A novel in-fiber whispering-gallery mode (WGM) microsphere resonator-based integrated device is reported. It is fabricated by placing a silica microsphere into an embedded dual-core hollow fiber (EDCHF). Using a fiber tapering method, a silica microsphere can be placed and fixed in the transition section of the hollow core of the EDCHF. The transmitted light from the tapered-input single-mode fiber is coupled into the embedded silica microsphere via the two suspended fiber cores, and hence effectively excites the WGMs. A -factor of is achieved over the wavelength range of 1100–1300 nm. The polarization and temperature dependence of the in-fiber WGM microsphere resonator device is also investigated experimentally. This integrated photonics device provides greatly improved mechanical stability, compared with the traditional tapered fiber-coupled WGM microresonator devices. Additional advantages include ease of fabrication, compact structure, and low cost. This novel in-fiber WGM resonator integrated device is ideally positioned to access a wide range of potential applications in optical sensing and microcavity lasing.
© 2018 Optical Society of America
In recent years, basic micro–nano integrated photonic building blocks, variants of microresonators, have been widely explored and developed for a wide range of applications, such as optical communication (e.g., narrowband optical resonant, modulator, comb filter [1–3]), sensing [4,5], slow light , ultralow threshold microcavity lasers , microcavity quantum electrodynamics , nonlinear optics , and other important optical research areas. Microsphere resonators have drawn significant attention due to their ultrahigh -factor (up to ) and very small mode volume as well as other advantages compared with other microresonators, including ease of fabrication and relatively low cost.
To date, several coupling methods for coupling light into and out of the microresonator to effectively excite the whispering gallery modes (WGMs) of the microsphere have been reported, e.g., prism coupling , tapered fiber , half-block , and angle-polished fiber . The most common and efficient approach is to use a fiber taper with a very thin waist where the evanescent field is strongly exposed . The microsphere–taper coupling approach therefore offers a nearly critical coupling condition [11,14]. However, this microsphere–taper coupling system always suffers from environmental turbulence due to the poor mechanical stability. Hence the robustness of such microsphere–taper systems can be improved using an UV curable epoxy packaging technique , and a -factor as high as has been realized . Unfortunately, the UV curable epoxy used for packaging the microsphere–taper usually comprises toxic, volatile polymeric fluorinated polymers, and has other disadvantages, e.g., poor thermal stability, large absorption loss, and limited lifetime after the UV curing process. In order to improve the mechanical stability of the fiber–microsphere structure, significant enhancements have been achieved using a number of specialty optical fibers for encapsulating and exciting the WGMs of a microsphere placed inside light guides, such as the capillary–microsphere structure , photonic crystal fiber–microsphere structure  and microstructured optical fiber–microsphere structure . Many of the fabrication techniques for making in-fiber WGM devices demand a chemical etching process using hydrofluoric acid [17,18], which is highly unsafe to handle, and thus increases the difficulty of fabrication and potential manufacturing costs. Therefore, an easy, simple, and efficient in-fiber coupling system is potentially required.
In the work of this investigation, an in-fiber WGMs microsphere resonator integrated device has been developed and successfully fabricated, as shown schematically in Fig. 1. The unique hollow structure of embedded dual-core hollow fiber (EDCHF) allows the microsphere to be easily placed into the optical fiber without the need for any additional chemical etching process, and the WGMs can be effectively excited by the light propagation through the embedded cores on the fiber inner wall. A taper is necessarily included at the splicing points between the EDCHF and input/output of the single-mode fibers (SMFs). Therefore, by adjusting the tapering parameters, the microsphere is encapsulated inside the gap between the embedded cores. The polarization and temperature dependence of this integrated device have also been characterized. The WGM device presented in this Letter is much more robust and compact, and also this structure provides a means to realize an alignment-free and package-free integrated “lab-in-fiber” platform for the excitation of the WGMs and therefore accesses a wide range of new sensing and lasing applications, compared to established tapered optical fiber-based WGMs.
The coupling mechanism between the embedded cores and the microsphere can be explained as follows. The inset of Fig. 1 illustrates the propagation of light. Taking the coupling of one of the fiber’s cores and the microsphere as an example, when the light in the core reaches the contact point with the microsphere, a part of the light couples into the microsphere and the remainder continues to propagate along the core. The light coupled into the microsphere will excite the resonance of the WGMs if the wavelength satisfies the resonance condition. Furthermore, a portion of the light coupled in the microsphere is also in turn recoupled into the core and reaches the output. The matrix equation for the electric fields at various points in the figure can be expressed as 2 shows the electric field distribution of microspheres in different profiles using a commercial simulation software (Rsoft Fullwave) based on the well-known finite-difference time-domain method. It can be seen in the simulation that the input light is coupled into the microsphere and excites the resonance of the WGMs.
The EDCHF structure used in this investigation is shown in Fig. 3(a). The optical fiber consists of an annular cladding with diameter , a central air hole with diameter , and two semielliptical cores. The two cores with a major axis and a minor axis are embedded symmetrically in the annular cladding. A thin cladding exists as a layer between the air and cores and its thinnest section () is about 1.85–2 μm. The refractive index of the cladding and cores are 1.462 and 1.466, respectively. In the first step of the fabrication, the EDCHF was spliced with a standard 125 μm diameter SMF (SMF-28, Thorlabs) using a commercial splicer machine (Fujikura, ARC master FSM-100P+). Compared with traditional fusion splicers, this machine has a variety of arc discharge modes to splice various types of specialty fibers and a SWEEP motor, which can move the fiber fixed by the left and right fiber holders along the fiber axis. Using these functions, the splicing parameters were modified several times to produce a minimum insertion loss. Here, the arc current and the duration of the fusion splicer were manually controlled to approximately 13.2 mA and 2 s, respectively. The alignment method adopted for fusion splicing in the experiments was cladding alignment. Then the joint section of SMF-EDCHF was tapered using a filament heating method and yielding a microtaper (40 μm), which plays a role of light beam coupler that splits the input light from the SMF into the two cores of EDCHF. In this case, the diameter of the microsphere is approximately the same as the EDCHF cone so that the microsphere could be naturally fixed within the EDCHF.
A number of taper fibers with a taper waist diameter of 10–30 μm were subsequently prepared by heating a strand of 1060XP SMF with a ceramic microheater (CMH-7-19, NTT-AT), and these were cut in the middle to produce half-tapered optical fibers. The SMF-EDCHF was placed on a three-dimensional micrometer stepper and fixed in place using a fiber holder. Figure 3(c) shows the micrograph of the microspheres, and it can be observed that the microspheres exhibit regular circular shape, uniform size, and a smooth surface. One microsphere was observed to have a diameter of 47 μm, which was measured by aligning it with a half-tapered fiber and attaching it to the tip of the fiber by utilizing the electrostatic force that exists between them and subsequently placing it on another stepper.
By adjusting the three-dimensional stepper under microscope observation, the microsphere was successfully placed and fixed inside the air hole of the EDCHF. Figures 4(a) and 4(b) show the process and result of the micro-operation. Finally, the other end of the EDCHF was spliced and tapered with a SMF. The length of the EDCHF should not be too short. During splicing, the EDCHF internally generated pressure will cause the fiber to deform, and this can potentially lead to increased loss. It should also not be too long in order to prevent the microsphere from moving too far inside the EDCHF, which can result in the surface of microsphere becoming worn, leading to a significant decrease in the -factor. Following several experiments, the optimum length of the hollow fiber was determined to be 3 mm, and this produced a measurable spectral output, as shown in Fig. 4(c). The resonance dip exhibits an extinction ratio (ER) of the resonance shape of circa 15 dB and a full width at half-maximum of 0.22 nm. The -factor was calculated to be , which is limited by the scattering loss caused by the minor degradation of the smoothness of the microsphere surface due to the effect of contact of the microsphere surface with the optical fiber, and the optical loss of microsphere material itself.
Figure 5 shows the experimental setup for measuring both the polarization and the temperature dependence. Combined with the enlarged optical structure shown in Fig. 1, we can see light from a supercontinuum light source (SC YSL SC-series) was transmitted via a taper at the splicing point to the two cores suspended on the inner wall of the EDCHF. The resonance of the WGMs that satisfies the resonance conditions was transmitted through two cores of the EDCHF and coupled into the SMF via another cone area. The transmitted spectrum was collected using an optical spectrum analyzer (OSA YOKOGAWA, AQ-6370C). The fiber structure was placed in the climate chamber (ESPEC SH-222).
First, by adjusting the polarization angle, the polarization characteristics of the resonance of the WGMs were investigated. As the polarization direction of the input light changes, the direction of its electric field in the resonator also changes. According to the boundary continuity condition of the electromagnetic field, the TE and TM modes, respectively, have their own eigenfrequencies, as shown in Fig. 6(a). By varying the polarization state of the incident broadband light source using a linear polarizer, we could resolve a small difference as high as 0.14 nm in the resonance positions. The deviation of the resonant wavelength is attributed to the polarization-dependent coupling from the asymmetric boundary conditions at the interfaces between the embedded fiber cores and the microsphere. The effect of the fabrication parameters, i.e., the ellipticity of the fiber core, the thickness of the fiber cladding, the coupling length, and a design for a polarization-independent EDCHF–microsphere will be further investigated in due course by the authors.
The temperature characteristics were measured by varying the temperature of the climate chamber. As the temperature was changed, the resonant peak position also shifted in wavelength. The resonances of the WGMs are susceptible to external thermal fluctuations caused by either the surrounding environmental temperature changes or the absorption of circulating light within the cavity material. From the resonant condition, resonance shift can be expressed as 22]. When the temperature was increased, the position of the resonance peak shifts toward the longer wavelength. The resonant wavelength was observed to shift around 1199.5 nm for every 4°C temperature increase in the range 20°C to 60°C, and the result was characterized using a linear fit, which produced an average temperature sensitivity of 25.5 pm/°C with a linear fitting coefficient of 0.9824. This is a relatively high temperature sensitivity value and covers a relatively large temperature range when compared with other silica microsphere-based devices used for temperature sensing. From the output spectrum of Figs. 6(b) and 6(c), it is clear that the ER of the resonance also changed, which can be explained by the fact that the microsphere expands slightly when the temperature increases. The coupling between the microsphere and the fiber cores is also altered, which in turn affects the geometry of the structure. The same sample was also tested for changes in resonant wavelength as the temperature decreased, which resulted in a temperature sensitivity of 21.9 pm/°C, and it is close to the value obtained within the same temperature range during the warming process.
A novel in-fiber structure for the excitation of the WGMs of microsphere resonators has been successfully fabricated and tested as a proof of concept. The fabrication of the in-fiber integrated device is simple and involves only fusion splicing. Light is directly coupled via the EDCHF cores to the microsphere enclosed within the silica hollow core. The measured -factor was , which is limited by the quality of the microsphere and the imperfect coupling condition between the fiber cores and the microsphere. The polarization and the temperature dependence were also measured. This in-fiber microsphere resonator system is compact, robust, and has a great potential in high-performance sensing and lasing applications.
National Natural Science Foundation of China (NSFC) (61575050); National Key R&D Program (2016YFE0126500); Key Program for Natural Science Foundation of Heilongjiang Province of China (ZD2016012); National Natural Science Foundation Youth Science Fund of China (11704086); Open Fund of the State Key Laboratory on Integrated Optoelectronics (IOSKL2016KF03); 111 Project (B13015); Recruitment Program for Young Professionals (The Young Thousand Talents Plan).
1. M. A. Popovíc, T. Barwicz, M. R. Watts, P. T. Rakich, L. Socci, E. P. Ippen, F. X. Kärtner, and H. I. Smith, Opt. Lett. 31, 2571 (2006). [CrossRef]
2. Q. Xu, B. Schmidt, S. Pradhan, and M. Lipson, Nature 435, 325 (2005). [CrossRef]
3. T. J. Kippenberg, R. Holzwarth, and S. A. Diddams, Science 332, 555 (2011). [CrossRef]
4. T. Aoki, B. Dayan, E. Wilcut, W. P. Bowen, A. S. Parkins, T. J. Kippenberg, K. J. Vahala, and H. J. Kimble, Nature 443, 671 (2006). [CrossRef]
5. A. M. Armani, R. P. Kulkarni, S. E. Fraser, R. C. Flagan, and K. J. Vahala, Science 317, 783 (2007). [CrossRef]
6. C. Liu, Z. Dutton, C. H. Behroozi, and L. V. Hau, Nature 409, 490 (2001). [CrossRef]
7. S. M. Spillane, T. J. Kippenberg, and K. J. Vahala, Nature 415, 621 (2002). [CrossRef]
8. S. M. Spillane, T. J. Kippenberg, O. J. Painter, and K. J. Vahala, Phys. Rev. Lett. 91, 043902 (2003). [CrossRef]
9. D. J. Moss, R. Morandotti, A. L. Gaeta, and M. Lipson, Nat. Photonics 7, 597 (2013). [CrossRef]
10. M. L. Gorodetsky and V. S. Ilchenko, Opt. Commun. 113, 133 (1994). [CrossRef]
11. J. C. Knight, G. Cheung, F. Jacques, and T. A. Birks, Opt. Lett. 22, 1129 (1997). [CrossRef]
12. B. E. Little, J. P. Laine, and H. A. Haus, J. Lightwave Technol. 17, 704 (1999). [CrossRef]
13. V. S. Ilchenko, X. S. Yao, and L. Maleki, Opt. Lett. 24, 723 (1999). [CrossRef]
14. M. Cai, O. Painter, and K. J. Vahala, Phys. Rev. Lett. 85, 74 (2000). [CrossRef]
15. P. Wang, M. Ding, T. Lee, G. S. Murugan, L. Bo, Y. Semenova, Q. Wu, D. Hewak, G. Brambilla, and G. Farrell, Appl. Phys. Lett. 102, 131110 (2013). [CrossRef]
16. Y. Z. Yan, C. L. Zou, S. B. Yan, F. W. Sun, Z. Ji, J. Liu, Y. G. Zhang, L. Wang, C. Y. Xue, W. D. Zhang, Z. F. Han, and J. J. Xiong, Opt. Express 19, 5753 (2011). [CrossRef]
17. X. Zhang, Y. Yang, H. Bai, J. Wang, M. Yan, H. Xiao, and T. Wang, Photon. Res. 5, 516 (2017). [CrossRef]
18. R. Wang, M. Fraser, J. Li, X. Qiao, and A. Wang, Opt. Lett. 40, 308 (2015). [CrossRef]
19. K. Kosma, G. Zito, K. Schuster, and S. Pissadakis, Opt. Lett. 38, 1301 (2013). [CrossRef]
20. X. Zhang, D. Huang, and X. Zhang, Opt. Express 15, 13557 (2007). [CrossRef]
21. J. M. Ward, N. Dhasmana, and S. N. Chormaic, Eur. Phys. J. 223, 1917 (2014). [CrossRef]
22. T. J. Kippenberg, S. M. Spillane, and K. J. Vahala, Phys. Rev. Lett. 93, 083904 (2004). [CrossRef]