Open Access
Add to BibSonomy
Abstract
We propose an effective method for biomolecular detection based on an external referencing optofluidic microbubble resonator system (EROMBRS), which possesses good long-term stability and low noise. In this study, EROMBRSs were used for nonspecific detection of bovine serum albumin (BSA) molecules and specific detection of D-biotin molecules. Ultra-low practical detection limits of 1 fg/mL for nonspecific and specific biomolecular detection were achieved.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Whispering gallery mode (WGM) resonators, which possess high quality (Q) factors and small mode volumes, have been widely applied in many scientific and industrial fields [1–6]. WGM resonance arises from total internal reflection of light propagating along the circular surface of the resonator. Any change in the surrounding environment of the resonator will cause a WGM resonance shift [7,8]. Therefore, WGM resonators have great potential for applications in physical and biological/chemical sensing [9–14]. Several types of WGM resonators exist, including microdisks, microspheres, microbubbles, and microrings [15–21]. Of these types, hollow-core microcapillary or microbubble resonators provide excellent optofluidic platforms for biomolecular detection in aqueous media [18–22]. Currently, microcapillary and microbubble resonators can detect biomolecules at the levels of several fg/mL and pM [22–24]. One study detected a low concentration of BSA biomolecules of 10 fg/mL by using a self-referenced microbottle [22]. A single partially packaged microbubble can detect D-biotin at a concentration of 100 fg/mL (0.41 pM) [24]. However, system instabilities caused by environmental disturbances deteriorate the effective detection limits in highly sensitive biomolecular detection [25–27]. Several effective techniques (e.g., using a self-reference resonator) have been proposed to suppress environmental noise [22,28]. Nevertheless, because of the small spacing between splitting modes (typically several picometers), the detection range of the self-referencing system is limited. In addition, the long-term stability of a whole packaged optofluidic microbubble resonator is affected by the fluctuation of the tunable laser source.
In this study, we present an effective external referencing optofluidic microbubble resonator system (EROMBRS) to detect biomolecules. In this sensing scheme, sealed whole packaged optofluidic microbubble resonators (OMBRs) were fabricated by fixing the coupling position between the fiber taper and microbubble. An external referencing OMBR was added to the detection system that monitors fluctuations in the surrounding environment and the stability of the tunable laser source. The final sensing data were corrected with data from the external referencing OMBR, effectively suppressing the noise. We utilized EROMBRS for nonspecific detection of bovine serum albumin (BSA) and specific detection of small biomolecules (< 1000 Da), such as D-biotin, at ultralow concentrations. We determined the concentrations of biomolecules at the fg/mL level, which is far below the minimum concentration required for disease diagnostics.
2. Fabrication and experimental setup
A miniature quartz tube (TSP075150, Polymicro Technologies) was blown to form a spherical microbubble in a fiber fusion splicer (FSU-975, Ericsson, Sweden) with an air injector. The experimental setup, which is illustrated in Fig. 1(a), consisted of a tunable laser (TLB6700, 838-853 nm, New Focus), a polarization controller (FPC023, Thorlabs), two whole packaged OMBRs, two photodetectors (DET10C and APD430A/M, Thorlabs), a camera (BC2000, Bocheng, China), a computer and a data acquisition card (PCIe 6351, National Instruments). The fiber taper was suspended on a glass scaffold and placed perpendicular to the microbubble. The geometric size of the microbubble and the fiber taper shown in Fig. 1(b) was characterized by the following parameters: microbubble outer diameter D = 300 µm; wall thickness t = 3 µm; and fiber taper diameter d = 2 µm. A schematic of the interaction between light and bound D-biotin molecules on the inner surface of the microbubble is presented in Fig. 1(c). The spectral signals were collected by the photodetectors and recorded by the acquisition card. The computer was used to locate the positions (minima) of the spectra after Lorentzian fitting to determine the exact resonance wavelengths of the WGMs. Figure 1(d) displays a transmission spectrum of an OMBR when the microbubble was empty, and the inset shows one of the WGMs with a Q factor of 1.8 × 106. The Q factor decreased to 1.1 × 106 when the microbubble was filled with deionized water (DI water), as shown in Fig. 1(e).
Fig. 1 Experimental setup of the EROMBRS. (a) Experimental instruments and optical path setup. (b) The fused fiber taper was coupled at the equatorial surface of the microbubble photographed with a vertical camera. The diameter of the microbubble was 300 µm and had a wall thickness of 3 µm. (c) Schematic of the interaction between light and bound D-biotin molecules in the microbubble interior. (d) and (e) are fine scans of the transmission spectra at approximately 850 nm when the microbubble was empty or filled with deionized water, respectively. The Q factor was 1.8 × 106 and 1.1 × 106 in the empty and water-filled OMBR, respectively.
3. Results and discussion
3.1 Performance test of the EROMBRS
Figures 2(a) and 2(b) show schematics of the partially and whole packaged OMBRs, respectively. The gray piece that fixes the microbubble on the glass scaffold was UV glue (NOA68, Norland), and the cerulean piece that wraps the fiber taper was low-index UV glue (My 133, n = 1.33, My Polymer, Israel). The microbubble and tapered fiber were in complete contact and fixed to the glass scaffold. For the partially packaged OMBR, the fiber taper was simply fixed to the groove of the glass scaffold, which keeps the position of the fiber taper relatively static. The whole packaged OMBR was wrapped with low-index UV glue. Consequently, the position between the microbubble and fiber taper was immobilized so that the OMBR was more immune to environmental disturbances. Scattering loss and absorption occurs as a result of the coating of low-index UV glue around the ends of the fiber taper, which in turn causes the intensity of the transmission light after packaging to diminish [24,29]. To measure the long-term stability in a conventional manner, a tunable laser was scanned while continuously monitoring the wavelength position of the WGM. Figure 2(c) presents the wavelength shifts of the different OMBR systems in real time without using the external referencing OMBR. The red line in Fig. 2(c) presents the drift of the sensing system in real time when the microbubble and fiber taper were not packaged. By contrast, the modes of the partially (blue line) and whole packaged (orange line) OMBRs did not shift substantially over time. All stability test experiments were conducted at room temperature (approximately 22°C and 40% humidity). All the OMBRs was filled with water. The standard deviations of the nonpackaged, partially packaged, and whole packaged OMBRs were 0.516, 0.158, and 0.095 pm, respectively. These data revealed that the whole packaged OMBR systems have lower noise. With the whole packaged technology, an OMBR sample, for example, can be packaged into a Petri dish and transported to other laboratories for measurements. In the following experiments, the whole packaged OMBRs are utilized for measurement unless otherwise specified.
Fig. 2 Schematic of an OMBR and stability of the EROMBRS. (a) Partially packaged OMBR and (b) whole packaged OMBR. The gray and cerulean pieces are UV and low-index UV glues, respectively. (c) Wavelength shifts of the sensing systems for the nonpackaged OMBR (red line), partially packaged OMBR (blue line), and whole packaged OMBR (orange line) during a detection period of up to 2500 s. (d) Optical transmission spectra of the test and external referencing OMBRs in biomolecular detection of D-biotin. Positions 1 and 2 correspond to the WGMs that we tracked. (e) The mode of the test (pink line) and external referencing (violet line) resonators as a function of time, and the result after correction (green line). Inset: Wavelength shifts caused by the specific binding of D-biotin solutions. Parts E, G, and I are the cases of PBS rinsing. Parts F and H are two consecutive measurements at two bulk concentrations of 10 and 100 fg/mL, respectively.
Despite the previous results, correcting the wavelength shift caused by fluctuations of the tunable laser source [Part 1 in Fig. 2(c)] and wavelength perturbations caused by the surrounding environment [Part 2 in Fig. 2(c)] is difficult. Figures 2(d) and 2(e) show the data from a typical biomolecular detection of D-biotin based on the EROMBRS shown in Fig. 1(a). Both OMBRs were extracted at the same flow rate to overcome the cross sensitivity of the liquid flow rate and pressure change caused by pump extraction. Optical transmission spectra of the test and external referencing OMBRs at approximately 850 nm are presented in Fig. 2(d). Positions 1 and 2 were utilized to locate the positions (minima) of the WGMs after Lorentzian fitting. The distance between Positions 1 and 2 was approximately 11 pm. Note that obtaining two identical OMBRs experimentally is difficult. Differences in the wall thickness and radius of an OMBR affect the transmission spectrum. The OMBRs are also very sensitive to pressure changes, caused by pump extraction, stress-induced refractive index change, and strain-induced size expansion resulting from pressure change [30]. OMBRs with different sizes and walls possess different pressure sensitivities. We chose a reference OMBR with a size and wall as close as possible to those of the test OMBR to ensure that the signals generated by the pressure changes of the two OMBRs were consistent. Another factor determining whether two OMBRs can be used for test and external referencing OMBRs is the bulk refractive index sensitivities (BRISs) of the OMBRs. BRISs have similar values indicate that the WGM (dip) from the test OMBR responses to the change are similar to the WGM (dip) of the external referencing OMBR.
As shown in Fig. 2(e), the wavelength blueshift caused by fluctuations of the tunable laser source (Part A, cyan region) was detected both in the test (pink line) and external referencing (violet line) OMBRs. In Part B (blue region), the test OMBR detected the effective signal derived from the binding of D-biotin. The incidental mode laser hopping (Part C, pink region) resulted in simultaneous WGM resonance shifts both in the test and referencing OMBRs. Part D (yellow region) shows the sensing signal from the dissociation phases of D-biotin. The green line shows the data corrected by the test OMBR data subtracted from those external referencing OMBR. Clearly, the corrected signal is immune to the fluctuations and incidental mode laser hopping of the laser source. The standard deviations of the test OMBR, external referencing OMBR and corrected data were 0.749, 0.595, and 0.144 pm, respectively. Therefore, the EROMBRS possessed good long-term stability and low noise. The insert in Fig. 2(e) shows two consecutive measurements at the two bulk concentrations of 10 and 100 fg/mL. This verified the reproducibility of the sensor system. Note that bioactive binding is saturable and reversible and is related to the probe concentration. In the dissociation phase, the analyte solution was replaced with a phosphate buffered saline (PBS), which decreased the analyte concentration. Then, some biomolecules were dissociated from the immobilized targets to achieve a new dynamic equilibrium state, which corresponds to a decline in the binding curves.
3.2 Optical characterization of the EROMBRS
The BRIS of the sensor is characterized by monitoring the WGM spectral shift in response to a change in the refractive indices of solutions, such as, dimethyl sulfoxide (DMSO)-water mixtures. The BRIS characterization indicates the refractometric sensing ability of the EROMBRS. DMSO solutions were diluted using DI water to 1%, 2%, 3%, 4%, and 5%, respectively. The refractive indices of DMSO-water mixtures were proportional to the volume ratio of DMSO due to the Lorentz-Lorenz equation, as shown in Fig. 3(a) [31]. The DI-water and the DMSO-water mixtures were placed at room temperature for an extended period prior to use to minimize temperature-induced WGM changes. Before the DMSO-water mixtures were extracted, a stable detection baseline of the DI water was first established to ensure that the microbubble filled with DI water reached thermal equilibrium.
Fig. 3 Simulated and experimental BRISs of the EROMBRSs. (a) Relationship between the refractive index of the DMSO-water mixtures and DMSO (volume fraction). (b) Ladder-like wavelength shift sensorgram when DI water, 1-5% DMSO-water mixtures and DI water again were extracted into the microbubble. (c) Wavelength shifts as a function of refractive index. The experimental BRIS of the whole packaged OMBR was approximately SExp = 11.39 nm/RIU. The simulated BRIS with a third-order radial mode was SSim = 11.16 nm/RIU. (d) Normalized electric field intensity distributions of the WGMs with different radial orders.
Upon extraction, the spectral position of the WGM shifted gradually in response to the homogeneous refractive index change. After the shift in the WGM was tracked, a ladder-like sensorgram was obtained, as shown in Fig. 3(b). Six segments with approximately horizontal lines existed that corresponded to DI water, 1–5% DMSO-water mixtures and DI water. Because of the spherical structure of the microbubble, a small amount of the high-refractive-index DMSO-water mixture may not be completely rinsed with DI water in a short time. This will delay the return of the wavelength shift to the baseline. The wavelength shifts of the WGMs were proportional to the refractive index of the DMSO-water mixtures in Fig. 3(c). A change in the refractive index (δn) of the surrounding environment of the WGM can result in a shift of the resonant wavelength (δλ), which can be expressed as:
where S denotes the BRIS of the EROMBRS [32,33]. In our study, the BRIS was SExp = 11.39 nm/RIU based on a linear fit from an actual measurement. The standard deviation was σ = 0.21 pm, and the theoretical detection limit (DL) was estimated as DL = 3σ/SDMSO = 5.5 × 10−5 RIU.We employed the commercial finite element method software (COMSOL Multiphysics 5.2a) to conduct a simulation of the whole packaged OMBR at a resonance wavelength of approximately 850 nm. In this case, the OMBR had a diameter of 300 μm and a wall thickness of 3.2 μm, which was a similar in size to that of the experimental resonator. The BRIS values of the fundamental, second-order radial, and third-order radial modes are approximately 0.23, 3.86, and 11.16 nm/RIU, respectively. Thus, we identified the mode of the experimental OMBR as a third-order radial mode. When the refractive index was larger than 1.34, the WGM continued to redshift with an increase in the refractive index. The normalized electric field intensity distributions of the WGMs for the fundamental, second-order radial and third-order radial modes are presented in Fig. 3(d). The detection range of our system was determined by the fine scanning range of the tunable laser, which was ± 30 GHz (approximately 144 pm) and was much larger than that of self-reference mode splitting (only a dozen pm). This also indicated that the choice of exciting higher order radial WGMs could achieve higher sensitivities and better detection limits.
3.3 Nonspecific detection of BSA
When the WGM interacts with an ambient medium, the spectral position will shift because of the change in the external refractive index on the sensing surface. In addition, when molecules or nanoparticles are captured or attached to the sensing region of the inner surface of the microcavity, the immobilized analytes form a (possibly nonhomogeneous) layer on the biosensor surface. The dielectric constant of this “biological layer” is slightly higher than that of the analyte solution. This in turn affects the WGMs and can be monitored for biosensing [34,35].
In our study, conventional silanization of the microbubble inner surface was achieved by using 3-glycidoxypropyltrimethoxysilan (GOPTS, Sinopharm Chemical Regent Company, China) solution [24,36,37]. The functionalized microbubbles formed epoxy groups on the inner surface, which could form covalent bonds with the amino groups of proteins. All operations are shown in Fig. 4(a). Here, bovine serum albumin (BSA, A1933-5G, Sigma Aldrich) solutions were diluted using PBS (diluted to 1 × , P5493-1L, Sigma Aldrich) to 1, 10, and 100 fg/mL, 1, 10, and 100 pg/mL and to 1, 10, and 100 ng/mL, respectively. A stable detection baseline was established first by extracting PBS for 30 min. In this study, nonspecific binding of BSA molecules to surface-immobilized GOPTS caused a redshift of the WGM, as shown in Fig. 4(b). Real-time data were recorded from the beginning of the baseline to the end of the dissociation phase with a time resolution of 1 s. We continuously added different concentrations of BSA solution. Each BSA solution was extracted at a constant speed of 2 µL/min for 400 s. Figure 4(b) shows a result of real-time BSA measurements at different concentrations, where a relatively saturated concentration was obtained at 100 ng/mL. Combining GOPTS and BSA is accomplished by amine-based adsorption. Many free amine groups are present on the surface of BSA, which can quickly be adsorbed by the GOPTS. Therefore, the response of BSA changes and reaches the steady condition very quickly. The BRIS was 5.89 nm/RIU in this case.
Fig. 4 Nonspecific detection of bovine serum albumin solutions. (a) Silanization of the microbubble surface and the process of nonspecifically identifying bovine serum albumin. (b) Wavelength shifts caused by the binding of bovine serum albumin solutions. (c) Final calculated molecular surface density (MSD) as a function of different bovine serum albumin concentrations.
In general, a linear relationship exists between the wavelength shift of a WGM and the molecular surface density (MSD) on the internal surface of the OMBR, which provides a detailed quantitative analysis for the detection of molecules using an OMBR [33,38]:
where σp is the MSD (molecules/cm2) given the number of molecules per unit area; αex is the excess polarizability of the analyte; n2 and n3 are the refractive indices of the microbubble wall (1.45) and the aqueous medium in the microbubble core (similar to 1.33), respectively; ε0 is the vacuum permittivity; δλ is the WGM spectral shift caused by the biomolecules binding or being captured on the surface of the microbubble and λ is the wavelength of the WGM (approximately 850 nm) [33,38]. Equation (2) allows us to quantify σp on the inner surface of the OMBR by normalizing the wavelength shift to the BRIS [39]. The excess polarizability αex is approximately proportional to the mass of the molecule and was determined to be αex = 4πε0(3.85 × 10−21) cm3 for BSA [32,40–42]. With Eqs. (1) and (2), we finally calculated the MSD corresponding to the different concentrations of the BSA solutions. Each error bar represents the standard deviation for three independent measurements. In Fig. 4(c), the MSD (σp) is nearly log-linear with the BSA concentration from 1 fg/mL to 100 pg/mL. The linear regression equation is expressed as σp = 5.446 × 1010 × lg (BSA concentration) + 7.936 × 1010 (R2 = 0.996). In Fig. 4(b), the standard deviation of the PBS section (analytical blank) is σ = 0.051 pm in Fig. 4(b), which corresponds to 5.119 × 109 molecules/cm2 through the Eq. (2). The fitted sensitivity is SBSA = 4.153 × 1010 (molecules/cm2)/(fg/mL) (calibration curve) in the case of Fig. 4(b). The theoretical DL can be estimated as DL = 3σ/SBSA = 0.370 fg/mL.3.4 Specific detection of small biomolecules
Specific detection can be achieved by utilizing biological recognition elements or molecular probes such as antibodies or oligonucleotides [7,24,43,44]. To demonstrate the ability of the EROMBRS to detect biomolecules, an approach to functionalize the sensor surface and then specifically detect D-biotin (244.31 Da, 47868, Sigma-Aldrich) is the focus of the remainder of this section. The functionalization processes for specific detection of D-biotin are illustrated schematically in Fig. 5(a).
Fig. 5 Specific detection of the small molecule D-biotin. (a) Silanization of the microbubble surface and the process of specifically identifying the D-biotin biomolecule. (b) Final wavelength shift curves for the detection of 4.1 fM (1 fg/mL) of D-biotin. The OMBR was rinsed and soaked in PBS for several seconds to establish a stable detection baseline. The D-biotin solution was introduced into the OMBR for 1200 s and was then rinsed with PBS once again. (c) Final calculated MSD as a function of different D-biotin concentrations.
The silanization process used to coat the silica microbubble surfaces was previously described above. Surface functionalization continued with exposure to 0.5 mg/mL streptavidin (SA, Thermo Fisher Scientific) for 2 h. Note that SA can be combined with GOPTS to serve as a specific probe for identifying proteins (D-biotin). After rinsing with PBS, a 0.5 mg/mL BSA solution was extracted and left in the microbubble for 30 min. Here, the BSA solution did not react with the biological protein of the substance being tested but closed the site of GOPTS without binding to SA. This process is called surface blocking [39]. Residual BSA was rinsed with PBS. Then, specific recognition of the D-biotin solution was formally initiated. D-biotin solutions were diluted using PBS to 1, 10, and 100 fg/mL, 1, 10, and 100 pg/mL, respectively. A stable detection baseline was established first by extracting PBS for 30 min, and a D-biotin solution was administered for 20 min, and then rapidly replaced by PBS. Specific binding of the small D-biotin molecule to surface-immobilized SA caused a redshift of the WGM, as shown in Fig. 5(b), which was recorded from the beginning of the baseline to the end of the dissociation phase with a time resolution of 1 s.
Figure 5(b) shows the final curves for the detection of D-biotin with a concentration of 1 fg/mL. Three independent detections (including the functionalization and binding procedures) were implemented with three different test OMBRs. The wavelength redshift occurred when D-biotin was bound to SA. The fluctuations in the curves were due to slight alternate associations and dissociations of D-biotin and SA. Nonspecific binding of D-biotin to BSA or GOPTS may have also occurred during this period, and it was more obvious in the case of ultralow concentrations. During the PBS rinsing process, a certain amount of wavelength blueshift came from the dissociation phase and then the WGM position reached an equilibrium state. The stable wavelength shift (t > 1500 s) relative to the starting point (zero shift) was the net wavelength shift (δλnet). The corresponding BRIS (S) was also detected. The excess polarizability (αex) was determined to be αex = 4πε0(1.42 × 10−23) cm3 for D-biotin. The net wavelength shift (δλnet), BRIS (S) and MSD (σp) of three independent measurements are listed in Table 1. The mean MSD (σp) was 1.603 × 1013 molecules/cm2. Three independent measurements were performed to detect the D-biotin at a bulk concentration from 1 fg/mL to 100 pg/mL. Figure 5(c) shows the final calculated MSD as a function of different D-biotin concentrations. Each error bar represents the standard deviation in three independent measurements. The MSD nearly log-linear with the concentration ranging from 1 fg/mL to 100 pg/mL. The linear regression equation is expressed as σp = 3.016 × 1013 × lg (D-biotin concentration) + 1.086 × 1013 (R2 = 0.972).
Table 1. Experimental Data of Three Independent Measurements.
The MSD difference is mainly due to nonspecific binding and the error transfer of the preceding steps. When the experimental detection approaches limitation, the influence of these errors is more remarkable, which leads to a great difference between the two independent measurements. Nevertheless, this does not conceal the specific binding signal. Therefore, the measurement results were valid and effective. In our study, the molar concentrations of BSA and D-biotin at a bulk concentration of 1 fg/mL were 15 aM and 4.1 fM, respectively. At the same bulk concentration and unit volume, the number of molecules of D-biotin was much greater than that of BSA. During ultralow concentration detection (when the number of biomolecules extracted into OMBR was insufficient to fully cover the functionalized surface of OMBR), the number of D-biotin adsorbed on the surface of OMBR was greater than that of BSA. Therefore, the average signal generated by D-biotin binding was much greater than that of BSA. Another major factor was that each SA possessed four binding sites per molecule and could bind to four biotin molecules [45]. This feature can be used to construct a multi-level signal amplification system. Therefore, the wavelength shift for D-biotin detection was greater than that for BSA.
Previous studies showed that the DL of a self-referencing optofluidic ring resonator to detect BSA is 1 pg/mL [28]. In addition, a partially packaged microbubble that can determine D-biotin is 100 fg/mL (0.41 pM) [24], an air-slot SiNx microdisk that detects SA is 30 ± 2 ng/mL [19], and a polymer microring to detect BSA is approximately 5.3 pg/mm2 and 55.9 fg/mm2 in wavelength-shift and intensity-variation schemes, respectively [46]. Compared to the DL of similar sensors, our system can detect BSA and D-biotin at a bulk concentration of 1 fg/mL, which revealed that the ultra-low effective detection limits proposed in this study. Biomarkers, such as, proteins are usually found in concentrations ranging from several ng/mL [47] to hundreds of µg/mL [48]. As an example, based on p53 quantification, for a typical tumor suppressor and transcription factor in cancer biology, cancer diagnostics require a DL as low as pg/mL [35,49,50]. In our study, the detection ability of the EROMBRS for biomolecules was confirmed by experiments, providing a gateway to the basic processes of biology at the nanoscale. The lowest concentration of biomolecule solutions had already reached the 1-fg/mL level, which was far below the concentration required for disease detection and medical diagnosis.
4. Conclusions
In this study, we proposed an effective label-free optofluidic microbubble resonator system for biomolecular detection with ultralow concentrations. The lowest detected concentration of BSA and D-biotin was 1 fg/mL (15 aM and 4.1 fM, respectively). Compared to similar studies, the whole packaged OMBR has good long-term stability and can be stored for long periods; the whole packaged method can greatly reduce the impact of environmental vibrations and improve the portability of the OMBR sample and DL of the system. In a future study, we intend to optimize the condition of every step to further improve the signal-to-noise ratio of the detection. We will also investigate the selective excitation of high-order radial WGMs with higher sensitivity, which can achieve better detection limits. An alternative method is to fabricate an Au or Ag layer on the inner surface of an OMBR to obtain high-Q plasmonic/WGM hybrid modes, which can improve the performance of the current system as well. The EROMBRS has great application potential for medical diagnosis, disease treatment and drug screening.
Funding
Special Project of National Key Technology R&D Program of the Ministry of Science and Technology of China (2016YFC0201401); National Key Basic Research Program of China (Program 973, 2015CB352006); National Natural Science Foundation of China (NSFC) (61378080, 61327008, 61505032 and 61705039).
Acknowledgments
We thank Zihao Li for his guidance and help during the experiment. Zhihe Guo would like to sincerely thank Junyu Liu for her support and encouragement during writing.
References
1. K. J. Vahala, “Optical microcavities,” Nature 424(6950), 839–846 (2003). [CrossRef] [PubMed]
2. A. B. Matsko and V. S. Ilchenko, “Optical resonators with whispering gallery modes I: basics,” IEEE J. Sel. Top. Quantum Electron. 12(1), 3–14 (2006). [CrossRef]
3. V. S. Ilchenko and A. B. Matsko, “Optical resonators with whispering-gallery modes-part II: applications,” IEEE J. Sel. Top. Quantum Electron. 12(1), 15–32 (2006). [CrossRef]
4. T. Reynolds, N. Riesen, A. Meldrum, X. Fan, J. M. Hall, T. M. Monro, and A. François, “Fluorescent and lasing whispering gallery mode microresonators for sensing applications,” Laser Photonics Rev. 11(2), 1600265 (2017). [CrossRef]
5. F. Vollmer and S. Arnold, “Whispering-gallery-mode biosensing: label-free detection down to single molecules,” Nat. Methods 5(7), 591–596 (2008). [CrossRef] [PubMed]
6. S. Zhu, L. Shi, B. Xiao, X. Zhang, and X. Fan, “All-optical tunable microlaser based on an ultrahigh-Q erbium-doped hybrid microbottle cavity,” ACS Photonics 5(9), 3794–3800 (2018). [CrossRef]
7. A. Giannetti, A. Barucci, S. Berneschi, A. Cosci, F. Cosi, D. Farnesi, G. N. Conti, S. Pelli, S. Soria, S. Tombelli, C. Trono, G. C. Righini, and F. Baldini, “Optical micro-bubble resonators as promising biosensors,” Proc. SPIE 9506, 950617 (2015). [CrossRef]
8. J. L. Nadeau, V. S. Ilchenko, D. Kossakovski, G. H. Bearman, and L. Maleki, “High-Q whispering-gallery mode sensor in liquids,” Proc. SPIE 4629, 172–180 (2002). [CrossRef]
9. F. Vollmer, D. Braun, A. Libchaber, M. Khoshsima, I. Teraoka, and S. Arnold, “Protein detection by optical shift of a resonant microcavity,” Appl. Phys. Lett. 80(21), 4057–4059 (2002). [CrossRef]
10. C. H. Dong, L. He, Y. F. Xiao, V. R. Gaddam, S. K. Özdemir, Z. F. Han, G. C. Guo, and L. Yang, “Fabrication of high-Q polydimethylsiloxane optical microspheres for thermal sensing,” Appl. Phys. Lett. 94(23), 231119 (2009). [CrossRef]
11. Y. Sun and X. Fan, “Optical ring resonators for biochemical and chemical sensing,” Anal. Bioanal. Chem. 399(1), 205–211 (2011). [CrossRef] [PubMed]
12. A. Arbabi and L. L. Goddard, “Measurements of the refractive indices and thermo-optic coefficients of Si3N4 and SiO(x) using microring resonances,” Opt. Lett. 38(19), 3878–3881 (2013). [CrossRef] [PubMed]
13. M. R. Foreman, J. D. Swaim, and F. Vollmer, “Whispering gallery mode sensors,” Adv. Opt. Photonics 7(2), 168–240 (2015). [CrossRef] [PubMed]
14. S. C. Yang, Y. Wang, and H. D. Sun, “Advances and prospects for whispering gallery mode microcavities,” Adv. Opt. Mater. 3(9), 1136–1162 (2015). [CrossRef]
15. X. Li, Z. Zhang, Q. Yin, M. Qiu, and Y. Su, “Ultra-compact parallel label-free biosensors based on concentric micro-ring resonators in silicon-on-insulator,” in Proceedings of the Asia Optical Fiber and Optoelectronic 2008 Exposition and Conference (IEEE, 2008). [CrossRef]
16. H. Zhu, P. S. Dale, and X. Fan, “Optofluidic ring resonator sensor for sensitive label-free detection of breast cancer antigen CA15-3 in human serum,” Proc. SPIE 7322, 732204 (2009). [CrossRef]
17. H. K. Hunt, J. L. Dahmen, and C. E. Soteropulos, “Interfacing whispering gallery mode microresonators for environmental biosensing,” Proc. SPIE 8960, 89600O (2014). [CrossRef]
18. L. M. Freeman and A. M. Armani, “Photobleaching of Cy5 conjugated lipid bilayers determined with optical microresonators,” IEEE J. Sel. Top. Quantum Electron. 18(3), 1160–1165 (2012). [CrossRef]
19. S. Lee, S. C. Eom, J. S. Chang, C. Huh, G. Y. Sung, and J. H. Shin, “Label-free optical biosensing using a horizontal air-slot SiNx microdisk resonator,” Opt. Express 18(20), 20638–20644 (2010). [CrossRef] [PubMed]
20. J. Zhu, Ş. K. Özdemir, 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(1), 46–49 (2010). [CrossRef]
21. M. Sumetsky, Y. Dulashko, and R. S. Windeler, “Optical microbubble resonator,” Opt. Lett. 35(7), 898–900 (2010). [CrossRef] [PubMed]
22. X. Zhang, L. Liu, and L. Xu, “Ultralow sensing limit in optofluidic micro-bottle resonator biosensor by self-referenced differential-mode detection scheme,” Appl. Phys. Lett. 104(3), 033703 (2014). [CrossRef]
23. Y. N. Zhang, T. Zhou, B. Han, A. Zhang, and Y. Zhao, “Optical bio-chemical sensors based on whispering gallery mode resonators,” Nanoscale 10(29), 13832–13856 (2018). [CrossRef] [PubMed]
24. Z. Li, C. Zhu, Z. Guo, B. Wang, X. Wu, and Y. Fei, “Highly sensitive label-free detection of small molecules with an optofluidic microbubble resonator,” Micromachines (Basel) 9(6), 274 (2018). [CrossRef] [PubMed]
25. R. Madugani, Y. Yang, V. H. Le, J. M. Ward, and S. N. Chormaic, “Linear laser tuning using a pressure-sensitive microbubble resonator,” IEEE Photonics Technol. Lett. 28(10), 1134–1137 (2016). [CrossRef]
26. S. Liu, W. Sun, Y. Wang, X. Yu, K. Xu, Y. Huang, S. Xiao, and Q. Song, “End-fire injection of light into high-Q silicon microdisks,” Optica 5(5), 612–616 (2018). [CrossRef]
27. Y. Zhi, X.-C. Yu, Q. Gong, L. Yang, and Y.-F. Xiao, “Single nanoparticle detection using optical microcavities,” Adv. Mater. 29(12), 1604920 (2017). [CrossRef] [PubMed]
28. M. Li, X. Wu, L. Liu, X. Fan, and L. Xu, “Self-referencing optofluidic ring resonator sensor for highly sensitive biomolecular detection,” Anal. Chem. 85(19), 9328–9332 (2013). [CrossRef] [PubMed]
29. T. Tang, X. Wu, L. Liu, and L. Xu, “Packaged optofluidic microbubble resonators for optical sensing,” Appl. Opt. 55(2), 395–399 (2016). [CrossRef] [PubMed]
30. Y. Yang, S. Saurabh, J. M. Ward, and S. Nic Chormaic, “High-Q, ultrathin-walled microbubble resonator for aerostatic pressure sensing,” Opt. Express 24(1), 294–299 (2016). [CrossRef] [PubMed]
31. T. M. Aminabhavi, “Use of mixing rules in the analysis of data for binary liquid mixtures,” J. Chem. Eng. Data 29(1), 54–55 (1984). [CrossRef]
32. S. Arnold, M. Khoshsima, I. Teraoka, S. Holler, and F. Vollmer, “Shift of whispering-gallery modes in microspheres by protein adsorption,” Opt. Lett. 28(4), 272–274 (2003). [CrossRef] [PubMed]
33. H. Zhu, I. M. White, J. D. Suter, P. S. Dale, and X. Fan, “Analysis of biomolecule detection with optofluidic ring resonator sensors,” Opt. Express 15(15), 9139–9146 (2007). [CrossRef] [PubMed]
34. A.-M. Haughey, C. Foucher, B. Guilhabert, A. L. Kanibolotsky, P. J. Skabara, G. Burley, M. D. Dawson, and N. Laurand, “Hybrid organic semiconductor lasers for bio-molecular sensing,” Faraday Discuss. 174, 369–381 (2014). [CrossRef] [PubMed]
35. L. Liang, L. Jin, Y. Ran, L.-P. Sun, and B.-O. Guan, “Fiber light-coupled optofluidic waveguide (flow) immunosensor for highly sensitive detection of p53 protein,” Anal. Chem. 90(18), 10851–10857 (2018). [CrossRef] [PubMed]
36. A. Macanovic, C. Marquette, C. Polychronakos, and M. F. Lawrence, “Impedance-based detection of DNA sequences using a silicon transducer with PNA as the probe layer,” Nucleic Acids Res. 32(2), e20 (2004). [CrossRef] [PubMed]
37. D.-K. Kim, H.-G. Lee, H.-I. Jung, and S.-H. Kang, “Single-protein molecular interactions on polymer-modified glass substrates for nanoarray chip application using dual-color TIRFM,” Bull. Korean Chem. Soc. 28(5), 783–790 (2007). [CrossRef]
38. H. Zhu, I. M. White, J. D. Suter, M. Zourob, and X. Fan, “Opto-fluidic micro-ring resonator for sensitive label-free viral detection,” Analyst (Lond.) 133(3), 356–360 (2008). [CrossRef] [PubMed]
39. H. Zhu, P. S. Dale, C. W. Caldwell, and X. Fan, “Rapid and label-free detection of breast cancer biomarker CA15-3 in clinical human serum samples with optofluidic ring resonator sensors,” Anal. Chem. 81(24), 9858–9865 (2009). [CrossRef] [PubMed]
40. J. Wen, T. Arakawa, and J. S. Philo, “Size-exclusion chromatography with on-line light-scattering, absorbance, and refractive index detectors for studying proteins and their interactions,” Anal. Biochem. 240(2), 155–166 (1996). [CrossRef] [PubMed]
41. F. Vollmer and L. Yang, “Label-free detection with high-Q microcavities: a review of biosensing mechanisms for integrated devices,” Nanophotonics 1(3-4), 267–291 (2012). [CrossRef] [PubMed]
42. K. Wilson and F. Vollmer, “Whispering gallery mode resonator biosensors,” in Encyclopedia of Nanotechnology, J. Fagerberg, D. Mowery, and R. Nelson, eds. (Springer, 2012) pp. 2837–2849.
43. M. Baaske and F. Vollmer, “Optical resonator biosensors: molecular diagnostic and nanoparticle detection on an integrated platform,” ChemPhysChem 13(2), 427–436 (2012). [CrossRef] [PubMed]
44. S. Berneschi, F. Baldini, A. Barucci, A. Cosci, F. Cosi, D. Farnesi, G. N. Conti, G. Righini, S. Soria, S. Tombelli, C. Trono, S. Pelli, and A. Giannetti, “Localized biomolecules immobilization in optical microbubble resonators,” Proc. SPIE 9727, 972719 (2016). [CrossRef]
45. E. P. Diamandis and T. K. Christopoulos, “The biotin-(strept)avidin system: principles and applications in biotechnology,” Clin. Chem. 37(5), 625–636 (1991). [PubMed]
46. X. Tu, S. L. Chen, C. L. Song, T. Y. Huang, and L. J. Guo, “Ultrahigh Q polymer microring resonators for biosensing applications,” IEEE Photonics J. 11(2), 1–10 (2019). [CrossRef]
47. J. Hernández and I. M. Thompson, “Prostate-specific antigen: a review of the validation of the most commonly used cancer biomarker,” Cancer 101(5), 894–904 (2004). [CrossRef] [PubMed]
48. K. R. Kozak, F. Su, J. P. Whitelegge, K. Faull, S. Reddy, and R. Farias-Eisner, “Characterization of serum biomarkers for detection of early stage ovarian cancer,” Proteomics 5(17), 4589–4596 (2005). [CrossRef] [PubMed]
49. M. Hasanzadeh, N. Shadjou, and M. de la Guardia, “Current advancement in immunosensing of p53 tumor suppressor protein based on nanomaterials: Analytical approach,” Trends Analyt. Chem. 89, 13–20 (2017). [CrossRef]
50. S. Senturk, Z. Yao, M. Camiolo, B. Stiles, T. Rathod, A. M. Walsh, A. Nemajerova, M. J. Lazzara, N. K. Altorki, A. Krainer, U. M. Moll, S. W. Lowe, L. Cartegni, and R. Sordella, “p53Ψ is a transcriptionally inactive p53 isoform able to reprogram cells toward a metastatic-like state,” Proc. Natl. Acad. Sci. U.S.A. 111(32), E3287–E3296 (2014). [CrossRef] [PubMed]
