Open Access
Add to BibSonomy
Abstract
A new type of tunable broadband fiber-optic acousto-optic sensor was experimentally demonstrated by utilizing a bubble-on-fiber (BoF) interferometer. A single micro-bubble was generated by injecting a heating laser at λ = 980 nm on the metalized facet of an optical fiber. The BoF formed a spherical micro-cavity in water whose acoustic deformation was precisely detected by using a narrowband DFB laser at 1550 nm. The heating light and the interrogating light were fed into a single fiber probe by wavelength division multiplexing (WDM) realizing a small footprint all-fiber configuration. The diameter of the BoF was stabilized with a variation less than 0.5 nm by fast servo-control of the heating laser power. The stabilized BoF served as a Fabry-Pérot cavity that can be deformed by acoustic perturbation, and a minimum detectable pressure level of as low as ~1 mPa/Hz1/2 was achieved in a frequency range of over 60 kHz in water at room temperature. Our proposed BoF technology can provide a tunable, flexible and all-fiber solution to detect minute acoustically driven perturbations combining high-precision interferometry. Due to the very small form-factor, the technique can find applications of liquid-immersible in situ measurements in bio-molecular/cell detection and biochemical phenomena study.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Fiber-optic acoustic sensors (FOASs) have shown rapid development for acoustic/ultrasonic detection in both military and biomedical imaging applications because of their high sensitivity, light weight, immunity to electromagnetic interference (EMI), and the capability of remote detection and multiplexing [1]. In particular, FOASs based on Fabry-Perot (FP) cavities constructed by attaching thin deflectable diaphragms to the fiber tips, have drawn increasing attention because their small form-factor probe structures are well suited for various space-limited applications [2–7]. Diaphragms made of silica [2], silicon [3], polymer [4] and metal [5] have been employed to construct these FP-cavities on optical fiber tips. Recently, few atomic layers of two-dimensional materials such as graphene [6] and MoS2 [7] have been applied to form ultra-thin diaphragms achieving record-breaking sensitivities. However, these nanoscale diaphragms have severely suffered from mechanical damages and irreversible wrinkle formation when immersed in water for acoustic/ultrasonic detection. Sophisticated preparation and transfer of these 2D material films will further hinder practical applications. All of the prior diaphragm FP acoustic sensors do not have spherical symmetry in their structures and have an inherent problem that the sensing could show a directional dependence especially in highly localized environment.
Microbubbles have been extensively exploited in biomedical and microfluidic applications including contrast agents for ultrasound imaging [8], valves in microfluidic chips [9], and particle manipulation [10]. However, prior bubble techniques have lacked stable diameter control schemes and previous studies have mainly focused on the microbubble dynamics including the nucleation, growth, transport and collapse of the microbubbles [11–15]. If stabilized in the diameter and the spatial position, these microbubbles in aqueous environment can show both a high robustness and an intrinsically isotropic sensitivity to acoustic/ultrasonic perturbations. Moreover, the deformations of their air/water interfaces can be directly interrogated by high-accuracy optic interferometry if built at the optical fiber tip.
Here, we proposed a new scheme to utilize intrinsic spherical symmetry on the air-water interface of the single bubble on fiber (BoF), and its high sensitivity in the acoustic sensing was explored by combining all-fiber reflection interferometry, for the first time to the best knowledge of the authors. Note that the proposed BoF is already integrated to the optical fiber obviating all the complex diaphragm fabrication processes, and the bubble diameter can be readily varied by tuning the heating laser power allowing a unique reconfigurable acoustic frequency response, which was not possible in prior fixed diaphragm FP sensors.
2. Experimental results
The configuration of BoF probe is schematically shown in Fig. 1(a). A conventional single mode fiber (SMF) was cleaved at the right angle and its facet was coated with gold (Au) film with a uniform thickness of ~5 nm using a sputtering system. A continuous wave (CW) laser diode (HY-TPLM-980, Hoyatek) at λ = 980 nm was used as a heating laser and the light was delivered to the Au film. The film absorbs the light and generates a localized heat distribution near the core, which vaporizes the surrounding water initiating the bubble nucleation. Gases dissolved in the water then quickly diffuse into it and expand the bubble diameter [14], as shown in Fig. 1(a).
Fig. 1 (a) Schematic of the bubble on fiber (BoF) for acoustic waves detection; (b) Temporal evolution of the bubble growth in BoF for heating light powers of 150 and 200 mW; (c) Decay process of the bubble when the heating laser is switched off.
Temporal evolution of the bubble growth was monitored at various heating light powers and the results are summarized in Fig. 1(b). The bubble radius was directly estimated from the microscope images (inset of Fig. 1(b)). The bubble grew faster at a higher heating power (200 mW) and the bubble radius can be continuously tuned by changing either the laser power or the heating time. When the heating light was switched off at a certain bubble radius, the bubble monotonically shrank and eventually disappeared. For the bubbles with the radii of ~52, 79 and 101 μm, the temporal evolution of the bubble decay was experimentally measured and the results are summarized in Fig. 1(c). The lifetime of the bubble τ, defined as the time during which the bubble fully dissolves, showed a quadratic dependence on the bubble radius r as in the inset of Fig. 1(c), which was consistent to prior reports [16, 17]. Although these bubbles have relatively long lifetimes of order of minutes, a further measure is required to utilize these BoFs for stable acoustic sensing. Based upon bubble growth and decay dynamics in Fig. 1, we tried to search for an optimal heating light power reaching an equilibrium for the gas molecular diffusion through the gas/water interface [14] and eventually stabilizing the bubble diameter for practical sensing applications. Therefore, instead of immediately switching off the laser, we adjusted the power of the heating light at λ = 980 nm using an electronically variable optical attenuator (EVOA), as shown in Fig. 2. The fast response time of 1 ms and sufficiently large power dynamic range of 0 to 100 mW of EVOA (EVOA800A, Thorlabs) enabled the real-time fast servo-control of the laser power to stabilize the bubble diameter. Schematic diagram for the proposed BoF acoustic sensor is shown in Fig. 2. The heating light after passing the EVOA then travelled through the 980/1550nm wavelength division multiplexer (WDM), and reached the fiber tip. Meanwhile, a 1550 nm narrowband interrogation light from a distributed feedback (DFB) laser (BasiK E15, NKT Photonics) with a power of 10 mW was delivered to the bubble through an optical circulator and the WDM. Its reflection from the bubble was detected by a photodiode (PD, 2053-FC-M, Newport).
Fig. 2 Schematic of a servo-control system to stabilize the bubble diameter and the reflection interferometry using a narrowband DFB laser source for acoustic sensing.
The detailed servo-control process is described as follows: the reflection spectrum of a 60 μm-radius bubble (see Fig. 3(a)) will shift with time. This spectrum shift will change the reflected intensity of the interrogation light at 1550 nm and thus the DC voltage output from the PD. Therefore, this DC voltage output can serve as the feedback signal for servo-control of the heating light power to stabilize the bubble. For example, when the reflection spectrum shifts to a short wavelength after 30 second as shown in Fig. 3(a), the PD voltage output decreases. Then the heating light power is increased via the EVOA controlled by a data acquisition (DAQ) device to maintain PD voltage the same as the initial value. The time to complete one cycle of the PID algorithm based servo-control program including reading the DC voltage output from the PD and setting the voltage for the EVOA is estimated to be ~100 ms. This duration is much longer than the EVOA response time and determines the speed of our servo-control scheme. Figure 3(b) compares the DC voltages from the PD for a duration of 30 minutes w/ and w/o servo-control. The standard deviation (σ) of the output PD voltage with the servo-control was ~8 mV, which corresponded to ~0.5 nm variation of the bubble diameter. Further improvement of the bubble diameter stability might be achieved by servo-controlling the duty cycle of a pulsed 980 nm laser, instead of the CW laser power. It needs to point out that this servo-control process cannot only stabilize the bubble, but also reduce the low-frequency environmental perturbations to the bubble radius such as sporadic temperature or pressure fluctuations.
Fig. 3 (a) Reflection spectrums of a bubble over 1 minute w/o servo-control; (b) The output voltages of the bubble interrogated by a 1550 nm laser light w/ and w/o servo-control for 30 minutes.
The reflection spectrum was measured by delivering a broadband incoherent light (1525-1575 nm) to the bubble and the periodic fringes results from the inference of the reflected lights between the fiber-bubble interface and bubble-water interface [6]. Due to the low reflectivity at the air/water interface (∼2% at ∼1550 nm) and the transmission loss of light when travelling through the bubble [6], the reflection spectra exhibit a relatively low level of intensity. Thermal effects might also degrade the adhesion of the Au film to the fiber end and cause further reduced intensity of the reflected light. Therefore, the bubble can be considered as a low-finesse Fabry-Perot cavity, and its radius r of the bubble can be estimated from the relationship r = λ1λ2/4nair(λ2-λ1), where λ1 and λ2 are the adjacent peak/valley wavelengths on the reflection spectra, and nair is the refractive index of the air [6].
3. Acoustic response
The stabilized BoF was then utilized in the acoustic signal detection as shown in the acoustic testing part in Fig. 2. The BoF and a calibration hydrophone (B&K8104) were placed at two symmetry points to the central axis of the speaker in a water tank. The acoustic waves were generated by a loudspeaker driven by a signal generator and calibrated by the hydrophone. The electrical signals from the hydrophone were amplified by a charge amplifier and then received by an oscilloscope. The AC voltages from the PD or equivalently the acoustic response of the bubble were measured by a signal analyzer (MS2692A, Anritsu). To maximize the acoustic sensitivity, the maximum-slope point of the bubble reflection spectrum was matched to the wavelength (1550 nm) of the interrogation light, by adjusting the heating light power to tune the bubble radius, which obviated an expensive tunable DFB laser used in previous research [18].
Figure 4(a) shows the PD output voltage signals from a 39 μm-radius BoF when subjected to a 46 kHz sinusoidal acoustic wave with various pressure amplitudes. The BoF exhibited a linear response to the pressure up to 137 Pa, which is the maximum output from the speaker. Assuming that the gas in the bubble behaves polytropically, the product is a constant, where P is the pressure inside the bubble, V is the volume of the bubble, and κ is the polytropic coefficient [19]. If the acoustic pressure Pa is negligible compared to the pressure inside the bubble P, the pressure sensitivity S, defined as the ratio of the bubble diameter change to the acoustic pressure, can then be derived as, S = 2Δr/Pa = 2r/3κP. For a bubble with a radius of 39 μm, the pressure P inside the bubble can be estimated by the formula P = PL + 2σs/r ≈1.03 × 105 Pa, where σs = 0.072 N/m is the air-water surface tension at room temperature, and PL is the pressure in the liquid outside the bubble approximate to the atmospheric pressure P0 ≈1.01 × 105 Pa. If adiabatic condition is assumed, the polytropic coefficient κ is estimated to be 1.4, subsequently the pressure sensitivity to be~0.2 nm/Pa, which is equivalent to the highest reported value for a diaphragm with the same radius as the bubble [3]. It needs to mention that the reflectivity change of the bubble air/water interface, caused by acoustically modulated gas and water refractive index, can also contribute to the pressure sensitivity of the BoF. This effect was examined by theoretical calculation using Fresnel equation and the coefficients accounting for pressure-induced refractive index change of air and water [20, 21], and is negligible to the bubble sensitivity.
Fig. 4 (a) Output voltage from PD as a function of the acoustic pressure. Inset: time-domain signals from the hydrophone and BoF when subjected to a 46 kHz and 100 Pa acoustic wave. (b) Measured and fitted frequency responses in the frequency range from 2 to 98 kHz for a 39 μm-radius BoF. Inset: the frequency spectrum corresponds to a 72 kHz and 9 Pa acoustic wave.
The frequency response of the BoF measured by varying the driving frequency of the loudspeaker from 2 kHz to 98 kHz is shown in Fig. 4(b). For a 39 μm-radius bubble, the noise equivalent pressure (NEP) of the sensor at the frequency of 72 kHz is estimated to be 290 μPa/Hz1/2 (see inset of Fig. 4(b)). In the flat frequency range below 60 kHz, the averaged NEP is ~1 mPa/Hz1/2. The NEP in the test is limited by the laser power dependent shot/intensity noise [22]. The heat induced by the water absorption of the 1550 nm laser light was found to have negligible effect on the noise level. This was verified by injecting light from an additional 1555 nm laser to purely heat the water surrounding the bubble. To avoid the interference between the 1550 nm and 1555 nm lights, a narrow band-pass filter centered at 1550 nm was used before the reflected lights from the BoF being detected by the PD. The water absorption induced heat, however, causes the red-shift of the BoF reflection spectrum, which can be compensated by the increase of the sever-controlled 980 nm heating power. By modelling the pulsating bubble as a forced and damped linear oscillator, the amplitude of the bubble radius oscillation Δr(f) to a sinusoidal acoustic pressure wave can be described by the following equation [19],
where f is the frequency of the acoustic pressure wave, r0 is the equilibrium radius of the pulsating bubble, ρ is the density of water, δtot is the total dimensionless damping coefficient, and f0 is the fundamental resonant frequency. For large bubble where the surface tension can be neglected, f0 = (3κPL/ρ)1/2/2πr0. If the bubbles is under hydrostatic pressure PL equal to the atmospheric pressure P0, a simplified form as f0·r0 ≈ 3 m·Hz can be obtained. Accordingly, the theoretical resonant frequency (f0) of the 39 μm-radius bubble is ~ 76.9 kHz, showing a good agreement with experiments in Fig. 4(b). The total dimensionless damping coefficient δtot can be approximated by,where the right-hand three terms account for the contributions to the damping due to the thermal conductivity, the acoustic radiation and the liquid viscosity, respectively. γ is heat capacity ratio of air, g is the dimensionless multiplicative constant which accounts for the effects of surface tension on the bubble stiffness, and c and η are the acoustic wave speed and the dynamic viscosity of water, respectively. The parameter Gth is equal to 3γPL/4πρDair, where Dair is the thermal diffusivity of air. For the bubble with a radius r0 of several tens of micrometers, g ≈1. At room temperature, γ ≈1.4, c ≈1500 m/s, η ≈0.9 cP, and Dair ≈1.9 × 10−5 m2s−1. The total damping constant δtot for a pulsating bubble with an equilibrium radius r0 of 39 μm can then be calculated and is equal to ~0.12, mainly contributed by the damping due to the thermal conductivity. Based on Eq. (1), the frequency response of the bubble S(f) can be described by S(f) = K × Δr(f)/Pa, where K is the ratio of the PD output voltage to the bubble radius change, and depends on multiple parameters including the laser power, the fringe contrast of the bubble interference spectrum as well as the responsibility and the gain setting of the PD [6]. Here, the value of K is obtained by fitting the measured data with the calculated Δr(f)/Pa based on the least square method. The calculated frequency response S(f) is plotted in Fig. 4(b) and agrees with the measured data. The discrepancy between the measured data and the fitted curve may be caused by the fluctuations of the measured data due to the interference of the acoustic waves reflected/scattered from the container wall and the hydrophone. The interference may be diminished using a better-designed cylindrical water container for acoustic test [3]. On the other hand, Eq. (1) describes the frequency response of freely oscillating bubbles. The generated bubbles are actually attached to the fiber end, which might affect the oscillating behavior and cause the discrepancy. Further investigation regarding the effect of the fiber end on the bubble acoustic response will be conducted in the future.We further measured the frequency responses of two BoFs with the radii of 23 and 74 μm as shown in Fig. 5(a). During the test, we firstly measured the frequency response of the 23 μm-radius BoF, and then simply raised the heating light power to obtain the 74 μm-radius bubble. Therefore, the frequency bandwidth of BoF can be photothermally reconfigured onsite. To demonstrate the broadband tunability of the proposed BoF, the resonant frequencies of bubbles with various radii are summarized in Fig. 5(b). Despite that a 4 μm-radius bubble can be well stabilized, only resonant frequencies of bubbles with the radius larger than 30 μm was acquired due to the limited bandwidth (~100 kHz) of the hydrophone. Nevertheless, the good agreement between the measured and theoretical resonant frequencies (see Fig. 5(b)) indicates the potential of this bandwidth-tunable BoF for various applications. More specifically, a 50 μm-radius BoF with a resonant frequency of 60 kHz may be used for underwater acoustic detection in 3 kHz to 50 kHz frequency range [3], while a 5 μm-radius BoF with a 0.6 MHz resonant frequency may be used for ultrasonic or photoacoustic imaging [23] using one fiber optic probe.
Fig. 5 (a) Frequency responses of bubbles with radii of 74 μm and 23 μm ; (b) Resonant frequency as a function of the bubble radius.
In obtaining the results shown in Figs. 4 and 5(a), the bubble was aligned to the central axis of the speaker for maximum sensitivity. Figure 6(a) shows the normalized acoustic sensitivities of BoF for various alignment angles. The tested bubbles had the radii of 29 and 56 μm and the acoustic pressure waves had a frequency of 40 kHz. Due to the spherical symmetry, the sensor shows a nearly isotropic response, enabling the probe to be oriented arbitrarily for practical acoustic sensing.
Fig. 6 (a) Normalized acoustic sensitivity of BoFs for acoustic pressure from various angles. Inset: schematic showing the alignment angle between the bubble and the speaker; (b) The output DC voltages from the PD for the bubble stabilized by servo-control in the distilled water, 0.5 wt% and 1.0 wt% NaCl solutions for 30 minutes. The curves were vertically offset for clarity.
It needs to be mentioned that the light power used for the bubble generation and stabilization was liquid-dependent. In addition to the distilled water, we attempted sodium chloride (NaCl) aqueous solutions. The concentrations of the NaCl solutions were 0.5 wt% and 1.0 wt% and Fig. 6(b) indicates that the stabilized bubble can work in liquids with various physical nature, which is being investigated by the authors.
4. Conclusion
In summary, we demonstrated a built-in tunable broadband acousto-optic sensor for liquid-immersible in situ measurements of acoustic waves. With fast servo-control of the heating laser power, the diameter of BOF was stabilized with a variation less than 0.5 nm. The stabilized BOF demonstrated a minimum detectable pressure level of as low as ~1 mPa/Hz1/2 in a frequency range of over 60 kHz in water at room temperature. By tuning the heating laser power, the BoF diameter can be readily varied allowing a dynamically reconfigurable frequency response. Integrated with high-accuracy and compact fiber-optic interferometry, this stable BoF, together with advantages of easy fabrication, low cost, compact size, and reconfigurable feature, is promising as an ultrasensitive platform for a number of applications [23–26], such as ultrasound/photoacoustic imaging, and bio-molecular/cell detection and biochemical phenomena study.
Funding
National Science Foundation of China (No. 61705082); Natural Science Foundation of Guangdong Province (No. 2017A030313361).
Acknowledgment
We thank Prof. Wei Jin in the Department of Electrical Engineering, The Hong Kong Polytechnic University and Prof. Kai Chen in the Institute of Photonics Technology, Jinan University for their proof reading of the manuscript.
References and links
1. C. K. Kirkendall and A. Dandridge, “Overview of high performance fibre-optic sensing,” J. Phys. D Appl. Phys. 37(18), R197–R216 (2004). [CrossRef]
2. B. Yu, D. W. Kim, J. Deng, H. Xiao, and A. Wang, “Fiber Fabry-Perot sensors for detection of partial discharges in power transformers,” Appl. Opt. 42(16), 3241–3250 (2003). [CrossRef] [PubMed]
3. O. Kilic, M. J. F. Digonnet, G. S. Kino, and O. Solgaard, “Miniature photonic-crystal hydrophone optimized for ocean acoustics,” J. Acoust. Soc. Am. 129(4), 1837–1850 (2011). [CrossRef] [PubMed]
4. L. Chen, C. Chan, X. Ang, W. Yuan, P. Zu, W. Wong, Y. Zhang, and K. C. Leong, “Miniature in vivo chitosan diaphragm-based fiber-optic ultrasound sensor,” IEEE J. Sel. Top. Quantum Electron. 18(3), 1042–1049 (2012). [CrossRef] [PubMed]
5. F. Xu, J. Shi, K. Gong, H. Li, R. Hui, and B. Yu, “Fiber-optic acoustic pressure sensor based on large-area nanolayer silver diaghragm,” Opt. Lett. 39(10), 2838–2840 (2014). [CrossRef] [PubMed]
6. J. Ma, H. Xuan, H. L. Ho, W. Jin, Y. Yang, and S. Fan, “Fiber-optic Fabry-Pérot acoustic sensor with multilayer graphene diaphragm,” IEEE Photonics Technol. Lett. 25(10), 932–935 (2013). [CrossRef]
7. F. Yu, Q. Liu, X. Gan, M. Hu, T. Zhang, C. Li, F. Kang, M. Terrones, and R. Lv, “Ultrasensitive pressure detection of few-layer MoS2,” Adv. Mater. 29(4), 1603266 (2017). [CrossRef] [PubMed]
8. C. Errico, J. Pierre, S. Pezet, Y. Desailly, Z. Lenkei, O. Couture, and M. Tanter, “Ultrafast ultrasound localization microscopy for deep super-resolution vascular imaging,” Nature 527(7579), 499–502 (2015). [CrossRef] [PubMed]
9. K. Zhang, A. Jian, X. Zhang, Y. Wang, Z. Li, and H. Y. Tam, “Laser-induced thermal bubbles for microfluidic applications,” Lab Chip 11(7), 1389–1395 (2011). [CrossRef] [PubMed]
10. R. Taylor and C. Hnatovsky, “Trapping and mixing of particles in water using a microbubble attached to an NSOM fiber probe,” Opt. Express 12(5), 916–928 (2004). [CrossRef] [PubMed]
11. X. Xing, J. Zheng, C. Sun, F. Li, D. Zhu, L. Lei, X. Cai, and T. Wu, “Graphene oxide-deposited microfiber: a new photothermal device for various microbubble generation,” Opt. Express 21(26), 31862–31871 (2013). [CrossRef] [PubMed]
12. C. Zhang, Y. Gong, W. Zou, Y. Wu, Y. Rao, G. Peng, and X. Fan, “Microbubble-Based fiber optofluidic interferometer for sensing,” J. Lightwave Technol. 35(13), 2514–2519 (2017). [CrossRef]
13. C. D. Ohl, M. Arora, R. Dijkink, V. Janve, and D. Lohse, “Surface cleaning from laser-induced cavitation bubbles,” Appl. Phys. Lett. 89(7), 074102 (2006). [CrossRef]
14. G. Baffou, J. Polleux, H. Rigneault, and S. Monneret, “Super-heating and micro-bubble generation around plasmonic nanoparticles under cw illumination,” J. Phys. Chem. C 118(9), 4890–4898 (2014). [CrossRef]
15. X. Xing, J. Zheng, F. Li, C. Sun, X. Cai, D. Zhu, L. Lei, T. Wu, B. Zhou, J. Evans, and Z. Chen, “Dynamic behaviors of approximately ellipsoidal microbubbles photothermally generated by a graphene oxide-microheater,” Sci. Rep. 4, 6086 (2006). [CrossRef] [PubMed]
16. R. S. Taylor and C. Hnatovsky, “Growth and decay dynamics of a stable microbubble produced at the end of a near-field scanning optical microscopy fiber probe,” J. Appl. Phys. 95(12), 8444–8449 (2004). [CrossRef]
17. P. S. Epstein and M. S. Plesset, “On the stablity of gas bubble in liquid-gas solutions,” J. Chem. Phys. 18(11), 1505–1509 (1950). [CrossRef]
18. X. Mao, X. Zhou, and Q. Yu, “Stabilizing operation point technique based on the tunable distributed feedback laser for interferometric sensors,” Opt. Commun. 361, 17–20 (2016). [CrossRef]
19. T. G. Leighton, The Acoustic Bubble (Academic, 1994).
20. M. Ranusawud, P. Limsuwan, T. Somthong, and K. Vacharanukul, “Effects of the environment on refractive index of air in long gauge block interferometer,” Precis. Eng. 37, 782–786 (2013). [CrossRef]
21. R. L. Phillips, “Proposed fiber-optic acoustical probe,” Opt. Lett. 5(7), 318–320 (1980). [CrossRef] [PubMed]
22. A. Dandridge, A. B. Tveten, R. O. Miles, and T. G. Giallorenzi, “Laser noise in fiber-optic interferometer systems,” Appl. Phys. Lett. 37(6), 526–528 (1980). [CrossRef]
23. D. Liu, Y. Liang, L. Jin, H. Sun, L. Cheng, and B. O. Guan, “Highly sensitive fiber laser ultrasound hydrophones for sensing and imaging applications,” Opt. Lett. 41(19), 4530–4533 (2016). [CrossRef] [PubMed]
24. X. Jiang, L. Shao, S.-X. Zhang, X. Yi, J. Wiersig, L. Wang, Q. Gong, M. Lončar, L. Yang, and Y.-F. Xiao, “Chaos-assisted broadband momentum transformation in optical microresonators,” Science 358(6361), 344–347 (2017). [CrossRef] [PubMed]
25. J. Zhu, S. K. Ozdemir, Y.-F. Xiao, L. Li, L. He, D.-R. Chen, and L. Yang, “On-chip single nanoparticle detection and sizing by mode splitting in an ultrahigh-Q microresonator,” Nat. Photonics 4, 46–49 (2010). [CrossRef]
26. S. H. Huang, S. Sheth, E. Jain, X. Jiang, S. P. Zustiak, and L. Yang, “Whispering gallery mode resonator sensor for in situ measurements of hydrogel gelation,” Opt. Express 26(1), 51–62 (2018). [CrossRef] [PubMed]
