CuSO₄·5H₂O, L-serine, olivetol, sodium hydroxide, paraffin oil, and graphite were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). The double distilled water was supplied for the preparation of aqueous solutions in experiments. All reagents were of analytical grade and used as received.

As Afzali reported, the CuO nanoparticles were synthesized with some modifications [26]. 4.0 g of CuSO₄·5H₂O was dissolved in 100 mL of water, forming a transparent blue solution. Then, 100 mL of NaOH solution (0.3 mol/L) was slowly added into the copper sulfate solution under magnetic stirring at 80 °C. The pH value of this mixture would be adjusted to around 12, and dark brown precipitates would appear. After the incubation at 80 °C for 2 h, the mixture was washed with water by centrifugation. Finally, the product was dried under 105 °C for 8 h in the oven and ground into powders in a mortar for further use. The CuO nanoparticles were characterized by scanning electron microscope (SEM, Apreo 2 Thermo Scientific, Waltham, MA, USA) and X-ray Diffraction (XRD, Rigaku SmartLab 9, Tokyo, Japan) using Cu-Kα radiation at 40 kV and 40 mA.

The CPE was prepared according to Ghalkhani’s report with some modifications [27]. 70.0% (w/w) of graphite powder and 30.0% (w/w) of paraffin oil were hand mixed in an agate mortar until a homogenous paste was obtained. Some amount of this paste was transferred and firmly packed into a cavity at the end of a Teflon tube (3.0 mm inner diameter and 2.0 mm long). A copper wire was fixed in the tube, which connected to the paste, to provide an external electrical contact. Before the usage of CPE, a small portion of the paste was squeezed out, and the surface was polished with weighing paper.

Copper oxide modified carbon paste electrode (CuO/CPE) was prepared as follow. 65.0% (w/w) of graphite powder, 10.0% (w/w) of CuO nanoparticles, and 25.0% (w/w) of paraffin oil were hands mixed adequately in an agate mortar and pestle for 30 min until a thick homogenous paste was obtained. The paste was packed into a cavity at the end of a Teflon tube and treated as the same procedure.

The p-L-serine/CuO/CPE was fabricated by an electrochemical polymerization process using cyclic voltammetry. The CuO/CPE was immersed in 0.2 mol L⁻¹ phosphate buffer solution (pH 6.5) containing 1.0 mmol L⁻¹ of L-serine and scanned between −0.6 and 1.8 V for 20 cycles. Finally, the electrode surface was washed with water to remove the physically adsorbed L-serine.

The electrochemical measurements were carried out by using a CHI 660E electrochemical workstation (Shanghai Chenhua Co., Ltd., Shanghai, China) in the three electrodes system. The CPE and modified CPE were used as the working electrode (3.0 mm diameter). A platinum wire was applied as the auxiliary electrode, and an Ag/AgCl was used as the reference electrode.

The cyclic voltammetry (CV), differential pulse voltammetry (DPV), and electrochemical impedance spectroscopy (EIS) scan modes were applied, and the corresponding voltammograms were recorded. The electrochemical determination of olivetol was performed in an electrolytic cell containing a supporting electrolyte solution (0.01 mol L⁻¹ of phosphate buffer saline containing 0.005 mol L⁻¹ of potassium chloride, pH 8.0). CV and DPV were recorded from 0 V to 1.0 V with a scan rate of 0.05 V s⁻¹. The amplitude was set at 0.05 V for DPV. EIS was scanned in 5.0 mmol L⁻¹ of K₃[Fe(CN)₆]/K₄[Fe(CN)₆] and 0.1 mol L⁻¹ of potassium chloride. The potential was set at 0.283 V. The amplitude was set at 0.005 V with a frequency range of 0.01 to 10⁵ Hz. All experiments were carried out in three duplicates at 25 ± 2 °C. After each measurement, the modified electrode was transferred to the water by stirring for 20 s to obtain a renewed electrode surface.

The SEM image and XRD pattern of CuO nanoparticles are presented in Figure 1. The SEM photograph (Figure 1a) showed the morphology of CuO nanoparticles was agglomerated and irregular in shape, and the size ranged from 50 nm to 100 nm. The diffraction peaks obtained in the XRD pattern of CuO nanoparticles (Figure 1b) at the 2θ values of 33.2°, 35.4°, 38.6°, 48.6°, 52.4°, 57.9°, 61.5°, 66.3°, 67.9°, 72.2°, and 75.2° could be attributed to (110), (111/002), (111/200), (202), (020), (202), (113), (311/002), (220), (311), and (222) planes indicating the monoclinic lattice structure of CuO nanoparticles through the comparing with standard card (JCPDS file: 05-0661) [28,29].

To optimize the proportion of CuO nanoparticles in the preparation of CuO/CPE, CV was used to investigate the influence on the electrochemical responses of olivetol solution [30]. The CV curves and trends are disclosed in Figure 2a,b. The peak current increased with the proportion of CuO nanoparticles from 5% to 10% and then decreased at a greater proportion (from 15% to 20%). This might be because of the augment in the number of active sites in CPE introduced by the addition of CuO nanoparticles. However, the excess addition of CuO nanoparticles in the paste would lead to a decrease in conductivity [31]. Therefore, 10.0% (w/w) of CuO nanoparticles were adopted through the preparation of CuO/CPE and p-L-serine/CuO/CPE.

The electropolymerization of L-serine was carried on the surface of CuO/CPE using cyclic voltammetry in the existence of 0.1 mol L⁻¹ NaOH and cyclically scanned between −0.6 to 1.8 V for 20 cycles. As elucidated in Figure 2c, the peak current in the CV plot gradually changed with increasing cycles, which endorsed the growth of polymeric films on the electrode. The electropolymerization cycles of obtained p-L-serine/CuO/CPE were investigated by changing the cycle numbers from 4 to 24 cycles (Figure 2d) and submitted to the electrochemical detection of olivetol. It could be observed that the peak current was enhanced up to 20 cycles and decreased slightly after that (the CV curves were provided in Figure S1). The p-L-serine membrane was formed and deposited on the surface of CuO/CPE, and the thickness of the layer might affect the electrocatalytic property of the forming p-L-serine/CuO/CPE, which was closely related to the electropolymerization cycles [24]. Therefore, 20 cycles of electropolymerization were selected as the optimum condition for the fabrication of the p-L-serine/CuO/CPE. After electropolymerization, the modified electrode was carefully rinsed with water and stored for further use.

The electrochemical responses of the bare CPE, CuO/CPE, and p-L-serine/CuO/CPE were studied in 5 mmol L⁻¹ of potassium ferricyanide containing 0.1 mol L⁻¹ of KCl as the supporting electrolyte. A comparison of the corresponding peak currents in CV is given in Figure 3a. The peak currents of potassium ferricyanide at both CuO/CPE and p-L-serine/CuO/CPE increased as compared to those at the bare CPE. Though the potential difference of p-L-serine/CuO/CPE was bigger than that of CuO/CPE, the response was also higher. The result showing the modification of CPE enhanced the electrochemical activity of the electrode. Then, the responses of modified electrodes on PBS and olivetol solutions were further compared. As shown in Figure 3b, both bare CPE and modified CPE had no response in CV curves (black, brown, and green plots). Meanwhile, a couple of redox peaks could be found in CuO/CPE and p-L-serine/CuO/CPE when analyzed in olivetol solution, which was located at around 0.65 V (blue and red plots). According to these characterizations, p-L-serine/CuO/CPE could be used in the electrochemical analysis of olivetol by enhanced signals.

The EIS plots of bare CPE, CuO/CPE, and p-L-serine/CuO/CPE were investigated to find the change of charge transfer resistance (Rct) of electrodes during modifications. The diameter of the semicircle in the Nyquist plots reflected the value of Rct, and the diameters of modified electrodes were smaller than that of bare CPE, indicating a decrease in resistance [23]. Through the fitting of EIS data with the equivalent circuit, the values of Rct were stimulated as 130 KΩ, 66 KΩ, and 1896 Ω for bare CPE, CuO/CPE, and p-L-serine/CuO/CPE, respectively. The change of Rct values was in accordance with that of the diameters in the Nyquist plots. Moreover, the effective area of p-L-serine/CuO/CPE was calculated by the Randles-Sevcik equation (I_p = 2.69 × 10⁵ A D^1/2 n^3/2 υ^1/2 C), where A is the effective area (cm²), D is the diffusion coefficient of potassium ferricyanide (7.6 × 10⁻⁶ cm² s⁻¹), n is the transfer number of electrons (1), υ is the scan rate (0.05 V s⁻¹), and C is the concentration of potassium ferricyanide (5 mmol L⁻¹) [32]. In this study, the effective area of p-L-serine/CuO/CPE was 0.039 cm².

To investigate the experimental conditions in terms of the pH of supporting electrolytes, various pH of supporting electrolytes (7.0, 7.5, 8.0, 8.5, and 9.0) containing olivetol were electrochemically detected in CV using p-L-serine/CuO/CPE (Figure 4a). The results showed that the pH affected the peak currents of olivetol. The peak current of olivetol increased with increasing pH from 7.0 to 8.0, and then higher pH decreased the peak current sharply. Thus, the pH of the supporting electrolyte was adjusted to pH 8.0 for the following experiments.

The influence of scan rates on the peak current of olivetol (pH 8.0) was studied using p-L-serine/CuO/CPE in the CV method. Figure 4 demonstrated the resulting CV curves at scan rates varying from 0.01 V s⁻¹ to 0.3 V s⁻¹. and the corresponding linear plots of peak currents and square roots of scan rates. As shown in Figure 4b, the oxidation peak potentials slightly shifted to positive values with the increasing scan rate, and the peak currents increased as well. With the increase in scan rate, the peak current increased linearly (Figure 4c), and the slope was 156.2 µA s V⁻¹ (r² = 0.979). Moreover, the plot in Figure 4d demonstrated the linear relationship between the peak current and the square root of the scan rate, which would be expressed as Ip = 104.1 v^1/2 − 7.45 (r² = 0.981). These trends suggested that the electrode reaction on the p-L-serine/CuO/CPE is controlled by diffusion as a rate-determining step [33].

As reported in S.T. Nandibewoor, the number of electrons involved in the reaction was 2, which was equal to the number of protons. Based on these earlier papers, a possible mechanism is shown in Figure 5.

As a sensitive electrochemical method, the DPV measurements were studied for the determination of olivetol under optimum conditions. Figure 6 provided the DPV plots and calibration curves under increasing concentrations ranging from 20 to 100 μmol L⁻¹. It could be found that the oxidation peak appeared at 0.6 V as well, and the peak current increased with increasing concentrations of olivetol. After fitting and calculation, the DPV response of olivetol expressed good linearity, which could be explained as Ip = 0.514C − 0.519 with r² = 0.999. The Limit of detection (LOD) and quantification (LOQ) was calculated as 3 times S/N and 10 times S/N, where N is the slope of the standard curve and S is the standard deviation of blank currents [34]. The LOD and LOQ in this study were calculated as 1.04 and 3.47 μmol L⁻¹. The existing reported electrochemical detection of olivetol was reported by S.T. Nandibewoor in 2017 [35]. Some of the parameters in the method were compared in Table 1. Although the reported gold electrode had good sensitivity for olivetol (0.1–1.5 μmol L⁻¹), the proposed p-L-serine/CuO/CPE demonstrated good linearity and a wide quantitative analysis range from 20 to 100 μmol L⁻¹. The sensitivity was a typical advantage of the gold electrode. And many gold sensors and modified gold electrodes were applied to the trace level of target samples. However, the advantages of CPE were cheap and easy, which could enlarge the application ranges. Considering the materials and preparation of the modified CPE, this sensor could be applied in the rapid analysis in the field and cost control.

For the reproducibility test, five independently modified electrodes were used to determine the same sample and the resulting relative standard deviation (RSD) was 2.57%. The repeatability of the modified electrode was obtained by testing five times a day, and the obtained RSD was 2.42%. For the stability test, the modified electrode was stored in the laboratory at room temperature, and the signal only decreased by 3.00% of its initial value over a week. The interference study of the modified electrode was studied in the presence of interferents. The concentrations of inorganic ions and organic compounds were 10-fold higher than that of the target compound, and the concentration of orcinol was the same as that of the target compound. It was found that the addition of Mg²⁺, Ca²⁺, Al³⁺, Na⁺, K⁺, phenol and orcinol influenced the signal in bearable degrees in the determination of olivetol (Figure 7). The relative errors during the test were kept at less than 5.0%, indicating that these compounds were not potential interferences for this analysis [36].