Open Access
Add to BibSonomy
Abstract
Surface plasmon resonance (SPR) sensors are powerful tools for optical sensing of refractive index (RI) and bio-molecules due to their high sensitivity. In this article, we demonstrate a beam-angle-scanning SPR system using a combined galvanometer mirror and relay lens optics. Use of a photodetector in the galvanometer mirror scanning of the incident beam angle enables both high precision and rapid data acquisition. RI resolution of 2.306×10−5 refractive index unit (RIU) and RI accuracy of 8.984×10−5 RIU were achieved at a data acquisition rate of 100 Hz. Furthermore, we performed real-time monitoring of the avidin-biotin antigen-antibody reaction. The results show the high potential of this beam-angle-scanning SPR system.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Surface plasmon resonance (SPR) [1,2] is an optical technique that measures the change of refractive index (RI) near a metal surface via a compression wave based on the collective vibration of free electrons localized on the metal surface when the light is incident on a glass prism coated with a metal thin film under the condition of total internal reflection. SPR sensors are widely used as biosensors by surface modification of a molecular recognition layer on the metal surface [3,4], not to mention RI sensors [5,6]. A spectral dip caused by SPR, known as the SPR dip, appears in the angular spectrum or the wavelength spectrum of the reflected light, and therefore SPR sensors are categorized as angle-SPR mode [5–9] and wavelength-SPR mode [10,11]. The angle-SPR mode provides a high-resolution angular spectrum of the SPR dip with a simple configuration including a monochromatic light source and an angle-scanning prism. On the other hand, the wavelength-SPR mode enables real-time acquisition of the optical spectrum of the SPR dip by a combination of a broadband light source and a multi-channel spectrometer. Sensitivity of the angle-SPR mode and/or the wavelength-SPR mode were enhanced by use of whether nanomaterials [12] or topological insulator [13]. Furthermore, a combination of the angle-SPR mode with the wavelength-SPR mode allows free sensor surface selection for optimum material-sensing and biosensing [14]. From the cost-effective point of view, the angle-SPR mode is more widely used for RI sensing and biosensing than the wavelength-SPR mode.
The angle-SPR mode is further categorized into two types: the prism-scanning angle-SPR mode [6–8] and the multi-channel angle-SPR mode [5,9]. In the prism-scanning angle-SPR mode, mechanical angle scanning of the prism provides an angular SPR spectrum with high resolution; however, such mechanical scanning results in a long acquisition time. The real-time data acquisition capability can be given by measuring the intensity of the reflected light at a fixed angle of the prism because the angle shift in the linear slope of the SPR dip is converted into the change of optical intensity at the fixed prism angle. However, the angular spectrum of SPR dip is often shifted along both axes of the angle and the intensity due to the temperature drift. This results in the limitation of RI sensing performance. In the multi-channel angle-SPR mode, a line-focused beam of monochromatic light is incident on the prism as a collimated beam with various incident angles; then the reflected light from the metal thin film is imaged onto a camera by a lens. The obtained camera image corresponds to an angular spectrograph of the SPR dip along the lined-focused beam, and the angular SPR spectrum can be extracted from the camera image without the need for mechanical scanning of the prism. Therefore, the multi-channel angle-SPR mode benefits from rapid data acquisition at the camera frame rate. However, its sensing performance is hampered by the following three factors: (1) the number of camera pixels determines the angular resolution of the angle SPR spectrum, (2) the brightness resolution of the camera limits the dynamic range of the light intensity, and (3) the uneven sensitivity of the camera pixels leads to background noise on the angle SPR spectrum. In this way, existing angle-SPR modes have both advantages and disadvantages.
If the rapid data acquisition, high resolution, wide dynamic range, and reduced background noise can be simultaneously achieved in the angle-SPR mode, the performance of RI sensing and biosensing will be further enhanced. Unfortunately, it is challenging to further improve the camera performance in the multi-channel angle-SPR mode. One promising approach for the enhanced performance is to introduce rapid angle scanning of the incident beam as an alternative means to the prism mechanical scanning in the angle-SPR mode. Recently, a digital micro-mirror device (DMD) was used for angle scanning of the incident beam [15]; however, the data acquisition time of SPR spectrum was remained 450 ms when 45 angle points were sampled in total. In the study described in this article, we demonstrated angle-SPR mode by rapid, fine angle scanning of an incident laser beam with a galvanometer mirror and relay lenses; namely, a beam-scanning angle-SPR mode. This configuration enabled us to combine the merits of both prism-scanning angle-SPR and multi-channel angle-SPR, in other words, high precision and rapid data acquisition.
2. Experimental setup
We used the Kretschmann configuration for the beam-scanning angle-SPR mode. Figure 1 shows a schematic diagram of the experimental setup. A He-Ne laser (OSK-6328-5P, Thorlabs Inc., wavelength = 632.8 nm, optical power = 5 mW, linear polarization, beam diameter = 0.8 mm) was used as a monochromatic light source. The diameter of the output light was decreased to 0.4 mm by a pair of lenses (L1, focal length = 100 mm, diameter = 30 mm; L2, focal length = 50 mm, diameter = 30 mm); then, its polarization was set to a linear p-polarization by a Glan-Taylor polarizer (P, GT5-A, Thorlabs Inc., wavelength = 350 ∼ 700 nm, clear aperture = 5 mm, extinction ratio = 100,000:1). The resulting laser beam was mechanically scanned within an optical scan angle range of ±3° at 100 Hz at maximum by a combination of a single-axis galvanometer mirror (GM, GVSM001-JP/M, Thorlabs Inc., optical scan angle range = ±20°, resolution = 0.0008°, bandwidth of triangular wave = 175 Hz) with a pair of relay lenses (RL1, RL2, focal length = 100 mm, diameter = 40 mm). We formed a gold (Au) thin film (thickness = 50 nm) on a glass right-angle prism (RPB-30-2L, Thorlabs Inc., glass material = BK7, length = 30 mm, uncoated) by using a chromium (Cr) thin film (thickness < 5 nm) as an adhesive layer. This optical configuration enabled us to rapidly scan the collimated beam within the selected range of incident angles at a fixed position of the prism/gold interface. The reflected light passed through another pair of relay lenses (RL3, RL4, focal length = 100 mm, diameter = 40 mm), and then was incident on a photodetector (PD, PDA36A-EC, Thorlabs Inc., wavelength = 350–1100 nm, bandwidth = DC ∼10 MHz, active area = 3.6 mm by 3.6 mm). The output voltage signal from the photodetector and the driving voltage signal for the GM from a waveform generator (WG) were acquired by a data acquisition board (DAQ, USB-6361, National Instruments Corp., ADC resolution = 16 bit, maximum sample rate = 2 MSample/s). The angular resolution of 0.0008° in the galvanometer mirror is corresponding to that of 0.0005° in the prism considering the refraction of the laser beam at the incidence of the prism. On the other hand, the angular sampling interval was determined by dividing the whole range of beam scanning angle in the prism (= 6° in air sample and 3.2° in water solution sample) by the number of angle sampling points (= 10,000 at maximum). Benefitting from dual configurations of relay lenses optics in the incident light and the reflected light, the beam spot (diameter = 0.4 mm) on the sample (5 mm circle) and the detector (3.6 mm square) is fixed during the beam-angle scanning.
Fig. 1. Experimental setup. Laser, He-Ne laser; L1 and L2, lenses; P, Glan-Taylor polarizer; GM, single-axis galvanometer mirror; RL1, RL2, RL3, and RL4, relay lenses; PD, photodetector; DAQ, data acquisition board; WG, waveform generator.
3. Results
3.1 Basic performance
We evaluated the basic performance of the beam-scanning angle-SPR system when no samples were placed on the gold thin film, corresponding to the case where air was regarded as a sample with RI of 1 refractive index unit (RIU). From theoretical calculation of the BK7/Cr/Au/air layer model [16], the SPR dip will appear around an incident angle of 43.8°, as shown by a black line in Fig. 2. To observe the SPR dip of air, we set the range of incident angles to be 40.6°–46.6°. Red plots in Fig. 2 show experimental data measured (scanning range of incident angle = 6°, angle sampling points =250, angular sampling interval = 0.024°). The experimental data was in good agreement with the theoretical curve around the center of the SPR dip. There is a small difference between them at both ends of the SPR dip. We consider that the reason for this difference is due to non-uniform thickness of the Cr and/or Au layer.
Fig. 2. Comparison of SPR dip spectra between experimental data and theoretical curve. A sample is air (RI = 1 RIU).
We investigated the frequency response of the present system when the incident angle was rapidly scanned. Figure 3 compares the experimental data of the SPR spectrum when the scanning rate was set to be 1 Hz, 10 Hz, and 100 Hz (scanning range of incident angle = 6°, angle sampling points =250, angular sampling interval = 0.024°). All spectra were in good agreement with each other around the center of the SPR dip. The small difference among them may imply a difference of the frequency response in the GM. Importantly, a high signal-to-noise ratio and a good angular resolution was achieved in the SPR spectrum even though the scanning rate was set to be 100 Hz.
Fig. 3. Angular spectra of SPR dip at a data acquisition rate of 1 Hz, 10 Hz, and 100 Hz when a sample is air (RI = 1 RIU).
3.2 RI sensing of ethanol/water samples
Mixtures of ethanol and pure water were used as standard samples for RI sensing. The sample RI was adjusted by changing the mixing ratio between ethanol and water. Unfortunately, commercialized RI instruments such as Abbe's refractometer has insufficient performance for high-precision RI sensing below 10−5. Therefore, we simply calculated the sample RI from the water RI (= 1.3317) [17] and the ethanol RI (= 1.3604) [18] with a volume ratio between them. The relationship between the ethanol concentration EC (unit: EtOH%) and the sample RI (unit: RIU) is given by
Correctness of this equation was confirmed by previous studies of high-precision RI sensing with optical frequency comb [19]. We performed RI sensing of ethanol/water samples with different mixing ratios (= 0–10 EtOH%, corresponding to 1.3317–1.3346 RIU) when the scanning rate was set to be 100 Hz. The 0 EtOH% sample was first introduced into the sample cell and then was exchanged into another sample with a syringe in order. Solid lines in Fig. 4(a) show the experimental data of the angle-SPR spectrum (scanning range of incident angle = 3.2°, angle sampling points =1,000, angular sampling interval = 0.0032°) in the 0 EtOH% (pure water) sample (RI = 1.3317 RIU, green color), the 2.5 EtOH% sample (RI = 1.3324 RIU, blue color), the 5 EtOH% sample (RI = 1.3331 RIU, red color), the 7.5 EtOH% sample (RI = 1.3339 RIU, purple color), and the 10 EtOH% sample (RI = 1.3346 RIU, orange color). We confirmed the RI dependence of the SPR dip in these samples.Fig. 4. (a) Angle-SPR spectrum, (b) the corresponding sensorgram, and (c) the corresponding relation between sample RI and θSPR in RI sensing of ethanol/water samples with different mixing ratios. (d) Reproducibility of RI sensing when five different ethanol/water samples were measured.
We performed curve fitting analysis of these experimental data with the quintic polynomial function to determine the center angle θSPR of the SPR dip. Figure 4(b) shows the corresponding sensorgram of θSPR when the sample RI was increased from 0 EtOH% to 10 EtOH%. A step-like change of θSPR was clearly confirmed. The impulse-like change of θSPR immediately after changing the sample RI was caused by unstable temperature and/or flow in the sample cell. We consider that very steep increase in Fig. 4(b) is mainly due to temperature drift. We determined the mean and the standard deviation of the measured θSPR with respect to different sample RIs in Fig. 4(b). Figure 4(c) shows a relation between the sample RI and the θSPR. As a linear relation was confirmed between them, we determined the linear relation to be θSPR = 142.3RI-117.1 with a coefficient of determination (R2) of 0.992. The resulting RI sensitivity was determined to be 142.3 deg/RIU from the slope coefficient. The RI resolution was estimated to be 2.306×10−5 RIU from the mean of the standard deviation of θSPR [unobservable in each plot of Fig. 4(c) due to too small amount] and the RI sensitivity. On the other hand, the RI accuracy was estimated to be 8.984×10−5 RIU when it was defined as the root mean square error (RMSE) between the experimental data [see red plots in Fig. 4(c)] and the linear approximation [see blue line in Fig. 4(c)]. Even at a scan rate of 100 Hz, high resolution and accuracy in RI sensing were achieved in the RI sensing. We evaluated the reproducibility of RI sensing. To this end, we determined the mean and the standard deviation of the measured θSPR with respect to different sample RIs when five measurements were repeated for different sets of sample. Figure 4(d) shows a relation between the sample RI and θSPR. The RI reproducibility was less than the RI accuracy in the present system.
3.3 Biosensing of antigen-antibody reaction
The results in Fig. 4 indicate the potential of beam-scanning angle-SPR mode for optical transducers in biosensors for the specific detection of biomolecules associated with antigen-antibody reactions. Next, we performed monitoring of the avidin-biotin interaction, which is a representative non-covalent interaction between protein and coenzyme and has one of the strongest dissociation constants. Here we give a brief description of the biotin modification on the sensor surface. First, the surface of the sputtered gold thin film on the prism was cleaned and organics on it was removed by UV ozone. Second, a biotin terminated self-assembled monolayer (SAM) was formed on the cleaned Au surface via the covalent bond of Au with thiol group by immersing in 50 µM of biotin-terminated-alkanethiol dissolved in ethanol. Third, the biotin was immobilized on the SAM. We prepared avidin/PBS solutions with different molar concentrations as the samples.
We acquired the angle-SPR spectrum with scanning range of incident angle of 3.2°, angle sampling points of 1,000, and angular sampling interval of 0.0032°, and determined θSPR with the curve fitting analysis of those experimental data with the quintic polynomial function. Figure 5 shows a sensorgram of θSPR in the avidin-biotin interaction when the molar concentration of avidin was increased from 1 nM to 10 µM. The step-like change of θSPR reflects the change of molar concentration. The temporal behavior of θSPR at 100 nM implies that the avidin-biotin interaction was not completed and was still in progress due to the concentration dependence of SPR sensitivity, association constant, and/or complicated events at the sensor surface. The molar concentration of 10 nM could be detected using the beam-scanning angle-SPR system. This result is comparable to the sensitivity limit of existing angle-SPR modes with the slower acquisition rate.
Fig. 5. Sensorgram of θSPR in the avidin-biotin interaction when the molar concentration of avidin was increased from 1 nM to 10 µM.
4. Discussion
We discuss a comparison of the basic performance among the GM-based beam-scanning angle-SPR mode (present work), the DMD-based beam-scanning angle-SPR mode [15], and a commercialized multi-channel angle-SPR mode (Kyushu Keisokki Co., Ltd., Japan, Kretschmann configuration, LED wavelength = 650 nm, Au film thickness = 50 nm, angle resolution = 0.00001°, angular sampling interval = 0.0078°, ADC resolution = 10 bit, measurement rate = 25 Hz). Table 1 summarizes a specification of them. From the comparison among them, GM-based beam-scanning angle-SPR mode is highlighted in the angular sampling interval and the acquisition rate.
Table 1. Comparison of specification among the GM-based beam-scanning angle-SPR mode, the DMD-based beam-scanning angle-SPR mode, and a commercialized multi-channel angle-SPR mode
We further measured the angle-SPR spectrum of pure water (RI = 1.3317 RIU) using the GM-based beam-scanning angle-SPR mode and the commercialized multi-channel angle-SPR mode. We extracted a single angle-SPR spectrum from multiple angle-SPR spectra measured by the multi-channel angle-SPR mode. Figure 6 compares the measured angle-SPR spectrum (see red line) measured using (a) the GM-based beam-scanning angle-SPR mode and (b) the multi-channel angle-SPR mode. We performed curve fitting analysis of the experimental data with the quintic polynomial function as shown in blue lines in Figs. 6(a) and 6(b), similar to above. The measured spectra in both modes were in good agreement with the fitting curves; however, there is a significant difference of signal-to-noise ratio (SNR) between them. The increased noise in Fig. 6(b) was due to the uneven sensitivity of the camera pixels because the intensity of the optical signal in the multi-channel angle-SPR system was sufficiently high. Residuals between them in the GM-based beam-scanning angle-SPR mode and the multi-channel angle-SPR mode are shown in Figs. 6(c) and 6(d), respectively. The standard deviation of those residuals was 1.16×10−3 for the GM-based beam-scanning angle-SPR mode and 1.09×10−2 for the multi-channel angle-SPR mode. This comparison highlights the advantage of the GM-based beam-scanning angle-SPR mode although one has to consider the difference of experiment setup (for example, optical power of a light source) between them.
Fig. 6. Angle-SPR spectrum of the pure water measured using (a) the GM-based beam-scanning angle-SPR mode and (b) the multi-channel angle-SPR mode. Experimental data and the corresponding fitting curve were indicated as a red line and a blue one, respectively. Residual of reflectance between the experimental data and the fitting curve in (c) the GM-based beam-scanning angle-SPR mode and (d) the multi-channel angle-SPR mode.
5. Conclusion
We demonstrated beam-scanning angle-SPR mode based on a combination of a galvanometer mirror and relay lenses. The RI resolution and the RI accuracy were respectively 2.306×10−5 RIU and 8.984×10−5 RIU. Then, the beam-scanning angle-SPR mode was effectively applied for the real-time monitoring of the avidin-biotin antigen-antibody reaction with the molar concentration of 10 nM to 10 µM. Furthermore, compared with the multi-channel angle-SPR mode, the better performance of the beam-scanning angle-SPR mode was highlighted in the SNR of the angle SPR spectrum.
Although the beam-scanning angle-SPR mode was performed at the wavelength of visible light in this article, the use of the near-infrared (NIR) light (for example, 1550 nm) in angle-SPR has the potential to further boost the sensing performance because the SPR dip is significantly sharper in the NIR region compared with the visible region [10]. On the other hand, the high cost and poor performance of NIR cameras (for example, InGaAs CCDs) make it difficult to use them for multi-channel angle-SPR at those wavelengths. Fortunately, as the performance of NIR photodetectors (for example, InGaAs photodiodes) is comparable to that of visible photodetectors (for example, Si photodiodes), the effectiveness of beam-scanning angle-SPR mode will become more apparent in RI sensing and bio-sensing at NIR wavelengths. Work is in progress to perform beam-scanning angle-SPR at a wavelength of 1550 nm while considering the wavelength dependency of SPR decay length.
Funding
Japan Agency for Medical Research and Development (20he0822006j00).
Acknowledgments
The authors thank Ms. Asaka Murakami of Tokushima University for her help in English proofreading of the manuscript.
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. P. Pattnaik, “Surface plasmon resonance,” Appl. Biochem. Biotechnol. 126(2), 079–092 (2005). [CrossRef]
2. J. Homola, S. S. Yee, and G. Gauglitz, “Surface plasmon resonance sensors,” Sens. Actuators, B 54(1-2), 3–15 (1999). [CrossRef]
3. J. Homola, “Present and future of surface plasmon resonance biosensors,” Anal. Bioanal. Chem. 377(3), 528–539 (2003). [CrossRef]
4. 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]
5. K. Matsubara, S. Kawata, and S. Minami, “Optical chemical sensor based on surface plasmon measurement,” Appl. Opt. 27(6), 1160–1163 (1988). [CrossRef]
6. W. Y. W. Yusmawati, H. P. Chuah, and M. Y. W. Mahmood, “Optical properties and sugar content determination of commercial carbonated drinks using surface plasmon resonance,” Am. J. Appl. Sci. 4(1), 1–4 (2007). [CrossRef]
7. H. Liang, H. Miranto, N. Granqvist, J. W. Sadowski, T. Viitala, B. Wang, and M. Yliperttula, “Surface plasmon resonance instrument as a refractometer for liquids and ultrathin films,” Sens. Actuators, B 149(1), 212–220 (2010). [CrossRef]
8. K. Tiwari and S. C. Sharma, “Surface plasmon based sensor with order-of-magnitude higher sensitivity to electric field induced changes in dielectric environment at metal/nematic liquid-crystal interface,” Sens. Actuators, A 216, 128–135 (2014). [CrossRef]
9. C. E. H. Berger, T. A. M. Beumer, R. P. H. Kooyman, and J. Greve, “Surface plasmon resonance multisensing,” Anal. Chem. 70(4), 703–706 (1998). [CrossRef]
10. R. Ziblat, V. Lirtsman, D. Davidov, and B. Aroeti, “Infrared surface plasmon resonance: a novel tool for real time sensing of variations in living cells,” Biophys. J. 90(7), 2592–2599 (2006). [CrossRef]
11. A. Shalabney and I. Abdulhalim, “Figure-of-merit enhancement of surface plasmon resonance sensors in the spectral interrogation,” Opt. Lett. 37(7), 1175–1177 (2012). [CrossRef]
12. S. Zeng, D. Baillargeat, H.-P. Ho, and K.-T. Yong, “Nanomaterials enhanced surface plasmon resonance for biological and chemical sensing applications,” Chem. Soc. Rev. 43(10), 3426–3452 (2014). [CrossRef]
13. Y. Zhao, S. Gan, G. Zhang, and X. Dai, “High sensitivity refractive index sensor based on surface plasmon resonance with topological insulator,” Results Phys. 14, 102477 (2019). [CrossRef]
14. D. Lakayan, J. Tuppurainen, M. Albers, M. J. van Lint, D. J. van Iperen, J. J. A. Weda, J. Kuncova-Kallio, G. W. Somsen, and J. Kool, “Angular scanning and variable wavelength surface plasmon resonance allowing free sensor surface selection for optimum material- and bio-sensing,” Sens. Actuators, B 259, 972–979 (2018). [CrossRef]
15. D. Wang, F.-C. Loo, H. Cong, W. Lin, S. K. Kong, Y. Yam, S.-C. Chen, and H. P. Ho, “Real-time multi-channel SPR sensing based on DMD-enabled angular interrogation,” Opt. Express 26(19), 24627–24636 (2018). [CrossRef]
16. R. V. Andaloro, H. J. Simon, and R. T. Deck, “Temporal pulse reshaping with surface waves,” Appl. Opt. 33(27), 6340–6347 (1994). [CrossRef]
17. G. M. Hale and M. R. Querry, “Optical constants of water in the 200-nm to 200-µm wavelength region,” Appl. Opt. 12(3), 555–563 (1973). [CrossRef]
18. J. Rheims, J. Köser, and T. Wriedt, “Refractive-index measurements in the near-IR using an Abbe refractometer,” Meas. Sci. Technol. 8(6), 601–605 (1997). [CrossRef]
19. R. Oe, S. Taue, T. Minamikawa, K. Nagai, K. Shibuya, T. Mizuno, M. Yamagiwa, Y. Mizutani, H. Yamamoto, T. Iwata, H. Fukano, Y. Nakajima, K. Minoshima, and T. Yasui, “Refractive-index-sensing optical comb based on photonic radio-frequency conversion with intracavity multi-mode interference fiber sensor,” Opt. Express 26(15), 19694–19706 (2018). [CrossRef]
