Abstract
We report the fabrication of a novel high nonlinear fiber made of Ge–Sb–Se chalcogenide glasses with high numerical aperture (), where the core and the cladding glasses consist of and (mol. %), respectively. The nonlinear refractive index () of the core glass is at 1.55 μm, and its laser-induced damage threshold under irradiation of 3.0 μm fs laser is approximately . By pumping a 20-cm-long fiber with a core diameter of 23 μm using 150 fs pulses at 6.0 μm, supercontinuum spanning from to was generated.
© 2016 Optical Society of America
Mid-infrared (MIR) supercontinuum (SC) sources are characterized by an attractive combination of high brightness and broad bandwidth that makes them ideal for spectroscopy, species identification, optical imaging, and laser frequency metrology [1,2]. Of the various media including fluoride, tellurite, and chalcogenide glass (ChG) fibers that have been used to generate MIR SC, both fluoride and tellurite are transparent in a wavelength range of fewer than 5 μm, and thus may not be used in a longer wavelength. By contrast, ChG can transmit to even longer wavelengths and also offer higher nonlinearities than either fluoride or telluride, making it promising for MIR SC generation [3,4].
Recently, ultrabroadband MIR SC generations have been achieved in ChG fibers pumped by a tunable optical parametric amplifier (OPA) [5–11]. For example, an SC spectrum spanning from 1.8 to 10 μm with pulses at 4.0 μm has been generated from an 11-cm-long step-index Ge–As–Se–ChG fiber [10], and an SC generation up to 13.3 μm has been generated from an 85-mm-long fiber with an core surrounded by a cladding pumped with MW pulses at 6.3 μm [11]. However, most ChG fiber media for SC generation, such as and , contain the toxic element arsenic. Thus, security issues may occur during glass preparation, fiber drawing, and performance testing [12–14]. Especially when the toxic materials are exposed to a laser power that is beyond the damage threshold of the materials, they could be burnt or evaporated. Obviously, these evaporated gases are extremely harmful to health.
Widely used or glass has a relatively low glass transition temperature (Tg) at 185°C and 178°C, respectively [15]. By contrast, Ge–Sb–Se glasses exhibit higher thermal and mechanical durability and have been demonstrated to be suitable for molded infrared-transmitting lenses. The replacement of highly toxic arsenic with antimony makes the glasses more environmentally favorable. Furthermore, optical nonlinearity of Ge–Sb–Se glass is greater than that of Ge–As–Se because of the replacement of As by more metallic Sb [16]; thus, the ultrabroad SC spectrum is expected to be generated in Ge–Sb–Se glass fibers.
In this study, we demonstrated a proof-of-concept step-index optical fiber fabricated by a rod-in-tube technique. Specific care was paid to the design and fabrication of two glasses with excellent thermomechanical properties that are suitable for fiber drawing. By optimizing the compositions of the Ge–Sb–Se glass system, two types of glasses, namely, and , were determined to be used as core and cladding glasses, respectively. A multimode fiber configuration with high numerical aperture (NA, ) based on Ge–Sb–Se glass was fabricated. We further investigated the nonlinearities and laser damage characteristics of the glass in comparison with those of glass. Finally, we achieved an ultrabroad SC spectrum of 1.8–14 μm based on Ge–Sb–Se fiber excited by a 6.0 μm fs laser source.
ChGs were prepared from purified raw materials by the melt-quenching method [17]. The thermal characteristics of the core and cladding glasses were investigated by differential scanning calorimetry (DSC). Tg and Tx (crystallization temperature) of (core) were 236°C and 350°C, respectively, while those for (cladding) were 227°C and 375°C, respectively. Both values of (Tx–Tg) of the two glasses were more than 100°C, implying the excellent thermal stability of the glasses against crystallization during the fiber drawing [18]. The coefficients of thermal expansion for and , determined by thermal mechanical analysis, were 162 and 149 () from room temperature to 200°C, respectively. A slightly higher thermal expansion coefficient of the core glass was beneficial for a more compact core cladding interface, although overlarge expansion coefficient could also result in built-in strain or laser-induced strain upon irradiation [19].
The multimode step-index ChG fiber was fabricated by the standard rod-in-tube technique [20]. No crystallization or chemical reaction was found during the fiber-drawing process. Finally, a fiber with a core diameter of 23 μm and cladding diameter of 307 μm was used for the optical measurement and SC generation.
The loss of the fiber was measured with the cutback method. Figure 1 shows the attenuation curve of the core/cladding multimode step-index fiber, and the minimum of losses was near 5 dB/m at 6.0 μm. Absorption bands at 2.92, 3.53, and 12.8 μm were assigned to O-H, Se-H, and Ge-O bonds, respectively. The losses may be caused by the intrinsic absorptions of the materials, the scattering induced by microinhomogeneity in the glass, and the imperfect interfaces between the core and cladding. These could be further reduced through improved purification, composition optimization, and better control of the parameters in the fiber drawing [18].
The linear refractive indices of the core and cladding glasses as a function of wavelength were measured by ellipsometry; the results are shown in Fig. 2(a). The NA of the fiber was calculated to be , which can significantly increase coupling efficiency, reduce confinement loss, and enable the use of a simple and compact ultrafast laser to pump the fiber to generate broadband SC [20].
The normalized mode-field intensity and dispersion of the step-index GhG fiber were simulated using commercial software (Lumerical MODE Solution) with full-vectorial mode-solver technology. Figure 2(b) shows the obtained chromatic dispersion curve, where the zero-dispersion wavelength (ZDW) of the fiber is .
We investigated the optical nonlinearity of the fiber material using the -scan method. The nonlinearity was excited using 150 fs laser pulses at 1550 nm. The measurements were repeated 10 times at different places on the sample surface to minimize experimental error [21]. The results showed that the value of the glass is , or approximately 2 times higher than that of glass [22–24]. The high of the glass is associated with a high Sb component, and the strong ionic characteristic of Sb element is believed to induce a larger optical nonlinearity [25].
We assumed that the fiber length and laser power threshold could be reduced in a material with a high optical nonlinearity for the SC generation. However, nonlinear damage can also be induced simultaneously. We therefore investigated the laser-induced damage threshold (LIDT) of and glass. Two pieces of the glasses, each 2 mm thick, were thoroughly polished because surface quality of the glass has an effect on LIDT measurement [26]. The root mean square (RMS) surface roughness and the peak-to-valley surface flatness of glass samples measured by an interferometer are approximately and , respectively. The glasses were shined with a fs laser from OPA with a repetition rate of 1 kHz, a center wavelength of 3000 nm (approximate photoenergy of 0.41 eV), and a pulse width of 150 fs. During the measurements, the laser was focused on a spot size of 250 μm on the front surface of the sample by a lens with a focal length of 50 mm. The results indicated that visible damage arises at a beam intensity of for the glass bulk, and at for glass. The latter value is similar to that reported in the literature [27]. The surface damage morphologies of two samples after irradiation with 30 mW () and 20 s duration are shown in Fig. 3. Both damaged spots typically appeared in the form of a crater on the surface. However, glass presents an image with deep craters produced on the surface, which is in sharp contrast to shallow melting, as observed in the glass, indicating that has a higher LIDT.
We further measured SC generation based on the Ge–Sb–Se fiber. A tunable OPA system (Mirra 900 + Legend Elite + OperA Solo) was used as the exciting source. The pump pulse had a duration of (full width at half-maximum) and a repetition rate of 1 kHz. The beam from the OPA was coupled via a calcium fluoride lens with a focal length of 75 mm into the fiber. The SC output from the fiber was collected by the input slit of a monochromator. A liquid nitrogen-cooled HgCdTe (MCT) detector with a wavelength range of 1–16 μm was used to measure the output of the monochromator.
Figure 4 shows the evolution of the SC spectra excited at 6.0 μm with increasing pump power. To analyze these results, we compared the measured spectra with numerical simulations. For the current fiber, the ZDW was as shown in Fig. 2(b), and the group velocity dispersion (GVD) parameter was at a pump wavelength of 6.0 μm. Based on the calculated dispersion and nonlinearity, the generalized nonlinear Schrödinger equation was solved using the split-step Fourier method up to 20th-order dispersion [10,28]. Thus, as the coupled peak power increased, the spectral broadening was mainly driven by the self-phase modulation with the Raman shift extending to the long wavelength edge. Experimentally, 150 fs pulses at 6.0 μm emitted from an OPA system were injected into a 20-cm-long fiber with a core diameter of 23 μm. MIR SC spanning from to was obtained. The reason for the increased flatness of the simulation results is that part of the light was unavoidably coupled to the cladding and higher-order core modes. The spectral dips at and , as indicated by the arrows in Fig. 4(a), might be caused by switching the long-pass filters, since four long-pass filters (1.9, 3.6, 7.0, and 10.0 μm) were applied as order-sorting filters to eliminate a high-order signal [9]. The spectral peaks at and are caused by the dispersive wave of higher-order modes in the long wavelength, since the fiber we used is multimoded where the higher-order mode should start to function in the SC generation. Similar peaks have been reported in Ref. [11]. The spectral peaks at 3.5 μm might be caused by the dispersive wave of the short wavelength and the peak at 5 μm is due to the fission of higher-order solitons, which lead to the soliton self-frequency shift (SSFS) by pumping at 6.0 μm [29].
In conclusion, we report the fabrication of a high NA, As-free step-index GhG fiber with the core and cladding made of and glasses, respectively. These two glasses are thermally compatible and can be drawn into step-index fibers with a very high NA through the rod-in-tube drawing technique. By pumping a 20-cm-long fiber with a core diameter of 23 μm with 150 fs pulses at 6.0 μm, SC spanning from to can be achieved, indicating the significant potential of the Ge–Sb–Se fibers as nonlinear media for a practical, bright MIR SC source.
Funding
National Natural Science Foundation of China (NSFC) (61377099, 61435009); Ningbo University.
Acknowledgment
We thank Dr. Yinsheng Xu, Dr. Bin Zhang, and Dr. Zhiyong Yang for the useful discussion.
REFERENCES
1. P. Falk, M. H. Frosz, and O. Bang, Opt. Express 13, 7535 (2005). [CrossRef]
2. I. Savelii, O. Mouawad, J. Fatome, B. Kibler, F. Désévédavy, G. Gadret, J. CJules, P. Y. Bony, H. Kawashima, W. Gao, T. Kohoutek, T. Suzuki, Y. Ohishi, and F. Smektala, Opt. Express 20, 27083 (2012). [CrossRef]
3. L. Petit, N. Carlie, K. Richardson, A. Humeau, S. Cherukulappurath, and G. Boudebs, Opt. Lett. 31, 1495 (2006). [CrossRef]
4. J. Sanghera, C. Florea, L. Shaw, P. Pureza, V. Nguyen, M. Bashkansky, Z. Dutton, and I. Aggarwal, J. Non-Cryst. Solids 354, 462 (2008). [CrossRef]
5. O. Mouawad, J. Picot-Clémente, F. Amrani, C. Strutynski, J. Fatome, B. Kibler, F. Désévédavy, G. Gadret, J. C. Jules, and D. Deng, Opt. Lett. 39, 2684 (2014). [CrossRef]
6. D. D. Hudson, M. Baudisch, D. Werdehausen, B. J. Eggleton, and J. Biegert, Opt. Lett. 39, 5752 (2014). [CrossRef]
7. I. Kubat, C. S. Agger, U. Møller, A. B. Seddon, Z. Tang, S. Sujecki, T. M. Benson, D. Furniss, S. Lamrini, and K. Scholle, Opt. Express 22, 19169 (2014). [CrossRef]
8. F. Theberge, N. Thire, J. F. Daigle, P. Mathieu, B. E. Schmidt, Y. Messaddeq, R. Vallee, and F. Legare, Opt. Lett. 39, 6474 (2014). [CrossRef]
9. T. L. Cheng, K. Nagasaka, T. H. Tuan, X. J. Xue, M. Matsumoto, H. Tezuka, T. Suzuki, and Y. Ohishi, Opt. Lett. 41, 2117 (2016). [CrossRef]
10. Y. Yu, B. Zhang, X. Gai, C. Zhai, S. Qi, W. Guo, Z. Yang, R. Wang, D. Y. Choi, S. Madden, and B. Luther-Davies, Opt. Lett. 40, 1081 (2015). [CrossRef]
11. C. R. Petersen, U. Møller, I. Kubat, B. Zhou, S. Dupont, J. Ramsay, T. Benson, S. Sujecki, N. Abdel-Moneim, Z. Tang, D. Furniss, A. Seddon, and O. Bang, Nat. Photonics 8, 830 (2014). [CrossRef]
12. T. Kohoutek, X. Yan, T. W. Shiosaka, S. N. Yannopoulos, A. Chrissanthopoulos, T. Suzuki, and Y. Ohishi, J. Opt. Soc. Am. B. 28, 2284 (2011). [CrossRef]
13. F. W. Glaze, D. H. Blackburn, J. S. Osmalov, D. Hubbard, and M. H. Black, J. Res. Natl. Bur. Stand. 59, 83 (1957). [CrossRef]
14. R. J. Curry, S. W. Birtwell, A. K. Mairaj, X. Feng, and D. W. Hewak, J. Non-Cryst. Solids 351, 477 (2004). [CrossRef]
15. G. Snopatin, V. Shiryaev, V. Plotnichenko, E. Dianov, and M. Churbanov, Inorg. Mater. 45, 1439 (2009).
16. W. Wei, R. Wang, X. Shen, L. Fang, and B. L. Davies, J. Phys. Chem. C 117, 16571 (2013).
17. B. Zhang, W. Guo, Y. Yu, C. Zhai, S. Qi, A. Yang, L. Li, Z. Yang, R. Wang, and D. Tang, J. Am. Ceram. Soc. 98, 1389 (2015). [CrossRef]
18. Z. Yang, T. Luo, S. Jiang, J. Geng, and P. Lucas, Opt. Lett. 35, 3360 (2010). [CrossRef]
19. J. H. Butterworth, D. Jayasuriya, Q. Q. Li, D. Furniss, N. A. Moneim, E. Barney, S. Sujecki, T. M. Benson, J. S. Sanghera, and A. B. Seddon, Proc. SPIE 8938, 89380W (2014). [CrossRef]
20. T. Cheng, Y. Kanou, D. Deng, X. Xue, M. Matsumoto, T. Misumi, T. Suzuki, and Y. Ohishi, Opt. Express 22, 13322 (2014). [CrossRef]
21. S. Dai, F. Chen, Y. Xu, Z. Xu, X. Shen, T. Xu, R. Wang, and W. Ji, Opt. Express 23, 1300 (2015). [CrossRef]
22. M. Olivier, J. C. Tchahame, P. Němec, M. Chauvet, V. Besse, C. Cassagne, G. Boudebs, G. Renversez, R. Boidin, E. Baudet, and V. Nazabal, Opt. Mater. Express 4, 525 (2014).
23. T. Wang, X. Gai, W. Wei, R. Wang, Z. Yang, X. Shen, S. Madden, and B. Luther-Davie, Opt. Mater. Express 4, 1011 (2014).
24. E. R. Barney, N. S. Abdel-Moneim, J. J. Towey, J. Titman, and A. B. Seddona, Phys. Chem. Chem. Phys. 17, 6314 (2015). [CrossRef]
25. B. Qiao, S. Dai, Y. Xu, P. Zhang, X. Shen, T. Xu, Q. Nie, W. Ji, and F. Chen, Opt. Mater. Express 5, 2359 (2015).
26. C. B. Schaffer, A. Brodeur, and E. Mazur, Meas. Sci. Technol. 12, 1784 (2001). [CrossRef]
27. E. Romanova, A. Konyukhov, S. Muraviov, and A. Andrianov, in 12th International Conference on Transparent Optical Networks, Munich, Germany (2010).
28. S. Shabahang, M. P. Marquez, G. Tao, M. U. Piracha, D. Nguyen, P. J. Delfyett, and A. F. Abouraddy, Opt. Lett. 37, 4639 (2012). [CrossRef]
29. K. Sakamaki, M. Nakao, M. Naganuma, and M. Izutsu, IEEE J. Sel. Top. Quantum Electron. 10, 876 (2004). [CrossRef]