Low threshold mid-infrared supercontinuum generation in short fluoride-chalcogenide multimaterial fibers

Xia Li; Wei Chen; Tianfeng Xue; Juanjuan Gao; Weiqing Gao; Lili Hu; Meisong Liao

doi:10.1364/OE.22.024179

1. Introduction

Supercontinuum generation (SCG) has been intensively studied [1–6]. As far as the spectral region is concerned, many achievements in SCG in silica fibers covering the visible and near-infrared wavelength regions have been accomplished and SCG in the mid-infrared wavelength region continues to attract significant research attention. Many molecules such as H₂O, O₃, CO, NO, CH₄ and N₂O have strong characteristic absorption through vibrational transitions in the mid-infrared region. Thus, the molecular spectroscopy requires a mid-infrared source [7]. Mid-infrared Doppler lidar has already been utilized to provide real-time displays of wind and aerosol backscattering data [8]. The 3-5 μm atmosphere window allows trace gas sensing for security and industrial applications [9].

Intrinsic material loss caused by the multiphonon absorption edge limits the SCG at longer wavelengths in silica-based fibers [10]. To enlarge the application in the mid-infrared region, soft glasses such as fluoride, tellurite and chalcogenide glasses are used to fabricate fibers [11–14]. Fluoride glass fibers, with a wide transparency range up to ~5 μm, are most widely used in mid-infrared region [15–17]. SCG extended to 4.5 μm has been realized from fluoride fibers [18, 19]. The average power of SCG approached 1.3 W in [20]. On account of low level nonlinear effects (Table 1), fluoride fibers of several meters or even longer are necessary in these experiments. Tens of thousands of watts pump peak power with multi-stage amplifiers are required to extend the spectra to the mid-infrared region [21–23]. As the propagation distance increased, coherence degradation of SCG increases and may also cause problems [24]. To utilize SCG in spectroscopy and imaging, high coherence of the spectrum is also required [25].

Table 1. Thermal and optical properties of different glasses

View Table

The requirement for cheap and compact supercontinuum mid-infrared sources leaves the chalcogenide glass more opportunities, which has a high nonlinearity and a broad transparency range in mid-infrared region. To get a wide SCG, the nonlinear fiber is preferred to be pumped in the anomalous group velocity dispersion (GVD) region close to the zero dispersion wavelength (ZDW). However, the chalcogenide glasses always suffer from large normal GVD and the ZDWs are ~5μm or above. Therefore, strong waveguide dispersion is necessary to compensate for material dispersion and shorten the ZDW to match those commercially available laser sources, the center wavelengths of which are around 1.5 or 2 μm [26, 27]. Typically, fiber tapers, microstructured optical fibers (MOF) and planer waveguides are applied [28–30]. A spectrum from 2.4 to 4.6 μm was acquired from a 19 mm long tapered As₂S₃ glass fiber [31]. The tapering process included resistive heating and subsequent pulling. A suspended-core chalcogenide fiber has been investigated for SCG and a broadband spectrum nearly reaching 5 μm was obtained by 1.32 kW pumping [32]. The ~3 μm core structures were suspended by three delicate glass struts bridging 30-50 μm air regions to provide proper wave guidance. The researchers can also get broad band SCG from compact devices by tailoring the dispersion in planer waveguides [30, 33, 34]. However, there are some problems for the above-mentioned structures. For example, the tapering process requires accurate coordination for the heating and dragging units to acquire an expected tapered waist to get an appropriate dispersion for SCG. The tapered waist is very fragile and must be handled cautiously. During drawing process, the MOF with air holes may suffer from some thermo physical effects such as non-uniform heat dissipation, change of pressure and fluctuation of surface tension, all of which add complexity in maintaining the air-hole structure [35]. Besides, it remains an issue for conservation and maintenance of the MOF. In the waveguide system, complex OPO system is required to generate pump pulse above 3 μm [34]. In some other situations in planer waveguides, it is difficult for the spectra to extend to beyond 2 μm by using 1550 nm pump [30, 33]

Recently, multimaterial fibers (MMFs) combined with polymer, semiconductor and different glasses have been studied to get different optical, electronic and thermomechanical properties [36–38]. In [39, 40], chalcogenide glass has been composed with different glasses. In this paper, we investigate MMFs composed of chalcogenide glass and fluoride glass. Ge₂₀Sb₁₅Se₆₅ (GSS) and ZnF₄-BaF₂-LaF₃-AlF₂-NaF (ZBLAN) are selected. GSS is used to generate large nonlinear effect. In order to adjust the dispersion profile without employing fiber tapers and air hole MOFs, fluoride and tellurite glasses are the candidates to compose the cladding of the chalcogenide core MMF. The refraction index of tellurite glass is ~2.0 and the ZDW of bulk tellurite glass is about 2 μm. Consequently, the ZDW of MMFs composed of GSS and tellurite glass should be larger than 2 μm. The second candidate fluoride glass ZBLAN has a lower refraction index, as shown in [41]. Dispersion can be more flexibly controlled with higher refraction contrast. The glass transition temperatures of GSS and ZBLAN are closely matched (Table 1), which imply similar fiber-drawing temperatures. The thermal expansion coefficient is ~20 × 10⁻⁶/K from room temperature to 180 °C for an analogous chalcogenide composite (Ge_22.5Sb_7.5Se₇₀) and ~18 × 10⁻⁶/K for ZBLAN at the same temperature region [42, 43]. Considering the very short fiber length in the application for SCG, the difference is not so large. Therefore the tension due to the mismatch in thermal expansion coefficient is slight. It is compatible for the core and clad not to crack or shatter during the cooling period in fiber-drawing process. Minimizing the drawing temperature with inert gas atmosphere maintained is preferred during the drawing process to improve the interface quality and prevent the glass from crystallization. Coating can be applied to enhance the fiber strength. With these material properties and fiber-drawing strategies, the feasibility of fiber fabrication is quite high. We investigate into step-index MMFs composed of GSS and ZBLAN firstly. With a much lower refraction index ZBLAN cladding, ZDW of the step-index MMF is able to be shifted to ~2 μm with all solid configurations. Since supercontinuum generated in all normal dispersion fiber usually features better coherence property, we propose the all-solid microstructured multimaterial fibers (MS-MMFs). All normal dispersion can be obtained in the MS-MMFs, which is difficult to realize in the step-index MMFs. Dispersion and nonlinear properties are considered in the fiber design and the SCG capacity is simulated. A 175 W peak power pulse centered at 1.95 μm is applied to generate a high coherence supercontinuum extending from 1250 nm to 2750 nm within a 1 cm-long MS-MMF. The spectrum can also be extended to 5 μm as the launched peak power increased to 1400W. Overall, the proposed MMFs can benefit from the high refraction index contrast and high nonlinear coefficient of the material. This design makes it possible to adjust the dispersion of the chalcogenide glass expediently without rely on air holes and fiber tapers. The SCG can extend to 5 μm by using readily accessible 2 μm fiber laser in a short MMF. The fabricating process of this kind of all-solid fiber is not complicated with conventional fiber drawing method applied. The all-solid scheme provides easier post-processing such as coupling and maintenance.

2. Numerical simulation

The linear refractive index n for both glasses is given by the Sellmeier dispersion equation:

n (λ) = \sqrt[2]{1 + \frac{b_{1} \times λ^{2}}{λ^{2} - c_{1}} + \frac{b_{2} \times λ^{2}}{λ^{2} - c_{2}} + \frac{b_{3} \times λ^{2}}{λ^{2} - c_{3}}}

where λ is the wavelength and the fitting coefficients b_1,2,3 and c_1,2,3 are taken from [47, 48].

We rely on the full-vector finite element method (FEM) with perfect match layer boundary condition employed to calculate the effective index that yields the propagation constant β. Thus, Taylor series expansion is applied to β, which can be used to obtain the group velocity and chromatic dispersion in the fiber:

D (λ) = - \frac{2 π c}{λ^{2}} β_{2} = - \frac{λ}{c} \frac{d^{2} n_{e f f (λ)}}{d λ^{2}}

c is light speed, β₂ is the second order Taylor expansion coefficient for propagation constant, n_eff represents for effective index.

To simulate pulse propagation dynamics in the nonlinear fiber, the split-step Fourier method is applied to numerically solve the generalized nonlinear Schrodinger equation (GNLSE) considering dispersion and nonlinear terms, including self-phase modulation (SPM), self-steepening and Raman response:

\begin{array}{l} \frac{\partial A}{\partial z} = \frac{α}{2} A - \sum_{n \geq 2} β_{n} \frac{i^{n - 1}}{n!} \frac{\partial^{n}}{\partial t^{n}} A + i γ (1 - f_{R}) ({| A |}^{2} A - \frac{2 i}{ω_{0}} \frac{\partial}{\partial t} ({| A |}^{2} A)) \\ {^{^{}}}^{}_{}^{}^{} + i γ f_{R} (1 + \frac{i}{ω_{0}} \frac{\partial}{\partial t}) (A \int_{0}^{\infty} h_{R} (τ) {| A (t - τ) |}^{2} d τ) \end{array}

A is slowly varying envelope approximation for the light field. z is transmission distance. α is the transmission loss which is regarded as zero in the simulation. The term β_n denotes GVD. t is time. f_R represents the Raman contribution. The right hand side of the equation represents the gain, dispersion, SPM and self-steeping, Raman response respectively. Nonlinear refraction index n₂ = 15 × 10⁻¹⁸ m²/W for the chalcogenide glass [49] is taken as an approximate value to determine the nonlinear index of the fiber by using:

γ = \frac{2 π}{λ} \frac{n_{2}}{A_{e f f}} = \frac{2 π}{λ} \frac{{\iint n_{2} (x, y) | F (x, y) |}^{4} d x d y}{{({\iint n_{2} (x, y) | F (x, y) |}^{2} d x d y)}^{2}}

F(x, y) is the modal distribution, which is reckoned based on FEM. Raman response function h_R(t) is commonly expressed as the following analytical form:

h_{R} (t) = \frac{τ_{1}^{2} + τ_{2}^{2}}{τ_{1} τ_{2}^{2}} \exp (- \frac{t}{τ_{2}}) \sin (\frac{t}{τ_{1}})

The parameter τ₁ relates to the phonon oscillation frequency while τ₂ defines the characteristic damping time of the network of vibrating atoms. These two parameters should be optimized to yield the best fit of h_R(t). Meanwhile, Raman gain

g (ω) = (2 ω_{p} / c) n_{2} f_{R} I m [H_{R} (ω)]

, should be measured to get H_R(w), which is determined by applying Kramers-Kronig relations on Im[H_R(w)], as the Fourier transformation of h_R(t). Restricted by the experiment conditions, the parameter used in our simulation is taken from [50, 51], as τ₁ = 23.1 fs, τ₂ = 195 fs and f_R = 0.1.

The degree of coherence is studied by calculating the spectra with random noise seeds with 1% of the pulse intensity which is defined as [24]:

| g_{12}^{(1)} (λ, t_{1} - t_{2}) | = | \frac{〈 E_{1}^{*} (λ, t_{1}) E_{2} (λ, t_{2}) 〉}{{[〈 {| E_{1} (λ, t_{1}) |}^{2} 〉 〈 {| E_{2} (λ, t_{2}) |}^{2} 〉]}^{1 / 2}} |

where E₁ and E₂ are continuum spectra generated with input quantum noise in the simulation. The angle brackets denote the ensemble average of over 100 independently generated pairs of SCG. The injected pump pulse is set as a Sech² profile with an average power less than 10 mW and a duration of femtosecond magnitude at 1.97 μm, which can be achieved with a commercially available laser system. Taylor series expansion coefficients of up to 12 orders are included when modeling the dispersion property to achieve precise molding. The theoretical simulation has been compared with [2] to ensure our code works well.

3. SCG in fluoride-chalcogenide step-index MMFs

The material dispersions of GSS and ZBLAN are shown in Fig. 1(a). The ZDW of GSS is located at above 5 μm, far from the conventional pulse pump source while that of ZBLAN is 1.65 μm. Step-index MMFs with chalcogenide core and ZBLAN cladding are first studied. Reducing the core geometry introduces heavy waveguide dispersion and reduces the ZDW, as can be seen in Fig. 1(b). The core radius changes from 1.2 μm to 0.5 μm while increased GVD below 2 μm is observed and the first ZDW shifts towards shorter wavelength. On the long wavelength side of the curves, GVD decreases for smaller cores and strong normal waveguide dispersion causes a second ZDW point, which turns towards shorter wavelength as the core radius decreases. Due to the strong material dispersion, it is difficult to shift the ZDW to shorter wavelength.

Fig. 1 (a) Material dispersion of GSS and ZBLAN; (b) Total GVD of the step-index fiber with a chalcogenide core and fluoride cladding. The radius of core reduces from 1.2 μm to 0.5 μm.

Download Full Size | PDF

We select a step-index MMF with r = 0.9 μm featuring a ZDW at 1.95 μm for further study. The effective fundamental mode area is 1.553 μm² and the nonlinear effect is 31.1 (W × m)⁻¹ at the ZDW point. The third order dispersion is also calculated to evaluate the flatness of the dispersion profile. It is 0.523 ps/nm²/km for the ZBLAN cladding fiber at ZDW point. As a comparision, it is 0.551 ps/nm²/km for the suspended core fiber with the same ZDW, which is assumed to have chalcogenide core and air clad in the simulation. In addition to a flatter GVD profile, the all-solid end face of the MMF possesses higher laser damage threshold. Besides, the all-solid MMFs are promising to splice with fiber laser pump source to increase the couple efficiency, which is not possible for the air-hole-contained suspended core fiber.

Figure 2 shows the evolution process as the pulse propagates along the fiber. The center wavelength is set to 1970 nm. The pulse width is 50 fs in Figs. 2(a) and 2(b); 100 fs in Figs. 2(c) and 2(d); 300 fs in Figs. 2(e) and 2(f) while the peak power is fixed to 175 W. The upper insets Figs. 2(a), 2(c) and 2(e) exhibit the pulse distribution in the time domain. The lower insets Figs. 2(b), 2(d) and 2(f) show light field changes corresponding to the frequency domain. It takes 0.2 cm in the 50 fs pulse condition to obtain a steady spectrum after a rapid extension. A soliton-like structure is recognizable in the time domain graph. The 100 fs pumping source requires about 0.3 cm to reach its full broadened capacity. In some wavelength bands, the spectrum is stronger than the other bands. Overall, the disparity of the intensity at different wavelengths comes within about 20dB. The 300 fs pulse requires longer fibers approximately 0.6 cm to broaden sufficiently.

Fig. 2 Evolution of the pulse in the time and frequency domains with (a, b) 50 fs, (c, d) 100 fs and (e, f) 300 fs pump pulse duration. The peak power is maintained at 175 W. The upper graphs show the temporal evolution and the bottom graphs show spectral evolution.

Download Full Size | PDF

Lacking experimental date, α in (3) is not considered in our study. This approximation is also used in other work such as [52]. The confinement loss for the fiber we used was also calculated but not shown here. Because confinement loss is negligible and it is not adequate to evaluate the total optical loss, which comprises the scattering loss, the imperfection loss, the ultraviolet and infrared absorption loss, the confinement loss and the absorption loss of other impurities. The experimental issues such as interface quality and stress induced by thermal expansion mismatch may cause additional optical loss. This is always a problem in MMFs. The transmission distance required in our fibers does not exceed 1 cm. Such a short fiber length helps to reduce optical loss efficiently. The high nonlinear coefficient counteracts the disadvantage of the high optical losses of MMF. For example, the measured propagation loss in a chalcogenide-tellurite composite fiber as shown in [39] is 18.3 dB/m at 1550 nm and the total loss should be higher in longer wavelength. But they still get broad band SCG because of the short interaction distance.

At the first stage of propagation, SPM domains the spectrum broaden and stimulate Raman scatting and four-wave mixing take charge later on. Considering the characteristic dispersive length scale $L_{D} = T_{0}^{2} / | β_{2} |$ and nonlinear length $L_{N L} = 1 / (γ P_{0})$ , 4.24 cm and 0.018 cm are respectively determined in 50 fs duration circumstance. The dispersion lengths, which equal to 16 and 144 cm, are respectively obtained under 100 and 300 fs pulse duration. Meanwhile the nonlinear length L_NL keeps unchanged. Pulses with a wider time scale require longer fibers to expand but the spectra are able to extend broader because the order of solitons is higher.

Figure 3 shows the spectrum and degree of coherence of the SCG with a 175 W peak power. The result is based on calculating 100 pairs of pulses propagating along 1 cm fiber using Eq. (6). Figure 3(a) shows the results of 300 fs pulse pumping and 3(b) shows the results of 100fs pumping. Under shorter pulse duration, the coherence is better in short wavelengths. Through the entire bandwidth from 1.2 μm to 3 μm, the coherence curve fluctuates strongly because soliton fission breaks the injected pulse, which is sensitive to input fluctuations and pump laser shot noise. Fiber length and pump conditions must match well to maintain acceptable coherence, which may limit further utilization of the supercontinuum source. Soliton dynamics can be avoided by using an all-normal dispersion fiber where the SPM plays a more significant role and modulation instability does not occur. Thus, the coherent property of the supercontinuum spectra can be maintained under rational fiber lengths using different ultra-fast pump sources. From Fig. 1(b), we can find that it is difficult to achieve all normal dispersion as the core diameter is changed. So MS-MMF is demonstrated in the next section.

Fig. 3 (a) Spectrum and coherence properties of SCG in the step-index MMF with 175 W peak power and 300 fs pumping. (b) Spectrum and coherence properties of SCG in the step-index MMF with 175 W peak power and 100 fs pumping

Download Full Size | PDF

4. SCG in all-solid fluoride-chalcogenide MS-MMFs

To achieve better coherence of SCG, we propose triangular MS-MMFs with six rings of ZBLAN holes. The six layers have structural ratios of f_i (i = 1 to 6), which is defined as d_i/Ʌ (d_i is the diameter of the hole in the ith layer and Ʌ is the pitch between the nearest hole centers. See Fig. 4.) [53]. Core diameter 2Ʌ-d₁ is used to assign the geometric dimensions of the MS-MMF to obtain an approximate assessment of the mode area.

Fig. 4 Cross section of the MS-MMF. d_i (i = 3 to 6) are the same in our study.

Download Full Size | PDF

Figure 5(a) shows the dispersion curves obtained when the structural ratio of different layers is maintained at f_i (i = 1 to 6) = 0.9, but the core diameter is increased from 0.9 μm to 1.2 μm. Figure 5(b) illustrates the dispersion curves obtained when core diameter set as 1μm while f_i (i = 1 to 6) is reduced from 0.95 to 0.8. Waveguide dispersion imposes an obvious effect on the curve at wavelengths longer than 1.6 μm in both conditions. The right wing shrinks and the top of the profile lowers as f diminishes, leading to a flatter roof and smaller slope. Similar results are obtained as the core diameter decreases. The mode field decreases from 0.93 μm² to 0.87 μm² as f increases as shown in the insert in Fig. 5(b). Figures 5(c) and 5(d) reveal the effect of variations in the structural ratio of different layers on GVD. f₂ and f₃ have been changed respectively with all the other characteristics kept constant, that is, Ʌ = 0.91 μm and all other layers have a structural ratio of 0.9. The dispersion profile can be adjusted to meet different requirements with changing f_i. In Fig. 5(c), by reducing only f₂ from 0.95 to 0.75, the GVD becomes more negative above 1.9 μm. Figure 5(d) depicts the same trend in the wavelength longer than 2.5 μm as bring f₂ back to the original value but f₃ changes. Comparing Fig. 5(c) and 5(d), it is obvious that the outer layers will not influence the dispersion profile in short wavelength as much as the inner layers. Since large f_i would contribute to confine the pulse energy in the core area, we fix f_i (i = 3 to 6) to 0.95 to reduce the confinement loss without influence the dispersion profile excessively. f₁ and f₂ are changed together to obtain an all normal GVD. f₂ is always kept the same value as f₁ to reduce the production difficulty.

Fig. 5 (a) Dispersion profile with f_i (i = 1 to 6) = 0.9, core = (0.9, 1.0, 1.1 and 1.2) μm. (b) Dispersion profile with core of 1 μm and f_i (i = 1 to 6) changing from 0.95 to 0.8. Insert in (b) shows the calculated effective area of the fundamental mode. (c) Dispersion profile with f₂ changed while all the other geometric parameters kept constant. (d) Dispersion profile with f₃ changed while all the other geometric parameters kept constant.

Download Full Size | PDF

As demonstrated in Fig. 6(a), the fiber with the black dispersion profile was selected for further study. Figure 6(b) shows the mode field distribution at the pumping wavelength. The calculated nonlinear coefficient γ is 54.9 W⁻¹/m. The high nonlinear coefficient helps to reduce the required launched power. The core size is 1 μm. In previous work, pump power could be coupled into fibers with much smaller cores in sub-micron grade with a coupling efficiency of ~10% to ~12% [54–56]. The numerical aperture is 1 for our fiber due to the large refraction index contrast, which will contribute to improve the coupling efficiency. The all-solid MMFs demonstrated here has some coupling superiority over traditional air-hole-contained MOFs. The all solid fiber surface can endure higher pump power comparing with air-hole-contained fibers. It can be spliced with tapered output fiber of pump source if necessary. However, the pump laser should still be coupled attentively into the end face of the MMF with a proper method in the practical experiment to enhance the coupling efficiency. The time and frequency evolution processes of three time scales at a peak power of 175 W are shown in Fig. 7. Figures 7(a) and 7(b) indicate 50 fs pumping. Figure 7(c) and 7(d) indicate 100 fs pumping. Figures 7(e) and 7(g) indicate 300 fs pumping. Figures 7(a), 7(c) and 7(e) show the temporal pulse distribution. Figures 7(b), 7(d) and 7(f) show spectral changes in the corresponding light fields. The all-normal GVD of the MS-MMF results in a phenomena that the shorter wavelength part transfers slower than longer wavelength components of the spectrum. Thus, the red-shifted components overtake blue-shifted components, and rapid oscillation structures can be observed in the temporal graph as shown in Fig. 7(c). The effects of high order dispersion, self-steeping and Raman scatting are in inverse proportion to the pulse duration. As the pulse width increases, the effects reduce. So under longer pulse widths, the SCG becomes a little narrower in Fig. 7. This is different from step-index MMFs shown in section 3. In Fig. 2, the soliton fission in step-index MMFs alleviates the peak power of the pulse. With shorter pulse width, the soliton fission takes place faster and the pump power decreases earlier, which reduces the width of the SCG. Besides, SCG in MS-MMFs is smoother and flatter than that in step-index fiber. It is the special dispersion profiles make these differences when the light propagates along the fibers.

Fig. 6 (a) Slight tailoring of the MS-MMF to realize all normal GVD. The black line with a core of 1.0 μm and the structural ratio of the two inner layers f₁ = f₂ = 0.88 and the outer structural ratio f_i (i = 3 to 6) = 0.95 is chosen for further study. (b) Mode field distribution at the pumping wavelength.

Download Full Size | PDF

Fig. 7 Evolution of the pulse in the time and frequency domains with (a, b) 50 fs, (c, d) 100 fs and (e, f) 300 fs pump pulse duration. The peak power is maintained at 175 W. The upper graphs show the temporal evolution and the bottom graphs show spectral evolution.

Download Full Size | PDF

Figure 8(a) shows the coherence and spectrum along the wavelength. It is a 175 W 100 fs pumping. In the SCG range, the coherence curve slightly increases as the wavelength increases and distributed between 0.8 and 1, which implies better coherence in the all-normal dispersion MS-MMF than that observed in the step-index MMF. The high nonlinear effect of GSS leads to short nonlinear lengths with low peak power. When fluoride fibers are applied for mid-infrared SCG, the peak power launched into the fibers often reaches several kilowatts because of the low nonlinearity [19, 20]. In this MS-MMF, the spectrum can extend to beyond 2.5 μm when the input peak power is 175 W with short fibers. It is possible to get wider spectrum with increased pump peak power if needed. Figure 8(b) illustrates the spectra by increasing the pump power while the pump pulse duration is held constant at 100 fs. The propagation distance required for SCG varies in Fig. 8(b). The spectra can extend to 5 μm as the pump peak power increased to 1400 W. It takes shorter than 1 mm in this condition for the spectrum to cover mid-infrared region. The intrinsic loss in ZBLAN glass may limit the extension of the spectra to wavelength beyond 5μm.

Fig. 8 (a) Degree of coherence and spectrum for 100 fs and 175W pump conditions. (b) Spectra obtained under different pump powers while keeping the pulse duration as 100 fs.

Download Full Size | PDF

5. Conclusion

We investigate MMFs combined with GSS and ZBLAN glass for SCG in the mid-infrared region. Compared to the fibers composed of single type of glasses, the refraction contrast between the two types of glasses allows more flexibility for dispersion control. The step-index MMF is designed to have a ZDW at ~1.95 μm to match the commercial available pumping sources. Benefitting from the high nonlinearity of chalcogenide glass, efficient SCG in the mid-infrared region can be achieved with short fibers. Relatively high level of optical loss in the MMFs can be counteracted by the short length of the fibers required for SCG. To obtain better coherence, all normal dispersion MS-MMFs without air holes for easier post processing are studied. Pumping MS-MMFs shorter than 1 cm with a femtosecond pulse laser with a peak power of 175 W at 1.95 μm yields broadband SCG with high coherence ranging from 1250 nm to 2750 nm. Low level pump powers make it possible to get the SCG from seed laser directly without complicated amplifiers. The spectrum obtained can be extended to 5 μm as the launched peak power increased.

Acknowledgments

This research was supported by “Hundred Talents Program” of the Chinese Academy of Sciences, the National Natural Science Foundation of China (NSFC) (No. 11374084 and 61307056), the “Pujiang Talent Plan” (No. 14PJ1409200) and the “Joint Research Project of Chinese Academy of Science and Japan Society for the Promotion of Science” (No. GJHZ1412).

References and links

1. J. K. Ranka, R. S. Windeler, and A. J. Stentz, “Visible continuum generation in air-silica microstructure optical fibers with anomalous dispersion at 800 nm,” Opt. Lett. 25(1), 25–27 (2000). [CrossRef] [PubMed]

2. J. M. Dudley, G. Genty, and S. Coen, “Supercontinuum generation in photonic crystal fiber,” Rev. Mod. Phys. 78(4), 1135–1184 (2006). [CrossRef]

3. S. T. Cundiff and J. Ye, “Colloquium: Femtosecond optical frequency combs,” Rev. Mod. Phys. 75(1), 325–342 (2003). [CrossRef]

4. 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]

5. A. F. Fercher and E. Roth, “Ophthalmic laser interferometry,” Proc. SPIE 658, 48–51 (1986). [CrossRef]

6. A. Labruyere, P. Leproux, V. Couderc, V. Tombelaine, J. Kobelke, K. Schuster, H. Bartelt, S. Hilaire, G. Huss, and G. Melin, “Structured-core GeO₂-doped photonic-crystal fibers for parametric and supercontinuum generation,” IEEE Photon. Technol. Lett. 22(16), 1259–1261 (2010). [CrossRef]

7. A. Schliesser, N. Picqué, and T. W. Hänsch, “Mid-infrared frequency combs,” Nat. Photonics 6(7), 440–449 (2012). [CrossRef]

8. G. J. Koch, J. Y. Beyon, B. W. Barnes, M. Petros, J. Yu, F. Amzajerdian, M. J. Kavaya, and U. N. Singh, “High-energy 2μm Doppler lidar for wind measurements,” Opt. Eng. 46(11), 116201 (2007). [CrossRef]

9. M. Ebrahim-Zadeh and I. T. Sorokina, Mid-infrared coherent sources and applications, NATO Science for Peace and Security, Series B: Physics and Biophysics. (Springer Verlag, The Netherlands, 2008).

10. T. Izawa, N. Shibata, and A. Takeda, “Optical attenuation in pure and doped fused silica in the IR wavelength region,” Appl. Phys. Lett. 31(1), 33–35 (1977). [CrossRef]

11. J. S. Sanghera, L. Brandon Shaw, and I. D. Aggarwal, “Chalcogenide glass-fiber-based mid-IR sources and applications,” IEEE J. Sel. Top. Quantum Electron. 15(1), 114–119 (2009). [CrossRef]

12. M. Liao, W. Gao, T. Cheng, Z. Duan, X. Xue, H. Kawashima, T. Suzuki, and Y. Ohishi, “Ultrabroad supercontinuum generation through filamentation in tellurite glass,” Laser Phys. Lett. 10(3), 036002 (2013). [CrossRef]

13. S. Guan, Y. Tian, Y. Guo, L. Hu, and J. Zhang, “Spectroscopic properties and energy transfer processes in Er³⁺/Nd³⁺ co-doped tellurite glass for 2.7-μm laser materials,” Chin. Opt. Lett. 10(7), 71603–71607 (2012). [CrossRef]

14. F. Huang, X. Liu, W. Li, L. Hu, and D. Chen, “Energy transfer mechanism in Er doped fluoride glass sensitized by Tm³⁺ or Ho³⁺ for 2.7-μm emission,” Chin. Opt. Lett. 12(5), 51601–51604 (2012). [CrossRef]

15. O. P. Kulkarni, V. V. Alexander, M. Kumar, M. J. Freeman, M. N. Islam, F. L. Terry Jr, M. Neelakandan, and A. Chan, “Supercontinuum generation from ~1.9 to 4.5 μm in ZBLAN fiber with high average power generation beyond 3.8 μm using a thulium-doped fiber amplifier,” J. Opt. Soc. Am. B 28(10), 2486–2498 (2011). [CrossRef]

16. J. Swiderski, M. Michalska, and G. Maze, “Mid-IR supercontinuum generation in a ZBLAN fiber pumped by a gain-switched mode-locked Tm-doped fiber laser and amplifier system,” Opt. Express 21(7), 7851–7857 (2013). [CrossRef] [PubMed]

17. X. Zhu and R. Jain, “10-W-level diode-pumped compact 2.78 microm ZBLAN fiber laser,” Opt. Lett. 32(1), 26–28 (2007). [CrossRef] [PubMed]

18. D. Tran, G. Sigel, and B. Bendow, “Heavy-metal fluoride glasses and fibers: A review,” J. Lightwave Technol. 2(5), 566–586 (1984). [CrossRef]

19. C. Xia, M. Kumar, O. P. Kulkarni, M. N. Islam, F. L. Terry Jr, M. J. Freeman, M. Poulain, and G. Mazé, “Mid-infrared supercontinuum generation to 4.5 microm in ZBLAN fluoride fibers by nanosecond diode pumping,” Opt. Lett. 31(17), 2553–2555 (2006). [CrossRef] [PubMed]

20. C. Xia, M. Kumar, M.-Y. Cheng, R. S. Hegde, M. N. Islam, A. Galvanauskas, H. G. Winful, F. L. Terry Jr, M. J. Freeman, M. Poulain, and G. Mazé, “Power scalable mid-infrared supercontinuum generation in ZBLAN fluoride fibers with up to 1.3 watts time-averaged power,” Opt. Express 15(3), 865–871 (2007). [CrossRef] [PubMed]

21. G. S. Qin, X. Yan, C. Kito, M. S. Liao, C. Chaudhari, T. Suzuki, and Y. Ohishi, “Ultra-broadband supercontinuum generation from ultraviolet to 6.28 µm in a fluoride fiber,” in Optical Fiber Communication Conference, Technical Digest (CD) (Optical Society of America, 2010), paper OTuJ6.

22. L. Liu, G. Qin, Q. Tian, D. Zhao, and W. Qin, “Numerical investigation of mid-infrared supercontinuum generation up to 5 μm in single mode fluoride fiber,” Opt. Express 19(11), 10041–10048 (2011). [CrossRef] [PubMed]

23. I. Kubat, C. R. Petersen, U. V. Møller, A. Seddon, T. Benson, L. Brilland, D. Méchin, P. M. Moselund, and O. Bang, “Thulium pumped mid-infrared 0.9-9μm supercontinuum generation in concatenated fluoride and chalcogenide glass fibers,” Opt. Express 22(4), 3959–3967 (2014). [CrossRef] [PubMed]

24. J. M. Dudley and S. Coen, “Coherence properties of supercontinuum spectra generated in photonic crystal and tapered optical fibers,” Opt. Lett. 27(13), 1180–1182 (2002). [CrossRef] [PubMed]

25. H. Tu and S. A. Boppart, “Coherent fiber supercontinuum for biophotonics,” Laser Photon Rev 7(5), 628–645 (2013). [CrossRef] [PubMed]

26. L. B. Shaw, R. R. Gattass, J. Sanghera, and I. Aggarwal, “All-fiber mid-IR supercontinuum source from 1.5 to 5 μm,” Proc. SPIE 7914, 79140P (2011). [CrossRef]

27. J. Yang, Y. Tang, and J. Xu, “Development and applications of gain-switched fiber lasers [Invited],” Photon. Res. 1(1), 52–57 (2013). [CrossRef]

28. H. Chen, Z. Chen, X. Zhou, and J. Hou, “Cascaded PCF tapers for flat broadband supercontinuum generation,” Chin. Opt. Lett. 10(12), 120603 (2012). [CrossRef]

29. J. Hou, J. Zhao, C. Yang, Z. Zhong, Y. Gao, and S. Chen, “Engineering ultra-flattened-dispersion photonic crystal fibers with uniform holes by rotations of inner rings,” Photon. Res. 2(2), 59–63 (2014). [CrossRef]

30. S. Shabahang, G. Tao, M. P. Marquez, H. Hu, T. R. Ensley, P. J. Delfyett, and A. F. Abouraddy, “Nonlinear characterization of robust multimaterial chalcogenide nanotapers for infrared supercontinuum generation,” J. Opt. Soc. Am. B 31(3), 450–457 (2014). [CrossRef]

31. A. Marandi, C. W. Rudy, N. C. Leindecker, V. G. Plotnichenko, E. Dianov, K. Vodopyanov, and R. L. Byer, “Mid-Infrared Supercontinuum Generation from 2.4 µm to 4.6 µm in Tapered Chalcogenide Fiber,” in CLEO: Science and Innovations, Technical Digest (online) (Optical Society of America, 2012), paper: CTh4B.5.

32. W. Gao, M. El Amraoui, M. Liao, H. Kawashima, Z. Duan, D. Deng, T. Cheng, T. Suzuki, Y. Messaddeq, and Y. Ohishi, “Mid-infrared supercontinuum generation in a suspended-core As₂S₃ chalcogenide microstructured optical fiber,” Opt. Express 21(8), 9573–9583 (2013). [CrossRef] [PubMed]

33. N. D. Psaila, R. R. Thomson, H. T. Bookey, S. Shen, N. Chiodo, R. Osellame, G. Cerullo, A. Jha, and A. K. Kar, “Supercontinuum generation in an ultrafast laser inscribed chalcogenide glass waveguide,” Opt. Express 15(24), 15776–15781 (2007). [CrossRef] [PubMed]

34. X. Gai, D.-Y. Choi, S. Madden, Z. Yang, R. Wang, and B. Luther-Davies, “Supercontinuum generation in the mid-infrared from a dispersion-engineered As₂S₃ glass rib waveguide,” Opt. Lett. 37(18), 3870–3872 (2012). [CrossRef] [PubMed]

35. R. Buczynski, D. Pysz, I. Kujawa, P. Fita, M. Pawlowska, J. Nowosielski, C. Radzewicz, and R. Stepien, “Silicate all-solid photonic crystal fibers with a glass high index contrast,” Proc. SPIE 6588, 658802 (2007). [CrossRef]

36. A. F. Abouraddy, M. Bayindir, G. Benoit, S. D. Hart, K. Kuriki, N. Orf, O. Shapira, F. Sorin, B. Temelkuran, and Y. Fink, “Towards multimaterial multifunctional fibres that see, hear, sense and communicate,” Nat. Mater. 6(5), 336–347 (2007). [CrossRef] [PubMed]

37. F. Sorin, A. F. Abouraddy, N. Orf, O. Shapira, J. Viens, J. Arnold, J. D. Joannopoulos, and Y. Fink, “Multimaterial photodetecting fibers: a geometric and structural study,” Adv. Mater. 19(22), 3872–3877 (2007). [CrossRef]

38. G. Tao, A. M. Stolyarov, and A. F. Abouraddy, “Multimaterial fibers,” I. J. Appl. Glass Sci. 3(4), 349–368 (2012). [CrossRef] [PubMed]

39. 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]

40. T. Cheng, Y. Kanou, D. Deng, X. Xue, M. Matsumoto, T. Misumi, T. Suzuki, and Y. Ohishi, “Fabrication and characterization of a hybrid four-hole AsSe₂-As₂S₅ microstructured optical fiber with a large refractive index difference,” Opt. Express 22(11), 13322–13329 (2014). [CrossRef] [PubMed]

41. C. Chaudhari, M. S. Liao, T. Suzuki, and Y. Ohishi, “Chalcogenide Core Tellurite Cladding Composite Microstructured Fiber for Nonlinear Applications,” J. Lightwave Technol. 30(13), 2069–2076 (2012). [CrossRef]

42. 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, “Towards mid-infrared supercontinuum generation: Ge-Sb-Se mid-infrared step-index small-core optical fiber,” Proc. SPIE 8938, 89380W (2014). [CrossRef]

43. J. M. Jewell, C. Askins, and I. D. Aggarwal, “Interferometric method for concurrent measurement of thermo-optic and thermal expansion coefficients,” Appl. Opt. 30(25), 3656–3660 (1991). [CrossRef] [PubMed]

44. M. Naftaly, S. Shen, and A. Jha, “Tm³⁺-doped tellurite glass for a broadband amplifier at 1.47 µm,” Appl. Opt. 39(27), 4979–4984 (2000). [CrossRef] [PubMed]

45. A. Lin, A. Zhang, E. J. Bushong, and J. Toulouse, “Solid-core tellurite glass fiber for infrared and nonlinear applications,” Opt. Express 17(19), 16716–16721 (2009). [CrossRef] [PubMed]

46. T. Kohoutek, X. Yan, T. W. Shiosaka, S. N. Yannopoulos, A. Chrissanthopoulos, T. Suzuki, and Y. Ohishi, “Enhanced Raman gain of Ge–Ga–Sb–S chalcogenide glass for highly nonlinear microstructured optical fibers,” J. Opt. Soc. Am. B 28(9), 2284–2290 (2011). [CrossRef]

47. P. Klocek and L. Colombo, “Index of refraction, dispersion, bandgap and light scattering in GeSe and GeSbSe glasses,” J. Non-Cryst. Solids 93(1), 1–16 (1987). [CrossRef]

48. F. Gan, “Optical properties of fluoride glasses: a review,” J. Non-Cryst. Solids 184, 9–20 (1995). [CrossRef]

49. L. Petit, N. Carlie, H. Chen, S. Gaylord, J. Massera, G. Boudebs, J. Hu, A. Agarwal, L. Kimerling, and K. Richardson, “Compositional dependence of the nonlinear refractive index of new germanium-based chalcogenide glasses,” J. Solid State Chem. 182(10), 2756–2761 (2009). [CrossRef]

50. J. Hu, C. R. Menyuk, L. B. Shaw, J. S. Sanghera, and I. D. Aggarwal, “Raman response function and supercontinuum generation in chalcogenide fiber,” in Conference on Lasers and Electro-Optics, (Optical Society of America, 2008), paper CMDD2. [CrossRef]

51. B. Ung and M. Skorobogatiy, “Chalcogenide microporous fibers for linear and nonlinear applications in the mid-infrared,” Opt. Express 18(8), 8647–8659 (2010). [CrossRef] [PubMed]

52. S. Wang, J. Hu, H. Guo, and X. Zeng, “Optical Cherenkov radiation in an As₂S₃ slot waveguide with four zero-dispersion wavelengths,” Opt. Express 21(3), 3067–3072 (2013). [CrossRef] [PubMed]

53. F. Poli, A. Cucinotta, S. Selleri, and A. H. Bouk, “Tailoring of ﬂattened dispersion in highly nonlinear photonic crystal ﬁbers,” IEEE Photon. Technol. Lett. 16(4), 1065–1067 (2004). [CrossRef]

54. M. Liao, G. Qin, X. Yan, T. Suzuki, and Y. Ohishi, “A tellurite nanowire with long suspended struts for low-threshold single-mode supercontinuum generation,” J. Lightwave Technol. 29(2), 194–199 (2011). [CrossRef]

55. M. Liao, C. Chaudhari, G. S. Qin, X. Yan, T. Suzuki, and Y. Ohishi, “A highly nonlinear tellurite microstructured fiber pumped by picosecond pulse for supercontinuum generation,” in Proceedings of IEEE Optoelectronics and Communications Conference (IEEE, 2010), pp. 160–161.

56. N. Granzow, M. A. Schmidt, W. Chang, L. Wang, Q. Coulombier, J. Troles, P. Toupin, I. Hartl, K. F. Lee, M. E. Fermann, L. Wondraczek, and P. S. J. Russell, “Mid-infrared supercontinuum generation in As₂S₃-silica “nano-spike” step-index waveguide,” Opt. Express 21(9), 10969–10977 (2013). [CrossRef] [PubMed]

Glass type		Tg (°C)	Transparency Range		Refraction Index	Nonlinear Refraction ( × 10⁻²⁰)
Tellurite [44, 45]	ZnO-TeO₂	302	0.35-5 μm	~2.0	~2.05	38
Fluoride	ZnF₄-BaF₂	265	0.3-5 μm	1.65	1.5	3.3
Chalcogenide [46]	Ge₂₀Sb₁₅Se₆₅	264	1-14 μm	5.0	~2.5	1500

Low threshold mid-infrared supercontinuum generation in short fluoride-chalcogenide multimaterial fibers

Abstract

1. Introduction

2. Numerical simulation

3. SCG in fluoride-chalcogenide step-index MMFs

4. SCG in all-solid fluoride-chalcogenide MS-MMFs

5. Conclusion

Acknowledgments

References and links

Cited By

Figures (8)

Tables (1)

Equations (6)

Optics Express