Abstract

The research paper describes the design and simulation of an optical biosensor based on a ring resonator with a strip waveguide and a subwavelength grating waveguide. The main goal of the device is to function in low-loss regions and prevent bending losses in ring resonator configuration. The finite-difference-time-domain analysis is performed to simulate the device and to observe the sensing parameters. The biosensor is intended to be used for sensing biological molecules. The working principle of the sensor is to detect changes in the resonant wavelength caused by changes in the refractive index of biomaterials. By observing the shift in the resonant wavelength, the sensor can provide sensitive and reliable detection of biomolecules in solution. The evaluated sensitivity of the device is 206.3 nm/RIU, with a limit of detection of the order of 10−5 RIU. The device is also tested for the detection of glucose concentration in blood and found to be well-suited for on-chip applications. This makes it a promising candidate for use in a wide range of biosensing applications, from medical diagnostics to food safety testing.

1. Introduction

Biosensors based on integrated photonics have become increasingly popular in the last decade [1, 2]. These sensors offer several advantages over traditional biosensing methods, such as high sensitivity and precision [3]. Photonic biosensors, encompassing both fiber in-line and photonic on-chip sensors, provide exceptional sensitivity and label-free detection capabilities for precise analysis of biomolecular interactions, thereby facilitating diverse applications in health care, diagnostics, and environmental monitoring [4, 5]. In the field of optical sensing, on-chip sensors have emerged as a promising alternative to existing fiber in-line interferometers, offering advantages such as improved sensitivity, miniaturization potential, and multiplexing capability [6]. These on-chip sensors enable enhanced detection of minute variations in refractive index (RI), seamless integration into compact devices, and the simultaneous detection of multiple analytes or parameters, thereby paving the way for advanced and versatile sensing applications [7].

Silicon photonics-based biosensors, in particular, have gained attention due to their ability to use the same CMOS manufacturing facilities as electronic devices [8]. This allows for the integration of lasers and detection units on the same chip, which can reduce costs and improve performance compared to other biosensing technologies [9]. In addition to medical and biosensing applications, integrated optical devices such as the one described in this paper can also be used in other fields. For example, they could be used for environmental monitoring [10, 11], food security [12], health monitoring [13], and security surveillance. Silicon photonics utilizing a silicon-on-insulator (SOI) platform is a beneficial method for providing high confinement of the optical field in the core region of the waveguide, this is due to the advantages it offers, such as high sensitivity and resolution for these types of applications [14, 15].

Optical confinement in relied on the RI difference between the core and cladding of the strip waveguide hence are suitably used for optical device. However, the term SOI is elaborately used for the design of the device using silicon as core and silicon dioxide as lower cladding. This makes it possible to create efficient and effective photonic devices with great sensitivity and resolution. It is evident, that a slight portion of the optical field—known as an evanescent field—always seeps outside of the waveguide. This field can be used for sensing applications, such as detecting the presence of biomolecules in solution. Hence for sensing applications high intensity of the evanescent field is desired to get the interpretation of the small change in the biomaterial present in the surrounding medium typically called as upper cladding region [16, 17]. The device’s design can be improved for biosensing applications to enhance the percentage of the optical field that is present on the sensor’s surface [18]. The biorecognition layer is specifically designed to bind to the targeted molecules, allowing for their detection on the surface of the sensor. This layer enables the biosensor to be highly specific and selective for the desired target. Consequently, to improve the light–matter interaction in biosensing applications, various modifications have been proposed in the literature. One potential solution is to employ a strip and slot waveguide arrangement. Strip waveguides are often utilized for modal confinement and low-propagation losses but the sensing performance is affected due to obvious reasons for less field–matter interaction. Though, the slot waveguide is separated by the silicon rails separated by the small gap region often called as gap where the majority of optical field is concentrated. As a result, resonator structures based on slot waveguides can significantly improve the biosensing characteristics of the device [14, 1921]. But the major limitation arises in terms of propagation loss slot waveguide [19]. Perez et al. [22] reported a subwavelength grating (SWG) waveguide that is suitably utilized for biosensing applications due to its feature of flexible tuning feature based on the dimensions corresponding to a low-loss region. The configuration of SWG comprises a periodic arrangement of two different materials with different refractive indices which can be approximated to avoid the diffraction limit and allow for flexible control of the waveguide’s propagation characteristics. This makes the SWG waveguide a promising alternative to slot waveguides for resonator structures in the biosensing applications.

The ring resonator, which is frequently investigated for biosensing, is made up of a bus waveguide and a ring that can function in both add-drop and all-pass modes. In all-pass and add-drop systems, one bus waveguide is utilized. In a paper, Sahba et al. [11] discussed the development of a ring resonator for biosensing applications using an ultrathin waveguide with a sensitivity of 133 nm/RIU. Multiple rings linked to a single bus waveguide were investigated by Bruns et al. [12] to enable multiplexed detection. The utilization of the Vernier effect to combine the resonance and optical absorption characteristics of biomaterials for improved biosensing has also been investigated by other researchers. These methods show how adaptable and useful photonic devices are for biosensing applications [8, 2326]. To detect various quantities of ethanol, Lei et al. [24] employed a cascade of two rings and obtained a Q-factor of 2 × 104 and a limit of detection (LOD) of 1.15 × 10−2 RIU. Wang et al. [27], claim that a compound resonator–interferometer may attain a sensitivity of 688.3 nm/RIU. A hybrid structure surface plasmon-based Mach–Zehnder interferometer (MZI) and SOI chip to increase the sensitivity around 102 nm/RIU is reported in [28]. As previously discussed the inherent capability of a label-free sensor using a slot waveguide is demonstrated using a ring resonator design to quantify the sensitivity value of 300 nm/RIU [29]. Similarly whispering gallery mode using a microring resonator is harnessed to report the sensitivity of 120 nm/RIU [30]. Since the concentration of the optical field is increased by using the narrower waveguide hence can enhance the sensitivity of the label-free sensor [16, 31, 32], the reported values are 100 nm/RIU and achieved LOD of 5 × 10−4 RIU [32].

To evaluate the performance of a label-free biosensing device, a hybrid ring resonator architecture that includes an SWG waveguide and a strip waveguide is presented in this work. The bus waveguide is composed of a low-loss SWG waveguide, while the ring is comprised of a strip waveguide. Section 2 presents a theoretical examination of the SWG waveguide, while Section 3 discusses sensor design and operation. Section 4 investigates the device’s usage for glucose sensing and concludes with a summary and references.

2. Modeling of SWG Waveguide

An SWG is a type of optical waveguide that utilizes a periodic 1D structure consisting of a repeated arrangement of regions of high- and low-refractive indices. These high- and low-index regions create a diffraction grating that enables the waveguide to guide light at subwavelength scale. The grating-period (Λ) and duty-cycle (dc) are two important parameters that define the performance of an SWG waveguide. The Λ is the distance between the centers of two consecutive high- or low-RI regions, and the dc is the ratio of the width of the high-RI region to the total grating period. The Λ and dc can be adjusted to optimize the waveguide for a specific application, such as reducing loss, increasing the confinement of the guided mode, or increasing the device’s sensitivity. The SWG waveguide, which exhibits periodicity in the z-direction, is schematically depicted in Figure 1. When a wave is propagating, its electric field satisfies the Bloch–Floquet mechanism, which has the formula is the wave equation and the propagating wave is called Bloch wave [22, 33]. The behavior of the wave in a 1D periodic structure is affected by the phase-matching condition, which is dependent on the Λ and the Bragg wavelength (λB), where λB = 2·Λ·neff. The stopband in a Bragg grating waveguide may be seen in the transmission spectrum at the Bragg wavelength (λB) [34]. Similar to this, the SWG waveguide Λ has two distinct uses. The waveguide is ideal for designing a fiber-to-waveguide coupler when Λ is greater than λB/(2·neff), at the instant the propagating wave becomes weak as most of the signal radiated out by striking the wall of the grating [34]. Second, the low-loss region when the value of Λ is smaller than λB/(2·neff), at this condition the SWG behaves as a normal waveguide by suppressing the diffraction limit and with the flexibility of altering the propagation characteristics [22, 35].

The effective index method can be used to analyze a planar waveguide for biosensing applications. However, since there is no analytical model available for representing subwavelength waveguides, the finite-difference-time-domain (FDTD) method must be used to evaluate the effective index of a waveguide [36]. The analysis of 1D photonic crystals relies on the calculation of their band structure within the Brillouin zone defined by the range , where kz is the wave vector. The dispersion diagram in Figure 2 shows the bandgap and low-loss region of the 1D periodic structure. To design a SWG waveguide, the grating period must be determined based on the low-loss region of the spectrum corresponding to the desired wavelength.

3. Device Modeling Using SWG and Strip Waveguide

Monitoring blood glucose levels plays a vital role in the effective management of diabetes among individuals. The continuous and accurate monitoring of glucose enables timely modifications in medication, diet, and lifestyle choices, thereby mitigating the risks associated with short-term complications like hypoglycemia as well as long-term complications such as cardiovascular diseases, neuropathy, and retinopathy [37]. This manuscript focuses on highlighting the significance of blood glucose monitoring with the help of compact ring resonator-based label free biosenor. The microring resonator device comprises SWG and strip waveguide on an SOI platform of 2 mm thickness. The core (Si) of the waveguide is having RI value of 3.47, and the lower cladding (SiO2) is having a RI of 1.444 at wavelength of 1,550 nm. The core of both of the waveguides is having dimensions of 220 nm in thickness and 600 nm in width. For the implication of the properties of analytes present in the human body, the top cladding is considered to have a RI of 1.33. Figure 3 depicts the variation of the effective index of the SWG on a variation of the wavelength (in nm) which is evaluated by performing FDTD analysis over the SWG dimension.

In Figures 4(a) and 4(b) of the research paper, an add-drop configuration of the device is illustrated, where the SWG waveguide is placed corresponding to the two parallel channels. This add-drop configuration allows for the selective filtering of certain optical signals, enabling the device to add or drop specific wavelength channels as desired. However, the centralized ring section of the device is composed of a strip waveguide. This design allows for the efficient coupling of light which is critical for the device’s performance. The combination of the SWG waveguide and the strip waveguide in the design allows for the precise manipulation of light and the ability to selectively filter certain wavelengths, making it well-suited for a range of applications such as sensing, filtering, and modulation. From the fabrication results the propagation loss of the SWG waveguide is considered to be 2.1 dB/cm [38]. However, using the SWG waveguide for the curved section of the ring increases radiation losses and reduces the overall field intensity. To avoid these losses, the ring segment is instead designed using a strip waveguide. Four triangular-shaped segments are added to the ends of the SWG waveguides to increase the efficiency of the transmission of signals from the source to the waveguide. These segments, also known as bridge segments, are around 8 μm long overall. By using bridge segments, it is possible to strengthen and more reliably transmit the signal by increasing the coupling of the signal between the source and the waveguide. The other relevant parameters of the device are listed as; the radius of the ring section is 10 µm, the gaps between SWG and ring are 250 nm. This parameter is important as it defines the coupling strength between the ring resonator and the SWG waveguide. It also affects the device performance by determining the amount of light that can be transmitted. The value of Λ is 300 nm and dc is 60%. These parameters are specific to the SWG waveguide and influence the waveguide’s diffraction efficiency, confinement, and loss.

Figure 5(a) shows the modification in the transmission spectra of the device for different values of the analyte refractive indices present over the upper cladding of the device. By employing the high-resolution spectra analyzer the shift in spectra is quantified for detection of the presence of the analytes. In the transmission spectrum, the region of operation is fixed, and the dip in the transmission spectra corresponding to a specific wavelength value is called the resonant wavelength (λres). In the range of refractive indices from 1.33 to 1.37, a significant shift in the transmission spectra is observed.

Sensitivity (S) and detection limit (LOD) are the basic metrics used to assess the performance of photonic biosensors. The definition of sensitivity corresponds to the shift of on account of the change in the analyte RI (), mathematically represented as . The sensitivity of the sensor is found by examining the linear relationship shown in Figure 5(b), which in this device is calculated to be 206.3 nm/RIU.

4. Device Application for Detection of Blood Glucose Concentration

When our blood glucose levels are too high, such as in the case of diabetes, it can lead to long-term complications such as damage to the blood vessels, nerves, and organs [39]. On the other hand, when our blood glucose levels are too low, it can result in symptoms such as weakness, dizziness, confusion, and even loss of consciousness [4043]. The biosensing capabilities of the device shown in Figure 4 are tested for the detection of glucose concentration in blood samples. In this test, blood is applied to the device’s top cladding area. To obtain the practical results, the solution used for testing consisted of blood glucose, ethylene-diamine-tetra-acetic acid (EDTA), and sodium fluoride (NaF) in proportions of 0.1, 0.8, and 0.1, respectively. The RI of EDTA () is 1.363 and NaF () is 1.3194 at 1550 nm. In the solution, EDTA acts as an anticoagulant, and NaF functions as a preservative [4446]. Lorenz–Lorentz mixing rule is used for finding the value of RI of the mixture () which is given by Tasic et al. [47] as follows:where is the RI of blood glucose.

The value of can be calculated using the Gladstone–Dale mixing rule. This rule states that the RI of a mixture is given by [47],where , c is the concentration of the glucose in blood [46], and the RI of the blood () is evaluated using Sellemier’s equation [48],where the values of , , , and are 0.83423, 0.04296, 10775.45, and 6.136 × 106, respectively. Hence the value of RI of the depends on the c and λ. Figure 6 shows a plot of the resonant wavelength and c, which exhibits an approximately linear relationship. As the concentration of glucose increases, the resonance wavelength shifts to higher values. As shown in Figure 4(b), the test samples can be added into the sensing window via a fluidic system [4951].

To test the ability of the glucose sensor to distinguish between hyperglycemia and hypoglycemia, concentration values of 70, 110, and 140mg/dl are used [46]. The corresponding resonant wavelengths were found to be 1.56735, 1.56785, and 1.5683 µm, respectively. These findings show that the sensor can precisely identify a variation in glucose concentrations.

The LOD is determined to identify the smallest quantifiable value required to observe λres of the sensor. The relationship between the glucose concentration (c) and the resonant wavelength (λres) is shown in Figure 6, based on this relationship, the LOD is estimated to be in the range of 10−5 RIU. Table 1 corresponds to a comparison of performance metrics for different label-free biosensors. This suggests that the device is appropriate for both lab-on-chip applications and usage as a glucose sensor.

5. Conclusion

Using a strip waveguide and an SWG waveguide, we have demonstrated a ring resonator structure in this study. The SWG waveguide’s parameters, such as the waveguide width, grating period, and duty cycle, may be adjusted to modulate the optical properties. The performance of the device was evaluated using the computational FDTD method. The shift in wavelength when the RI of a biomaterial changed is used to assess the device’s sensing capacity. The device’s detection limit is in the range of 10−5 RIU, and its sensitivity is determined to be 206.3nm/RIU. The device is also tested for its ability to detect the concentration of glucose in the blood. Overall, our results indicate that the device is well-suited for applications as a glucose sensor and in lab-on-chip systems.

Data Availability

No public data are involved in the study. The results presented in the manuscript were obtained through experimentation in the lab.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This study is a collaborative effort between the faculties of GGITS Jabalpur and MNIT Jaipur. The authors would also like to acknowledge the administrative bodies of the institutions for providing the necessary resources.