Open Access
Add to BibSonomy
Abstract
We report on the design, fabrication, and characterization of mass-producible, sensitive, intensity-detection-based planar waveguide sensors for rapid refractive index (RI) sensing; the sensors comprise suspended glass planar waveguides on glass substrates, and are integrated with microfluidic channels. They are facilely and cost-effectively constructed via vacuum-less processes. They yield a high throughput, enabling mass production. The sensors respond to solutions with different RIs via variations in the transmitted optical power due to coupling loss in the sensing region, facilitating real-time and simple RI detection. Experiments yield a good resolution of 5.65 × 10−4 RIU. This work has major implications for several RI-sensing-based applications.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Refractive index (RI) sensing is a classic analytical method which can be used to monitor molecular interactions and quantitatively analyze chemical reactions. In contrast to other measuring methods, RI sensing offers the advantage of highly sensitive, label-free, and real-time detection [1]. Sensors based on optofluidic refractometry, resulting from the synergistic integration of micro- and nano-optical systems with microfluidics technology, have enjoyed an unparalleled technological development and revealed tremendous potential for a broad spectrum of applications ranging from clinical diagnosis and chemical analysis to adulteration and food safety. Enhanced sensitivity and resolution, reduced device dimensions and cost are some of the unique characteristics that have resulted from this combination of technologies [2]. Given the advantages, significant efforts have been made toward the development of RI sensors based on different structures and sensing mechanisms [3–30].
To achieve sensitive RI sensing, different RI sensors have been developed, such as guided-mode-resonance sensors [3–9], prism sensors [10–12], fiber-optic nanoplasmonic sensors [13–16], waveguide (WG) sensors [17–19], Mach-Zehnder interferometer sensors [20–22], surface plasmon resonance sensors [23–26] and photonic crystal sensors [27–30]. Although these technologies exhibit excellent RI sensing, they lack certain characteristics which are desirable for commercialization such as cost-effectiveness, mass-producible fabrication methods and rapid readout systems. Aforementioned technologies usually require processes such as vacuum thin-film deposition, dry etching and E-beam lithography for sensor fabrication, which are complicated and time-consuming and hence increase the production time and reduce producibility [2]. Another limitation is the use of wavelength or polarization-based sensing approaches requiring complex post-processes for obtaining the solution RI, failing to provide real-time detection capabilities. Lastly, the optical read-out systems usually demand precise and costly instruments and components, such as high-stability lasers, high-resolution spectrometers, and high-precision translation/rotation stages, leading to bulky and more expensive detection systems. Thus, it is highly desirable to develop simple RI sensors favoring mass production, real-time detection, cost effectiveness and compact detection systems for practical applications.
The WG is a well-known optical device for light propagation [31,32]. The field distribution of the guided modes is closely related to the RI of the core and cladding layers. Thus, changes in the RI of the cladding layer can lead to modification of the guided modes in WGs. In this study, we proposed and developed a simple, cost-effective, and mass-producible optofluidic WG-type RI sensor. The proposed WG-type sensor is composed of a suspended planar WG structure on a glass substrate with a microfluidic module incorporated as the sensing region. The WG RI sensors were constructed without using any lithography or vacuum processes, which therefore makes them suitable for high-throughput mass-production. When light is coupled into the WG from one side, it travels through the sensing region to the other side of the WG. The optical power of the propagating light is a function of the RI of the WG and different samples that are injected into the sensing region of the sensor. As sample solutions with varying concentration are introduced into the channel, the intensity of the output light is modified due to variation in the RI. Monitoring variations in output light intensity enables sensitive RI detection in real time. The system used for output light detection employs a high-stability, inexpensive light source in the form of a light-emitting diode (LED), and a photodetector (PD) as the accurate optical receiver and recorder, thereby allowing easy, real-time RI detection. RI experiments reveal a good RI resolution of $\textrm{5}\textrm{.65} \times \textrm{10}_{}^{ - 4}\textrm{ RIU}$. We then performed numerical simulations to study the sensitivity in terms of the WG’s geometry. These results demonstrate a new WG RI sensor for rapid and cost-effective RI sensing for a wide range of applications.
2. Design and sensing principle of optofluidic WG RI sensors
Figure 1 shows the schematic representation and sensing principle of the proposed RI sensor based on a planar WG structure. Figure 1(a) shows the structure of our proposed WG sensor consisting of (1) a low-cost commercially available glass slab substrate (2) a borosilicate glass cover slip of thickness t as the WG material, and (3) a cyclic olefin copolymer (COC) injection-molded microfluidic module for injecting solutions. To achieve waveguiding, the WG must be surrounded by cladding layers with a lower RI. Figure 1(b) depicts the top view of the sensor. The most important feature of this design is its suspended structure which facilitates the waveguiding of light into the core. Unlike the conventional etching method used to partially or completely remove the sacrificial layer below the WG core [33–34], we obtained the suspended structure by gluing the WG to the glass substrate from two sides. This allowed air to get trapped between the substrate and the core, making the WG layer suspended, as shown in Fig. 1(c). As a result, the significant RI contrast between the WG and air improves the optical confinement for the WG layer and ensures minimal leakage of light into the substrate. A microfluidic channel is also integrated at the center of the sensing region of the WG to promote light-analyte interaction and further improve the stability and accuracy of the sensor. Figure 1(d) depicts the sensing mechanism. The light coupled into the WG from the far field, with intensity Iin, propagates through the WG in three distinct regions, namely region 1, 2, and 3. In region 1 and region 3, the upper cladding layer is made of COC with a larger RI of $n = 1.56$ as shown in Fig. 1(d), while in region 2 the upper cladding layer is the solution with a smaller RI in the range of $n = 1.333 - 1.373$. As a result, the evanescent wave extending into the cladding layer in regions 1 and 3 is more significant than region 2. Therefore, as light reaches the interface between region 1 and region 2, a significant coupling loss is induced due to the mode mismatch. As the RI of the solution changes, the distribution of guiding modes in region 2 is changed correspondingly. The coupling loss due to the mode mismatch is thus a function of the RI of the solution, and hence so is the magnitude of received output light intensity Iout. Therefore, changes in RI due to analyte concentration can be converted into a variation of transmitted light intensity which is obtained at the output of the platform, thereby providing the basis for the sensor. Note that the sensitivity of the sensor is independent of the waveguide length because the sensing mechanism relies on the coupling loss at the interfaces between region 1 and region 2, and region 2 and region 3.
Fig. 1. Diagrammatic representation of the proposed optofluidic waveguide refractive index (WG RI) sensor, including a glass substrate, a glass WG and a microfluidic channel. (a) 3D view, (b) top view, (c) side view, and (d) sensing mechanism of the WG RI sensor. COC: cyclic olefin copolymer.
3. Fabrication of optofluidic WG RI sensors
The optofluidic WG RI sensors were constructed using vacuum-less and lithography-less techniques. The fabrication flow is shown in Fig. 2. Low cost glass with a size of $25.4 \times 76.2\textrm{ mm}$ was used as the substrate. In order to ensure good adhesion of the WG to the glass substrate, the latter was cleaned with DI water and air dried by a nitrogen gun, followed by an oven bake of 1 h. The same cleaning steps were adopted for cleaning the cover slip WG and microfluidic channel. The size of the coverslip WG was chosen as $24 \times 50\textrm{ mm}$ with a thickness of 0.2 mm. After cleaning, the borosilicate glass cover slip was glued only at two sides of the substrate and irradiated by a UV lamp for 15 min to allow the glue to harden. Extreme care was taken while gluing the cover slip to the substrate to avoid spreading of the glue toward the sensing region. A COC microfluidic module containing a fluidic channel with dimensions of $32\textrm{ mm (length)} \times 3\textrm{ mm (width)} \times 0.2\textrm{ mm (thickness)}$ was processed by injection molding techniques [30] and bonded to the sensor to manage the fluidic sample solutions. Finally, the sensor was completed by connecting two flexible tubes to the microfluidic channel for the inlet and outlet of the fluids. An optical image of the fabricated WG RI sensor is shown in Fig. 3. The simple fabrication technique avoids the costly and lengthy lithography process, thereby yielding highly practical RI sensors suitable for high-quantity mass production. The approximate cost of the chip is less than one USD.
Fig. 2. Diagram for the fabrication flow of optofluidic WG RI-sensors integrated with a microfluidic module.
4. Characterization of WG RI sensors and discussion
The setup of the optical detection system for taking RI measurements of the fabricated WG sensor is shown in Fig. 4. The set-up is realized on the basis of the NI 9215 data acquisition system of National Instruments [35]. The control of the set-up and the measurement procedures is executed in the LabVIEW environment of graphical programming. A highly stable, low-cost, commercially available green LED operating at $\lambda = 532$ nm was used as the light source. The signal-to-noise ratio was enhanced using a band-pass filter with the aid of a 1-kHz square wave with a 50% duty cycle, generated by a homemade LED driver to drive and directly modulate the LED. The emitted light was collimated using a lens and then coupled to one facet of the slab WG via a 20X objective. A three-axis translation stage was used for fine adjustment of the chip to ensure minimal losses while coupling light into the sensor chip. The light transmitted via the WG chip was filtered by means of an adjustable iris, converged by a lens, and finally focused onto a silicon photodetector for converting the output light intensity into a photocurrent. Later, a homemade current amplifier with a band-pass filter was used to amplify the generated photocurrent, which was then converted to digital signals using an analog-to-digital converter. A lock-in program was used for demodulating the recorded real-time digital signals.
Fig. 4. Diagram of the optical detection system for taking refractive index measurements using the fabricated waveguide sensor.
Experiments were carried out in order to characterize the RI sensing performance. Solutions with different RI $n = 1.333 - 1.373$ were prepared by varying the concentration of sucrose in each. First, DI water was injected into the chip as a blank solution. After that, the various sucrose solutions were injected into the WG RI sensor successively, followed by a final injection of DI water with a fluid flow rate of ∼0.25 mL/s. The data acquisition system was utilized to synchronously record the transmitted light intensity (I) as described above. The normalized real-time optical response of the RI sensor with different solutions is shown in Fig. 5(a). The system noise $\sigma = \frac{{SD}}{{{I_0}}} = \frac{1}{{SNR}}$ was obtained from the results by calculating the standard deviation (SD) and average intensity (I0) measured from the blank solution (DI water) light intensity output. Here SNR stands for signal-to-noise ratio of the system. From our experiment, we obtained $\sigma = 0.017\%$ and SNR=5882, indicating good power stability, which is attributed to the use of a highly-stable LED as the light source as opposed to lasers which usually have $\sigma = 0.1\%$ [5–8].
Fig. 5. (a) Real-time optical response of waveguide refractive index (RI) sensor with different RI solutions. (b) Normalized average intensity as a function of RI of the solutions.
In addition, Fig. 5(a) also shows that an increase in the RI of the sample solutions resulted in a change in the output light intensity, due to the decreasing optical loss in the planar sensing region. The average output light intensities (${I}$avg) for each RI resolution were extracted from the real-time optical responses. The normalized average light intensity output $\left( {\frac{{{I_\textrm{avg}}}}{{{I_0}}}} \right)$ as a function of the RI of the solutions is represented in Fig. 5(b). The normalized sensitivity (Sn) and sensor RI resolution (Rs), representing the minimum detectable change in the RI of the solution, can be evaluated as,
Linear fitting of the experimental results gives ${S_n} = 0.30242\;\textrm{RIU}_{}^{ - 1}$ with a linear correction coefficient $R_{}^2 = 0.98444$, revealing good linearity of the present sensor over a wide range of $\textrm{0}\textrm{.004 }\:\textrm{RIU}$. The RI resolution for the present sensing system is found to be $\textrm{5}\textrm{.65} \times \textrm{10}_{}^{ - 4}\;\textrm{RIU}$. The results obtained demonstrate the feasibility of the proposed sensor for use in a variety of applications.5. Numerical simulation and discussion
Numerical simulation was carried out to better understand the sensing capacity of the developed WG RI sensor and exploit all the possible dependencies of the output signal on various structural and optical parameters of the sensor. In particular, the sensitivity of planar WG sensors highly depends on the thickness of the WG, which significantly affects the evanescent wave distribution in the cladding layer. We performed finite element method (FEM) analysis to study effect of the thickness of the WG on the sensitivity of the sensor. A plane wave at 532 nm was directed into the WG. Perfectly matched layers were used to terminate the simulation domain to absorb the outgoing waves completely. The transmittance (T(n)) of the planar WG structure was then calculated as a function of t. The RIs of the materials used in the FEM simulations are $n_\textrm{s}$ = 1 for air, $n_\textrm{w}$ = 1.52 for higher-RI coverslip WG, and $n_\textrm{coc}$ = 1.56 for COC at λ = 532 nm.
Figure 6(a) shows the calculated normalized squared electric field distribution of the planar WG structure for $t = 3\textrm{ }\mathrm{\mu}\textrm{m}$, and Fig. 6(b) shows the electric field distribution of the TE fundamental mode of the WG. From the results, it is clear that light can be well confined in the WG due to the RI difference between the WG and the cladding layer (solution). In addition, as the incident light travels from the COC to the sensing region, part of the light escapes to the COC region. This is due to coupling loss between the COC and core resulting from the low RI difference. Since only a fraction of the input light reaches the output, attenuation in the output light intensity is observed. As the RI of the sample is increased, the coupling loss decreases due to the reduced mode mismatch, resulting in an increase of the transmittance.
Fig. 6. (a) Simulated electric field distribution of the planar waveguide (WG) structure for $t = 3\textrm{ }\mathrm{\mu}\textrm{m}$ and $w = 60\textrm{ }\mathrm{\mu}\textrm{m}$. (b) Simulated electric field distribution of the TE fundamental mode. (c) Simulated normalized transmittance as a function of refractive index for sample solutions with different WG thickness. (d) Simulated normalized sensitivity with different WG thicknesses. COC: cyclic olefin copolymer.
From the transmittance values of the sensor, normalized transmittance values $\left( {\frac{T}{{{T_0}}}} \right)$ were calculated, where T0 represents the transmittance for the case of DI water $n = 1.333$; Fig. 6(c) shows the calculated $\left( {\frac{T}{{{T_0}}}} \right)$ as a function of the RI of the solution with different WG thickness. It can be seen from figure that for each thickness of the WG, $\left( {\frac{T}{{{T_0}}}} \right)$ increases with increasing RI of the solution. This is because of the reduced coupling loss which results from the reduction of RI gradient in the cladding (solution) region. These observations confirm that a change in sample concentration leads to a modification of the RI of the solution, which can be converted into a modulation of the output light intensity and facilitate real-time intensity-detection-based RI sensing. The value of Sn was calculated from the normalized transmittance data using the formula
Figure 6(d) depicts Sn as a function of t, extracted from Fig. 5(b). The value of Sn of $\textrm{0}\textrm{.04052 RIU}_{}^{ - 1}$ was obtained for $t = 3\textrm{ }\mathrm{\mu}\textrm{m}$. The normalized sensitivity decreased slowly as t increased, because of the reduced evanescent wave intensity in the cladding layer. The analysis indicates that a thinner WG leads to the enhancement of the normalized sensitivity. However, care has to be taken in coupling light properly into the WG, because decreasing t makes it more difficult to couple light into the WG from the far field, which could weaken the light intensity output, degrade the SNR, and limit the RI sensing ability. Hence, a trade-off between the system noise and normalized sensitivity exists and an optimum choice of thickness, which reduces the system noise without compromising the sensitivity, becomes crucial for satisfactory sensor performance. The discrepancies between the simulated and experimental results may because the non-idealities associated with the experiments like roughness of waveguide and interface that contribute to additional coupling loss, leading to increased sensitivity.6. Conclusion
We have presented mass-producible, intensity-detection-based glass WG RI sensors. The sensors were constructed using simple, rapid and vacuum-less processes, offering unique advantages such as low cost and high throughput: necessary requirements for bulk production. Owing to the higher lifespans of the sensor constituent materials, the minimum lifetime of the sensor is expected to be 1 year. The measurement system employs a higher-stability, inexpensive green LED light source and a photodetector as the optical receiver, which makes the system more compact, cost-effective, and applicable for use in portable detection systems. A good RI resolution of $\textrm{5}\textrm{.65} \times \textrm{10}_{}^{ - 4}\textrm{ RIU}$ is achieved over a wide range of $\textrm{0}\textrm{.04 RIU}$. Numerical simulations show that the sensing mechanism is attributed to the coupling loss in the sensing region. The above work has major implications for the fabrication of small, mass-producible sensors for low cost, rapid detection for a wide range of applications.
Funding
Ministry of Science and Technology, Taiwan (MOST 108-2119-M-194-002, MOST 108-2218-E-194-001, MOST 109-2218-E-194-006).
Acknowledgments
The authors thank Prof. Tsung-Heng Tsai at CCU for providing the home-made LED driver and readout circuits in the experiments.
Disclosures
The authors declare no conflicts of interest.
References
1. C. Li, G. Bai, Y. Zhang, M. Zhang, and A. Jian, “Optofluidics refractometers,” Micromachines 9(3), 136 (2018). [CrossRef]
2. I.-C. Liu, P.-C. Chen, L.-K. Chau, and G.-E. Chang, “Optofluidic refractive-index sensors employing bent waveguide structures for low-cost, rapid chemical and biomedical sensing,” Opt. Express 26(1), 273–283 (2018). [CrossRef]
3. J. Vörös, J. J. Ramsden, G. Csúcs, I. Szendrő, S. M. De Paul, M. Textor, and N. D. Spencer, “Optical grating coupler biosensors,” Biomaterials 23(17), 3699–3710 (2002). [CrossRef]
4. W. Zhang, N. Ganesh, I. D. Block, and B. T. Cunningham, “High sensitivity photonic crystal biosensor incorporating nanorod structures for enhanced surface area,” Sens. Actuators, B 131(1), 279–284 (2008). [CrossRef]
5. H.-Y. Li, W.-C. Hsu, K.-C. Liu, Y.-L. Chen, L.-K. Chau, S. Hsieh, and W.-H. Hsieh, “A low cost, label-free biosensor based on a novel double-sided grating waveguide coupler with sub-surface cavities,” Sens. Actuators, B 206, 371–380 (2015). [CrossRef]
6. Y.-F. Ku, H.-Y. Li, W.-H. Hsieh, L.-K. Chau, and G.-E. Chang, “Enhanced sensitivity in injection-molded guided-mode-resonance sensors via low-index cavity layers,” Opt. Express 23(11), 14850–14859 (2015). [CrossRef]
7. Y.-C. Lin, W.-H. Hsieh, L.-K. Chau, and G.-E. Chang, “Intensity-detection-based guided-mode-resonance optofluidic biosensing system for rapid, low-cost, label-free detection,” Sens. Actuators, B 250, 659–666 (2017). [CrossRef]
8. S.-F. Lin, F.-C. Chang, Z.-H. Chen, C.-M. Wang, T.-H. Yang, W.-Y. Chen, and J.-Y. Chang, “A polarization control system for intensity-resolved guided mode resonance sensors,” Sensors 14(3), 5198–5206 (2014). [CrossRef]
9. P. G. Hermannsson, K. T. Sørensen, C. Vannahme, C. L. C. Smith, J. J. Klein, M.-M. Russew, G. Grützner, and A. Kristensen, “All-polymer photonic crystal slab sensor,” Opt. Express 23(13), 16529–16539 (2015). [CrossRef]
10. P. Domachuk, I. C. M. Littler, M. Cronin-Golomb, and B. J. Eggleton, “Compact resonant integrated microfluidic refractometer,” Appl. Phys. Lett. 88(9), 093513 (2006). [CrossRef]
11. Z.-M. Qi, N. Matsuda, J. H. Santos, A. Takatsu, and K. Kato, “Prism-coupled multimode waveguide refractometer,” Opt. Lett. 27(9), 689–691 (2002). [CrossRef]
12. T. Okamoto, M. Yamamoto, and I. Yamaguchi, “Optical waveguide absorption sensor using a single coupling prism,” J. Opt. Soc. Am. A 17(10), 1880–1886 (2000). [CrossRef]
13. W.-T. Hsu, W.-H. Hsieh, S.-F. Cheng, C.-P. Jen, C.-C. Wu, C.-H. Li, C.-Y. Lee, W.-Y. Li, L.-K. Chau, C.-Y. Chiang, and S.-R. Lyu, “Integration of fiber optic-particle plasmon resonance biosensor with microfluidic chip,” Anal. Chim. Acta 697(1-2), 75–82 (2011). [CrossRef]
14. C.-W. Wu, C.-Y. Chiang, C.-H. Chen, C.-S. Chiang, C.-T. Wang, and L.-K. Chau, “Self-referencing fiber optic particle plasmon resonance sensing system for real-time biological monitoring,” Talanta 146, 291–298 (2016). [CrossRef]
15. G. Liang, Z. Luo, K. Liu, Y. Wang, J. Dai, and Y. Duan, “Fiber optic surface plasmon resonance-based biosensor technique: Fabrication, advancement, and application,” Crit. Rev. Anal. Chem. 46(3), 213–223 (2016). [CrossRef]
16. L.-K. Chau, Y.-F. Lin, S.-F. Cheng, and T.-J. Lin, “Fiber-optic chemical and biochemical probes based on localized surface plasmon resonance,” Sens. Actuators, B 113(1), 100–105 (2006). [CrossRef]
17. J. Hu, V. Tarasov, A. Agarwal, L. Kimerling, N. Carlie, L. Petit, and K. Richardson, “Fabrication and testing of planar chalcogenide waveguide integrated microfluidic sensor,” Opt. Express 15(5), 2307–2314 (2007). [CrossRef]
18. C. S. Burke, L. Polerecky, and B. D. MacCraith, “Design and fabrication of enhanced polymer waveguide platforms for absorption-based optical chemical sensors,” Meas. Sci. Technol. 15(6), 1140–1145 (2004). [CrossRef]
19. F. Bahrami, M. Maisonneuve, M. Meunier, J. S. Aitchison, and M. Mojahedi, “An improved refractive index sensor based on genetic optimization of plasmon waveguide resonance,” Opt. Express 21(18), 20863–20872 (2013). [CrossRef]
20. S. Dante, D. Duval, B. Sepúlveda, A. B. González-Guerrero, J. R. Sendra, and L. M. Lechuga, “All-optical phase modulation for integrated interferometric biosensors,” Opt. Express 20(7), 7195–7205 (2012). [CrossRef]
21. M. I. Lapsley, I.-K. Chiang, Y. B. Zheng, X. Ding, X. Mao, and T. J. Huang, “A single-layer, planar, optofluidic Mach-Zehnder interferometer for label-free detection,” Lab Chip 11(10), 1795–1800 (2011). [CrossRef]
22. V. Toccafondo and C. J. Oton, “Robust and low-cost interrogation technique for integrated photonic biochemical sensors based on Mach-Zehnder interferometers,” Photonics Res. 4(2), 57–60 (2016). [CrossRef]
23. R. Robelek and J. Wegener, “Label-free and time-resolved measurements of cell volume changes by surface plasmon resonance (SPR) spectroscopy,” Biosens. Bioelectron. 25(5), 1221–1224 (2010). [CrossRef]
24. L. Malic, M. G. Sandros, and M. Tabrizian, “Designed biointerface using near-infrared quantum dots for ultrasensitive surface plasmon resonance imaging biosensors,” Anal. Chem. 83(13), 5222–5229 (2011). [CrossRef]
25. H. H. Nguyen, J. Park, S. Kang, and M. Kim, “Surface plasmon resonance: A versatile technique for biosensor applications,” Sensors 15(5), 10481–10510 (2015). [CrossRef]
26. P. Singh, “SPR biosensors: Historical perspectives and current challenges,” Sens. Actuators, B 229, 110–130 (2016). [CrossRef]
27. Y. Nazirizadeh, U. Bog, S. Sekula, T. Mappes, U. Lemmer, and M. Gerken, “Low-cost label-free biosensors using photonic crystals embedded between crossed polarizers,” Opt. Express 18(18), 19120–19128 (2010). [CrossRef]
28. J. Jágerská, H. Zhang, Z. Diao, N. L. Thomas, and R. Houdré, “Refractive index sensing with an air-slot photonic crystal nanocavity,” Opt. Lett. 35(15), 2523–2525 (2010). [CrossRef]
29. A. D. Falco, L. O’Faolain, and T. F. Krauss, “Chemical sensing in slotted photonic crystal heterostructure cavities,” Appl. Phys. Lett. 94(6), 063503 (2009). [CrossRef]
30. S. Surdo, S. Merlo, F. Carpignano, L. M. Strambini, C. Trono, A. Giannetti, F. Baldini, and G. Barillaro, “Optofluidic microsystems with integrated vertical one-dimensional photonic crystals for chemical analysis,” Lab Chip 12(21), 4403–4415 (2012). [CrossRef]
31. S. L. Chuang, Physics of Photonic Devices (Wiley, 2009) 2nd ed.
32. G. T. Reed and A. P. Knights, Silicon Photonics: An Introduction (Wiley, 2004).
33. T. H. Stievater, M. W. Pruessner, W. S. Rabinovich, D. Park, R. Mahon, D. A. Kozak, J. B. Boos, S. A. Holmstrom, and J. B. Khurgin, “Suspended photonic waveguide devices,” Appl. Opt. 54(31), F164–F173 (2015). [CrossRef]
34. J. S. Penadés, C. A. Ramos, A. Z. Khokhar, M. Nedeljkovic, L. A. Boodhoo, A. O. Moñux, I. M. Fernández, P. Cheben, and G. Z. Mashanovich, “Suspended SOI waveguide with sub-wavelength grating cladding for mid-infrared,” Opt. Lett. 39(19), 5661–5664 (2014). [CrossRef]
35. C. Y. Chiang, T. T. Huang, C. H. Wang, C. J. Huang, T. H. Tsai, S. N. Yu, Y. T. Chen, S. W. Hong, C. W. Hsu, T. C. Chang, and L. K. Chau, “Fiber optic nanogold-linked immunosorbent assay for rapid detection of procalcitonin at femtomolar concentration level,” Biosens. Bioelectron. 151, 111871 (2020). [CrossRef]
