Surface Plasmon Resonance Sensor Based on Polypyrrole–Chitosan–BaFe 2 O 4 Nanocomposite Layer to Detect the Sugar

: The surface plasmon resonance sensor was used to detect and measure low concentrations of sugar. A polypyrrole–chitosan–BaFe 2 O 4 nanocomposite layer was prepared to improve the surface of the gold layer for the detection of glucose, fructose, and sucrose using the surface plasmon resonance technique. The polypyrrole–chitosan–BaFe 2 O 4 was synthesized using the electrodeposition method in di ﬀ erent thicknesses. The functional group, crystal structure, and morphology of the layer were investigated with Fourier transform infrared spectroscopy, X-ray di ﬀ raction technique, and ﬁeld emission electron microscopy. Consequently, the BaFe 2 O 4 was scattered on the surface of the polymer, and the a ﬃ nity of polypyrrole–chitosan–BaFe 2 O 4 to bond with glucose is higher than that for the other sugars. The sensor limit was 0.005 ppm.


Introduction
In the last decade, many researchers focused on the detection and measurement of the low concentration of sugar such as glucose, fructose, and sucrose [1]. Sugars are found in the blood, food, and beverages. Numerous methods have been used to measure glucose, fructose, and sucrose levels, including colorimetric [2], infrared [3], Raman [4], and electrochemistry spectroscopy [5]. Other analytical methods such as gas chromatography and mass spectroscopy were considered to investigate sugar levels in human and natural products with a limit in the range of mM to µM [6]. The optical method is an interesting technique to evaluate chemical and biochemical properties [7]. Hence, the detection and measurement of low concentrations of glucose, fructose, and sucrose were carried out using a laser interferometer [8] Ultraviolet-visible (UV-Vis) absorption spectroscopy, fluorescence spectroscopy, the optical polarization properties of medium and heterodyne polarimetry [9,10], reflected THz radiation [11], and a surface plasmon resonance sensor [12][13][14]. Recently, glucose biosensors have been presented based on enzymatic catalysts in three versions of enzymatic glucose biosensors [15].
The limitation of these sugar biosensors depends on the performance of these sensors-that is, the inherent instability [16] and the sensitivity of these type of biosensors depends on pH, the temperature of samples, and concentration of oxygen [17,18].
The ratio of signal to noise and immobilization of sugar to the sensing layer are the significant parameters to design the application of glucose, fructose, and sucrose sensors [19,20]. Therefore, the nanoparticles (Au, Ag, Fe 2 O 3 , and ZnO) and conductive polymer (polypyrrole) were used to increase and improve the ratio of signal to noise and immobilization in sugar sensors. Polypyrrole has the potential of electron transfer, because the electron transfer between sugar and the sensing layer is slow [1]. Hence, the coupling of nanoparticles and polypyrrole can improve the electron transfer in the procedure of sugar detection (sensor) [1,16], and it can also enhance the selectivity and sensitivity of the sensor.
Moreover, the surface plasmon resonance (SPR) sensor is a versatile and accurate method to detect the chemical materials and sugars. The SPR phenomenon occurs at the interface of two mediums with opposite dielectric signs. Normally, the gold layer is used for the SPR sensors, and the sensing layer has been coated on the surface of the gold layer to improve the sensitivity and selectivity of the sensor. Recently, the nanocomposite layers have been used for the sensing layer. The gold nanoparticle, carbon-based nanostructure [1], and magnetic nanoparticles were used to improve the sensitivity of the sugar sensor. Magnetic nanoparticles have some advantages, such as high catalytic efficiency, good stability, and monodispersion [2]. Hence, many researchers used the magnetic nanoparticles such as ZnFe 2 O 4 [2], Fe 3 O 4 [21], and polypyrrole-chitosan/Fe 2 O 3 nanocomposite [16] to detect glucose and other sugars; the detection limit was in the range of 1 to 3 µM, and the ratio of signal to noise was about 3. Consequently, the magnetic nanoparticles are a considerable nanostructure to design the sugar sensor.
In this study, a polypyrrole-chitosan-BaFe 2 O 4 (PPy-Chi-BaFe 2 O 4 ) nanocomposite layer was prepared using the electrochemical method. The prepared layer was characterized using the field emission scanning electron microscopy (FE-SEM), the X-ray diffraction (XRD), and the Fourier transform infrared spectroscopy (FT-IR). The characterized layer was used to detect the glucose, fructose, and sucrose using the surface plasmon resonance technique.

Materials
To prepare the PPy-Chi-BaFe 2 O 4 , commercial chemical components were used. Iron nitrate (Fe (NO 3 ) 3 .9H 2 O) and barium nitrate (Ca(NO 3 ) 2 .6H 2 O), polyvinyl alcohol (PVA, MW = 31,000 g/mol), glucose, sucrose, fructose, chitosan (medium molecular weight (75-85% deacetylated), lithium perchlorate, and potassium dihydrogen phosphate were purchased from Sigma Aldrich company. The materials were in high quality, and had high purity of about 99%. The saturated calomel electrodes (SCE) was used as a reference electrode from BASi Company. To synthesize the PPy-Chi-BaFe 2 O 4 nanocomposite layer, a PVA/BaFe 2 O 4 nanocomposite was prepared using the thermal treatment method [22][23][24]. Ba(NO 3 ) 2 H 2 O, Fe(NO 3 ) 3 H 2 O, DDW, and polyvinyl alcohol (PVA) were used to precursors, solvent, and capping agent, respectively [24]. First, 3.5 g of PVA was dissolved in 100 mL of DDW at 353 K. After that, 0.2 mmol Fe (NO 3 ) 3 H 2 O and 0.1 mmol Ca(NO 3 ) 2 H 2 (Fe:Ba=2:1) were added into the PVA solution, and the mixture was stirred constantly for 1.5 h, and a clear solution was obtained. The clear mixture was heated at 362 K for 24 h to remove the solvent (deionized distilled water, or DDW), and the solid BaFe 2 O 4 was reminded. The product was grounded in a mortar to obtain the uniform powder, and it was heated at 773 K for crystallization of the nanocrystal and decomposition of the PVA. The final powder was used to prepare the PPy-Chi-BaFe 2 O 4 .

Synthesis of PPy-Chi-BaFe 2 O 4 Composite Layer
The PPy-Chi-BaFe 2 O 4 was synthesized using the electrodeposition method. The gold-coated glass slide was prepared using a sputtering coating method, and the pyrrole was polymerized on the surface of the gold layer in the presence of chitosan and BaFe 2 O 4 using the cyclic voltammetry method.
To prepare the electrolyte solution for the electrodeposition of PPy-Chi-BaFe 2 O 4 , 0.2 g of chitosan powder was dissolved in 50 ml aqueous solution of 0.3M oxalic acid, and it was stirred 4 h prior to its use for the preparation of the layer. The monomer solution contains mixed electrolytes in a 0.1 M phosphate buffer solution (pH 7.2), 0.1 M pyrrole monomer, 0.1 M LiClO 4 as a dopant, 0.1 M chitosan, and 0.05 g of BaFe 2 O 4 . The electrolyte was stirred during electrodeposition to make sure there is no precipitation. The working electrode, reference electrode, and counter electrode were a gold-coated glass slide, saturated calomel electrodes (SCE), and platinum electrode, respectively. The polymerization of pyrrole was carried out at the potential range of 0 V to 0.8 V in the presence of chitosan and BaFe 2 O 4 nanoparticles at different cycles from 1 to 50. The prepared samples were put in the oven for 2 h to dry the layers. The PPy-Chi-BaFe 2 O 4 composite layers were characterized using the FT-IR spectrometer (model: NEXUS), XRD spectrometer (WITec, Alpha 300R with Shimadzu diffractometer: model XRD6000 and Cu, Ka (0.154 nm)), and the FE-SEM (NOVA NANOSEM 230)

Surface Plasmon Resonance Setup
The SPR setup based on the Kretschmann configuration was used to measure the low concentration of glucose, fructose, and sucrose. Figure 1 shows the SPR setup with a high index prism based on angular modulation, and the variation of laser beam intensity was detected at a different angle using a silicon photodetector. The PPy-Chi-BaFe 2 O 4 composite layer was deposited on the gold side of the glass slide, and it was attached to a high index prism (SF52, Foctek) using the index matching gel (F-IMF-105, Newport, USA) [12,24,25]. The prism was placed on a precision rotation stage, and the fluid holder was contacted to the sensing layer using an O-ring. The glucose, fructose, and sucrose were contacted separately to the sensing layer. The rotation stage was rotated up to 14 • at increments of 0.01 • . The intensity of the laser beam and the angle of rotation were registered when the rotation stage was stopped momentarily.
The experiment was repeated separately for each sugar about 10 times. The SPR signal was analyzed using the Fresnel's theory based on the matrix method for a multi-layer system [26].

Preparation of Sugar Solution
First, 0.1 g of glucose, sucrose, and fructose were separately used to prepare the high concentration of sugar solution in 100 ml of the deionized distilled water (DDW). Other concentrations of the sugar solution, including 0.005 ppm (0.0005 mg/dL), 0.05 ppm (0.005 mg/dL), 0.5 ppm (0.05 mg/dL), 5 ppm (0.5 mg/dL), 15 ppm (1.5 mg/dL), and 25 ppm (2.5 mg/dL), were systematically prepared by dissolving the sugars solution in the DDW.
In this research, glucose, fructose, and sucrose (three sugars) in six concentrations were used to do the experiment, and the SPR experiment was repeated 10 times for each concentration of sugar.    (120), respectively. They confirm that BaFe 2 O 4 has a face-centered cubic structure [30]. The XRD pattern of PPy-Chi-BaFe 2 O 4 confirms that the BaFe 2 O 4 also formed in the composite layer in the cubic structure. During the electrodeposition process, monomer molecules and metal ions at the electrode-electrolyte interface deposited on the surface of the working electrode by reaching the proper applied potential, and the mechanism is as follows [28]:  [16]. BaFe 2 O 4 nanoparticles (NPs) were scattered on the surface of layer, and it impressed the morphology of the PPy-Chi-BaFe 2 O 4 nanocomposite layer [31].

Results and Discussion
The essential charges for organizing the PPy were the electrons and anions that LiClO 4 and BaFe 2 O 4 provided them. Therefore, the role of LiClO 4 and BaFe 2 O 4 were as the electron and anions charges to create the counterbalance of the PPy-Chi-BaFe 2 O 4 composite layer. Hence, the polymer electropolymerized at the surface of the working electrode, containing BaFe 2 O 4 nanoparticles [32].
The gold/PPy-Chi-BaFe 2 O 4 system layer was prepared in different thickness. The thickness of the gold layer was 48.3 nm, and the thickness of the PPy-Chi-BaFe 2 O 4 composite layers was tested using a profilometer (AMBIOS, XP-200) in the range of 3 ± 1 nm to 94 ± 1 nm, and Figure 3a shows that the thickness of the layers increased as the cycle increases. The system layer was coated on the microscope glass slide, and it was attached to the prism using index matching gel. The SPR signal was separately registered for each thickness in the presence of DDW (n = 1.3323). Figure 3b,c shows the SPR results to test the refractive index of the layers and baseline. The refractive index of the layer was achieved by analysis of the SPR signal using Fresnel's theory for the multilayer system when the refractive index of the gold layer was 0.237 + 3.335i.
Fresnel's theory predicts the reflection coefficient (r) from dielectric and metal medium such as the gold layer. The matrix form that was presented in ref. [33,34] is used for the multilayer system. The reflectance (R) is equal to r × r. If the refractive index (n) and accurate thickness (t) of layers are unknown, they can obtain from the minimum root square of the difference between the theory and experimental value of reflectance as follows [26,27]: where R Theory and R Exp are the theoretical and experimental value of reflectance, respectively.  According to the SPR signals analysis, the accurate thickness of layers was in the range of 3.6 nm to 94.2 nm, and the refractive index of the PPy-Chi-BaFe 2 O 4 was in the range of 1.67571 + 0.139i to 1.57072 + 0.164i. Figure 3c shows the SPR signal at the baseline when the baseline was obtained at 53.948 using the PPy-Chi-BaFe 2 O 4 layer with 7.3 nm of thickness.
The variation of real (n) and imaginary (k) parts of the refractive index were demonstrated in Figure 4a,b, respectively. Consequently, the real (n) part of the refractive index decreased, and the imaginary (k) part of the PPy-Chi-BaFe 2 O 4 increased when the thickness of the layer increased. The pertinent parameters were listed in Table 1.  The PPy-Chi-BaFe 2 O 4 with 7.3 nm thickness was used to measure and detect the low concentration of glucose, sucrose, and fructose. The experiment was separately carried out in the presence of 0.005 ppm (0.0005 mg/dL), 0.05 ppm (0.005 mg/dL), 0.5 ppm (0.05 mg/dL), 5 ppm (0.5 mg/dL), 15 ppm (1.5 mg/dL), and 25 ppm (2.5 mg/dL) of glucose, sucrose, and fructose. The SPR signals were registered, and the variation of resonance angle shift was achieved during 420 s. Figure 5a-c shows the resonance angle shift with time (sensorgram) for the glucose, fructose, and sucrose, respectively. As a result, the experimental values fit well with the first order Langmuir formula (∆θ = ∆θ sat (1 − exp(−k a t))) [35,36].
The concentration of sugars that attached the sensing layer was tested using UV-vis spectroscopy. The PPy-Chi-BaFe 2 O 4 sensing layer was separately immersed in the glucose, fructose, and sucrose solutions in the concentration of 15 ppm. The concentrations of glucose, fructose, and sucrose solution were measured before and after the experiment. The degree of adsorption was calculated from [37]: where C i and C f were the initial and final concentration of the sugar. Figure 6a1,a2,b1,b2,c1,c2 shows the UV-visible spectra of the glucose, fructose, and sucrose in the concentrations of 2 ppm (0.2 mg/dL), 6 ppm (0.6 mg/dL), 15 ppm (1.5 mg/dL), 20 ppm (2 mg/dL), and 50 ppm (5 mg/dL). The main peak occurred at about 280 nm for them. The intensity peaks were driven for each concentration of the glucose, fructose, and sucrose. Figure 6a3, b3, c3 shows the variation of peak intensity with each concentration of the sugar as a calibration curve. Figure 6a2,b2,c2 show the UV-vis spectra of the sugar solutions after they contacted with the sensing layer and the concentration of glucose, fructose, and sucrose after contact with the PPy-Chi-BaFe 2 O 4 sensing layer was achieved from the calibration curve. As a result, the concentration of glucose was lower than that of the other sugars, and it means that the affinity of the PPy-Chi-BaFe 2 O 4 nanocomposite layer to the adsorption of the glucose is much higher than that of the other sugars. The concentration value of the sugar and the degree of adsorption are listed in Table 2.   Magnetic nanoparticles and chitosan can be used for sugar sensor application, because they contribute to transfer the electron between a receptor (sensing layer) and sugar [1,2,16]. The BaFe 2 O 4 NPs can enhance the electron transfer between glucose, fructose, and sucrose with the sensing layer. The molecule of glucose contains one hydroxyl group (OH) and one hydroxymethyl group (CH 2 OH) at the edge of the molecule plane, and the hydroxyl group can interact with the BaFe 2 O 4 -NPs in the sensing layer. The difference between glucose and fructose is in the number of hydroxyl groups at the edge of molecule planes so that fructose contains two hydroxymethyl groups (CH 2 OH) at the edge of molecule plane, and they cause the space moment between fructose and BaFe 2 O 4 NPs. Moreover, the number of hydroxyl groups (OH) in glucose is higher than those in fructose. Therefore, the tendency of glucose to exchange the electron with a sensing layer is higher than fructose [2,38]. The molecule of sucrose is a combination of glucose and fructose, and it is very heavy and stable. Hence, the tendency of sucrose to exchange the electron with BaFe 2 O 4 NPs is very weak. Consequently, the degree of adsorption for glucose is higher than the degree of adsorption of fructose and sucrose. Therefore, the PPy-Chi-BaFe 2 O 4 nanocomposite layer is sensitive to bind the low concentration of the glucose.

Conclusions
The PPy-Chi-BaFe 2 O 4 nanocomposite layer was prepared using electrodeposition technique. The cyclic voltammetry method was used in a different cycle to control the thickness of the layer, which was in the range of 3.6 nm to 94.2 nm. The refractive indices of the PPy-Chi-BaFe 2 O 4 composite layers were measured using the SPR method, and they ranged from 1.67571 + 0.139i to 1.57072 + 0.164i, and the resonance angle shifted from 53.126 • to 68.131 • . The PPy-Chi-BaFe 2 O 4 nanocomposite layer was used to detect the sugar. As a result, the affinity of the sensing layer to bind the glucose was higher than that of sucrose and fructose, and the limit sensor was about 0.005 ppm (0.0005 mg/dL). Consequently, PPy-Chi-BaFe 2 O 4 nanocomposite layer tends to interact with sugar, and it can adsorb the glucose well.