Porous Silicon One-Dimensional Photonic Crystals for Optical Signal ...
1514
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 12, NO. 6, NOVEMBER/DECEMBER 2006
Porous Silicon One-Dimensional Photonic Crystals for Optical Signal Modulation Sharon M. Weiss, Member, IEEE, and Philippe M. Fauchet, Fellow, IEEE
Abstract—Porous silicon one-dimensional (1-D) photonic crystal microcavities are fabricated as building blocks for optical modulators. The microcavities consist of porous silicon layers with alternating low and high porosity. Since the pore size is much smaller than the incident wavelength of light, each porous silicon layer is treated as an effective medium with a specific effective refractive index that corresponds to the given porosity. Active tuning of the microcavity resonance is accomplished through electrical actuation of liquid crystals infiltrated into the pores. As the refractive index of the liquid crystals change in response to the external stimulus, the effective refractive index of the porous silicon layers changes and the resonance wavelength shifts accordingly. The resonance wavelength shift creates an optical signal intensity modulation at a particular wavelength. Extinction ratios greater than 10 dB are achievable for these devices. The performance characteristics of the porous silicon photonic crystal microcavities can be tailored to specific applications based on the choice of infiltrated optically active material and the quality factor of the device. Index Terms—Liquid crystal, modulation, photonic crystal, silicon.
I. INTRODUCTION HE past few years have marked widespread progress in the area of silicon photonics. Specifically, with regard to components for optical interconnects, there have been demonstrations of silicon Raman lasers [1], [2], efficient and compact optical coupling schemes [3], electrically driven silicon modulators [4], [5], low-loss silicon-on-insulator waveguides [6], and high speed detectors [7]. In order to continue to meet the demand for increasing information bandwidth while maintaining high speed, low power consumption, and low-cost components, long-term solutions to the interconnect bottleneck need to be addressed on several levels, from interboard communication to intrachip connections. Silicon has not traditionally been considered a viable optical material due to its indirect bandgap, which inhibits light emission efficiency, and its low electro-optic coefficient. However, due to the extensive infrastructure that exists for silicon processing, the potential advantages derived from achieving optical functionality on a silicon platform are a strong motivation to invest substantial resources into investigating new methods of “siliconizing” photonics [8].
T
Manuscript received November 17, 2005; revised August 21, 2006. This work was supported in part by the Air Force Office of Scientific Research under Grant F 49620-02-1-0376 and in part by Intel Corporation. S. M. Weiss was with the Institute of Optics, University of Rochester, Rochester, NY 14627 USA. She is now with the Department of Electrical Engineering, Vanderbilt University, Nashville, TN 37235 USA (e-mail: [email protected]). P. M. Fauchet is with the Department of Electrical and Computer Engineering, University of Rochester, Rochester, NY 14627 USA (e-mail: [email protected]). Digital Object Identifier 10.1109/JSTQE.2006.884083
Photonic crystals are engineered optical structures that have been demonstrated as means of controlling light propagation. Demonstrations of silicon photonic crystals as waveguides [9], add/drop filters [10], and modulators [11] form the basis of future optical interconnect building blocks. Since photonic crystals localize electromagnetic fields, compact photonic components can be realized. Moreover, the capability to infiltrate optically active materials into the air holes of photonic crystal structures enables enhanced optical functionality beyond that of the traditional silicon devices. Several silicon-based material systems have been explored for optical applications. Intel’s efforts, for example, have been focused on standard planar silicon device configurations for light sources [2] and modulators [4], whereas silicon germanium is the popular choice for detectors at near-infrared wavelengths [7]. Several groups have also explored the possibility of enhancing silicon’s optical capabilities by reducing the feature size. Silicon nanocrystals are being studied for their light emission properties [12], [13], and the large achievable refractive index range of porous silicon is being exploited for controlling the reflectance and transmission profiles of optical modulator building blocks [11]. II. POROUS SILICON AS AN OPTICAL MATERIAL A. Fabrication by Electrochemical Etching Porous silicon has been investigated as an optical material since 1990 when Canham first reported visible photoluminescence [14]. The most common method of porous silicon formation is by electrochemical etching in a hydrofluoric acid-based electrolyte. Depending on the substrate doping and orientation, the electrolyte composition, and the applied current density, porous silicon with pore sizes ranging from a few nanometers to several micrometers can be fabricated [15], [16]. When the pore size is much smaller than the wavelength of incident light, porous silicon acts like an effective medium. In this way, the volume of void space within the silicon matrix, governed by the pore size and the density, determines the refractive index of the porous silicon, as illustrated in Fig. 1. A change in applied current density directly corresponds to a change in porous silicon porosity. For a given silicon substrate and electrolyte, a porosity range of approximately 30%–80% is attainable [17], [18]. Based on the Bruggeman effective medium approximation, this corresponds to a refractive index range of 1.32 to 2.72 at 1.5 µm [19]. This large refractive index contrast is the basis for multilayer porous silicon structures with tailored reflectance and transmission spectra that are useful for silicon photonic devices.
1077-260X/$20.00 © 2006 IEEE
WEISS AND FAUCHET: POROUS SILICON ONE-DIMENSIONAL PHOTONIC CRYSTALS FOR OPTICAL SIGNAL MODULATION
1515
Fig. 1. Illustration of effective medium approximation. Since the pore size is much smaller than the wavelength of light, porous silicon can be represented by an effective refractive index that is a function of the porosity, refractive index of silicon, and refractive index of the material inside the pores.
B. Porous Silicon One-Dimensional (1-D) Photonic Crystal 1-D photonic crystals have a refractive index that is periodic in one direction, creating a range of frequencies in which no light can propagate (i.e., photonic bandgap). Resonant photonic crystal structures are created by introducing a break in the refractive index periodicity, which localizes the electromagnetic field and causes a resonance in the optical spectrum. For example, in a multilayer stack of alternating high and low refractive index layers, a thicker layer is inserted into the middle of the stack to give rise to an allowed mode in the photonic bandgap. Resonant porous silicon 1-D photonic crystals designed for operation near 1.5 µm are formed with two different size-scale pores. Mesoporous silicon photonic crystals, with a typical pore diameter of 20 nm, are formed on highly doped p-type silicon (∼ 0.01 Ω·cm) in a solution of 15% hydrofluoric acid in ethanol. Current densities of 5 and 50 mA/cm2 are applied to create alternating layers with refractive indexes of 2.15 and 1.43, respectively. Macroporous silicon photonic crystals, with a typical pore diameter of 150 nm, are formed on highly doped n-type silicon (∼ 0.01 Ω·cm) using an electrolyte composed of H2 O:HF (17:1) and a few drops of surfactant [18]. For these structures, current densities of 15 and 35 mA/cm2 are applied to create alternating layers with refractive indexes of 2.15 and 1.32, respectively. The electrochemically formed mesoporous and macroporous silicon layers and a sketch of the resonant photonic crystal structure are shown in [11]. C. Active Porous Silicon Photonic Structures In order to dynamically tune the reflectance and transmission spectra of the porous silicon structures, an optically active material can be infiltrated inside the pores. When the refractive
Fig. 2. Simulated curves indicating relationship between microcavity Q-factor and extinction ratio for a refractive index change of ∆n = 0.01. (a) Smaller Q-factor devices. (b) Higher Q-factor devices. Since smaller Q-factor devices have broader resonances than higher Q-factor devices, a smaller extinction ratio is obtained for a given refractive index change in smaller Q-factor devices. Larger refractive index changes lead to larger resonance wavelength shifts. Therefore, smaller Q-factor devices require larger refractive index changes to perform satisfactorily.
index of the optically active material changes in response to an external stimulus (e.g., electric field, temperature, pressure), the refractive index of the porous silicon layers also changes based on the effective medium consideration. Fig. 2 illustrates how the change in refractive index causes the resonance wavelength to shift and leads to an extinction ratio that depends on the Q-factor of the resonance. Higher Q-factor devices require smaller refractive index changes to achieve large extinction ratios. In this paper, E7 liquid crystals are chosen as the active material based on their large birefringence of approximately 0.2 at near-infrared wavelengths (no ≈ 1.5, ne ≈ 1.7) [20]. The size ˚ wide and of the liquid crystal molecule is approximately 5 A 2 nm long. The porous silicon photonic crystals are first slightly oxidized to provide a stabilized, hydrophilic internal surface. Then, the liquid crystals are infiltrated into porous silicon in vacuum to ensure a uniform distribution throughout the depth of the multilayer structures. While the liquid crystal can be infiltrated into both mesoporous and macroporous silicon photonic crystals, they are considered to be in a confined geometry, where the influence of surface anchoring is significant [21]. The resulting composite structure forms the basis of an electrically tunable
1516
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 12, NO. 6, NOVEMBER/DECEMBER 2006
decreases as a function of distance from the surface, and liquid crystals beyond 10 nm from the surface rotate much more freely [22]. Therefore, 4 nm represents a reasonable distance from the surface for which strong surface anchoring is to be assumed. 2) The pores are cylinders completely filled with liquid crystals. Using these assumptions for the mesoporous silicon with rbig = 10 nm, it can be calculated from (1) that 64% of the liquid crystals are fixed and 36% of the liquid crystals can rotate. Vfixed =
Fig. 3. Schematic illustration of assumptions involved in estimating percentage of liquid crystals that contribute to effective birefringence. Liquid crystals in the shaded region, within 4 nm of the pore wall, are assumed to be fixed. Liquid crystals in the center of the pore can freely rotate. It is also assumed that liquid crystals fill the entire volume of the pore.
silicon photonic device. An ac electric field at 1 kHz is applied between indium-tin-oxide (ITO)-coated glass that is attached on top of the porous silicon photonic crystal and the highly doped crystalline silicon substrate below the porous silicon structure. III. DESIGN CONSIDERATIONS OF POROUS SILICON OPTICAL SIGNAL MODULATOR For the porous silicon photonic crystals, it is important to understand the theoretical relationship between the pore size, porosity, and the device performance. In practice, device size and fabrication complexity must also be considered. The following discussion and analysis will demonstrate the relative impact of each of these characteristics. A. Pore Size Pore size affects the ultimate modulator device performance by determining the effective birefringence of the liquid crystals in the porous silicon matrix. A larger refractive index change leads to larger extinction ratio devices. The refractive index change of liquid crystals is governed by their physical rotation. Surface anchoring, which limits the liquid crystal rotation and hence refractive index change, plays a larger role in smaller pores. An estimate of the percentage of liquid crystals that rotate and contribute to the effective refractive index change in both mesoporous and macroporous silicon can be calculated based on the following assumptions. Fig. 3 illustrates these assumptions. 1) Liquid crystals within 4 nm of the pore walls (e.g., equivalent to two liquid crystals touching end-to-end on the long axis or eight on the short axis) are fixed and do not rotate when the electric field is applied. In general, the first few monolayers of liquid crystals are considered to be very strongly anchored, the surface anchoring energy
2 2 − rsmall )h π(rbig 2 πrbig h
(1)
For macroporous silicon with rbig = 75 nm, it can be calculated that 10% of the liquid crystals are fixed and 90% of the liquid crystals can rotate. Clearly, the larger pore size gives rise to a larger liquid crystal effective birefringence and consequently, a larger optical signal contrast. Experimental results also support this claim, as will be shown in Section IV. B. Porosity The porosity of the porous silicon layers directly influences the volume of liquid crystals incorporated into each device. Layers with higher porosity can accommodate a larger quantity of liquid crystals and, in turn, enable a larger refractive index change. The porosity, and hence refractive index, contrast between constituent layers of the porous silicon photonic crystals governs the Q-factor of the structure for a given number of porous silicon layers. A higher Q-factor can be obtained for a given number of layers when the refractive index contrast between the layers is larger. Therefore, for the practical consideration of a smaller device size, it is desirable to minimize the number of layers required to achieve acceptable device operation. Simulations based on the transfer matrix methodology are performed to quantitatively determine the impact of porosity on both the Q-factor and the effective birefringence achievable for the porous silicon photonic crystals. As shown in Fig. 4, there is a negligible difference in modulator performance as the porosity contrast is decreased from 30%/80% to 50%/80%. This implies that a porous silicon optical signal modulator with 30% and 80% porosity layers would be the preferred choice since the performance is not compromised by the lower overall porosity profile. Assuming a modest effective liquid crystal birefringence of 0.01, a 30%/80% device with thickness of 4.2 µm is required to achieve 16 dB attenuation, while a 6.6-µm-thick 50%/80% device is needed to achieve an extinction ratio of 18 dB. As the porosity contrast is further reduced beyond 50%/80%, the device thickness continues to increase without a substantial increase in device performance. Before making a final determination of the preferred porosity profile for the porous silicon photonic crystal modulators, practical fabrication constraints also need to be considered. In the case of mesoporous silicon, low temperature etching is required to achieve the largest porosity contrast of 30%–80%
WEISS AND FAUCHET: POROUS SILICON ONE-DIMENSIONAL PHOTONIC CRYSTALS FOR OPTICAL SIGNAL MODULATION
Fig. 4. Effect of porosity contrast on device performance based on simulation. The square data points represent 4, 6, 8, and 10 periods and the circular data points represent 3, 4, 5, 6, and 7 periods in order of increasing Q-factor. There is negligible difference in the extinction ratio of porous silicon photonic crystals with a moderate difference in porosity contrast. Consequently, other design parameters such as pore size and ease of fabrication are more significant when determining the optimal device specifications.
porosity [17]. While etching near −20 ◦ C increases the achievable porosity contrast and reduces millimeter-scale roughness, implementing low temperature etching adds complexity to the fabrication method [23]. For macroporous silicon photonic crystals, there is a direct correlation between porosity and pore size. Therefore, reducing the porosity to 30% compromises the pore size and limits the liquid crystal rotation, as explained in the previous section. For these practical reasons, the devices that are fabricated for this study consist of macroporous silicon layers with porosities of 50% and 80%. The mesoporous silicon photonic crystals that are also investigated consist of layers with porosities of 50% and 75%. IV. PERFORMANCE OF POROUS SILICON OPTICAL SIGNAL MODULATOR A. Experimental Results Fig. 5 shows the relative resonance shifts of the mesoporous and macroporous silicon photonic crystals as a function of the applied voltage. The larger pores of the macroporous silicon photonic crystals lead to larger resonance shifts, as expected from the discussion in Section III-A. It can be calculated based on the magnitude of the resonance shift that the maximum effective birefringence of the liquid crystals in macroporous silicon photonic crystals is 0.01, while the birefringence is only 0.005 in the mesoporous silicon photonic crystals. Surface anchoring and the initial liquid crystal alignment in the pores significantly reduce the effective liquid crystal birefringence compared to the theoretically obtainable value of 0.2. As calculated from (1), only a fraction of the liquid crystals, whose rotation is not significantly inhibited by surface anchoring, can contribute to the birefringence. A quantitative analysis of the influence of initial liquid crystal alignment can be performed based on the measured birefringence and estimation of surface anchoring from (1). With the application of the elec-
1517
Fig. 5. Relative resonance wavelength shift as a function of voltage for mesoporous and macroporous silicon photonic crystals based on experimental measurements. The resonance wavelength is near 1.5 µm for both designs.
tric field, the effective liquid crystal refractive index inside the mesoporous silicon photonic crystals is given by −field neff = 0.64nfixed + 0.36nE aligned
= 0.64nfixed + 0.36ncenter initial − ∆nmeasured
(2)
−field nE aligned
= 1.5 for positive anisotropy E7 liquid crystals where aligned with the applied electric field, ncenter initial relates to the initial alignment of the liquid crystals in the center of the pores that are free to rotate, and ∆nmeasured = 0.005 is the magnitude of the maximum effective birefringence of the liquid crystals in mesoporous silicon that has been experimentally measured based on Fig. 5. Solving (2), it can be shown that the initial refractive index of the liquid crystals in the center of the mesopores is approximately 1.514. Then, by using the index ellipse in (3), the initial liquid crystal alignment in the center of the pores is calculated to be 17◦ with respect to the pore walls 1 cos2 (θ) sin2 (θ) = + . n2 (θ) n2o n2e
(3)
A similar analysis can be followed for the macroporous silicon photonic crystals to show that the liquid crystals in the center of the pores are aligned at an angle of 15◦ with respect to the pore walls. However, the wavelength shift of the macroporous silicon photonic crystal resonance is not saturated with the applied voltage in Fig. 5. Therefore, it is anticipated that the liquid crystals will continue to rotate if a larger voltage is applied. This would imply that the liquid crystals in the center of the macropores are actually aligned at a slightly larger angle with respect to the pore walls than indicated by the present calculation. Clearly, the initial liquid crystal alignment significantly reduces the effective liquid crystal birefringence in both the mesoporous and macroporous silicon photonic crystals. B. Importance of Q-Factor on Device Performance Given the resonance wavelength shift as a function of voltage, as shown in Fig. 5, it is possible to determine the effective liquid
1518
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 12, NO. 6, NOVEMBER/DECEMBER 2006
V. CONCLUSION Porous silicon photonic crystals have been demonstrated as optical signal modulators. The design of the porous silicon structures plays an important role in the ultimate device performance. Larger pore sizes, as opposed to large porosity contrasts, have the largest impact on achievable extinction ratios. For porous silicon photonic crystals with a Q-factor of 1000, 10 dB and 20 dB attenuations are obtained for electrically actuated mesoporous and macroporous structures, respectively. ACKNOWLEDGMENT
Fig. 6. Potential porous silicon optical modulator device performance based on experimentally measured results. Realistic devices with Q-factors of the order of 103 can achieve 10-dB attenuation with mesoporous silicon and 20-dB attenuation with macroporous silicon.
crystal birefringence as a function of voltage, and thus, the device performance can be evaluated for various Q-factors. Larger Q-factors imply sharper resonances for a given operating wavelength. Therefore, smaller shifts of the resonance wavelength will correspond to larger extinction ratios for higher Q-factor devices. Fig. 6 shows the potential porous silicon optical modulator performance for mesoporous and macroporous silicon devices with Q-factors of the order of 102 , 103 , and 104 based on the measurement results, as depicted in Fig. 5. The liquid crystals in the center of the pores are assumed to be oriented at an angle of 17◦ with respect to the pore walls, as shown in Section IV-A. A Q-factor of the order of 100 is not sufficient for most modulator applications. However, an achievable Q-factor of the order of 1000 does lead to acceptable performance for many applications. Mesoporous silicon photonic crystals can operate with greater than 10 dB attenuation for an applied voltage of 75 V, and macroporous silicon photonic crystals can operate with greater than 20 dB attenuation at 50 V. Obtaining porous silicon photonic crystals with Q-factors of the order of 104 is not realizable at the present time, and would only provide a modest improvement in the extinction ratio, albeit at lower voltages. It is important to note that the threshold applied electric field is determined based on the thickness of the porous silicon photonic crystal and the thickness of the liquid crystal-filled gap that exists between the porous silicon and the ITO-coated glass top contact. The turn-on voltage can, therefore, be reduced by minimizing the voltage drop across the top liquid crystal layer, which currently reduces the effective field felt by the liquid crystals inside the porous silicon photonic crystals. Thus, practical devices can be realized given the limitations of the E7 liquid crystal birefringence in porous silicon. Further efforts towards minimizing the effects of surface anchoring and the thickness of the liquid crystals on top of the photonic crystals, and investigation of alternative optically active materials infiltrated into the pores will lead to lower operating voltage and enhanced performance of the porous silicon optical modulators.
The authors would like to thank H. Ouyang of the University of Rochester for her technical assistance with the macroporous silicon structures. REFERENCES [1] O. Boyraz and B. Jalali, “Demonstration of a silicon Raman laser source,” Opt. Express, vol. 12, pp. 5269–5273, 2004. [2] H. Rong, R. Jones, A. Liu, O. Cohen, D. Hak, A. Fang, and M. Paniccia, “A continuous-wave Raman silicon laser,” Nature, vol. 433, pp. 725–728, 2005. [3] V. R. Almeida, R. R. Panepucci, and Michal Lipson, “Nanotaper for compact mode conversion,” Opt. Lett., vol. 28, pp. 1302–1304, 2003. [4] A. Liu, R. Jones, L. Liao, D. Samara-Rubio, D. Rubin, O. Cohen, R. Nicolaescu, and M. Paniccia, “A high-speed silicon optical modulator based on a metal–oxide–semiconductor capacitor,” Nature, vol. 427, pp. 615–618, 2004. [5] Q. Xu, B. Schmidt, S. Pradhan, and M. Lipson, “Micrometre-scale silicon electro-optic modulator,” Nature, vol. 435, pp. 325–327, 2005. [6] H. Rong, A. Liu, R. Nicolaescu, and M. Paniccia, “Raman gain and nonlinear optical absorption measurements in a low-loss silicon waveguide,” Appl. Phys. Lett., vol. 85, pp. 2196–2198, 2004. [7] M. R. Reshotko, D. L. Kencke, and B. Block, “High-speed CMOS compatible photodetectors for optical interconnects,” in Proc. SPIE, vol. 5564, 2004, pp. 146–155. [8] M. Paniccia and S. Koehl, “The silicon solution,” IEEE Spectr., vol. 42, no. 10, pp. 38–43, Oct. 2005. [9] M. Tokushima, H. Kosaka, A. Tomita, and H. Yamada, “Lightwave propagation through a 120◦ sharply bent single-line-defect photonic crystal waveguide,” Appl. Phys. Lett., vol. 76, pp. 952–954, 2000. [10] S. Noda, A. Chutinan, and M. Imada, “Trapping and emission of photons by a single defect in a photonic bandgap structure,” Nature, vol. 407, pp. 608–610, 2000. [11] S. M. Weiss, H. Ouyang, J. Zhang, and P. M. Fauchet, “Electrical and thermal modulation of silicon photonic bandgap microcavities containing liquid crystals,” Opt. Express, vol. 13, pp. 1090–1097, 2005. [12] L. Pavesi, L. Dal Negro, C. Mazzoleni, G. Franzo, and F. Priolo, “Optical gain in silicon nanocrystals,” Nature, vol. 408, pp. 440–444, 2000. [13] J. Ruan, P. M. Fauchet, L. Dal Negro, M. Cazzanelli, and L. Pavesi, “Stimulated emission in nanocrystalline silicon superlattices,” Appl. Phys. Lett., vol. 83, pp. 5479–5481, 2003. [14] L. T. Canham, “Silicon quantum wire array fabricated by electrochemical and chemical dissolution of wafers,” Appl. Phys. Lett., vol. 57, pp. 1046– 1048, 1990. [15] L. Canham, Ed., Properties of Porous Silicon. London, U.K.: Inst. Electr. Eng.-INSPEC, 1997. [16] V. Lehmann, Electrochemistry of Silicon: Instrumentation, Science, Materials and Applications. Weinheim, Germany: Wiley-VCH, 2002. [17] W. H. Zheng, P. Reece, B. Q. Sun, and M. Gal, “Broadband laser mirrors made from porous silicon,” Appl. Phys. Lett., vol. 84, pp. 3519–3521, 2004. [18] H. Ouyang, M. Christophersen, R. Viard, B. L. Miller, and P. M. Fauchet, “Macroporous silicon microcavities for macromolecule detection,” Adv. Funct. Mater., vol. 15, pp. 1851–1859, 2005. [19] W. Theiß, “Optical properties of porous silicon,” Surf. Sci. Rep., vol. 29, pp. 91–192, 1997.
WEISS AND FAUCHET: POROUS SILICON ONE-DIMENSIONAL PHOTONIC CRYSTALS FOR OPTICAL SIGNAL MODULATION
[20] B. Bahadur, R. K. Sarna, and V. G. Bhide, “Refractive indices, density and order parameter of some technologically important liquid crystalline mixtures,” Mol. Cryst. Liq. Cryst., vol. 72, pp. 139–145, 1982. ˇ [21] G. P. Crawford and S. Zumer, Eds., Liquid Crystals in Complex Geometries Formed by Polymer and Porous Networks. London, U.K.: Taylor & Francis, 1996, pp. 21–52. [22] J. Cognard, Alignment of Nematic Liquid Crystals and Their Mixtures. New York: Gordon and Breach, 1982, pp. 56–63. [23] G. Lerondel, P. Reece, A. Bruyant, and M. Gal, “Strong light confinement in microporous photonic silicon structures,” in Proc. Mater. Res. Soc. Symp., 2004, vol. 797, pp. W1.7.1–W1.7.6.
Sharon M. Weiss (S’04–M’05) received the B.S., M.S., and Ph.D. degrees from the Institute of Optics, University of Rochester, Rochester, NY, in 1999, 2001, and 2005, respectively. Since August 2005, she has been an Assistant Professor of electrical engineering and physics at Vanderbilt University, Nashville, TN. Her current research interests include silicon-based biosensing and optical components, and white-light LEDs.
1519
Philippe M. Fauchet (A’79–M’82–SM’89–F’00) received the Engineer’s degree from the Faculte Polytechnique de Mons, Mons, Belgium, in 1978, the M.S. degree in engineering from Brown University, Providence, RI, in 1980, and the Ph.D. degree in applied physics from Stanford University, Stanford, CA, in 1984. He spent one year at Stanford University as an IBM Post-Doctoral Fellow and Acting Assistant Professor, before holding a faculty position in electrical engineering with Princeton University, Princeton, NJ. He is currently a Distinguished Professor of electrical and computer engineering at the University of Rochester, Rochester, NY, where he also is a Professor of optics, biomedical engineering, and materials science, and a Senior Scientist at the Laboratory for Laser Energetics. His current research interests include nanoscience and nanotechnology with silicon quantum dots, silicon-based LEDs and lasers, semiconductor optoelectronics and devices, OIs, silicon-based biosensors, the silicon/biology interface, and optical diagnostics. In 1998, he founded the Center for Future Health, a multidisciplinary laboratory where engineers, physicians, and social scientists develop affordable technology that can be used at home to keep individuals independent and healthy. He has authored over 350 publications, edited nine books. Dr. Fauchet has chaired many conferences, including the 2005 IEEE Group Four Photonics Conference. He is a Fellow of the Optical Society of America and the American Physical Society. He received an IBM Faculty Development Award in 1985, an NSF Presidential Young Investigator Award in 1987, an Alfred P. Sloan Research Fellowship in 1988, and the 1990–1993 Prix Guibal & Devillez for his work on porous silicon.
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 12, NO. 6, NOVEMBER/DECEMBER 2006
Porous Silicon One-Dimensional Photonic Crystals for Optical Signal Modulation Sharon M. Weiss, Member, IEEE, and Philippe M. Fauchet, Fellow, IEEE
Abstract—Porous silicon one-dimensional (1-D) photonic crystal microcavities are fabricated as building blocks for optical modulators. The microcavities consist of porous silicon layers with alternating low and high porosity. Since the pore size is much smaller than the incident wavelength of light, each porous silicon layer is treated as an effective medium with a specific effective refractive index that corresponds to the given porosity. Active tuning of the microcavity resonance is accomplished through electrical actuation of liquid crystals infiltrated into the pores. As the refractive index of the liquid crystals change in response to the external stimulus, the effective refractive index of the porous silicon layers changes and the resonance wavelength shifts accordingly. The resonance wavelength shift creates an optical signal intensity modulation at a particular wavelength. Extinction ratios greater than 10 dB are achievable for these devices. The performance characteristics of the porous silicon photonic crystal microcavities can be tailored to specific applications based on the choice of infiltrated optically active material and the quality factor of the device. Index Terms—Liquid crystal, modulation, photonic crystal, silicon.
I. INTRODUCTION HE past few years have marked widespread progress in the area of silicon photonics. Specifically, with regard to components for optical interconnects, there have been demonstrations of silicon Raman lasers [1], [2], efficient and compact optical coupling schemes [3], electrically driven silicon modulators [4], [5], low-loss silicon-on-insulator waveguides [6], and high speed detectors [7]. In order to continue to meet the demand for increasing information bandwidth while maintaining high speed, low power consumption, and low-cost components, long-term solutions to the interconnect bottleneck need to be addressed on several levels, from interboard communication to intrachip connections. Silicon has not traditionally been considered a viable optical material due to its indirect bandgap, which inhibits light emission efficiency, and its low electro-optic coefficient. However, due to the extensive infrastructure that exists for silicon processing, the potential advantages derived from achieving optical functionality on a silicon platform are a strong motivation to invest substantial resources into investigating new methods of “siliconizing” photonics [8].
T
Manuscript received November 17, 2005; revised August 21, 2006. This work was supported in part by the Air Force Office of Scientific Research under Grant F 49620-02-1-0376 and in part by Intel Corporation. S. M. Weiss was with the Institute of Optics, University of Rochester, Rochester, NY 14627 USA. She is now with the Department of Electrical Engineering, Vanderbilt University, Nashville, TN 37235 USA (e-mail: [email protected]). P. M. Fauchet is with the Department of Electrical and Computer Engineering, University of Rochester, Rochester, NY 14627 USA (e-mail: [email protected]). Digital Object Identifier 10.1109/JSTQE.2006.884083
Photonic crystals are engineered optical structures that have been demonstrated as means of controlling light propagation. Demonstrations of silicon photonic crystals as waveguides [9], add/drop filters [10], and modulators [11] form the basis of future optical interconnect building blocks. Since photonic crystals localize electromagnetic fields, compact photonic components can be realized. Moreover, the capability to infiltrate optically active materials into the air holes of photonic crystal structures enables enhanced optical functionality beyond that of the traditional silicon devices. Several silicon-based material systems have been explored for optical applications. Intel’s efforts, for example, have been focused on standard planar silicon device configurations for light sources [2] and modulators [4], whereas silicon germanium is the popular choice for detectors at near-infrared wavelengths [7]. Several groups have also explored the possibility of enhancing silicon’s optical capabilities by reducing the feature size. Silicon nanocrystals are being studied for their light emission properties [12], [13], and the large achievable refractive index range of porous silicon is being exploited for controlling the reflectance and transmission profiles of optical modulator building blocks [11]. II. POROUS SILICON AS AN OPTICAL MATERIAL A. Fabrication by Electrochemical Etching Porous silicon has been investigated as an optical material since 1990 when Canham first reported visible photoluminescence [14]. The most common method of porous silicon formation is by electrochemical etching in a hydrofluoric acid-based electrolyte. Depending on the substrate doping and orientation, the electrolyte composition, and the applied current density, porous silicon with pore sizes ranging from a few nanometers to several micrometers can be fabricated [15], [16]. When the pore size is much smaller than the wavelength of incident light, porous silicon acts like an effective medium. In this way, the volume of void space within the silicon matrix, governed by the pore size and the density, determines the refractive index of the porous silicon, as illustrated in Fig. 1. A change in applied current density directly corresponds to a change in porous silicon porosity. For a given silicon substrate and electrolyte, a porosity range of approximately 30%–80% is attainable [17], [18]. Based on the Bruggeman effective medium approximation, this corresponds to a refractive index range of 1.32 to 2.72 at 1.5 µm [19]. This large refractive index contrast is the basis for multilayer porous silicon structures with tailored reflectance and transmission spectra that are useful for silicon photonic devices.
1077-260X/$20.00 © 2006 IEEE
WEISS AND FAUCHET: POROUS SILICON ONE-DIMENSIONAL PHOTONIC CRYSTALS FOR OPTICAL SIGNAL MODULATION
1515
Fig. 1. Illustration of effective medium approximation. Since the pore size is much smaller than the wavelength of light, porous silicon can be represented by an effective refractive index that is a function of the porosity, refractive index of silicon, and refractive index of the material inside the pores.
B. Porous Silicon One-Dimensional (1-D) Photonic Crystal 1-D photonic crystals have a refractive index that is periodic in one direction, creating a range of frequencies in which no light can propagate (i.e., photonic bandgap). Resonant photonic crystal structures are created by introducing a break in the refractive index periodicity, which localizes the electromagnetic field and causes a resonance in the optical spectrum. For example, in a multilayer stack of alternating high and low refractive index layers, a thicker layer is inserted into the middle of the stack to give rise to an allowed mode in the photonic bandgap. Resonant porous silicon 1-D photonic crystals designed for operation near 1.5 µm are formed with two different size-scale pores. Mesoporous silicon photonic crystals, with a typical pore diameter of 20 nm, are formed on highly doped p-type silicon (∼ 0.01 Ω·cm) in a solution of 15% hydrofluoric acid in ethanol. Current densities of 5 and 50 mA/cm2 are applied to create alternating layers with refractive indexes of 2.15 and 1.43, respectively. Macroporous silicon photonic crystals, with a typical pore diameter of 150 nm, are formed on highly doped n-type silicon (∼ 0.01 Ω·cm) using an electrolyte composed of H2 O:HF (17:1) and a few drops of surfactant [18]. For these structures, current densities of 15 and 35 mA/cm2 are applied to create alternating layers with refractive indexes of 2.15 and 1.32, respectively. The electrochemically formed mesoporous and macroporous silicon layers and a sketch of the resonant photonic crystal structure are shown in [11]. C. Active Porous Silicon Photonic Structures In order to dynamically tune the reflectance and transmission spectra of the porous silicon structures, an optically active material can be infiltrated inside the pores. When the refractive
Fig. 2. Simulated curves indicating relationship between microcavity Q-factor and extinction ratio for a refractive index change of ∆n = 0.01. (a) Smaller Q-factor devices. (b) Higher Q-factor devices. Since smaller Q-factor devices have broader resonances than higher Q-factor devices, a smaller extinction ratio is obtained for a given refractive index change in smaller Q-factor devices. Larger refractive index changes lead to larger resonance wavelength shifts. Therefore, smaller Q-factor devices require larger refractive index changes to perform satisfactorily.
index of the optically active material changes in response to an external stimulus (e.g., electric field, temperature, pressure), the refractive index of the porous silicon layers also changes based on the effective medium consideration. Fig. 2 illustrates how the change in refractive index causes the resonance wavelength to shift and leads to an extinction ratio that depends on the Q-factor of the resonance. Higher Q-factor devices require smaller refractive index changes to achieve large extinction ratios. In this paper, E7 liquid crystals are chosen as the active material based on their large birefringence of approximately 0.2 at near-infrared wavelengths (no ≈ 1.5, ne ≈ 1.7) [20]. The size ˚ wide and of the liquid crystal molecule is approximately 5 A 2 nm long. The porous silicon photonic crystals are first slightly oxidized to provide a stabilized, hydrophilic internal surface. Then, the liquid crystals are infiltrated into porous silicon in vacuum to ensure a uniform distribution throughout the depth of the multilayer structures. While the liquid crystal can be infiltrated into both mesoporous and macroporous silicon photonic crystals, they are considered to be in a confined geometry, where the influence of surface anchoring is significant [21]. The resulting composite structure forms the basis of an electrically tunable
1516
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 12, NO. 6, NOVEMBER/DECEMBER 2006
decreases as a function of distance from the surface, and liquid crystals beyond 10 nm from the surface rotate much more freely [22]. Therefore, 4 nm represents a reasonable distance from the surface for which strong surface anchoring is to be assumed. 2) The pores are cylinders completely filled with liquid crystals. Using these assumptions for the mesoporous silicon with rbig = 10 nm, it can be calculated from (1) that 64% of the liquid crystals are fixed and 36% of the liquid crystals can rotate. Vfixed =
Fig. 3. Schematic illustration of assumptions involved in estimating percentage of liquid crystals that contribute to effective birefringence. Liquid crystals in the shaded region, within 4 nm of the pore wall, are assumed to be fixed. Liquid crystals in the center of the pore can freely rotate. It is also assumed that liquid crystals fill the entire volume of the pore.
silicon photonic device. An ac electric field at 1 kHz is applied between indium-tin-oxide (ITO)-coated glass that is attached on top of the porous silicon photonic crystal and the highly doped crystalline silicon substrate below the porous silicon structure. III. DESIGN CONSIDERATIONS OF POROUS SILICON OPTICAL SIGNAL MODULATOR For the porous silicon photonic crystals, it is important to understand the theoretical relationship between the pore size, porosity, and the device performance. In practice, device size and fabrication complexity must also be considered. The following discussion and analysis will demonstrate the relative impact of each of these characteristics. A. Pore Size Pore size affects the ultimate modulator device performance by determining the effective birefringence of the liquid crystals in the porous silicon matrix. A larger refractive index change leads to larger extinction ratio devices. The refractive index change of liquid crystals is governed by their physical rotation. Surface anchoring, which limits the liquid crystal rotation and hence refractive index change, plays a larger role in smaller pores. An estimate of the percentage of liquid crystals that rotate and contribute to the effective refractive index change in both mesoporous and macroporous silicon can be calculated based on the following assumptions. Fig. 3 illustrates these assumptions. 1) Liquid crystals within 4 nm of the pore walls (e.g., equivalent to two liquid crystals touching end-to-end on the long axis or eight on the short axis) are fixed and do not rotate when the electric field is applied. In general, the first few monolayers of liquid crystals are considered to be very strongly anchored, the surface anchoring energy
2 2 − rsmall )h π(rbig 2 πrbig h
(1)
For macroporous silicon with rbig = 75 nm, it can be calculated that 10% of the liquid crystals are fixed and 90% of the liquid crystals can rotate. Clearly, the larger pore size gives rise to a larger liquid crystal effective birefringence and consequently, a larger optical signal contrast. Experimental results also support this claim, as will be shown in Section IV. B. Porosity The porosity of the porous silicon layers directly influences the volume of liquid crystals incorporated into each device. Layers with higher porosity can accommodate a larger quantity of liquid crystals and, in turn, enable a larger refractive index change. The porosity, and hence refractive index, contrast between constituent layers of the porous silicon photonic crystals governs the Q-factor of the structure for a given number of porous silicon layers. A higher Q-factor can be obtained for a given number of layers when the refractive index contrast between the layers is larger. Therefore, for the practical consideration of a smaller device size, it is desirable to minimize the number of layers required to achieve acceptable device operation. Simulations based on the transfer matrix methodology are performed to quantitatively determine the impact of porosity on both the Q-factor and the effective birefringence achievable for the porous silicon photonic crystals. As shown in Fig. 4, there is a negligible difference in modulator performance as the porosity contrast is decreased from 30%/80% to 50%/80%. This implies that a porous silicon optical signal modulator with 30% and 80% porosity layers would be the preferred choice since the performance is not compromised by the lower overall porosity profile. Assuming a modest effective liquid crystal birefringence of 0.01, a 30%/80% device with thickness of 4.2 µm is required to achieve 16 dB attenuation, while a 6.6-µm-thick 50%/80% device is needed to achieve an extinction ratio of 18 dB. As the porosity contrast is further reduced beyond 50%/80%, the device thickness continues to increase without a substantial increase in device performance. Before making a final determination of the preferred porosity profile for the porous silicon photonic crystal modulators, practical fabrication constraints also need to be considered. In the case of mesoporous silicon, low temperature etching is required to achieve the largest porosity contrast of 30%–80%
WEISS AND FAUCHET: POROUS SILICON ONE-DIMENSIONAL PHOTONIC CRYSTALS FOR OPTICAL SIGNAL MODULATION
Fig. 4. Effect of porosity contrast on device performance based on simulation. The square data points represent 4, 6, 8, and 10 periods and the circular data points represent 3, 4, 5, 6, and 7 periods in order of increasing Q-factor. There is negligible difference in the extinction ratio of porous silicon photonic crystals with a moderate difference in porosity contrast. Consequently, other design parameters such as pore size and ease of fabrication are more significant when determining the optimal device specifications.
porosity [17]. While etching near −20 ◦ C increases the achievable porosity contrast and reduces millimeter-scale roughness, implementing low temperature etching adds complexity to the fabrication method [23]. For macroporous silicon photonic crystals, there is a direct correlation between porosity and pore size. Therefore, reducing the porosity to 30% compromises the pore size and limits the liquid crystal rotation, as explained in the previous section. For these practical reasons, the devices that are fabricated for this study consist of macroporous silicon layers with porosities of 50% and 80%. The mesoporous silicon photonic crystals that are also investigated consist of layers with porosities of 50% and 75%. IV. PERFORMANCE OF POROUS SILICON OPTICAL SIGNAL MODULATOR A. Experimental Results Fig. 5 shows the relative resonance shifts of the mesoporous and macroporous silicon photonic crystals as a function of the applied voltage. The larger pores of the macroporous silicon photonic crystals lead to larger resonance shifts, as expected from the discussion in Section III-A. It can be calculated based on the magnitude of the resonance shift that the maximum effective birefringence of the liquid crystals in macroporous silicon photonic crystals is 0.01, while the birefringence is only 0.005 in the mesoporous silicon photonic crystals. Surface anchoring and the initial liquid crystal alignment in the pores significantly reduce the effective liquid crystal birefringence compared to the theoretically obtainable value of 0.2. As calculated from (1), only a fraction of the liquid crystals, whose rotation is not significantly inhibited by surface anchoring, can contribute to the birefringence. A quantitative analysis of the influence of initial liquid crystal alignment can be performed based on the measured birefringence and estimation of surface anchoring from (1). With the application of the elec-
1517
Fig. 5. Relative resonance wavelength shift as a function of voltage for mesoporous and macroporous silicon photonic crystals based on experimental measurements. The resonance wavelength is near 1.5 µm for both designs.
tric field, the effective liquid crystal refractive index inside the mesoporous silicon photonic crystals is given by −field neff = 0.64nfixed + 0.36nE aligned
= 0.64nfixed + 0.36ncenter initial − ∆nmeasured
(2)
−field nE aligned
= 1.5 for positive anisotropy E7 liquid crystals where aligned with the applied electric field, ncenter initial relates to the initial alignment of the liquid crystals in the center of the pores that are free to rotate, and ∆nmeasured = 0.005 is the magnitude of the maximum effective birefringence of the liquid crystals in mesoporous silicon that has been experimentally measured based on Fig. 5. Solving (2), it can be shown that the initial refractive index of the liquid crystals in the center of the mesopores is approximately 1.514. Then, by using the index ellipse in (3), the initial liquid crystal alignment in the center of the pores is calculated to be 17◦ with respect to the pore walls 1 cos2 (θ) sin2 (θ) = + . n2 (θ) n2o n2e
(3)
A similar analysis can be followed for the macroporous silicon photonic crystals to show that the liquid crystals in the center of the pores are aligned at an angle of 15◦ with respect to the pore walls. However, the wavelength shift of the macroporous silicon photonic crystal resonance is not saturated with the applied voltage in Fig. 5. Therefore, it is anticipated that the liquid crystals will continue to rotate if a larger voltage is applied. This would imply that the liquid crystals in the center of the macropores are actually aligned at a slightly larger angle with respect to the pore walls than indicated by the present calculation. Clearly, the initial liquid crystal alignment significantly reduces the effective liquid crystal birefringence in both the mesoporous and macroporous silicon photonic crystals. B. Importance of Q-Factor on Device Performance Given the resonance wavelength shift as a function of voltage, as shown in Fig. 5, it is possible to determine the effective liquid
1518
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 12, NO. 6, NOVEMBER/DECEMBER 2006
V. CONCLUSION Porous silicon photonic crystals have been demonstrated as optical signal modulators. The design of the porous silicon structures plays an important role in the ultimate device performance. Larger pore sizes, as opposed to large porosity contrasts, have the largest impact on achievable extinction ratios. For porous silicon photonic crystals with a Q-factor of 1000, 10 dB and 20 dB attenuations are obtained for electrically actuated mesoporous and macroporous structures, respectively. ACKNOWLEDGMENT
Fig. 6. Potential porous silicon optical modulator device performance based on experimentally measured results. Realistic devices with Q-factors of the order of 103 can achieve 10-dB attenuation with mesoporous silicon and 20-dB attenuation with macroporous silicon.
crystal birefringence as a function of voltage, and thus, the device performance can be evaluated for various Q-factors. Larger Q-factors imply sharper resonances for a given operating wavelength. Therefore, smaller shifts of the resonance wavelength will correspond to larger extinction ratios for higher Q-factor devices. Fig. 6 shows the potential porous silicon optical modulator performance for mesoporous and macroporous silicon devices with Q-factors of the order of 102 , 103 , and 104 based on the measurement results, as depicted in Fig. 5. The liquid crystals in the center of the pores are assumed to be oriented at an angle of 17◦ with respect to the pore walls, as shown in Section IV-A. A Q-factor of the order of 100 is not sufficient for most modulator applications. However, an achievable Q-factor of the order of 1000 does lead to acceptable performance for many applications. Mesoporous silicon photonic crystals can operate with greater than 10 dB attenuation for an applied voltage of 75 V, and macroporous silicon photonic crystals can operate with greater than 20 dB attenuation at 50 V. Obtaining porous silicon photonic crystals with Q-factors of the order of 104 is not realizable at the present time, and would only provide a modest improvement in the extinction ratio, albeit at lower voltages. It is important to note that the threshold applied electric field is determined based on the thickness of the porous silicon photonic crystal and the thickness of the liquid crystal-filled gap that exists between the porous silicon and the ITO-coated glass top contact. The turn-on voltage can, therefore, be reduced by minimizing the voltage drop across the top liquid crystal layer, which currently reduces the effective field felt by the liquid crystals inside the porous silicon photonic crystals. Thus, practical devices can be realized given the limitations of the E7 liquid crystal birefringence in porous silicon. Further efforts towards minimizing the effects of surface anchoring and the thickness of the liquid crystals on top of the photonic crystals, and investigation of alternative optically active materials infiltrated into the pores will lead to lower operating voltage and enhanced performance of the porous silicon optical modulators.
The authors would like to thank H. Ouyang of the University of Rochester for her technical assistance with the macroporous silicon structures. REFERENCES [1] O. Boyraz and B. Jalali, “Demonstration of a silicon Raman laser source,” Opt. Express, vol. 12, pp. 5269–5273, 2004. [2] H. Rong, R. Jones, A. Liu, O. Cohen, D. Hak, A. Fang, and M. Paniccia, “A continuous-wave Raman silicon laser,” Nature, vol. 433, pp. 725–728, 2005. [3] V. R. Almeida, R. R. Panepucci, and Michal Lipson, “Nanotaper for compact mode conversion,” Opt. Lett., vol. 28, pp. 1302–1304, 2003. [4] A. Liu, R. Jones, L. Liao, D. Samara-Rubio, D. Rubin, O. Cohen, R. Nicolaescu, and M. Paniccia, “A high-speed silicon optical modulator based on a metal–oxide–semiconductor capacitor,” Nature, vol. 427, pp. 615–618, 2004. [5] Q. Xu, B. Schmidt, S. Pradhan, and M. Lipson, “Micrometre-scale silicon electro-optic modulator,” Nature, vol. 435, pp. 325–327, 2005. [6] H. Rong, A. Liu, R. Nicolaescu, and M. Paniccia, “Raman gain and nonlinear optical absorption measurements in a low-loss silicon waveguide,” Appl. Phys. Lett., vol. 85, pp. 2196–2198, 2004. [7] M. R. Reshotko, D. L. Kencke, and B. Block, “High-speed CMOS compatible photodetectors for optical interconnects,” in Proc. SPIE, vol. 5564, 2004, pp. 146–155. [8] M. Paniccia and S. Koehl, “The silicon solution,” IEEE Spectr., vol. 42, no. 10, pp. 38–43, Oct. 2005. [9] M. Tokushima, H. Kosaka, A. Tomita, and H. Yamada, “Lightwave propagation through a 120◦ sharply bent single-line-defect photonic crystal waveguide,” Appl. Phys. Lett., vol. 76, pp. 952–954, 2000. [10] S. Noda, A. Chutinan, and M. Imada, “Trapping and emission of photons by a single defect in a photonic bandgap structure,” Nature, vol. 407, pp. 608–610, 2000. [11] S. M. Weiss, H. Ouyang, J. Zhang, and P. M. Fauchet, “Electrical and thermal modulation of silicon photonic bandgap microcavities containing liquid crystals,” Opt. Express, vol. 13, pp. 1090–1097, 2005. [12] L. Pavesi, L. Dal Negro, C. Mazzoleni, G. Franzo, and F. Priolo, “Optical gain in silicon nanocrystals,” Nature, vol. 408, pp. 440–444, 2000. [13] J. Ruan, P. M. Fauchet, L. Dal Negro, M. Cazzanelli, and L. Pavesi, “Stimulated emission in nanocrystalline silicon superlattices,” Appl. Phys. Lett., vol. 83, pp. 5479–5481, 2003. [14] L. T. Canham, “Silicon quantum wire array fabricated by electrochemical and chemical dissolution of wafers,” Appl. Phys. Lett., vol. 57, pp. 1046– 1048, 1990. [15] L. Canham, Ed., Properties of Porous Silicon. London, U.K.: Inst. Electr. Eng.-INSPEC, 1997. [16] V. Lehmann, Electrochemistry of Silicon: Instrumentation, Science, Materials and Applications. Weinheim, Germany: Wiley-VCH, 2002. [17] W. H. Zheng, P. Reece, B. Q. Sun, and M. Gal, “Broadband laser mirrors made from porous silicon,” Appl. Phys. Lett., vol. 84, pp. 3519–3521, 2004. [18] H. Ouyang, M. Christophersen, R. Viard, B. L. Miller, and P. M. Fauchet, “Macroporous silicon microcavities for macromolecule detection,” Adv. Funct. Mater., vol. 15, pp. 1851–1859, 2005. [19] W. Theiß, “Optical properties of porous silicon,” Surf. Sci. Rep., vol. 29, pp. 91–192, 1997.
WEISS AND FAUCHET: POROUS SILICON ONE-DIMENSIONAL PHOTONIC CRYSTALS FOR OPTICAL SIGNAL MODULATION
[20] B. Bahadur, R. K. Sarna, and V. G. Bhide, “Refractive indices, density and order parameter of some technologically important liquid crystalline mixtures,” Mol. Cryst. Liq. Cryst., vol. 72, pp. 139–145, 1982. ˇ [21] G. P. Crawford and S. Zumer, Eds., Liquid Crystals in Complex Geometries Formed by Polymer and Porous Networks. London, U.K.: Taylor & Francis, 1996, pp. 21–52. [22] J. Cognard, Alignment of Nematic Liquid Crystals and Their Mixtures. New York: Gordon and Breach, 1982, pp. 56–63. [23] G. Lerondel, P. Reece, A. Bruyant, and M. Gal, “Strong light confinement in microporous photonic silicon structures,” in Proc. Mater. Res. Soc. Symp., 2004, vol. 797, pp. W1.7.1–W1.7.6.
Sharon M. Weiss (S’04–M’05) received the B.S., M.S., and Ph.D. degrees from the Institute of Optics, University of Rochester, Rochester, NY, in 1999, 2001, and 2005, respectively. Since August 2005, she has been an Assistant Professor of electrical engineering and physics at Vanderbilt University, Nashville, TN. Her current research interests include silicon-based biosensing and optical components, and white-light LEDs.
1519
Philippe M. Fauchet (A’79–M’82–SM’89–F’00) received the Engineer’s degree from the Faculte Polytechnique de Mons, Mons, Belgium, in 1978, the M.S. degree in engineering from Brown University, Providence, RI, in 1980, and the Ph.D. degree in applied physics from Stanford University, Stanford, CA, in 1984. He spent one year at Stanford University as an IBM Post-Doctoral Fellow and Acting Assistant Professor, before holding a faculty position in electrical engineering with Princeton University, Princeton, NJ. He is currently a Distinguished Professor of electrical and computer engineering at the University of Rochester, Rochester, NY, where he also is a Professor of optics, biomedical engineering, and materials science, and a Senior Scientist at the Laboratory for Laser Energetics. His current research interests include nanoscience and nanotechnology with silicon quantum dots, silicon-based LEDs and lasers, semiconductor optoelectronics and devices, OIs, silicon-based biosensors, the silicon/biology interface, and optical diagnostics. In 1998, he founded the Center for Future Health, a multidisciplinary laboratory where engineers, physicians, and social scientists develop affordable technology that can be used at home to keep individuals independent and healthy. He has authored over 350 publications, edited nine books. Dr. Fauchet has chaired many conferences, including the 2005 IEEE Group Four Photonics Conference. He is a Fellow of the Optical Society of America and the American Physical Society. He received an IBM Faculty Development Award in 1985, an NSF Presidential Young Investigator Award in 1987, an Alfred P. Sloan Research Fellowship in 1988, and the 1990–1993 Prix Guibal & Devillez for his work on porous silicon.
