Detecting changes in the refractive index is important for sensing and imaging applications. Surface plasmon resonances (SPRs) are the collective oscillations of the surface electrons in a structure in response to an incident electromagnetic field. By monitoring SPRs in a metal film, we can detect the refractive index variations of local dielectric environments situated nearby.1 The technique is label-free and allows real-time investigation of the effects of biophysical stimuli exerted on cells and how cells respond to such cues. However, there are still a number of issues to be resolved. The SPR system requires optical couplers such as prisms and gratings, has a limited range of operation—typically within visible frequencies—and, most critically, refractive index changes can only be detected within a short distance of a couple of hundreds of nanometers.2 These limitations impede progress in integrating SPR sensors with low-cost, real-time, and high-throughput biochips for rapid bioanalytical measurements of quantity-limited samples.
We have used a plasmonic sensor based on an array of split-ring resonators (SRRs) (see Figure 1) to improve the sensitivity and detection distance, while preserving the advantages of conventional SPRs. No optical coupler is required to excite the resonances. More importantly, the SRR structures exhibit multiple reflectance peaks (see Figure 2). The spectral positions of these peaks are sensitive to the local dielectric environment and can be quantitatively described by the standing-wave plasmonic resonance model (SWPR),3 providing a design rule for this multi-mode refractive index sensor.4 By applying thin dielectric layers with different thicknesses on the SRR array, we can quantitatively calibrate the sensing performance (such as sensitivity and detection length) of each resonance mode in the multi-resonance reflectance spectra. Our nanostructured SRRs can be used for coupler-free, scalable, and multi-mode refractive index sensing.
Figure 1. Schematic of the designed split-ring resonator (SRR) unit cell (inset) and scanning electron microscopy image of a fabricated planar SRR array.
The normalized reflectance spectra of five different-sized planar SRRs from 100nm (d100) to 750nm (d750) without a layer of poly(methyl methacrylate) (PMMA). Different panels show different excitation modes: (a) 1∥mode, (b) 2⊥
mode, (c) 3∥, and (d) 4⊥
mode. The odd modes (1∥, 3∥), and even modes (2⊥
) correspond to two orthogonal polarization directions of the electric field indicated on the arrays in the insets in red (E∥ and E⊥
), respectively. (a) and (b) are measured within the mid-IR region and (c) and (d) are measured within the near-IR by micro Fourier-transform IR spectroscopy. Here no optical coupler is required to excite plasmonic resonance.5
Figure 2 shows the measured spectra. They have multiple reflectance peaks whose resonance wavelengths can be determined from the SWPR model.3 The effective refractive index of the dielectric environment, neff, stems from the collective contribution of substrates (nsub), analytes (na), and surroundings (nsur). We assume a linear combination so that neff is expressed as: where xi and ni represent the fraction and the refractive index of the species i, respectively. While varying the analyte on the SRR plasmonic sensor, we introduce a fluctuation of the effective refractive index, ∂neff, leading to a wavelength shift in reflectance peaks ∂λm. Thus, the corresponding sensitivity, S, of the SRR sensor is given by: where L denotes the total length of the SRR, λm is the resonance wavelength, m is the resonance mode, and λ0 depends on the geometric structure. For such a multi-mode refractive index sensor, this derived model clearly provides a quantitative description of the sensitivity with respect to the resonance mode (m) and the size of the SRR structure (L). The linear dependence of the sensitivity on the resonance mode indicates that the SRR structure has great potential for practical refractive index sensing applications.6
Normalized reflectance spectrum of planar SRRs. Red and black curves represent the responses with and without a 500nm layer of PMMA film. Different panels show different modes of excitation: (a) 1∥and 3∥ modes, (b) 2⊥
mode, (c) 5∥ mode, and (d) 4⊥
mode. Insets indicate different field orientations with respect to the array. Data in (a) and (b) are measured within the mid-IR region. The bottom panels are measured within the near-IR by micro Fourier-transform IR spectroscopy. Here no optical coupler is required to excite plasmonic resonance.5
We apply a 500nm layer of poly(methyl methacrylate) (PMMA) on top of the planar SRRs as an effective change in the refractive index of the surrounding medium. We measure the reflectance spectra of the multiple resonance modes of the SRR with and without the PMMA layer to examine this sensitivity formula (see Figure 3). The sensitivity depends linearly on the reciprocal of the resonance modes, and the sensor is most sensitive in the primary resonance mode as predicted in Equation (2) (see Figure 4). Moreover, the SRR samples are more sensitive than conventional plasmonic sensors as a result of the longer resonance wavelength. The measurements indicate an excellent value for the change in resonance wavelength of 2700nm per unit refractive index change (per RIU) at the first resonance. The performance of the SRR plasmonic biosensor equals or even exceeds that of other refractive index biosensors. For example, the sensitivity of prism coupler-based surface plasmon polariton (SPP) biosensors ranges from 970 to 13,800nm/RIU, depending on the resonance wavelength,1 and that of localized surface plasmon resonance (LSPR) biosensors is between 1207 and 500nm/RIU.8
Figure 4. Linear relationship between resonance wavelength (λm)and the reciprocal of resonance mode (1/m). The slope is proportional to the total length of the SRR (L) and a correction factor (xa). The different size parameters of the designed SRR structure are indicated by d450 and d510.
Simulated results of the varied detection lengths with respect to resonance modes 1–5. (a) There is little effect on the resonance frequency as the PMMA layer thickness increases for mode 1, 2, and 3 (1∥, 2⊥
, and 3∥ resonance modes). (b) There is a larger shift in the resonance frequency as the PMMA layer thickness increases for modes 4 and 5 (4⊥
and 5∥ resonance modes).5
In addition to quantitatively demonstrating the sensitivity of the SRR plasmonic biosensor, we have also investigated its detection length by gradually increasing the thickness of the dielectric PMMA layer applied on top of the SRR until the corresponding wavelength shift saturated. For lower resonance modes 1∥, 2⊥, and 3∥, the shift of wavelength saturates as the thickness of PMMA layers is about 200–500nm: see Figure 5(a). The odd modes (1∥, 3∥, 5∥) and even modes (2⊥, 4⊥) correspond to two orthogonal E-field polarization directions (E∥ and E⊥), respectively. In contrast, for higher resonance modes 4⊥ and 5∥, no saturation effect is observed even as the thickness is increased to 2μm: see Figure 5(b). The detection length of the sensor is much greater than conventional SPR sensors.2
In terms of sensitivity and detection length, the SRR-based optical sensors promise a real-time, tunable, and multi-mode solution for biological and chemical detection, drug delivery, and other applications. The proposed quasi-quantitative standing-wave plasmonic resonance mode also provides a good guideline for the sensing behavior. We will focus our future research on constructing a compact microscopic platform based on SRRs to image whole cells for applications in intracellular investigation.
Ta-Jen Yen, Yueh-Chun Lai
National Tsing Hua University
1. J. Homola, S. S. Yee, G. T. Gauglitz, Surface plasmon resonance sensor: review, Sens. Act. B 54, pp. 3-15, 1999.
2. H. Raether, Surface Plasmons on Smooth and Rough Surfaces and on Gratings, Springer, 1988.
3. C. Y. Chen, S. C. Wu, T. J. Yen, Experimental verification of standing-wave plasmonic resonances in split-ring resonators, Appl. Phys. Lett. 93, pp. 034110, 2008.
4. T. Driscoll, G. O. Andreev, D. N. Basov, S. Palit, S. Y. Cho, N. M. Jokerst, D. R. Smith, Tuned permeability in terahertz split-ring resonators for devices and sensors, Appl. Phys. Lett. 91, pp. 062511, 2007.
5. Y. T. Chang, Y. C. Lai, C. T. Lee, C. K. Chen, T. J. Yen, A multi-functional plasmonic biosensor,
Opt. Express 18, pp. 9561-9569, 2010.
6. A. Cunningham, Introduction to Bioanalytical Sensors, John Wiley & Sons, 1998.
7. J. J. Mock, D. R. Smith, S. Schultz, Local refractive index dependence of plasmon resonance spectra from individual nanoparticles, Nano Lett. 3, pp. 485-491, 2003.
8. M. M. Miller, A. A. Lazarides, Sensitivity of metal nanoparticle plasmon resonance band position to the dielectric environment as observed in scattering, J. Opt. A: Pure Appl. Opt. 8, pp. S239-S249, 2006.