In this stage, we utilized an electroanalyzer made by SAMA 500, Isfahan, Iran for chronoamperometric and voltammetric measurements and EIS. This 3-electrode cell was made up of a platinum wire as the counter electrode, Ag/AgCl (saturated KCl) as the reference electrode, and a carbon paste electrode as a working electrode. Moreover, X-ray powder diffraction (XRD) data were gathered using a Philips PC-APD X-ray diffractometer made in the Netherlands and the modifier was described by energy dispersive spectroscopy (EDS) and scanning electron analysis (EM 3200 SEM and KYKY, provided from China). Finally, we used a digital pH meter (ELICO LI 120) for the pH value measurement.

Graphite powder, ethylene glycol, zinc nitrate, sodium dihydrogen phosphate, disodium hydrogen phosphate, paraffin oil, terbium chloride, sodium hydroxide, polyvinylpyrrolidone, and phosphoric acid, made by Merck Co., were the chemicals chosen for this research. It should be noted that each chemical had high analytical purity (≥99% with the exception of phosphoric acid with 85% purity). We applied double-distilled water to prepare the solution.

Electrochemical measurements were done using a 3-electrode cell in a computer-connected electrochemical system, which included a platinum wire as the auxiliary electrode, unmodified and modified CPEs as the working electrodes, and an Ag/AgCl electrode as the reference electrode.

As mentioned above, we employed double-distilled water in each stage of the experiments for washing dishes and preparation of the aqueous solution. Then, the weighted KYN and TRP were added to fresh distilled water in volumetric flasks to prepare various concentrations of KYN and TRP. Furthermore, we prepared fresh buffer and drug solutions daily and a specific concentration of the above materials was diluted using a buffer solution at a given pH to prepare the sample solution of the blood serum, and the standard addition method was utilized to transfer a certain content to the electrochemical cells for voltammetric measurements. Then, we initially cleaned the whole container with a solution of nitric acid and water (1: 1), and the glassware was washed with deionized and distilled water.

To provide the flower-like NSs of ZnO doped with Tb, 0.1 g of terbium chloride and 2 g of zinc nitrate and PVP were dissolved in 100 mL of ethylene glycol. After stirring, we transported the obtained mixture to an autoclave at 200°C for 12 h. In the next step, filtration of the precipitate and washing with ethanol and water were performed 3 times. The calcination of the precipitate was done at 500°C for 4 h, and then it was completely dried at 80°C for a 12 h period.

Upon the preparation of the blood serum or urine samples, we poured 0.9 mL of 15 w/v % solution of Zn sulfate/acetonitrile into 1 mL of the sample of human plasma or urine and kept the test tube at 40°C for 15 min. Furthermore, solution centrifugation was performed to settle the proteins to reach a fully transparent blood serum or urine sample. Afterward, the buffer was poured for a 5-fold dilution of the blood serum or urine sample, and various amounts of KYN, TRP, and standard solution were poured into the final diluted blood serum. Finally, the DPVs were shown, and the percentage of the recovery of KYN and TRP was specified using the standard addition approach.

Cyclic voltammetry (CV), chronoamperometry (CHA), and differential pulse voltammetry (DPV) were applied for the electrochemical study and quantification of KYN and TRP, respectively.

CV was performed in a phosphate buffer (0.1 M, pH 7.0) with and without the presence of KYN and TRP, starting with the equilibrium potential in the anodic direction using a potential window of −0.5–1.2 V at different scan rates. Anodic peaks were analyzed in order to establish the relationship between the maximum current intensity of the anodic peaks and the scan rate.

Under optimized conditions, CHA experiments were carried out at an applied potential of 0.40 and 0.86 V versus SCE using different concentrations of KYN and TRP, respectively.

In order to achieve the higher analytical response (anodic current), the optimal conditions for DPV measurements were as follows: PBS, pH of 7.0, modulation amplitude of 0.02505 V, modulation time of 30 ms, interval time of 200 ms, step potential of 10 mV, initial potential = 100 mV, and end potential of 950 mV. To achieve the DP voltammograms of KYN and TRP, appropriate volumes of the stock solutions of drugs were added to the cell containing supporting electrolytes to a total volume of 25 mL.

The XRD in Figure 1 shows the crystalline nature of the nanoflowers in that a 2θ peak was observed at 32.56° (100), 34.83° (002), 36.415° (101), 47.14° (102), 56.22° (110), 63.59° (103), 67.75° (200), 68.63° (112), 69.73° (201), 71.70° (004), and 77.22° (202); however, we did not observe any characteristic peak for other impurities such as Zn (OH)₂ and metallic Zn, reflecting the product purity. There was a complete correlation between the peaks and polycrystalline hexagonal wurtzite-structured ZnO; that is, three pronounced peaks (100), (002), and (101) at 2θ = 32.56°, 34.83°, and 36.41° that could be compared to the typical XRD pattern of the standard ZnO (JCPDS 89–7,102) (SudapalliShimpi, 2023). It should be noted that ZnO NF had higher intensity and narrower peaks, resulting in greater crystallinity. In Figure 1, a clear shift of the diffraction peaks towards a higher angle than pure ZnO is shown, demonstrating the significantly greater ionic radius of Tb (237 p.m.) in comparison to Zn (139 nm) that may be due to the minor doping of Tb ions into the ZnO lattice, leading to an improvement in the lattice parameter in the Tb-doped ZnO crystallites. Researchers have estimated these minor changes in the Tb replacing Zn ions in the lattice without any variations in the crystal lattice (Taherizadeh et al., 2023).

In this step, FESEM was used for the characterization of the morphology of the as-synthesized Tb-doped ZnO nanoflowers. Figure 2A shows the flower-like 3D morphology of the ZnO nanoflakes formed in the petals in the FESEM images at a lower magnification. Consequently, high-density growth was observed because of the nanoflakes’ self-assembly. Moreover, Figures 2B, C shows FESEM images at high magnification, containing uniform nanoflowers. As shown in Figures 2A, C single nano-flower dimension was ∼1–3 µm with multiple nanoflakes with radial growth from the center in symmetry so that most of the nanoflowers were associated with each other.

EDAX analysis, shown in Figure 3, proved the presence of Tb, O, and Zn in the ZnO nanoflowers. As shown in Figure 3, Tb, O, and Zn had matching elemental mapping, with a smooth distribution of these three elements in the sample according to the results obtained from the elemental mapping.

According to our research design, EIS was applied to show the variations in the electrochemical behaviors of the modified electrode surface. Studies in the field have shown the ability of polymers, semiconductor materials, and nanomaterials coated on the surface of the electrode to change the transport resistance and double–layer capacitance of the corresponding electrode. Hence, this method could be used to achieve more data on the variations in the surface impedance of the electrodes in the course of the modification process of the electrode. The features of the electrochemical impedance of the modified FL-NS ZnO/CPE, FL-NS Tb³⁺/ZnO/CPE, and unmodified CPE in the solution with 1 mM of redox pair KCl 0.1 M and [Fe(CN)₆]³⁻/[Fe(CN)₆]⁴⁻ ranged between 0.1 Hz and 100 kHz, shown as the Nyquist curve (Z_re versus Z_im) in Figure 4. The equivalent circuit was designed and implemented to understand and evaluate the individual components of the EIS system. The resistance of the solution (R_s), double layer capacitance at the surface of the electrode (C_dI), charge transfer resistance (R_ct), and Warburg resistance (W) were simplified in the Randles equivalent circuits, as shown in Figure 4. Warburg resistance is the result of a diffusion process occurring at the electrode–electrolyte interface. As can be observed in the Nyquist diagram, the high resistance of the charge transfer for the unmodified CPE (Figure 4 (curve a)) resulted from the reduced transfer rate of the charge and mass on the surface of the electrode (792 Ω). Moreover, the electron transfer kinetics in FL-NS ZnO [Figure 4 (curve b)] were enhanced in the presence of the modifier in the electrodes and decreased the transport resistance of the charge (408 Ω). Additionally, the charge transfer was considerably enhanced around the modified electrode with the FL-NS ZnO mixture and Tb³⁺ (Figure 4 (curve c)) which could related to the higher conductivity of the modifier (218 Ω).

According to the research design, the Randles-Sevcik equation for a quasi-reversible electrochemical process (Eq. 1) and the CV were employed to calculate the effective surface area (in cm²) of electrode A. Hence, we used the CV of the soluble 0.3 mM of [Fe(CN)₆]^−3/−4 with the diffusion coefficient 7.6 × 10⁻⁶ (Bard and Faulkner, 2001): I p = ± 2.63 × 10 5   n 3 / 2   A   D 1 / 2   C   v 1 / 2 (1) where n stands for the number of electrons transported in the reduction and oxidation processes of [Fe(CN)₆]^−3/−4 (=1), D refers to the diffusion coefficient in cm²/s, and C represents the [Fe(CN)₆]^−3/−4 concentration (mol/cm³). The scanning rate and current are represented by V/s and A. From the slope of the plot of I_p vs. ν^1/2, the surface area of the unmodified CPE was found to be 0.112 cm² and for the FL-NS ZnO/CPE and FL-NS Tb³⁺/ZnO/CPE, the calculated surface areas were 0.228 cm² and 0.301 cm² (Figure 5; Table 1). The surface area of the FL-NS Tb³⁺/ZnO/CPE was more significant, which can be attributed to the presence of Tb³⁺ and FL-NP ZnO that led to high electrical conductivity and the specific surface area of the modified electrode.

For this step, CV was applied to assess the electrochemical behaviors of KYN and TRP over the surface of the unmodified CPE, FL-NS Tb³⁺/ZnO/CPE, and FL-NS ZnO/CPE in 0.1 M PBS with a pH of 7.0 (Figure 6). Moreover, a solution of KYN (250.0 µM) and TRP (150.0 µM) in the PBS was added to the electrochemical cells, separately. See Figure 6 for the oxidation peak current of KYN and TRP with the peak oxidation potential of the solution over the electrodes’ surfaces. As seen in Figure 6, these voltammograms exhibited an oxidation peak for KYN and TRP at 0.54 V and 1.001 V for the unmodified CPE, which shifted to 0.350 V and 0.809 V for FL-NS Tb³⁺/ZnO/CPE, respectively. Different currents were observed for the various electrodes. The currents for the FL-NP Tb³⁺/ZnO/CPE and FL-NP ZnO/CPE were 6.7 and 4.86 times greater compared to the unmodified CPE. The results indicate the simultaneous effect of Tb³⁺ and FL-NP ZnO as modifiers in improving the KYN and TRP peak currents and thus increasing the sensitivity of the proposed method.

To achieve the best response in the electrochemical measurement of KYN and TRP, the amount of FL-NS Tb³⁺/ZnO used as a modifier in preparing carbon paste electrodes was optimized. For this purpose, electrodes with values of 0.002–0.006 g of FL-NS Tb³⁺/ZnO were prepared and placed in a 0.1 M phosphate buffer solution at a pH of 7.0 to measure 40.0 μM of KYN solution and 35.0 μM of TRP solution by DPV. The results indicated that 0.004 g of FL-NS Tb³⁺/ZnO has the highest current in the measurement of KYN and TRP.

It should be noted that the potential scan rate has been introduced as a major variable employed in electrochemistry for investigating the oxidation of the samples, the reaction mechanism of reduction, or obtaining the kinetic variable. In this step, we adjusted the potential scan rate of the KYN and TRP oxidation over the FL-NS Tb³⁺/ZnO/CPE surface at 10–900 mV/s but we used CV to determine its effect in 0.1 M PBS (pH = 7.0). See Figures 7A,C for more information. With the enhancement in the scanning rate, it was found that the current increased, and Figures 7B, D shows the current curve to ν^1/2 (square of the scan rate), depicting a linear relationship based on Eqs 2, 3. As seen, the samples’ diffusion limits the oxidation reaction of KYN and TRP, respectively. I p = 1.5434   ν 1 / 2 − 0.0302   R 2 = 0.9992 (2) I p = 0.741 ν 1 / 2 − 2.0752 R 2 = 0.9993 (3)

Experts in the field use chronoamperometry to measure the diffusion coefficient of electroactive samples. According to this approach, a potential is applied to the static electrode with a specific floating area in a solution with an electro-active compound at a given concentration. Then, the working electrode potential is applied to the diffusion platform of the electroactive samples, and a current–time dependence occurs. There was a relationship between the current–time curve and variations in the concentration gradient near the electrode surface. As time increased, the diffusion layer widened and inhibited the electroactive compound from approaching the surface of the electrode by decreasing the concentration slope of the profile. Hence, the current intensity in a flat electrode decreased with time (Cottrell Equation). Based on the Cottrell Equation, the emission current from the electrochemical reaction of an electroactive sample would be followed by Eq. 4 (Bard and Faulkner, 2001): I = n F A D 1 / 2 C b π − 1 / 2 t − 1 / 2 (4) where D represents the diffusion coefficient (cm²/s), C refers to the concentration of the electroactive compound in terms of mol/cm³, and n refers to the stoichiometric number of electrons involved in the reaction (Supplementary Figure S1). In the case of plotting the flow changes with the reverse square time for a chronoamperogram, the slope equals nFACD^1/2/π^1/2, in which knowledge of C, F, n, and A can help to calculate the diffusion coefficient. We also examined the oxidations of KYN and TRP on the FL-NS Tb³⁺/ZnO/CPE surface through chronoamperometry (Figures 8A, D) and the current diagram in terms of t^−1/2 was found to be linear for various concentrations (Figures 8B, E). Figures 8C, F depict the charts’ slope for diverse concentrations, with diffusion coefficients of 1.02 × 10⁻⁶ cm²/s and 1.15 × 10⁻⁶ cm²/s for KYN and TRP, respectively.

As mentioned above, we employed DPV for the evaluation and determination of small concentrations of KYN and TRP. Figure 9 represents the DP voltammograms of the FL-NS Tb³⁺/ZnO/CPE surface in the optimum conditions that ranged from 0.001 to 700.0 μM. Eqs 5, 6 give the linear relationship of I_pa and the concentrations: I   μ A = 0.0575 C KYN   μ M + 0.0595   R 2 = 0.9999 (5) I   μ A = 0.0488   C TRP   μ M + 0.0417   R 2 = 0.9998 (6)

The limit of detection (LOD) of this new approach, obtained from Eq. LOD = 3S_b/M [so that M represents the slope of the calibration curve, S_b stands for the blank standard deviation (SD)], was found to be 0.34 nM and 0.22 nM for KYN and TRP, respectively. Hence, this performance and greater sensitivity resulted from the synergistic effect of FL-NS Tb³⁺ and ZnO, increasing the transport rate of electrons and detecting the limited levels of KYN and TRP.

In this step, we determined the impact of numerous interference samples (α) in the real samples (Table 2) on the potential and peak current of the DP voltammograms for investigating the FL-NS Tb³⁺/ZnO/CPE selectivity to KYN and TRP. Then, the DPV of 50.0 μM KYN and TRP were recorded five times in the presence of α with 100 times the concentration in comparison with KYN and TRP. The next step addressed the comparison of the average of the peaks E_p and I_P with the KYN and TRP peaks E_p and I_P in the absence of α. In the case of ˂5% E_p and I_P in the presence of α compared to its absence, the interference samples and the above stages continued to reach a 5% difference. The interference samples were chosen according to the common molecules and ions in the drug samples and biological fluids. The α samples were not interfered with when measuring KYN and TRP, revealing the very good selectivity of FL-NS Tb³⁺/ZnO/CPE to both drugs and their probable measurement in the real samples.

It is widely accepted that simultaneous measurements of the medicines using a modern technique by a new electrode could exhibit acceptable functions in real samples. Hence, we assessed the function of the FL-NS Tb³⁺/ZnO/CPE electrode in simultaneous measurement of KYN and TRP (Figure 10A). With regard to Figures 10B, C, concentration of KYN and TRP were changed from 20.0-100.0 μM for KYN and 15.0-60.0 for TRP. As shown by Figure 10A, no changes were observed in the potential of KYN and TRP and peak current in the presence of each of them. Thus, an FL-NS Tb³⁺/ZnO/CPE electrode could be used for measuring samples of both KYN and TRP without interfering with each other.

To examine the reproducibility of the preparation process of the electrode, we conducted measurements in 50.0 μM solutions of KYN and TRP using five modified electrodes with the same DPV and the value of the relative standard deviation (RSD) of the anodic peak current (mean of 3 measurements/electrode) equaled 2.84 and 2.52 (Supplementary Figure S1). The findings indicated the acceptable reproducibility of the modified electrode in the voltammetric measurements and electrode production. Then, we put the electrode into a buffer medium at room temperature for 4 weeks and analyzed 50.0 μM solutions of KYN and TRP to evaluate the modified electrode stability. Ultimately, the peak currents (3.12% and 2.26%) declined, demonstrating the acceptable stability of the prepared electrode (Supplementary Figure S2).

The comparison of analytical efficacy between the as-fabricated electrode and other electrochemical and non-electrochemical methods was performed for KYN and TRP (Table 3). As shown in Table 3, the detection limit and linear range of the as-fabricated sensor were better than electrochemical and non-electrochemical methods (Zhao et al., 2010; Wu et al., 2023; Kambale and Lokhande, 2023; Zhao et al., 2023; Taherizadeh et al., 2023; Guo et al., 2023). When comparing chromatography methods with electrochemical methods, these methods are expensive, sophisticated, and multi-process techniques, with the need for sample preparation, pre-filtration, extraction, and temperature monitoring. The as-fabricated sensor is potentially able to determine the trace amounts of KYN and TRP in various media. In addition, it is noteworthy that the voltammetry method determines KYN and does not utilize simultaneous determination with TRP. As seen in Table 3, the as-fabricated electrode for electrochemically sensing KYN and TRP generally showed admirable properties for measurement speed, sensitivity, detection limit, and linear range when compared to other methods reported in the literature.

The new electrode was employed for determining KYN and TRP in blood serum and urine samples to evaluate the capability of FL-NS Tb³⁺/ZnO/CPE for oxidation of the analytes. The standard addition procedure was applied for the measurements because of the complexity of the real samples’ matrices. Based on the findings, the recovery percentage for the blood serum samples ranged from 97.7%–101.2% and for the urine samples from 98.7%–103.6% (Table 4). The acceptable percentage of recovery and RSD values of ˂5% demonstrate the high functionality of the FL-NS Tb³⁺/ZnO/CPE to measure KYN and TRP in samples with real matrices.