Facile Non-Enzymatic Electrochemical Sensing for Glucose Based on Cu2O–BSA Nanoparticles Modified GCE

Transition-metal nanomaterials are very important to non-enzymatic glucose sensing because of their excellent electrocatalytic ability, good selectivity, the fact that they are not easily interfered with by chloride ion (Cl−), and low cost. However, the linear detection range needs to be expanded. In this paper, Cu2O–bovine serum albumin (BSA) core-shell nanoparticles (NPs) were synthesized for the first time in air at room temperature by a facile and green route. The structure and morphology of Cu2O–BSA NPs were characterized. The as-prepared Cu2O–BSA NPs were used to modify the glassy carbon electrode (GCE) in a Nafion matrix. By using cyclic voltammetry (CV), the influence from scanning speed, concentration of NaOH, and load of Cu2O–BSA NPs for the modified electrodes was probed. Cu2O–BSA NPs showed direct electrocatalytic activity for the oxidation of glucose in 50 mM NaOH solution at 0.6 V. The chronoamperometry result showed this constructing sensor in the detection of glucose with a lowest detection limit of 0.4 μM, a linear detection range up to 10 mM, a high sensitivity of 1144.81 μAmM−1cm−2 and reliable anti-interference property to Cl−, uric acid (UA), ascorbic acid (AA), and acetaminophen (AP). Cu2O–BSA NPs are promising nanostructures for the fabrication of non-enzymatic glucose electrochemical sensing devices.


Introduction
Diabetes mellitus is a group of lifelong metabolic diseases characterized by hyperglycemia caused by multiple causes. According to the International Diabetes Federation (IDF) [1], in 2017, there is one diabetic patient for every 11 people aged 20-79, and 425 million people suffering from diabetes worldwide. It is estimated that the number of diabetic patients in this age group will rise to 629 million by 2045. At present, there is no cure for diabetes mellitus. People can only use blood sugar detection, diet therapy, exercise therapy, and drug treatment to control blood sugar level. To control blood sugar concentration, people must monitor blood sugar frequently, so blood sugar detection plays a vital role in the treatment of diabetes mellitus. This encourages researchers to explore the field of glucose biosensors.
It is known that the catalytic performance of the nanomaterials depends on their surface areas, surface atomic structures, and morphologies. By controlling the shapes, sizes, and coating agents of the nanomaterials, catalysts with large surface areas and abundant catalytic active sites can be obtained, thereby improving catalytic performance. As a model globular protein, bovine serum albumin (BSA) is widely used in biochemical studies [51]. BSA is composed of 583 amino acid residues, several sulfur, oxygen, and nitrogen atoms [51]. In recent years, BSA were generally used to stabilize the inorganic NPs [51][52][53][54][55][56]. BSA-stabilized Au and Ag NPs have been widely explored for biomedical [54][55][56][57][58][59] and sensing applications [60]. The fabrication of metal-BSA NPs is a green reaction process. The products demonstrated excellent conductivity, biocompatibility, and multifunctionality, and were environmentally friendly and highly stable in both aqueous and solid form. Hu et al. [61] prepared Pt-BSA nanocomposite as an enzyme-labeled glucose-sensing interface.
In this paper, Cu 2 O-BSA NPs with size of 180-220 nm was prepared via a green route and characterized by TEM and XRD. Taking advantage of as-prepared Cu 2 O-BSA NPs as catalyst and microscopic network structure of Nafion to stabilize the modified electrode, a novel non-enzymatic glucose sensor was constructed. Results showed that the Cu 2 O-BSA NPs exhibit a remarkable electrocatalytic activity and electrochemical sensing matrix for the oxidation of glucose, free from loading any enzymes. The modified electrode indicated wide linear detection range of up to 10 mM, high sensitivity, low detection limit, fast response time, and strong anti-interference property to Cl − , AA, UA, and AP. Our results indicated that the Cu 2 O-BSA NPs is an excellent material for non-enzyme electrochemical biosensors.

Reagents and Characterization
Copper nitrate trihydrate (Cu(NO 3 ) 2 ·3H 2 O), sodium hydroxide (NaOH), sodium chloride (NaCl), potassium ferricyanide, potassium ferrocyanide, glucose, ethanol, AA, UA, and AP are analytical pure and were bought from Sinopharm Chemical Reagent Co., Ltd. BSA was purchased from Aladdin with purity of 96%. Nafion 117 solution is analytical purity, produced by Sigma-Aldrich Company, USA. All of reagents were used without further purification.
The crystalline structure of Cu 2 O-BSA NPs was detected by an X-ray diffractometer (Bruker D8 ADVANCE) with Cu-Kα radiation (λ = 1.5406 Å). A type of H-7650 TEM (Japan) was used to observe the size and morphology of the Cu 2 O-BSA NPs. A CHI 690 electrochemical workstation (Chenhua, Shanghai, China) was assumed to conduct the electrochemical experiments including CV and amperometry. A conventional three-electrode electrochemical system, which consists of a working electrode, a Pt foil auxiliary electrode, and Ag/AgCl reference electrode, was used for all electrochemical experiments.

Synthesis of the Cu 2 O-BSA NPs
70 mg BSA was dissolved in 26.4 mL deionized water with magnetic stirring for 10 min. 1.4 mL 0.2 M Cu(NO 3 ) 2 solution was added to BSA solution and stirred for 1 min, then 2.2 mL 1 M NaOH solution was added to the above solution and stirred for 5 min. 5 mL 0.1 M AA was added dropwise into the solution for reduction of Cu 2+ . After stirring for 10 min, the solution became orange colloid and the color was stable, which indicated Cu 2 O-BSA NPs were generated. The colloid was aged for half an hour then centrifuged three times at 8000 rpm for 10 min. The final Cu 2 O-BSA NPs were freeze dried in a vacuum freeze dryer and sealed in a small brown glass bottle, then preserved in a refrigerator at 4 • C for later use.

Preparation Modified Electrode of Nafion/Cu 2 O-BSA/GCE
Typically, 5 mg Cu 2 O-BSA NPs was added into 1 mL ethanol and sonicated for 15 min. 10 µL dispersion liquid of Cu 2 O-BSA NPs was dropped on the surface of bare treated GCE (diameter, 3 mm). After drying in air at room temperature, Cu 2 O-BSA NPs were evenly coated on the surface of GCE. Then 5 µL 0.1 wt. % Nafion/ethanol solution was dropped on the coating of Cu 2 O-BSA NPs. After natural drying, the modified electrode was sealed and stored for later use. The as-prepared electrode was named Nafion/Cu 2 O-BSA/GCE. For comparison, a Nafion/GCE without modified Cu 2 O-BSA NPs was also prepared by the same process.  Figure 1A). So, the structure of the Cu 2 O-BSA NP is a core-shell structure. The NPs have good monodispersity. Figure

Electrochemical Characterization of the Sensor
It is reported that the Cu 2 O/MWCNTs modified electrode [63] works well in an alkaline environment for electrochemical oxidation glucose. Here 50 mM NaOH solution was used to probe electrocatalytic activities of the Nafion/Cu 2 O-BSA/GCE electrode towards the oxidation of glucose. The result is shown in Figure 2A, in which curve a and b represent 0 and 2 mM glucose in the solution, respectively. The scan rate is 5 mV/s. In the absence of glucose (curve a), the valence changes of copper ion is represented by the five peaks that can be determined according to [64], with Peak 1 near to −0.375 V representing the oxidation of Cu(0) to Cu(I). Peak 1 is very low because the material modified GCE is Cu 2 O-BSA NPs, the content of copper (0) is zero. Peak 2 at −0.110 V refers to the oxidation process of Cu(I) to Cu(II). This peak is also low, which indicates that the transformation of Cu(I) to Cu(II) is slow. The formation of peak 3 at 0.25 V is related to the adsorption of OH − on the electrode and the formation of soluble substances from Cu 2 O-based solids. However, the potential of peak 3 is much higher than that of reference [65] and reference [66]. We speculate the formation of Peak 3 may be related to the negative charge of BSA, which leads to the voltage of the electrode-adsorbed OH − shifting a lot towards the right. According to reference [54], in high-concentration sodium hydroxide solution, Cu(III) begins to form at 0.6 V. Peak 4 at 0.6 V attributes to the reduction process from Cu(III) to Cu(II). Peak 5 indicates the reduction from Cu(II) to Cu(I). In the whole cycle, no Cu(I) is converted to Cu(0), which means that no Cu(0) was formed during the cycle. When 2 mM glucose was added into solution, Peak 1 became obvious, which indicates Cu (0) was formed during the cycling process. Peak 2 is very low and almost the same as the absence of 2 mM glucose, which indicates that a small quantity of Cu(0) was generated before this stage, and most Cu(0) has been oxidized to Cu(I) during the period of peak 1 to peak 2. The disappearance of peak 3 may be the formation of complex (Cu(I)-glucose) on the surface of the electrode by the adsorption of glucose, which hinders the formation of soluble substances between Cu 2 O and OH − . A new oxidation peak at about 0.52 V is due to the desorption of glucose from the Cu(I)-glucose complex and the simultaneous oxidation of Cu(I) to Cu(II). The reason of the disappearance of peak 4 is the consumption of Cu(III) by electric oxidation of glucose. Peaks 5 and 6 are the reduction processes of Cu(II) to Cu(I) and Cu(I) to Cu(0), respectively. The electrocatalytic effect of the Nafion/Cu 2 O-BSA/GCE was further compared with the Nafion/GCE for oxidation of glucose, as shown in Figure 2B. For the Nafion/Cu 2 O-BSA modified GCE, curve a and b show the CV results of absence and presence 5 mL glucose. The peak current density ratio of curve b and curve a is 1.54, which indicates the current density increase when glucose was catalyzed on the surface of Nafion/Cu 2 O-BSA/GCE. For the only Nafion-modified GCE, curve c and d in inset indicate the results of CV without and with 5 mL glucose. The peak current density ratios of curve d and curve c is 1.2, which shows the current density of only Nafion-modified GCE also increases when glucose was catalyzed on the surface of Nafion/ GCE. However, the extent of increase of Nafion/GCE is far less than that of the Nafion/Cu 2 O-BSA/ GCE because the current density of the Nafion/Cu 2 O-BSA/GCE is 60 times of the Nafion/GCE. The above results indicate Cu 2 O-BSA NPs greatly increase the catalytic ability of the GCE for glucose, which is ascribed to high catalytic active sites for the glucose oxidation offered by the Cu 2 O NPs and good conductivity of BSA.
It is known that the scanning rate, concentration of NaOH in electrolyte, and load of Cu 2 O NPs on the GCE have effects on the current density of CVs. To obtain an optimal result, the above factors were probed, respectively. Figure 3A displayed the CVs of the Nafion/Cu 2 O-BSA/GCE for glucose oxidation recorded in 50 mM NaOH solution with the scanning speed in the range of 5-300 mV/s. According to Figure 3A, the peak current densities of oxidation (red line) and reduction (black line) and their change with the scanning speeds are shown in Figure 3B. The oxidative and reduction peak current densities are proportional to the scan rates in the range of 5-300 mV/s, with correlation coefficients (R 2 ) of 0.9878 and 0.9750 (see Figure 3B), indicating that the electrocatalytic oxidation of glucose by the Nafion/Cu 2 O-BSA/GCE under alkaline conditions is a surface-controlled process similar to the previous report [67]. 50 mV/s was selected as the scanning speed in later experiments. Figure 3C shows the CV results when the concentrations of NaOH are 5, 10, 20, and 50 mM with same concentration of glucose (2 mM) and at the same scan rate (50 mV/s). The current densities of CVs increase with the concentrations of NaOH. This is because the conductivity of electrolyte increases at higher pH and the mutarotation of glucose becomes easy with the increase of pH [63], which speeds up the electrocatalytic oxidation of glucose, and more Cu(III) was produced. 50 mM NaOH was selected as the fixed concentration for later experiments. In our experiments, 30-125 µg load of Cu 2 O-BSA NPs was used to modify the GCEs. With the increase of the NPs loads, the current densities increase first then decrease. When the load of Cu 2 O NPs is 50 µg, the maximum current density was obtained. The reason for the above results is that more absorption sites and catalyzing sites are supplied with the increase of Cu 2 O NPs load, but excessive load of NPs on the GCE leads to the increase of electrode resistance and the decrease of current. The optimal load of Cu 2 O NPs of 50 µg was assumed in later experiments. Figure 4A shows a chronoamperometric diagram of the Nafion/Cu 2 O-BSA/GCE responding to successive addition of glucose at a work potential of 0.6 V (vs. Ag/AgCl). The concentrations of glucose in the electrolyte are in the range or 200 nM−11.67 mM. For clarity, the chronoamperometric curve when the concentrations of glucose are in the range of 11.1 µM−216.1 µM is shown in the inset. Figure 4B presents the calibration curve (black curve) for the Nafion/Cu 2 O-BSA/GCE based on the result of Figure 4A. The red line is the Langmuir isothermal fitting curve [68]. One can see the linear range is up to 10 mM and the sensitivity is 1144.81 µAcm −2 mM −1 . The corresponding regression equation is I (µA/cm 2 ) = 1144.81C (mM) + 80.98 (µA/cm 2 ) (R 2 = 0.98528). When the concentration of glucose is very low, the current of the electrode is also very low, and the noise is very high. Therefore, a separate study was carried out. Figure 4C displays an enlarged chronoamperometric curve of the Nafion/Cu 2 O-BSA/GCE when the concentration of glucose is as low as 200 nM. One can see from Figure 4C that the lowest detection limit is 400 nM (S/N = 3), and the response time is about 7 s. Therefore, the Nafion/Cu 2 O-BSA/GCE has a wide linear detection range, a high sensitivity, a low detection limit, and a quick response time to glucose. In comparison to the performance of Cu 2 O-modified GCE in recent years (Table 1), the Cu 2 O-B SA NPs modified GCE have high sensitivity and wide linear range. The higher sensitivity is attributed to high electrocatalytic activity of Cu 2 O-BSA NPs with large surface area, BSA has good adsorption ability to glucose and good conductivity to electrons [58,59,61]. In comparison with our previous hollow Cu 2 O-PVP/Nafion-modified GCE [50], the linear detection range of Cu 2 O-BSA increases by 267 times.   The storage stability of Nafion/Cu 2 O-BSA/GCE was conducted over six days of consecutive testing, with 2 mM glucose in 50 mM NaOH solution for 6 min, and the voltage fixed at 0.6 V. After testing, the electrode was dipped in pure water for 3 min, three times, then taken out and dried in air. The electrode was stored at room temperature when not in use. The results are shown in Figure 6. For the same Nafion/Cu 2 O-BSA/GCE, the current response decreased over time. On the sixth day, the current density fell to 80% of that of the first day. In our previous experiment [50], Cu 2 O-PVP/Nafion-modified GCE had good stability for five weeks. We guess this result may be relative to the coating agent of BSA. BSA is a type of protein, which is easier influenced by environment than PVP. In addition, a group of six Nafion/Cu 2 O-BSA/GCEs were prepared by the same method and used to do the chronoamperometry experiments, a relative standard deviation (RSD) of 6.9% was obtained, which indicates good fabrication reproducibility of these sensing electrodes. The Cu 2 O-BSA NPs we prepared make good material for the fabrication of sensitive, specific, and rapid amperometric sensors for the non-enzymatic detection of glucose.

Conclusions
In this paper, Cu 2 O-BSA NPs were synthesized in air at room temperature by Cu(NO 3 ) 2 ·3H 2 O as a precursor, BSA as a coating agent, and AA as a reductant. The sphere NPs with size of 180-220 nm have good monodispersity. The as-prepared Cu 2 O-BSA NPs were used to modify the GCE. By using CV and chronoamperometry, Cu 2 O-BSA NPs showed direct electrocatalytic activity for the oxidation of glucose in 50 mM NaOH solution at 0.6 V. The Nafion/Cu 2 O-BSA NPs film provides rich active sites for the sensing of glucose. This constructed sensing electrode was used in the detection of glucose with the lowest detection limit of 0.4 µM, a linear detection range up to 10 mM, a high sensitivity of 1144.81 µAmM −1 cm −2 and a quick response time of 7 s. The common interfering contents of Cl − , UA, AA, and AP existing in the physiological environment have almost no interference in the oxidation of glucose for the Nafion/Cu 2 O-BSA/GCE. For the green synthesis of Cu 2 O-BSA NPs and facile modification of GCE, there was excellent performance of the Nafion/Cu 2 O-BSA/GCE for glucose, and Cu 2 O-BSA NPs will become promising nanostructures for the fabrication of non-enzymatic glucose electrochemical sensing devices.