TDM-controlled ring resonator arrays for fast , fixed-wavelength optical biosensing

A novel control concept for serial ring resonator arrays based on a time-division multiplex (TDM) approach is presented. It allows fast sampling rates in terms of biological kinetics. The novelty consists of using both thermal tuning of the effective refractive index and thermo-optical multiplexing for the silicon-on-insulator (SOI) ring resonator arrays, without the need for a tunable laser source. Using a fixed wavelength, fast read-out rates of 100 Hz are demonstrated for each ring. © 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement OCIS codes: (230.0230) Optical devices; (130.3990) Micro-optical devices. References and links 1. A. J. Qavi, A. L. Washburn, J. Y. Byeon, and R. C. Bailey, “Label-free technologies for quantitative multiparameter biological analysis,” Anal. Bioanal. Chem. 394(1), 121–135 (2009). 2. W. Lukosz, “Integrated optical chemical and direct biochemical sensors,” Sens. Actuators B Chem. 29(1-3), 37– 50 (1995). 3. X. Fan, I. M. White, S. I. Shopova, H. Zhu, J. D. Suter, and Y. Sun, “Sensitive optical biosensors for unlabeled targets: A review,” Anal. Chim. Acta 620(1-2), 8–26 (2008). 4. K. Tiefenthaler and W. Lukosz, “Sensitivity of grating couplers as integrated-optical chemical sensors,” J. Opt. Soc. Am. B 6(2), 209–220 (1989). 5. B. J. Luff, J. S. Wilkinson, J. Piecher, U. Hollenbach, J. Ingenhoff, and N. Fabricius, “Integrated optical MachZehnder biosensor,” J. Lightwave Technol. 16(4), 583–592 (1998). 6. R. Heideman, R. Kooyman, and J. Greve, “Performance of a highly sensitive optical waveguide Mach-Zehnder interferometer immunosensor,” Sens. Actuators B Chem. 10(3), 209–217 (1993). 7. F. Prieto, B. Sepúlveda, A. Calle, A. Llobera, C. Domínguez, and L. M. Lechuga, “Integrated Mach-Zehnder interferometer based on ARROW structures for biosensor applications,” Sens. Actuators B Chem. 92(1–2), 151– 158 (2003). 8. J. Homola, S. Yee, and G. Gauglitz, “Surface plasmon resonance sensors: review,” Sens. Actuators B Chem. 54(1–2), 3–15 (1999). 9. www.biacore.com 10. C. Y. Chao and L. J. Guo, “Design and optimization of microring resonators in biochemical sensing applications,” J. Lightwave Technol. 24(3), 1395–1402 (2006). 11. A. L. Washburn and R. C. Bailey, “Photonics-on-a-chip: recent advances in integrated waveguides as enabling detection elements for real-world, lab-on-a-chip biosensing applications,” Analyst (Lond.) 136(2), 227–236 (2011). 12. M. S. Luchansky and R. C. Bailey, “High-Q optical sensors for chemical and biological analysis,” Anal. Chem. 84(2), 793–821 (2012). 13. S. TalebiFard, S. Schmidt, W. Shi, W. Wu, N. A. F. Jaeger, E. Kwok, D. M. Ratner, and L. Chrostowski, “Optimized sensitivity of silicon-on-insulator (SOI) strip waveguide resonator sensor,” Biomed. Opt. Express 8(2), 500–511 (2017). 14. I. M. White and X. Fan, “On the performance quantification of resonant refractive index sensors,” Opt. Express 16(2), 1020–1028 (2008). 15. J. Hu, X. Sun, A. Agarwal, and L. C. Kimerling, “Design guidelines for optical resonator biochemical sensors,” J. Opt. Soc. Am. B 26(5), 1032–1041 (2009). 16. M. Soltani, S. Yegnanarayanan, Q. Li, and A. Adibi, “Systematic engineering of waveguide-resonator coupling for silicon microring/microdisk/racetrack resonators: theory and experiment,” IEEE J. Quantum Electron. 46(8), 1158–1169 (2010). Vol. 26, No. 17 | 20 Aug 2018 | OPTICS EXPRESS 22356 #332404 Journal © 2018 https://doi.org/10.1364/OE.26.022356 Received 29 May 2018; revised 18 Jul 2018; accepted 19 Jul 2018; published 15 Aug 2018 17. A. Densmore, D.-X. Xu, P. Waldron, S. Janz, P. Cheben, J. Lapointe, A. Delge, B. Lamontagne, J. H. Schmidt, and E. Post, “A silicon-on-insulator photonic wire based evanescent field sensor,” IEEE Photonics Technol. Lett. 18(23), 2520–2522 (2006). 18. W. Bogaerts, P. De Heyn, T. Van Vaerenbergh, K. De Vos, S. K. Selvaraja, T. Claes, P. Dumon, P. Bienstman, D. Van Thourhout, and R. Baets, “Silicon microring resonators,” Laser Photonics Rev. 6(1), 1–27 (2011). 19. J. H. Wade and R. C. Bailey, “Applications of optical microcavity resonators in analytical chemistry,” Annu. Rev. Anal. Chem. (Palo Alto, Calif.) 9(1), 1–25 (2016). 20. K. De Vos, J. G. Molera, T. Claes, Y. De Koninck, S. Popelka, E. Schacht, R. Baets, and P. Bienstman, “Multiplexed antibody detection with an array of silicon-on-insulator microring resonators,” IEEE Photonics J. 1(4), 225–235 (2009). 21. D.-X. Xu, M. Vachon, A. Densmore, R. Ma, A. Delâge, S. Janz, J. Lapointe, Y. Li, G. Lopinski, D. Zhang, Q. Y. Liu, P. Cheben, and J. H. Schmid, “Label-free biosensor array based on silicon-on-insulator ring resonators addressed using a WDM approach,” Opt. Lett. 35(16), 2771–2773 (2010). 22. M. Iqbal, M. A. Gleeson, B. Spaugh, F. T. Tybor, W. G. Gunn, M. Hochberg, T. Baehr-Jones, R. C. Bailey, and L. C. Gunn, “Label-free biosensor arrays based on silicon ring resonators and high-speed optical scanning instrumentation,” IEEE J. Sel. Top. Quantum Electron. 16(3), 654–661 (2010). 23. K. De Vos, I. Bartolozzi, E. Schacht, P. Bienstman, and R. Baets, “silicon-on-insulator microring resonator for sensitive and label-free biosensing,” Opt. Express 15(12), 7610–7615 (2007). 24. T. Claes, W. Bogaerts, and P. Bienstman, “Experimental characterization of a silicon photonic biosensor consisting of two cascaded ring resonators based on the Vernier-effect and introduction of a curve fitting method for an improved detection limit,” Opt. Express 18(22), 22747–22761 (2010). 25. T. Claes, W. Bogaerts, and P. Bienstman, “Vernier-cascade label-free biosensor with integrated arrayed waveguide grating for wavelength interrogation with low-cost broadband source,” Opt. Lett. 36(17), 3320–3322 (2011). 26. Customer Service, Keysight Technology, 130 Herrenberger Straße, Boeblingen, 71034, Germany (personal communication, 2018) 27. M. Jäger, D. Volkmann, J. Bruns, and K. Petermann, “Multiplexed single wavelength biosensor for low cost applications,” Adv. Photonics, OSA Technical Digest (online), paper SeT1C.4 (2015). 28. G. Cocorullo, F. Della Corte, I. Rendina, and P. M. Sarro, “Thermooptic effect exploitation in silicon microstructures,” Sens. Actuators A Phys. 71(1), 19–26 (1998). 29. J. E. Saunders, C. Sanders, H. Chen, and H. P. Loock, “Refractive indices of common solvents and solutions at 1550 nm,” Appl. Opt. 55(4), 947–953 (2016).


Introduction
The development of biosensors, e.g. for clinical approaches, water quality screenings and food industries, has been an active research area for over 20 years [1][2][3].Particularly relevant are optical methods of label-free sensing, since they do not require signal transducing markers, which often bias the results and warrant time-intensive methods [4][5][6][7][8].Today commercially available optical label-free biosensors exist in free-space beam configurations as well as in chip-integrated form.Well known examples are SPR (surface plasmon resonance) biosensors by different companies and the chip-integrated detection system "Maverick" by Genalyte [9].Within the field of optical integrated biosensors, evanescent field sensors are particularly promising.Light is guided in the high refractive index waveguide core, surrounded by low index cladding.A fraction of light, the evanescent optical field, extends some tens of nanometers into the surrounding media.The existence of analytes changes the refractive index close to the waveguide surface, and, in turn, the effective index of the guided mode.This sensing scheme facilitates the label-free detection [10].Several different concepts for these devices were developed [11]: grating coupled, interferometric, photonic crystal, surface plasmon resonance and ring resonator based sensors were investigated for lab-on-a-chip analytical applications.A comprehensive review can be found in [12,13].Among these different concepts, ring resonator based biosensors as a chip integrated approach are considered as very capable devices.Their main advantages are a high sensitivity and the small device size.Thus a large number of sensor elements can be placed on a single chip, enabling parallel multiplexed measurements.
Design guidelines and schemes for optimization were given in [14][15][16].In particular, SOI is a promising material for biosensors because silicon wire waveguides are notably sensitive to surface adsorption.The optical mode can be strongly localized near the waveguide surface due to the high refractive index contrast between silicon and the cladding material [17].Ring resonators can be designed with minimum bending radii down to 5 microns for the waveguide geometry used [18].Chips can be fabricated in high volumes using CMOS-compatible processes.This can make the sensor chips disposable, avoiding complicated cleaning procedures.
Many SOI-based biosensors have been demonstrated in the last years for a variety of applications [19][20][21][22][23][24][25] and several solutions for multiplexed and massive parallel detecting sensor chips were presented.All these concepts require an expensive tunable laser source (TLS), which limits the measurement speed of the sensor system due to the scanning repetition speed, meaning the number of full sweeps per second, which is usually in the range of some Hz for commercial TLS [26].Alternatively, it is propose here to use a fixed wavelength source and to apply thermo-optical tuning of the SOI ring resonators.In [27], we already presented a fixed wavelength operation of a ring resonator array.In this paper a novel modulation technique that enables fast tuning and measuring of multiple ring resonators at a fixed wavelength is presented.

Fixed wavelength operation
A simple model for a ring resonator coupled to an access waveguide is based on an ideal lossless coupler with field transmission factor  and field coupling factor  (* denotes the conjugated complex value) as shown in Fig. 1.
with the intensity at the pass port pass I and the input port in I .At resonance condition, light coupled into the ring sees a phase shift   of multiples of 2 and interferes constructively: with the effective refractive index eff n and m being an integer describing the resonance order.Environmental changes near the waveguides surface, for e.g.molecule binding, affect the effective refractive index of the ring, thus causing a change in resonance condition.
If a tunable laser is used to detect the resonance condition, the corresponding resonance wavelength is given as:

Modulation and resonance condition
The working principle of the sensor concept is illustrated in Fig. 2. It shows the transmission characteristics of a single ring resonator as a function of eff n for several resonance orders.Around the bias point ,0 eff n , the effective index of the ring resonator is modulated using the thermo-optical effect of silicon [28].A harmonic voltage ( ) ~sin( )  results in an electrical power  (5)

Multiplexing and read-out mechanism
A general ring resonator based biosensor consists of n sensor elements, which are individually addressable and tunable in the electronic regime using a TDM-like control system.An AC tuning signal and a switching unit that divides the signal in equally spaced time slots provide the modulation signal for the sensor array.Each slot is connected to one specific ring.Fig. 3. Schematic ring resonator system with TDM-like control.An AC signal for tuning is applied to the TDM system, which divides it into equally spaced time slots.Each slot is connected to one sensor and evokes its effective index change in a fix switching scheme.
In this approach the thermo-optical effect in silicon is used for four rings.A sinusoidal voltage is applied to metal heating electrodes adjacent to the rings as shown in Fig. 3.This evokes a local thermal change which tunes the effective refractive index eff n of each ring separately.During one measurement, all rings of the array are modulated serially in a fix switching scheme.The transmission of the sensor device with all rings is measured and A/Dconverted.Additionally, the modulation signal connected to ring 1 acts as a timing reference.
After the transmission of the sensor array has been recorded during the experiment, the positions of the resonances in the common spectrum are extracted from the continuous data stream for each time slot.The modulation signal of ring 1 is used to identify the corresponding rings.Since each ring is tuned over more than res n  , the spectrum of each time slot contains two resonances which are time-dependent, as shown in Fig. 4. By plotting the resonance position for each ring separately, a number of n rings results in n separate binding curves, which enables multi-analyte or multi-parameter measurements.

Sample design and fabrication
The sensor chips were fabricated using commercially available 200 mm SOI wafer.Top silicon layer thickness was 220 nm, buried oxide thickness 2 µm.Deep ultraviolet (DUV) lithography at 248 nm was used for structuring and decoupled plasma processing for etching.In-and out-coupling of the optical signal was achieved by one-dimensional grating couplers, supporting the fundamental transverse electrical (TE) mode.Triangular ring resonators with a circumference of 570 µm were realized on a chip, consisting of single-mode nanowire waveguides of 450 nm width and 220 nm height.These structures possess a bulk sensitivity of 90 nm/RIU without, and 28 nm/RIU with cladding (apart from the circular opening for sensing), a Q-factor around 1.3 × 10 5 and a FSR of 1 nm.The circumference of the resonator results in a tuning range of res n  = 0.0027 for two adjacent peaks and a constant wavelength of 1550 nm.To cover this range, a temperature change of 15°C in the silicon waveguide is required.A schematic view and SEM pictures of a sensor element is depicted in Fig. 5.
The triangular shape of the ring allows for a spatial separation of interaction areas for coupling, heating and sensing, which keeps reciprocal influences at a minimum.For refractive index tuning and modulation, aluminum heaters are structured on top of one side of the triangle in order to make use of the thermo-optical effect in silicon.A buffer oxide layer of 1 µm thickness is used between heaters and waveguides to prevent optical losses, and a protective layer of 1 µm on top to prevent electrochemical influences and intemperate temperature changes on the chip surface due to the applied modulation voltage.The sensing area of the ring is defined by a circular opening etched into the buffer oxide on the sensors left side.Each sensor consists of five ring resonators coupled to a common bus waveguide using directional couplers.The coupling coefficients are chosen close to critical coupling to ensure a sufficient extinction for the detection of the resonances.On four rings, the top SiO2 layer is partially opened to allow for sensing, while the fifth ring is completely covered with SiO2 and acts as a reference to control for ambient temperature variations.The sensor chip contains many different sensors, as shown in Fig. 6.The access waveguides end in grating couplers allowing for a connection to fiber optics.The fundamental TE mode was chosen using the polarizing property of the grating couplers and optimizing the input polarization for maximum transmission.Five ring resonators are coupled to a common bus waveguide, only four rings are open to the environment (circular opening).The fifths sensor acts as a reference.In the middle of the picture the electrical contact pads are connected using a probe head.

Experimental setup
The measurement setup consists of an Agilent 8164B frame and 81960A tunable laser source operating in fixed wavelength mode at 1550 nm.The sensor chip is mounted on a temperature controlled stage, which regulates the temperature to 25°C.In-and out-coupling of light is done with the help of standard single mode fibers and grating couplers.An optical power meter (Agilent 81636B) is used to detect the transmitted light intensity and a Zurich Instruments MFLI performs A/D-conversion and recording of the data.The AC signal is provided by a TOE 7706 signal generator.For tuning and multiplexing of the ring resonators on chip, a TDM-like control system has been built up.It provides four output channels, so that four ring resonators can be controlled during the experiments (see Fig. 3).

Experimental procedure
A first demonstration of the sensor system is given by a diffusion experiment.The refractive index change caused by a droplet of saline solution diffusing in an initial drop of water is spatially resolved.The saline solution consists of 5g sodium chloride dissolved in 100 ml of deionized water.According to [29] this solution possesses a refractive index of 1.3241 at room temperature, whereas deionized water has a refractive index of 1.3164.The implementation of the experiment is divided into two steps.At first the surface of the chip is initialized.Therefore a drop of 5 µl deionized water is placed on the chip surface with a chemical dropper and expanded over the whole ring resonator array, while the optical coupling regions for the input and output are not wetted.The separation of the deionized water and the electrical contacts are ensured by thin barriers of silicone adhesive stripes No.
5302A from Nitto Denko.The second step begins with the start of the data recording.The transmission and therefore the refractive index on the surface of each ring is measured 100 times per second, applying a modulation signal with 4.5 V peak-voltage.A sample of 5 µl saline solution is added on the left side of the array to the deionized water and the refractive index change is recorded.Figure 7 shows the referencing and read out scheme of the sensor array used for the experiment as well as a picture of the sensor wetting during an experiment.

Results and discussion
It is expected that a temporal difference in sensor responses due to diffusion of the saline solution occurs, so that the change in refractive index starts at ring 1 and ends at ring 3. Furthermore every ring of the ring resonator array should detect the same total change in refractive index in the end because the sensing area of each ring is designed equally. is highest for ring one, followed by ring 2 and ring 3, which is correlated with a dilution of the saline solution with time.The level overshoot of ring 1 and 2 might be caused by the high level of saline solution on the left side of the array due to the added drop at the beginning of the experiment, which is diluted as is spreads rightwards through the deionized water and leads to a decreasing refractive index over time at the position of ring 1 and 2. According to diffusion constants in the literature [29] a gradient used in this experiment compensates within the order of seconds in aqueous solutions for a distance of one millimeter.Ring 3, the very right ring of the array, shows a slower and steady increase in refractive index.Furthermore, the time delay of sensor responses shows a nonlinear relation to their spatial distance.Though there is no significant variation in ring resonator design, the resonator responses show slight deviations.While the responses of ring 1 and ring 3 converge after approximately 6 seconds, ring 2 mainly keeps below these levels, which is attributed to the simple set-up of this experiment.
Besides the sensor characteristics, particularly important factors that limit the measurement accuracy is the A/D-conversion of the photodiodes signal.Because noise decreases with increasing sampling rate, we chose a sampling rate of 200 kSa/s (which is the maximum sampling rate of the proposed system), causing a standard deviation of 5 × 10 6 • , mol n compared to a standard deviation of ~4 × 10 5 • mol n for 50 kSa/s.Additionally, the resolution of the A/D-conversion results in a quantization of the signal, limiting the least detectable change in sensor response.Thermal crosstalk must be considered as another source of noise during thermo-optical tuning.Considering the applied modulation voltage, the structure of the sensors and the distance of 1400 µm between the resonators, this influence is negligible.
Theoretical considerations based on a thermal equivalent circuit of the material stack of our chips show that a maximum scanning frequency of 20 kHz is possible for thermo-optic modulation.However, current limitations in terms of the sampling rate of our ADC restricts the maximum scanning frequency to 100 Hz.This seems comparable to fast TLS scanning speeds of up to 200 nm/s, but taking into account scanning repetition speeds of some Hz for such devices, the proposed TDM system combines both, fast scanning and repetition at once.

Conclusion
A bio sensor system based on a TDM-like modulation to drive and read out serial arrays of ring resonators realized in SOI technology is demonstrated.This approach enables a fast multi-parameter analysis and operates at a fixed wavelength.The system only requires cheap

Fig. 1 . 22 1
Fig. 1.Coupling model of a single ring resonator in all-pass configuration.

.
effective refractive index of the ring is used to detect the resonance condition at res n , the sensor is operating at a fixed wavelength 0

2 (
heating up the silicon, causing a change in refractive index.

Fig. 2 .
Fig. 2. Principle of a fixed wavelength modulation approach on a single ring: the resonances in the transmission spectrum of the ring resonator are shown for several higher orders .m An AC voltage V(t) is applied to the electrode of the ring resonator, where the resulting power P(t) modulates the effective refractive index until two resonances are detected.Changes in refractive index due to molecular binding mol n  can be estimated by tracking the resonance

Fig. 4 .
Fig. 4. Measured data stream containing two resonances for each ring in a fix scheme and the modulation signal of ring 1. Slight variations in extinction, FSR and Q-factors between the rings are caused by technological deviations of the structures, while variations in res n  are due

Fig. 5 .
Fig. 5. Schematic (a) and SEM image (b) of a triangular ring resonator: Filler structures, metal layer for heaters and buffer oxide on top of the waveguide structures, which show through the backend stack.The sensing region (circular opening) and the folded heating wire (right) are visible.The remainder of the ring resonator is hidden under the back-end.(c) SEM image of the heating regions cross-section, showing the aluminum electrodes placed above the waveguide.(d) SEM image of the nanowire waveguide cross-section in the sensing region, cladded with air.

Fig. 6 .
Fig. 6.Top view of a sensor chip: One chip possesses several sensors with different geometries and waveguide types.Here, sensor array marked with no. 8 is the device under test.Optical coupling occurs through fibers using grating structures on the far left and right of the chip.Five ring resonators are coupled to a common bus waveguide, only four rings are open to the environment (circular opening).The fifths sensor acts as a reference.In the middle of the picture the electrical contact pads are connected using a probe head.

Fig. 7 .
Fig. 7. Left: spatial and read out configuration as used during the diffusion experiments, right: picture of a device under test taken through the eyepiece of an optical microscope.The DIwater drop for initialization can easily be seen.
Figure 8 shows the refractive index change of ring 1 to 3 caused by the drop of saline solution.The fourth ring acts as the reference and has no direct contact to the liquid under test due to its oxide cladding.The data was smoothed by an 8 point FFT filter.Ring 1 is the first to show a step response, followed by ring 2 after approx.one second with an overshooting refractive index level.The change / mol d n dt 

Fig. 8 .
Fig. 8. Left: Refractive index change for one sensor array under test.The leveling at zero was achieved by an initial drop of deionized water with a volume of 5 µl, which was then mixed with a drop of saline solution of concentration and 5µl volume.Right: zoom, reference clearing and offset for a better view onto the sensor response delay.