TiO2 Nanoparticles Decorated Graphene Nanoribbons for Voltammetric Determination of an Anti-HIV Drug Nevirapine

In the present study, electrochemical behavior of nevirapine on a glassy carbon electrode (GCE) modified with TiO2 nanoparticles decorated graphene nanoribbons was investigated. Characterization of different components used for modifications was achieved using Fourier transform infrared spectroscopy (FT-IR) and scanning electronmicroscopy (SEM).*e electrochemical behavior of nevirapine on the modified electrodes was examined using cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), chronoamperometry (CA), and differential pulse voltammetry (DPV). A considerable oxidation potential decrease of +352mV for nevirapine in 0.1M phosphate-buffered saline (PBS), pH 11.0, was achieved due to synergy offered by graphene nanoribbons and TiO2 compared to graphene nanoribbons (+252mV) and TiO2 (−37mV), all with respect to the glassy carbon electrode. Under optimized conditions, DPV gave linear calibrations over the range of 0.020–0.14 μM. *e detection limit was calculated as 0.043 μM. *e developed sensor was used for determination of nevirapine in a pharmaceutical formulation successfully.


Introduction
e need for reliable, fast, cheap sensing devices for monitoring biomolecules is important in everyday life. Nevirapine (NVP) (11-cyclopropyl-4-methyl-5, 11-dihydro-6Hdipyrido [3,2-b:2′,3′-e] [1,2] diazepin-6-one) is a nonnucleoside reverse transcriptase inhibitor. Its uses have been highlighted [1,3]. e continued use of nevirapine in treatment and management of HIV/AIDS in the health sector thus calls for a great need to improve on its reported analytical work. Several analytical methods, such as reversed phase high-performance liquid chromatography (RP-HPLC) [4] and capillary electrophoresis [2], have been used for quantifying NVP in pharmaceutical combinations and some real samples. e growing demand for NVP biomolecule stimulates a search for new and even more effective monitoring techniques, which give better insight on their reported analytical work. erefore, it is appropriate to develop new and/or improve on the existing analytical techniques regarding their qualitative and quantitative determination in various matrices, namely, human serum, urine, breast milk, and pharmaceutical formulations. e development of sensors with high sensitivity to promote safety and efficiency during administration to patients who depend on these different biomolecules for life support is important. For example, nevirapine in HIV patients' treatment can coexist together with other biomolecules; hence, simultaneous determination is critical, helping management of antiretroviral treatment by minimizing drug-food interactions. Furthermore, nevirapine is also easily oxidized; hence development of an electrochemical sensor is significant.
Electrochemical techniques based on modification of electrode surfaces can grant some remarkable advantages on applications due to provision of high selectivity and improved sensitivity and stability [5][6][7]. Among the various modifiers, those based on carbon nanomaterials have been shown to be highly promising in electrochemical sensing [8,9]. In recent years, graphene nanoribbons (GNRs), a onedimensional form of graphene strip, have shown promising application in fabrication of composites, batteries, supercapacitors, and fuel cells as shown by a high length-to-width ratio, intrinsic energy band gap, and straight edges, availability of larger surface area, and higher electrical conductivity [10][11][12]. e electron confinement in GNRs has been reported to offer good electronic properties, hence transforming semimetallic to semiconducting properties [13]. Interestingly, GNRs' attractiveness in a variety of electrochemical applications is owed to their outstanding electronic, catalytic, charge transport and surface passivation properties in sensing various organic and inorganic compounds [14,15]. Oxygen functionalities at the edges of nanoribbons after synthesis reportedly offer homogeneous distribution of metal oxide nanoparticles on the surface of carbon [16].
Metallic oxide nanoparticles have been reported to show favorable properties in electrocatalysis [17,18]. In this regard, metal oxide nanoparticles such as TiO 2 have been used in sensor fabrication [19][20][21][22] due to low cost, nontoxicity, large surface area, biocompatibility, strong adsorptive ability, high uniformity, and excellent catalytic activity [22]. Based on the aforementioned properties, the hybridization of GNRs with metal oxide nanoparticles can provide nanocomposite with synergic properties [23].
In the present study, we have fabricated a simple, cheap glassy carbon electrode (GCE) modified with TiO 2/ GNRs as an electrochemical sensor for nevirapine. In our investigation, the atypical properties of GNRs as a good support for making nanoparticle dispersions caused by large surface area, high electrical conductivity, and electrochemical stability in acidic and alkaline electrolytes were an attractive factor. Based on the attractive properties of GNRs and TiO 2 , synthesis of TiO 2 , GNRs, and a TiO 2/ GNRs nanocomposite through environmental friendly methods was carried out. e results showed that the developed sensor had good performances such good reproducibility and good selectivity, owing to the synergic effects of catalysis characters of TiO 2 and GNRs. Besides, to evaluate the applicability of the proposed sensor, it was used to determine nevirapine quantities in a pharmaceutical sample.

Chemicals and Solutions.
All chemicals used were of analytical grade. Phosphate-buffered saline (PBS) solutions (as supporting electrolyte) with different pH values (6)(7)(8)(9)(10)(11)(12) were prepared by mixing standard stock solutions of 0.10 M Na 2 HPO 4 and 0.10 M NaH 2 PO 4 . NaNO 3 , K 3 Fe(CN) 6 2 ] and ethanol (C 2 H 5 OH) were obtained from Associated Chemical Enterprises (South Africa). All solutions were prepared using ultra-Millipore water from Milli-Q Water Systems (Millipore Corp., Bedford, MA, USA). e MWCNTs were purified to remove metal oxide catalysts as reported [18]. e stock solution of appropriate solution of nevirapine was prepared by weighing and dissolving the drug in a 1 : 1 mixture of distilled water and ethanol. e working solutions were prepared by dilution of the stock standard solution with PBS (0.1 M, pH 11). A solution of glucose was prepared by dissolving appropriate amount of glucose in ultrapure water and left for 24 h for mutarotation at room temperature.

Equipment.
Fourier transform infrared spectroscopy (FT-IR, Nicolet 6700 model) was used in IR characterization. e scanning electron microscopy (SEM) image was obtained using a TESCAN Vega TS 5136LM Electron microscope. Cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), chronoamperometry, and differential pulse voltammetry (DPV) were performed using an Autolab potentiostat PGSTAT 302N (Eco Chemie, Utrecht, Netherlands) equipped with NOVA 1.10 software. Sonicator model KQ-250B was used for agitation of samples. e pH of the solutions was measured and adjusted by a ermo Scientific Orion Star A211 pH meter.

Preparation of Graphene Nanoribbons.
Graphene nanoribbons (GNRs) were prepared as described in literature with some little modifications [24,25]. Briefly, a suspension was formed by dissolving 100 mg of MWCNT in 3.4 mL of 98% H 2 SO 4 and then homogenized via ultrasonication for 1 h. ereafter, the suspension solution was placed in an ice bath accompanied with vigorous stirring and 75 mg of NaNO 3 was then added. Next, 450 mg of KMnO 4 was added to the suspension. Upon completion of reaction, 20 mL of 5% sulphuric acid solution was added and reaction allowed to cool. As soon as bubble formation started, 2 mL of 30% H 2 O 2 was added dropwise. After approximately 30 min, centrifugation and washing with 5% nitric acid three times and deionized water five times were carried out followed by filtration and drying in an oven at 90°C for 12 h under vacuum. e prepared GNRs were ascertained to contain oxygen-containing functional groups.

Preparation of TiO 2
Nanoparticles. e sol-gel process was used for the synthesis of TiO 2 nanoparticles at 80-90°C. Briefly, 2.195 g of titanium acetate dihydrate was dissolved in 100 mL ethanol followed by stirring in ambient atmosphere. 1.122 g KOH was dissolved in 10 mL distilled water and then added to the titanium acetate dihydrate-ethanol solution dropwise under continuous stirring. e mixed solution turned into a jelly form and a milky white solution was obtained after a few minutes. Furthermore, the mixture was then heated for 3 h at 80-90°C without stirring. Centrifugation was applied to the resulting suspension to obtain intended product. Finally, the mixture was washed with ultra-Millipore water in an ultrasonic bath and then the powder dried at 70°C overnight.

Preparation of TiO 2 /GNR and TiO 2 /GNR/GCE.
e suspension of TiO 2 (1 mg/mL) was added to GNRs (2 mg/mL) DMF solution (1 : 2) and sonicated for 4 h at room temperature to obtain the TiO 2 /GNR nanocomposite homogeneous suspension. e GCE was polished to a mirror finish with alumina slurry (0.3 μm). e electrode was sonicated in Millipore water for 5 min during the three successive cleaning stages and finally dried in a stream of nitrogen. e drop dry technique was employed for electrode modification. e optimized volume of the composite, 5 µL of TiO 2 /GNR suspension, was drop-casted on the surface of GCE and dried in the oven to obtain TiO 2 /GNRs/ GCE. e fabrication procedure of the electrochemical sensor is shown in Scheme 1.

Assay of Nevirapine Tablets.
Five nevirapine tablets (each containing 200 mg per tablet) were obtained from local commercial sources. e drugs were crushed to obtain a finely homogenized powder using the mortar. A portion of the powder suitable to prepare 1 mM was weighed and transferred into a 10 mL volumetric flask containing water, sonicated to allow dissolution, and then diluted to the mark. Appropriate amounts of solutions were taken and analyzed by DPV method.

Characterization.
FT-IR spectra of TiO 2 , GNR, and TiO 2 /GNR composite are shown in Figure 1(a). e characteristic peaks in (b) were assigned as follows: 1654 cm −1 attributed to the stretching vibrations of carboxyl (-COOH) and at 1130 cm −1 corresponding to C-O-H bending and C-OH. e presence of such peaks can be used to explain the hybrid structure of graphene nanoribbons. After making the composite, the small peaks observed in (a) disappeared and peak intensity of C � O in GNR/TiO 2 shows some changes as a result of incorporation of nanoparticles ( Figure 1(c)). e wavenumber region 3400 cm −1 is the stretching vibration of the hydroxyl group in all spectra. e morphology of prepared materials was observed using SEM ( Figure 1(b)). As shown in Figure 1(a) TiO 2 nanospheres are aggregated in larger agglomerates. Figure 2(b) shows long curved rod-like structures of GNR [26]. e TiO 2 structures can be seen to be closely and homogeneously grown on the GNR (Figure 2(c)). For comparison, the GCE was included (Figure 1(d)).
EIS studies were done in the [Fe(CN) 6 ] 3−/4− redox system to deduce the resistance to electron transfer. Nyquist plots ( Figure 2) are shown by semicircle portions at high frequencies, while diffusion-limiting steps of the electrochemical process are shown by linear parts at lower frequencies. e equivalent circuit model used to fit impedance data into R et values is shown as inset in Figure 2. From Table 1, it can be seen that modification of the GCE showed lowering of R et except for TiO 2 with a larger diameter (2.2 kΩ) suggesting that TiO 2 acted as an insulating layer and barrier [18]. e incorporation of GNRs significantly lowered R et values of GCE by 33.3% and that of TiO 2 in composite by 77.8%. e order as deduced from impedance values is TiO 2 /GCE > GCE > GNR/GCE > TiO 2 /GNR/GCE. e changes in R et suggested proper modification of modified electrodes; hence, the electrode based on TiO 2 / GNR composite was used throughout the study. CV involving modified electrodes was performed in 1 mM [Fe(CN) 6 ] 3−/4− and 0.1 M KCl electrolyte at a scan rate of 100 mV/s (Figure 2(b)). Redox peaks were observed on all tested electrodes with different peak current (i p ) and change in peak potential separation (ΔE P ) ( Table 1). e effect of different modifiers was shown by ΔE P , with lower ΔE P showing better electron transfer ability. After drop-casting TiO 2 /GNR on surface of GCE, i pa , increased i pc , and reduced E p were observed compared to i pa , i pc , and E p of GCE. e observation is attributed to TiO 2 /GNR particles offering a large surface and good conductivity. Furthermore, the GNR acted as a suitable pathway to shuttle electrons; hence, improved peak currents were displayed. TiO 2 /GCE showed a sluggish electron transfer process. e electrodes gave peak potential differences with the following trend: where i p is the peak current, n is the number of electrons transferred (n � 1) during the redox couple Fe (II)/Fe (III), Dis the diffusion coefficient of the analyte in solution (7.6 × 10 −6 cm 2 s −1 ), C is the solution concentration in mol/cm 3 , A eff is the effective surface area, and v is the scan rate (V/s). Voltammograms at different scan rates (50-300 mV/s) were run and gave a linear plot for i p versusv 1/2 . From the slope, TiO 2 /GNR/GCE had an effective surface area of 0.208 cm 2 relative to the GCE area of 0.0712 cm 2 [27]. e electrochemical reaction of a biomolecule is usually affected by pH; hence, influence of pH on i pa and E pa was examined by cyclic voltammetry in the pH range of 7-12 in phosphate-buffered solutions at 100 mV/s (Figure 3). From the voltammograms in Figure 3, it can be seen that increasing pH results in biomolecule being easily oxidized, causing peak current and peak sharpness to increase as well. It should be noted that voltammograms in acidic media were not included, since no oxidation could be observed as a result of protonation on biomolecule. e repulsive forces between the biomolecule and the electrode surface of TiO 2 /GNR/GCE slowed down arrival of biomolecules to the surface. As depicted in the inset, the current increases up to maximum of pH 11 and then decreases. erefore, pH 11 was chosen as optimum pH for further studies. e peak potential was affected by pH as shown in the inset (Figure 3). e E pa values of nevirapine were shifted to less positive values with increasing pH, showing deprotonation in the oxidation process at higher pHs [18]. A linear relationship was observed between E pa and pH for the biomolecule with the following regression equations: E pa (mV) � −63.9 pH + 1123; R 2 � 0.9508 for nevirapine. e slope for nevirapine (63.9 mV per pH)≈59.6 mV per pH indicates involvement of equal number of protons and electrons during oxidation reactions. is is in agreement with previous reports [5,6,28].

Cyclic Voltammetry Behavior of Nevirapine.
It is pertinent to study the behavior of modified electrodes in PBS alone before any determinations (Figure 4(a)).
ere was an increase in the peak background corrected current from GCE to TiO 2 /GNR/GCE. Graphene nanoribbons assisted in faster electron transport as well as providing a large surface area compared to TiO 2 alone. e electrochemical oxidation behavior of nevirapine was investigated by CV in 0.1 M PBS (pH 11.0) at modified electrodes (Figure 4(b)). e obtained voltammograms indicated electroactiveness of the compound and irreversible behavior. As shown in Figure 4(b), i pa increased for nevirapine following modification on GCE; TiO 2 /GCE < GNR/GCE < TiO 2 /GNR (Table 1). e GCE exhibits very low i pa at different potentials, whereas with the TiO 2 modified glassy carbon electrode under same conditions, i pa increases compared to that of GCE (Figure 4(b)). On the other hand, when GNR was deposited on GCE, i pa of nevirapine was found to be 2.2 times greater compared to GCE (Figure 4(c)). is may be attributed to fast electron transfer as well as high surface area available on the GNR for electrooxidation of nevirapine.
e electrocatalytic oxidation of nevirapine with TiO 2 /GNR composite modified GCE showed improved behaviors. An increase of i pa , 186% for nevirapine, was deduced (Figure 4(d)), accompanied with a decrease in overpotential (352 mV) compared to glassy carbon electrode. e decrease in overpotential is caused by the synergetic effect of GNRs and TiO 2 . TiO 2 nanoparticles help in electrocatalytic oxidation, whereas the GNRs provide larger surface area for TiO 2 as well as for nevirapine and then allow faster electrode kinetics. Hence, the TiO 2 /GNR/GCE was used for the determination of nevirapine in the presence of interferent glucose in 0.1 M PBS (pH 11.0) using CV (Figure 4(c)). CV shows that it is possible for the biomolecules to be determined simultaneously with no overlap as shown by ΔE pa of 410 mV.
In order to study nature of the electrode process, the influence of scan rate (v) on i pa and E pa for a mixture of 0.1 mM glucose (interferent) and nevirapine in a 0.1 M PBS (pH 11.0) was examined by CV with scan rate ranging from 50 to 400 mV/s ( Figure 5). From Figure 5 inset (a, b), i pa is directly proportional to scan rate (v) with a correlation coefficient of 0.9991 (nevirapine) and 0.9948 (glucose) showing electrochemical behavior of biomolecules on TiO 2 /GNR modified electrode as adsorption controlled processes [4,29,30]. e adsorption mechanism process has also been reported for nevirapine [5,6]. e presence of functional groups at edges of GNR in the composite might have facilitated adsorption of nevirapine by π-π stacking, hydrogen bonding, and covalent interactions [31]. Furthermore, plot of log i pa against log v ( Figure 5(c), inset) gave a linear plot with the following equation: log i pa (µA) � 0.804 log v (mV/s) + 0.493 for nevirapine. e slope of nevirapine (0.804) is greater than the theoretical value of 0.5 V/s, which further confirms that electrochemical oxidation exhibits mixed behavior [32]. Additionally, from voltammograms, both biomolecules show i pa increasing and E pa shifting in the less positive value with increase in scan rate. e obtained results depict a totally irreversible electrochemical process.

Catalytic Rate Constants.
Catalytic rate constants for nevirapine at TiO 2 /GNR/GCE ( Figure 6) were determined by chronoamperometry based on favorable oxidation results from voltammetry. e rate constants were evaluated using the following equation [33]: Journal of Chemistry where I cat and I buf are currents on TiO 2 /GNR/GCE in presence and absence of nevirapine, respectively, c � kC o t (C o is the bulk concentration of nevirapine), and erf is the error function. e error function is almost equal to 1 when c exceeds 2 and hence equation (2) reduces to the following equation: where k is the catalytic rate constant (cm 3 /mol/s) and t is the time elapsed in (s). e catalytic rate constants for nevirapine were calculated based on information obtained from chronoamperometry ( Figure 6). A plot of I cat /I buf versus t 1/2 for oxidation of nevirapine gave a linear plot.
e calculated values of catalytic constants were 7.9 × 10 5 cm 3 /mol/s. e calculated values further elucidate sharp features of catalytic i pa of nevirapine at the surface of TiO 2 /GNR/GCE.

Nevirapine Determination by Differential Pulse
Voltammetry. Determination of nevirapine in the absence of interferents was investigated (Figure 7(a)). In all cases, i pa increased with increase in concentration of analyte as shown by voltammograms and the analytical parameters shown in Table 2.

Behavior of Nevirapine in the Presence of Interferent.
e effect of different concentrations of analyte on electrochemical response of nevirapine and interferent glucose on TiO 2 /GNR modified electrode was investigated (Figures 8(a)-8(c)). Different cases were studied, where the concentration of only one compound was varied, while the concentration of the other compound was kept constant. In the first case, the concentration of nevirapine was changed,    Journal of Chemistry while the concentration of glucose was kept constant (Figure 8(a)). As shown in Figure 8(a), DPP was obtained with different concentrations (0.20-1.4 µM) of nevirapine in presence of 0.20 µM glucose on TiO 2 /GNR modified electrode. e oxidation peak currents increased linearly when the bulk concentration of nevirapine was also increased. Furthermore, the oxidation peak currents of glucose remained the same as the numbers of cycles were increased. e inset (Figure 8(a)) shows a plot of i pa versus different concentrations of nevirapine.
On the other hand, Figure 8(b) shows DPVs obtained in different concentrations (0.20-1.4 µM) of glucose in presence of 0.20 µM nevirapine. e same trend was observed with the glucose molecule, where i pa corresponding to oxidation of glucose showed linearity with an increase in the bulk concentration of glucose, and i pa for the oxidation of nevirapine remained the same as the number of cycles increased. A plot of i pa versus different concentrations of glucose is shown as inset (Figure 8(b)). e analytical parameters are shown in Table 2. Figure 8(c) shows DPVs obtained at TiO 2 /GNR modified electrode when equal concentrations of nevirapine and glucose were simultaneously changed. As shown in Figure 8(c), clearly resolved voltammetric signals were observed for addition of increasing concentrations with calibration curves as insets.
e analytical parameters of the calibration plots are listed in Table 2. From the use of TiO 2 /GNR modified electrode, it is interesting to note the detection of glucose and nevirapine is independent as shown by detection limits and clear peak potential separation. A comparison of the detection limits of the present method with those reported in recent years at other surfaces is tabulated in Table 3. e values of limit of detection (LOD) were calculated using 3s/k (where s is the standard deviation of the blank and k is the slope of the calibration curve) and using Figures 7, 8(a)-8(c), and 9. It can be clearly seen that the LOD observed using TiO 2 / GNR is much lower. e synergic effects between TiO 2 and excellent physicochemical properties of GNR make the composite highly suitable for nevirapine and glucose sensing.
3.7. Impedimetric Determination. EIS was also employed to study the oxidation behavior of nevirapine at TiO 2 /GNR/ GCE (Figure 9(a)). e results showed that, in absence of nevirapine, Nyquist diagram comprises a large semicircle at high frequencies due to large charge transfer resistance (2.5 kΩ). In the presence of interferent glucose and biomolecule nevirapine, diameter of the semicircle decreases, confirming the electrocatalytic capability of composite. However, glucose showed a smaller R et value of 0.5 kΩ, while nevirapine had a large value of 0.75 kΩ. e lower R et in glucose showed that TiO 2 /GNR provided more favorable orientation and conductive pathway for transfer of electrons. Overall, it can be concluded that TiO 2 /GNR sensor was specific towards oxidation of glucose and nevirapine.
e EIS responses of fabricated sensors to different concentrations of nevirapine were carried out further (Figure 9(b)). It can be seen in Figure 9(a) that R et decreased with increasing nevirapine concentration. e observations can be attributed to TiO 2 /GNR offering a suitable platform for oxidation of the biomolecule. Increasing the concentrations of biomolecule, as shown Figure 9(b), results in generation of more electrons (nevirapine 2e − ), causing R et values to decrease markedly, consequently enhancing the electrode kinetics. Under the optimized conditions, the electrochemical sensor gave parameters as shown in Table 2.
e LODs are comparable to both single detections of analytes when one is constant and in equal concentrations of the analytes. e results obtained clearly elucidate possible impedimetric detection of nevirapine using GNR and TiO 2 in the composite.       Journal of Chemistry 3.9. Reproducibility Studies. When developing sensors, reproducibility is a significant process to be investigated. e TiO 2 /GNR sensor was investigated through five repetitive measurements of 0.1 mM of nevirapine in PBS (pH 11) by CV ( Figure 10). e relative standard deviation of i pa responses to nevirapine sensing was less than 3.5%, indicating good reproducibility.

Nevirapine in Pharmaceutical Formulations.
e results obtained from the developed sensor for determining amount of nevirapine in commercially available tablets are shown in Table 4. ese were calculated after running working solutions prepared after suitable dilution with DPV. e results of analysis were found to be satisfactory as shown by high recovery values.

Conclusions
A selective nevirapine electrochemical sensor was developed based on the modification of a GCE with metal oxide and graphene nanoribbons nanocomposite. e successful preparation of GNR and TiO 2 was confirmed by FT-IR and SEM. EIS confirmed superior electrochemical properties of the prepared TiO 2 /GNR/GCE in comparison to bare GCE. It was also illustrated that TiO 2 /GNR/GCE offered a low potential during detection by cyclic voltammetry. e obtained TiO 2 /GNR/GCE composite exhibited good reproducibility during analysis. e developed method was successfully applied to the quantification of nevirapine in a pharmaceutical formulation. e present study provides a general strategy for monitoring drug-food behavior during electrochemical applications.

Data Availability
Data will be shared through the authors' library repository if accepted.