Soliton self-frequency shift (SSFS) and third-harmonic generation (THG) are observed in a four-hole As2S5 chalcogenide microstructured optical fiber (MOF). The As2S5 MOF is tapered to offer an ideal environment for SSFS. After tapering, the zero-dispersion wavelength (ZDW) shifts from 2.02 to 1.61 μm, and the rate of SSFS can be enhanced by increasing the energy density of the pulse. By varying the average input power from 220 to 340 mW, SSFS of a soliton central wavelength from 2.206 to 2.600 μm in the mid-infrared is observed in the tapered segment, and THG at 632 nm is observed in the untapered segment.
©2014 Optical Society of America
Soliton self-frequency shift (SSFS) in optical fibers is a phenomenon in which soliton pulses, propagating with anomalous group velocity dispersion, experience a continuous redshift because of intrapulse stimulated Raman scattering (SRS) from optical phonons [1–4]. Since its initial discovery by Mitschke et al. in 1986 , SSFS has attracted much attention because of its potential applications [6–10]. More recently, many studies on SSFS have been conducted involving various optical fibers, such as tapered microstructured optical fibers (MOFs) [11, 12], photonic crystal fibers (PCFs) , hollow-core photonic bandgap fibers , higher-order mode fibers , and tellurite and chalcogenide MOFs [16, 17]. On the other hand, third-harmonic generation (THG) in optical fibers with various structures and materials has also been widely reported since its initial discovery by Gabriagues in 1983. THG mainly depends on two factors: third-order susceptibility (χ(3)) and phase-matching (PM) of the fundamental (ω) to the third-harmonic (TH, 3ω) waves [18–22]. Recently, soft-glass optical fibers, such as tellurite, fluoride, and chalcogenide optical fibers, have attracted much attention because of their high nonlinearity and wide transmission window [23, 24]. In particular, chalcogenide glasses with varying compositions may lead to χ(3) 100-1000 times that of silica, which can greatly enhance SSFS and THG [23, 25].
In this paper, we experimentally observed SSFS and THG in a four-hole As2S5 chalcogenide MOF. When a soliton pulse propagated in this As2S5 MOF, a continuous soliton wavelength shift from 2.206 to 2.600 μm occurred in the tapered segment, which was as short as 1.5 cm in length. THG at 632 nm was concurrently observed in the untapered segments at the input and output ends of the As2S5 MOF.
2. As2S5 chalcogenide MOF
Figure 1(a) shows the transmission spectra of As2S5 and As2S3 glasses with a thickness of 1 mm. It is clear that the As2S5 glass exhibits higher transmission than As2S3 glass in the spectral range from 0.5 to 9.9 μm. The As2S5 and As2S3 glasses exhibit transmissions of 68.7% and 64.5% at ~1900 nm, respectively. In addition, the As2S5 glass has a shorter cut-off wavelength in the visible range than the As2S3 glass. As shown in the inset of Fig. 1(a), the As2S5 glass cuts off at ~526 nm while the As2S3 glass cuts off at ~562 nm. Therefore, the As2S5 glass used in this work is more effective in the spectral range from 0.5 to 9.9 μm than the As2S3 glass. The four-hole geometry was selected for the As2S5 chalcogenide MOF because of its easy fabrication and low confinement loss.
The four-hole As2S5 chalcogenide MOF was fabricated with the rod-in-tube technique. First, the As2S5 glass was prepared in the form of rods (outer diameter of ~12 mm) by a direct synthesis in an evacuated silica ampoule from 5N elements at a temperature of 650 °C. The resulting glass rods were annealed near the glass transition temperature for 2 hours to stabilize their structure and to relieve internal stress . The As2S5 glass transition temperature, Tg = 140 °C, was obtained with differential scanning calorimeter (DSC) measurements. Second, one of the As2S5 rods (8 cm in length) was ultrasonically drilled to form a structured rod with four air holes (~2 mm diameter) surrounding its center. The other two rods were drilled to form tubes with a central hole of 3 mm diameter. Third, the structured rod was elongated to match its inner diameter to that of the chalcogenide tubes (3 mm). The softening temperature, Ts = 226 °C, was defined by the temperature of maximal expansion, which was measured by thermal mechanical analysis (TMA). The elongated structured rod was inserted into one of the As2S5 tubes and they were elongated together to obtain a preform with outer diameter ~3 mm. Finally, the preform was inserted into another As2S5 tube and drawn into a MOF. During the fiber-drawing process, a positive pressure of nitrogen gas was added to maintain the holes in the preform.
A 15 cm-long As2S5 MOF was put inside the furnace of an elongation apparatus for tapering. 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 tension can be controlled by varying the drawing speed. In turn, the expected core diameter of the MOF can be inferred from the tension. The temperature was gradually increased to 183 °C, which was chosen upon the basis of numerous experimental trials. During the tapering process, the tapered segment maintained the same geometry as the original. Tapering has several advantages, such as, increasing the nonlinearity of As2S5 MOF, obtaining anomalous dispersion and enhancing the rate of SSFS by increasing the energy density of the pulse. These advantages offer an ideal environment for SSFS in the tapered segment of the As2S5 MOF. On the other hand, the untapered segment also has advantages, such as easy coupling and relief of the pump power limitation imposed by the surface damage threshold.
The structure of the tapered As2S5 MOF is shown in the inset of Fig. 1(b), which was observed with an optical microscope. The core and cladding diameters of the tapered segment were ~1.42 μm and ~62.2 μm, and were ~2.62 μm and ~115.8 μm for the untapered segments. The lengths of the untapered input end, tapered segment and untapered output end were ~1.8 cm, 1.5 cm and 4.5 cm, respectively. Parts I and II were short non-uniform waists (about ~2 mm) and can be neglected. The coupling light loss from the untapered segment to the tapered segment was small, because the fiber structure did not collapse. In order to establish the refractive index of As2S5 glass at random wavelengths, the refractive index of As2S5 glass was measured at different wavelengths from 500 nm to 4200 nm using prisms, and the measured values were then fitted to the Sellmeier formula to obtain Sellmeier coefficients, as shown in Table 1. The fundamental mode (HE11), effective refractive indices (neff), and the chromatic dispersions (D) of the tapered and untapered segments were calculated based on the Sellmeier coefficients for As2S5 glass. After tapering, longitudinally varying dispersion obtained, and the zero-dispersion wavelength (ZDW) shifted from 2.02 μm to 1.61 μm. An 8 m-long untapered As2S5 MOF was used to measure the loss, following the cut-back technique. The loss was ~1.8 dB/m at 1900 nm, mainly due to the impurity of the raw materials in the As2S5 glass. Based on the nonlinear refractive index of ~3 × 10−18 m2 W−1 , the calculated nonlinear coefficients of the tapered and untapered segments at 1900 nm were ~6463 and ~2584 km−1 W−1, respectively.
3. Experimental results
The experimental setup for observing SSFS and THG in the As2S5 MOF is shown in Fig. 2. A laser pulse with ~200 fs duration and repetition rate of ~80 MHz generated from an optical parametric oscillator (OPO) was used as the pump light. The pulse was coupled into the core of a 7.8 cm-long tapered As2S5 MOF by a lens with a focal length of ~4.5 mm and numerical aperture (NA) of ~0.47. The output signal from the optical fiber was butt-coupled into a 0.3 m long large-mode-area (LMA) fluoride (ZBLAN) fiber with the core diameter of ~105 μm and transmission window from 0.4 to 5 μm. The LMA ZBLAN fiber was connected to two optical spectrum analyzers (OSAs; Yokogawa AQ6375 and Yokogawa AQ6373) and a Fourier-transform infrared (FT-IR) spectrometer.
SSFS spectra for a pump wavelength of ~1900 nm with different pump average powers are shown in Fig. 3. As the power increases and exceeds the Raman threshold, the soliton pulse changes its central frequency due to SRS. By varying the pump average power from 220 to 340 mW (corresponding to the peak powers ~2.75, 3.5, 3.9 and 4.25 kW), we observed a fundamental soliton with carrier frequency shift Δν from 22.1 to 42.5 THz (~306 to 700 nm in wavelength). At the maximum power of ~340 mW, the ratio of the pump power and the fundamental soliton power was 5:1. Because the amount of solitons depends on the pump peak power and the fiber length, the increasing pump power will exceed what is needed to form the fundamental soliton, explaining the appearance of high-order soliton fission with several peaks. The spectral evolution for the redshift was mainly due to the soliton dynamics, and the blueshift was mainly due to self-phase modulation (SPM) and the dispersive wave (DW) emitted by the solitons. In Fig. 4, the soliton central wavelength shift is shown as a function of the pump average power (220 mW, 280 mW, 310 mW and 340 mW). At high pump average powers, SSFS was so large that the loss of pulse energy to excited phonons (owing to the Raman nature of the process) weakens SSFS . The output signal did not extend above 2.80 μm, and one of the reasons is that the As2S5 glass core has a strong absorption round 2.80 μm due to OH.
In this experiment, solitons were only generated from the 1.5 cm-long tapered segment, which is much shorter than previous reports [11, 12], and not from the untapered input and output ends. This effect is because the pump wavelength was set in the normal dispersion regime of untapered segment and soliton cannot be formed in this segment. On the other hand, when the pump wavelength was set in the anomalous dispersion regime of the tapered segment, the excitation of solitons was possible. In order to confirm that SSFS was generated from the tapered segment, and not the untapered segments, a 7.8 cm-long untapered As2S5 MOF with λZDW = 2.02 μm was also measured at the pump average power of ~340 mW while other parameters stayed unchanged. The measured spectrum is shown in Fig. 5. The spectrum was broadened mainly by SPM in the normal dispersion regime, and no solitons were observed.
Solitons generated from the tapered segment will transmit naturally in the untapered output end. However, if the soliton central wavelength is shorter than the ZDW of the untapered segment (λZDW = 2.02 μm), a soliton pulse will be transmitted in the normal dispersion regime of the untapered segment. As a result it would be impossible to observe the solitons from the untapered output end.
Because the As2S5 MOF has a large core-cladding refractive index difference, the PM condition between the fundamental and TH waves is easily satisfied . THG spectrum for the untapered segment was measured with OSA at a pump wavelength of ~1900 nm with pump average powers of ~280, 310, and 340 mW, as shown in Fig. 6(a). The central wavelength was ~632 nm and it remained stable in spite of increasing pump average power (or, variation of the Raman-shifted soliton position). This observation excluded the possibility that the spectrum was composed of DWs that emitted from solitons. Similar results can also be obtained from the 7.8 cm long untapered As2S5 MOF with the same pump wavelength. The pump light loss in the tapered segment was larger than that in the untapered segment, due to its smaller diameter. Meanwhile, solitons and DWs shared most of the pump power, so the weak pump light in the tapered segment cannot cause THG. Although the pump power in the untapered segment was strong enough to cause THG, the TH signal was still weak and no obvious TH wave was observed, even at a power of 280 mW. Several explanations are possible for this effect: fiber loss from glass impurities, unsatisfactory PM in the untapered segment (Δβ = 0.32 μm−1), and a low launched peak power compared with the damage threshold of the As2S5 MOF. These influences can be reduced with the following changes: reducing the fiber loss by improving the As2S5 glass purity, satisfying the PM condition by changing the core size of the fiber, and increasing the launched peak power within the damage threshold of the fiber by improving the coupling efficiency or using a high-power laser.
Figure 6(b) shows the red light of a TH wave. The red light was obvious at the untapered input and output ends but not at the tapered segment. This clearly shows that THG occurred at the untapered segments, and not tapered segment.
In summary, a four-hole As2S5 MOF was tapered in order to offer an ideal environment for SSFS. In the lab, we observed a fundamental soliton with carrier frequency shift Δν from 22.1 to 42.5 THz in a relatively short tapered segment (1.5 cm in length) while the pump average power increasing from 220 to 340 mW. Meanwhile, THG with a central wavelength of ~632 nm was observed from the untapered input (1.8 cm) and output end (4.5 cm).
This work is supported by MEXT, the Support Program for Forming Strategic Research Infrastructure (2011-2015).
References and links
3. J. H. Lee, J. van Howe, X. Liu, and C. Xu, “Soliton Self-Frequency Shift: Experimental Demonstrations and Applications,” IEEE J. Sel. Top. Quantum Electron. 14(3), 713–723 (2008). [CrossRef] [PubMed]
4. L. Liu, X. Meng, F. Yin, M. Liao, D. Zhao, G. Qin, Y. Ohishi, and W. Qin, “Soliton self-frequency shift controlled by a weak seed laser in tellurite photonic crystal fibers,” Opt. Lett. 38(15), 2851–2854 (2013). [CrossRef] [PubMed]
7. M. E. Masip, A. A. Rieznik, P. G. König, D. F. Grosz, A. V. Bragas, and O. E. Martinez, “Femtosecond soliton source with fast and broad spectral tunability,” Opt. Lett. 34(6), 842–844 (2009). [CrossRef] [PubMed]
8. B. Barviau, O. Vanvincq, A. Mussot, Y. Quiquempois, G. Mélin, and A. Kudlinski, “Enhanced soliton self-frequency shift and CW supercontinuum generation in GeO2-doped core photonic crystal fibers,” J. Opt. Soc. Am. B 28(5), 1152–1160 (2011). [CrossRef]
9. H. Lim, J. Buckley, A. Chong, and F. W. Wise, “Fibre-based source of femtosecond pulses tunable from 1.0 to 1.3 μm,” Electron. Lett. 40(24), 1523–1525 (2004). [CrossRef]
10. A. M. Al-kadry and M. Rochette, “Mid-infrared sources based on the soliton self-frequency shift,” J. Opt. Soc. Am. B 29(6), 1347–1355 (2012). [CrossRef]
11. X. Liu, C. Xu, W. H. Knox, J. K. Chandalia, B. J. Eggleton, S. G. Kosinski, and R. S. Windeler, “Soliton self-frequency shift in a short tapered air-silica microstructure fiber,” Opt. Lett. 26(6), 358–360 (2001). [CrossRef] [PubMed]
12. A. C. Judge, O. Bang, B. J. Eggleton, B. T. Kuhlmey, E. C. Mägi, R. Pant, and C. M. de Sterke, “Optimization of the soliton self-frequency shift in a tapered photonic crystal fiber,” J. Opt. Soc. Am. B 26(11), 2064–2071 (2009). [CrossRef]
13. I. G. Cormack, D. T. Reid, W. J. Wadsworth, J. C. Knight, and P. S. J. Russell, “Observation of soliton self-frequency shift in photonic crystal fiber,” Electron. Lett. 38(4), 167–169 (2002). [CrossRef]
14. D. G. Ouzounov, F. R. Ahmad, D. Müller, N. Venkataraman, M. T. Gallagher, M. G. Thomas, J. Silcox, K. W. Koch, and A. L. Gaeta, “Generation of megawatt optical solitons in hollow-core photonic band-gap fibers,” Science 301(5640), 1702–1704 (2003). [CrossRef] [PubMed]
15. J. van Howe, J. H. Lee, S. Zhou, F. Wise, C. Xu, S. Ramachandran, S. Ghalmi, and M. F. Yan, “Demonstration of soliton self-frequency shift below 1300 nm in higher-order mode, solid silica-based fiber,” Opt. Lett. 32(4), 340–342 (2007). [CrossRef] [PubMed]
16. X. Yan, G. Qin, M. Liao, T. Suzuki, and Y. Ohishi, “Transient Raman response effects on the soliton self-frequency shift in tellurite microstructured optical fiber,” J. Opt. Soc. Am. B 28(8), 1831–1836 (2011). [CrossRef]
17. J. Fatome, B. Kibler, M. Amraoui, J. C. Jules, G. Gadret, F. Desevedavy, and F. Smektala, “Mid-infrared extension of supercontinuum in a chalcogenide suspended core fibre through soliton gas pumping,” Electron. Lett. 47(6), 398–400 (2011). [CrossRef]
19. T. Lee, Y. Jung, C. A. Codemard, M. Ding, N. G. R. Broderick, and G. Brambilla, “Broadband third harmonic generation in tapered silica fibres,” Opt. Express 20(8), 8503–8511 (2012). [CrossRef] [PubMed]
21. Y. Tamaki, K. Midorikawa, and M. Obara, “Phase-matched third-harmonic generation by nonlinear phase shift in a hollow fiber,” Appl. Phys. B 67(1), 59–63 (1998). [CrossRef]
22. G. Qin, M. Liao, C. Chaudhari, X. Yan, C. Kito, T. Suzuki, and Y. Ohishi, “Second and third harmonics and flattened supercontinuum generation in tellurite microstructured fibers,” Opt. Lett. 35(1), 58–60 (2010). [CrossRef] [PubMed]
23. T. Kohoutek, S. Mizuno, T. Suzuki, Y. Ohishi, M. Matsumoto, and T. Misumi, “Third-harmonic generation measurement of nonlinear optical susceptibility χ (3) of Ge-Ga-Sb-S chalcogenide glasses proposed for highly nonlinear photonic fibers,” J. Opt. Soc. Am. B 28(2), 298–305 (2011). [CrossRef]
24. T. L. Cheng, Z. C. Duan, M. S. Liao, W. Q. Gao, D. H. Deng, T. Suzuki, and Y. Ohishi, “A simple all-solid tellurite microstructured optical fiber,” Opt. Express 21(3), 3318–3323 (2013). [CrossRef] [PubMed]
25. A. C. Judge, S. A. Dekker, R. Pant, C. M. de Sterke, and B. J. Eggleton, “Soliton self-frequency shift performance in As2S3 waveguides,” Opt. Express 18(14), 14960–14968 (2010). [CrossRef] [PubMed]
26. M. El-Amraoui, J. Fatome, J. C. Jules, B. Kibler, G. Gadret, C. Fortier, F. Smektala, I. Skripatchev, C. F. Polacchini, Y. Messaddeq, J. Troles, L. Brilland, M. Szpulak, and G. Renversez, “Strong infrared spectral broadening in low-loss As-S chalcogenide suspended core microstructured optical fibers,” Opt. Express 18(5), 4547–4556 (2010). [CrossRef] [PubMed]
27. A. Efimov, A. J. Taylor, F. G. Omenetto, J. C. Knight, W. J. Wadsworth, and P. St. J. Russell, “Phase-matched third harmonic generation in microstructured fibers,” Opt. Express 11(20), 2567–2576 (2003). [CrossRef] [PubMed]