July 1, 2014 / Vol. 39, No. 13 / OPTICS LETTERS
3845
Iron-oxide nanoparticles embedded silica microsphere resonator exhibiting broadband all-optical wavelength tunability Ping Zhao,1 Lei Shi,1,* Yang Liu,1 Zheqi Wang,1 Shengli Pu,2 and Xinliang Zhang1,3 1
Wuhan National Laboratory for Optoelectronics & School of Optical and Electronic Information, Huazhong University of Science and Technology, Wuhan 430074, China 2 College of Science, University of Shanghai for Science and Technology, Shanghai 200093, China 3 e-mail: [email protected] *Corresponding author: [email protected]
Received March 27, 2014; revised May 19, 2014; accepted May 25, 2014; posted May 27, 2014 (Doc. ID 208898); published June 23, 2014 In this Letter, a novel silica microsphere resonator (MSR) embedded with iron-oxide nanoparticles, which possesses broadband all-optical wavelength tunability, is demonstrated. It is generated by using in-line 1550 nm laser ablation of a microfiber with the assistance of magnetic fluid. To the best of our knowledge, this simple method of fabricating such MSRs is reported for the first time. Prominent photothermal effect is realized by the iron-oxide nanoparticles absorbing light pumped via the fiber stem, leading to a wavelength shift of over 13 nm (1.6 THz). Moreover, a linear tuning efficiency up to 0.2 nm∕mW is realized. With excellent robustness and being fiberized, the spheres can be attractive elements in building up novel micro-illuminators, point heaters, optical sensors, and fiber communication modules. © 2014 Optical Society of America OCIS codes: (230.5750) Resonators; (190.4870) Photothermal effects; (060.2340) Fiber optics components. http://dx.doi.org/10.1364/OL.39.003845
Due to high-quality factors, optical microsphere resonators (MSRs) have been of great interest in basic research and engineering applications [1–3], including nonlinear optics [4], quantum optics [5], optical sensing [6], optical signal processing [7], and microwave photonics [8]. Wavelength tuning of MSRs is of great importance in enhancing the flexibility of MSRs. Mechanical wavelength tuning of silica microspheres has been previously demonstrated [9–11]. However, the wavelength-tunable range is less than 1 nm, and the pressing of microspheres unavoidably results in noncircle geometry deformations of the microspheres. As a result, contactless methods of changing the resonance wavelength, such as electrical or thermal tuning, were investigated [12–15]. The elastic strain of an MSR arises under the force exerted by an electrostatic field, while the low piezoelectric modulus leads to a wavelength-tuning range of only tens of picometers [12]. On the other hand, thermal wavelength adjustment has a large tuning range [13,15]. The temperature of conventional microelectric heaters can hardly reach a high level in order to induce a prominent wavelength shift of MSRs. Heating with lasers can overcome such a problem and is quite flexible [13]. Tapalian et al. reported thermo-optical modulators using polymer-coated MSRs [13]. Nevertheless, polymer matrix is usually subjected to poor environmental durability. Hence solid glass MSRs with high stabilities are favored. However, the thermo-optical effect of pure silica MSRs is weak. Doping in silica is an effective approach of enhancing this effect. An all-optical thermal wavelength tuning for erbium: ytterbium-doped glass MSRs was demonstrated by Watkins et al. [15]. However, the pump wavelength was limited due to the absorption of such glass. What’s worse, the signal-to-noise ratio of optical signals at 1550 nm may be degraded due to the spontaneousemission noise generated along the pumping processes. 0146-9592/14/133845-04$15.00/0
In this Letter, we propose and demonstrate broadband all-optical wavelength tuning of a novel MSR. The key component is a silica microsphere embedded with ironoxide nanoparticles. These particles can perform highly efficient photon-to-heat conversion. The sphere has a core-shell similar structure: the nanoparticles are mainly concentrated in the core area rather than near the surface. Consequently, the light wave traveling through the core is strongly absorbed, while the signal propagating as whispering-gallery modes (WGMs) experiences little loss. With a 1550 nm pump, a wavelength shift of more than 13 nm (1.6 THz) is achieved in the C-band of optical fiber communications with respect to the fabricated MSR. A broad linear-tuning wavelength range of more than one terahertz is obtained, and the corresponding power tuning efficiency reaches up to 0.2 nm∕mW. The proposed wavelength tuning of the MSRs also exhibits excellent robustness and high flexibility. The silica microsphere embedded with iron-oxide nanoparticles was fabricated by an in-line infrared-laser ablation of optical microfiber surrounded by magnetic fluid. The microfiber was tapered from a standard single-mode fiber (SMF) (YOFC, China) with a flame-heated technique [16,17]. To fabricate microspheres, commercial water-based magnetic fluid (Beijing Sunrise Ferrofluid Tech. Co. Ltd., China) was utilized. Figure 1(d) shows the image of a sample of 2 mL magnetic fluid. It is composed of iron-oxide nanoparticles uniformly dispersed in deionized water. Due to strong wideband optical absorption and magnetic properties [18], this kind of nanoparticle has raised plenty of interest in optics [19–22]. The diameter of the nanoparticle is about 8 nm, and the concentration is 33 w.t. %. Figures 1(a)– 1(c) show the fabrication process of such a microsphere. First, a drop of magnetic fluid was dropped on the end facet of a cleaved glass rod with a diameter of 3 mm. © 2014 Optical Society of America
3846
OPTICS LETTERS / Vol. 39, No. 13 / July 1, 2014
Fig. 1. Fabrication process of microspheres. (a) A microfiber is immerged into a drop of magnetic fluid. SMF, single-mode fiber. (b) The microfiber is melted and snapped under in-line magnetic-fluid-assisted laser ablation. (c) The pump power is increased up to 200 mW. An iron-oxide nanoparticles embedded silica microsphere with a smooth surface is generated after the further ablation. The inset is a scanning electron microscope image of the resulting microsphere. The scale bar is 10 μm. (d) 2 mL magnetic fluid sample.
Due to the surface tension, a liquid semi-sphere with a diameter of about 3 mm was formed as shown by Fig. 1(a). An optical microfiber was then emerged into the liquid drop. The diameter of the microfiber can be from submicrometer to tens of micrometers. Next, a 1550 nm light wave from a semiconductor laser was injected into the microfiber through the SMF after being amplified by an erbium-doped fiber amplifier. The initial ablation power was around 30 mW. The evanescent wave along the microfiber acts as optical tweezers and attracts nanoparticles [23]. These nanoparticles were then attached onto the microfiber surface, strongly absorbing the infrared rays [24]. Simultaneously, prodigious amounts of heat were generated. As presented by Fig. 1(b), the local temperature dramatically increased, resulting in the melting and breakage of the microfiber. Due to the surface tension, the end of the microfiber was quickly fused into a sphere. Some iron-oxide nanoparticles were blended in the melted sphere. Next, we further increase the ablation power up to about 200 mW to generate a smoother microsphere [Fig. 1(c)]. The sphere expanded under the further heating. During the expansion of the sphere, the magnetic nanoparticles moved toward the core area in the sphere. It is believed that the movements are due to the mutual magnetic attraction between these particles. When the heat dissipation and generation were balanced, the sphere geometry was then maintained. After the withdrawal of the optical pump, a smooth microsphere was harvested, as shown by the inset of Fig. 1(c). The nanoparticles were mainly distributed near the core of the sphere. On the contrary, there were few iron-oxide nanoparticles under the wall of the microsphere. It is inspired that the WGMs may suffer low loss, although the light wave running through the sphere core is strongly absorbed; that is, selective absorption sensitive to the areas in such microspheres can be implemented. Figure 2(a) shows an optical microscope (AxioLab A1, ZEISS, Germany) image of an iron-oxide nanoparticles embedded silica microsphere with a diameter of 40 μm. It was ablated from an 11 μm diameter microfiber. As can be seen, the sphere exhibits a smooth surface. This implies that the sphere is able to construct an optical microresonator. Besides, visible fluorescence was always observed during the fabrication of microspheres. Pumped by a 1550 nm laser via the microfiber stem, visible light is emitted by the sphere, as shown by Fig. 2(b). The pump power was 200 mW. The uniform fluorescence illustrates that the device may find applications in the area of microscope illumination. The corresponding
fluorescence spectrum is depicted by Fig. 2(c), measured by an optical spectrometer (Princeton Instruments SP2750 coupled with a CCD PIXIS 100B). The FWHM of the spectrum is about 230 nm. This wide spectral range implies that the fluorescence is attributed to thermalinduced emission. The WGM in the microsphere was excited with a widely adopted fiber taper coupling approach [25]. The 1550 nm pump was injected into the microsphere via the fiber stem. The fiber taper was in contact with the microsphere. Figure 3(a) shows the optical transmission spectra of an MSR with a diameter of about 22 μm. The blue line is a spectral curve obtained without an optical pump. Transmission extinction ratios up to 13 dB are harvested, with a free spectral range of 27.7 nm for the fundamental WGM. The shallow dips are due to highorder WGM resonances. As can be seen, insertion loss of the MSR is down to about 0.66 dB. The low loss suggests that the nanoparticle concentration of the area under the sphere surface is quite low. The Q-factor of the fundamental WGM is about 103 . It can be improved by optimizing the gap between the taper waist and the sphere [26]. When the pump power was increased to 15 mW, the transmission spectrum (green curve) is red-shifted up by 2.8 nm. An increase of cavity temperature δT results in not only a radial expansion δR of the microsphere but also a change of mode indice δn. Both changes then show the impact on the resonance wavelength λ0 . The variation of resonance wavelength δλ0 of the same resonance order against the temperature change δT can be estimated from [27] 1 ∂n 1 ∂R · · ∂T; n ∂T R ∂T
δλ0 λ0
(1)
where n and R are constant values and are referenced to the room temperature. ∂n∕n · ∂T and ∂R∕R · ∂T are
Fig. 2. Optical microscope images of an iron-oxide nanoparticles embedded silica microsphere (a) without and (b) with pump. (c) Corresponding fluorescence spectrum of the microsphere under the pump.
July 1, 2014 / Vol. 39, No. 13 / OPTICS LETTERS
3847
Fig. 3. (a) Transmission spectra of a MSR. The blue and green curves were recorded under 0 and 15 mW 1550 nm pump, respectively. (b) Local transmission spectra of the MSR near 1540 nm when the pump power (P) continuously increases from 0 to 100 mW. (c) Resonance wavelengths near 1540 nm varying with the pump power. The square-dotted solid and circle-dotted dashed lines are for increasing and decreasing the pump power, respectively.
the change ratio with respect to temperature for the refractive index of silica and thermal expansion of the microsphere. Here ∂n∕n · ∂T is about 1.1 × 10−5 K−1 [28]. It is about two orders of magnitude higher than the thermal expansion coefficient (approximate 5.5 × 10−7 K−1 ). The signs of both coefficients are positive. Hence it can be predicted that the resonance wavelength will be redshifted when the temperature rises, and this is in good agreement with the experimental demonstrations. Moreover, continuous optical tuning of the MSR spectrum was performed as well. The pump power varied from 0 to 100 mW. The power step was about 1 mW in the 0–40 mW region and about 2 mW in the 40–100 mW region. Local transmission spectra near 1540 nm is shown by Fig. 3(b). As can be seen in Fig. 3(b), the spectrum shape is well preserved in the continuous tuning process. The fluctuation in transmission extinction ratio is less than 0.55 dB, corresponding to a relative variation of 4%. During the tuning process, the 3 dB bandwidth fluctuates near 2.43 nm, with a variation of 0.11 nm. The small variations in the MSR parameters imply the good quality of wavelength tuning. Figure 3(c) presents the resonance wavelength as a function of the pump power. The square-dotted solid blue curve is obtained by increasing the pump power. Define the optical tuning efficiency (η) as wavelength change ratio with respect to pump power, i.e., Δλ0 ∕ΔP. As can be seen, η can be nearly 0.2 nm∕mW in the linear region. The linear tuning of over 8 nm (1 THz) is achieved. When the pump power is more than 40 mW, the tuning efficiency gradually drops. This is believed to be due to the fact that the thermo-optical effect tends to be saturated under high temperature [29]. The total range of tuning the resonance wavelength can be over 13 nm (1.6 THz). Besides, shifting the resonance wavelength via decreasing the pump power is presented by the circle-dotted dashed curve, which is consistent with the blue line. Hence the proposed approach of wavelength tuning of the novel resonator demonstrates high robustness, which paves the way for practical applications of the device. In our experiments, it was found that two-stage in-line ablation is more efficient to fabricate microspheres: (1) low power (25–50 mW) was utilized to break the microfiber in the first stage; (2) high power (about 200 mW) was used to further ablate the microfiber end in the following step. Figure 4 shows scanning electron
microscope images of two typical microfiber ends generated with different initial ablation powers: (a) 20 mW and (b) 400 mW. Low-power pump slightly above the ablation threshold facilitates the deposition of nanoparticles along the microfiber. Since the amount of the heat is not large enough to break the microfiber, more and more particles are deposited on the microfiber and improve the optical-to-thermal transfer efficiency. This leads to a microfiber end with large amounts of nanoparticles, as shown in Fig. 4(a). The inset of Fig. 4(a) presents a close-up image of the microfiber end. The nanoparticles are clustered on the end, yielding a rough surface whose quality can be greatly improved by a second laser ablation. Instead, if the initial power is considerably high, huge heat is generated once the light wave enters the magnetic fluid. The microfiber then immediately crack at the air-fluid interface. This leads to a microfiber end with few nanoparticles, as demonstrated in Fig. 4(b). The microfiber end can hardly be fused into a microsphere in the second step even under higher pump power due to limited energy transfer. In summary, iron-oxide nanoparticles embedded MSRs, of which the resonance wavelengths can be broadly tuned using an all-optical approach, are demonstrated. The microsphere is fabricated by magnetic-fluidassisted in-line infrared-laser ablation of silica microfibers and exhibits optical absorption, which is sensitive to the region in the sphere. With an optical pump, a robust wavelength shift over 1.6 THz is realized, corresponding to a linear tuning efficiency up to 0.2 nm∕mW. Owing to the ultrawide absorption spectrum of iron-oxide nanoparticles, it is believed that a light wave with an arbitrary wavelength from ultraviolet radiation to far-infrared rays can be the pump of such a resonator. With advantages of
Fig. 4. Scanning electron microscope images of microfiber ends after first-stage ablation. (a) The initial ablation power was 20 mW. (Inset) Close-up image of the end with a scale bar of 5 μm. (b) The initial ablation power was 400 mW.
3848
OPTICS LETTERS / Vol. 39, No. 13 / July 1, 2014
a small footprint, ease of fabrication, and high flexibility, the proposed resonators can be attractive elements in areas of optical sensing, fiber communications, and long-range distributing tuning. The authors are deeply grateful to Dr. Yongyue Xu from the Laser & Terahertz Technology Division of Wuhan National Laboratory for Optoelectronics for measuring the fluorescence spectrum of microspheres. The work was partially supported by the National Science Foundation for Distinguished Young Scholars of China (grant no. 61125501), the Major International Joint Research Project of the National Science Foundation of China (grant no. 61320106016), the National Natural Science Foundation of China (grant no. 61307075), the Specialized Research Fund for the Doctoral Program of Higher Education of China (grant no. 20120142120067), and the Fundamental Research Funds for the Central Universities (HUST: 2014TS019). References 1. K. J. Vahala, Nature 424, 839 (2003). 2. V. S. Ilchenko and A. B. Matsko, IEEE J. Sel. Top. Quantum Electron. 12, 15 (2006). 3. J. Ward and O. Benson, Laser Photon. Rev. 5, 553 (2011). 4. I. H. Agha, Y. Okawachi, and A. L. Gaeta, Opt. Express 17, 16209 (2009). 5. D. W. Vernooy, A. Furusawa, N. P. Georgiades, V. S. Ilchenko, and H. J. Kimble, Phys. Rev. A 57, R2293 (1998). 6. M. Manzo, T. Ioppolo, U. K. Ayaz, V. LaPenna, and M. V. Otugen, Rev. Sci. Instrum. 83, 105003 (2012). 7. K. Totsuka, N. Kobayashi, and M. Tomita, Phys. Rev. Lett. 98, 213904 (2007). 8. V. S. Ilchenko, X. S. Yao, and L. Maleki, Proc. SPIE 3611, 190 (1999). 9. V. S. Ilchenko, P. S. Volikov, V. L. Velichansky, F. Treussart, V. Lefèvre-Seguin, J. M. Raimond, and S. Haroche, Opt. Commun. 145, 86 (1998).
10. W. von Klitzing, R. Long, V. S. Ilchenko, J. Hare, and V. Lefèvre-Seguin, Opt. Lett. 26, 166 (2001). 11. R. Madugani, Y. Yang, J. M. Ward, J. D. Riordan, S. Coppola, V. Vespini, S. Grilli, A. Finizio, P. Ferraro, and S. N. Chormaic, Opt. Lett. 37, 4762 (2012). 12. T. Ioppolo, U. Ayaz, and M. V. Ötügen, Opt. Express 17, 16465 (2009). 13. H. C. Tapalian, J. P. Laine, and P. A. Lane, IEEE Photon. Technol. Lett. 14, 1118 (2002). 14. J. M. Ward and S. N. Chormaic, Appl. Phys. B 100, 847 (2010). 15. A. Watkins, J. Ward, and S. N. Chormaic, J. Appl. Phys. 51, 052501 (2012). 16. G. Brambilla, V. Finazzi, and D. J. Richardson, Opt. Express 12, 2258 (2004). 17. P. Zhao, J. Zhang, Y. Yu, J. Dong, L. Shi, Y. Liu, and X. Zhang, Opt. Express 21, 8231 (2013). 18. A. Schlegel, S. F. Alvarado, and P. Wachter, J. Phys. C 12, 1157 (1979). 19. S. Pu, X. Chen, C. Yuping, Y. Xu, W. Liao, L. Chen, and Y. Xia, J. Appl. Phys. 99, 093516 (2006). 20. S. Pu, H. Wang, N. Wang, and X. Zeng, Appl. Phys. B 112, 223 (2013). 21. S. Pu, X. Chen, L. Chen, W. Liao, C. Yuping, and Y. Xia, Appl. Phys. Lett. 87, 021901 (2005). 22. H. Wang, S. Pu, N. Wang, S. Dong, and J. Huang, Opt. Lett. 38, 3765 (2013). 23. L. N. Ng, B. J. Luff, M. N. Zervas, and J. S. Wilkinson, Opt. Commun. 208, 117 (2002). 24. J. Tang, M. Myers, K. A. Bosnick, and L. E. Brus, J. Phys. Chem. B 107, 7501 (2003). 25. J. C. Knight, G. Cheung, F. Jacques, and T. A. Birks, Opt. Lett. 22, 1129 (1997). 26. M. Cai, O. Painter, and K. J. Vahala, Phys. Rev. Lett. 85, 74 (2000). 27. Z. Cai, A. Chardon, H. Xu, P. Féron, and G. Michel Stéphan, Opt. Commun. 203, 301 (2002). 28. M. J. Weber, Handbook of Optical Materials (CRC Press, 2002). 29. U. Haken, O. Humbach, S. Ortner, and H. Fabian, J. Non-Cryst. Solids 265, 9 (2000).
3845
Iron-oxide nanoparticles embedded silica microsphere resonator exhibiting broadband all-optical wavelength tunability Ping Zhao,1 Lei Shi,1,* Yang Liu,1 Zheqi Wang,1 Shengli Pu,2 and Xinliang Zhang1,3 1
Wuhan National Laboratory for Optoelectronics & School of Optical and Electronic Information, Huazhong University of Science and Technology, Wuhan 430074, China 2 College of Science, University of Shanghai for Science and Technology, Shanghai 200093, China 3 e-mail: [email protected] *Corresponding author: [email protected]
Received March 27, 2014; revised May 19, 2014; accepted May 25, 2014; posted May 27, 2014 (Doc. ID 208898); published June 23, 2014 In this Letter, a novel silica microsphere resonator (MSR) embedded with iron-oxide nanoparticles, which possesses broadband all-optical wavelength tunability, is demonstrated. It is generated by using in-line 1550 nm laser ablation of a microfiber with the assistance of magnetic fluid. To the best of our knowledge, this simple method of fabricating such MSRs is reported for the first time. Prominent photothermal effect is realized by the iron-oxide nanoparticles absorbing light pumped via the fiber stem, leading to a wavelength shift of over 13 nm (1.6 THz). Moreover, a linear tuning efficiency up to 0.2 nm∕mW is realized. With excellent robustness and being fiberized, the spheres can be attractive elements in building up novel micro-illuminators, point heaters, optical sensors, and fiber communication modules. © 2014 Optical Society of America OCIS codes: (230.5750) Resonators; (190.4870) Photothermal effects; (060.2340) Fiber optics components. http://dx.doi.org/10.1364/OL.39.003845
Due to high-quality factors, optical microsphere resonators (MSRs) have been of great interest in basic research and engineering applications [1–3], including nonlinear optics [4], quantum optics [5], optical sensing [6], optical signal processing [7], and microwave photonics [8]. Wavelength tuning of MSRs is of great importance in enhancing the flexibility of MSRs. Mechanical wavelength tuning of silica microspheres has been previously demonstrated [9–11]. However, the wavelength-tunable range is less than 1 nm, and the pressing of microspheres unavoidably results in noncircle geometry deformations of the microspheres. As a result, contactless methods of changing the resonance wavelength, such as electrical or thermal tuning, were investigated [12–15]. The elastic strain of an MSR arises under the force exerted by an electrostatic field, while the low piezoelectric modulus leads to a wavelength-tuning range of only tens of picometers [12]. On the other hand, thermal wavelength adjustment has a large tuning range [13,15]. The temperature of conventional microelectric heaters can hardly reach a high level in order to induce a prominent wavelength shift of MSRs. Heating with lasers can overcome such a problem and is quite flexible [13]. Tapalian et al. reported thermo-optical modulators using polymer-coated MSRs [13]. Nevertheless, polymer matrix is usually subjected to poor environmental durability. Hence solid glass MSRs with high stabilities are favored. However, the thermo-optical effect of pure silica MSRs is weak. Doping in silica is an effective approach of enhancing this effect. An all-optical thermal wavelength tuning for erbium: ytterbium-doped glass MSRs was demonstrated by Watkins et al. [15]. However, the pump wavelength was limited due to the absorption of such glass. What’s worse, the signal-to-noise ratio of optical signals at 1550 nm may be degraded due to the spontaneousemission noise generated along the pumping processes. 0146-9592/14/133845-04$15.00/0
In this Letter, we propose and demonstrate broadband all-optical wavelength tuning of a novel MSR. The key component is a silica microsphere embedded with ironoxide nanoparticles. These particles can perform highly efficient photon-to-heat conversion. The sphere has a core-shell similar structure: the nanoparticles are mainly concentrated in the core area rather than near the surface. Consequently, the light wave traveling through the core is strongly absorbed, while the signal propagating as whispering-gallery modes (WGMs) experiences little loss. With a 1550 nm pump, a wavelength shift of more than 13 nm (1.6 THz) is achieved in the C-band of optical fiber communications with respect to the fabricated MSR. A broad linear-tuning wavelength range of more than one terahertz is obtained, and the corresponding power tuning efficiency reaches up to 0.2 nm∕mW. The proposed wavelength tuning of the MSRs also exhibits excellent robustness and high flexibility. The silica microsphere embedded with iron-oxide nanoparticles was fabricated by an in-line infrared-laser ablation of optical microfiber surrounded by magnetic fluid. The microfiber was tapered from a standard single-mode fiber (SMF) (YOFC, China) with a flame-heated technique [16,17]. To fabricate microspheres, commercial water-based magnetic fluid (Beijing Sunrise Ferrofluid Tech. Co. Ltd., China) was utilized. Figure 1(d) shows the image of a sample of 2 mL magnetic fluid. It is composed of iron-oxide nanoparticles uniformly dispersed in deionized water. Due to strong wideband optical absorption and magnetic properties [18], this kind of nanoparticle has raised plenty of interest in optics [19–22]. The diameter of the nanoparticle is about 8 nm, and the concentration is 33 w.t. %. Figures 1(a)– 1(c) show the fabrication process of such a microsphere. First, a drop of magnetic fluid was dropped on the end facet of a cleaved glass rod with a diameter of 3 mm. © 2014 Optical Society of America
3846
OPTICS LETTERS / Vol. 39, No. 13 / July 1, 2014
Fig. 1. Fabrication process of microspheres. (a) A microfiber is immerged into a drop of magnetic fluid. SMF, single-mode fiber. (b) The microfiber is melted and snapped under in-line magnetic-fluid-assisted laser ablation. (c) The pump power is increased up to 200 mW. An iron-oxide nanoparticles embedded silica microsphere with a smooth surface is generated after the further ablation. The inset is a scanning electron microscope image of the resulting microsphere. The scale bar is 10 μm. (d) 2 mL magnetic fluid sample.
Due to the surface tension, a liquid semi-sphere with a diameter of about 3 mm was formed as shown by Fig. 1(a). An optical microfiber was then emerged into the liquid drop. The diameter of the microfiber can be from submicrometer to tens of micrometers. Next, a 1550 nm light wave from a semiconductor laser was injected into the microfiber through the SMF after being amplified by an erbium-doped fiber amplifier. The initial ablation power was around 30 mW. The evanescent wave along the microfiber acts as optical tweezers and attracts nanoparticles [23]. These nanoparticles were then attached onto the microfiber surface, strongly absorbing the infrared rays [24]. Simultaneously, prodigious amounts of heat were generated. As presented by Fig. 1(b), the local temperature dramatically increased, resulting in the melting and breakage of the microfiber. Due to the surface tension, the end of the microfiber was quickly fused into a sphere. Some iron-oxide nanoparticles were blended in the melted sphere. Next, we further increase the ablation power up to about 200 mW to generate a smoother microsphere [Fig. 1(c)]. The sphere expanded under the further heating. During the expansion of the sphere, the magnetic nanoparticles moved toward the core area in the sphere. It is believed that the movements are due to the mutual magnetic attraction between these particles. When the heat dissipation and generation were balanced, the sphere geometry was then maintained. After the withdrawal of the optical pump, a smooth microsphere was harvested, as shown by the inset of Fig. 1(c). The nanoparticles were mainly distributed near the core of the sphere. On the contrary, there were few iron-oxide nanoparticles under the wall of the microsphere. It is inspired that the WGMs may suffer low loss, although the light wave running through the sphere core is strongly absorbed; that is, selective absorption sensitive to the areas in such microspheres can be implemented. Figure 2(a) shows an optical microscope (AxioLab A1, ZEISS, Germany) image of an iron-oxide nanoparticles embedded silica microsphere with a diameter of 40 μm. It was ablated from an 11 μm diameter microfiber. As can be seen, the sphere exhibits a smooth surface. This implies that the sphere is able to construct an optical microresonator. Besides, visible fluorescence was always observed during the fabrication of microspheres. Pumped by a 1550 nm laser via the microfiber stem, visible light is emitted by the sphere, as shown by Fig. 2(b). The pump power was 200 mW. The uniform fluorescence illustrates that the device may find applications in the area of microscope illumination. The corresponding
fluorescence spectrum is depicted by Fig. 2(c), measured by an optical spectrometer (Princeton Instruments SP2750 coupled with a CCD PIXIS 100B). The FWHM of the spectrum is about 230 nm. This wide spectral range implies that the fluorescence is attributed to thermalinduced emission. The WGM in the microsphere was excited with a widely adopted fiber taper coupling approach [25]. The 1550 nm pump was injected into the microsphere via the fiber stem. The fiber taper was in contact with the microsphere. Figure 3(a) shows the optical transmission spectra of an MSR with a diameter of about 22 μm. The blue line is a spectral curve obtained without an optical pump. Transmission extinction ratios up to 13 dB are harvested, with a free spectral range of 27.7 nm for the fundamental WGM. The shallow dips are due to highorder WGM resonances. As can be seen, insertion loss of the MSR is down to about 0.66 dB. The low loss suggests that the nanoparticle concentration of the area under the sphere surface is quite low. The Q-factor of the fundamental WGM is about 103 . It can be improved by optimizing the gap between the taper waist and the sphere [26]. When the pump power was increased to 15 mW, the transmission spectrum (green curve) is red-shifted up by 2.8 nm. An increase of cavity temperature δT results in not only a radial expansion δR of the microsphere but also a change of mode indice δn. Both changes then show the impact on the resonance wavelength λ0 . The variation of resonance wavelength δλ0 of the same resonance order against the temperature change δT can be estimated from [27] 1 ∂n 1 ∂R · · ∂T; n ∂T R ∂T
δλ0 λ0
(1)
where n and R are constant values and are referenced to the room temperature. ∂n∕n · ∂T and ∂R∕R · ∂T are
Fig. 2. Optical microscope images of an iron-oxide nanoparticles embedded silica microsphere (a) without and (b) with pump. (c) Corresponding fluorescence spectrum of the microsphere under the pump.
July 1, 2014 / Vol. 39, No. 13 / OPTICS LETTERS
3847
Fig. 3. (a) Transmission spectra of a MSR. The blue and green curves were recorded under 0 and 15 mW 1550 nm pump, respectively. (b) Local transmission spectra of the MSR near 1540 nm when the pump power (P) continuously increases from 0 to 100 mW. (c) Resonance wavelengths near 1540 nm varying with the pump power. The square-dotted solid and circle-dotted dashed lines are for increasing and decreasing the pump power, respectively.
the change ratio with respect to temperature for the refractive index of silica and thermal expansion of the microsphere. Here ∂n∕n · ∂T is about 1.1 × 10−5 K−1 [28]. It is about two orders of magnitude higher than the thermal expansion coefficient (approximate 5.5 × 10−7 K−1 ). The signs of both coefficients are positive. Hence it can be predicted that the resonance wavelength will be redshifted when the temperature rises, and this is in good agreement with the experimental demonstrations. Moreover, continuous optical tuning of the MSR spectrum was performed as well. The pump power varied from 0 to 100 mW. The power step was about 1 mW in the 0–40 mW region and about 2 mW in the 40–100 mW region. Local transmission spectra near 1540 nm is shown by Fig. 3(b). As can be seen in Fig. 3(b), the spectrum shape is well preserved in the continuous tuning process. The fluctuation in transmission extinction ratio is less than 0.55 dB, corresponding to a relative variation of 4%. During the tuning process, the 3 dB bandwidth fluctuates near 2.43 nm, with a variation of 0.11 nm. The small variations in the MSR parameters imply the good quality of wavelength tuning. Figure 3(c) presents the resonance wavelength as a function of the pump power. The square-dotted solid blue curve is obtained by increasing the pump power. Define the optical tuning efficiency (η) as wavelength change ratio with respect to pump power, i.e., Δλ0 ∕ΔP. As can be seen, η can be nearly 0.2 nm∕mW in the linear region. The linear tuning of over 8 nm (1 THz) is achieved. When the pump power is more than 40 mW, the tuning efficiency gradually drops. This is believed to be due to the fact that the thermo-optical effect tends to be saturated under high temperature [29]. The total range of tuning the resonance wavelength can be over 13 nm (1.6 THz). Besides, shifting the resonance wavelength via decreasing the pump power is presented by the circle-dotted dashed curve, which is consistent with the blue line. Hence the proposed approach of wavelength tuning of the novel resonator demonstrates high robustness, which paves the way for practical applications of the device. In our experiments, it was found that two-stage in-line ablation is more efficient to fabricate microspheres: (1) low power (25–50 mW) was utilized to break the microfiber in the first stage; (2) high power (about 200 mW) was used to further ablate the microfiber end in the following step. Figure 4 shows scanning electron
microscope images of two typical microfiber ends generated with different initial ablation powers: (a) 20 mW and (b) 400 mW. Low-power pump slightly above the ablation threshold facilitates the deposition of nanoparticles along the microfiber. Since the amount of the heat is not large enough to break the microfiber, more and more particles are deposited on the microfiber and improve the optical-to-thermal transfer efficiency. This leads to a microfiber end with large amounts of nanoparticles, as shown in Fig. 4(a). The inset of Fig. 4(a) presents a close-up image of the microfiber end. The nanoparticles are clustered on the end, yielding a rough surface whose quality can be greatly improved by a second laser ablation. Instead, if the initial power is considerably high, huge heat is generated once the light wave enters the magnetic fluid. The microfiber then immediately crack at the air-fluid interface. This leads to a microfiber end with few nanoparticles, as demonstrated in Fig. 4(b). The microfiber end can hardly be fused into a microsphere in the second step even under higher pump power due to limited energy transfer. In summary, iron-oxide nanoparticles embedded MSRs, of which the resonance wavelengths can be broadly tuned using an all-optical approach, are demonstrated. The microsphere is fabricated by magnetic-fluidassisted in-line infrared-laser ablation of silica microfibers and exhibits optical absorption, which is sensitive to the region in the sphere. With an optical pump, a robust wavelength shift over 1.6 THz is realized, corresponding to a linear tuning efficiency up to 0.2 nm∕mW. Owing to the ultrawide absorption spectrum of iron-oxide nanoparticles, it is believed that a light wave with an arbitrary wavelength from ultraviolet radiation to far-infrared rays can be the pump of such a resonator. With advantages of
Fig. 4. Scanning electron microscope images of microfiber ends after first-stage ablation. (a) The initial ablation power was 20 mW. (Inset) Close-up image of the end with a scale bar of 5 μm. (b) The initial ablation power was 400 mW.
3848
OPTICS LETTERS / Vol. 39, No. 13 / July 1, 2014
a small footprint, ease of fabrication, and high flexibility, the proposed resonators can be attractive elements in areas of optical sensing, fiber communications, and long-range distributing tuning. The authors are deeply grateful to Dr. Yongyue Xu from the Laser & Terahertz Technology Division of Wuhan National Laboratory for Optoelectronics for measuring the fluorescence spectrum of microspheres. The work was partially supported by the National Science Foundation for Distinguished Young Scholars of China (grant no. 61125501), the Major International Joint Research Project of the National Science Foundation of China (grant no. 61320106016), the National Natural Science Foundation of China (grant no. 61307075), the Specialized Research Fund for the Doctoral Program of Higher Education of China (grant no. 20120142120067), and the Fundamental Research Funds for the Central Universities (HUST: 2014TS019). References 1. K. J. Vahala, Nature 424, 839 (2003). 2. V. S. Ilchenko and A. B. Matsko, IEEE J. Sel. Top. Quantum Electron. 12, 15 (2006). 3. J. Ward and O. Benson, Laser Photon. Rev. 5, 553 (2011). 4. I. H. Agha, Y. Okawachi, and A. L. Gaeta, Opt. Express 17, 16209 (2009). 5. D. W. Vernooy, A. Furusawa, N. P. Georgiades, V. S. Ilchenko, and H. J. Kimble, Phys. Rev. A 57, R2293 (1998). 6. M. Manzo, T. Ioppolo, U. K. Ayaz, V. LaPenna, and M. V. Otugen, Rev. Sci. Instrum. 83, 105003 (2012). 7. K. Totsuka, N. Kobayashi, and M. Tomita, Phys. Rev. Lett. 98, 213904 (2007). 8. V. S. Ilchenko, X. S. Yao, and L. Maleki, Proc. SPIE 3611, 190 (1999). 9. V. S. Ilchenko, P. S. Volikov, V. L. Velichansky, F. Treussart, V. Lefèvre-Seguin, J. M. Raimond, and S. Haroche, Opt. Commun. 145, 86 (1998).
10. W. von Klitzing, R. Long, V. S. Ilchenko, J. Hare, and V. Lefèvre-Seguin, Opt. Lett. 26, 166 (2001). 11. R. Madugani, Y. Yang, J. M. Ward, J. D. Riordan, S. Coppola, V. Vespini, S. Grilli, A. Finizio, P. Ferraro, and S. N. Chormaic, Opt. Lett. 37, 4762 (2012). 12. T. Ioppolo, U. Ayaz, and M. V. Ötügen, Opt. Express 17, 16465 (2009). 13. H. C. Tapalian, J. P. Laine, and P. A. Lane, IEEE Photon. Technol. Lett. 14, 1118 (2002). 14. J. M. Ward and S. N. Chormaic, Appl. Phys. B 100, 847 (2010). 15. A. Watkins, J. Ward, and S. N. Chormaic, J. Appl. Phys. 51, 052501 (2012). 16. G. Brambilla, V. Finazzi, and D. J. Richardson, Opt. Express 12, 2258 (2004). 17. P. Zhao, J. Zhang, Y. Yu, J. Dong, L. Shi, Y. Liu, and X. Zhang, Opt. Express 21, 8231 (2013). 18. A. Schlegel, S. F. Alvarado, and P. Wachter, J. Phys. C 12, 1157 (1979). 19. S. Pu, X. Chen, C. Yuping, Y. Xu, W. Liao, L. Chen, and Y. Xia, J. Appl. Phys. 99, 093516 (2006). 20. S. Pu, H. Wang, N. Wang, and X. Zeng, Appl. Phys. B 112, 223 (2013). 21. S. Pu, X. Chen, L. Chen, W. Liao, C. Yuping, and Y. Xia, Appl. Phys. Lett. 87, 021901 (2005). 22. H. Wang, S. Pu, N. Wang, S. Dong, and J. Huang, Opt. Lett. 38, 3765 (2013). 23. L. N. Ng, B. J. Luff, M. N. Zervas, and J. S. Wilkinson, Opt. Commun. 208, 117 (2002). 24. J. Tang, M. Myers, K. A. Bosnick, and L. E. Brus, J. Phys. Chem. B 107, 7501 (2003). 25. J. C. Knight, G. Cheung, F. Jacques, and T. A. Birks, Opt. Lett. 22, 1129 (1997). 26. M. Cai, O. Painter, and K. J. Vahala, Phys. Rev. Lett. 85, 74 (2000). 27. Z. Cai, A. Chardon, H. Xu, P. Féron, and G. Michel Stéphan, Opt. Commun. 203, 301 (2002). 28. M. J. Weber, Handbook of Optical Materials (CRC Press, 2002). 29. U. Haken, O. Humbach, S. Ortner, and H. Fabian, J. Non-Cryst. Solids 265, 9 (2000).
