Glucose Sensor Based on Ellipsometry and Circular Dichroism in Achiral Plasmonic Structure

Various efforts have been made to detect minimum value of glucose in any medium like water or body buffer solutions with high-sensitivity, accurate, and low-cost sensors in order to enhance life style. Therefore, the present study was done to investigate reliability of two-dimensional plasmonic structure by circular dichroism (CD) and ellipsometry tools in different concentrations of glucose. Our results confirmed a dependency of the CD signal on glucose concentrations and also a very good sensitivity based on the phase difference between each polarization in ellipsometry parameters with the help of an achiral plasmonic structure.


Introduction
Diabetes mellitus is a chronic metabolic disorder that, according to a statistical report by International Diabetes Federation, affects about 400 million people worldwide per year. In the United States, an estimated cost of 245 billion dollars puts a huge burden on the country [1].
The increasing prevalence of this chronic disease among children and adolescents is regarded as a serious threat to the future security of communities. Among the many causes known for diabetes, its most important symptom is a high blood glucose. Therefore, by controlling the blood glucose levels, all the long-term related complications such as developing cardiovascular disease, chronic renal failure, and diabetic retinopathy can consequently be controlled. Up to now, various devices have been proposed for the control of the blood glucose levels.
Accordingly, most of these blood glucose measuring (BGM) devices use bio-electrochemical designs based on glucose oxidase (GOx) or immobilized glucose dehydrogenase (GDH) on the surface of a disposable electrode [2]. Moreover, these monitoring schemes are nontoxic and with high selectivity. However, these methods are invasive, so in order to use them, blood must be taken from diabetic persons. However, to solve this problem, saliva and interstitial fluids are used to determine the patients' glucose level, because it has been proved that there is a direct relationship between their glucose level and glucose in the blood [3]. Recently, many studies have been performed on optical methods as non-invasive and high-accuracy methods, instead of electrochemical methods, in order to detect the amount of glucose in interstitial fluids. In addition, some minimally invasive or noninvasive techniques were studied for the blood glucose monitoring, including infrared (IR) spectroscopy [4], terahertz time domain spectroscopy [5], Raman spectroscopy [6], and surface plasmon resonance (SPR). However, the results should still be correlated with the direct blood glucose measurements, and the sensitivity and reliability are also limited by spectral signal-to-noise ratio (SNR) level and skin thickness.
Among the optical methods, an optical sensor based on the SPR is very popularly used to determine very low concentrations of any disease with high sensitivity based on the changes in metal refractive index as well as dielectric adjacent environment and also to detect the amount of glucose in the human body. For example, a plasmonic cuvette that is dye chemistry coupled to plasmonic interferometry for glucose sensing [7]; use of gold nanoparticles along with the stimulation of available plasmon polaritons in which high-molecular-weight dextran coated nanoparticles are aggregated with concanavalin A (Con A), which results in a significant shift as well as broadening of the gold plasmon absorption [8]; and a SPR system used for the measurement of glucose in aqueous solution [9].
In the above-mentioned studies, the manufacturing process due to the modification of the surface of the sensor chip or the labeling of nanoparticles, is very complex. As well, some of them have low accuracy for detecting glucose in interstitial fluids. However, it is obvious that the phase sensitivity of SPR systems is much higher than their intensity sensitivity. In this regard, in our earlier works, we established and approved that enhancing SNR can help in detecting very low concentrations diseases at early stage by phase sensitive methods in tissue [10,11], texture or in membrane activity [12]. Correspondingly, this phase sensitive method, named as Plasmonic ellipsometry, is an optical technique that measures the amplitude and phase changes of linear polarizations in reflecting light radiated from the surface of a sensor chip. In general, this technique can be applied to characterize material properties, including composition, roughness, thickness (depth), crystalline nature, doping concentration, electrical conductivity, etc. [13].
In spite of phase differences, circular dichroism (CD), as the difference between right and left circular polarizations' transmission, is widely used for biophotonic applications [14] such as the investigation of circular dichroism-active interactions between Fipronil and Neuronal cells in a study by Xiuxiu Wang et al. [15]. However, the problem of low SNR still remains a major obstacle to the CD sensing. Up to now, various methods have been proposed for SNR amplification, including the use of chiral near-field, which being chiral means that, this plasmonic near-field is not compatible with the mirror image of itself and creates a nanostructure with extrinsic chirality. Mohammadi et al. in their study theoretically investigated the amplification of the CD signal by both chiral plasmonic and dielectric nanostructures [16]. By the use of helical plasmonic nanostructures as prototypical Chiral nearfield Sources, Martin Schäferling et al. created electromagnetic fields with intrinsic chirality, which then enhanced their interaction with chiral molecules [17]. As well, Maria C. di Gregorio et al. studied the interaction between silver and glutathione nanocubes, in order to investigate the amplification of chiroptical effects on plasmon-molecule interactions [18]. In another study, using LSPR from gold chiral nanohooks, Gunnar Klös et al. designed a CD sensor with the enhanced refractive index sensitivity due to the reduction of substrate noise [19].
Therefore, in the present research, we proposed ellipsometry and CD methods based on the phase changes of the polarization of both reflective and transmission beams relative to the incident beam as a high accuracy, online measurement, low-cost, and label free sensor, in order to measure low concentrations of glucose.

II. Material and methods
In this study, we applied soft nano lithography to fabricate an achiral periodic sample using polydimethylsiloxane (PDMS) substrate and 2D charge coupled device (CCD), as the main stamp [20]. This PDMS perforated substrate is covered by a 35 nm thick gold layer using the thermal vapor deposition method to produce two dimensional plasmonic sample. The patterned PDMS, that serves as a 2D grating with periodicity of 3.11 μm, as estimated from scanning electron microscopy (SEM) is shown in Fig. 1(a). In order to investigate the detection of glucose in the phosphate buffered saline (PBS) biological solution in the transmission measurement setup, we designed the fluidic channel ( Fig. 1(b)). Moreover, we designed a transparent flow cell with two inlet channels with the same size: one of them for the equal entry of different concentrations of glucose in the PBS solution and the other one for washing the sensor cell after each measurement of glucose with di-ionized (DI) water using laser incisions on a transparent plexiglass sheet of 2 mm thickness. In addition, the generated sensor chip glued to the surrounding cube, which was embedded inside the flow cell, so that its gold-coated surface was adjacent to the material passing through the flow cell. Thereafter, in order to demonstrate the capability of the sensor chip in accurately detecting different concentrations of glucose, we firstly prepared the PBS biological solvent. Since the wave vector of surface plasmon polaritons has a non-zero imaginary part, in linear polarized light reflection from the plasmonic sample, according to the Airy formula for sequential systems, the phase difference and intensity between the reflected beams with s, p polarization are created which are described by ∆ and Ψ parameters, respectively. Therefore, any change in the properties of the metal and its adjacent dielectric environment in the plasmon sample causes ∆ and Ψ changes. [13,21].
In addition, plasmonic structures have been used to increase the intensity of the circular dichroism signal and also to enhance the chirality of adjacent chiral molecules or structures due to the presence of near-fields obtained from the plasmonic chiral localized dipoles.
Correspondingly, this interpretation was generally expressed by the following relationship regarding the transition matrix approach of multilayer structures of substrates, metal nanoparticles, and chiral molecules by utilizing the Poynting vector relationship of passing fields as follows [16]: where A indicates the transmission light through the structure, and describe the density of electrical and magnetic energy and and are equivalent to the imaginary part of the polarizability of electric and magnetic dipoles, respectively. Additionally, is an intrinsic parameter of the structure, known as the polarizability of the electric-magnetic dipole, which describes its chirality. The optical chirality parameter shows the result of inductive chirality in electromagnetic field, which can be calculated as follows: where E and B are the complex electric and magnetic fields, respectively. Notably, parameter С which depends on the circular polarizations of the incident light, describes the ability of the incident electromagnetic light in being coupled with the chiral structure [17,22]. However, the difference between our approach and the one mentioned earlier was that our two-dimensional plasmonic structure was completely achiral and the parameter was non-zero due to either external or extrinsic chirality resulted from both the stimulation and interference of surface

III. Result and Discussion
When an overlap occurs between the localize surface plasmon polaritons resonances (LSPRs) resulted from the nanowires of each unit cell of square lattice (350 *100 nm) with lattice diffraction order in a 2D lattice structure, a phenomenon of surface lattice resonance (SLR) occurs [23]. These plasmonic SLRs arising from LSPRs of individual metal nanowires are shown as stars in Fig. 1 with diffractive orders presented in a periodic array.  The measurements were repeated for three larger concentrations of 50, 75, and 100 mg/dl using ellipsometry method as shown in Figs.2 (d to f). It can be seen in Fig. 2d that at the nearest peak to the SLR wavelength (668nm), the ∆ has considerably increased along with concentration increasing, so that for the concentrations of 50, 75, and 100mg/dl, ∆ levels were 15.9, 17.8, and 19, respectively. Accordingly, ∆ values for different concentrations are shown in Fig. 2e. In addition, the sensitivity parameter of this measurement was then calculated by the slope of this diagram, the value of which was S = 0.06. The diagram of the changes in the Ψ parameter is plotted in Fig. 2f, which shows a red-shift for increasing glucose concentrations, similar to the measurements of the previous three concentrations. 10  Wavelength (nm) Here, after measuring plasmonic ellipsometry, to detect the baseline values of glucose in the saliva and interstitial fluid simulator environment of the human body, we use another detection method, namely the circular dichroism signal sensing method in 2D achiral plasmonic structure as shown in Fig. 3. To do this, it is necessary to record the transmission spectrum of the circular polarized incident light at a normal angle to the surface of the sensor chip in the adjacent of the dielectric medium. Fig. 3 shows the left-handed incident light spectrum of the plasmonic sample adjacent of glucose in concentrations of 50, 75, and 100 mg / dL. with a square array achiral structure using the FDTD module of Lumerical software . As shown in Fig. 4, the results of the electric field distribution for normal radiation showed that due to the interference of SLRs from the nanowires of each lattice point and the fact that in each angle, some parts of the structural points projection create an asymmetry in the structure as well as external chirality. In this study, we used two sensing methods with high sensitivity which are compared with other optical methods in Table I, both of which were based on the changes of surface plasmon polaritons phase caused by the changes in glucose concentration. The ellipsometry method was used to measure the phase changes of SPPs directly and to assure us that this phase change has occurred. But circular dichroism was applied to measure the transmission intensity of the two circular polarizations resulted from the external chirality that was created by the same phase changes of SPPs.

IV. Conclusion:
In sum, the achiral plasmonic structure was used in this study as a sensitive and low-cost sensor to detect glucose by transmitted and reflected beams. Our results showed that changes in glucose concentrations caused changes in the refractive index of the surrounding medium of the 2D plasmonic substrate. Resonant frequency of SLR in the main sample was altered by this refractive index change and both methods as CD and ellipsometry based on the extrinsic chirality and phase difference were successfully implemented.