- Email: [email protected]
A planar split-ring resonator-based microwave biosensor for label-free detection of biomolecules
Sensors and Actuators B 169 (2012) 26–31
Contents lists available at SciVerse ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
A planar split-ring resonator-based microwave biosensor for label-free detection of biomolecules Hee-Jo Lee a,1 , Jung-Hyun Lee b,1 , Hui-Sung Moon c , Ik-Soon Jang d , Jong-Soon Choi d,e , Jong-Gwan Yook f , Hyo-Il Jung b,g,∗ a
Graphene Research Institute, Sejong University, Seoul, Republic of Korea National Core Research Center for Nanomedical Technology, Yonsei University, Seoul, Republic of Korea c Bio Lab, Emerging Tech. R&D Center, Samsung Advanced Institute of Technology, Suwon, Republic of Korea d Division of Life Science, Korea Basic Science Institute, Daejon, Republic of Korea e Graduate School of Analytical Science and Technology, Chungnam National University, Daejon, Republic of Korea f School of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of Korea g School of Mechanical Engineering, Yonsei University, Seoul, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 21 November 2011 Received in revised form 5 January 2012 Accepted 15 January 2012 Available online 23 January 2012 Keywords: Biosensor Cortisol Prostate specific antigen Radio-frequency Label-free
a b s t r a c t In this study, a planar split-ring resonator (SRR)-based RF biosensor was developed for label-free detection of biomolecules such as the prostate cancer marker, prostate specific antigen (PSA), and cortisol stress hormone. The biosensor has a resonance-assisted transducer and is excited by a time-varying magnetic field component of a local high-impedance microstrip line. The resulting device exhibits an intrinsic S21 resonance with a quality-factor (or Q-factor) of 50. For the biomolecular interaction, anti-PSA and anticortisol were immobilized on the gold surface of the resonator by a protein-G mediated bioconjugation p process and corresponding frequency shifts of f1 = 30 ± 2 MHz (for anti-PSA) and f1c = 20 ± 3 MHz (for anti-cortisol) were observed. The additional frequency shift of each PSA and cortisol antigen with a 100 pg/ml concentration was about 5 ± 1.5 MHz and 3 ± 1 MHz, respectively. From the experimental results, we confirmed that our device is very effective RF biosensor with a limit of detection (LOD) of 100 pg/ml and has sufficiently feasibility as a label-free biosensing scheme. Crown Copyright © 2012 Published by Elsevier B.V. All rights reserved.
1. Introduction In the past years, many approaches using electrochemical and electrical signals for the detection of biomolecule have been reported, including micro-cantilever [1,2], surface plasmon resonance (SPR) [3], and nanowire-based biosensors [4]. In particular, a change in the resonant frequency generated by an antigen-antibody interaction was detected in the micro-cantilever-based biosensor approach [1]. The biosensing system could operate at low frequencies, unlike the conventional SPR-based biosensor in the optical regime. Although these biosensing methodologies with mechanical, electrical, and optical transducers could be used to successfully achieve sensitive, real-time, and label-free detection, such biosensors still require sophisticated and complex fabrication processes or expensive instrumentation.
∗ Corresponding author at: Laboratory of Biochip Technology, Yonsei University, Seoul, Republic of Korea. Tel.: +82 2 2123 5814; fax: +82 2 312 2159. E-mail address: [email protected] (H.-I. Jung). 1 These authors contributed equally to this work.
In this work, to allow rapid and label-free biosensing in the microwave region, a cost-efficient and miniaturized biosensor with a simple and direct transducer scheme is proposed. In particular, we consider a planar split-ring resonator (SRR) as a transducer that allows the detection of biomolecules such as prostate-specific antigen (PSA) and cortisol. PSA is a serine protease produced by the prostate epithelium to maintain liquefaction of seminal fluid, and is the best and most widely available serum biomarker for the detection of prostate cancer [5]. The steroid hormone cortisol is another important biomarker for numerous diseases and functions in the regulation of blood pressure, glucose levels, and carbohydrate metabolism at normal physiological levels [6–8]. Cortisol has been widely investigated as an indicator of the stress-load of the human body. We therefore selected PSA and cortisol as biomarkers for verifying the feasibility of our biosensor. The SRR has attracted a great deal of interest in recent years because it can be used as an artificial magnetic element (or inclusion) of electromagnetic metamaterials to produce negatively effective magnetic permeability in the RF/microwave [9–11] to THz [12–15] range. Since the dimensions of the SRR are much smaller than the free space wavelength of the incident electromagnetic fields, production of a compact and miniaturized RF/microwave
0925-4005/$ – see front matter. Crown Copyright © 2012 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2012.01.044
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
27
Fig. 1. Schematic of the complete biosensing device and the surface current distribution of the simulated SRR element. (a) The device consists of an SRR and a high-impedance microstrip line with a masking layer; (b) the surface current distribution and intensity of the SRR element excited by a high-impedance microstrip line at resonant frequency.
device using SRR arrays is feasible [16–18]. Moreover, provided that the loss characteristics of the resonator itself can be further diminished, the design of a high-Q resonator in the quasi-optical region is plausible. The characteristics of the SRR element would allow it to be used as a resonator for sensing applications. For example, a nanoscaled-plasmonic SRRs-based biosensor has been used for DNA sensing at optical frequencies [19]. In addition, we previously identified changes in the resonant frequency due to biotin-streptavidin binding on the surface of the SRRs array at microwave frequencies [20]. Although the biosensing device showed a remarkable frequency shift upon biomolecular binding, we used a relatively high-concentration on the SRRs array and the procedure was somewhat complex and cumbersome due to the use of a liquid wall tank for surface biological processing of the pattern, thus presenting difficulties in testing numerous samples. In this study, our findings show that an antigen-antibody binding system can be effectively and rapidly detected by the SRR-based RF biosensor in a local high-impedance microstrip line system. 2. Materials and methods 2.1. Design, fabrication, and measurement of the split-ring resonator
We have designed a biosensing device for biomolecule detection (Fig. 1a). It consists of a microstrip line, which is a kind of planar transmission line for transmitting electromagnetic energy, with local high-impedance and an SRR element. The high-impedance microstrip line functions to effectively and intensively excite the resonator because the high current produced by the narrow line enhances the time-varying magnetic field component. The characteristic impedance of the line was about 80 and the other parts of the microstrip line (input and output port) were matched at 50 . With this topology, we expected that strong circular current flows could be produced on the surface of the inner and outer rings (Fig. 1b). The resonator element can be modeled as input impedance seen from the microstrip line using the following equation:
ZSRR = R + j ωL −
1 ωC
overall area of the microstrip line was 20 by 10 mm2 . The electric permittivity (εr ) and loss tangent (tan ı) of the dielectric layer were 9.7 and 0.0036, respectively, and its thickness was 0.76 mm. We performed a full-wave electromagnetic simulation based on the finite element method (FEM) to predict the resonant frequency and the surface current distribution of the resonator. From the resulting simulation, it is worth noting that the intensity of the current is highly strengthened at the outer ring of the resonator. The device was fabricated with well-defined printed circuit board (PCB) technology (Fig. 2a, right) [21]. To construct a biolayer on the surface of the resonator, the copper pattern was coated with a nickel (Ni) adhesion layer at a thickness of 3–5 m, followed by top-coating with a gold (Au) layer at a thickness of 0.05 m. To confine biomolecular binding to the surface of the resonator itself, the entire device, excluding the active sensing region and the edge contact pads used for measurement, was coated with a masking layer at a thickness of 10–20 m. The RF measurement system for the fabricated sample is shown in Fig. 2a (left). This system comprised an RF test fixture associated with a two-port vector network analyzer (VNA) system. From the VNA system, we could obtain the scattering parameters (S-parameters) of the biosensing device. The magnitude of the S-parameters is defined as
(1)
where R, L, and C are the resistance, inductance, and capacitance components, respectively and ω = 2f is the angular frequency. The width and length of the high-impedance line were 0.2 mm and 3.56 mm (including two tapered sections on both sides), respectively, and the width and length of the other two parts of the matching line were 0.78 mm and 16.44 mm, respectively. The
S11 (= S22 ) = 20 log10
V1− V1+
,
S21 (= S12 ) = 20 log10
V2−
V1+
(2)
where S11 and S21 are the reflection (ratio of the reflected voltage wave to the incident voltage wave: V1− /V1+ ) and transmission coefficients (ratio of the transmitted voltage wave to the incident voltage wave: V1− /V1+ ), respectively. The S21 parameter of the resonator was used as a discriminating signal for biomolecule detection due to its resonant frequency. Before biological application, we measured the resonant frequency and magnitude of the bare samples. The measured results had a lower frequency of 180 MHz and a higher magnitude of 6.48 dB relative to the simulated results (Fig. 2b). In addition, the simulated resonator had a Q-factor of 80 at the resonant frequency, while the measured Q-factor decreased to 50 due to various losses such as metal, radiation, and substrate loss. However, this discrepancy was not significant, since the biosensing procedure is based purely on relative frequency and magnitude shift. 2.2. Surface modification of the split-ring resonator In this work, PSA and cortisol as biomarkers for verifying the feasibility of our clinical biosensor were selected. In the sensor surface, cysteine (Cys) 3-mediated protein-G was first immobilized on a gold surface to allow binding to the Fc (constant fragment) region
28
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
Fig. 2. Schematic of the RF measurement system. (a) The sample fixture system associated with a network analyzer (left) and fabricated sample (right); SRR dimensions are w = 0.19 mm, s = 0.19 mm, d = 0.08 mm, g = 0.1 mm, and a = 1.63 mm; (b) the S21 resonance of simulated and measured SRR samples.
Fig. 3. Immobilization of anti-PSA and cortisol on the surface of the SRR device and binding experiments. (a) Experimental group with anti-PSA (or anti-cortisol); (b) control group without anti-PSA (or anti-cortisol). Fluorescent microscopic images of the surface of the SRR biosensor treated with PSA-Cy3 (c) without anti-PSA (or anti-cortisol) and (d) with anti-PSA (or anti-cortisol).
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
29
2.3. Observation of a specific interaction between anti-PSA and PSA using fluorescent dye
10 – 15 ± 2 100 – 20 ± 3 1 – 10 ± 2 10 – 20 ± 3 a
PBS
– ∼0 –
Materials on the sensor surface
Concentration (ng/ml) f1 (MHz) f2 (MHz)
The frequency shifts of the interactions of PSA and anti-PSA were investigated (Fig. 4). For the antibody immobilized on the surface of the sensor, there was a shift to the lower frequency region and the S21 resonance level changed slightly compared to the bare sensors (see footnote a in Table 1). When different PSA concentrations (100 ng/ml, 10 ng/ml, 1 ng/m, and 100 pg/ml) were applied, the resonant frequency underwent an additional shift to lower frequency regions with increased PSA concentration, from 30 ± 2 MHz (100 ng/ml) to 5 ± 1.5 MHz (100 pg/ml). The Cys3-protein-G molecular cross-linker through which antibody can be immobilized did not show a significant frequency shift (1–2 MHz; Table 1). Our biosensor could distinguish between anti-PSA binding and the bare surface of the sensor, reflecting the proper immobilization of
Table 1 Summary of experimental results.
3.2. Frequency shift measurements
The average value of resonant frequency for 10 bare samples is 10.48 ± 0.22 GHz.
100 – 30 ± 2 500 20 ± 5 –
Prior to the frequency shift measurements, fluorescent microscopic images were obtained to investigate the surface modification due to the presence of anti-PSA and the subsequent binding of Cy3-labeled PSA labeled. A very strong red fluorescence image was observed when fluorescently-labeled PSA was added to the antibody-coated biosensor (Fig. 3d). The absence of any fluorescence when the labeled PSA was added to the bare surface of the biosensor (i.e., no antibody coating) (Fig. 3c) indicated that the interaction of PSA with the antibody on the sensor surface was specific.
Antibody (PSA)
3.1. Optical detection of PSA binding to anti-PSA antibody immobilized on the sensor
1 × 106 1–2 ± 0.5 –
3. Results and discussion
Cys-3-mediated Protein G
Prostate specific antigen (PSA)
0.1 – 5 ± 1.5
Fluorescent staining was used to examine the specific interaction between anti-PSA antibody and PSA. PSA molecules were fluorescently labeled by dissolving fluorolinker Cy3 mono-reactive dye (100 ng/mL, Amersham Biosciences, USA) in 1 M sodium bicarbonate buffer (pH 9.3) and allowing it to react with the PSA (100 ng/ml, Abcam, UK) for 1 h at room temperature. The free dye was then separated from the labeled PSA using dye removal resin in fluorescent dye-removal columns (Thermo, USA). The absorbance and emission wavelengths of Cy3 are 550 nm and 570 nm, respectively. To verify that the fluorescence resulted from the specific binding of PSA to the antibody, the SRR sensor without anti-PSA was also treated with Cy3-PSA and then rinsed with PBS buffer. Both samples were then observed under a fluorescence microscope.
500 18 ± 4 –
Antibody (Cortisol)
Cortisol-BSA conjugate
1 – 5±2
0.1 – 3±1
of the antibody [22–24]. PSA and cortisol biomolecules were then introduced (Fig. 3). The biological process is in detail as follows: the gold surface of the sensor was treated with 10 g/ml Cys3-mediated protein-G in phosphate buffered saline (PBS) solution at pH 7.4 for approximately 1 h. Next, 500 ng/ml of anti-PSA solution was added to the sensor, followed by BSA (about 1 mg/ml) to prevent nonspecific surface binding. Finally, four different PSA and cortisol-BSA concentrations (100 ng/ml, 10 ng/ml, 1 ng/ml, and 100 pg/ml) were used in the assay (Fig. 3a). Binding experiments were conducted in PBS solution (pH 7.4) for about 90 min. For the control experiment (Fig. 3b), the sensor surface was treated with protein-G and BSA under the same conditions in the absence of anti-PSA solution. PSA was used at a concentration of 100 g/ml. The sensor surface was thoroughly rinsed with PBS solution between each step and then immediately measured using the RF measurement system.
30
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
a concentration of 100 pg/ml (Table 1). The minimum detectable level of our device is lower level than conventional SPR-based biosensor for PSA in sample volume of 130 m [25]. In this work, the resonant frequency shift for biomolecular binding can be easily predictable. According to quasi-static analysis [26], the resonant frequency can be predicted since the proposed biosensing scheme essentially behaves as an LC circuit, as follows: ω02 =
2 , rav LC
(3)
L = Lm + Li + Lo ,
Fig. 4. Frequency shift according to the buffer solution (PBS) and various concentrations of PSA.
eff
anti-PSA on the sensor surface, which generated a reasonable frequency shift. As a reference experiment, PBS solution without any biomolecules was applied to the sensor and after complete drying, no frequency shift was observed, indicating that the signal was generated from the specific biomolecular interactions. The molecular weight of cortisol is very small and we expected that cortisol would not produce a significant signal in our device. Thus, bovine serum albumin (BSA)-conjugated cortisol (cortisol-BSA) was used to amplify the signal. Indeed, when free cortisol was applied to the anti-cortisol antibody-coated surface of the sensor, a very small signal was observed (data not shown). The BSA without cortisol did not show any signal either, suggesting that the frequency shift upon addition of cortisol-BSA originated solely from the interaction of anti-cortisol antibody with cortisol. The frequency shifts of cortisol antigen-coupled BSA binding to its antibody is shown in Fig. 5. A similar shift to lower resonant frequency regions was observed, but the changes in frequency were smaller than those of the four different PSA concentrations. The variation in frequency shift and magnitude may result from uncertainties in the preparation of the biolayers. Because the samples were prepared by a biological process in which the biosensors were left to soak in static liquid PBS with biomolecules, the biolayer on the gold surface of the sensors may not have been uniformly coated with biomolecules. Based on the results, the sensing limitation of our device was estimated as
rav =
(ri + ro ) 2
(4)
where rav is the average radius between rings and L and C are the total inductance and capacitance of the structure, respectively. In Eq. (4), the total inductance consists of the mutual inductance (Lm ) per unit length between rings, the self inductance (Li ) of the inner ring and the self inductance (Lo ) of the outer ring. Similarly, the total capacitance is comprised of the mutual capacitance (Cm ) per unit length between rings, the split capacitance (Ci ) of the inner ring and the split capacitance (Co ) of the outer ring. Since the biosensing samples are immobilized with Cys3-modified protein-G coupled with PSA antibody and are then bound to the PSA, additional layers due to biomolecular binding can be modeled as two nano-sized parallel dielectric layers with a different permittivity on the gold surface. The total effective capacitance for the two layers can be expressed as C ∝ εr1
Fig. 5. Frequency shift according to the buffer solution (PBS) and various concentration of cortisol-coupled BSA.
C = Cm + Ci + Co ,
Apul1 dpul1
eff
+ εr2
Apul2 dpul2
.
(5)
In this equation, the first term is the effective capacitance produced by the immobilization layer of protein-G coupled to PSA antibody, while the second term is the effective capacitance proeff eff duced by the PSA binding layer, where εr1 and εr2 are the effective permittivities of the first and second dielectric layer, respectively. Consequentially, although the frequency shifts for the biomolecular binding resulted from a change in inductance components, the change in the capacitance component is predominantly due to the electrical permittivity of the biomolecule and the electric field distribution between the inner and outer rings. As indicated by Eq. (5), the overall effective capacitance component increases as biomolecular binding progresses and, therefore, the resonant frequency shifts toward the lower frequency region. 4. Conclusion We have clearly demonstrated the use of our planar SRRbased RF biosensor excited by a local high-impedance microstrip line as a label-free biosensor with a resonant-assisted transduction scheme. In terms of biomolecular detection, such as PSA and cortisol antigen-antibody binding systems, our device has a very simple, direct, and sensitive transduction scheme at microwave frequencies. In addition, it requires a small volume of biomolecular specimen, has a fast response time, is easy to operate and requires no labeling process. We therefore suggest that our biosensing approach is a good candidate for a RF/microwave and THz biosensor by using a more sophisticated microfabrication process. We believe that this seminal work may pave the way to realizing a label-free RF biosensing transducer for simple and direct detection of human specimens at the desired radiation frequencies. Acknowledgments This study was supported by grants from the National R&D Program for Cancer Control, Ministry of Health & Welfare (No.1120290), the National Research Foundation of Korea (NRF)
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
(No.2011-0016731) and the Korean Basic Science Institute (No. T3278B). References [1] J.H. Lee, K.S. Hwang, J. Park, K.H. Yoon, D.S. Yoon, T.S. Kim, Immunoassay of prostate-specific antigen (PSA) using resonant frequency shift of piezoelectric nanomechanical microcantilever, Biosens. Bioelectron. 20 (2005) 2157–2162. [2] K.S. Hwang, J.H. Lee, J. Park, D.S. Yoon, J.H. Park, T.S. Kim, In situ quantitative analysis of a prostate-specific antigen (PSA) using a nanomechanical PZT cantilever, Lab. Chip 4 (2004) 547–552. [3] W.D. Wilson, Analyzing biomolecular interactions, Science 295 (2002) 2103–2105. [4] G. Zheng, F. Patolsky, Y. Cui, W.U. Wang, C.M. Lieber, Multiplexed electrical detection of cancer markers with nanowire sensor arrays, Nat. Biotechnol. 23 (2005) 1294–1301. [5] R.B. Nadler, P.A. Humphrey, D.S. Smith, W.J. Catalona, T.L. Ratliff, Effect of inflammation and benign prostatic hyperplasia on elevated serum prostate specific antigen levels, J. Urol. 154 (1995) 407–413. [6] J.C. Zhou, M.H. Chuang, E.H. Lan, B. Dunn, P.L. Gillman, S.M. Smith, Immunoassays for cortisol using antibody-doped sol–gel silica, J. Mater. Chem. 14 (2004) 2311–2316. [7] S.S.C. Tai, M.J. Welch, Development and evaluation of a candidate reference method for the determination of total cortisol in human serum using isotope dilution liquid chromatography/mass spectrometry and liquid chromatography/tandem mass spectrometry, Anal. Chem. 76 (2004) 1008–1014. [8] R.C. Stevens, S.D. Soelberg, S. Near, C.E. Furlong, Detection of cortisol in saliva with a flow-filtered, portable surface plasmon resonance biosensor system, Anal. Chem. 80 (2008) 6747–6751. [9] D.R. Smith, Willie J. Padilla, D.C. Vier, S.C. Nemat-Nasser, S. Schultz, Composite medium with simultaneously negative permeability and permittivity, Phys. Rev. Lett. 84 (2000) 4184–4187. [10] R.A. Shelby, D.R. Smith, S. Schultz, Experimental verification of a negative index of refraction, Science 292 (2001) 77–79. [11] A.A. Houck, J.B. Brock, I.L. Chuang, Experimental observations of a left-handed material that obeys Snell’s law, Phys. Rev. Lett. 90 (2003) 137401. [12] S. Linden, C. Enkrich, M. Wegener, J. Zhou, T. Koschny, C.M. Soukoulis, Magnetic response of metamaterials at 100 terahertz, Science 306 (2004) 1351–1353. [13] H.O. Moser, B.D.F. Casse, O. Wilhelmi, B.T. Saw, Terahertz response of a microfabricated rod-split-ring-resonator electromagnetic metamaterial, Phys. Rev. Lett. 94 (2005) 063901. [14] H.-T. Chen, W.J. Padilla, J.M.O. Zide, A.C. Gossard, A.J. Taylor, R.D. Averitt, Active terahertz metamaterial devices, Nature 444 (2006) 597–600. [15] 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 (2007) 062511. [16] J. Garcia-Garcia, F. Martin, F. Falcone, J. Bonache, J.D. Baena, I. Gil, E. Amat, T. Lopetegi, M.A.G. Laso, J.A.M. Iturmendi, Microwave filters with improved stopband based on sub-wavelength resonators, IEEE Trans. Microwave Theory Tech. 53 (2005) 1997–2006. [17] J. Bonache, I. Gil, J. Garcia-Garcia, F. Martin, Novel microstrip bandpass filters based on complementary split-ring resonators, IEEE Trans. Microwave Theory Tech. 54 (2006) 265–271. [18] I. Bulu, H. Caglayan, K. Aydin, E. Ozbay, Compact size highly directive antennas based on the SRR metamaterial medium, New J. Phys. 7 (2005) 223. [19] A.W. Clark, A. Glidle, D.R.S. Cumming, J.M. Cooper, Plasmonic split-ring resonators as dichroic nanophotonic DNA biosensors, J. Am. Chem. Soc. 131 (2009) 17615–17619. [20] H.J. Lee, J.G. Yook, Biosensing using split-ring resonators at microwave regime, Appl. Phys. Lett. 92 (2008) 254103. [21] H.J. Lee, H.S. Lee, K.H. Yoo, J.G. Yook, DNA sensing using split-ring resonator alone at microwave regime, J. Appl. Phys. 108 (2010) 014908. [22] W. Lee, B.K. Oh, W.H. Lee, J.W. Choi, Immobilization of antibody fragment for immunosensor application based on surface plasmon resonance, Colloids Surf. B: Biointerfaces 40 (2005) 143–148. [23] Z. Shen, G.A. Stryker, R.L. Mernaugh, L. Yu, H. Yan, X. Zeng, Single-chain fragment variable antibody piezoimmunosensors, Anal. Chem. 77 (2005) 797–805.
31
[24] J.M. Lee, H.K. Park, Y. Jung, J.K. Kim, S.O. Jung, B.H. Chung, Direct immobilization of protein G variants with various numbers of cysteine residues on a gold surface, Anal. Chem. 79 (2007) 2680–2687. [25] D.A. Healy, C.J. Hayes, P. Leonard, L. McKenna, R. O’Kennedy, Biosensor developments: application to prostate-specific antigen detection, Trends Biotechnol. 25 (2007) 125–131. [26] R. Marqués, F. Medina, R. Rafii-El-Idrissi, Role of bianisotropy in negative permeability and left-handed metamaterials, Phys. Rev. B 65 (2002) 144440.
Biographies Hee-Jo Lee received the M.Sc. (Physics and Applied Physics) and Ph.D. (Electrical and Electronic Engineering) degrees from Yonsei University, in 2004 and 2010, respectively, Seoul, South Korea. During his Ph.D. degree and first Post Doctoral Researcher at BK21 Institute of Information and Technology in Yonsei University, he was dedicated to the metamaterials for biosensing, carbon nanotube-based RF nanobiosensors, and electromagnetic field theory. Since November 2010, he joined the Graphene Research Institute (GRI), Sejong University, Seoul, Korea, where he is now a Post Doctoral Research Fellow. His research interests include RF circuit modeling and characterization of graphene, graphene-based RF field-effect transistors (FETs)/flexible RF devices/RF nanobiosensors, and computational electromagnetics. Jung-Hyun Lee received the B.Sc. degree (Bio & Environmental Technology) from Seoul Women’s University, South Korea in 2009. She is currently a MS degree student at the National Core Research Center for Nanomedical Technology, Yonsei University. Her research interests include the development of analytical system for biomedical and environmental monitoring. Hui-Sung Moon received the B.Sc. and Ph.D. (Mechanical Engineering) degrees from Yonsei University, South Korea in 2005 and 2011, respectively. During his Ph.D. degree, he pursued his research on microfluidic cell manipulation. He is currently a research staff in Samsung Advanced Institute of Technology (SAIT). Ik-Soon Jang received the B.Sc. degrees from Sungkyunkwan University in 1995, and M.Sc. and the Ph.D. (Biochemistry and Molecular Biology) degree from the Seoul National University, College of Medicine in 1997 and 2001. He is currently a senior scientist at the Korea Basic Science Institute (KBSI), and his research interests in Aging Mechanism by using proteomics and development of membrane protein association. Jong-Soon Choi received the B.Sc. and M.Sc. (Biology) degrees from Yonsei University in 1986 and 1990, respectively, and the Ph.D. (Microbiology) degree from the Korea Advanced Institute of Science and Technology (KAIST) in 1999. He is the principal researcher in the Division of Life Science of Korea Basic Science Institute. Also, he is the professor in the Graduate School of Analytical Science and Technology of Chungnam National University. His main research is oriented in the analytical forensic science using functional proteomics. Jong-Gwan Yook received the B.Sc. and M.Sc. (Electronic Engineering) degrees from the Yonsei University in 1987 and 1989, respectively, and the Ph.D. (Electrical and Computer Engineering) degree from the University of Michigan, Ann Arbor, USA in 1996. He is currently a professor at the School of Electrical and Electronic Engineering, Yonsei University, and his research interests include the areas of theoretical/numerical electromagnetic modeling and characterization of microwave/millimeter-wave circuits and components, design of RF integrated circuits (RFICs) and monolithic microwave integrated-circuits (MMICs), and analysis and optimization of high-frequency high-speed interconnects, including RF microelectromechanical system (MEMS), based on frequency as well as time-domain full-wave methods. His research team has also been engaged in the development of biosensors, such as CNTs RF biosensor for nanometer-sized antigen-antibody detection as well as remote wireless vital signal monitoring sensors. Hyo-Il Jung received the B.Sc. and M.Sc. (Biotechnology) degrees from the Korea Advanced Institute of Science and Technology (KAIST) in 1993 and 1995, respectively, and the Ph.D. (Physical Biochemistry) degree from the University of Cambridge, United Kingdom. He is currently an associate professor at the School of Mechanical Engineering, Yonsei University, and his research interests center on microfluidic biosensors and cancer diagnosis.
Contents lists available at SciVerse ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
A planar split-ring resonator-based microwave biosensor for label-free detection of biomolecules Hee-Jo Lee a,1 , Jung-Hyun Lee b,1 , Hui-Sung Moon c , Ik-Soon Jang d , Jong-Soon Choi d,e , Jong-Gwan Yook f , Hyo-Il Jung b,g,∗ a
Graphene Research Institute, Sejong University, Seoul, Republic of Korea National Core Research Center for Nanomedical Technology, Yonsei University, Seoul, Republic of Korea c Bio Lab, Emerging Tech. R&D Center, Samsung Advanced Institute of Technology, Suwon, Republic of Korea d Division of Life Science, Korea Basic Science Institute, Daejon, Republic of Korea e Graduate School of Analytical Science and Technology, Chungnam National University, Daejon, Republic of Korea f School of Electrical and Electronic Engineering, Yonsei University, Seoul, Republic of Korea g School of Mechanical Engineering, Yonsei University, Seoul, Republic of Korea b
a r t i c l e
i n f o
Article history: Received 21 November 2011 Received in revised form 5 January 2012 Accepted 15 January 2012 Available online 23 January 2012 Keywords: Biosensor Cortisol Prostate specific antigen Radio-frequency Label-free
a b s t r a c t In this study, a planar split-ring resonator (SRR)-based RF biosensor was developed for label-free detection of biomolecules such as the prostate cancer marker, prostate specific antigen (PSA), and cortisol stress hormone. The biosensor has a resonance-assisted transducer and is excited by a time-varying magnetic field component of a local high-impedance microstrip line. The resulting device exhibits an intrinsic S21 resonance with a quality-factor (or Q-factor) of 50. For the biomolecular interaction, anti-PSA and anticortisol were immobilized on the gold surface of the resonator by a protein-G mediated bioconjugation p process and corresponding frequency shifts of f1 = 30 ± 2 MHz (for anti-PSA) and f1c = 20 ± 3 MHz (for anti-cortisol) were observed. The additional frequency shift of each PSA and cortisol antigen with a 100 pg/ml concentration was about 5 ± 1.5 MHz and 3 ± 1 MHz, respectively. From the experimental results, we confirmed that our device is very effective RF biosensor with a limit of detection (LOD) of 100 pg/ml and has sufficiently feasibility as a label-free biosensing scheme. Crown Copyright © 2012 Published by Elsevier B.V. All rights reserved.
1. Introduction In the past years, many approaches using electrochemical and electrical signals for the detection of biomolecule have been reported, including micro-cantilever [1,2], surface plasmon resonance (SPR) [3], and nanowire-based biosensors [4]. In particular, a change in the resonant frequency generated by an antigen-antibody interaction was detected in the micro-cantilever-based biosensor approach [1]. The biosensing system could operate at low frequencies, unlike the conventional SPR-based biosensor in the optical regime. Although these biosensing methodologies with mechanical, electrical, and optical transducers could be used to successfully achieve sensitive, real-time, and label-free detection, such biosensors still require sophisticated and complex fabrication processes or expensive instrumentation.
∗ Corresponding author at: Laboratory of Biochip Technology, Yonsei University, Seoul, Republic of Korea. Tel.: +82 2 2123 5814; fax: +82 2 312 2159. E-mail address: [email protected] (H.-I. Jung). 1 These authors contributed equally to this work.
In this work, to allow rapid and label-free biosensing in the microwave region, a cost-efficient and miniaturized biosensor with a simple and direct transducer scheme is proposed. In particular, we consider a planar split-ring resonator (SRR) as a transducer that allows the detection of biomolecules such as prostate-specific antigen (PSA) and cortisol. PSA is a serine protease produced by the prostate epithelium to maintain liquefaction of seminal fluid, and is the best and most widely available serum biomarker for the detection of prostate cancer [5]. The steroid hormone cortisol is another important biomarker for numerous diseases and functions in the regulation of blood pressure, glucose levels, and carbohydrate metabolism at normal physiological levels [6–8]. Cortisol has been widely investigated as an indicator of the stress-load of the human body. We therefore selected PSA and cortisol as biomarkers for verifying the feasibility of our biosensor. The SRR has attracted a great deal of interest in recent years because it can be used as an artificial magnetic element (or inclusion) of electromagnetic metamaterials to produce negatively effective magnetic permeability in the RF/microwave [9–11] to THz [12–15] range. Since the dimensions of the SRR are much smaller than the free space wavelength of the incident electromagnetic fields, production of a compact and miniaturized RF/microwave
0925-4005/$ – see front matter. Crown Copyright © 2012 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2012.01.044
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
27
Fig. 1. Schematic of the complete biosensing device and the surface current distribution of the simulated SRR element. (a) The device consists of an SRR and a high-impedance microstrip line with a masking layer; (b) the surface current distribution and intensity of the SRR element excited by a high-impedance microstrip line at resonant frequency.
device using SRR arrays is feasible [16–18]. Moreover, provided that the loss characteristics of the resonator itself can be further diminished, the design of a high-Q resonator in the quasi-optical region is plausible. The characteristics of the SRR element would allow it to be used as a resonator for sensing applications. For example, a nanoscaled-plasmonic SRRs-based biosensor has been used for DNA sensing at optical frequencies [19]. In addition, we previously identified changes in the resonant frequency due to biotin-streptavidin binding on the surface of the SRRs array at microwave frequencies [20]. Although the biosensing device showed a remarkable frequency shift upon biomolecular binding, we used a relatively high-concentration on the SRRs array and the procedure was somewhat complex and cumbersome due to the use of a liquid wall tank for surface biological processing of the pattern, thus presenting difficulties in testing numerous samples. In this study, our findings show that an antigen-antibody binding system can be effectively and rapidly detected by the SRR-based RF biosensor in a local high-impedance microstrip line system. 2. Materials and methods 2.1. Design, fabrication, and measurement of the split-ring resonator
We have designed a biosensing device for biomolecule detection (Fig. 1a). It consists of a microstrip line, which is a kind of planar transmission line for transmitting electromagnetic energy, with local high-impedance and an SRR element. The high-impedance microstrip line functions to effectively and intensively excite the resonator because the high current produced by the narrow line enhances the time-varying magnetic field component. The characteristic impedance of the line was about 80 and the other parts of the microstrip line (input and output port) were matched at 50 . With this topology, we expected that strong circular current flows could be produced on the surface of the inner and outer rings (Fig. 1b). The resonator element can be modeled as input impedance seen from the microstrip line using the following equation:
ZSRR = R + j ωL −
1 ωC
overall area of the microstrip line was 20 by 10 mm2 . The electric permittivity (εr ) and loss tangent (tan ı) of the dielectric layer were 9.7 and 0.0036, respectively, and its thickness was 0.76 mm. We performed a full-wave electromagnetic simulation based on the finite element method (FEM) to predict the resonant frequency and the surface current distribution of the resonator. From the resulting simulation, it is worth noting that the intensity of the current is highly strengthened at the outer ring of the resonator. The device was fabricated with well-defined printed circuit board (PCB) technology (Fig. 2a, right) [21]. To construct a biolayer on the surface of the resonator, the copper pattern was coated with a nickel (Ni) adhesion layer at a thickness of 3–5 m, followed by top-coating with a gold (Au) layer at a thickness of 0.05 m. To confine biomolecular binding to the surface of the resonator itself, the entire device, excluding the active sensing region and the edge contact pads used for measurement, was coated with a masking layer at a thickness of 10–20 m. The RF measurement system for the fabricated sample is shown in Fig. 2a (left). This system comprised an RF test fixture associated with a two-port vector network analyzer (VNA) system. From the VNA system, we could obtain the scattering parameters (S-parameters) of the biosensing device. The magnitude of the S-parameters is defined as
(1)
where R, L, and C are the resistance, inductance, and capacitance components, respectively and ω = 2f is the angular frequency. The width and length of the high-impedance line were 0.2 mm and 3.56 mm (including two tapered sections on both sides), respectively, and the width and length of the other two parts of the matching line were 0.78 mm and 16.44 mm, respectively. The
S11 (= S22 ) = 20 log10
V1− V1+
,
S21 (= S12 ) = 20 log10
V2−
V1+
(2)
where S11 and S21 are the reflection (ratio of the reflected voltage wave to the incident voltage wave: V1− /V1+ ) and transmission coefficients (ratio of the transmitted voltage wave to the incident voltage wave: V1− /V1+ ), respectively. The S21 parameter of the resonator was used as a discriminating signal for biomolecule detection due to its resonant frequency. Before biological application, we measured the resonant frequency and magnitude of the bare samples. The measured results had a lower frequency of 180 MHz and a higher magnitude of 6.48 dB relative to the simulated results (Fig. 2b). In addition, the simulated resonator had a Q-factor of 80 at the resonant frequency, while the measured Q-factor decreased to 50 due to various losses such as metal, radiation, and substrate loss. However, this discrepancy was not significant, since the biosensing procedure is based purely on relative frequency and magnitude shift. 2.2. Surface modification of the split-ring resonator In this work, PSA and cortisol as biomarkers for verifying the feasibility of our clinical biosensor were selected. In the sensor surface, cysteine (Cys) 3-mediated protein-G was first immobilized on a gold surface to allow binding to the Fc (constant fragment) region
28
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
Fig. 2. Schematic of the RF measurement system. (a) The sample fixture system associated with a network analyzer (left) and fabricated sample (right); SRR dimensions are w = 0.19 mm, s = 0.19 mm, d = 0.08 mm, g = 0.1 mm, and a = 1.63 mm; (b) the S21 resonance of simulated and measured SRR samples.
Fig. 3. Immobilization of anti-PSA and cortisol on the surface of the SRR device and binding experiments. (a) Experimental group with anti-PSA (or anti-cortisol); (b) control group without anti-PSA (or anti-cortisol). Fluorescent microscopic images of the surface of the SRR biosensor treated with PSA-Cy3 (c) without anti-PSA (or anti-cortisol) and (d) with anti-PSA (or anti-cortisol).
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
29
2.3. Observation of a specific interaction between anti-PSA and PSA using fluorescent dye
10 – 15 ± 2 100 – 20 ± 3 1 – 10 ± 2 10 – 20 ± 3 a
PBS
– ∼0 –
Materials on the sensor surface
Concentration (ng/ml) f1 (MHz) f2 (MHz)
The frequency shifts of the interactions of PSA and anti-PSA were investigated (Fig. 4). For the antibody immobilized on the surface of the sensor, there was a shift to the lower frequency region and the S21 resonance level changed slightly compared to the bare sensors (see footnote a in Table 1). When different PSA concentrations (100 ng/ml, 10 ng/ml, 1 ng/m, and 100 pg/ml) were applied, the resonant frequency underwent an additional shift to lower frequency regions with increased PSA concentration, from 30 ± 2 MHz (100 ng/ml) to 5 ± 1.5 MHz (100 pg/ml). The Cys3-protein-G molecular cross-linker through which antibody can be immobilized did not show a significant frequency shift (1–2 MHz; Table 1). Our biosensor could distinguish between anti-PSA binding and the bare surface of the sensor, reflecting the proper immobilization of
Table 1 Summary of experimental results.
3.2. Frequency shift measurements
The average value of resonant frequency for 10 bare samples is 10.48 ± 0.22 GHz.
100 – 30 ± 2 500 20 ± 5 –
Prior to the frequency shift measurements, fluorescent microscopic images were obtained to investigate the surface modification due to the presence of anti-PSA and the subsequent binding of Cy3-labeled PSA labeled. A very strong red fluorescence image was observed when fluorescently-labeled PSA was added to the antibody-coated biosensor (Fig. 3d). The absence of any fluorescence when the labeled PSA was added to the bare surface of the biosensor (i.e., no antibody coating) (Fig. 3c) indicated that the interaction of PSA with the antibody on the sensor surface was specific.
Antibody (PSA)
3.1. Optical detection of PSA binding to anti-PSA antibody immobilized on the sensor
1 × 106 1–2 ± 0.5 –
3. Results and discussion
Cys-3-mediated Protein G
Prostate specific antigen (PSA)
0.1 – 5 ± 1.5
Fluorescent staining was used to examine the specific interaction between anti-PSA antibody and PSA. PSA molecules were fluorescently labeled by dissolving fluorolinker Cy3 mono-reactive dye (100 ng/mL, Amersham Biosciences, USA) in 1 M sodium bicarbonate buffer (pH 9.3) and allowing it to react with the PSA (100 ng/ml, Abcam, UK) for 1 h at room temperature. The free dye was then separated from the labeled PSA using dye removal resin in fluorescent dye-removal columns (Thermo, USA). The absorbance and emission wavelengths of Cy3 are 550 nm and 570 nm, respectively. To verify that the fluorescence resulted from the specific binding of PSA to the antibody, the SRR sensor without anti-PSA was also treated with Cy3-PSA and then rinsed with PBS buffer. Both samples were then observed under a fluorescence microscope.
500 18 ± 4 –
Antibody (Cortisol)
Cortisol-BSA conjugate
1 – 5±2
0.1 – 3±1
of the antibody [22–24]. PSA and cortisol biomolecules were then introduced (Fig. 3). The biological process is in detail as follows: the gold surface of the sensor was treated with 10 g/ml Cys3-mediated protein-G in phosphate buffered saline (PBS) solution at pH 7.4 for approximately 1 h. Next, 500 ng/ml of anti-PSA solution was added to the sensor, followed by BSA (about 1 mg/ml) to prevent nonspecific surface binding. Finally, four different PSA and cortisol-BSA concentrations (100 ng/ml, 10 ng/ml, 1 ng/ml, and 100 pg/ml) were used in the assay (Fig. 3a). Binding experiments were conducted in PBS solution (pH 7.4) for about 90 min. For the control experiment (Fig. 3b), the sensor surface was treated with protein-G and BSA under the same conditions in the absence of anti-PSA solution. PSA was used at a concentration of 100 g/ml. The sensor surface was thoroughly rinsed with PBS solution between each step and then immediately measured using the RF measurement system.
30
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
a concentration of 100 pg/ml (Table 1). The minimum detectable level of our device is lower level than conventional SPR-based biosensor for PSA in sample volume of 130 m [25]. In this work, the resonant frequency shift for biomolecular binding can be easily predictable. According to quasi-static analysis [26], the resonant frequency can be predicted since the proposed biosensing scheme essentially behaves as an LC circuit, as follows: ω02 =
2 , rav LC
(3)
L = Lm + Li + Lo ,
Fig. 4. Frequency shift according to the buffer solution (PBS) and various concentrations of PSA.
eff
anti-PSA on the sensor surface, which generated a reasonable frequency shift. As a reference experiment, PBS solution without any biomolecules was applied to the sensor and after complete drying, no frequency shift was observed, indicating that the signal was generated from the specific biomolecular interactions. The molecular weight of cortisol is very small and we expected that cortisol would not produce a significant signal in our device. Thus, bovine serum albumin (BSA)-conjugated cortisol (cortisol-BSA) was used to amplify the signal. Indeed, when free cortisol was applied to the anti-cortisol antibody-coated surface of the sensor, a very small signal was observed (data not shown). The BSA without cortisol did not show any signal either, suggesting that the frequency shift upon addition of cortisol-BSA originated solely from the interaction of anti-cortisol antibody with cortisol. The frequency shifts of cortisol antigen-coupled BSA binding to its antibody is shown in Fig. 5. A similar shift to lower resonant frequency regions was observed, but the changes in frequency were smaller than those of the four different PSA concentrations. The variation in frequency shift and magnitude may result from uncertainties in the preparation of the biolayers. Because the samples were prepared by a biological process in which the biosensors were left to soak in static liquid PBS with biomolecules, the biolayer on the gold surface of the sensors may not have been uniformly coated with biomolecules. Based on the results, the sensing limitation of our device was estimated as
rav =
(ri + ro ) 2
(4)
where rav is the average radius between rings and L and C are the total inductance and capacitance of the structure, respectively. In Eq. (4), the total inductance consists of the mutual inductance (Lm ) per unit length between rings, the self inductance (Li ) of the inner ring and the self inductance (Lo ) of the outer ring. Similarly, the total capacitance is comprised of the mutual capacitance (Cm ) per unit length between rings, the split capacitance (Ci ) of the inner ring and the split capacitance (Co ) of the outer ring. Since the biosensing samples are immobilized with Cys3-modified protein-G coupled with PSA antibody and are then bound to the PSA, additional layers due to biomolecular binding can be modeled as two nano-sized parallel dielectric layers with a different permittivity on the gold surface. The total effective capacitance for the two layers can be expressed as C ∝ εr1
Fig. 5. Frequency shift according to the buffer solution (PBS) and various concentration of cortisol-coupled BSA.
C = Cm + Ci + Co ,
Apul1 dpul1
eff
+ εr2
Apul2 dpul2
.
(5)
In this equation, the first term is the effective capacitance produced by the immobilization layer of protein-G coupled to PSA antibody, while the second term is the effective capacitance proeff eff duced by the PSA binding layer, where εr1 and εr2 are the effective permittivities of the first and second dielectric layer, respectively. Consequentially, although the frequency shifts for the biomolecular binding resulted from a change in inductance components, the change in the capacitance component is predominantly due to the electrical permittivity of the biomolecule and the electric field distribution between the inner and outer rings. As indicated by Eq. (5), the overall effective capacitance component increases as biomolecular binding progresses and, therefore, the resonant frequency shifts toward the lower frequency region. 4. Conclusion We have clearly demonstrated the use of our planar SRRbased RF biosensor excited by a local high-impedance microstrip line as a label-free biosensor with a resonant-assisted transduction scheme. In terms of biomolecular detection, such as PSA and cortisol antigen-antibody binding systems, our device has a very simple, direct, and sensitive transduction scheme at microwave frequencies. In addition, it requires a small volume of biomolecular specimen, has a fast response time, is easy to operate and requires no labeling process. We therefore suggest that our biosensing approach is a good candidate for a RF/microwave and THz biosensor by using a more sophisticated microfabrication process. We believe that this seminal work may pave the way to realizing a label-free RF biosensing transducer for simple and direct detection of human specimens at the desired radiation frequencies. Acknowledgments This study was supported by grants from the National R&D Program for Cancer Control, Ministry of Health & Welfare (No.1120290), the National Research Foundation of Korea (NRF)
H.-J. Lee et al. / Sensors and Actuators B 169 (2012) 26–31
(No.2011-0016731) and the Korean Basic Science Institute (No. T3278B). References [1] J.H. Lee, K.S. Hwang, J. Park, K.H. Yoon, D.S. Yoon, T.S. Kim, Immunoassay of prostate-specific antigen (PSA) using resonant frequency shift of piezoelectric nanomechanical microcantilever, Biosens. Bioelectron. 20 (2005) 2157–2162. [2] K.S. Hwang, J.H. Lee, J. Park, D.S. Yoon, J.H. Park, T.S. Kim, In situ quantitative analysis of a prostate-specific antigen (PSA) using a nanomechanical PZT cantilever, Lab. Chip 4 (2004) 547–552. [3] W.D. Wilson, Analyzing biomolecular interactions, Science 295 (2002) 2103–2105. [4] G. Zheng, F. Patolsky, Y. Cui, W.U. Wang, C.M. Lieber, Multiplexed electrical detection of cancer markers with nanowire sensor arrays, Nat. Biotechnol. 23 (2005) 1294–1301. [5] R.B. Nadler, P.A. Humphrey, D.S. Smith, W.J. Catalona, T.L. Ratliff, Effect of inflammation and benign prostatic hyperplasia on elevated serum prostate specific antigen levels, J. Urol. 154 (1995) 407–413. [6] J.C. Zhou, M.H. Chuang, E.H. Lan, B. Dunn, P.L. Gillman, S.M. Smith, Immunoassays for cortisol using antibody-doped sol–gel silica, J. Mater. Chem. 14 (2004) 2311–2316. [7] S.S.C. Tai, M.J. Welch, Development and evaluation of a candidate reference method for the determination of total cortisol in human serum using isotope dilution liquid chromatography/mass spectrometry and liquid chromatography/tandem mass spectrometry, Anal. Chem. 76 (2004) 1008–1014. [8] R.C. Stevens, S.D. Soelberg, S. Near, C.E. Furlong, Detection of cortisol in saliva with a flow-filtered, portable surface plasmon resonance biosensor system, Anal. Chem. 80 (2008) 6747–6751. [9] D.R. Smith, Willie J. Padilla, D.C. Vier, S.C. Nemat-Nasser, S. Schultz, Composite medium with simultaneously negative permeability and permittivity, Phys. Rev. Lett. 84 (2000) 4184–4187. [10] R.A. Shelby, D.R. Smith, S. Schultz, Experimental verification of a negative index of refraction, Science 292 (2001) 77–79. [11] A.A. Houck, J.B. Brock, I.L. Chuang, Experimental observations of a left-handed material that obeys Snell’s law, Phys. Rev. Lett. 90 (2003) 137401. [12] S. Linden, C. Enkrich, M. Wegener, J. Zhou, T. Koschny, C.M. Soukoulis, Magnetic response of metamaterials at 100 terahertz, Science 306 (2004) 1351–1353. [13] H.O. Moser, B.D.F. Casse, O. Wilhelmi, B.T. Saw, Terahertz response of a microfabricated rod-split-ring-resonator electromagnetic metamaterial, Phys. Rev. Lett. 94 (2005) 063901. [14] H.-T. Chen, W.J. Padilla, J.M.O. Zide, A.C. Gossard, A.J. Taylor, R.D. Averitt, Active terahertz metamaterial devices, Nature 444 (2006) 597–600. [15] 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 (2007) 062511. [16] J. Garcia-Garcia, F. Martin, F. Falcone, J. Bonache, J.D. Baena, I. Gil, E. Amat, T. Lopetegi, M.A.G. Laso, J.A.M. Iturmendi, Microwave filters with improved stopband based on sub-wavelength resonators, IEEE Trans. Microwave Theory Tech. 53 (2005) 1997–2006. [17] J. Bonache, I. Gil, J. Garcia-Garcia, F. Martin, Novel microstrip bandpass filters based on complementary split-ring resonators, IEEE Trans. Microwave Theory Tech. 54 (2006) 265–271. [18] I. Bulu, H. Caglayan, K. Aydin, E. Ozbay, Compact size highly directive antennas based on the SRR metamaterial medium, New J. Phys. 7 (2005) 223. [19] A.W. Clark, A. Glidle, D.R.S. Cumming, J.M. Cooper, Plasmonic split-ring resonators as dichroic nanophotonic DNA biosensors, J. Am. Chem. Soc. 131 (2009) 17615–17619. [20] H.J. Lee, J.G. Yook, Biosensing using split-ring resonators at microwave regime, Appl. Phys. Lett. 92 (2008) 254103. [21] H.J. Lee, H.S. Lee, K.H. Yoo, J.G. Yook, DNA sensing using split-ring resonator alone at microwave regime, J. Appl. Phys. 108 (2010) 014908. [22] W. Lee, B.K. Oh, W.H. Lee, J.W. Choi, Immobilization of antibody fragment for immunosensor application based on surface plasmon resonance, Colloids Surf. B: Biointerfaces 40 (2005) 143–148. [23] Z. Shen, G.A. Stryker, R.L. Mernaugh, L. Yu, H. Yan, X. Zeng, Single-chain fragment variable antibody piezoimmunosensors, Anal. Chem. 77 (2005) 797–805.
31
[24] J.M. Lee, H.K. Park, Y. Jung, J.K. Kim, S.O. Jung, B.H. Chung, Direct immobilization of protein G variants with various numbers of cysteine residues on a gold surface, Anal. Chem. 79 (2007) 2680–2687. [25] D.A. Healy, C.J. Hayes, P. Leonard, L. McKenna, R. O’Kennedy, Biosensor developments: application to prostate-specific antigen detection, Trends Biotechnol. 25 (2007) 125–131. [26] R. Marqués, F. Medina, R. Rafii-El-Idrissi, Role of bianisotropy in negative permeability and left-handed metamaterials, Phys. Rev. B 65 (2002) 144440.
Biographies Hee-Jo Lee received the M.Sc. (Physics and Applied Physics) and Ph.D. (Electrical and Electronic Engineering) degrees from Yonsei University, in 2004 and 2010, respectively, Seoul, South Korea. During his Ph.D. degree and first Post Doctoral Researcher at BK21 Institute of Information and Technology in Yonsei University, he was dedicated to the metamaterials for biosensing, carbon nanotube-based RF nanobiosensors, and electromagnetic field theory. Since November 2010, he joined the Graphene Research Institute (GRI), Sejong University, Seoul, Korea, where he is now a Post Doctoral Research Fellow. His research interests include RF circuit modeling and characterization of graphene, graphene-based RF field-effect transistors (FETs)/flexible RF devices/RF nanobiosensors, and computational electromagnetics. Jung-Hyun Lee received the B.Sc. degree (Bio & Environmental Technology) from Seoul Women’s University, South Korea in 2009. She is currently a MS degree student at the National Core Research Center for Nanomedical Technology, Yonsei University. Her research interests include the development of analytical system for biomedical and environmental monitoring. Hui-Sung Moon received the B.Sc. and Ph.D. (Mechanical Engineering) degrees from Yonsei University, South Korea in 2005 and 2011, respectively. During his Ph.D. degree, he pursued his research on microfluidic cell manipulation. He is currently a research staff in Samsung Advanced Institute of Technology (SAIT). Ik-Soon Jang received the B.Sc. degrees from Sungkyunkwan University in 1995, and M.Sc. and the Ph.D. (Biochemistry and Molecular Biology) degree from the Seoul National University, College of Medicine in 1997 and 2001. He is currently a senior scientist at the Korea Basic Science Institute (KBSI), and his research interests in Aging Mechanism by using proteomics and development of membrane protein association. Jong-Soon Choi received the B.Sc. and M.Sc. (Biology) degrees from Yonsei University in 1986 and 1990, respectively, and the Ph.D. (Microbiology) degree from the Korea Advanced Institute of Science and Technology (KAIST) in 1999. He is the principal researcher in the Division of Life Science of Korea Basic Science Institute. Also, he is the professor in the Graduate School of Analytical Science and Technology of Chungnam National University. His main research is oriented in the analytical forensic science using functional proteomics. Jong-Gwan Yook received the B.Sc. and M.Sc. (Electronic Engineering) degrees from the Yonsei University in 1987 and 1989, respectively, and the Ph.D. (Electrical and Computer Engineering) degree from the University of Michigan, Ann Arbor, USA in 1996. He is currently a professor at the School of Electrical and Electronic Engineering, Yonsei University, and his research interests include the areas of theoretical/numerical electromagnetic modeling and characterization of microwave/millimeter-wave circuits and components, design of RF integrated circuits (RFICs) and monolithic microwave integrated-circuits (MMICs), and analysis and optimization of high-frequency high-speed interconnects, including RF microelectromechanical system (MEMS), based on frequency as well as time-domain full-wave methods. His research team has also been engaged in the development of biosensors, such as CNTs RF biosensor for nanometer-sized antigen-antibody detection as well as remote wireless vital signal monitoring sensors. Hyo-Il Jung received the B.Sc. and M.Sc. (Biotechnology) degrees from the Korea Advanced Institute of Science and Technology (KAIST) in 1993 and 1995, respectively, and the Ph.D. (Physical Biochemistry) degree from the University of Cambridge, United Kingdom. He is currently an associate professor at the School of Mechanical Engineering, Yonsei University, and his research interests center on microfluidic biosensors and cancer diagnosis.