Open Access
Add to BibSonomyAbstract
In this paper, we propose and analyze novel ring resonator based bio-chemical sensors on silicon nanowire optical waveguide (SNOW) and show that the sensitivity of the sensors can be increased by an order of magnitude as compared to silicon-on-insulator based ring resonators while maintaining high index contrast and compact devices. The core of the waveguide is hollow and allows for introduction of biomaterial in the center of the mode, thereby increasing the sensitivity of detection. A sensitivity of 243 nm/refractive index unit (RIU) is achieved for a change in bulk refractive index. For surface attachment, the sensor is able to detect monolayer attachments as small as 1 Å on the surface of the silicon nanowires.
©2011 Optical Society of America
1. Introduction
Optical biosensors have attracted considerable attention in the last decade because of their promise to contribute to major advances in medical diagnosis, environmental monitoring, drug development, quality control, and homeland security [1]. Compared to electrical transducers, optical sensors provide significant advantages because of their small size, immunity to electromagnetic interference, ease of multiplexing using wavelength encoding, and capability of remote sensing [2]. Optical sensors can be broadly characterized in two categories: fluorescence based detectors and label-free detectors. In fluorescence based detectors, the target molecules are labeled with fluorescent tags such as dyes and the fluorescence is detected in presence of the targeted molecule. This allows for extremely sensitive detection down to a single molecule [3]. However, the process is laborious and may also affect the function of the biomolecules. Further, precise quantitative measurements are difficult as the number of flourophores attached to the targeted molecules cannot be controlled [4]. In contrast, in label-free detection the targeted molecules are detected in their natural form. The targeted molecules are surface attached to the optical sensor using probe molecules and the attachment is detected by measuring the change in optical properties for the sensor. Sensors based on silicon-on-insulator (SOI) photonic wire waveguides have attracted considerable attention because of compatibility with CMOS fabrication and possibility of integrating detection and decision on the same chip leading to ”laboratories on a chip” [2]. Further, guided-wave sensors allow for integration of multiple sensors on a single chip. As such, different sensors based on directional couplers [5], Mach-Zehnder interferometers [6], Bragg grating based Fabry-Perot resonators [7], microdisks [8], microtoroids [9], photonics crystal cavities [10], microring resonators [11], and slot waveguides [12] have been demonstrated. In these sensors, the targeted molecule is probed by light guided through a solid medium using the evanescent field. The lower refractive index surrounding medium (typically water with refractive index ∼ 1.3253) is displaced by higher refractive index organic molecules (n ∼ 1.45 – 1.6) changing the effective index of the propagating mode resulting in a spectral shift of the resonant cavity which can be measured directly. Mach-Zehnder interferometers, Fabry-Perot resonators, microdisks, photonic crystal cavities, and ring resonators have been demonstrated using silicon photonics. Mach-Zehnder interferometers suffer from low Q and the requirement of large interaction length to increase the sensitivity. Bragg-reflectors for Fabry-Perot resonators are difficult to fabricate in high-index contrast materials resulting in high insertion losses. Photonics crystal cavities are also difficult to produce with low propagation losses and it is difficult to couple light in and out of these waveguides reproducibly. Microdisks have higher whispering gallery modes which can overlay on the fundamental characteristics making detection difficult. As such, ring resonators offer most attractive solution as they provide low insertion loss, single mode cavities, and small form factors. Ring resonators offer a compelling solution as multiplexing different sensors on a single chip using wavelength is possible by simply changing the diameter of the ring in the resonator. Biochemical sensors based on SOI photonic waveguides have been studied extensively. In [11], a 70 nm/RIU sensitivity was achieved for bulk changes of refractive index and a 625 pm shift of wavelength was achieved for label-free sensing of proteins. The sensitivity can be further increased by using a slot-waveguide which is an optical waveguide guiding light in a subwavelength-scale low refractive index region sandwiched between two ridges of high index. This enhances the transverse electric field in the slot thereby increasing the interaction of the optical field with the targeted molecules. An increased sensitivity of 298 nm/RIU was achieved with a foot-print of 13 μm×10 μm [12]. However, the problem with slot waveguides is the difficulty with introduction of fluids within the slot region. Further since the slot waveguide works in the quasi-TE mode, it is lower modal index waveguide compared to SOI waveguides.
In a previous work [13], we have proposed and analyzed a new kind of optical waveguide consisting of arrays of silicon nanowires (SiNWs) where the diameter of the SiNW is smaller than 75 nm for 1550 nm wavelength. A fabricated waveguide is shown in Fig. 1. A SOI wafer is etched into vertical SiNWs. For this sample, it consists of 9 rows of 40 nm diameter nanowires with a pitch of 100 nm. The length of the nanowires is 800 nm. If the diameter is less than 75 nm, the diffraction of light through the SiNWs is limited (provided the electric field is polarized along the length of the nanowires) and the medium starts to behave like an effective-index medium, thereby guiding light through the structure with very low propagation losses (< 0.2 cm −1). Vertical confinement is provided by the refractive index contrast between the silicon and the insulator. Unlike, photonic bandgap structures, the nanowires do not need to be aligned in a crystal and can be randomly arranged with minimal excess losses [13]. Further, the scattering due to side wall roughness results in minimal excess losses [14]. We have also shown that waveguides with bends of radii smaller than 5 μm can be designed on the SNOW structure with low loss (less than 0.06 dB per 360° turn) [14]. Such geometries have been achieved using electron-beam lithography and BOSCH etching of silicon [15]–[16] mainly for solar cells. Further, we have been able to fabricate SNOW structures with nanowire diameters and pitch as small as 15 nm and 75 nm respectively. While we are still in process of measuring optical properties of these waveguides, previous experimental works show that it should be feasible to guide light through these structures. In [18], Bock et. al. demonstrated light guidance through sub-wavelength grating structures over a waveguide crossing. Optical loss as low as 0.023 dB/crossing was achieved with a waveguide which worked as an effective-index medium, similar to SNOW. Further in [19], gain and stimulated emission was observed in nanopatterned silicon. The device consisted of 100 nm thick arrayed subwavelength structures on a SOI wafer cleaved into 1 mm long devices, again working like an effective index waveguide. Fabry-Perot characteristics in the emission demonstrated guidance of light in the 100 nm patterned silicon layer of the device and corresponded well with the effective index calculations. These experiments along with experimental fabrication of SNOW strongly suggest that guidance of light should be possible with the structures.
Even over a bend, the effective index approximation works well. This allows for designing and building of ring resonators on the SNOW especially for biochemical sensors. The advantage of SNOW is apparent from the fact that it is a hollow core waveguide and thus it is possible to introduce the bio-chemical agents in the region of highest optical field intensity. In this paper, we propose a ring resonator structure with SNOW in the ring excited by a SOI bus waveguide. We show that the sensitivity is increased by an order of magnitude compared to the SOI waveguides while achieving a compact structure.
2. Sensor structure
In order to be able to fabricate ring resonators, it is important for the SNOW region to support waveguide bends. Figure 2 shows the radiation loss through a 360° turn in the SNOW region as the radius of the bend is changed. For this simulation, the polarization is along the length of SiNWs and the wavelength of operation is 1550 nm. The diameter of the silicon nanowires is 50 nm and the pitch between them is 75 nm, similar to what we had previously proposed [13]. The height of the nanowires is 700 nm. The SNOW region has an effective index of 2.2 when air is surrounding the medium [13]. The width of the SNOW region is 650 nm. At this width, the second order mode in the ring is supported but does not get excited because of the symmetry rules. Further, the radiation loss for the second order mode over the bend is appreciably higher compared to the fundamental mode. Also plotted in the figure, is the radiation loss when the effective-index approximation shown in [13] is used and SNOW is approximated by a rib waveguide. All the simulations are done using the finite-difference time domain (FDTD) method with a grid size of 2 nm for SNOW and a grid size of 10 nm for the effective-index approximation. The simulations were done in 2-dimensions by using effective index method in the vertical direction to decompose a 3-D structure into planar waveguides. The method was tested by comparing results using 3-D simulations with 2-D for few samples. For all the simulations, the electric field was polarized along the length of the nanowires which corresponds to quasi-TM polarization for conventional waveguides. At 700 nm waveguide height, the optical mode is highly confined in the vertical direction and the effective index approximation works well. From Fig. 2 it is clear that the loss through the bend is mainly dominated by the caustic radiation and not by the radiation due to scattering from the individual nanowires. For a radius of 5 μm, the radiation loss over a 360° turn is 4.6 × 10−4 dB. These simulations show the appropriateness of using SNOW for bends and allow for fabricating ring resonators on the structure.
Fig. 2 (Color online) Radiation loss for various radii of bend SNOWs over a 360° turn when the electric field is parallel to the length of nanowires at wavelength of 1550 nm.
The proposed structure is shown in Fig. 3. A SOI waveguide is used as a bus waveguide feeding into a ring consisting of SNOW with parameters described above. Use of SOI waveguides allows for conventional input and output optical coupling into the structure resulting in low insertion losses. The bus waveguide has a width of 100 nm and has the same height (700 nm) as the SNOW ring. At this width, the bus waveguide is purely single mode. Separation between the waveguides is adjusted to 100 nm between the end of the bus waveguide and the first nanowire in the SNOW and achieves critical coupling condition for the ring resonator.
Figure 4 shows the lateral cut of the FDTD propagation of the electric field through the ring resonator at a wavelength of 1550 nm (not the resonance wavelength) over a few cycles within the ring. One can clearly see that the SNOW ring is guiding the electric field with very little radiation happening in the structure.
Fig. 4 (Color online) Lateral cut of the FDTD propagation of electric field through the SNOW ring resonator.
Figure 5 shows the lateral cut of the electric field through the SNOW structure for the parameters defined above. The lateral cuts for a straight SNOW and over a bend are shown. Lateral cut for the 200 nm wide straight SOI waveguides is also shown. Within the effective-index waveguide, the confinement factor for the SNOW is 94 % resulting in the low bend losses as shown in Fig. 2. The confinement factor for the optical mode within silicon in the SNOW region is only 32 % whereas for a 200 nm SOI waveguide, it is 76 %. The optical mode is better guided in the SNOW region as compared to the 200 nm SOI waveguide, and the modal power in the surrounding region is larger in the SNOW as compared to the 200 nm SOI. Thus, it should be possible to increase the sensitivity while still achieving compact devices.
Fig. 5 (Color online) Lateral electric field cuts for the SNOW with width of 650 nm for both bend, with a radius of 5 μm, and straight structures and the silicon waveguide with width of 200 nm at wavelength of 1550 nm.
3. Sensor characteristics for bulk refractive index change
We first calculated the response of sensor for bulk change in the refractive index of the surrounding medium. The SNOW structure is compared with a SOI waveguide ring resonator where the width of the SOI waveguide is 200 nm, similar to [11]. The geometric parameters for the two compared devices are shown in Table I. For the first set of simulations, the refractive index of the surrounding medium was changed and the effective index of the guided optical mode through SNOW was calculated. Figure 6 shows the change of effective index of the optical mode as a percentage with respect to the value of effective index as the surrounding refractive index is changed from 1.0 to 1.6 for the SNOW and the 200 nm SOI waveguide at a wavelength of 1550 nm. For the SNOW, the effective index changes by a factor of approximately 4 larger as compared to a 200 nm SOI waveguide for the same change in surrounding refractive index. In a ring resonator the change of the resonance wavelength is approximately given by [12]:
where Δneff is the change of the effective index due to the change of the refractive index of the surrounding medium, λ is the initial resonance wavelength and ng is the group index. This suggests that an improvement in sensitivity of 4 is expected if one uses a SNOW ring resonator as compared to a 200 nm wide SOI waveguide resonator.Fig. 6 (Color online) Change of effective-index as a percentage for SOI and SNOW ring resonator as the surrounding index is changed.
Figure 7 shows the response of the ring resonators for SNOW and SOI when the surrounding index is changed from 1.0 to 1.05. A wavelength shift of 12.2 nm is achieved for the SNOW ring resonator resulting in a sensitivity of 243 nm/RIU. For the 200 nm SOI waveguide, the wavelength shift for the same refractive index change is 3.14 nm resulting in a sensitivity of 63 nm/RIU. This compares well with the experimental value of 70 nm/RIU for a slightly higher bulk refractive index [11]. An improvement by a factor of 3.9 is seen in the sensitivity for the SNOW ring resonator compared to the 200 nm wide SOI waveguide for bulk change of refractive index.
Fig. 7 (Color online) (a) Shift of resonance wavelength for SOI ring resonator when the surrounding index is changed from 1 to 1.05. (b) Shift of resonance wavelength for SNOW ring resonator when the surrounding index is changed from 1 to 1.05.
We also studied the effect on sensitivity as the width of the SNOW region is changed. The ring resonator coupling was adjusted individually to achieve critical coupling. Figure 8 shows the change of sensitivity for the SNOW ring resonator as the width of the effective waveguide is varied. The diameter and pitch for the SiNWs are kept the same. An increase of sensitivity is observed when the waveguide width is decreased, reaching a value of 335 nm/RIU for a width of 300 nm. The surrounding index is again changed from 1 to 1.05. An improvement by a factor of 5.3 is observed as compared to the SOI waveguide. The behavior exhibited by the SNOW ring resonator is similar to that of the SOI ring resonators as the width is decreased. This is because of the increased evanescent field as the width is decreased.
Fig. 8 (Color online) Change of sensitivity as width of SNOW is changed. Sensitivity for a 200 nm SOI is also shown.
4. Sensor characteristics for surface attachments
In optical sensors, surface sensing plays an important role for a wide range of biochemical applications including DNA hybridization, antigen-antibody reactions, protein attachments etc. A layer of receptor molecules is surface attached to optical sensor and selective attachment is done for the targeted molecule. Since the refractive index of the molecules is different from the surrounding medium which is typically water based, a change of index happens at the surface of the sensor which is measured for detecting the presence of the molecule. The SNOW ring resonator was simulated for surface attachment of the molecules. A molecule layer with the test thickness was assumed to be attached the surface of the SiNWs. Water was considered as the surrounding medium with a refractive index of 1.325 at a wavelength of 1550 nm [20]. The refractive index of the molecule attached is considered to be 1.6, similar to 3-aminopropyltriethoxysilane (APTES) which we have measured previously [17] and controllably attached different thickness on the surface. Structures summarized in Table I were compared. Figure 9(b) shows the response of the SNOW sensor when 0.1 nm and 1 nm of the molecule layers are uniformly attached to the surface of the nanopillars. Wavelength shift of 0.35 nm and 3.1 nm and is achieved with a 0.1 nm and 1 nm attachment of the molecule. For these thin layers, the surface attachment increases linearly with the thickness of the molecule layer. Figure 9(a) shows the response if the 200 nm SOI waveguide was used for the detection. For the SOI waveguides, surface attachment was assumed over all the exposed surfaces of silicon including the sides and the top of the waveguide. Only a 1 nm layer attachment was considered. A wavelength shift of 0.15 nm is achieved for the attachment of 1 nm layer thickness. This shows an improvement by a factor of 20.5 with the SNOW ring resonator.
Fig. 9 (Color online) (a) Wavelength shift as 1 nm of layer is surface attached to the SOI waveguide (b) Wavelength shift as 1 nm of layer is surface attached to the SNOW ring. Wavelength shift due to a layer attachment of 0.1 nm is also shown.
Table I. Device parameters for simulated SNOW and standard SOI ring resonators.
The dependence of the width for the SNOW was also considered. Figure 10 shows the percentage change in the effective index of the SNOW structure as the waveguide width is changed from 300 nm to 1000 nm for a 1 nm thickness of the attached molecule layer. As opposed to the change in bulk refractive index, the behavior is different and the sensitivity increases as the width is increased. This is because the sensor is not working in the evanescent field but within the core of the optical mode. As the width is increased, the optical mode gets more confined within the SNOW region resulting in higher interaction with the surface attached material.
Fig. 10 (Color online) Change in the percentage of the effective modal index for 1 nm thickness of attached layers as the SNOW width is increased.
5. Conclusion
We have proposed a new optical sensor based on SNOW ring resonator and carried out a detailed investigation of its modal properties. The SNOW ring resonator consists of a SOI bus waveguide coupled into a ring waveguide consisting of closely etched silicon nanowires acting as an effective index waveguide. We have compared the proposed sensor to the SOI photonic wire based sensors. For bulk refractive index changes, an improvement by a factor of 5.3 is achieved with the SNOW sensor compared to that of the SOI. For surface attachment, an improvement by a factor of 20.5 is achieved for the SNOW sensor.
Acknowledgments
This work was supported by the Canadian National Science and Engineering Research Council (NSERC), Ontario Centres of Excellence (OCE) and DALSA Corp.
References and links
1. 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, 8–26 (2008). [CrossRef] [PubMed]
2. S. Balslev, A. M. Jorgensen, B. Bilenberg, K. B. Mogensen, D. Snakenborg, O. Geschke, J. P. Kutter, and A. Kristensen, “Lab-on-a-chip with integrated optical transducers,” Lab Chip 6, 213–217 (2006). [CrossRef] [PubMed]
3. W. E. Moerner, “New directions in single-molecule imaging and analysis,” Proc. Natl. Acad. Sci. U.S.A. 104, 12596–12602 (2007). [CrossRef] [PubMed]
4. W. G. Cox and V. L. Singer, “Fluorescent DNA hybridization probe preparation using amine modification and reactive dye coupling,” Biotechniques 36, 114–122 (2004). [PubMed]
5. B. J. Luff, R. D. Harris, J. S. Wilkinson, R. Wilson, and D. J. Schiffrin, “Integrated optical directional coupler biosensor,” Opt. Lett. 21, 618–620 (1996). [CrossRef] [PubMed]
6. F. Prieto, B. Sepúlveda, A. Calle, A. Llobera, C. Domínguez, A. Abad, A. Montoya, and L. M. Lechuga, “An integrated optical interferometric nanodevice based on silicon technology for biosensor applications,” Nanotechnology 14, 907–912 (2003). [CrossRef]
7. W. C. L. Hopman, P. Pottier, D. Yudistira, J. V. Lith, P. V. Lambeck, R. M. De La Rue, A. Driessen, H. J. W. M. Hoekstra, and R. M. de Ridder, “Quasi-one-dimensional photonic crystal as a compact building-block for refractometric optical sensors,” IEEE J. Sel. Top. Quantum Electron. 11, 11–16 (2005). [CrossRef]
8. E. S. Hosseini, S. Yegnanarayanan, A. H. Atabaki, M. Soltani, and A. Adibi, “High quality planar silicon nitride microdisk resonators for integrated photonics in the visible wavelength range,” Opt. Express 17, 14543–14551 (2009). [CrossRef] [PubMed]
9. A. M. Armani and K. J. Vahala, “Heavy water detection using ultra-high-Q microcavities,” Opt. Lett. 31, 1896–1898 (2006). [CrossRef] [PubMed]
10. L. Rindorf, J. B. Jenson, M. Dufva, L. H. Pedersen, P. E. Hiby, and O. Bang, “Photonic crystal fiber long-period gratings for biochemical sensing,” Opt. Express 14, 8224–8231 (2006). [CrossRef] [PubMed]
11. K. D. Vos, I. Bartolozzi, E. Schacht, P. Bienstman, and R. Baets, “Silicon-on-insulator microring resonator for sensitive and label-free biosensing,” Opt. Express 15, 7610–7615 (2007). [CrossRef] [PubMed]
12. T. Claes, J. G. Molera, K. D. Vos, E. Schacht, R. Baets, and P. Bienstman, “Label-free biosensing with a slot-waveguide based ring resonator in silicon on insulator,” IEEE Photon. J. 1, 197–204 (2009). [CrossRef]
13. M. Khorasaninejad and S. S. Saini, “Silicon nanowire optical waveguide (SNOW),” Opt. Express 18, 23442–23457 (2010). [CrossRef] [PubMed]
14. M. Khorasaninejad and S. S. Saini, “Bend-waveguides on silicon nanowire optical waveguide (SNOW),” IEEE Photon. J. 3, 696–702 (2011). [CrossRef]
15. M. D. Henry, S. Walavalkar, A. Homyk, and A. Scherer, “Alumina etch masks for fabrication of high-aspect-ratio silicon micropillars and nanopillars,” Nanotechnology 20, 1–4 (2009). [CrossRef]
16. Y. J. Hung, S. L. Lee, and L. A. Coldren, “Deep and tapered silicon photonic crystals for achieving anti-reflection and enhanced absorption,” Opt. Express 18, 6841–6852 (2010). [CrossRef] [PubMed]
17. S. S. Saini, C. Stanford, S. M. Lee, J. Park, P. DeShong, W. E. Bentley, and M. Dagenais, “Monolayer detection of biochemical agents using etched-core fiber Bragg grating sensors,” IEEE Photon. Technol. Lett. 19, 1341–1343 (2007). [CrossRef]
18. P. J. Bock, P. Cheben, J. H. Schmid, J. Lapointe, A. Delge, D. X. Xu, S. Janz, A. Densmore, and T. J. Hall, “Subwavelength grating crossings for silicon wire waveguides,” Opt. Express 18, 16146–16155 (2010). [CrossRef] [PubMed]
19. S. G. Cloutier, P. A. Kossyrev, and J. Xu, “Optical gain and stimulated emission in periodic nanopatterned crystalline silicon,” Nat. Mater. 4, 887–891 (2005). [CrossRef] [PubMed]
20. E. S. Larsen, J. R. Meyrowitz, and A. J. C. Wilson, “Measurement of refractive index,” in International Tables for Crystallography (2006), Vol. C, Chap. 3.3, pp. 160–161. [CrossRef]
