Open Access
Add to BibSonomy
Abstract
High-sensitivity optical sensors based on the cascaded reflective MZIs and the microring resonators are investigated theoretically and experimentally. The free spectral ranges of the microring and the reflective MZI are designed to be similar in order to produce Vernier effect. A high sensitivity of 1892dB/RIU of the optical sensor was achieved for the intensity interrogation. The measurement average power shows 13 dB increment comparing to that of the traditional two cascaded microring resonators. It’s meaningful for the realization of the low-cost portable systems and application in biochemical sensing.
© 2017 Optical Society of America
1. Introduction
Recently, competitive research work has been investigated about optical RI sensors with various structures, such as ring resonators [1,2], surface plasmon resonance (SPR) [3], waveguide Bragg gratings [4], long-period fiber grating (LPFG) [5] and Mach–Zehnder (MZI) [6,7]. Among them, the ring resonator has been regarded as a promising solution for biological recognition and chemical analysis. The light is coupled into the resonator under resonant condition and confined near the waveguide surface of the resonators with an evanescent field decaying into the surrounding medium. The resonant wavelength is influenced by the refractive index of the reagent contact to the waveguide surface. The real advantage of the microring resonator configurations with respect to many others is probably the very small footprint which allows multiplexing. Recently, several research groups have demonstrated multiplex biosensor using silicon photonic microring resonators [8–10]. High sensitivity with a large measurement range has been achieved using ring-coupled Mach–Zehnder interferometer for sensing [6,7]. However, the limitation of small detection power still exists. In certain complicated biochemical reactions, the absorption of the analyte molecules and the intrinsic waveguide loss will lead to output spectra detection failure in the drop port of the microring. Typically, surface functionalization approaches like molecularly imprinted membrane technology are required to modify the membrane on the sensing surface. In some molecular diagnostics and nanoparticle detection process, particles will even need to be deposited on the external part of the resonance ring. These appearances hinder the operation of high-Q micro-ring resonators in biochemical sensing. In our previous work, we proposed and demonstrated a highly sensitive optical sensor based on the cascaded the drop ports of the double-ring resonators (DRR) using Vernier effect [1,11,12]. However, the system has a low output power, because the light passing through the cascaded DRR is launched from the drop port when the resonant conditions of the DRR are satisfied. This result in the low output power. In order to get a photo detector with a high sensitivity, the costs of the sensor system have to be increased. In this paper, we cascaded reflective MZI with the through port of a single ring resonator to produce a Vernier effect with a high output power. Experiment results showed the output power of this sensor is 13dB higher than that of traditional cascaded DRR for the intensity interrogation.
2. The schematic of cascaded reflective MZI and a microring
A schematic illustration of the cascaded reflective MZI and a microring (RMM) is shown in Fig. 1. As shown in Fig. 1(a), the sensor comprised of two main components: the reflective MZIs and a microring. The reflective MZI contain a fiber coupler (2 × 2, 50% coupling ratio) with 200nm Au coated on the end of the two output fibers (port 2 and port 3) act as the two arms of the reflective MZI. In Fig. 1(b), the microring chip was fabricated on a silicon on insulator (SOI) platform with a 220 nm thick silicon top layer and a 2 μm thick buried oxide and was embedded in the trench of a microfluidic channel. The entire sensor chip was covered by a SU-8 upper cladding layer except for the sensing ring. For the microring, the widths of all the straight waveguides are designed to be 0.5 μm with a ridge height of 0.22 μm and the ring diameter of 124 μm. The width of ring (n) is 0.55 μm and the width of gap between straight waveguide and ring (m) is 0.5 μm (in Fig. 1(a)). To measure the spectral response of the sensor, a tunable laser source (Agilent 81606A) was used as the light source. A polarization controller was used to rotate the polarization of the input light for maximum coupling to the TM mode. The light is coupled into the input port (port 1) of fiber coupler, and split into the two arms of the reflective MZI. The reflective light from port 2 interferes with the one from port 3 at port 4. Finally, the light is coupled into the sensing ring by the grating coupler (port 5). The output signal from the through port (port 6) is finally received by the power sensor (Agilent 81634A).
Fig. 1 Schematic image of the cascaded microring and MZIs. (a). the whole structure; (b). the microring resonator.
The sensing principle can be described as followed. Based on the transfer matrix method (TMM), the spectrum T1 of the through port 6 of sensing ring is defined as [13]:
Where t1 and t2 are the transmission coefficient of the two different coupling regions, respectively (labeled A and B in Fig. 1(a)), β is the propagation constant in the microring and the bus waveguides, αR is the loss index of the microring, αL is the loss index of the bus waveguide, r is the radius of the ring and L is the distance from the coupling point to the output port.
According to the interference formulas, the output interference spectrum T2 of the reflective MZIs can be determined as [14]:
Where h is the difference of the length in the two output fibers of the coupler, n1 is the refractive index of the fiber and λ is the incident wavelength. Since the ends of port 3 and 4 of the coupler are coated with Au, the reflectivity R was approximated to be 1 and the loss in the fiber can be ignored.
The total transmission coefficient T = T1 × T2 and based on the Vernier-effect, when the RI of the liquid sample changes, the shift of the transmission peak is proportional to [1]:
Where FSR1 and FSR2 are the free spectral range of the microring and the reflective MZIs, respectively. Herein, FSR1 is determined by the radius of the ring r, while FSR2 is related to the length difference h of the two output fibers in the reflective MZIs. This indicates that the high sensitivity of the sensing system can be achieved by modifying the values of r and h.
3. Simulation results
In the following section, we theoretically demonstrate that the RMM can obtain much higher transmission power compared to that of the traditional DRR. According to the TMM, the simulated transmission spectra of the drop and the through port of a microring are shown in Fig. 2(a). In the calculation, the power coupling ratio (defined as K = κ2 = 1-t2) is set to be 0.42, the radius of ring r is set to be 124 μm, and the loss index of the ring is α = 13.2 dB/cm, all of which are estimated from practical tests. The solid lines represent the spectrum from the drop port and the dashed lines depict the through port spectrum. From Fig. 2(a), we can see that most of the incident light output from through port and only the light with the resonance wavelength was coupled into the ring and output from drop port. Moreover, the reflective fiber MZI is a discrete device, which will not be affected by the sensing region and has low loss. This part can thus maintain constant and high transmission power. Therefore, the design of the cascading through port of the microring and the reflective MZIs can be used to solve the problem of low output power when cascading the two drop ports of the DRR.
Fig. 2 (a). Simulated transmission spectra from the drop port (solid line) and from the through port (dashed line) of a microring. (b). Transmission spectra of the RMM with the effective index changing from 1.830 to 1.831. (c). Simulated wavelength shifts versus refractive index change. (d). Simulated output power versus the effective refractive index changing from 1.830 to 1.8325.
When the effective index changes from 1.830 to 1.831, the transmission spectrum of the cascaded RMM is shown in Fig. 2(b). The power coupling ratio K is set to be 0.42, the radius of the ring r is 124 μm and the loss ratio of ring is α = 13.2 dB/cm. The resonance peak envelope shift is significantly amplified by 35 times when the effective index changing changes by 0.001. Accordingly, the extracted refractive index sensitivity reaches 14000 nm/RIU as depicted in Fig. 2(c). When the effective refractive index changes from 1.830 to 1.831, the simulated output powers of the DRR and the RMM with the input power 0 dBm are compared in Fig. 3(d). In the simulation, the radii of double microrings are both set to be 124 μm to achieve the highest sensitivity. The central wavelength of the broadband source is 1550 nm with the 3 dB bandwidth 8 nm. The simulation results show that the detected output power of the RMM higher 14.2 dB than that of the DRR.
Fig. 3 (a) Optical microscope image of the microring. (b) SEM image of the input grating coupler. (c) SEM image of the direction coupler.
4. Experiment and results
For experimental verification of our numerical simulations, the MZI and the microring were fabricated. Fabrication of the microring is conducted on a SOI platform. The waveguide structure was patterned by photolithography, prior to inductively coupled plasma (ICP) etching of the silicon layers. An optical microscope image and some SEM images of the microring structure are presented in Figs. 3(a)-3(c). The experimentally measured transmission spectra of resonance single microring and the output spectrum of the fiber coupler are shown in Figs. 4(a) and 4(b) respectively. FSR of the single ring is about 0.75 nm and FSR of the reflective MZI is around 0.8 nm. The slight difference of FSRs between the single microring and the reflective MZI produces Vernier effect.
When injecting NaCl solution with different volume concentrations, the transmission spectra of the RMM sensor are depicted in Fig. 5. When the concentration of NaCl (aq) changes from 0% to 1.8%, the resonance wavelength red shifts remarkably from 1550.93 nm to 1561.08 nm. Since the effects of dispersion and the input/output grating couplers are not taken into consideration in the simulation, the simulation result is different from that of the experimental results. We repeated the experiment for three times, and the fluctuated wavelength shifts are recorded in Fig. 6(a). The measured sensitivity for the wavelength interrogation reaches 3048 nm/RIU in terms of the fitting line (△λ = 3048△n, R2 = 0.996), which is much higher than that of the traditional single ring (165 nm/RIU) [1]. The error bars showed in Fig. 6(a) indicate good reproducibility of the sensing system and the calculated relative standard deviation (the ratio of the standard deviation σ to the mean μ) is around RSD = 0.67%. The measured refractive sensitivity (3048 nm/RIU) is lower than that of the simulation result (14000 nm/RIU). This is because the amplification ratio of the peak shift in the experiment (10) is lower than that in simulation (45).
Fig. 5 Transmission spectra for different NaCl solutions. (a). distilled water (b). 1% NaCl solution (c). 1.4% NaCl solution (d). 1.8% NaCl solution.
Fig. 6 (a). Measured wavelength shift versus refractive index change of NaCl solutions with different concentrations and fitting curve. (b). Measured power ratio versus refractive index change of NaCl solutions with different concentrations and fitting curve.
For intensity interrogation, a low-cost broadband source with the central wavelength of 1585nm and 3 dB bandwidth of 8nm was used as the input source. The output powers were received by a power meter. The ratio of the output power with the input power 0 dBm varying with the refractive index change is shown in Fig. 6(b). For each NaCl solution, the sample was tested three times, and the least squares method [15] was adopted to fit the curve. In Fig. 6(b), the highest sensitivity of the sensor system in the linear part can be as high as 1892dB/RIU, which is quite high compared with 1000 dB/RIU for the cascaded Fabry-Perot laser and microring reported in [16], Song, J et al. The repeatable experimental results with low error (RSD = 0.31%) prove the high stability of the label free sensor system. It’s worth mentioning that the TEC was used to maintain the sensor at a constant temperature during the measurement.
The performance of our proposed sensor is listed in Table 1. This was compared to that of the double ring resonators and cascaded FP laser and microring for a clearer picture of the improvement in performance. It shows that the normalized output power of the RMM is 13dB higher than that of the DRR. The intensity sensitivity of RMM is smaller than that of DRR, while the signal to noise ratio of the RMM is higher than that of the DRR. Besides the sensitivity and average loss, the limit of detection (LOD) is an essential parameter when designing an optical sensor. The LODs of the RMM and DRR are calculated using the method of intensity interrogation [17]. We can see that the LOD of our system is about three times larger than the detection limit of the DRR. Due to its high power and signal to noise ratio, the RMM has a great potential for many practical applications.
Table 1. The performance of our proposed sensor
5. Conclusion
In conclusion, we have proposed a novel sensor structure composed of cascaded reflective fiber MZIs and a microring. The high-sensitivity and high output power of the sensor have been demonstrated theoretically and experimentally. The measured sensitivity is as high as 1892dB/RIU and the output power has dramatically increased about 13 dB compared to that of a DRR. The reported SOI based microrings of the sensors are scalable to be fabricated into arrays of devices with different functions to detect different types of sample. In addition to the advantage of multiplexing possibilities, the improvement of transmission power is demanded in low-cost portable systems and biochemical sensing. Therefore, this sensor has great potential for investigating biochemical reactions and monitoring the environment.
Acknowledgments
This work was supported by National Science Foundation (NSF) (61535010); National High Technological Research and Development Program of China (2014AA06A504); Science and Technology Department of Zhejiang Province (2014C31030 and 2014C31088); Natural Science Foundation of Zhejiang Province (LY16F050001). The authors acknowledge the wafer preparation and technical assistant by the Integrated Circuit Advanced Process Center (ICAC) of Institute of Microelectronics of Chinese Academy of Sciences (IMECAS).
References and links
1. L. Jin, M. Li, and J. J. He, “Optical waveguide double-ring sensor using intensity interrogation with a low-cost broadband source,” Opt. Lett. 36(7), 1128–1130 (2011). [CrossRef] [PubMed]
2. H. T. Kim and M. Yu, “Cascaded ring resonator-based temperature sensor with simultaneously enhanced sensitivity and range,” Opt. Express 24(9), 9501–9510 (2016). [CrossRef] [PubMed]
3. B. Schwarz, P. Reininger, D. Ristanić, H. Detz, A. M. Andrews, W. Schrenk, and G. Strasser, “Monolithically integrated mid-infrared lab-on-a-chip using plasmonics and quantum cascade structures,” Nat. Commun. 5, 4085 (2014). [CrossRef] [PubMed]
4. N. N. Klimov, S. Mittal, M. Berger, and Z. Ahmed, “On-chip silicon waveguide Bragg grating photonic temperature sensor,” Opt. Lett. 40(17), 3934–3936 (2015). [CrossRef] [PubMed]
5. W. Yu, T. Lang, J. Bian, and W. Kong, “Label-free fiber optic biosensor based on thin-core modal interferometer,” J. Sensor. Actuat. Biol. Chem. 228, 322–329 (2016).
6. X. Jiang, Y. Chen, F. Yu, L. Tang, M. Li, and J. J. He, “High-sensitivity optical biosensor based on cascaded Mach-Zehnder interferometer and ring resonator using Vernier effect,” Opt. Lett. 39(22), 6363–6366 (2014). [CrossRef] [PubMed]
7. J. Wang and D. Dai, “Highly sensitive Si nanowire-based optical sensor using a Mach-Zehnder interferometer coupled microring,” Opt. Lett. 35(24), 4229–4231 (2010). [CrossRef] [PubMed]
8. I. A. Estrada, R. W. Burlingame, A. P. Wang, K. Chawla, T. Grove, J. Wang, S. O. Southern, M. Iqbal, L. C. Gunn, and M. A. Gleeson, “Multiplex detection of pathogen biomarkers in human blood, serum, and saliva using silicon photonic microring resonators,” Proc. SPIE 9490, 94900E (2015). [CrossRef]
9. J. H. Wade, A. T. Alsop, N. R. Vertin, H. Yang, M. D. Johnson, and R. C. Bailey, “Rapid, Multiplexed Phosphoprotein Profiling Using Silicon Photonic Sensor Arrays,” ACS Cent Sci 1(7), 374–382 (2015). [CrossRef] [PubMed]
10. H. Cai, M. A. Stott, D. Ozcelik, A. Hawkins, and H. Schmidt, “On-chip Wavelength Multiplexing Detection of Cancer DNA Biomarkers in Blood Serum,” in C. CLEO: Science and Innovations (2016), paper SW4G.5.
11. X. Jiang, J. Ye, J. Zou, M. Li, and J. J. He, “Cascaded silicon-on-insulator double-ring sensors operating in high-sensitivity transverse-magnetic mode,” Opt. Lett. 38(8), 1349–1351 (2013). [CrossRef] [PubMed]
12. Y. Chen, F. Yu, C. Yang, J. Song, L. Tang, M. Li, and J. J. He, “Label-free biosensing using cascaded double-microring resonators integrated with microfluidic channels,” Opt. Commun. 344, 129–133 (2015). [CrossRef]
13. A. Ylein, “universal relations for coupling of optical power between microresonactors and dielectric waveguides,” Electron. Lett. 36, 12–15 (2000).
14. P. I. Nikitin, B. G. Gorshkov, M. V. Valeiko, and S. I. Rogov, “Spectral-phase interference method for detecting biochemical reactions on a surface,” J. Quantum Electron. 30(12), 1099–1104 (2000). [CrossRef]
15. A. Charnes, E. L. Frome, and P. L. Yu, “The equivalence of generalized least squares and maximum likelihood estimates in the exponential family,” J. J. Am. Stat. Assoc. 71(353), 169–171 (1976). [CrossRef]
16. J. Song, L. Wang, L. Jin, X. Xia, Q. Kou, S. Bouchoule, and J. J. He, “Intensity-Interrogated Sensor Based on Cascaded Fabry–Perot Laser and Microring, Resonator,” J. Lightwave Technol. 30(17), 2901–2906 (2012). [CrossRef]
17. L. Jin, M. Li, and J. J. He, “Analysis of wavelength and intensity interrogation methods in cascaded double-ring sensors,” J. Lightwave Technol. 30(12), 1994–2002 (2012). [CrossRef]
