A current detection surface plasmon resonance (SPR) sensor with an Au grating on an n-Si wafer was proposed. SPR excitation light is illuminated from the backside of the device and diffracted by the grating. Since the diffraction provides matching conditions, SPR can be coupled to the Au/analyte interface. Since the coupled SPR excites free electrons on the Au surface, the SPR can be detected as a current signal by a Schottky barrier diode formed on the Au/n-Si interface. The obtained angular current spectrum showed clear agreement with SPR coupling theory, thereby confirming that the sample on the Au surface can be electrically detected using the proposed sensor.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Surface plasmon resonance (SPR) chemical sensors provide highly sensitive and label-free chemical detection with a simple optical system. Miniaturization of the sensor system has therefore been a pursuit in field applications such as gas sensors [1–3] and biosensors [4,5]. A grating coupling-type SPR sensor is particularly suitable for miniaturization because it does not require a bulky prism to excite the SPR and enables changing the SPR coupling angle with a grating pitch. A grating coupling-type SPR sensor conventionally introduces the excitation light to the metal grating surface through the sample analyte. Since this optical configuration provides unwanted interaction between the excitation light and sample analyte, SPR excitation by backside illumination of the grating has recently been studied [6–8]. In these studies, the sensors are composed of a flat Au film and a grating buried beneath the Au film; the two criteria of optical design freedom and noninteraction between the excitation light and analyte are satisfied. This sensor configuration, however, still requires an optical system to detect SPR, which is recognized as an obstacle to miniaturizing SPR sensors . Removal of optical systems from backside illumination-type sensors will therefore improve the practicality of SPR sensors.
In this paper, we propose an electrical detection SPR sensor with backside illumination. SPR is electrically detected by signal transduction at a Schottky barrier formed on an interface of a metal film and a semiconductor substrate [9–11]; since a reflection optical system is no longer required [12–14], SPR electrical detection has recently been attracting attention for further miniaturizing SPR sensors. In this study, we applied backside illumination to an electrical detection SPR sensor with a grating. In most backside illumination-type SPR sensors, dielectric gratings are formed under a flat metal film. Although in a theoretical model in , a several-hundred-nanometer-thick Si film with a nanometer-depth-trench is employed as a contacting material with a flat metal film, such trench fabrication on a very thin Si film is still very difficult. Here, we estimated that the metal film surface of the SPR sensor should not always be flat if the surface roughness amplitude is well below the wavelength . We therefore formed a Au thin film on a Si grating with a height of 100 nm. Since the grating height is low enough to support SPR, the excitation light is diffracted at the backside of the Au film and is able to excite SPR at the analyte/Au interface. In addition, since the Si grating is opaque in the visible range, we employed near-infrared light as the excitation light. Since the energy of near-infrared light is smaller than the energy bandgap of Si (~1.1 eV), the incident light propagating through the Si substrate does not attenuate and is able to reach the Au film from the backside. When the SPR is excited, electrons in Au receive photon energy and are excited. The Au/n-Si Schottky barrier height is ~0.8 eV, so the electrons excited by the near-infrared light can overcome the barrier to generate a photocurrent. The Si therefore behaves as both a light-guiding medium and an electrical SPR detector.
In this paper, we present the fabrication, electrical characterization, and SPR response measurement of the device to confirm that the SPR behavior is altered depending on the permittivity of the analyte in order to check whether the proposed device is applicable to SPR sensing.
2. SPR detection
2.1. SPR coupling and electrical detection principle
The near-infrared excitation light of the SPR is incident to the Au/n-Si grating interface through the n-Si substrate in our proposed device (Fig. 1(a)). The incident plane of the SPR optics was taken to be perpendicular to the running direction of the grating. The polarization of the excitation light should be transverse magnetic (TM). Since the n-Si substrate is transparent in the near-infrared region, the light does not attenuate during propagation through the n-Si substrate. When the near-infrared excitation light irradiates the Au film from the backside with an angle of incidence θAu, it is diffracted by the grating. When the diffracted light reaches the analyte/Au interface (i.e., the upper surface of the grating), SPR is excited if the matching condition expressed by the following equation is satisfied ,Fig. 1(b)) and flow toward the conduction band of the n-Si substrate, the SPR is detected as an electric signal such as a photocurrent Iph.
2.2. Device configuration
The substrate of the device was a double-polished 750-μm-thick n-Si wafer whose resistivity was 16 Ωˑcm. One-dimensional gratings with a 3.4-μm pitch (denoted as a in the figure) and a height of 100 nm were formed on the front surface of the n-Si substrate. The surface of the grating was coated with a 20-nm-thick Au film; the upper surface of the grating functions as both a measurement site of the analyte and an anode electrode. On the front side of the n-Si surface, an Al film was also formed next to the Au grating to serve as a cathode electrode. Although high doping of impurity is usually necessary to form an electrical contact between the Al and the n-Si, it was experimentally confirmed beforehand that the n-Si with this doping level could provide an ohmic contact. Only a Au/n-Si interface therefore worked effective as a Schottky barrier in this device. The fabricated device chip mounted on a printed circuit board (PCB) is shown in Fig. 1(c). The sensor chip was 7.0 mm × 7.0 mm, the grating area was formed to be 6.6 mm × 4.9 mm, and the Al electrode area was 6.6 mm × 1.6 mm aside from the grating area.
3. Experimental setups
The electrical characteristics concerning the SPR detection capability by backside illumination were experimentally verified. First, the I-V characteristics of the device shown on the left side of Fig. 2(a) were obtained using a commercial source meter (Keithley 2614B, USA). The Schottky barrier height ΦB was then estimated to be 0.77 eV by parameter fitting for a forward current (Fig. 2(b)) according to the method in . Since the corresponding cutoff wavelength is 1605 nm, it was confirmed that the SPR excited by the near-infrared light could be measured with the proposed device.
Second, for the SPR detection experiment, the device chip was mounted on a PCB, at the center of which a hole was formed to introduce the excitation light from the backside of the device. In addition, a small cavity was constructed over a sensing Au grating area such that liquid analyte was maintained there. The cavity was composed of a silicone elastomer sidewall capped with a cover glass bonded by UV resin. The photocurrent Iph accompanying the SPR was converted to voltage by an IV conversion amplifier. The SPR device chip and the amplifier were installed into a shield box for noise elimination, which had an opening for the excitation light irradiation. In the experiment, no bias voltage was applied between anode and cathode electrodes of the SPR device chip.
The device was mounted on a rotational stage such that the angle of incidence θAu can be changed. The monochromatic TM polarized excitation light was emitted from a wavelength tunable laser (SC-450, Fianium, GBR), and the wavelength of the excitation light was changed from 1200 to 1500 nm. The intensities of the excitation light for different wavelengths were measured by a power meter (S122C, Thorlabs, USA) before the experiment. The sensor output was monitored by a source meter (6242, ADCMT, Japan).
4. Experimental results
4.1. Responses for front-side illumination
Before the experiments regarding the backside illumination, photocurrents accompanying SPR were measured using a conventional front-side illumination configuration usually used in the grating coupling method. In this experiment, we used a device with device configurations that were slightly different from those of the device proposed in the fabrication section: the grating pitch was 3.46 μm, the Au thickness was 100 nm, and the other conditions were the same. The purpose of this experiment was to investigate whether the fabricated device correctly worked as an SPR coupler and was able to detect the SPR as an electrical signal. The experiments were carried out in an air environment, so the air was regarded as an analyte (Fig. 3(a)). The excitation wavelength was scanned from 1200 to 1500 nm with a wavelength interval of 10 nm. The stage was rotated to change the angle of incidence θAu from −20 to 20° with an angular resolution of 0.1°.
The photocurrent curves obtained for different wavelengths are plotted in Fig. 3(b). The vertical axis of the plots is presented as responsivity, which is defined as the photocurrent Iph divided by the excitation light intensity at each wavelength. The uppermost photocurrent curve corresponds to the experimental results with excitation light λ = 1200 nm. The responsivity level monotonically decreased with an increase in the wavelength of the excitation light with an interval of 10 nm. Several distinctive peaks were found in the whole responsivity curves. For example, in the photocurrent curve for λ = 1200 nm, four distinctive peaks were found at angles of incidence θAu = −18.8°, −0.6°, 0.4°, 18.4°. The peak positions continuously shifted as the excitation wavelength increased; the angular positions of the peaks at θAu = 0.4° and 18.4° shifted toward larger and smaller angular positions, respectively, as the excitation wavelength increased, crossed at λ = 1400 nm, and diverged from each other at λ > 1400 nm. In addition, because the optical configuration is symmetric about the angle of incidence θAu = 0°, the photocurrent curves are line symmetric about θAu = 0°.
Since these peaks accompany SPR, these peak angles are denoted as θSPR in the following text. Multiple SPR peaks for excitation light of a single wavelength correspond to different diffraction orders in the equation above. The calculation using the equation supported that these observed SPR peaks were generated by ± 2nd and ± 3rd order diffraction. The peak angular positions θSPR for the different wavelengths were extracted, and theoretical values are also plotted in Fig. 3(c). In the calculation, the permittivity of air was 1.0, and the dispersive permittivity values of Au were taken from . Since the dispersion of Au is gentle in the wavelength range used, the theoretical SPR positions exhibit linear lines in Fig. 3(c). The experimental SPR peak positions showed significant consistency with the theoretical ones. The experiments with front-side SPR excitation light illumination revealed that the fabricated device can generate SPR and that its behavior completely followed SPR coupling theory. The photocurrent detection using the Schottky barrier was also confirmed to function as intentioned.
4.2. Responses for backside illumination for air condition
As a next step, we performed experiments using backside illumination and verified whether SPR photocurrent responses change with the species of the analyte, in this case, air and water. First, the cavity over the measurement area was kept as atmosphere such that the air became the analyte (Fig. 4(a)). The excitation light from the backside of the device refracted at the air/n-Si interface, and it reached the Au/n-Si interface with an angle of incidence θAu. The angle of incidence θAu varied from −8° to 8°. The photocurrent curves for the excitation light with λ = 1200, 1300, 1400 and 1500 nm are plotted in Fig. 4(b). The photocurrent curves were line symmetric around θAu = 0°, as in the previous experiments. Compared with those in the previous experiments, the photocurrent peaks were not clear. There were still distinctive peaks on the curves. For example, four peaks at θAu = −4.6°, −2.5°, 2.5°, and 4.7° were observed for the photocurrent curve with an excitation light of λ = 1500 nm. In addition, the peak positions θAu systematically shifted as the wavelength of the excitation light changed. The peak positions θAu were extracted from the experimental data in Fig. 4(c). Extraction procedures are below. The photocurrents were regarded as functions of the angle of incidence θ, and were differentiated with respect to θ. The SPR peak points were extracted by finding a zero crossing angle where the differential values changed from positive to negative. However, since the noise was relatively large in the measured data, moving average for nine neighboring data points was performed before the differentiation. It should also be noted that the data points which are likely to be due to noise were considered to be misdetection, and excluded from the SPR peak angle plots. In addition, the theoretical peak positions θAu calculated by the equation above are also plotted in Fig. 4(c).
In the calculation, the dispersion of n-Si was considered and was taken from . Although peaks around 0° for the wavelength ranging from 1250 to 1500 nm and peaks around ± 7° for the wavelength ranging from 1300 to 1500 nm cannot be attributed to SPR coupling, other peaks showed high consistency with the calculated results, ± 2nd, ± 3rd, and ± 4th order diffraction SPR couplings. In both the experimental and theoretical data plots, the peak lines crossed around θAu = ± 3.2°. The obtained results confirmed that SPR excitation could be performed using backside illumination for air as the analyte.
4.3. Responses for backside illumination for water condition
In the final experiment, we changed the analyte from air to water by filling the cavity with water and measured the SPR photocurrent response (Fig. 5(a)); the obtained results are shown in Fig. 5(b). The dispersive permittivity values of water at a temperature of 22°C were taken from . Two photocurrent peaks were found for excitation light with λ = 1200 nm at θAu = −4.7° and 4.8°, which were attributed to SPR. Compared with the photocurrent curves for the excitation light with λ = 1300 nm, 1400 nm and 1500 nm, the peak positions θSPR systematically shifted. Notably, there are also some gentle increases in the photocurrent curves around θAu = ± 4° for λ = 1500 nm. These increases can be attributed to the SPR excited at the Au/n-Si interface instead of the Au/analyte interface based on theoretical considerations. Since the resonant point of SPR occurring on the Au/n-Si interface is not altered by the analyte permittivity change, SPR measurements should be conducted around the angular positions where no Au/n-Si SPR exists.
The obtained experimental peak positions θSPR considered to have been excited on an Au/analyte interface are plotted in Fig. 5(c) along with theoretical plots. The extraction of the peak positions θSPR followed the procedure in the previous section. Theoretical calculation indicated that the SPR couplings on a Au/analyte interface corresponding to ± 2nd, ± 3rd, ± 4th and ± 5th order diffraction were generated on the device. At almost all wavelengths, SPR peaks corresponding to the ± 3rd order diffraction were experimentally observed. The SPR peaks corresponding to ± 4th order diffraction were sometimes difficult to observe, so there were several wavelengths where the SPR peaks were missed. The angular shift behavior of the SPR peaks with ± 4th order diffraction, however, approximately showed consistency with the theoretical angular behavior. On the other hand, the SPR peak with ± 2nd and ± 5th order diffraction was not observed, which can be attributed to the weak SPR coupling efficiency for the diffraction order. Since the obtained SPR photocurrent behaviors are basically dependent on the permittivity of the analyte on the Au grating surface, in principle, it is possible to employ the proposed device for chemical SPR sensing by improving the SP resonant performance of the proposed device for aqueous specimen. We calculated the sensitivity of the sensor by the angular shift amplitude per refractive index unit (RIU) change using the SPR responses for the diffraction order m = −3 at the wavelength of 1500 nm. The SPR angles can be found at 4.5° for air, and −0.5° for water. Since the refractive indices of air and water for the wavelength are 1.00 and 1.32, respectively, the angular shift sensitivity is calculated as 15.6°/RIU based on the procedure in .
In this paper, we verified that SPR excitation and electrical sensing can be performed with backside illumination using an electrically detectable SPR device equipped with a Au grating. This optical configuration can eliminate unwanted interactions between the excitation light and the analyte because the excitation light does not need to transmit through the analyte solution before reaching the Au grating. Many varieties of analyte liquid, including nontransparent solutions such as colloidal or opaque liquids, can therefore be applied for SPR sensing. Since the SPR coupling efficiency can be tuned by the design of the grating, such as the grating pitch and height, SPR measurements can be performed with clearer SPR curves. In addition, the grating design determines SPR coupling angles, and an optical arrangement, such as the normal incidence of the excitation light, will be possible. Then, the use of a planar light source such as a vertical-cavity surface-emitting laser (VCSEL) for the excitation light will allow elimination of the optical system and provide a one-chip SPR sensor. The proposed SPR measurement configuration can enable drastic downsizing of SPR sensors.
New Energy and Industrial Technology Development Organization (NEDO), Japan.
The EB direct writings were carried out using the EB lithography apparatus of the VLSI Design and Education Center (VDEC) of the University of Tokyo. The microfabrications were performed in a clean room of the Division of Advanced Research Facilities of the Coordinated Center for UEC Research Facilities of the University of Electro-Communications, Tokyo, Japan.
1. C. Nylander, B. Liedberg, and T. Lind, “Gas detection by means of surface plasmon resonance,” Sens. Actuators 3, 79–88 (1982). [CrossRef]
2. J. M. Bingham, J. N. Anker, L. E. Kreno, and R. P. Van Duyne, “Gas sensing with high-resolution localized surface plasmon resonance spectroscopy,” J. Am. Chem. Soc. 132(49), 17358–17359 (2010). [CrossRef] [PubMed]
3. B. Liedberg, C. Nylander, and I. Lunström, “Surface plasmon resonance for gas detection and biosensing,” Sens. Actuators 4, 299–304 (1983). [CrossRef]
5. E. Wijaya, C. Lenaerts, S. Maricot, J. Hastanin, S. Habraken, J.-P. Vilcot, R. Boukherroub, and S. Szunerits, “Surface plasmon resonance-based biosensors: From the development of different SPR structures to novel surface functionalization strategies,” Curr. Opin. Solid State Mater. Sci. 15(5), 208–224 (2011). [CrossRef]
6. N. C. Lindquist, T. W. Johnson, J. Jose, L. M. Otto, and S. H. Oh, “Ultrasmooth metallic films with buried nanostructures for backside reflection-mode plasmonic biosensing,” Ann. Phys. 524(11), 687–696 (2012). [CrossRef] [PubMed]
7. S. Pi, X. Zeng, N. Zhang, D. Ji, B. Chen, H. Song, A. Cheney, Y. Xu, S. Jiang, D. Sun, Y. Song, and Q. Gan, “Dielectric-Grating-Coupled Surface Plasmon Resonance From the Back Side of the Metal Film for Ultrasensitive Sensing,” IEEE Photonics J. 8, 4800207 (2016).
8. W.-K. Kuo, J. Tongpakpanang, P.-H. Kuo, and S.-F. Kuo, “Implementation and phase detection of dielectric-grating-coupled surface plasmon resonance sensor for backside incident light,” Opt. Express 27(4), 3867–3872 (2019). [CrossRef] [PubMed]
9. C. Clavero, “Plasmon-induced hot-electron generation at nanoparticle/metal-oxide interfaces for photovoltaic and photocatalytic devices,” Nat. Photonics 8(2), 95–103 (2014). [CrossRef]
10. A. Sobhani, M. W. Knight, Y. Wang, B. Zheng, N. S. King, L. V. Brown, Z. Fang, P. Nordlander, and N. J. Halas, “Narrowband photodetection in the near-infrared with a plasmon-induced hot electron device,” Nat. Commun. 4(1), 1643 (2013). [CrossRef] [PubMed]
11. W. Chen, T. Kan, Y. Ajiki, K. Matsumoto, and I. Shimoyama, “NIR spectrometer using a Schottky photodetector enhanced by grating-based SPR,” Opt. Express 24(22), 25797–25804 (2016). [CrossRef] [PubMed]
12. B. Park, S. H. Yun, C. Y. Cho, Y. C. Kim, J. C. Shin, H. G. Jeon, Y. H. Huh, I. Hwang, K. Y. Baik, Y. I. Lee, H. S. Uhm, G. S. Cho, and E. H. Choi, “Surface plasmon excitation in semitransparent inverted polymer photovoltaic devices and their applications as label-free optical sensors,” Light Sci. Appl. 3(12), e222 (2014). [CrossRef]
13. T. Tsukagoshi, Y. Kuroda, K. Noda, N. Binh-Khiem, T. Kan, and I. Shimoyama, “Compact surface plasmon resonance system with Au/Si schottky barrier,” Sensors (Basel) 18(2), 399 (2018). [CrossRef] [PubMed]
14. Y. Ajiki, T. Kan, K. Matsumoto, and I. Shimoyama, “Electrically detectable surface plasmon resonance sensor by combining a gold grating and a silicon photodiode,” Appl. Phys. Express 11(2), 022001 (2018). [CrossRef]
15. H. Raether, Surface Plasmons on Smooth and Rough Surfaces and on Gratings (Springer, 1988).
16. S. K. Cheung and N. W. Cheung, “Extraction of Schottky diode parameters from forward current‐voltage characteristics,” Appl. Phys. Lett. 49(2), 85–87 (1986). [CrossRef]
17. A. D. Rakić, A. B. Djurišić, J. M. Elazar, and M. L. Majewski, “Optical properties of metallic films for vertical-cavity optoelectronic devices,” Appl. Opt. 37(22), 5271–5283 (1998). [CrossRef] [PubMed]
18. R. Khanna, Handbook of Optical Constants of Solids (Academic Press, 1997).
19. P. Schiebener, J. Straub, J. M. H. Levelt Sengers, and J. S. Gallagher, “Refractive index of water and steam as function of wavelength, temperature and density,” J. Phys. Chem. Ref. Data 19(3), 677–717 (1990). [CrossRef]
20. J. Homola, I. Koudela, and S. S. Yee, “Surface plasmon resonance sensors based on diffraction gratings and prism couplers: sensitivity comparison,” Sens. Actuators B Chem. 54(1-2), 16–24 (1999). [CrossRef]