offset by SPM and XPM to a large extent. To obtain a broadband SC, the highly nonlinear fiber is expected to have a broad, flattened, and small normal dispersion [9,13,14]. Practically fiber with such a dispersion and an extremely high nonlinearity is not available so far. In our experiment, the tapered section functions as a compensation for dispersion. It mitigates the walk-off between pump pulse and SRS peaks. This scheme is similar to the dispersion compensation in optical fiber communications where fibers with different dispersions are connected and their total dispersion is close to zero.
SC covering 730-1700 nm pumped by several nJ picosecond pulses was demonstrated by using a tapered tellurite microstructured fiber. Though the shot-to-shot stability has not been tested, the spectra can be repeated very well when we measure at different time by the same pump power. For the first time five-order SRSs were observed for soft glass fibers. The mechanisms of SC broadening mainly are SRS, SPM, and XPM. The tapered segment not only increases the nonlinearity by several times, but also serves as a way of dispersion compensation, and increases the interaction lengths of SRS and XPM.
This research shows the possibility to obtain stable and broad SC by relative simple pump conditions. The scheme can be further optimized for a better result. For example, the core diameter of the microstructured fiber used for tapering can be larger to ensure the fifth order SRS is in normal dispersion region, and the fiber length after the taper segment can be shortened.
This research was supported by the Japan Society for the Promotion of Science (Postdoctoral Fellowship No. P 11059).
References and links
2. A. F. Fercher and E. Roth, “Ophthalmic laser interferometry,” Proc. SPIE 658, 48–51 (1986).
3. H. Takara, T. Ohara, K. Mori, K. Sato, E. Yamada, Y. Inoue, T. Shibata, M. Abe, T. Morioka, and K. I. Sato, “More than 1000 channel optical frequency chain generation from single supercontinuum source with 12.5 GHz channel spacing,” Electron. Lett. 36(25), 2089–2090 (2000). [CrossRef]
4. D. J. Jones, S. A. Diddams, J. K. Ranka, A. Stentz, R. S. Windeler, J. L. Hall, and S. T. Cundiff, “Carrier-envelope phase control of femtosecond mode-locked lasers and direct optical frequency synthesis,” Science 288(5466), 635–639 (2000). [CrossRef] [PubMed]
5. M. Nisoli, S. De Silvestri, and O. Svelto, “Generation of high energy 10 fs pulses by a new pulse compression technique,” Appl. Phys. Lett. 68(20), 2793–2795 (1996). [CrossRef]
6. K. L. Corwin, N. R. Newbury, J. M. Dudley, S. Coen, S. A. Diddams, K. Weber, and R. S. Windeler, “Fundamental noise limitations to supercontinuum generation in microstructure fiber,” Phys. Rev. Lett. 90(11), 113904 (2003). [CrossRef] [PubMed]
7. J. Herrmann, U. Griebner, N. Zhavoronkov, A. Husakou, D. Nickel, J. C. Knight, W. J. Wadsworth, P. S. J. Russell, and G. Korn, “Experimental evidence for supercontinuum generation by fission of higher-order solitons in photonic fibers,” Phys. Rev. Lett. 88(17), 173901 (2002). [CrossRef] [PubMed]
8. X. Gu, M. Kimmel, A. Shreenath, R. Trebino, J. Dudley, S. Coen, and R. Windeler, “Experimental studies of the coherence of microstructure-fiber supercontinuum,” Opt. Express 11(21), 2697–2703 (2003). [CrossRef] [PubMed]
9. J. M. Dudley, G. Genty, and S. Coen, “Supercontinuum generation in photonic crystal fiber,” Rev. Mod. Phys. 78(4), 1135–1184 (2006). [CrossRef]
12. A. M. Heidt, “Pulse preserving flat-top supercontinuum generation in all-normal dispersion photonic crystal fibers,” J. Opt. Soc. Am. B 27(3), 550–559 (2010). [CrossRef]
13. L. E. Hooper, P. J. Mosley, A. C. Muir, W. J. Wadsworth, and J. C. Knight, “Coherent supercontinuum generation in photonic crystal fiber with all-normal group velocity dispersion,” Opt. Express 19(6), 4902–4907 (2011). [CrossRef] [PubMed]
14. A. M. Heidt, A. Hartung, G. W. Bosman, P. Krok, E. G. Rohwer, H. Schwoerer, and H. Bartelt, “Coherent octave spanning near-infrared and visible supercontinuum generation in all-normal dispersion photonic crystal fibers,” Opt. Express 19(4), 3775–3787 (2011). [CrossRef] [PubMed]
15. S. Coen, A. H. L. Chau, R. Leonhardt, J. D. Harvey, J. C. Knight, W. J. Wadsworth, and P. S. J. Russell, “Supercontinuum generation by stimulated Raman scattering and parametric four-wave mixing in photoniccrystal fibers,” J. Opt. Soc. Am. B 19(4), 753–764 (2002). [CrossRef]
16. V. V. Kumar, A. K. George, W. H. Reeves, J. C. Knight, P. Russell, F. Omenetto, and A. Taylor, “Extruded soft glass photonic crystal fiber for ultrabroad supercontinuum generation,” Opt. Express 10(25), 1520–1525 (2002). [PubMed]
17. F. G. Omenetto, N. A. Wolchover, M. R. Wehner, M. Ross, A. Efimov, A. J. Taylor, V. V. R. K. Kumar, A. K. George, J. C. Knight, N. Y. Joly, and P. St. J. Russell, “Spectrally smooth supercontinuum from 350 nm to 3 mum in sub-centimeter lengths of soft-glass photonic crystal fibers,” Opt. Express 14(11), 4928–4934 (2006). [CrossRef] [PubMed]
18. M. Liao, C. Chaudhari, G. Qin, X. Yan, C. Kito, T. Suzuki, Y. Ohishi, M. Matsumoto, and T. Misumi, “Fabrication and characterization of a chalcogenide-tellurite composite microstructure fiber with high nonlinearity,” Opt. Express 17(24), 21608–21614 (2009). [CrossRef] [PubMed]
19. M. Liao, X. Yan, G. Qin, C. Chaudhari, T. Suzuki, and Y. Ohishi, “A highly non-linear tellurite microstructure fiber with multi-ring holes for supercontinuum generation,” Opt. Express 17(18), 15481–15490 (2009). [CrossRef] [PubMed]
20. D. Anderson, M. Desaix, M. Lisak, and M. L. Quiroga-Teixeiro, “Wave breaking in nonlinear-optical fibers,” J. Opt. Soc. Am. B 9(8), 1358–1361 (1992). [CrossRef]
21. D. J. Richardson, J. Nilsson, and W. A. Clarkson, “High power fiber lasers: current status and future perspectives,” J. Opt. Soc. Am. B 27(11), B63–B92 (2010). [CrossRef]
22. G. P. Agrawal, Nonlinear Fiber Optics, 4th ed. (Academic, 2006).