An optical microcavity based on pillar arrays has been fabricated in Si/SiO2 material system. Transmission measurement was taken and a quality factor as high as 27,600 was observed. This cavity was tested for sensing applications by immersing into optical fluids with accurate refractive indices. For refractive index change of 0.01, a resonance peak wavelength shift of 3.5 nm was measured. We also compare cavities consisting of pillars with different aspect ratios.
©2009 Optical Society of America
A 2-dimensional (2D) photonic crystal (PhC) has emerged as one of the most important platforms to fabricate a high quality factor (Q) optical microcavity [1, 2, 3, 4]. It can dramatically increase, beyond its physical size, the effective interaction length between the optical field and the materials inside or close to the mode volume. This feature enables ultra compact and highly sensitive label-free optical sensors to detect bio/chem reactions [5, 6]. Most of the reported 2D PhC based microcavities are formed from periodic air holes. To reduce the leakage into the substrate, the material underneath the cavity is usually selectively removed to form a free-standing membrane. To the contrary, a pillar array based 2D PhC has been rarely researched for micro-cavity and sensing applications since it is difficult to form a similar free-standing membrane. However as proven in our previous work , with careful design, the leakage into the substrate can be decreased and a high Q pillar-array cavity is feasible. In this paper, we continue exploring the usage of such a cavity for sensing applications.
Compared with hole-array based PhCs, pillar arrays have several advantages for sensing applications. Firstly, a pillar array has a much larger percentage of void space. In this paper, a cavity with more than 80% of air space was fabricated and tested. The big open space inside the cavity makes it much easier for the infiltration of the analytes. Secondly, as opposite to the hole arrays, in a pillar array cavity, the void space is connected, giving a flowing channel for the fluid carrying the analytes. Thirdly, due to the large and connect void space, the electric field of the confined mode tends to locates more out of the pillars, which will increase the overlapping with the analytes and then enhance the sensitivity.
In the following, we first describe the fabrication and measurement of a heterogenous pillar array based microcavity in Si/SiO2 material system. Optical fluids with accurate refractive indices were applied to the cavity. The peak wavelength shift was recorded and used to characterize the sensitivity of the cavity to the ambient refractive index change. In the last section, we analyze the dependence of the quality factor and the sensitivity of the cavity on the aspect ratio of the pillars.
2. Fabrication and theoretical calculation
Although seldom used as a microcavity, pillar-array based 2D PhCs have been successfully fabricated with top-down microfabrication process including advanced lithography and dry etching and used as optical interconnection components [8,9,10,11]. It is worth noting that high aspect ratio pillar array can also be grown with bottom-up methods such as chemical vapor deposition method  or metal-organic vapor phase epitaxy , catalyzed by metal particles . In this letter, we adopted top-down method with e-beam lithography and reactive ion etching (RIE).
The SEM images of the fabricated structure are shown in the Fig. 1. A buffer SiO2 layer (2 μm thick) and then a hydrogenated amorphous silicon (α-Si:H) guiding layer (730 nm thick) are deposited with plasma enhanced chemical vapor deposition on the crystalline Si substrate. The input/output waveguides and pillar array were defined by e-beam machine in a negative tone e-beam resist (MaN 2403) mask. The patterns were then transferred to the Si:H layer with inductively coupled plasma (ICP) RIE. SF6/C4F8 combined chemistry was used in the etching process and optimized to obtain smooth and vertical side walls for the pillars. After the etching, the e-beam resist mask was removed. The final device includes an array of finite-height (730 nm) pillars with diameter of about 242 nm, and two tapered coupling waveguides, which are tapered up from 2 μm to 4.5 μm to match the mode of the cavity.
A heterogenous design was employed to form a high Q cavity and reduce the vertical leakage. As shown in Fig. 1(a), the cavity consists of “core” and “wall” regions. The interface between them is perpendicular to ΓM direction. In the core region, it is a square lattice with lattice constant a around 483 nm. The period along the ΓM is √2a = 683 nm. In the wall region, along the interface the period keeps the same, while perpendicular to the interface, it is enlarged from 683 nm to (√2 + 0.2)a= 781 nm. As analyzed in [1, 3] and our previous work [7, 14], due to the misalignment of the band edge, the mode close to the band edge of the core region will be highly confined by the wall.
Three dimensional (3D) finite-difference time-domain (FDTD) method was used to calculate the mode pattern and Q of the fundamental mode confined in the above cavity. The vertical cross section of the schematic PhC slab used in the calculation is shown in Fig. 2(a). In the calculation, the spatial grid size is 1/16 a, where a is the period of the lattice. To absorb outgoing waves, a perfectly matched layer (PML) boundary condition is used. Regarding the details of the calculation, please refer to our previous work . The vertical and horizontal cross sections are shown in Fig. 2(b) and (c), from which it is clear that the mode is highly confined in the cavity both vertically and horizontally. The calculated Q is about 65,000.
3. Optical measurement
Optical transmission measurement was taken to characterize the fabricated microcavity. As shown in Fig. 3(a), tunable laser sources (JDS Uniphase: SWS15101 from 1460 to 1600 nm and SWS16101 from 1510 to 1650 nm) were used. The light was coupled into the input waveguide by a lens fiber. A polarization controller was used to maximize the TM component of the incident light. The transmitted light was collected by an 60× objective lens and filtered by a polarization beam splitter before entering a photodetector. The transmission spectrum through the cavity shown in Fig. 1 is displayed in the figure. The measured Q is 27,600, about 2.4 times smaller than the theoretical value. The discrepancy may come from two sources, the fabrication defects and the material loss of the amorphous Si:H .
To further measure the sensitivity of the cavity, we applied optical fluids (Cargille Laboratories) with accurate refractive indices to the cavity. The applied optical fluid has a coverage area and thickness, which are several orders of magnitude larger than the modal extent of the cavity mode. Therefore it can be viewed as a semi-infinite homogeneous background covering the pillar array and replacing air there. Since the optical field extends into the space around the pillars, it then overlaps with the optical fluid and the resonance wavelength will shift to longer position. The value of the shift will depend on the refractive index of the optical fluid. Six optical fluids with refractive indices from n = 1.392 to n = 1.442 in increments of Δn = 0.01 were used in sequence. Before another optical fluid was applied, the chip was rinsed in toluene to remove the previous one. It was then rinsed in acetone, methanol, isopropyl alcohol, and deionized water to remove the debris completely and dried with N 2 gas. The transmission spectra were measured and are shown in Fig. 3(b). Lorentzian fitting was used to determine the resonance peak positions, which are shown in Fig. 3(c). As seen from figure, the resonance peak shifts up linearly by about 3.5 nm for Δn = 0.01. which is a bit larger than the calculated value of about 3.0 nm. Since the full width at half maximum (FWHM) of the transmission peaks when applying optical fluids is about 0.1 nm, if we assume that it is possible to distinguish two peaks when they are out of each other’s width, with such a pillar array, we can approximately measure the ambient refractive index difference of about 3 × 10-4. After applying the optical fluid of n = 1.442, we cleaned the chip carefully and reapplied the optical fluid of n = 1.392. The transmission spectrum is shown in dashed line in Fig. 3(b). It reverts to almost exact wavelength position of the transmission peak when first applying optical fluid of n = 1.392, which testifies the effectiveness of the cleaning procedure.
4. Dependence on aspect ratio of pillars
When other parameters keep fixed and the aspect ratio of pillars increases, the electric field of the mode tends to locate more inside the pillars and the normalized frequency of the mode becomes smaller and further below the light line of the substrate. As a result, the vertical leakage will decrease and the quality factor will increase. However, as the quality factor increases, the cavity will become less sensitive. It is because that the overlap between the electric field of the mode and the dielectric materials introduced close to the pillars becomes smaller.
As an example, we can look at the cavities presented in Ref.  and above in Section 2. When normalized over the period of the lattice, these two cavities have similar parameters except for the aspect ratio of the pillars. As the aspect ratio increases from 1.7 to 3.0. The percentage of the electric field energy in pillars increases from 64.1% to 79.8%. We used 3D FDTD calculation to obtain the mode field pattern and the following formula to calculate the percentage of the electric field energy in pillars.
where the Vpillar refers to the space inside the pillars and V refers to whole mode volume. εr(r) and E(r) are the real part of the dielectric constant and electric field at spatial position r respectively. The measured quality factor increase from 7,350 to 27,600. When both of the cavities were immersed into an optical fluid with refractive index of 1.392, the shift of the peak wavelength was measured to be 167 nm and 130 nm for the aspect ratio of 1.7 and 3.0. It is clearly seen from this example that the quality factor and the sensitivity goes in opposite directions when aspect ratio changes. This trend can be further illustrated in Fig. 4 with aspect ratio increasing to 10.
We have demonstrated an ultra-sensitive optical sensor that is based on silicon pillar arrays. A Q as high as 2.7 × 104 was measured. The cavity has 80% connected void space and thus is suitable for sensing applications. The confined mode has a large percent of electric field energy locating in the void space and is very sensitive to the analyte introduced there. Based on the experiments, the cavity is capable to resolve a refractive index difference of 3 × 10-4 Increasing the height of the pillars can increase Q factor, while reducing the sensitivity of the cavity. For certain application, it is needed to find the balancing point between them.
We gratefully acknowledge financial support from NSERC, CIPI, OCE, CSA, AFOSR and ECTI.
References and links
2. K. Srinivasan, P. E. Barclay, O. Painter, J. Chen, A. Y. Cho, and C. Gmachl, “Experimental demonstration of a high quality factor photonic crystal microcavity,” Appl. Phys. Lett. 83, 1915–1917 (2003). [CrossRef]
3. T. Tanabe, M. Notomi, E. Kuramochi, A. Shinya, and H. Taniyama, “Trapping and delaying photons for one nanosecond in an untrasmall high-Q photonic-crystal nanocavity,” Nat. Photonics 1, 49–52 (2007). [CrossRef]
4. A. R. M. Zain, N. P. Johnson, M. Sorel, and R. M. De La Rue, “Ultra high quality factor one dimensional photonic crystal/photonic wire micro-cavities in silicon-on-insulator (SOI),” Opt. Express 16, 12084–12089 (2008). [CrossRef] [PubMed]
5. E. Chow, A. Grot, L.W. Mirkarimi, M. Sigalas, and G. Girolami, “Ultracompact biochemical sensor built with two-dimensional photonic crystal microcavity,” Opt. Lett. 29, 1093–1095 (2004). [CrossRef] [PubMed]
7. T. Xu, N. Zhu, M. Y.-C. Xu, L. Wosinski, J. S. Aitchison, and H. E. Ruda, “A pillar-array based two-dimensional photonic crystal microcavity,” Appl. Phys. Lett. 94, 241110 (2009). [CrossRef]
8. M. Tokushima, H. Yamada, and Y. Arakawa, “1.5-μm-wavelength light guiding in waveguides in square-lattice-of-rod photonic crystal slab,” Appl. Phys. Lett. 84, 4298 (2004). [CrossRef]
9. X. Ao, L. Liu, L. Wosinski, and S. He, “Polarization beam splitter based on a two-dimensional photonic crystal of pillar type,” Appl. Phys. Lett. 89, 171115 (2006). [CrossRef]
10. A. Kok, E. J. Geluk, B. Docter, J. van der Tol, R. Nözel, M. Smit, and R. Baets, “Transmission of pillar-based photonic crystal waveguides in InP technology,” Appl. Phys. Lett. 91, 201109 (2007). [CrossRef]
11. S. H. G. Teo, A. Q. Liu, J. B. Zhang, M. H. Hong, J. Singh, M. B. Yu, N. Singh, and G. Q. Lo, “Photonic bandgap crystal resonator enhanced, laser controlled modulations of optical interconnects for photonic integrated circuits,” Opt. Express 16, 7842 (2008). [CrossRef] [PubMed]
12. U. Philipose, T. Xu, S. Yang, P. Sun, H. E. Ruda, Y. Q. Wang, and K. L. Kavanagh, “Enhancement of band edge luminescence in ZnSe nanowires,” J. Appl. Phys. 100, 084316 (2006). [CrossRef]
13. T. Måtensson, M. Borgström, W. Seifert, B. J. Ohlsson, and L. Samuelson, “Fabrication of individually seeded nanowire arrays by vapour liquid solid growth,” Nanotechnology 14, 1255–1258 (2003). [CrossRef]
14. T. Xu, M. S. Wheeler, S. V. Nair, H. E. Ruda, M. Mojahedi, and J. S. Aitchison, “Highly confined mode above the light line in a two-dimensional photonic crystal slab,” Appl. Phys. Lett. 93, 241105 (2008). [CrossRef]
15. T. Xu, S. Yang, S. V. Nair, and H. E. Ruda, “Nanowire-array-based photonic crystal cavity by finite-difference time-domain calculations,” Phys. Rev. B 75, 125104 (2007). [CrossRef]