Cationically Modified Nanocrystalline Cellulose/Carboxyl-Functionalized Graphene Quantum Dots Nanocomposite Thin Film: Characterization and Potential Sensing Application

In this study, highly functional cationically modified nanocrystalline cellulose (NCC)/carboxyl-functionalized graphene quantum dots (CGQD) has been described. The surface of NCC was first modified with hexadecyltrimethylammonium bromide (CTA) before combining with CGQD. The CGQD, CTA-NCC and CTA-NCC/CGQD nanocomposites thin films were prepared using spin coating technique. The obtained nanocomposite thin films were then characterized by using the Fourier transform infrared spectroscopy (FTIR) which confirmed the existence of hydroxyl groups, carboxyl groups and alkyl groups in CTA-NCC/CGQD. The optical properties of the thin films were characterized using UV–Vis spectroscopy. The absorption of CTA-NCC/CGQD was high with an optical band gap of 4.127 eV. On the other hand, the CTA-NCC/CGQD nanocomposite thin film showed positive responses towards glucose solution of different concentration using an optical method based on surface plasmon resonance phenomenon. This work suggests that the novel nanocomposite thin film has potential for a sensing application in glucose detection.


Introduction
In recent years, there has been an interest in the production of nanocrystalline cellulose (NCC) from cellulosic material because of its biodegradability, renewability, abundance and excellent mechanical properties [1]. In this world, cellulose is one of the most numerous natural renewable and biodegradable polysaccharides. NCC is the nano-scaled of needle or rod-shaped crystalline which has hundreds of nanometers in length and 1-10 nm in width [2,3]. NCC is obtained when cellulose undergoes acid hydrolysis with conditions where the amorphous regions are selectively hydrolyzed [4]. Mineral acids including hydrochloric acid and sulfuric acid are used in the mixture of hydrolysis of cellulose to prepare NCC [5]. Thus, NCC is constitutively acidic and exhibits a lyotropic phase behavior depending on the concentration. NCC has the potential in various applications as a rheology modifier such as drilling fluids, consumer products, drug delivery, artificial tissue formation and injectable hydrogels [6][7][8].
As far as we know, the optical properties of the CTA cationically modified NCC/CGQD (CTA-NCC/CGQD) nanocomposite thin film and its potential application for detection of glucose using surface plasmon resonance technique (SPR) have yet to be reported. SPR is known as a simple optical method for surface studies of thin films and can act as a very sensitive spectroscopy for detection of a variety of targets [26][27][28][29][30][31][32][33][34][35][36][37][38][39][40][41][42]. Hence in this study, the fabrication of the CTA-NCC/CGQD nanocomposite thin film, its characterization and potential sensing application were explored.

Preparation of Chemicals
To prepare NCC solution, 1 g of NCC was diluted in 100 mL deionized water. Then, 0.2 g of CTA was diluted in 20 mL of deionized water to obtain CTA solution. NCC solution was then dropped into CTA solution drop by drop while heat stirred for 24 hours. The CTA-NCC solution was centrifuged at 3000 rpm for 15 minutes. Then, CTA-NCC/CGQD solution (0.05 wt%) was obtained by dispersing 1 mL of CGQD into 1 mL of CTA-NCC. The glucose solution was prepared by dissolving 9.91 mg of glucose with 100 mL of deionized water to produce 10 µM of glucose solution. To prepare glucose solution with various concentration, the 10 µM of glucose solution was diluted with deionized water based on the formula M 1 V 1 = M 2 V 2 to obtain 0.005, 0.01, 0.03, 0.05 and 0.1 µM of glucose [43][44][45].

Preparation of CTA-NCC/CGQD Nanocomposite Thin Film
Glass cover slips (24 mm × 24 mm × 0.1 mm) were used as the substrates. The glass slip was first sputtered with gold (SC7640 sputter coater machine) for 67 seconds to obtain 50 nm of gold thin film [46][47][48]. Then, spin coating technique was used to deposit the CTA-NCC/CGQD solution homogenously on the gold surface. About 1000 µL of CTA-NCC/CGQD solution was added on a gold coated glass slip and was spun at 3000 rev/min for 30 seconds using spin coater P-6708D to obtain around 12-15 nm thickness of the CTA-NCC/CGQD layer. The summarized flow chart for the preparation of CTA-NCC/CGQD is shown in Figure 1. sputtered with gold (SC7640 sputter coater machine) for 67 seconds to obtain 50 nm of gold thin film [46][47][48]. Then, spin coating technique was used to deposit the CTA-NCC/CGQD solution homogenously on the gold surface. About 1000 μL of CTA-NCC/CGQD solution was added on a gold coated glass slip and was spun at 3000 rev/min for 30 seconds using spin coater P-6708D to obtain around 12-15 nm thickness of the CTA-NCC/CGQD layer. The summarized flow chart for the preparation of CTA-NCC/CGQD is shown in Figure 1.

Characterization Instrument
The Fourier transform infrared (FTIR) spectrum of CGQD, CTA-NCC and CTA-NCC/CGQD solutions were analyzed using the Fourier Transform Infrared Spectrometer model spectrum 100 (PerkinElmer, Waltham, MA, USA), with wavelength set from 400 to 4000 cm −1 which is to determine the functional groups and the chemical interaction of the composites. Other than that, the purity of the compound can be obtained from the collection of the absorption band from the spectrum. For optical properties, the absorption of all samples with wavelength range from 220 nm to 500 nm was investigated using UV-Vis-NIR spectrometer (UV-3600 Shimadzu, Kyoto, Japan). The absorbance coated thin film was measured at room temperature. The energy band gap was determined by analyzing the graph of absorption peak against wavelength obtained using UV-Vis spectrometer.

Surface Plasmon Resonance
Surface plasmon resonance (SPR) is used to identify the potential of CTA-NCC/CGQD nanocomposite thin film for glucose detection. SPR is an optical process in which light satisfying resonance conditions excite a charge-density wave propagating along the interface between a dielectric material and metal by p-polarized and monochromatic light beam [49]. The reflected light intensity is reduced at a specific incident angle producing a sharp shadow due to the resonance occurs between surface plasmon wave and incident beam [50]. The SPR measurement was carried out by determining the reflected He-Ne laser beam (532.8 nm, 5 mW) [51]. Figure 2 shows the setup of SPR sensor. The SPR setup consisted of an He-Ne laser, a light attenuator, a polarizer and optical chopper (SR 540) and an optical stage driven by a stepper motor MM 3000 with a resolution of 0.001 • (Newport, CA, USA). The reflected beam was detected by photodiode and then processed by the lock-in-amplifier (SR530) [52][53][54][55].

Characterization Instrument
The Fourier transform infrared (FTIR) spectrum of CGQD, CTA-NCC and CTA-NCC/CGQD solutions were analyzed using the Fourier Transform Infrared Spectrometer model spectrum 100 (PerkinElmer, Waltham, MA, USA), with wavelength set from 400 to 4000 cm −1 which is to determine the functional groups and the chemical interaction of the composites. Other than that, the purity of the compound can be obtained from the collection of the absorption band from the spectrum. For optical properties, the absorption of all samples with wavelength range from 220 nm to 500 nm was investigated using UV-Vis-NIR spectrometer (UV-3600 Shimadzu, Kyoto, Japan). The absorbance coated thin film was measured at room temperature. The energy band gap was determined by analyzing the graph of absorption peak against wavelength obtained using UV-Vis spectrometer.

Surface Plasmon Resonance
Surface plasmon resonance (SPR) is used to identify the potential of CTA-NCC/CGQD nanocomposite thin film for glucose detection. SPR is an optical process in which light satisfying resonance conditions excite a charge-density wave propagating along the interface between a dielectric material and metal by p-polarized and monochromatic light beam [49]. The reflected light intensity is reduced at a specific incident angle producing a sharp shadow due to the resonance occurs between surface plasmon wave and incident beam [50]. The SPR measurement was carried out by determining the reflected He-Ne laser beam (532.8 nm, 5 mW) [51]. Figure 2 shows the setup of SPR sensor. The SPR setup consisted of an He-Ne laser, a light attenuator, a polarizer and optical chopper (SR 540) and an optical stage driven by a stepper motor MM 3000 with a resolution of 0.001° (Newport, CA, USA). The reflected beam was detected by photodiode and then processed by the lock-in-amplifier (SR530) [52][53][54][55].

FTIR Analysis
The FTIR spectrum of CGQD, CTA-NCC and CTA-NCC/CGQD solutions are shown in Figure  3. From the spectrum of the CGQD solution, the peak present at 3310 cm −1 represented O-H

FTIR Analysis
The FTIR spectrum of CGQD, CTA-NCC and CTA-NCC/CGQD solutions are shown in Figure 3. From the spectrum of the CGQD solution, the peak present at 3310 cm −1 represented O-H stretching. The peak at 2891 cm −1 was attributed to the C-H stretching. The characteristic band appearing at 1625 cm −1 corresponded to the stretching of C=O of the carboxylic group in graphene quantum dots and the peak at 1037 cm −1 represented the C-O stretching [18].
Crystals 2020, 10, x FOR PEER REVIEW 4 of 12 Next, in the spectrum of CTA-NCC solution, the peak at 3332 cm −1 corresponded to the O-H stretching. The peak at 1617 cm −1 was attributed to the stretching vibration of C-O and the peak at 1056 cm −1 corresponded to C-O stretching [56].
The spectra of CTA-NCC/CGQD solution displayed the properties similar to CGQD and CTA-NCC thin film where there was a broad absorption peak at 3277 cm −1 that was attributed to the O-H stretching vibration. The peak at 2885 cm −1 corresponded to C-H stretching. The peak at 1637 cm −1 can be assigned to C=O stretching and is similar to the peak for both spectrums of CGQD and CTA-NCC. The characteristics band that appeared at 1032 cm −1 corresponded to the stretching of C-O. From the results, it is successfully confirmed that the functional groups of O-H, C-H, C=O and C-O existed in the composite solution. CGQD that are rich with oxygen-containing groups might interact with the hydroxyl groups and oxygen atoms in CTA-NCC through the hydrogen bonding. Furthermore, the possible structure of the composite is also presented in Figure 4.   Next, in the spectrum of CTA-NCC solution, the peak at 3332 cm −1 corresponded to the O-H stretching. The peak at 1617 cm −1 was attributed to the stretching vibration of C-O and the peak at 1056 cm −1 corresponded to C-O stretching [56].
The spectra of CTA-NCC/CGQD solution displayed the properties similar to CGQD and CTA-NCC thin film where there was a broad absorption peak at 3277 cm −1 that was attributed to the O-H stretching vibration. The peak at 2885 cm −1 corresponded to C-H stretching. The peak at 1637 cm −1 can be assigned to C=O stretching and is similar to the peak for both spectrums of CGQD and CTA-NCC. The characteristics band that appeared at 1032 cm −1 corresponded to the stretching of C-O. From the results, it is successfully confirmed that the functional groups of O-H, C-H, C=O and C-O existed in the composite solution. CGQD that are rich with oxygen-containing groups might interact with the hydroxyl groups and oxygen atoms in CTA-NCC through the hydrogen bonding. Furthermore, the possible structure of the composite is also presented in Figure 4.

Optical Studies
The optical properties were analyzed using the absorbance spectrum of the thin film with wavelength range 220-500 nm. The absorbance curves for CGQD, CTA-NCC and CTA-NCC/CGQD thin film are presented in Figure 5.

Optical Studies
The optical properties were analyzed using the absorbance spectrum of the thin film with wavelength range 220-500 nm. The absorbance curves for CGQD, CTA-NCC and CTA-NCC/CGQD thin film are presented in Figure 5. As shown in Figure 5, the absorbance curve for the thin film is different. From the figure, the CTA-NCC/CGQD thin film shows the highest absorption spectra with the absorption peaks at 222.9 nm, 225.9 nm and 264.9 nm. The highest absorption was contributed by the modification of CGQD with CTA-NCC, which confirmed the presence plasmon resonance in carbonaceous material [57]. On the other hand, the lowest absorbance belongs to CGQD thin film. The characteristic peaks that appeared in the nanocomposite thin film can be attributed to the presence of π → π* bond transitions of the carbonyl groups [58]. In addition, it can be observed that the maximum absorption length can be determined from 263.04 nm to 266.63 nm. The results obtained are in the range of the absorption peaks for sulfur doped graphene quantum dots which are at 216-464 nm [59]. As shown in Figure 5, the absorbance curve for the thin film is different. From the figure, the CTA-NCC/CGQD thin film shows the highest absorption spectra with the absorption peaks at 222.9 nm, 225.9 nm and 264.9 nm. The highest absorption was contributed by the modification of CGQD with CTA-NCC, which confirmed the presence plasmon resonance in carbonaceous material [57]. On the other hand, the lowest absorbance belongs to CGQD thin film. The characteristic peaks that Crystals 2020, 10, 875 6 of 12 appeared in the nanocomposite thin film can be attributed to the presence of π → π* bond transitions of the carbonyl groups [58]. In addition, it can be observed that the maximum absorption length can be determined from 263.04 nm to 266.63 nm. The results obtained are in the range of the absorption peaks for sulfur doped graphene quantum dots which are at 216-464 nm [59].
To proceed with the determination of the optical band gap, the relationship between the absorbance and the intensities of the monochromatic light was used [60]. The absorbance, A of samples can be related with the ratio of the initial light intensity on the detector I 0 to the light intensity with the presence of the sample I t .
The absorbance coefficient is another quantity that can be measured. It is a very useful quantity which is used to compare samples of a varying thickness. The absorbance coefficient, α can be expressed as where t is the thin film thickness in meters and the α is in units of m −1 . The energy band gap of these composites has been figured out with the help of the absorption coefficient. To obtain the optical band gap from the absorption spectra, the Tauc relation is used: where k is a constant, h is the Plank's constant, v is the frequency of the incident photon, the multiplication of h and v, hv represents the incident photon energy, E g is the optical band gap and n is the state of transition. In this study we use n = 1/2 for direct transitions. Rearranging Equation (3) gives Based on Equation (4), a graph of (αhv) 2 against hv can be plotted using linear fitting techniques and the optical band gap of the thin films can be determined [61][62][63]. According to Abdulla and Abbo (2012), the intersection of the straight line on the x-axis is taken as the value of the optical band gap [64]. The graph of (αhv) 2 versus hv for CGQD thin film, CTA-NCC thin film and CTA-NCC/CGQD thin film are shown in Figures 6-8 where t is the thin film thickness in meters and the α is in units of m −1 . The energy band gap of these composites has been figured out with the help of the absorption coefficient. To obtain the optical band gap from the absorption spectra, the Tauc relation is used: where k is a constant, h is the Plank's constant, v is the frequency of the incident photon, the multiplication of h and v, hv represents the incident photon energy, Eg is the optical band gap and n is the state of transition. In this study we use n = 1/2 for direct transitions. Rearranging Equation (3) gives Based on Equation (4), a graph of (αhv) 2 against hv can be plotted using linear fitting techniques and the optical band gap of the thin films can be determined [61][62][63]. According to Abdulla and Abbo (2012), the intersection of the straight line on the x-axis is taken as the value of the optical band gap [64]. The graph of (αhv) 2 versus hv for CGQD thin film, CTA-NCC thin film and CTA-NCC/CGQD thin film are shown in Figures 6-8, respectively. As can be seen from the figures, the intersection of the linear fitted line on the x-axis gives the value of the optical band gap. The term band gap is denoting the energy difference between the top of the valence band to the bottom of the conduction band where electrons can jump from one band to another. It necessitates a specific minimum extent of energy for the transition to permit an electron to jump from a valence band to a conduction band and this energy is called as the band gap energy. The optical band gap energies of CGQD, CTA-NCC and CTA-NCC/CGQD are 3.867 eV, 4.143 eV and 4.127 eV, respectively. Based on the result, CTA-NCC had the highest band gap energy among energy band gap results of all three thin films. The variations of optical band gap for the composite thin films, probably due to the presence of CTA-NCC solution as the band gap of CTA-NCC/CGQD, is higher than CGQD. The CTA-NCC/CGQD has a higher energy band gap as compared to CGQD thin film, which is in good agreement with the work reported by , i.e., CTA-NCC increased the optical band gap of the composite [56].

Potential Sensing Analysis
The SPR experiment was first conducted using gold CTA-NCC/CGQD nanocomposite thin film in contact with deionized water (or 0 μM of glucose). The resonance angle for the first part of this experiment was obtained as 54.400°, where this value was used to compare the resonance angle for different concentrations of glucose solution. The SPR experiment was then continued for different concentrations of glucose solution that ranged from 0.005 μM to 0.1 μM. The glucose solution was injected into the cell one after another [65]. The reflectance as a function of incident angle of CTA-

Potential Sensing Analysis
The SPR experiment was first conducted using gold CTA-NCC/CGQD nanocomposite thin film in contact with deionized water (or 0 µM of glucose). The resonance angle for the first part of this experiment was obtained as 54.400 • , where this value was used to compare the resonance angle for different concentrations of glucose solution. The SPR experiment was then continued for different concentrations of glucose solution that ranged from 0.005 µM to 0.1 µM. The glucose solution was injected into the cell one after another [65]. The reflectance as a function of incident angle of CTA-NCC/CGQD thin film, in contact with different concentrations of glucose solutions is shown in Figure 7. From the curves, the resonance angle can be obtained for the glucose concentration of 0.005, 0.01, 0.03, 0.05 and 0.1 µM.
From Figure 9, it can be observed that the resonance angle was shifted to the right and it further increased with the increase of glucose concentration [66,67] nanocomposite thin film has affinity towards glucose, where its incorporation with SPR can be a potential sensor for glucose.

Conclusions
In this study, the CTA cationically modified NCC/CGQD nanocomposite thin film has been successfully fabricated. The functional groups that existed in the thin film were confirmed from the FTIR results. The absorbance value of CTA-NCC/CGQD was the highest with energy band gap of 4.127 eV. The studies of the CTA-NCC/CGQD nanocomposite thin film using the SPR technique have successfully shown that the novel thin film can detect various concentrations of glucose with the lowest detection of 5 nM. This study gives an important idea that the CTA-NCC/CGQD nanocomposite thin film has high potential as an application in sensing glucose when incorporated with the SPR technique and can be further investigated in future studies.

Conclusions
In this study, the CTA cationically modified NCC/CGQD nanocomposite thin film has been successfully fabricated. The functional groups that existed in the thin film were confirmed from the FTIR results. The absorbance value of CTA-NCC/CGQD was the highest with energy band gap of 4.127 eV. The studies of the CTA-NCC/CGQD nanocomposite thin film using the SPR technique have successfully shown that the novel thin film can detect various concentrations of glucose with the lowest detection of 5 nM. This study gives an important idea that the CTA-NCC/CGQD nanocomposite thin film has high potential as an application in sensing glucose when incorporated with the SPR technique and can be further investigated in future studies.