A Novel Plasmonic Sensor Based on Dual-Channel D-Shaped Photonic Crystal Fiber for Enhanced Sensitivity in Simultaneous Detection of Different Analytes

A dual-channel D-shaped photonic crystal fiber (PCF) based plasmonic sensor is proposed in this paper for the simultaneous detection of two different analytes using the surface plasmon resonance (SPR) technique. The sensor employs a 50 nm-thick layer of chemically stable gold on both cleaved surfaces of the PCF to induce the SPR effect. This configuration offers superior sensitivity and rapid response, making it highly effective for sensing applications. Numerical investigations are conducted using the finite element method (FEM). After optimizing the structural parameters, the sensor exhibits a maximum wavelength sensitivity of 10000 nm/RIU and an amplitude sensitivity of −216 RIU<inline-formula> <tex-math notation="LaTeX">$^{-{1}}$ </tex-math></inline-formula> between the two channels. Additionally, each channel of the sensor exhibits its unique maximal wavelength and amplitude sensitivities for different refractive index (RI) ranges. Both channels demonstrate a maximal wavelength sensitivity of 6000 nm/RIU. In the RI range of 1.31-1.41, Channel 1 (Ch1) and Channel 2 (Ch2) achieved their maximum amplitude sensitivities of −85.39RIU<inline-formula> <tex-math notation="LaTeX">$^{-{1}}$ </tex-math></inline-formula> and -304.52 RIU<inline-formula> <tex-math notation="LaTeX">$^{-{1}}$ </tex-math></inline-formula>, respectively, with a resolution of <inline-formula> <tex-math notation="LaTeX">${5}\times {10} ^{-{5}}$ </tex-math></inline-formula>. This sensor structure is noteworthy for its ability to measure both amplitude and wavelength sensitivity, providing enhanced performance characteristics suitable for various sensing purposes in chemical, biomedical, and industrial fields.

A Novel Plasmonic Sensor Based on Dual-Channel D-Shaped Photonic Crystal Fiber for Enhanced Sensitivity in Simultaneous Detection of Different Analytes close proximity to a metal surface [1].Due to its broad sensing range, simple light feeding, and control over propagation, the SPR-based sensing technique is more feasible than other sensing techniques [2].Kretschmann et al. first proposed a prism-based SPR technique to measure variations in the RI of surrounding medium(analyte).This method utilizes thin metal layers placed between a glass prism and the analyte to induce the plasmon effect.The commonly used plasmonic metals in this technique are gold and silver [3].Complete reflection happens when plane-polarized light hits a glass prism beyond a critical angle.At a specific angle and wavelength, an evanescent field is generated between the glass and metal layer, which is known as the resonance condition.At resonance condition, the incident light penetrates the metal surface, generating an evanescent field.The incident light excites free electrons in the metal-dielectric interface, leading to the formation of surface plasmons.As a result, the reflected light intensity decays exponentially, which is known as the SPR effect [4].
The propagation characteristics of surface plasmons are extremely sensitive to the RI of the analyte near the metal surface [5].If any minute RI changes in the analyte cause deviations in the phase or amplitude of the reflected light.Hence RI variation of the analyte near the metal surface, can be easily detected by using this SPR method.SPR sensors have diverse uses in sensing, such as chemical sensing [6], biosensing [7], environmental monitoring [8], and biochemical application [9].The main drawback of this method is it is not portable because of its bulky size, so it cannot be used in remote sensing applications, and the cost of this sensor is also high [10].
In order to address these problems, SPR sensors for remote sensing applications use optical fibers instead of prisms due to their compact size, exceptional sensitivity, and superior accuracy [11].The downside of this sensor is the metallization of the fiber structure.In order to accomplish the metallization, the cladding portion of conventional fibers should be polished almost to the core.It degrades the structure's integrity and results in mechanical breakdowns.It also increases the cost of fabrication.Hassani et al. proposed an initial SPR sensor using PCF as a remedy for these concerns, as metals are coated or filled in a specific air hole in PCF structures [12].PCFs have gained significant attention in the past few decades due to their unique optical properties, including high birefringence, low transmission loss, strong non-linearity, endless single mode, and the ability to control light propagation [13].These properties make PCFs ideal for various sensing applications, including chemical sensing [14], gas sensing [15], temperature sensing [16], and biosensing [17].In recent years, there has been a significant research focus on utilizing the SPR technique in combination with PCF for sensor applications due to its unique advantages.
The core mode and plasmon mode coupling at the metal-dielectric interface is the underlying principle of a SPR-based PCF sensor [18].This coupling creates leaky modes in the guided core mode, happening at a certain wavelength when the effective indices(n eff ) of both modes are almost the same, causing the plasmon mode to receive the majority of energy from the core mode, leading to the phase matching condition that generates the resonance wavelength [19].The RI of the analyte varies near the metal layer causing the resonant wavelength shift.This wavelength shift scheme provides higher sensitivity and accuracy.Depending on where the analyte is positioned within the sensor structure, sensing techniques are categorized as either internal sensing [20] or external sensing [21].Yang et al. created a PCF sensor utilizing SPR method with a gold-graphene configuration, where gold acted as the plasmonic substance and graphene was incorporated to enhance sensor efficiency.By situating the analyte exterior to the fiber structure, the sensor obtained a wavelength sensitivity of 4200 nm/RIU and an amplitude sensitivity of 450 RIU −1 , within a detection range of 1.32-1.41[22].Rahaman et al. designed a SPR sensor utilizing a PCF to estimate the glucose level in urine samples.A thin layer of gold was coated in the cladding region of the fiber to generate SPR effect, and the sensor was plunged in the sample for measurements.The SPR sensor's maximal achieved amplitude, and wavelength sensitivity were 152 RIU −1 and 2500 nm/RIU, respectively [23].
All of these sensors were developed to analyze one analyte at a time, which requires more time and resources.The multichannel/multi-analyte sensing approach has the advantage of simultaneously detecting multiple analytes in a same sensor structure, which minimizes the need for analyte filling, emptying, and cleaning stages and thus reduces time and cost.Zhang et al. developed the first-ever multi-channel micro structured fiber-based SPR sensor, which can simultaneously detect multiple analytes, thereby resolving the limitation of single-channel sensors.The sensor exhibited outstanding wavelength sensitivity of 1535 nm/RIU at a detection range of 1.33-1.36[24].PCF-based multi-channel SPR sensors have gained popularity among researchers for multi-analyte sensing applications in recent times.In 2015, Otupiri et al. introduced a multi-channel PCF sensor utilizing SPR method that combined circular and elliptical air holes in the cladding region of the PCF to attain great birefringence.The sensor demonstrated high wavelength sensitivity of 4600 nm/RIU for Ch1, and 2300 nm/RIU for Ch2 through the dual coating of gold and Ta 2 TO 5 on the inner wall of the cladding region [5].It will be challenging to fabricate both circular and elliptical air holes together.In 2019, Kaur et al. presented a dual-channel PCF-based SPR sensor that demonstrated exceptional wavelength sensitivity.Ch1 and Ch 2 displayed optimal wavelength sensitivity of 1000 nm/RIU and 3750 nm/RIU, respectively, between the RI values of 1.30-1.40.The sensor consisted of an analyte channel made up of two concentric rings coated with gold on the outside to create a plasmonic layer [25].Yasli et al. suggested a multi-channel SPR sensor in 2020, which obtained an optimal wavelength sensitivity of 2500 nm/RIU for Ch1 and 3083 nm/RIU for Ch2.The flat surface of the analyte channel was covered with a plasmonic coating made of gold and silver.Graphene was layered on top of the silver in order to safeguard it from oxidation and enhance sensitivity [26].In that year, Yasli et al. offered an PCF-based another multi channel SPR sensor with enhanced sensitivity of 4250 nm/RIU for Ch1 and 4200 nm/RIU for Ch2.The inner walls of Ch1, and Ch2 were coated with gold and silver, respectively, to act as plasmonic layers and generate plasmonic effects [27].Recently, Kamrunnahar et al. proposed a PCF based dual channel SPR sensor in 2022, where gold was coated as a plasmonic layer to induce SPR.The obtained highest sensitivity of Ch1 was 25000 nm/RIU and 3000 nm/RIU for Ch2 [28].
Moreover, in certain aforementioned studies, filling or coating metals in the inner wall of cladding is challenging, and it is observed that few channels performed unevenly with lower sensitivity.Therefore, further research is required to develop a multi-channel sensor with high sensitivity, quality factor and consistent performance across all channels.Thus, the proposed work presents a dual channel D-shape PCF sensor, utilizing the SPR technique for simultaneous detection of two different substances.The sensor structure involves removal of top and bottom circular parts of the fiber, forming a dual D-shape structure, and coating the cleaved surfaces with a slender layer of gold.FEM is utilized for fine-tuning the sensor structure for maximum sensitivity.The optimized sensor has a wavelength sensitivity of 10000 nm/RIU and an amplitude sensitivity of −216 RIU −1 between its two channels, Ch1 and Ch2.Both channels exhibit a peak wavelength sensitivity of 6000 nm/RIU, with a resolution of 5 × 10 −5 , across the detection range of 1.31 to 1.41.

II. PROPOSED SPR SENSOR STRUCTURE AND THEORY
Fig. 1(a) shows a visual depiction of a proposed SPR based dual channel D-shaped PCF sensor.The cladding section of the sensor has a hexagonal shape, and includes air holes arranged in a triangular pattern with a diameter (d) of 0.5 µm.The pitch or lattice constant ( ) between these holes is 0.8 µm.Dual core is used in this structure to analyze two different analytes at the same time and the core region is formed by removing two airhole rings in the top and one air hole ring in the bottom side.Birefringence induced by varying core size through reducing second ring of air holes by half of the initial size.This reduces the core size, as a large core permits multimode operation [29].A dual D-shape is formed by slicing off the top and bottom circular portions of the PCF, and thin plasmonic layer of gold (t Au ) with a thickness of 0.05 µm is layered on the cleaved section of the PCF to induce the SPR Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.effect.In the outer region, 1µm thick perfectly matched layer (t PML ) is used and that is ∼10% of the total diameter of this sensor i.e., 7.4µm.The analyte is placed between the gold layer and the PML, two sensing channels-Ch1 at the top and Ch2 at the bottom are used to detect two different analytes.Applying a scattering boundary condition on the exterior of the structure leads to a further reduction in the amount of reflected energy.
The proposed sensor uses silica as its core material and its RI of silica is determined by Sellmeier equation [20].
Gold is chosen as the plasmonic material due to its excellent optical characteristics, stability, and biocompatibility.The Drude model is commonly employed to predict the dispersion behavior of gold at different wavelengths.Mathematically, it can be represented as follows [19], where plasma frequency ω P = 9.06eV and damping rate P = 0.07eV.The inclusion of a polarizer helps to enhance the visibility of confinement loss for a particular polarization.By manipulating the RI of the analyte that is added to the sensor, there will be changes in the intensity or the wavelength of the confinement loss spectrum.These changes can be detected by connecting an optical spectrum analyzer (OSA) to a computer.The RI of the unknown analyte can be determined by analyzing the data received from the computer.
Numerical analyze are carried out using FEM method.For obtain precise results, a physics-controlled mesh model has been utilized to mesh the sensor's structure with extremely fine element sizes.The mesh consists of elements with a maximum element size of 0.52640 µm and a minimum element size of 0.00111 µm.The maximum element growth rate is set to 1.25, while the vertex elements have a size of 692.Ch1 is composed of 72009 total elements, including 5567 boundary elements, and has a minimum element quality of 0.4612.Ch2 comprises 84522 total elements, with 5602 boundary elements and a minimum element quality of 0.5385.The defective modes of the PCF are responsible for its convergence error, which are affected by perturbations with compact support [30].However, the convergence error can be significantly reduced by using the proposed SPR based dual channel D-shaped PCF structure.The error has been measured to be 1.5 × 10 −5 at 1.36 µm and Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.9.7 × 10 −4 at 1.46 µm, and it decreases with higher iteration numbers.
The proposed dual side polished D-shaped PCF based SPR sensor is fabricated using stack and draw method [31], the D-shape is obtained by using side polishing method [32], and the polished region is coated with a thin layer of gold using the chemical vapor deposition (CVD) method [33].Finally, an infiltration method is utilized to inject the analyte into the sensing channels [34].

III. RESULTS AND DISCUSSION
The effectiveness of the dual-channel D-shaped SPR sensor proposed in this study is closely tied to its geometric parameters, such as the diameter of the air holes, pitch size, and thickness of the gold layer.Through careful optimization of these parameters, the sensor's performance can be significantly enhanced.To evaluate the influence of each parameter, they are individually varied while maintaining the other structural parameters constant.Confinement loss is a crucial parameter because it can significantly impact the sensitivity and accuracy of the optical sensors.It measures the extent of light that is attenuated when it is propagate within the core region of the PCF [35].The confinement loss equation can be expressed as follows [19]: where Im (n eff ) -imaginary part of the effective mode index, λ -operating wavelength in µm.

A. Transmission Characteristics and Dispersion Relation of the Proposed Sensor
Transmission characteristics of the proposed SPR based D-shaped dual channel PCF sensor are investigated for the wavelength range of 1.0µm −1.8µm Figs. 2 (a, b) illustrates the relationship between the dispersion characteristics of the effective refractive index (n eff ) for Ch1 and Ch2, with respect to their real and imaginary components.
With increasing wavelength, the real part of the n eff for the core-guided mode decreases in both Ch1 and Ch2.At a specific wavelength of 1.36µm, energy is transferred to the plasmon mode from the core guided mode, leading to a slight decrease in real part of the Ch1.This suggests the initiation of the resonance condition, where the imaginary component of the n eff achieves its optimal value.Similarly, the resonance wavelength of Ch2 is at 1.46µm, where it reaches its maximum value.Ch1 allows the multi-mode propagation because of its core size is greater than Ch2's core size; as a result, Ch1's signal strength is lesser than Ch2's.As is evident from Fig. 2(a, b) the coupling efficiency of the Y-polarized modes is significantly lower than that of the X-polarized modes in both Ch1 and Ch2.As a result, for further analysis of both channels, the X-polarized mode has chosen over the Y-polarized mode.
Figs. 3(a, b) display the electric field distribution of the proposed sensor for Ch1 and Ch2, respectively.Fig. 3.a, b (i, ii) illustrates the X and Y-polarization, respectively, when the incident light is guided through the core.The coupling mode and plasmon mode are depicted in Fig. 3.a, b (iii, iv).The couple mode refers to the condition where the phase-matching criterion is fulfilled, resulting in the maximum transfer of energy from the core mode to the plasmon mode [36].

B. Optimization of Plasmonic Materials
Incorporating plasmonic materials, such as gold [23], silver [37], copper [38], aluminium [39], and titanium [40] with a PCF structure can alter the properties of the guided modes of the fiber through their interaction.This interaction can lead to an improvement in light confinement, a reduction in losses, and an enhancement of nonlinear effects, making it a promising technique for various applications.Among these materials, silver and gold exhibit favorable optical properties, low material losses, and high resonance peaks.However, gold Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply. is the preferred plasmonic material in most sensor applications due to its chemical stability, biocompatibility, sustained high sensitivity, sharp resonant peak, and excellent accuracy [41].On the other hand, silver is known to be unstable due to its susceptibility to oxidation, which can negatively impact its accuracy as a plasmonic material.Therefore, additional metal coatings are typically required to mitigate the oxidization issue, resulting in increased manufacturing costs and fabrication complexity [42].In recent years, there has been extensive research conducted on different plasmonic materials, with a particular focus on their suitability for sensing applications.Among these materials, gold has consistently emerged as the most appropriate choice, as supported by various studies [43], [44], [45], [46].One notable study conducted by Deepak kumar et al. in 2023 focused on investigating the properties of different plasmonic materials, including gold, silver, copper, aluminium, and a gold-tin alloy.The study specifically calculated the amplitude and wavelength sensitivities for each material, yielding the following results: silver exhibited a sensitivity of 1799nm/RIU, gold had a sensitivity of 1830.76nm/RIU,aluminium had a sensitivity of 1732nm/RIU, copper had a sensitivity of 1652nm/RIU, and the gold-tin alloy had a sensitivity of 1532.2nm/RIU.The study revealed that among the materials examined, gold exhibited the highest sensitivity, making it the preferred plasmonic material for many sensor applications [47].
In the proposed structure, the performance of the sensor in terms of confinement loss is being analyzed using three different metals: gold, silver, and copper.The confinement loss spectra for these plasmonic metals in relation to Ch1 and Ch2 are depicted in Figs. 4 (a,b).The results of the analysis indicate that among the three metals, gold exhibits the maximum confinement loss.This is attributed to the fact that gold provides the highest coupling between the core mode and plasmon mode, resulting in enhanced confinement loss characteristics.As a result, gold is chosen as the plasmonic metal in this proposed sensor based on its superior performance in terms of confinement loss.

C. Optimization of Geometrical Parameters
The objective of the analysis is to examine how various geometrical structural parameters, including air hole diameter (d), pitch size ( ), gold layer thickness (t Au ), and PML layer thickness (t PML ) affect the confinement loss.This wavelength-dependent measure is crucial in optimizing sensor performance.

1) Air Hole Diameter Optimization:
The sensing performance of the sensor is greatly influenced by the airhole diameter, as these air holes establish the pathway between the core and the metal surface, thereby playing a crucial role in the overall functionality.The investigation involves exploring the ideal air hole diameter by varying it within the range of 0.4µm, 0.5µm, and 0.6µm.Figs.5(a,b) depict the confinement loss spectra for Ch1 and Ch2 across different air hole diameters, plotted against the corresponding wavelengths.When larger air holes are employed, a significant portion of light is effectively guided within the core, leading to reduced interaction between the core mode and plasmon mode and consequently resulting in lower confinement loss.On the other hand, smaller air holes create a larger pathway between the core mode and plasmon mode, leading to a higher degree of confinement loss due to light leakage.However, it is important to note that, in comparison to both 0.5 µm and 0.4 µm air hole diameters, the 0.5 µm diameter air holes can generate a sharp spectrum.This sharp peak is particularly advantageous for efficient sensing in sensors.Hence, to achieve optimal sensitivity, the 0.5 µm air hole diameter is preferred.
2) Pitch Size Optimization: The pitch values are being adjusted to investigate how the pitch size affects the confinement loss characteristics while keeping the other structural parameters constant.Three different values of (0.78 µm, 0.8 µm, and 0.82 µm) are being considered for analyzing the confinement loss.The highest coupling between the core mode and plasmon mode is obtained at the pitch value of 0.8 µm, which gives a sharp peak.Figs. 6 (a, b) depicts the results of Ch1 and Ch2, respectively, illustrating the confinement loss characteristics at different pitch sizes.
3) Gold Layer Thickness Optimization: The gold layer thickness can significantly impact the plasmonic features and Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.efficacy of the sensor [41].The optimal gold layer thickness is investigated by varying it between 40nm, 50nm, and 60nm.
Figs. 7(a,b) display the confinement loss spectra for Ch1 and Ch2 at varying gold layer thicknesses, plotted against the wavelength.The graphs illustrate that increasing the thickness of the gold layer causes the resonance wavelength to shift towards shorter wavelengths.Moreover, there is a reduction in coupling efficiency.Maximum confinement losses of 498.00 dB/cm for Ch1 and 690.73 dB/cm for Ch2 are seen at a gold layer thickness of 50nm, hence, this thickness is used for further analysis.

4) PML Thickness Optimization:
The PML layer is used to minimize the reflection of light and reduce scattering losses at the PCF boundary.Increasing the percentage of the PML layer improves phase matching, leading to reduced reflection losses and enhanced PCF performance.A PML layer thickness of 5% to 15% of the PCF diameter enables better mode matching, resulting in decreased reflection losses at the boundary [48].This study examines various thicknesses of PML that correspond to 5%, 10%, and 15% of the total diameter.The objective is to assess their impact on confinement loss.The findings, depicted in Figs. 8 (a,b) for Ch1 and Ch2, clearly demonstrate that a PML thickness of 10% provides tight confinement, which enhances phase matching and results in higher confinement loss.These results suggest that opting for a PML layer with a thickness of approximately 10% (around 1 µm) can offer advantages in terms of achieving improved mode matching and minimizing reflection losses, particularly in sensing applications.

D. Analysis of Sensor Performance
The detection mechanism of the SPR based PCF sensor relies on the degree of coupling between the plasmon mode and core mode.The confinement loss and resonance  wavelength are fundamental characteristics that play a crucial role in detecting and identifying an unknown analyte [49].Sensitivity is a significant parameter in SPR sensing, and it measures the resonance wavelength change because of analyte's RI variation.PCF-based SPR sensors offer high sensitivity because they can confine light within the sensing layer, leading to a greater interaction area and improved sensitivity to RI changes.When the RI of the analyte varies, either the resonance peak magnitude or the resonance wavelength moves to a higher or lower wavelength.The efficacy of the sensor depends on its sensitivity, which can be evaluated using two sensing techniques: wavelength sensitivity and amplitude sensitivity.Below is the formula to calculate sensitivity based on wavelength [49], where, λ peak represents the peak wavelength shift, n a is the change of the analyte RI.The sensor's amplitude sensitivity is obtained using the equation below [50], where a λ , n a represent the loss spectrum difference between contiguous analyte RI values, a λ , n a is the overall maximum loss and n a is the change of the analyte RI.Ch2 is minimal at this point.On the other hand, Ch2 shows its maximum resonance peak at 1.46 µm, while Ch1 is minimal at this point.This single window detection method achieves high sensitivity and selectivity while minimizing background noise from other regions of the sensor.Ch1 exhibits a peak confinement loss of 498 dB/cm at 1.36 µm, while Ch2 demonstrates a peak confinement loss of 690.73 dB/cm at 1.46 µm.The channels attained a highest wavelength sensitivity of 10000 nm/RIU and an amplitude sensitivity of-216 RIU −1 between Ch1 and Ch2.
The proposed sensor's performance is evaluated by testing the RI variations in both channels.Initially, Ch1 is investigated for different analyte variations while keeping the RI value of Ch2 kept at 1.34.Subsequently, Ch2 is investigated with different RI values while keeping the RI value of Ch1 kept at 1.33.Figs.10(a, b) presents the confinement loss spectra of Ch1 and Ch2 for the proposed sensor, which were numerically investigated for a range of RI values from 1.31 to 1.41 in increments of 0.01.Ch1 showed a decrease in the magnitude of the resonance peak as the RI assorted from 1.31 to 1.36, followed by the shift to longer wavelengths.The maximum confinement loss of 505.23 dB/cm has observed at 1.36 µm for the RI value of 1.31.On the other hand, Ch2 exhibited an opposite trend, with the magnitude of the resonance peak increasing as the RI increased from 1.31 to 1.35 and then decreasing until 1.37.At 1.46 µm wavelength, the optimum confinement loss of 743.59 dB/cm is attained for the RI value of 1.35, whereas for RI values above 1.37, the resonance peak shifts towards longer wavelengths.Notably, the proposed sensor structure possesses the ability to measure both amplitude and wavelength sensitivity, thereby providing enhanced performance characteristics.Additionally, Ch1 and Ch2 individually achieved an optimal wavelength sensitivity of 6000 nm/RIU.Fig. 11 depicted wavelength sensitivity analysis of the proposed sensor for Ch1 and Ch2.
Figs. 12(a, b) illustrate the amplitude sensitivity with respect to changes in wavelength for different analyte RI values in Ch1 and Ch2.As the RI of analyte increases in Ch1, the resonance peak moves towards higher wavelengths, and observed an optimal amplitude sensitivity of −87.39 RIU −1 .Alternatively, Ch2 exhibits a shift in the resonance peak towards shorter wavelengths with increasing analyte RI, coupled with an optimal amplitude sensitivity of −304.52 RIU −1 .Table I illustrates the proposed sensor's performance analysis for both Ch1 and Ch2 with varying analyte RI values.
The sensor's resolution determines its ability to identify minor alterations in the analyte's RI.The following equation shows the measure of minimum observable RI limit [19], The spectral resolution is set at a minimum of 0.1 nm, indicated by λ min .Meanwhile, λ peak and n a represent the peak wavelength shift and analyte's RI change, respectively.The maximum resolution measured using the proposed sensor is 5 × 10 −5 RIU for both Ch1 and Ch2, it is evident that the proposed sensor can identify minor RI changes.
The figure of merit (FOM) for the sensor is a vital parameter that is determined by dividing the sensitivity with full width at half maximum (FWHM).The optimal FOM measured using the proposed sensor for Ch1, Ch2 is 125 RIU and 68.96 RIU, respectively.This value can be calculated using the following equation [19]: The performance of the proposed sensor demonstrated significant improvements compared to the existing literature.Table II Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.

TABLE II COMPARATIVE ANALYSIS OF THE PRESENTED SENSOR WITH RECENTLY PUBLISHED PCF BASED SPR SENSORS
presents a performance comparison between the presented sensor and the most recent publications.It is clear that our suggested sensor surpasses the existing sensors Additionally, the proposed sensor has a broad detection range, enabling it to detect various analyte RIs.

IV. CONCLUSION
This article proposed a novel dual-channel plasmonic sensor that utilizes a D-shaped PCF for detecting two different analytes simultaneously.The sensor design incorporates a dual-core structure to enable the analysis of two different analytes simultaneously.The core section is formed by eliminating one ring of air holes at the bottom and two rings at the top, while the birefringence is attained by halving the size of the second air hole rings in the bottom core region.
Additionally, the top and bottom circular portions of the fiber are removed to form a dual D-shaped structure.The SPR effect is initiated by depositing a 50 nm thick gold layer on both planar sides of the PCF, and the analyte is placed above it in Ch1 and Ch2.Numerical analyses are performed using FEM.After optimizing the sensor structure, it attains optimal wavelength sensitivity of 10000 nm/RIU and amplitude sensitivity of −216 RIU −1 between Ch1 and Ch2.Both channels exhibit distinct sensitivity ranges, where Ch1 shows a maximal amplitude sensitivity of −85.39 RIU −1 over the RI value of 1.31 to 1.40 and maximal wavelength sensitivity of 6000 nm/RIU with a resolution of 5 × 10 −5 in the range of 1.37 to 1.41, while Ch2 demonstrates a maximal amplitude sensitivity of −304.52 RIU −1 over the RI value of 1.31 to 1.40 and maximal wavelength sensitivity of 6000 nm/RIU with a resolution of 5×10 −5 in the range of 1.38 to 1.41.This novel sensor offers a wide detection range, making it appropriate for diverse sensing applications such as chemical, biological, and industrial sensing.

Fig. 1 .
Fig. 1.(a) Graphical representation of the proposed dual channel D-shaped PCF sensor using SPR; (b) Experimental arrangement of the proposed sensor.

Fig. 1 (
Fig. 1(b) illustrates the experimental arrangement of the dual channel D-shaped sensor proposed.In order to enhance coupling efficiency, a light source with a broad spectrum is introduced into the sensor through single mode fiber (SMF).The inclusion of a polarizer helps to enhance the visibility of confinement loss for a particular polarization.By manipulating the RI of the analyte that is added to the sensor, there will be changes in the intensity or the wavelength of the confinement loss spectrum.These changes can be detected by connecting an optical spectrum analyzer (OSA) to a computer.The RI of the unknown analyte can be determined by analyzing the data received from the computer.Numerical analyze are carried out using FEM method.For obtain precise results, a physics-controlled mesh model has been utilized to mesh the sensor's structure with extremely fine element sizes.The mesh consists of elements with a maximum element size of 0.52640 µm and a minimum element size of 0.00111 µm.The maximum element growth rate is set to 1.25, while the vertex elements have a size of 692.Ch1 is composed of 72009 total elements, including 5567 boundary elements, and has a minimum element quality of 0.4612.Ch2 comprises 84522 total elements, with 5602 boundary elements and a minimum element quality of 0.5385.The defective modes of the PCF are responsible for its convergence error, which are affected by perturbations with compact support[30].However, the convergence error can be significantly reduced by using the proposed SPR based dual channel D-shaped PCF structure.The error has been measured to be 1.5 × 10 −5 at 1.36 µm and

Fig. 2 .
Fig. 2. Dispersion relation between the real and the imaginary parts of n eff as a function of wavelength: (a) Ch1; (b) Ch2.

Figs. 9
(a, b) illustrate the confinement loss spectra and amplitude sensitivity of a proposed D-shaped dual channel PCF-based SPR sensor as a function of wavelength between Ch1 and Ch2.The RI values of Ch1(n a ) and Ch2(n b ) are kept at 1.33 and 1.34, respectively for the purpose of this analysis.Ch1 reached its highest resonance point at 1.36 µm, while Authorized licensed use limited to the terms of the applicable license agreement with IEEE.Restrictions apply.

Fig. 9 .Fig. 10 .
Fig. 9. (a) Confinement loss spectra of Ch1 and Ch2 as a function of wavelength; (b) Amplitude sensitivity between Ch1 and Ch2 as a function of wavelength.

TABLE I PERFORMANCE
ANALYSIS OF THE PROPOSED SENSOR FOR DIFFERENT RI VALUES ACCORDING TO CHANNELS Fig. 11.Wavelength sensitivity analysis of different analyte RI value for Ch1 and Ch2.