1004 IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 16, NO. 4, JULY/AUGUST 2010 Target-Localized Nanograting-Based Surface Plasmon Resonance Detection toward Label-free Molecular Biosensing Kyungjae Ma, Dong Jun Kim, Kyujung Kim, Seyoung Moon, and Donghyun Kim (Invited Paper) Abstract—We explore the sensitivity enhancement of label-free detection based on localized surface plasmon resonance using surface-relief nanograting structures. A nanograting structure was modeled, so that target molecular interactions are localized in hot spots of the near fields. The nanograting structure was op- timized numerically for the highest enhancement of sensitivity with hybridization between complementary strands of DNA as the model target interaction. Experimentally, angled evaporation was performed to fabricate the target-localized nanograting samples. Measured data confirm the numerical results that sensitivity en- hancement by an order of magnitude may be feasible on a per-unit-volume basis through target localization. Index Terms—Biosensor, DNA hybridization, field localization, label-free detection, nanostructure, surface plasmon resonance (SPR). I. INTRODUCTION D ETECTION of molecules and molecular interactions has been attempted by many approaches. Optical techniques for molecular detection have been mostly based on organic fluorescent dyes and/or inorganic agents, such as quantum dots. Molecular events are chemically targeted by the fluorescent labels and can be filtered out once the labels are optically excited via electron energy transfer processes. For example, recent introduction of stimulated emission depletion (STED) mi- croscopy and photoactivated localization microscopy (PALM) substantially enhances spatial resolution of fluorescence imag- ing, making acquisition of molecular events using simple optics close to a reality [1], [2]. The labels are approximately a few Manuscript received August 1, 2009; revised August 14, 2009 and Septem- ber 17, 2009; accepted September 30, 2009. Date of publication January 15, 2010; date of current version August 6, 2010. This work was supported in part by the Korea Science and Engineering Foundation (KOSEF) through National Core Research Center for Nanomedical Technology under Grant R15-2004-024- 00000-0 and Grant M10755020003-08N5502-00310, in part by the Korea Re- search Foundation (KRF) Grant funded by the Korean Government under Grant KRF-2008-331-D00389, in part by the Ministry of Knowledge Economy under Project “System IC 2010,” and in part by the Microelectromechanical Systems Research Center for National Defense under Grant 2009-MM-41-UD090024FD funded by the Defense Acquisition Program Administration. K. Ma and D. J. Kim are with the School of Electrical and Electronic Engineering, Yonsei University, Seoul 120–749, Korea (e-mail: economic@ yonsei.ac.kr; sana2jun@yonsei.ac.kr). K. Kim, S. Moon, and D. Kim (corresponding author) are with the Pro- gram for Nanomedical Science and Technology and School of Electrical and Electronic Engineering, Yonsei University, Seoul 120–749, Korea (e-mail: kjkimgo@yonsei.ac.kr; moons@yonsei.ac.kr; kimd@yonsei.ac.kr). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/JSTQE.2009.2034123 nanometers in size. Despite relative ease of selecting out specific target molecular interactions and potential for enhanced resolu- tion and sensitivity, use of labels are often limited by molecular interference, photobleaching, issues related to blinking in the case of quantum dots, and overall difficulty of quantification. In contrast to fluorescence-based imaging modalities, such as STED or PALM, for label-free detection techniques, specificity is mediated by linkers, e.g., antibodies, which bind specifically to target molecules. Whereas label-free detection is free of the concerns related to using labels, it suffers from low spatial res- olution and insufficient sensitivity, thus it usually delivers point measurement data, rather than providing images. Yet, molecu- lar sensing has been attempted, for example, using a microring resonator that detects changes in the whispering gallery mode caused by the binding of target molecules [3]. In this paper, we investigate the feasibility of molecular detection using surface plasmon resonance (SPR). Surface plasmon (SP) refers to a longitudinal electron concentration wave that is formed when TM-polarized light is incident on a metal–dielectric interface. Electronic dipoles associated with SPs create electromagnetic fields that are localized within 100–200 nm near the metal surface. The dispersion relation between SP momentum and energy can be calculated from Maxwell’s equations and is given by k SP = ω c  ε m ε d ε m + ε d = k 0 n s sin θ SP (1) where θ SP , ε m , ε d , and n s represent the angle of incidence at resonance, metal permittivity, target dielectric permittivity, and refractive index of a prism substrate, respectively, k SP is the SP wave number, and ω is the incident-light angular frequency. c and k 0 are for the speed of light and wave number in the free space, respectively. The derivation with other details of (1) can be found in [4]. k SP and θ SP in (1) are required to be complex, as ε m is complex. For gold or silver that is typically used to excite SPs, |Re(ε m )|≫|Im(ε m )|, so that sin θ of a measured reso- nance angle may be approximated as [ε m ε d /(ε m + ε d )] 1/2 /n s from (1). It is basically the change of ε d associated with biomolecular interactions that one measures using SPR. As the near field is quite sensitive to molecular interactions that may occur at surface, by measuring the changes of the resonance condition, one can quantify the interactions in real time. The discovery of SPR dates as early as 1900s. Wood ob- served dark and light bands of light anomaly reflected by a 1077-260X/$26.00 © 2010 IEEE