Open Access
Add to BibSonomy
Abstract
With technological advancement, photonic crystal fibers (PCFs) are effectively used to design miniaturized, flexible, and efficient biosensors. This paper proposes an exposed core PCF biosensor based on widely known surface plasmon resonance (SPR) phenomena. An external sensing mechanism is followed to characterize the sensing performance within the refractive index (RI) range between 1.28 and 1.40. Metal strip (gold (Au) and titanium dioxide (TiO2)) is deposited on the outer surface only along the four channels instead of the entire surface, which could decrease the difficulties associated with the metal deposition on the entire circular surface. Simulating the sensor using finite element method based COMSOL Multiphysics software, we find tremendous amplitude sensitivity of 7420.69 RIU−1 and wavelength sensitivity of 87,000 nm/RIU. In addition, the sensor offers the highest resolution of 7.7×10−6 RIU, the figure of merit of 1011.63 RIU−1, signal to noise ratio of 10.05 dB, the detection accuracy of 0.016598 nm−1, and detection limit of 102.23 nm. However, the promising sensing performance indicates that the proposed sensor could be implemented effectively to detect different biological and chemical substances.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
A device that combines biological elements and a physicochemical detector to identify chemical substances is known as a biosensor. Configuration of biosensors can be illustrated by combining numerous biological elements [1,2]. Nowadays, the demand for photonic crystal fiber (PCF) based surface plasmon resonance (SPR) biosensor is increasing because of having user-friendly, real-time sensing, portable, and lightweight properties. SPR is the phenomenon where the free electrons are collectively oscillated at the metal-dielectric interface [3]. If the free electrons become excited by the evanescent wave, an electromagnetic wave is generated, known as surface plasmon wave (SPW). When the frequencies of both received photons and free electrons are equal, it results in SPR excitation that promotes resonance conditions. Surface plasmons are exceptionally sensitive and any slightest change in the environmental refractive index (RI) causes the shifting of resonance of the surface plasmon polariton (SPP) mode. Among numerous types of bio-sensing devices, some are developed on photonic crystal fiber (PCF). They are regarded as promising optical sensors due to design flexibility, lightweight, immunity to electromagnetic interference, and remote sensing application. The combination of PCF with SPR gives rise to the numerous application fields with better detection feasibility, such as cancer cell recognition, food poisoning detection, bio-imaging, biochemical reaction monitoring, environmental and biological monitoring, thickness monitoring, chemical, biochemical, and organic chemical detection, gas sensing, bio-sensing, aqueous solution sensing and alcohol sensing [4–9] etc.
After designing a biosensor, choosing proper materials is crucial, and it influences the performance a lot [10]. Some materials are commonly used for biosensors in recent plasmonic research. For instance, gold, aluminum, copper, and silver are regarded as noble plasmonic materials (PMs) because these materials reveal comparatively low optical loss [11,12]. Among them, silver is considered the best metal for plasmonic because it has lower optical damping. It also has some other advantages like more conductivity with sharp and narrow-band resonance peaks. Although it has an excellent performance capability, it is also corrosive. It provides oxidation due to the existence of humidity in the environment [13]. Although copper is not that much expensive like silver and gold, it also oxidizes promptly and forms Cu$_2$O and CuO. On the contrary, gold is that type of material which is highly bio-compatible, chemically stable, and it offers less oxidation than other materials [6]. Again, it also shows greater resonance peak shifts. Due to the wider resonance peak, the performance and efficiency of the sensor can be affected. It may results in a lower sensing range, and false detection of the sensor [2,11]. Even after the constraint of gold, it has been chosen for the proposed model since it is the most non-reactive of all metals.
A slight coating of titanium dioxide (TiO$_2$) or graphene layer is used to limit the oxidation. Moreover, titanium nitride or indium tin oxide can also be used as an alternate material. Although this additional metallic layer increases not only fabrication complexity but also manufacturing cost [14]. Researchers all over the world nowadays are doing investigations to invent the most efficient PCF-based SPR sensors [15,16].
Paul et al. suggested a work [2] in which an air-core PCF-based SPR biosensor has been studied. This sensor demonstrated maximum amplitude sensitivity (AS), wavelength sensitivity (WS), and wavelength resolution (WR) as 159.7 RIU$^{-1}$, 11,700 nm/RIU and $8.55{\times }10^{-6}$ RIU respectively for the range of RI from 1.42 to 1.53. The sensing range of this paper is comparatively very low. Another RI sensor was analyzed by Guo et al. in which a single mode PCF-based SPR sensor was designed [6]. It offered WS (nm/RIU) of 1931.03 for the range of RI from 1.35 to 1.46, but no information was provided for AS. Al Mahfuz et al. proposed a hexagonal lattice structure containing circular air holes with gold and silver as PM [10]. For this structure, WS and WR were the same for both the materials and they were 12000 nm/RIU and $8.33{\times }{10}^{-6}$ RIU respectively. However, change was noticed in the case of obtaining maximum AS. In y-polarized mode, it was 1656 RIU$^{-1}$ for silver and for gold it decreased to 1086 RIU$^{-1}$. For both of these materials, it showed a lower range of AS. This type of investigation was also done by Han et al. in [17] with H-shaped PCF-based SPR sensor. However, only gold was used as PM and WS was obtained as 25,900 nm/RIU in this case. Another sensor with an external sensing mechanism was demonstrated by Li et al. with a gold-coated fiber. This work gave the highest AS and WS of 641 RIU$^{-1}$ and 11,000 nm/RIU respectively [18]. Other researchers had also investigated some sensors with external metal deposition. Haider et al. proposed an alphabetic-core based on PCF-SPR sensor, which showed 12,000 nm/RIU as WS in the analyte range of 1.33 to 1.40 [19]. Another PCF-SPR sensor was designed by Islam et al., and it obtained a maximum WS of 62000 nm/RIU invisible to the near-infrared spectral region [3]. However, there was a possibility of showing huge surface roughness that restrains the sensing performance because of the metal-covered outer surface. D-shaped PCF can be considered as a solution for this case with a polished flat surface for metal deposition. A D-shaped sensor was suggested with depositing an ITO film by Liu et al. with a spectral sensitivity of 15,000 nm/RIU in the range of 1.32 and 1.33 RI [20]. However, D-shaped PCF also enhances complications because it requires perfect polishing after fabrication. In addition, polishing also makes a PCF weaker and more fragile.
In this article, a PCF biosensor model with four symmetrical quadrants has been proposed based on the SPR phenomenon. The performance of the designed sensor is studied numerically through the finite element method (FEM). Various structural parameters like the thickness of gold and TiO$_2$ layer, radius, and length of the channel are optimized to obtain the best performance. In this paper, a simplified, feasible, and realistic biosensor has been studied with supreme sensing responses for sensing applications.
2. Structural design and numerical analysis
The geometrical view of the proposed sensor is illustrated in Fig. 1(a). The channel of the circular sensor is divided into four symmetrical quadrants with an equal radius. A circle surrounds these quadrants; the radius of this circle is 4 $\mathrm{\mu}$m. The designed circle is based on fused silica (SiO$_2$) and it is used as background material so that maximum performance can be obtained. The following equation is used to calculate the RI of SiO$_2$ [4]:
where $\lambda$ = wavelength ($\mathrm{\mu}$m), $\eta (\lambda )$ = effective RI of SiO$_2$ at $\lambda$. Again, P$_1$ = 0.696263, Q$_1$ = 0.0046914826 $\mathrm{\mu}\textrm {m}^2$, P$_2$ = 0.4079426, Q$_2$ = 0.0135120631 $\mathrm{\mu}\textrm {m}^2$, P$_3$ = 0.8974794, and Q$_3$ = 97.9340025 $\mathrm{\mu}\textrm {m}^2$.
Fig. 1. (a) Schematic-view of the proposed structure and (b) to (c) Development of proposed structure with the coating of gold and titanium dioxide layer (d) 3D view of the proposed sensor and (e) Block diagram showing the experimental sensing of the designed sensor.
The thickness of the solid fiber circle is 0.45 $\mathrm{\mu}$m which is denoted by d. The gap between two adjacent quadrants is 1.1 $\mathrm{\mu}$m which is the width of the channel and represented by w. Four arcs of gold and TiO$_2$ are placed on the top of the PCF structure to support SPR phenomenon. The adherence of gold depends on TiO$_2$, which is used in a thinned layer between the gold layer and silica glass. The dielectric constant of gold can be determined by Drude–Lorenz model using the following equation [3]:
As metal gratings help in localizing the plasmons hence it has been used here instead of using gold in the whole outer surface. An electromagnetic field is produced as well as metal surface electrons are excited as a result of propagating an axial ray in the down of the fiber.
Again, TiO$_2$ has a high value of RI, and it helps to generate a large number of surface electrons. Possessing a higher RI produces a huge number of electrons on the surface and creates a strong wave. This wave is known as the evanescent wave, and it helps to be attracted with the fields from the core so that interaction with SPP mode is possible. The following equation helps in determining the RI of TiO$_2$ [21]:
where $\eta _t$ represents the RI of TiO$_2$, A = 5.913, B = ${2.441\times {10}^{7}}$, C = ${0.803 \times {10}^{7}}$ and $\lambda$ is in Angstroms.The whole structure is surrounded by a perfectly matched layer (PML), and the analyte is placed in between PML and SiO$_2$ circle containing sensing layer. PML is used due to its absorbing nature that helps to study light propagation. As it absorbs radiated energy and makes the computation region finite, it is suitable for the application. PML thickness has no effects on evaluating sensor performance as the optical properties of both PML dielectric are similar. The optimized parameters are tabulated in Table 1. t$_{au}$ and t represent the thickness of the gold and TiO$_2$ layer, respectively. w depicts the width of the channel, and r represents the radius of the inner circle.
Table 1. Optimized parameters of a designed sensor
Though a lot of internally and externally metal-coated PCF sensors have been demonstrated, indeed, it is a tough task to achieve uniform metal film in practice. However, there are many physical methods that are frequently used for metal coating either internally or externally. For example, RF sputtering [22], thermal evaporation [23] and vacuum evaporation [24] are some physical methods to achieve this metallic coating. Also, various deposition techniques such as wet-chemistry deposition [25], atomic vapor deposition (AVD) [26] and chemical vapor deposition (CVD) [27] are used to metallic coating with high pressure. Among them, CVD possesses standard accuracy in the case of metal-coated PCF [28]. Again, the internal coating is very complex and hard to achieve than the external one. On the other hand, metal coating on the whole outer circular surface creates huge surface roughness [28]. Hence, metal coating only a portion instead of the entire surface could be effectively reducing the surface roughness. Klantsataya et al. has demonstrated an exposed core PCF where the electroless plating process is used successfully to fabricate thin metallic film [29]. A D-shaped PCF-based SPR sensor was experimentally studied where the metal coating on a flat plane had been accomplished by magnetron sputtering device [30]. Another experimental work has been demonstrated by Wei et al. where authors have coated both flat planes and split surfaces of the air holes with different and independent polishing depth and angle [31]. In the above-mentioned work, the metal coating process has been successfully carried out, and the authors are hopeful that the metal coating of the proposed sensor can be possible by any of the existing advanced metal coating processes.
The practical sensing method of the proposed sensor is shown by the block diagram in Fig. 1(e). After transmitting the optical power through the polarizer, it is launched to the single-mode fiber (SMF) to obtain linearly polarized light. Cleaning and filling the analyte channel can be maintained with the help of the INLET and OUTLET ports situated at the outer surface using pumping [32]. Due to the variation of analyte, RI of the test sample changes and as a result the resonance peak shifts from the previous one. In Fig. 1(e), the blue and red resonance peaks indicate the peak variation to a longer and shorter wavelength due to different RI sample. This process of shifting the resonance peak happens when the analyte molecules are attached to the ligand. Output power can be measured, and unknown analytes can be identified by the optical spectrum analyzer (OSA). The output spectrum is observed by the OSA, which is connected to a computer.
The confinement loss (CL) ($\alpha$) can be calculated by [33]:
Using the wavelength interrogation mode, WS can be measured by using the following equation: [34]:
Sensor length is another crucial parameter that can also be considered as an alternative way to measure CL. It can be calculated by following equation [3]:
From the equation, it is notable that the length of the sensor is only dependent on the CL. In addition, some other parameters; for example, full wave half maxima (FWHM), figure of merit (FOM), detection limit (DL), detection accuracy (DA), signal to noise ratio (SNR) have been calculated for this sensor using the following equation [34] [35]:
3. Effect of design parameters and comparative performance scrutiny
3.1 Thickness variation of gold
In this subsection, the effect of $t_{au}$ will be discussed on the overall performance of the proposed sensor. All other parameters are assumed as constant except RI. All parameters like CL, AS, WS, FWHM, FOM, etc., have been calculated through the equations mentioned above using the simulated data, and the summary of the calculation is shown in Table 2. Figure 2(a) has described how CL is changed with different wavelengths for different $t_{au}$. It can be seen from Fig. 2(a) that if the RI is increased, the peak of the CL has also been increased, and resonance wavelength has been shifted from left to right. For the change of RI from 1.33 to 1.34, CL has been increased into 0.0342, 0.0376, 0.0354, 0.0264 dB/cm for $t_{au}$ (nm) of 25, 30, 35, and 40 respectively. Again, Fig. 2(b) shows that the peak of the AS has been shifted from left to right when thickness has been increased. Also, a gradual rise is observed in the peak values of AS till 35 nm. WS has been increased by 1000 nm/RIU, and a gradual rise is also observed for FOM as tabulated in Table 2. If the data are monitored closely, it can be seen that when the thickness is 35 nm, the loss has decreased after a continuous rise. AS also has given its maximum output when $t_{au}$ is 35 nm. Therefore, considering all other values, 35 nm is regarded as optimum value for $t_{au}$.
Fig. 2. (a), (c) Confinement loss and (b), (d) Amplitude sensitivity for different thickness of gold layer $(t_{au})$ and titanium layer $(t)$ respectively.
Table 2. Optimization of design parameters with numerical records
3.2 Thickness variation of titanium dioxide
TiO$_2$ is used in addition with the gold as PM on the curved layer of this sensor. $t$ has been varied, keeping other parameters constant, excluding RI. The variation of CL and AS is described in Fig. 2(c) and 2(d) respectively. The behavioral change is investigated for 10, 15, and 20 nm thickness. CL and AS are observed for the different values of $t$ while RI is 1.33 and 1.34. It is seen from Fig. 2(c) that when RI turns into 1.34 from 1.33, the magnitude of the peak CL has been increased dramatically. The rise of the peak value of the CL is respectively for 10, 15, and 20 nm. All analyzed parameters are summarized in Table 2. AS (1/RIU) has been increased from −73.484 to −75.339 for the thickness of 10 and 15 nm, but then it has decreased to −75.323 when the thickness is 20 nm. A gradual decline has also been noticed in the case of FWHM. Therefore, 15 nm is considered the optimum value for $t$.
3.3 Variation of fiber diameter
In this subsection, a change of the fiber diameter has been observed for the proposed sensor keeping other parameters constant. Then the performance has been studied numerically with the variation of the RI. Figure 3(a) shows that CL has been decreased gradually with the increase of this radius and it is inversely proportional to the CL. The peak value of the AS (1/RIU) has been found as −582.36, −591.94, and −583.76, respectively, for the radius of the fiber of 0.4, 0.45, and 0.5 $\mathrm{\mu}$m as represented in Fig. 3(b). With the increase of the radius, FWHM also has been increased, but a gradual downward is noticed in the values of FOM. No change is seen in WS as the peak value of CL is very slightly changed. Also, the radius of this fiber has a minimal influence on shifting resonance for this sensor. All of these values are tabulated in Table 2. Analyzing the data, 0.45 $\mathrm{\mu}$m is considered the optimum value for the radius of the fiber.
3.4 Width variation of channel
Keeping $t_{au}$ and $t$ in optimum value and thickness of the first solid circle as 0.45 $\mathrm{\mu}$m, width of the channel (w) has been changed to observe the performance variation of this sensor. All other parameters were remained constant except RI and w. Figure 4(a) describes the behavioral change of CL due to the variation of w. Fig. 4(a) represents the proportional change of CL with w, and with the reduced width of the channel, the value of CL is also decreased. In the case of AS (1/RIU), as shown in Fig. 4(b), a dramatic increase has been observed with the increased value of the width of the channel. Again, with the increase of the width of the channel, the value of FWHM has been decreased, but the peak value of FOM has been obtained for the value of 1.1 $\mathrm{\mu}$m. Also, the peak of the AS has been found for w = 1.1 $\mathrm{\mu}$m. Therefore, w is taken as 1.1 $\mathrm{\mu}$m for maintaining a balance among all performance parameters.
Fig. 4. (a) Confinement loss and (b) Amplitude sensitivity with the variation of the width variation of the channel.
4. Numerical analysis of the sensor performance
Figure 5(a) shows the basic dispersion relation as a function of wavelength with RI = 1.38. Figure 5(b) and 5(c) show the SPP mode and core guided mode respectively for field distribution. Here, in the primary axis, the real part of the effective index of the core and SPP mode is plotted, and in the secondary axis, the CL of the core mode is plotted. Pink and violate lines indicate the real value of the SPP mode and core mode, respectively. This figure shows that CL for both y- and x -polarized mode is the same, which is represented by solid red lines and marked lines, respectively. Hence, for other calculations, values of y-polarized mode have been considered. The curve of CL starts to increase, and at a certain wavelength, it reaches its maximum point. After that, it starts to decrease again. This corresponding wavelength in which maximum loss occurs is known as resonance wavelength and phase matching occurs at this wavelength by transferring maximum energy. It shows the maximum CL as 2.54 dB/cm when RI is 1.38 at a wavelength of 0.8 $\mathrm{\mu}$m. At 0.8 $\mathrm{\mu}$m, the value of core mode and the real value of the effective index for SPP mode is same as well as maximum loss is found at this point.
Fig. 5. (a) Dispersion Relation and Optical Field Distribution for the (b) SPP Mode, and (c) Core Mode
4.1 Confinement loss
The geometrical structure as well as the leaky nature of the modes, causes CL in a PCF. If modes are kept in a little place like core, it creates the partial circulation of the modes out of the core of the fiber. CL is produced by this leakage occurrence. The CL is dependent on the structure because it can be influenced by the numerical value and dimension of air holes and wavelength for a fiber. To observe the changing pattern of CL, the value of RI is increased from 1.28 to 1.40 with an increment of 0.01.
From close observation of Fig. 6, it can be stated that with the increased value of RI, the magnitude of the peak of CL is rising, and the resonance wavelength is shifting from left to right. It gives the minimum and maximum peak of the CL, which are 0.101 dB/cm and 135.372 dB/cm at RI of 1.28 and 1.40, respectively. All the details values are tabulated in Table 3 and the average value of CL is 11.439 dB/cm.
Fig. 6. Confinement loss spectrum of the proposed sensor for core mode with analytes of different RI from 1.28 to 1.40.
Table 3. Performance analysis of the proposed sensor
4.2 Wavelength sensitivity
The variations in RI are measured by detecting the shift of the resonance peak. Figure 6 represents the CL curve with their operating wavelength for different values of RI. All the necessary data are summarized in Table 3. This table represents the shifting of the value of the resonant wavelength, which is 10, 10, 10, 10, 20, 20, 20, 30, 40, 70, 130, and 870 nm, respectively, for the values of RI from 1.28 to 1.40. It gives the minimum WS (nm/RIU) of 1000 and maximum WS as 87000. The average value of WS is 10333.33 (nm/RIU).
4.3 Amplitude sensitivity
Figure 7 presents maximum values of AS (${RIU}^{-1}$) in the negative axis. From the given figure, it can be described that the peak of the AS is increasing as well as it is making a right-shift with the increased value of RI. Peak values of the AS in the negative axis are 32.817, 39.445, 47.840, 58.385, 75.095, 98.002, 134.556, 193.8, 288.299, 524.32, 1118.12 and 7420.69 as tabulated in Table 3. These values are obtained for the values of RI from 1.28, 1.29, 1.30, 1.31, 1.32, 1.33, 1.34, 1.35, 1.36, 1.37, 1.38, 1.39, and 1.40 respectively. The peak value of the AS is 7420.69 (${RIU}^{-1}$) and the average value of AS is 835.9476 (${RIU}^{-1}$).
Fig. 7. Amplitude Sensitivity Amplitude sensitivity with respect to wavelength for the value of RI from 1.28 to 1.40.
4.4 Sensor resolution and sensor length
Sensor resolution can be defined as that parameter that is used to measure the minimum change of RI that can be identified by the sensor. It is necessary to find out the slightest change of RI accurately. Sensor resolution also indicates the resonant wavelength peak shift when a change in the analyte’s RI has occurred.
The curve fitting is represented in Fig. 8 as a function of RI. Here in primary axis the sensor length is plotted and in secondary axis CL is plotted. Figure 8 is plotted by using the given polynomial equation y = 2E+08x6 - 1E+09x5 + 5E+09x4 - 8E+09x3 + 8E+09x2 - 4E+09x + 8E+08. High linearity R$^2$ = 0.9049 indicates that the curve is fitted almost perfectly. Figure 8 also depicts the inverse relationship between length and the loss as when the length is on rising, the value of the loss is on a downtrend. From Table 3, the highest value of the length of the sensor can also be obtained which is 9.83 cm and the lowest length is 0.007 cm for the value of RI is 1.28 and 1.40 respectively.
4.5 FOM and SNR
In spite of having a proportional relation with FOM and WS, it varies inversely with the FWHM. With the increased value of RI, the value of FOM also increases. The maximum value of FOM has been obtained as 1011.63 (${RIU}^{-1}$) and the lowest value of FOM is 11.26 (${RIU}^{-1}$). The average value of FOM can be calculated as 130.77 (${RIU}^{-1}$). It is seen from the data that initially, the rate of increase of the values of FOM was quite slow but after a certain period when the value of RI is 1.34, the rate of increase of the values of FOM is noteworthy. Another important parameter is SNR which is also directly proportional to RI. The minimum value of SNR is −9.48 dB and the maximum value of SNR is 10.05 dB. All the details values are tabulated in Table 3.
4.6 DL and DA
By definition, DL is directly proportional to FWHM, and values of DL increase with the rise of RI. Maximum DL is obtained as 102.23 nm for RI = 1.28 and minimum DL is 32.14 nm for RI = 1.39. The average value of DL is 64.85 nm. DA is the other parameter that is inversely proportional to FWHM and RI. The maximum value of DA has been obtained as 0.01125 (${nm}^{-1}$) and the minimum value of DA is 0.006623 (${nm}^{-1}$).
From the above numerical study, the sensing region of our proposed sensor is in between 0.50 $\mathrm{\mu}$m to 2.1 $\mathrm{\mu}$m and the detection range of RI of our proposed sensor is from 1.28 to 1.40 for possible real time applications. There are a number of biological samples and biochemical solutions which have a RI within the detection range (1.28 to 1.40). For example various cancerous cells (1.36 to 1.40), glucose solution (1.3635), human liver (1.369), sucrose solution up to 40 percent (1.3330 to 1.3999), acetone (1.36), ethanol (1.361), human urine (1.3415 to 1.3464) and intestinal mucosa (1.329 to 1.338) [4,39] etc. So we can say that our proposed biosensor is a good candidate to detect different biological samples and biochemical solutions.
A comparative performance analysis of our proposed sensor is given in Table 4 with some relative existing works considering WS, AS, WR, FWHM, FOM and SNR. From this table, maximum AS of 116,000 nm/RIU is obtained in [38]. In maximum reported works, FWHM, FOM, and SNR is not measured except ref. [34]. Our proposed sensor shows a better performance in case of all performance parameters.
Table 4. Relative performance analysis of the suggested sensor with the detection range of RI
5. Conclusions
In this paper, a PCF biosensor sensor with four symmetrical quadrants has been considered probably for the first time. Based on the numerical results, it is seen that the proposed sensor can be used to detect analytes with a wide range RI that is from 1.28 to 1.40. The peak of the WS is 87,000 nm/RIU and AS is −7420.69 ${RIU}^{-1}$ for an analyte with RI of 1.39 for y- polarized mode. Chemically inactive gold and TiO$_2$ coating have been deployed in order to maximize its sensing applications with economically considerable. Moreover, the four-quadrant section makes it feasible to manufacture for different applications in biochemical industries.
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.
References
1. A. P. Turner, “Biosensors: Fundamentals and applications–historic book now open access,” Biosens. Bioelectron. 65, A1 (2015). [CrossRef]
2. A. K. Paul, M. S. Habib, N. H. Hai, and S. A. Razzak, “An air-core photonic crystal fiber based plasmonic sensor for high refractive index sensing,” Opt. Commun. 464, 125556 (2020). [CrossRef]
3. M. S. Islam, J. Sultana, A. A. Rifat, R. Ahmed, A. Dinovitser, B. W.-H. Ng, H. Ebendorff-Heidepriem, and D. Abbott, “Dual-polarized highly sensitive plasmonic sensor in the visible to near-ir spectrum,” Opt. Express 26(23), 30347–30361 (2018). [CrossRef]
4. M. A. Mollah, M. Yousufali, I. M. Ankan, M. M. Rahman, H. Sarker, and K. Chakrabarti, “Twin core photonic crystal fiber refractive index sensor for early detection of blood cancer,” Sens. Bio-Sensing Res. 29, 100344 (2020). [CrossRef]
5. M. A. Jabin, K. Ahmed, M. J. Rana, B. K. Paul, M. Islam, D. Vigneswaran, and M. S. Uddin, “Surface plasmon resonance based titanium coated biosensor for cancer cell detection,” IEEE Photonics J. 11(4), 1–10 (2019). [CrossRef]
6. Z. Guo, Z. Fan, X. Kong, and Z. Meng, “Photonic crystal fiber based wide-range of refractive index sensor with phase matching between core mode and metal defect mode,” Opt. Commun. 461, 125233 (2020). [CrossRef]
7. A. Nooke, U. Beck, A. Hertwig, A. Krause, H. Krüger, V. Lohse, D. Negendank, and J. Steinbach, “On the application of gold based spr sensors for the detection of hazardous gases,” Sensors Actuators B: Chem. 149(1), 194–198 (2010). [CrossRef]
8. K. Ahmed, M. J. Haque, M. A. Jabin, B. K. Paul, I. S. Amiri, and P. Yupapin, “Tetra-core surface plasmon resonance based biosensor for alcohol sensing,” Phys. B 570, 48–52 (2019). [CrossRef]
9. A. K. Paul, M. A. Mollah, M. Z. Hassan, N. Gomez-Cardona, and E. Reyes-Vera, “Graphene-coated highly sensitive photonic crystal fiber surface plasmon resonance sensor for aqueous solution: Design and numerical analysis,” Photonics 8(5), 155 (2021). [CrossRef]
10. M. Al Mahfuz, M. A. Mollah, M. R. Momota, A. K. Paul, A. Masud, S. Akter, and M. R. Hasan, “Highly sensitive photonic crystal fiber plasmonic biosensor: Design and analysis,” Opt. Mater. 90, 315–321 (2019). [CrossRef]
11. A. A. Rifat, R. Ahmed, A. K. Yetisen, H. Butt, A. Sabouri, G. A. Mahdiraji, S. H. Yun, and F. M. Adikan, “Photonic crystal fiber based plasmonic sensors,” Sens. Actuators, B 243, 311–325 (2017). [CrossRef]
12. H. Heidarzadeh, “Analysis and simulation of a plasmonic biosensor for hemoglobin concentration detection using noble metal nano-particles resonances,” Opt. Commun. 459, 124940 (2020). [CrossRef]
13. T. Wieduwilt, A. Tuniz, S. Linzen, S. Goerke, J. Dellith, U. Hübner, and M. A. Schmidt, “Ultrathin niobium nanofilms on fiber optical tapers–a new route towards low-loss hybrid plasmonic modes,” Sci. Rep. 5(1), 17060 (2015). [CrossRef]
14. M. B. Hossain, S. R. Islam, K. T. Hossain, L. F. Abdulrazak, M. N. Sakib, and I. Amiri, “High sensitivity hollow core circular shaped pcf surface plasmonic biosensor employing silver coat: A numerical design and analysis with external sensing approach,” Results Phys. 16, 102909 (2020). [CrossRef]
15. S. Islam, “Design and analysis of advanced photonic devices for electromagnetic transmission and sensing,” Ph.D. thesis, Faculty of Engineering, University of Adelaide, Australia, (2021).
16. F. Haider, M. Mashrafi, R. A. Aoni, R. Haider, M. Hossen, T. Ahmed, G. A. Mahdiraji, and R. Ahmed, “Multi-analyte detection based on integrated internal and external sensing approach,” IEEE Trans. on NanoBiosci. 21(1), 29–36 (2022). [CrossRef]
17. H. Han, D. Hou, L. Zhao, N. Luan, L. Song, Z. Liu, Y. Lian, J. Liu, and Y. Hu, “A large detection-range plasmonic sensor based on an h-shaped photonic crystal fiber,” Sensors 20(4), 1009 (2020). [CrossRef]
18. X. Li, S. Li, X. Yan, D. Sun, Z. Liu, and T. Cheng, “High sensitivity photonic crystal fiber refractive index sensor with gold coated externally based on surface plasmon resonance,” Micromachines 9(12), 640 (2018). [CrossRef]
19. F. Haider, R. A. Aoni, R. Ahmed, W. J. Chew, and G. A. Mahdiraji, “Alphabetic-core assisted microstructure fiber based plasmonic biosensor,” Plasmonics 15(6), 1949–1958 (2020). [CrossRef]
20. C. Liu, J. Wang, F. Wang, W. Su, L. Yang, J. Lv, G. Fu, X. Li, Q. Liu, T. Sun, and P. K. Chu, “Surface plasmon resonance (spr) infrared sensor based on d-shape photonic crystal fibers with ito coatings,” Opt. Commun. 464, 125496 (2020). [CrossRef]
21. M. S. Islam, C. M. Cordeiro, J. Sultana, R. A. Aoni, S. Feng, R. Ahmed, M. Dorraki, A. Dinovitser, B. W.-H. Ng, and D. Abbott, “A hi-bi ultra-sensitive surface plasmon resonance fiber sensor,” IEEE access 7, 79085–79094 (2019). [CrossRef]
22. L. Armelao, D. Barreca, G. Bottaro, G. Bruno, A. Gasparotto, M. Losurdo, and E. Tondello, “Rf-sputtering of gold on silica surfaces: evolution from clusters to continuous films,” Mater. Sci. Eng. C 25(5-8), 599–603 (2005). [CrossRef]
23. M. C. Barnes, D.-Y. Kim, H. S. Ahn, C. O. Lee, and N. M. Hwang, “Deposition mechanism of gold by thermal evaporation: approach by charged cluster model,” J. Cryst. Growth 213(1-2), 83–92 (2000). [CrossRef]
24. R. Klenk, T. Walter, H.-W. Schock, and D. Cahen, “A model for the successful growth of polycrystalline films of cuinse2 by multisource physical vacuum evaporation,” Adv. Mater. 5(2), 114–119 (1993). [CrossRef]
25. J. A. Sioss and C. D. Keating, “Batch preparation of linear au and ag nanoparticle chains via wet chemistry,” Nano Lett. 5(9), 1779–1783 (2005). [CrossRef]
26. M. B. Griffiths, P. J. Pallister, D. J. Mandia, and S. T. Barry, “Atomic layer deposition of gold metal,” Chem. Mater. 28(1), 44–46 (2016). [CrossRef]
27. P. J. Sazio, A. Amezcua-Correa, C. E. Finlayson, J. R. Hayes, T. J. Scheidemantel, N. F. Baril, B. R. Jackson, D.-J. Won, F. Zhang, E. R. Margine, V. Gopalan, V. H. Crespi, and J. V. Badding, “Microstructured optical fibers as high-pressure microfluidic reactors,” Science 311(5767), 1583–1586 (2006). [CrossRef]
28. A. A. Rifat, R. Hasan, R. Ahmed, and A. E. Miroshnichenko, “Microstructured optical fiber-based plasmonic sensors,” in Computational Photonic Sensors, (Springer, 2019), pp. 203–232.
29. E. Klantsataya, A. François, H. Ebendorff-Heidepriem, P. Hoffmann, and T. M. Monro, “Surface plasmon scattering in exposed core optical fiber for enhanced resolution refractive index sensing,” Sensors 15(10), 25090–25102 (2015). [CrossRef]
30. Y. Chen, Q. Xie, X. Li, H. Zhou, X. Hong, and Y. Geng, “Experimental realization of d-shaped photonic crystal fiber spr sensor,” J. Phys. D: Appl. Phys. 50(2), 025101 (2017). [CrossRef]
31. W. Luo, X. Li, J. Meng, Y. Wang, and X. Hong, “Surface plasmon resonance sensor based on side-polished d-shaped photonic crystal fiber with split cladding air holes,” IEEE Trans. Instrum. Meas. 70, 1–11 (2021). [CrossRef]
32. M. A. Mollah and M. S. Islam, “Novel single hole exposed-suspended core localized surface plasmon resonance sensor,” IEEE Sens. J. 21(3), 2813–2820 (2020). [CrossRef]
33. A. K. Paul, A. K. Sarkar, and A. Khaleque, “Dual-core photonic crystal fiber plasmonic refractive index sensor: a numerical analysis,” Photonic Sens. 9(2), 151–161 (2019). [CrossRef]
34. N. Jahan, M. M. Rahman, M. Ahsan, M. A. Based, M. M. Rana, S. Gurusamy, and J. Haider, “Photonic crystal fiber based biosensor for pseudomonas bacteria detection: A simulation study,” IEEE Access 9, 42206–42215 (2021). [CrossRef]
35. M. A. Mollah, H. Sarker, M. Ahsan, M. T. Elahi, M. A. Based, J. Haider, and S. Palani, “Designing highly sensitive surface plasmon resonance sensor with dual analyte channels,” IEEE Access (2021).
36. S. A. S. Sunny, T. Ahmed, S. M. Hiam, and A. K. Paul, “Highly sensitive externally metal coated plasmonic refractive index sensor based on photonic crystal fiber,” Optik 243, 167482 (2021). [CrossRef]
37. J. Dong, Y. Zhang, Y. Wang, F. Yang, S. Hu, Y. Chen, W. Zhu, W. Qiu, H. Guan, H. Lu, J. Yu, Y. Zhong, J. Zhang, Y. Luo, and Z. Chen, “Side-polished few-mode fiber based surface plasmon resonance biosensor,” Opt. Express 27(8), 11348–11360 (2019). [CrossRef]
38. E. Haque, S. Mahmuda, M. A. Hossain, N. H. Hai, Y. Namihira, and F. Ahmed, “Highly sensitive dual-core pcf based plasmonic refractive index sensor for low refractive index detection,” IEEE Photonics J. 11(5), 1–9 (2019). [CrossRef]
39. M. Al Mahfuz, M. A. Hossain, E. Haque, N. H. Hai, Y. Namihira, and F. Ahmed, “Dual-core photonic crystal fiber-based plasmonic ri sensor in the visible to near-ir operating band,” IEEE Sens. J. 20(14), 7692–7700 (2020). [CrossRef]
