Insights into the Design of An Enzyme Free Sustainable Sensing Platform for Efavirenz

Abstract

In this study, a new hybrid sensor was developed using titanium oxide nanoparticles (TiO₂-NPs) and nafion as an anchor agent on a glassy carbon electrode (GCE/TiO₂-NPs-nafion) to detect efavirenz (EFV), an anti-HIV medication. TiO₂-NPs was synthesized using Eucalyptus globulus leaf extract and characterized using ultraviolet–visible spectroscopy (UV–VIS), scanning electron microscopy (SEM), X-ray diffraction (XRD), and energy-dispersive spectroscopy (EDS). The electrochemical and sensing properties of the developed sensor for EFV were assessed using cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS). The current response of GCE/TiO₂-NPs-nafion electrode towards the oxidation of EFV was greater compared to the bare GCE and GCE/TiO₂-NPs electrodes. A linear dynamic range of 4.5 to 18.7 µM with 0.01 µM limit of detection was recorded on the electrode using differential pulse voltammetry (DPV). The electrochemical sensor demonstrated good selectivity and practicality for detecting EFV in pharmaceuticals (EFV drugs) with excellent recovery rates, ranging from 92.0–103.9%. The reactive sites of EFV have been analyzed using quantum chemical calculations based on density functional theory (DFT). Monte Carlo (MC) simulations revealed a strong electrostatic interaction on the substrate-adsorbate (GCE/TiO₂-NPs-nafion-EFV) system. Results show good agreement between the MC computed adsorption energies and the experimental CV results for EFV. The stronger adsorption energy of nafion onto the GCE/TiO₂-NPs substrate contributed to the catalytic role in the signal amplification for sensing of EFV. Our results provide an effective way to explore the design of new 2D materials for sensing of EFV, which is highly significant in medicinal and materials chemistry.

1. Introduction

Efavirenz (EFV) is known as Sustiva in the marketplace, and it is a cost-effective option for the first line of antiretroviral treatment [1]. As such, it plays a crucial role in HIV treatment strategies [2]. The combination of EFV with other drugs of abuse for recreational purposes triggers toxicity [3]. Additionally, EFV is generally used in combination with other drugs to prevent HIV infections. It is also known to undergo photoreactions that could be exploited for use in photodynamic therapies, however, there is limited information on the photo-catalytic reactivity of EFV. Therefore, the development of an analytical tool to measure qualitative and quantitative properties of pharmaceutical products such as EFV is highly important.

For the measurement of EFV, different analytical tests have been established, including chromatographic [1,4,5,6,7,8,9,10,11], spectrophotometric methods [2,12], and electrophoresis [13,14]. These assays, on the other hand, are expensive and necessitate lengthy and arduous experimental procedures, solvent extraction, sample pre-treatment, optimized instruments, and a qualified analyst [3,15]. In contrast, electrochemical methods have attracted considerable interest in the field of drug analysis due to their simplicity, low cost, rapid reaction times, and high sensitivity without the necessity of extensive extraction or pre-treatment steps. However, very few reports are available on the electroanalytical methods for the analysis of the EFV [3,16,17,18]. The use of the nanostructured materials to modify the electrodes’ surfaces results in significant signal amplification, enhancing the sensitivity, stability, and selectivity of the synthesized sensors/biosensors. These enhancements are attributed to the critical role of the nanomaterials, which can create the effect of synchronization between the catalytic activity, electrical conductivity, and biocompatibility based on accelerating the signal transduction [19].

The method employed in the synthesis of nanoparticles involves chemical and green (biological) assays. In recent years, the green route of synthesizing nanoparticles has attracted much attention because it is eco-friendly, less toxic, and cost-effective in contrast to chemical methods of synthesis [20,21,22]. As a result, green synthesis of TiO₂-NPs has been reported utilizing Jasmine flower extract [20], Syzygium cumini, and Azadirachta indica leaf extracts [23,24], black pepper (Piper nigrum), coriander (Coriandrum sativum) [25], clove (Syzygium aromaticum), and orange peel [26]. The synthesis of TiO₂-NPs has been reported using Eucalyptusglobulus leaf extract with no application to electrode modification for sensing of EFV [27].

TiO₂-NPs are among the most extensive metal oxide nanomaterials employed in modification of electrodes due to high surface area, thermal stability, biocompatibility, low toxicity, and a widespread band gap [28,29,30]. However, because of its great antifouling capability, high cation permeability, and strong adsorption ability, nafion has been widely used in electrode modification procedures [30,31]. Previous studies have shown that the integration of nafion with TiO₂-NPs improves conductivity and sensitivity of sensors for the detection of analyte [19,30].

In this study, we present, for the first time, a novel and cost-effective sensor for the detection of EFV based on bioinspired green synthesized TiO₂-NPs with nafion as an anchoring agent, supported on a glassy carbon electrode (GCE/TiO₂-NPs-nafion). In order to explain the interactions between EFV and GCE/TiO₂-NPs and GCE/TiO₂-NPs-nafion electrode surfaces, adsorption energies were computed from MC simulations. DFT calculations were also carried out to better understand the reactive sites and electronic characteristics of EFV [32]. A complementary approach using a computational perspective allowed for an integrated cross-disciplinary approach which combines in silico design with experimental measurements, presenting a powerful new paradigm that solves issues in the development of novel sensors. A combination of simulations with quantum mechanical DFT calculations were therefore used as a predictive tool to guide the design of the electrode surface modification.

2. Results and Discussion

2.1. Computational Results

2.1.1. Frontier Molecular Orbitals from DFT Calculations

The DFT calculations of the 3D modelled EFV structure used to obtain the frontier molecular orbital helped to determine which part of the molecule contributed to the oxidation or reduction reaction [32]. Therefore, to support the oxidation mechanism of EFV in a more accurate way and to evaluate the chemical reactivity of EFV, the energy difference between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO) of EFV was computed from DFT calculations. As seen in Figure 1, the HOMO and LUMO electron density is concentrated mostly on the benzoxazine ring with the lowest energy gap, E_gap = 5.171 eV, making it more reactive, suggesting the benzoxazine ring as the main active site of EFV. The HOMO orbital energies describe the ability of molecules to donate electrons in general. The smaller value of band gap energy (E_gap) in our situation provides a stronger reactivity for removing an electron from the highest occupied orbital to the lowest unoccupied orbital, implying that oxidation processes are more likely to occur because oxidation reactions tend to displace electrons from the HOMO. Therefore, less energy is required for the oxidation (loss of electrons) in comparison to the reduction (gain of electrons) reactions. Similarly, LUMO + 1 exhibited almost similar distribution of electron density. In contrast, LUMO + 2 and HOMO − 2 illustrated electron density over the whole molecule with only a small part localized on the oxazine ring for HOMO − 2 with the highest energy gap, ∆E_gap = 7.325 eV. The alkyne, cyclopropyl, and carbonyl functionality of the oxazine ring were preferred locations for the HOMO + 1 electron density.

2.1.2. Monte Carlo Simulation

In order to examine the interaction patterns of EFV, a simulated annealing job was applied using the adsorption locator module of the Monte Carlo (MC) simulations to the substrates GCE/TiO₂-NPs and GCE/TiO₂-NPs-nafion loaded with EFV as an adsorbate molecule. For finding a low energy adsorption site, a MC-based function was utilized to find the most favorable configuration.

Table 1. Calculated adsorption energy for the substrate-adsorbate systems.

Substrate-Adsorbate	Adsorption Energy/kcal mol
GCE/TiO2-NPs-EVF	−20.455
GCE/TiO2/nafion-EVF	−23.962

depicts the lowest adsorption energy values for the adsorbate-substrate system modelled in a systematic way under periodic boundary conditions.

Adsorption energies were all negative, indicating that adsorption occurs exothermically and spontaneously when the relaxed adsorbate components are adsorbed on the substrate, indicating that TiO₂-NPs interact strongly with the GCE substrate. GCE/TiO₂-NPs-nafion-EVF interaction is supported by a more highly negative adsorption energy (−23.962 kcal/mol) in contrast to a significantly higher value (−20.455 kcal/mol) in the absence of nafion.

As illustrated in Figure 2, a comparison of the electrode systems was performed to examine the adsorption behavior and energy differences in relation to electrode modification phases. The effect of intramolecular and intermolecular energies on the behavior of nafion adsorption was investigated using these models.

In Figure 2A, the EFV is positioned planarly above the TiO₂-NPs surface. This is due to the reduced van der Waals forces of interaction between EFV and the surface of TiO₂-NPs surface. Nafion, on the other hand, binds more strongly to TiO₂-NPs (Figure 2B). This accounts for the lower adsorption energy (−29.050 kcal/mol) and also supports a stronger EFV adsorption with the GCE/TiO₂-NPs-nafion substrate. This is demonstrated by the heat field maps presented in Figure 2C,D, with the more likely adsorption locations highlighted in red. As indicated by the “observed clouds”, the level of dispersion fields of the adsorbed molecule’s location correspond to the GCE surface.

2.2. Experimental Results

2.2.1. Spectroscopic and Morphological Characterization

Figure 3A shows the UV-vis spectrum of synthesized TiO₂-NPs with absorption bands observed at 365 nm, which is close to the band (356 nm) reported in literature [23]. The SEM morphology of TiO₂-NPs presented in Figure 3B revealed agglomeration of nearly spherical shaped grains of TiO₂-NPs on nanoclusters. The XRD pattern of TiO₂-NPs shown in Figure 3C was conducted in the range of 10–80^θ. Characteristics peaks of 29.487, 44.281, 56.427, 63.641, 64.965, and 74.270 indexed to 101, 210, 211, 204, 213, and 215 diffraction planes were recorded. The crystalline size (D) was calculated as 5.7 nm for TiO₂-NPs Anatase from 101 plane using the Debye–Scherrer equation, which is within the reported range (2.3–8.5 nm) [23,33].

The EDS spectra in Figure 3D presents the elemental composition of TiO₂-NPs, confirming the presence of carbon (C), oxygen (O), and titanium (Ti). The emission of C in TiO₂-NPs spectrum might have resulted from the leaf extract.

2.2.2. Electrochemical Behavior of Electrodes

Cyclic Voltametric Measurements

The charge transfer of (GCE, GCE/ TiO₂-NPs, and GCE/TiO₂-NPs-nafion) electrodes were investigated using CV in PBS (Figure 4A) and in a redox probe solution of [Fe(CN)₆]^3−/4− prepared in 0.1 M PBS, pH 7 (Figure 4B) at a scan rate of 25 mV/s. Figure 4B shows the CV responses at the electrodes with an decreased peak current noticed at GCE/TiO₂-NPs-nafion in contrast to other electrodes, suggesting an electron withdrawing effect due to the blocking of electrode surface area by nafion film [31].

Table 2. Summary of CV parameters recorded for the bare and modified electrodes in 5 mM [Fe (CN)₆]^3−/4−.

Electrodes	Epa (V)	Ipa (µA)	Epc (V)	Ipc (µA)	Epa-Epc (ΔE, V)	Ipa/Ipc	A (cm2)
GCE	0.2949	20.361	0.0874	−22.000	0.2075	−0.9255	0.0017
GCE/TiO2-NPs	0.2510	20.770	0.1387	−21.697	0.1123	−0.9572	0.0018
GCE/TiO2-NPs-nafion	0.3511	11.749	0.0288	−12.5823	0.3223	−0.9337	0.0010

summarizes the CV parameters, including peak-to-peak separation (∆E_p), ratio of anodic and cathodic peak current (I_pa/I_pc), active surface area (A), oxidation and reduction peak potential (Epa and Epc), and peak currents recorded at the electrodes. The I_pa/I_pc values were approximately unity, suggesting a reversible electrochemical process. The ∆E_p suggest more sluggish electron transport process at GCE/TiO₂-NPs-nafion.

EIS Measurement

The electrode materials’ interfacial charge transfer properties were evaluated by EIS in a [Fe (CN)6]^3−/4− redox probe solution produced in 0.1 M PBS, pH 7.0. Figure 4C shows the Nyquist plots obtained for the various electrodes at a fixed potential of 0.2 V in the frequency range of 0.1 Hz to 100 kHz, while Figure 4D shows the analogous circuit used in the fitting of EIS data. A high R_ct value was recorded at GCE/TiO₂-NPs-nafion electrode, suggesting a resistance to charge electron transport property at the electrode.

Table 3. EIS data obtained on various electrodes in 5 mM [Fe (CN)6]^3−/4− solution at +0.2 V fixed potential.

Electrode	Rs (Ω)	Rct (Ω)	Y° (µΩ*S^n)	n	χ²
GCE	172.31 (2.74)	3765 (2.26)	0.94 (11.29)	0.85 (1.54)	0.3610
GCE/TiO2-NPs	176.64 (2.04)	1279 (2.23)	1.50 (13.67)	0.85 (1.86)	0.2244
GCE/TiO2-NPs-nafion	217 (1.22)	14179 (1.25)	1.04 (3.74)	0.840 (0.57)	0.0896

summarizes the impedimetric data, with Y° denoting the magnitude of Q and n being the deviation with parameter accordingly. Depending on the homogeneity of the electrode surface, the behavior of electrodes as a capacitor, inductor, or insulator is proposed when n is 1, −1, or 0, respectively [34], with n < 1 indicating that the electrodes are pseudo capacitive.

2.2.3. Effect of Scan Rate Variation

To better understand the kind of electrode reaction, the influence of scan rate variation on the peak potential and current of the redox probe at GCE/TiO₂-NPs-nafion electrode was investigated. Figure 5A shows the CVs recorded at different scan rate in the range of 25 to 200 mV/s. An increased redox peak current and shifting of peak potentials of the redox probe with increase in scan rate was noticed. The relationship between peak current versus square root of scan rate (v^1/2) and log of peak current versus log of scan rate (v) (Figure 5B,C respectively) displayed an acceptable linear plot with 0.999 correlation coefficient (R²) value, suggesting a diffusion-controlled electrode reaction [34]. The slope value of Figure 5C is close to 0.5 for an ideal diffusion-controlled electrode process.

The plot of peak current (I_pa and I_pc) versus scan rate (v) depicted in Figure 5D revealed a linear pattern with a regression coefficient of 0.99. Based on the slope value of Figure 5D, the electrochemical surface coverage concentration (Γ) of the redox probe was estimated to be 1.49 × 10⁻³ mol/cm³ according to Equation (1).

$$I_p=\frac{n^2{F}^2\Gamma Av}{4RT}$$

(1)

where I_p, n, F, Γ, A, v, R, and T denotes peak current in amperes, number of electrons involved, faradays constant in coulombs, surface coverage concentration in molcm⁻², surface area of electrode in cm², scan rate in V/s, gas constant in J/mol/k, and temperature in Kelvin respectively. The investigation of apparent number of electrons (n) involved, and the charge transfer coefficient (α) of the GCE/TiO₂-NPs-nafion electrode, was conducted using the congruent changes in the redox peak potential (anodic and cathodic peak potentials) of Figure 5A as a function of the logarithm of scan rate. Figure 5E shows the plot of peak potentials versus logarithm of scan rates represented with two resulting linear slopes equal to $\frac{-2.3RT}{\mathsf{\alpha }\mathrm{nF}} \mathrm{and} \frac{2.3RT}{\left(1-\alpha \right)nF}$, based on the Laviron equation [35]. Based on the slope and employing the simultaneous equation, n and α were found to be 0.8186 (approximately 1) and 0.51, respectively. For an ideal diffusion-controlled reaction, α ≈ 0.50.

2.3. Electrocatalysis of EFV

2.3.1. Effect of pH Solution

The effect of supporting electrolyte pH solution on the sensitivity of the GCE/TiO₂-NPs-nafion electrode to 47.6 µM EFV was examined using CV, at 25 mV/s within potential window of −0.2 to 1.4 V in pH ranging from 5.0–8.5 (Figure 6A). Optimal current was obtained at pH 7.0. Thus, pH 7.0 was used through the study. Figure 6B presents the CV response in PBS alone at a scan rate of 25 mV/s at GCE/TiO₂-NPs-nafion electrode.

2.3.2. Electrochemical Behavior of EFV Bare and Modified Electrodes

The electrochemical behavior of 47.6 µM EFV on the GCE, GCE/TiO₂-NPs, and GCE/TiO₂-NPs-nafion electrode surfaces was examined using CVs and EIS. TiO₂-NPs modified electrode was coated with nafion to prevent electrode fouling and improve the catalytic activity of the TiO₂-NPs towards EFV. The CVs of 47.6 µM EFV in 0.1 M PBS at 25 mV/s had a peak potential of 1.2 V on GCE and GCE/TiO₂-NPs-nafion, and 1.1 V on GCE/TiO₂-NPs are shown in Figure 7A. According to the results, the electrocatalysis of EFV is an irreversible oxidation reaction, which is consistent with previous literature reports [3]. The amplified peak current observed at GCE/TiO₂-NPs-nafion electrode indicates an improved electrocatalytic capability of TiO₂-NPs nafion for oxidation of EFV in the presence of nafion with a rapid rate of electron transport and high current sensitivity.

The behavior of the electrode–electrolyte interface during the oxidation of 47.6 µM EFV was examined to gain a better understanding of the electron transport process.

Figure 5B shows the Nyquist plots obtained at the electrodes, whereas Figure 7C shows the Randle’s circuits ([R(QR)]) used in the fitting of EIS data after numerous circuit iterations. The parameter obtained on fitting the EIS Nyquist plots are presented on

Table 4. Impedance data obtained for electrodes in 47.6 µM EFV at a fixed potential of 1.2 V (vs. Ag/AgCl, standard KCl). The % errors of the data fitting are shown in parenthesis.

Electrodes	Rs (Ω)	Rct (KΩ)	Y (µΩ*S^n)	n	χ²
GCE	263(4.41)	712 (3.37)	0.36 (3.56)	0.76 (0.61)	0.42927
GCE/TiO2-NPs	257(4.32)	671 (4.31)	0.27(5.10)	0.86 (0.81)	0.8977
GCE/TiO2-NPs-nafion	222(5.22)	593 (7.17)	1.14 (4.58)	0.73 (0.92)	0.9142

, the values in parenthesis and the chi square values (χ²) confirmed successful fitting of the EIS data. The lowest R_ct value was achieved at GCE/TiO₂-NPs-nafion, which agrees with the CV measurements. The summary of EIS data of the electrode is presented in Table 4. As a result, all electrodes have n < 1, which indicates that the electrode–electrolyte interface exhibits a near capacitive response to EFV electro-oxidation.

According to literature, EFV undergoes one electron transfer (n), hence there is one proton involved in this reaction [4]. At GCE/TiO₂-NPs-nafion, the electrooxidation of EFV is a one-electron and one-proton process. Using literature reports and DFT calculations, the following reaction mechanism is proposed for the oxidation of EFV (Scheme 1).

2.3.3. Analytical Performance of GCE/TiO₂-NPs-Nafion Electrode

Figure 8A shows the DPV for various concentrations of EFV measured within the 0.8 and 1.4 V potential window. From 4.5 to 18.7 µM, the EFV oxidation peak current decreased linearly with increasing concentrations. For EFV, the computed values of LOD and LOQ of GCE/TiO₂-NPs-nafion were 0.01 and 0.03 µM, respectively. The LOD obtained with other EFV sensors is compared in

Table 5. Comparison present sensor with reported sensors.

Electrodes	Methods	Supporting Electrolyte	Peak Potential mV	Linearity (µM)	LOD (µM)	Ref
PGE/dsDNA	AdSV	pH 7.2 PBS	1001	6.33–7.60	1.9	[17]
Thin Hg-Film	AdSV	2.0 × 10−3 NaOH	−280		0.03	[18]
ErGO-Pt/Nafion/EPPG	SWV	PBS pH 7.2	1160	0.05–150	1.8 × 10−3	[16]
NiO–ZrO2/GCE	DPV	PBS 7.2	1200	0.01–10	1.36 × 10−3	[3]
GCE/TiO2-NPs-nafion	DPV	0.1 M PBS, pH 7	1010	4.5–18.7	0.01	This work

.

2.4. Real Sample Analysis

The developed sensor’s practical utility for determining EFV in real samples was tested using a pharmaceutical sample (Cipla efavirenz tablet) and the standard addition method with DPV. The developed sensor displayed satisfactory recovery from the range of 97–106% with RSDs in the range of 9.9–10.3%. The summary of the obtained results is presented in

Table 6. Percent recovery and RSD (%) of EFV on GCE/TiO₂-NPs-nafion in Cipla EFV sample.

Serial Number	Amount Added	Amount Found	Recovery (%)	RSD n = 3
1	8.3	8.8	106	9.9
2	11.5	11.2	97	10

.

2.5. Interference and Reproducibility Study

In order to test the selectivity of the developed sensor, the influence of several typical interfering species on the determination of 47.6 µM EFV in PBS pH 7.0 was evaluated using DPV and chronoamperometry. Figure 9A shows the DPV recorded on a GCE/TiO₂-NPs-nafion electrode in 0.1 M PBS containing mixtures of EFV, ascorbic acid (AA), and uric acid (UA) at the same concentrations (47.6 µM) for the first voltammogram and 90 µM for AA in the second voltammogram.

Two prominent peaks for UA, 0.55 V, and EFV, 1.19 V, and a suppressed signal for AA were noticed for the first voltammogram in black. The second DPV voltammogram displayed three distinct peaks with peak potentials for AA, UA, and EFV found at 0.31, 0.59, and 1.22 V, accordingly. The peak potential separation between AA and UA, UA and EFV, and AA and EFV were estimated to be 0.28, 0.63, and 0.91 V, respectively. The results suggest the possibility of detecting EFV in the presence of possible interfering species. Figure 9B shows chronoamperometric current signal of EFV, AA, and UA at working potential of 1.0 V and interval of 40 s. EFV (47.6 µM) signal was determined before and after adding 1 mL of AA (90.9 µM) and 0.5 mL UA (47.6 µM) into 10 mL PBS. The result shows non-interference of EFV signal after successive injection of AA and UA, indicating selectivity of the electrode.

The reproducibility of GCE/TiO₂-NPs-nafion was examined by analyzing the DPV responses of three independent electrodes in the 47.6 µM EFV and the RSD was found to be 6.7%, demonstrating acceptable reproducibility of the designed sensor for EFV detection (Figure 9C).

3. Materials and Methods

3.1. Experimental

3.1.1. Chemicals and Solutions

All analytical grade reagents were used for this study. Standard stock solutions Na₂HPO₄ and NaH₂PO₄.2H₂O were obtained from Capital Lab Supplies (Durban, South Africa). Alumina powder ≤3 μm were supplied by Metrohm (Durban, South Africa). Nitric acid (HNO₃), N, N-dimethyl formamide (DMF), Efavirenz (EFV), methanol, titanium tetra-isopropoxide (TTIP), and nafion (5 wt%) were obtained from Sigma-Aldrich (Johannesburg, South Africa). 0.1 M phosphate buffer solutions (PBS) were prepared by mixing stock solutions of 0.1 M NaH₂PO₄ and 0.1 M NaH₂PO₄.2H₂O. 5 mM [Fe(CN)₆]^3−/4− solution was prepared using suitable mass of K₃Fe(CN)₆ and K₄Fe(CN)₆.3H₂O in 0.1 M PBS. All reagent and samples were prepared using deionized water.

3.1.2. Preparation of Plant Extract

Harvested Eucalyptus globulus leaves from Eastern Cape were used in the synthesis of TiO₂-NPs. The extracts from the plant leaves, stems, roots, and seed have gained wide application, as these leaves act as a stabilizing or reducing agent [36]. The leaves were thoroughly washed with tap water, rinsed using deionised water, dried in the oven for 2 h at 50 °C, and ground to a powder. The dried ground leaf (25 g) was transferred into a beaker containing deionised water (250 mL) and the solution was heated whilst stirring at 50 °C for 30 min. Thereafter, it was cooled and filtered.

3.1.3. Synthesis of TiO₂ Nanoparticles

TiO₂-NPs were synthesized using titanium tetra-isopropoxide (TTIP) (0.1 M, 50 mL). Leaf extract (20 mL) was added dropwise to the above solution of TTIP. The mixture was stirred for 3 h at room temperature which resulted in a white precipitate. The change of colour from white precipitate to yellowish grey confirms the formation of TiO₂-NPs. The formed TiO₂-NPs were separated by centrifugation of the mixture at 5000 rpm for 20 min, washed with deionized water (DW) to remove impurities, and then dried in the oven overnight at 100 °C. It was then calcined in a muffle furnace at 500 °C for 3 h [37].

3.1.4. Electrode pre-Treatment and Modification

Alumina nano powder slurry was gently applied to the glassy carbon electrode (GCE) prior to modification. For the purpose of removing the residual alumina nano powder and producing a mirror-like surface, GCE was rinsed with distilled water, then kept in 50% methanol for 1 min, rinsed again with distilled water, and dried in an oven at 40 °C for 2 min. GCE was modified using a drop-drying method. Nanoparticle modified glassy carbon electrode (GCE/TiO₂-NPs) was prepared by weighing TiO₂-NPs (3 mg) into a glass vial followed by addition of 300 µL of DMF, and the mixture was stirred on a hot plate for approximately 10 min for complete homogenization. To make the working electrode (GCE/TiO₂-NPs), about 2.0 μL of the resultant paste was poured onto a GCE surface and oven-dried for one min at 50 °C. The modified electrode was allowed to cool at room temperature. Thereafter, 20 µL of 0.5% nafion was dropped on the modified electrode and allowed to dry at room temperature, yielding GCE/TiO₂-NPs-nafion electrode. Scheme 2 illustrates the design of the sensor for EFV detection.

3.1.5. Preparation of EFV Stock and Working Solutions

An EFV stock solution of 1 mM was prepared using 0.0078 g and dissolved in 25 mL pure methanol, wrapped in aluminum foil, and kept in a refrigerator prior to analysis. Working solutions for the CV, EIS, DPV, and chronoamperometry measurements were prepared from the stock solution by appropriate dilution of a known amount of stock solution. It was covered with foil and stored in a refrigerator.

3.1.6. Characterization of TiO₂-NPs

The morphology of TiO₂-NPs was examined on a field emission scanning electron microscopy (FESEM), using the Tescan MIRA SEM, while the energy dispersive X-ray spectra (EDX) was obtained employing the Thermo Fisher Nova NanoSEM, with an Oxford X-max 20 mm² detector and INCA software. The optical property and crystallinity of the synthesized nanomaterial was investigated using ultraviolet–visible spectrophotometry (UV-VIS) and X-ray diffractometer (XRD) on a PIXcel detector, X-ray source of cobalt (Co) with a wavelength of 1.79026 Å. The electrochemical characterization of the synthesized TiO₂-NPs as well as TiO₂-NPs-nafion were investigated using cyclic voltammetry (CV), and electrochemical impedance spectroscopy (EIS) in a 5 mM [Fe (CN)6]^3−/4− probe solution prepared in a 0.1 M of PBS at pH 7.0.

3.1.7. Electrochemical Studies

In order to conduct electrochemical experiments, an Autolab potentiostat PGSTAT 302 was interconnected to a 663VA computrace system (Metrohm, SA) in a three-electrode configuration. The system comprises the bare and modified working glassy carbon (GCE, GCE/TiO₂-NPs, and GCE/TiO₂-NPs-nafion), reference (silver/silver chloride saturated 3 M potassium chloride), and counter platinum electrodes in supporting electrolyte. Measurements were carried out by application of CV, and EIS using modified glassy carbon electrode (GCE) after bubbling high purity nitrogen for 2–4 min to remove oxygen. The electrocatalytic properties of the bare and modified electrode towards 4.7 × 10⁻² mM EFV was investigated using CV at 25 mV/s within potential windows of −0.2 to 1.4 V. EIS experiments were conducted within a frequency range of 100 kHz and 0.1 Hz at a constant applied potential of 1.2 V for EFV detection and 0.2 V for electrochemical characterization in a [Fe (CN)₆]^3−/4− solution. Quantitative detection of EFV was conducted using DPV with control conditions of 0.005 V step potential, 0.025 V modulation amplitude, and a modulation time of 0.05 s.

3.1.8. Preparation of Real Sample

Real samples were prepared using commercially available EFV (Cipla) tablet. The tablet (1.3120 g) was pulverized into fine homogenous powder using a mortar and pestle. 10 mg of EFV ground tablet was dissolved in a 10 mL volumetric flask containing methanol and made up to mark with methanol. The mixture was sonicated for complete homogenization and thereafter filtered to remove particulates. In an electrolytic cell containing 10 mL of PBS, 1 mL of the filtrate was transferred and spiked with different concentrations of the EFV standard, then analyzed using DPV. A calibration graph was constructed to estimate the recovery of analytes in the sample.

3.2. Computational Section

3.2.1. Model Building

The 3D molecular structures for EFV and nafion were retrieved from the PubChem database, while the construction of the nanostructures to mimic glassy carbon electrode (GCE) and TiO₂-NPs anatase were constructed using Materials Studio (MS) software program [38]. The Forcite module, as implemented in MS, was used to optimize the geometry of all 3D model structures. The Forcite program was used to observe the EFV molecule’s low energy configuration as well as the COMPASS force field’s ‘ultrafine’ quality.

3.2.2. Density Functional Theory (DFT) Calculations

In order to understand the chemical reactivity of EFV from its electronic properties, the (HOMO-LUMO) band gap energies were calculated. The quantum chemical calculations were done at the DFT level of theory. A full geometry optimization was performed on the EFV molecule in vacuum, using the B3LYP density functional [39,40] and the 6-311+G basis set using the Gaussian 09 software [41]. In order to make sure that the global minima were obtained during the optimization, we have ensured the absence of imaginary frequencies. These frequencies are obtained through calculating second derivatives of energy, which constitute the hessian matrix.

3.2.3. Monte Carlo (MC) Simulations

The atomistic interactions between the substrate and the adsorbate were studied using the Adsorption Locator (AL) module in the MS program on the layers of assembly set at 298 K, and this was done to understand the molecular interactions at an atomic level. With the forcefield method, AL was utilized as a preparation and screening tool to produce a ranking of the energies for each created configuration, indicating the preferable adsorption sites. Because the adsorbate can be adsorbed at several places on the GCE/TiO₂-NPs surface, the AL module was utilized to determine the optimal adsorption site on the surface with the lowest energy. The lowest adsorption energy conformers for GCE/TiO₂-NPs-EFV and GCE/TiO₂-NPs-nafion-EFV are presented in Table 1 and were separately optimized using the Forcite module to attain a stable conformation, as illustrated in Figure 2.

4. Conclusions

This study examined the electrochemical conductivity of EFV using a bioinspired green synthesized TiO₂-NPs modified electrode with nafion as an anchor agent. We demonstrated that nafion contributes to the improved performance of the modified sensor. MC simulation results indicate that EFV interacts strongly with the GCE/TiO₂-NPs-nafion electrode, supporting the CV and EIS results. As a result of the DFT calculations, the active sites of EFV could be predicted, resulting in the proposed mechanism. Additionally, we observed that this sensor has a low detection limit (0.01 µM), with a linear dynamic concentration range from 4.5 to 18.7 µM recorded on the electrode using DPV. In addition, the sensor showed good selectivity towards EFV amidst possible interfering species (AA and UA) and sensitivity (−0.00281 µA/µM). A GCE/TiO₂-NPs-nafion electrode was successfully applied to the determination of EFV in pharmaceutical samples. The electrochemical sensor demonstrated good positive selectivity with excellent recoveries ranging from 92.0–103.9%. MC simulations revealed a strong electrostatic interaction on the GCE/TiO₂-NPs-nafion-EFV (substrate-adsorbate) system, while the DFT based calculations represented the chemical reactivity for EFV, suggesting the benzoxazine ring as the active site. In this paper, a new and convenient electroanalytical method is described for the detection of EFV in the absence of enzymes.