Open Access
Add to BibSonomy
Abstract
This study experimentally demonstrated the effects of two polarization-maintaining fibers (PMFs) as a sensing region instead of conventional single-mode fiber (SMF) using Au-coated hetero-core optical fiber surface plasmon resonance (SPR) sensor. We experimentally observed that the SPR resonant wavelength is shifted toward longer wavelength with refractive index (RI) increasing from 1.332 to 1.396. A PMF sensor exhibits the broad SPR spectra, resulting in a higher sensitivity with a 1.5-fold change in the light intensity at 850 nm relative to RI, compared with the SMF case. Discussions most likely responsible for this effect are given by the lower angular distribution peak in the hetero-core region of PMFs.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Optical fiber sensors using surface plasmon resonance (SPR) [1–10] have been of interest in many sensing applications and biomedical analysis over the past decade because of their high sensitivity to refractive index (RI) changes and ability to sense inaccessible locations. Optical fibers can offer some attractive advantages, including the ability to remotely sense without electricity and the ability to be used in flammable and hazardous environments. Currently, several kinds of SPR-based optical fiber sensors are available, which include tapered fibers [2,3], unclad fibers [4,5], fiber Bragg grating (FBG) [6], and those with hetero-core fiber structures [7–10]. Tapered and unclad fibers are required to eliminate thick cladding layers to access the transmitted light in the core by evanescent waves, resulting in loss of mechanical strength. Meanwhile, FBG sensors have temperature and strain dependencies that should be compensated. Hetero-core fiber sensors require no temperature compensation, and their structures have high mechanical strength because a single-mode optical fiber piece is placed between two multimode fibers.
Compared to conventional techniques, SPR optical fiber sensors with a hetero-core structure can be easily fabricated with a simple structure without the need for cladding removal. The structure consists of an inserted single-mode fiber (SMF) segment that functions as a sensing region, namely, hetero-core section, whose core diameter is smaller than that of the transmission multimode fiber (MMF). Silver- and gold (Au)-coated SPR sensors have been reported to be attractive in terms of high sensitivity and high resolution in the order of 10−4–10−5 RIU for RIs of 1.333–1.398 RIU operating at a spectral resolution of 0.5 nm [7,8].
The performance of the optical fiber-based SPR sensor is determined by several factors, such as the thickness and dielectric permittivity of the plasmonic material layer, the bending effect, and the RI of the fiber surface in contact with the metal layer. Until now, the effects of bending on conventional SPR fiber-optic sensors have not been revealed because of the lack of flexibility in changing the curvature radius. Interestingly, non-SPR single U-bend plastic-clad silica fibers [11–13] have been reported to increase the sensitivity of such sensors owing to the increased penetration depth of evanescent waves. A theoretical model for the bending of SPR sensors has also been reported, showing that probe bending improves sensor sensitivity [14]. Calculations have shown that for a curvature radius as small as 10 mm, sensitivity to wavelength shift can be obtained under a constant power distribution mode even when the core diameter is as large as 600 µm. In addition, Hus et al. [15] theoretically and experimentally investigated the sensor sensitivity of curved D-type optical fiber sensor based on SPR for RIs (core diameter = 62.5 µm), by changing the radius curvature (480–630 mm) and the core-unpolished depth. The experimental results revealed that the curved D-type sensor with a 480-mm radius had a wide dynamic range of 1.33–1.43 RIs with sensitivities of 2.03 × 10–5 and 2.05 × 10–4 RIU theoretically and experimentally, respectively.
In our previous study [9], we experimentally demonstrated the bending effects of Au-coated, hetero-core optical fiber SPR sensors, taking advantage of the flexible bending capability of hetero-core sensors. It was found that a decrease in the radius of curvature (approximate curvature radius of 9.7 mm) resulted in a 2.8-fold increase in the SPR loss spectrum peak with a decrease in the spectral bandwidth by 75% and an additional 9-nm shift in the SPR resonance wavelength, compared to straight case. This is because that the distribution peak slightly shifts to a lower angle by about 1°. Experimental results indicated that shifting the power distribution near the critical angle can improve the sensitivity of the sensor. Therefore, if the incident light can be directed into a medium with lower RI than that of the cladding and then leaks into the cladding in the hetero-core region, the incident angular distribution can be pushed up near the critical angle based on Snell’s law. Instead of a conventional SMF element with a region of different RI between the cladding and core, commercially available polarization-maintaining fibers (PMFs) or photonic crystal fibers can be used as a hetero-core region.
In this study, we experimentally revealed the effect of two PMFs as a hetero-core region and applied it as a new type of RI sensor, with regard to the application of the SPR phenomenon to hetero-core structured optical fibers. To the best of our knowledge, this is first study of successfully changing the incident angular distribution of the sensor portion without the conventional fiber bending. A PMF fiber piece was sandwiched between two multimode fibers by thermally fused splicing. Two PMFs with core diameters similar to conventional SMF elements (core diameter = 3 µm) were tested, and their sensing behavior was investigated. The SPR spectra of the hetero-core portion in the PMF sensor were obtained for RI changes. Next, the sensitivity and capability of the hetero-core fiber RI sensors were compared with those of the SMF sensor element. A PMF sensor showed the broad SPR spectra, resulting in a higher sensitivity with a 1.5-fold change in the light intensity at 850 nm relative to RI, compared with the hetero-core portion of the SMF case. The experimental results are discussed in detail concerning the possible enhancement of the SPR excitation.
2. Materials and methods
2.1 Materials
Glycerol was purchased from FUJIFILM Wako Pure Chemical Corp., Japan. Deionized water with a resistivity of 18.2 MΩ cm was obtained from a water purification system (Direct Q UV 3, Merck KGaA., Germany). Single-mode step-index fiber (F-SA, core diameter = 2.8–4.1 µm), multimode GI fiber ((FutureGuide-MM50, core diameter = 50 µm), and two PMFs (PMS-450-XP, core diameter = 3.3–4.6 µm, stress rod diameter = 34 µm and PM460-HP, core diameter = 3.3 µm, stress rod diameter = 36 µm), called PMF-1 and PMF-2, were provided by Newport Corp., USA, Fujikura Ltd., Japan, and Thorlabs Inc., Japan, respectively. All fibers had a cladding diameter of 125 µm.
2.2 Sensor fabrication and operation principle
A hetero-core optical fiber structure using PMF as the hetero-core region is shown in Fig. 1(a). This structure consists of a transmission line MMF and a PMF insertion segment. Since the distance from the core center to the center of the stress rod is 27.5 µm [Fig. 1(b)], the MMF covers half of the stress rod. The sensing mechanism is the same as the SMF, as follows [7–10]. Owing to the difference in the core diameter, most of the light leaks into the cladding of the PMF element. An evanescent wave is generated at the cladding layer surface in the hetero-core portion when it is bounced off the boundary between the cladding region and the environment under conditions of total internal reflection. At the other end of the PMF element, a portion of the light is re-coupled into the core of the downstream multimode fiber. A thin metal film coating on the cladding surface provides an SPR optical arrangement similar to that of a Kretschmann configuration sensor. As shown in Figs. 1(c) and (d), in the presence or absence of stress rods, the PMF cladding may have two angles of incidence. As shown in Fig. 1(c), various incident angles centered on ${\theta _1}\; $ will occur. In contrast, as shown in Fig. 1(d), multiple incident angles centered on ${\theta _2}$ passing through the stress rod would be closer toward the critical angle than ${\theta _1}$, affecting the SPR conditions.
Fig. 1. Hetero-core optical fiber sensor. (a) Hetero-core region in the case of polarization-maintaining fiber (PMF). MMF: multimode fiber. (b) Cylindrical cross-section of the sensor portion. (c) A cross-sectional view along plane 1 in (b). (d) A cross-sectional view along plane 2 in (b).
All fiber samples were fabricated with a 15-mm segment of SMF or PMF (PMF-1 and PMF-2) inserted into the MMF transmission line. Segmented fiber insertion was accomplished using a thermal fusion splicer (FSM-100P+, Fujikura Ltd.) after cleaving each end of fiber with a fiber cleaver (CT-32, Fujikura Ltd.). The resultant typical insertion loss of the hetero-core region for SMF or PMF was measured as all 3.9 dB at the average wavelength of 450–1000 nm. The fabricated hetero-core region was then cylindrically coated with an Au thickness of 30 nm using a radio-frequency sputtering machine with a special rotation mechanism (CFS-4ES-231, Shibaura Mechatronics Corp., Kanagawa, Japan). Au was symmetrically deposited on the cladding surface [8–10].
2.3 Experimental setup
The experimental setup consisted of a halogen lamp with a wavelength of 450–1000 nm (HL-2000-LL, Ocean Optics, Tokyo, Japan) and a charge-coupled device-based spectrometer (CCS200, Thorlabs, Tokyo, Japan), as depicted in Fig. 2. The sensor was firmly mounted on a straight line with a stretching support to avoid undesirable spectral fluctuations due to fiber bending [Fig. 2(c)]. Glycerin water solutions were prepared as test solutions with concentrations from 0% (water) to 90% (w/w). The RIs of the solutions were measured using a refractometer (PR-RI, Atago Co., Ltd., Tokyo, Japan). The entire sensor was then immersed into the test solution at room temperature for spectral measurements.
Fig. 2. Experimental setup for transmission light spectrum measurements. (a) Schematic representation of the apparatus. (b) Overview of the apparatus without liquid reserver. PC: Laptop for data recording; LS: White light source; Sensor: The sensing portion tightly held with a stretching support. A liquid reserver is put on a jack; SP: Spectrometer. (c) Closeup view of the hetero-core portion and stretching support. (d) Fusion splicing view of the hetero-core portion for PMF along planes 1 and 2 in Fig. 1(b).
2.4 Data processing
The resonance wavelength with minimum intensity of the SPR spectrum was defined as λres. The full bandwidth at half maximum of the SPR loss spectra based on 0 dB is defined as WFWHM. In the case of PMFs, a Gaussian fit curve was applied to the obtained SPR spectrum of the hetero-core portion, and a definite λres and WFWHM were obtained using Origin Software (Lightstone Corp., Japan).
Theoretical calculations of SPR spectra were performed using a multilayered structure in which the transmitted light waves are reflected at an angle of $\theta $ at the boundary between the cladding surface and the metal layer, yielding total internal reflection in an optical arrangement in a given plane. In this simulation, the reflection coefficient of the multilayer structure was calculated using the transfer matrix method [16]. The dielectric function of Au was adopted from the literature [17]. The RIs of 1.4646, 1.4611, and 1.4648 were employed for the cladding of SMF, PMF-1, and PMF-2, respectively.
The power distribution ${P_\theta }$ at the incident angular distribution was calculated using Eq. (1). Detailed derivation of formula is described in [9].
3. Results and discussion
3.1 Incident angular distribution in the hetero-core region
To estimate the SPR effect of the incident angle due to the presence or absence of the stress rods for the hetero-core portion in the PMF, the theoretical SPR spectra were calculated using a three-layered structure (cladding-Au-sample) for various incident angles. Figure 3 shows the calculated SPR spectra as a function of the wavelength for several incident angles (θx = 78°, 80°, 82°, 84°, and 86°). Figures 3(a) and (b) show the cladding RI of 1.4611 (PMF-1) and 1.4648 (PMF-2) for a sample medium of 1.369 RIU with a 30-nm thick Au film, respectively. These spectra were normalized with the air spectrum given by a 30-nm-thick Au film. The decrease in incident angles from 86° to 78° caused the λres-value and spectral width to elongate and broaden, regardless of the cladding RI.
Fig. 3. Theoretical SPR spectra as a function of wavelength for an Au thickness of 30 nm for several incident angles: the cladding RI of (a) 1.4611 and (b) 1.4648 for the sample medium of 1.369 RIU.
Figures 4(a)–(c) show the experimentally obtained sensor responses for RIs of 1.332, 1.344, 1.357, 1.369, 1.383, and 1.396 as a function of wavelength, corresponding to glycerin concentrations of 0, 9.5, 19.5, 29.1, 39.3, and 48.5% (w/w). These results were normalized by the air environment spectrum. The cladding modes are essential in producing the SPR excitation by coating the cladding surface with thin metal film, similar to the Kretschmann configuration sensor case. When the metal film thickness is sufficiently thin, an evanescent filed is generated, extending toward the metal layer and reaching the tested medium. Consequently, evanescent waves can induce the SPR at the interface between the cladding-layers and Au film under the well-known matching conditions between the horizontal wave number vector of the incident light and that of the surface plasmon wave [7]. Since the wave number vector of the surface plasmon wave depends on the surrounding medium’s dielectric permittivity around the metal layer, the shifts of the SPR resonance wavelength were observed for the RI changes. Therefore, the RI of the sample medium can be measured based on the SPR resonant wavelength. As shown in Fig. 4(a), the SPR spectra obtained for glycerin solutions are agree excellently with the previous results [8]. Briefly, the typical minimum intensity due to λres was observed around 571, 585, 610, 633, 669, and 719 nm. The spectra obtained were broader in their resonant curve than those in the theoretical case of a single incident angle. Contrary to the theoretical case, the incident angle of cylindrical hetero-core cladding is not single but spreading to form an angular distribution around a certain angle somewhere between the critical angle and 90°. Therefore, the shape of SPR spectra became broader because each mode in the cylindrical wave guide reflects a corresponding incident angle, with giving a slightly shifted SPR resonant wavelength [7].
Fig. 4. Experimentally obtained SPR spectra (solid lines) and their Gaussian fit curves (dashed lines) normalized with the air spectrum for the hetero-core portion in (a) SMF, (b) PMF-1, and (c) PMF-2. The thickness of the Au film is 30 nm.
The hetero-core portion in the abovementioned case of the tested PMF is shown in Figs. 4(b) and (c). We successfully obtained the typical SPR curves in which the minimum intensities appear at a specific wavelength, although several peaks appear at the wavelength. These peaks were not observed for the hetero-core portion of SMF. The reason for this stem from the phase difference between two orthogonal polarization lights from the presence and absence of stress rods in PMF because the effective index is different between two orthogonal axes. The λres-values of both PMF-1 and PMF-2 elongate with increasing RI values. Additionally, the spectral width for the PMF-1 case slightly broadened, and λres elongated, compared with SMF and PMF-2. The reason for this would come from the lower incident angle of the hetero-core portion in PMF-1. In addition, the double peaks for PMF were observed by two different incident angles such as ${\theta _1}\; $ and ${\theta _2}\; $ shown in Fig. 1 due to different effective indices in the two orthogonal axes of PMF because the theoretically calculated SPR spectra depended on the incident angle (Fig. 3). As shown in Fig. 4(b), double peaks are observed at 615 and 652 nm, respectively, for PMF-1, indicating that they occurred at incident angles near 85° and 82°. In the case of PMF-2, for the hetero-core portion [Fig. 4(c)], the double peaks obtained are at 625 and 643 nm, corresponding to calculated incident angles around 83° and 82°. In the case of SMF [Fig. 4(a)], the experimental λres at 633 nm for 1.369 RIU gives an incident angle of about 83°. Considering that the spectrum becomes wider as the angle of incidence decreases, it can be seen that for PMF-1, the lower angle of incidence mainly dominates the sensor response. On the contrary, the spectral width of PMF-2 would be affected by a larger angle of incidence, as in the case of SMF. The experimental results suggest that the incident angular distribution with the lower incident angle may be formed in the hetero-core portion of PMF-1.
Plots in Figs. 5 and 6 show the full bandwidth WFWHM and the resonance wavelength λres, respectively, as a function of RI of the glycerin solutions for the hetero-core portion of SMF, PMF-1, and PMF-2, respectively. The λres and WFWHM of PMF sensors were obtained from the dashed line curves [Fig. 4(b) and (c)]. WFWHM is monotonically increased and the λres shifts toward longer wavelengths as the RI increases for all test fibers. It can be seen from Fig. 5 that the PMF-2 shows the narrower WFWHM compared with PMF-1 and SMF. Discussions most likely responsible for this effect are given by introducing the larger incident angles of the hetero-core portion of PMF-2. In this experiment, the shift rate vs. RI increases with increasing RI in the range 1.332–1.396. For PMF-1, the spectral sensitivities derived from the wavelength shift curve of the hetero-core portion were calculated to be 1258 nm/RIU and 5855 nm/RIU for RIs of 1.332 and 1.396, respectively. These values correspond to a detection resolution of 3.97 × 10−4 RIU and 8.53 × 10−5 RIU for RI, assuming a wavelength resolution of 0.5 nm in the spectral analyzer. The detection limits for SMF and PMF-1 were 5.69 × 10−4 RIU and 7.47 × 10−4 RIU for 1.332 RIU, and 1.37 × 10−4 RIU and 1.34 × 10−4 RIU for 1.396 RIU, respectively. This resolution is comparable to the previously reported SPR fiber sensors [1,2,15]. The sensitivity of the sensor can be improved by increasing the thickness of the gold film to 60 nm [8]. The shift ratio of PMF-1 is larger than that of SMF and PMF-2, indicating that PMF-1 is a superior spectral mode of operation in terms of spectral sensitivity characteristics.
Fig. 5. Full bandwidth of the SPR spectra, WFWHM, as a function of RI for hetero-core portions in SMF, PMF-1, and PMF-2.
Fig. 6. Resonance wavelength λres as a function of RI for hetero-core portions in SMF, PMF-1, and PMF-2.
Construction of the stable measurement system is important because the obtained SPR spectra for PMF case have the several peaks. The interference would be affected depending on the insertion length of the hetero-core portion. In fact, although other PMF sensors fabricated under the same conditions in Section 2.3 showed reproducible SPR spectra (Figs. S1 and S2), the interference was also induced due to the slight difference of PMF insertion length. A simple combination of an LED as a light source and an optical power detector is beneficial owing to its high-cost-performance, compared with a low-cost-effective measurement system, such as the spectral operation measuring the wavelength shift. It is considered that because general LEDs have a wide spectral width, the interference effect of PMF sensors can be neglected at the wavelength of 850 nm. Additionally, the obtained broad SPR spectra enabled us to measure the sensor signal using a single wavelength at the tail of the SPR spectrum such as 850 nm which is advantageous for practical use of the sensor with simple existing detectors for the near-infrared region [10]. Figure 7 demonstrates the intensity change extracted from spectral data at a center wavelength of 850-nm with the range of ±15 nm. Total attenuation losses including insertion losses of SMF, PMF-1, and PMF-2 are 5.4, 5.9, and 5.3 dB at the wavelength of 850 nm for the RI of 1.396, respectively. The light intensity ratio to RI also increases, with RI ranging from 1.332 to 1.396 for all fibers tested. As can be seen from the curves in Fig. 6, the intensity changes appeared differently with depending on the PMF-1 and PMF-2. Compared with SMF and PMF-2, the hetero-core portion in PMF-1 exhibits higher sensitivity with a 1.5-fold change in light intensity relative to RI. In addition, as shown in Figs. S1 and S2, PMF sensors can detect the RI changes with reproducible performance based on the spectral and amplitude operation.
Fig. 7. Experimental normalized intensity at a near-infrared wavelength of 850 nm with the range of ±15 nm as a function of RI for hetero-core portions in SMF, PMF-1, and PMF-2.
3.2 PM fiber effects found in a hetero-core optical fiber SPR sensor
Experiments with varying concentrations ranging approximately between 60% and 90% of glycerin solution were performed to investigate the existence of cladding modes in the hetero-core region, which could be excited by the incident light wave coupled from the upstream transmission GI fiber line, using a hetero-core coated with a nonmetal, similar to previous studies [9,18]. Figure 8 shows the losses obtained at average wavelengths of 450–800 nm for the hetero-core portions of SMF, PMF-1, and PMF-2 as a function of glycerin concentration (the horizontal axis also shows the RI of glycerin). The optical losses increased with the glycerin concentrations. The most coherent mechanism by which these losses occur can be found as a change in the critical angle condition due to an increase in RI at a given glycerin solution concentration. A well-known theory explains that as the RI of the medium around the hetero-core cladding increases, the critical angle of total internal reflection increases [9,18]. As the critical angle increases, light waves reflecting at the outermost cladding surface with the angle of incidence are no longer reflected and could disappear. Assuming that many cladding modes begin to build up through the total internal reflection with corresponding different incident angles in the hetero-core cladding, the loss variation as a function of critical angle allows one to deduce the possible incident angle distribution within the hetero-core cladding that dominates power transmission. Moreover, these modes were developed as a transient stage because the distances are too short for the hetero-core regions to form a well-developed multimode configuration; thus, these arguments enable to achieve the required sensing capability.
Fig. 8. Losses obtained at average wavelengths of 450–800 nm for the hetero-core portions of SMF, PMF-1, and PMF-2 as a function of glycerin concentration. The horizontal axis also shows the RI of glycerin. Data are expressed as mean ± SD (n = 3).
Looking at a group of plots in Fig. 8 for low glycerin concentrations, there appears to be no change in loss. In contrast, at higher concentrations, above 83% (w/w), the obtained losses are greater. In particular, the hetero-core portion of PMF-1 shows a sharp increase in optical loss at high RIs, although optical loss in PMF-2 sensor shows a similar response compared to SMF. Figures 9(a) and (b) show the schematic view of the hetero-core region of PMF-2 and the PMF-1 with stress rods [Fig. 1(d)], respectively. ${n_c}\; $ and ${n_s}\; $ denote the RIs of the cladding layer and environmental solution medium, respectively. ${\theta _{{n_s}}}$ denotes the critical angle given when the sensor is immersed in solution having an index ${n_s}$.
Fig. 9. Optical mode distribution for the hetero-core portion in (a) PMF-2 and (b) PMF-1, and its corresponding power density.
The incident angular distribution is indicated by the fan shape with a gray gradation in Fig. 9. In the case of the solution with a low RI, the critical angle (${\theta _{{n_s}}})$ is small and far away from the incident angular distribution, including most of the power. Thus, the power in the modes could be reserved if the PMF is applied to the hetero-core region. When RI increases, ${\theta _{{n_s}}}$ increases and approaches the incident angular distribution. Figure 9 shows that the power, even in the PMF-2 and SMF cases, shows the optical loss because ${\theta _{{n_s}}}$ increases continuously and exceeds the incident beam region. When the hetero-core portion is the PMF-1, it is presumed that the incident angular distribution could be more closely formed toward ${\theta _{{n_s}}}$, so that a larger power could be lost depending on the PMF-1 applied to the hetero-core region in a high-RI case, in contrast to that in the low-RI case.
On the basis of the abovementioned discussion, the power distribution ${P_\theta }$ in the incident angular distribution can be derived from the loss curves as a function of RI for various bending conditions. We plotted the power distribution as a function of the incident angle for all tested fibers normalized by ${P_{out}}$, as shown in Fig. 10, using the Eq. (1). The peak angle of the power distribution for PMF-1 appears at about 83°, whereas that of PMF-2 remains at a similar level of about 84° compared to that of SMF. Figure 10 shows that the distribution peak forms to a lower angle by ∼1° for the hetero-core portion in PMF-1. According to the SPR theoretical simulation shown in Fig. 3, a 1° shift can only produce a λres shift on the order of several nanometers, which agrees well with the experimentally observed shift in λres shown in Fig. 4.
Fig. 10. Power distribution as a function of incident angle calculated from Eq. (1) for hetero-core portions in SMF, PMF-1, and PMF-2.
The effective incident angle may be smaller range than the incident angular distribution shown in Fig. 10. When the range of incident angle becomes smaller, the SPR spectra would be narrower because of the reduction of the overlap of SPR spectra. Therefore, another possible reason for the sharp SPR spectra of PMF-2 may be associated with the smaller range of incident angles. However, it is not clear in the present work which range of incident angular distribution influences the SPR spectrum. The experimental result indicates that the shift in λres and broad spectra induced by the hetero-core portion for PMF-1 can be explained by the lower angular power $({{P_\theta }} )$ distribution.
4. Conclusions
The effectiveness of PMF as a sensing region was experimentally demonstrated using Au-coated hetero-core optical fiber SPR sensor instead of conventional SMF pieces. The PMF fiber was sandwiched between two multimode fibers using thermally fusion splicing. The advantageous feature of the proposed sensor is that it can be simply fabricated with commercially available fibers via thermally fusion splicing, which enables makes the sensing element easy to manufacture and robust. PMF sensors show attractive characteristics regrading the resonant wavelength and width of SPR spectra. Most interestingly, the broad spectra can enhance the sensitivity of the SPR sensor. The experimental results reported herein are discussed in detail, including the resonant wavelength and the intensity change at 850 nm. Our conventional simple measurement system based on the LED/PD [10] can be easily applied to the proposed sensor. Furthermore, a simple geometrical optics theory shows the different mode distribution formation in the hetero-core portion of PMF-1 so that the distribution could occur near the critical angle. This indicates that the proposed hetero-core SPR sensor using PMF can have two selectivities using a double-sided coating on the semicylinder surface of the sensor region, as exemplified in the detection of hydrogen and oxygen, as well as temperature and humidity.
Funding
Japan Society for the Promotion of Science (JP21K14173); Japan Science and Technology Agency, Core Research for Evolutional Science and Technology (JPMJCR17O2).
Disclosures
The authors declare no conflicts of interest.
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.
Supplemental document
See Supplement 1 for supporting content.
References
1. R. Nasirifar, M. Danaie, and A. Dideban, “Surface plasmon resonance biosensor using inverted graded index optical fiber,” Photonics and Nanostructures-Fundamentals and Applications 44, 100916 (2021). [CrossRef]
2. Y. Al-Qazwini, A. S. M. Noor, M. H. Yaacob, S. W. Harun, and M. A. Mahdi, “Experimental realization and performance evaluation of refractive index SPR sensor based on unmasked short tapered multimode-fiber operating in aqueous environments,” Sens. Actuators, A 236(1), 38–43 (2015). [CrossRef]
3. M.-C. Navarrete, N. D-Herrera, A. G-Cano, and Ó Esteban, “Surface plasmon resonance in the visible region in sensors based on tapered optical fibers,” Sens. Actuators, B 190, 881–885 (2014). [CrossRef]
4. Y. S. Dwivedi, A. K. Sharma, and B. D. Gupta, “Influence of design parameters on the performance of a surface plasmon sensor based fiber optic sensor,” Plasmonics 3(2-3), 79–86 (2008). [CrossRef]
5. Q. Wang, B. Sun, E. Hu, and W. Wei, “Cu/ITO-coated uncladded fiber-optic biosensor based on surface plasmon resonance,” IEEE Photonics Technol. Lett. 31(14), 1159–1162 (2019). [CrossRef]
6. T. Guo, “Fiber grating-assisted surface plasmon resonance for biochemical and electrochemical sensing,” J. Lightwave technol. 35(16), 3323–3333 (2017). [CrossRef]
7. M. Iga, A. Seki, and K. Watanabe, “Hetero-core structured fiber optic surface plasmon resonance sensor with silver film,” Sens. Actuators, B 101(3), 368–372 (2004). [CrossRef]
8. M. Iga, A. Seki, and K. Watanabe, “Gold thickness dependence of SPR-based hetero-core structured optical fiber sensor,” Sens. Actuators, B 106(1), 363–368 (2005). [CrossRef]
9. K. Takagi, H. Sasaki, A. Seki, and K. Watanabe, “Surface plasmon resonances of a curved optical hetero-core fiber sensor,” Sens. Actuators, A 161(1-2), 1–5 (2010). [CrossRef]
10. A. Hosoki, M. Nishiyama, H. Igawa, A. Seki, Y. Choi, and K. Watanabe, “A surface plasmon resonance hydrogen sensor using Au/Ta2O5/Pd multi-layers on hetero-core optical fiber structures,” Sens. Actuators, B 185, 53–58 (2013). [CrossRef]
11. B. D. Gupta, H. Dodeja, and A. K. Tomar, “Fibre-optic evanescent field absorption sensor based on a U-shaped probe,” Opt. Quantum Electron. 28(11), 1629–1639 (1996). [CrossRef]
12. S. K. Khijwania and B. D. Gupta, “Fiber optic evanescent field absorption sensor: effect of fiber parameters and geometry of the probe,” Opt. Quantum Electron. 31(8), 625–636 (1999). [CrossRef]
13. S. K. Khijwania, K. L. Srinivasan, and J. P. Singh, “An evanescent-wave optical fiber relative humidity sensor with enhanced sensitivity,” Sens. Actuators, B 104(2), 217–222 (2005). [CrossRef]
14. R. K. Verma and B. D. Gupta, “Theoretical modeling of a bi-dimensional U-shaped surface plasmon resonance based fiber optic sensor for sensitivity enhancement,” J. Phys. D: Appl. Phys. 41(9), 095106 (2008). [CrossRef]
15. J.-C. Hsu, S.-W. Jeng, and Y.-S. Sun, “Simulation and experiments for optimizing the sensitivity of curved D-type optical fiber sensor with a wide dynamic range,” Opt. Commun. 341(15), 210–217 (2015). [CrossRef]
16. D. Roy, “Surface plasmon resonance spectroscopy of dielectric coated gold and silver films on supporting metal layers: Reflectivity formulas in the kretschmann formalism,” Appl. Spectrosc. 55(8), 1046–1052 (2001). [CrossRef]
17. A. K. Sharma and B. D. Gupta, “On the performance of different bimetallic combinations in surface plasmon resonance based fiber optic sensors,” J. Appl. Phys. 101(9), 093111 (2007). [CrossRef]
18. M. Iga, A. Seki, Y. Kubota, and K. Watanabe, “Acidity measurements based on a hetero-core structured fiber optic sensor,” Sens. Actuators, B 96(1-2), 234–238 (2003). [CrossRef]
