In order to detect low level glucose concentration, an electrochemiluminescence (ECL) biosensor based on TGA-capped CdSe quantum dots (QDs) was fabricated by the immobilization of CdSe QDs after modifying the surface of a glassy carbon electrode (GCE) with 4-aminothiophenol diazonium salts by the electrochemical method. For the detection of glucose concentration, glucose oxidase (GOD) was immobilized onto the fabricated CdSe QDs-modified electrode. The fabricated ECL biosensor based on TGA-capped CdSe QDs was characterized using a scanning electron microscope (SEM), UV-vis spectrophotometry, transmission electron microscopy (TEM), a fluorescence spectrometer (PL), and cyclic voltammetry (CV). The fabricated ECL biosensor based on TGA-capped CdSe QDs is suitable for the detection of glucose concentrations in real human blood samples.

1. Introduction

Electrogenerated chemiluminescence (ECL) has become an important and valuable detection method in analytical chemistry in recent years [1]. ECL is well known because of its good stability against photo bleaching, high sensitivity, favorable spatial qualities, time controllability, low cost, and low background noise. Since the first detailed studies in the mid-1960s, over 1000 papers, patents, and book chapters have been published on ECL, ranging from the very fundamental to the very applied. Recently, ECL has been used extensively as a powerful analytical tool in many areas such as immunoassay [24], clinical diagnosis [5, 6], and the analysis of food and water [7]. Luminophores can be divided into organic and inorganic luminophores such as luminol, lucigenin, tris(bipyridine)ruthenium(II), and , and they have been widely used in immunoassay and DNA analysis because they result in light emission and allow for detection of the emitter at very low concentrations [8, 9].

Semiconductor nanocrystals such as quantum dots (QDs) have attracted considerable interest because of the advantages of their size-dependent electronic and optic properties, which have been extensively studied and applied in recognition and detection in biometry [10, 11]. A remarkable increase in ECL intensity has been reported for quantum dot (QD) composites with carbon nanotubes deposited on electrode surfaces, graphene, carbon paste electrodes, and nanoparticles. Among the detection techniques in the biomedical field, ECLs based on a QD system have been known for offering a wide range of advantages such as broad excitation spectra, narrow emission spectra, and low cost [1214]. In a previous paper, we also prepared an ECL biosensor using CdSe QDs as a luminophore for detection of acetylcholine in human blood. In this experiment, we used L-cystein as the capping chemicals of the QD surface [15]. However, with L-cystein as the capping chemicals, there is a decrease of ECL intensity because of quenching effects. Therefore, there is a need for advanced CdSe QDs with capping chemicals without quenching effects to enable the effective use of luminophore materials. Furthermore, we need an easy and simple preparation method of CdSe QDs.

In this study, a novel ECL biosensor based on thioglycolic acid- (TGA-) capped CdSe QDs was prepared to provide improved sensitivity, selectivity, serviceability, and stability along with low cost. First, the TGA-capped CdSe QDs were synthesized in an aqueous solution using thioglycolic acid (TGA). An ECL biosensor based on TGA-capped CdSe QDs was prepared by the modification of the GCE with multiwalled carbon nanotubes (MWNTs) and TGA-capped CdSe QDs, and then glucose oxidase (GOD) was immobilized on the surface for reaction with glucose. Furthermore, the fabricated ECL biosensor can be applied to the quantitative and qualitative analysis of glucose concentration.

2. Experimental

2.1. Reagents

MWNTs (95%; length 10 μm) were obtained from Hanwha Nanotech (South Korea). Cadmium sulfate hydrate (CdSO4·(3/8)H2O) was obtained from Junsei Chemical (South Korea), and selenium powder (Se powder), sodium nitrite (NaNO2), 4-aminothiophenol (4-ATP), sodium borohydride (NaBH4), D-(+)-glucose, and glucose oxidase (GOD, 17300 unit/g) were purchased from Sigma-Aldrich (USA). Phosphate buffer solution (PBS) was prepared by mixing stock solutions of NaH2PO4 and Na2HPO4 and then adjusting to pH = 7.0. All solutions for the experiments were prepared with water purified in a Milli-Q plus water purification system (Millipore, USA). Glassy carbon electrodes (GCEs) were purchased from Bioanalytical Systems (USA) (model MF-2012).

2.2. Preparation of Biosensor Based on TGA-Capped CdSe QDs for Detecting Glucose

The fabrication procedure of the glucose oxidase- (GOD-) immobilized biosensor based on TGA-capped CdSe QDs is exhibited in Scheme 1. The GCE was modified by the following steps. Prior to modification, the surface of the GCE was cleaned by polishing with 0.3 μm α-Al2O3 powder and then rinsed with distilled water and cleaned ultrasonically in acetonitrile for 3 min. MWNT was ultrasonicated at room temperature in a 3 : 1 mixture of sulfuric acid and nitric acid. The obtained MWNT was dispersed in a nafion solution, which was a mixture of 1.0 mL methanol and 40 μL nafion, to give an MWNT suspension. Then a 20 μL MWNT suspension was dropped onto a cleaned GCE (MWNT/GCE). Meanwhile, 0.07 g 4-aminothiophenol (4-ATP) and 0.2 g sodium nitride (NaNO2) were dissolved in 40 mL distilled water at 0~5°C. After the two reagents were mixed, 6.0 M cooled hydrochloride solution was slowly added to the mixture solution. After vigorous stirring, a bright yellow 4-aminothiophenol diazonium salt solution was obtained. The synthesized 4-aminothiophenol diazonium salts were deposited electrochemically on the surface of the MWNT/GCE electrode at constant potential of −0.7 V for 1200 s (named as SH/MWNT/GCE). To immobilize the TGA-capped CdSe QDs onto the surface of the SH/MWNT/GCE by self-assembly, the SH/MWNT/GCE was immersed for 10~15 h in 5 mL synthesized TGA-capped CdSe QD suspension (CdSe/SH/MWNT/GCE). Finally, 2.0 μL glucose oxidase (GOD, 3.0 mg/mL) was dropped on the surface of the CdSe/SH/MWNT/GCE to fabricate an ECL biosensor based on TGA-capped CdSe QDs for clinical diagnosis with a diabetic.

2.3. Preparation of TGA-Capped CdSe QDs

The preparation procedure of the thioglycolic acid- (TGA-) capped CdSe QDs using reduction reagents is shown in Scheme 2. Se powder (0.02 g) and NaBH4 (0.10 g) in ethanol (15.0 mL) were added in the left side round flask and stirred until dissolved. In the other round flask, CdCl2 (0.01 g) and TGA (0.01 g) with NaOH in 50 mL water were dissolved at 90°C. Next 0.1 M H2SO4 was dropped in the left side round flask using a dropping funnel in order to prepare H2Se gas. The preparation mechanism of the TGA-capped CdSe QDs is also shown in Scheme 2. The result was that we could obtain TGA-capped CdSe QDs of various sizes by controlling the reaction time.

2.4. Instrumentation

Cyclic voltammetry (CV) was performed using a potentiostat/galvanostat model 283 (Ametek PAR, USA) and a conventional three-electrode system comprising a composite-coated glassy carbon (diameter 2 mm) working electrode, a platinum wire counterelectrode, and an Ag/AgCl (saturated KCl) reference electrode. Electrochemical signals were recorded using an H7468-01 photomultiplier tube (PMT) system (Hamamatsu Photonics, Japan). The entire ECL cell was enclosed in a light-proof dark box. ECL measurements were carried out in 0.1 M PBS (pH 7). The applied working potential ranged from 0 to −2.0 V and the cycle scan rate was 100 mV/s. A high voltage power supply of 750 V was applied to the PMT and all experiments were performed 10 times.

3. Results and Discussion

3.1. Synthesis and Characterization of TGA-Capped CdSe QDs

Quantum dots are widely used not only in the photoelectrochemical field but also in biosensors, biological imaging, and bioconjugates because of their remarkable electronic and optic properties. First, TGA-capped CdSe QDs were easily synthesized using reducing agents. The morphology and particle size of the TGA-capped CdSe are shown in Figure 1. Overall, the TGA-capped CdSe QDs were spherical particles with an aggregation of small particles of 3~5 nm size. The UV-vis spectrum, photoluminescence (PL) spectrum, and DLS data of TGA-capped CdSe QDs are shown in Figure 2. The absorbance spectrum displays an excitation peak at 428 nm. The luminescence spectra at 610 nm are confirmed. In the UV spectrum, the band gap and the size of the TGA-capped CdSe QDs were calculated as 2.24 eV and 10.8 nm, respectively, as follows:

In the DLS data, the size of the TGA-capped CdSe QDs was 84 nm because of the aggregation of small QDs of 6 nm with each other. The results showed that the TGA-capped CdSe QDs in aqueous solution were aggregated at room temperature.

3.2. Fabrication and Characterization of a GOD-Immobilized Biosensor Based on CdSe QDs

In our previous study, a conjugated phenylene polymer with carboxylic acid was grafted onto the MWNT surface of an electrode by electrochemical polymerization. Here, the conjugated phenylene polymer-grafted MWNT can be also used as electron transfer supports on electrodes. Furthermore, the thiol group placed on the supports of the electrode can serve as a binding site for covalent attachment to crystal metal nanoparticles. The SEM surface images during the preparation process of the GOD-immobilized biosensor are displayed in Figure 3, where (a) is GCE surface, (b) is surface image of the MWNT/GCE electrode, (c) is the surface image of the SH/MWNT/GCE, (d) is the surface image of the CdSe/SH/MWNT/GCE, and (e) is the GOD-immobilized biosensor, namely, GOD/CdSe/SH/MWNT/GCE. After modification using MWNT suspension, the surface of the GCE appears as a crystal-like form (b), while the surface of the TH/MWNT/GCE (c), which means the conjugated phenylene polymer-grafted MWNT/GCE, shows a smooth form because of the formation of grafted polymer on the MWNT/GCE electrode. As a result of this step, there were no significant changes in terms of the immobilization of the CdSe QDs onto the surface of the SH/MWNT/GCE because of differences in nanosize; however, there was a significant change in morphology to a crystal solid form after the immobilization of the GOD as enzyme onto the CdSe/SH/MWNT/GCE (Figure 3(e)) because of the dry state GOD. The results confirmed that the GOD-immobilized ECL biosensor based on CdSe QDs had been prepared successfully for the detection of glucose.

The contact angle images of bare GCE (a), MWNT/GCE (b), SH/MWNT/GCE (c), CdSe/SH/MWNT/GCE (d), and GOD/CdSe/SH/MWNT/GCE (e) are shown in Figure 4. From an examination of the water contact angles (Figure 4), the value of the contact angle of the GCE was determined to be 64.3° (Figure 4(a)) and that of the MWNT/GCE was determined to be 64.2° (Figure 4(b)). After electrochemical reduction of 4-aminothiophenol on the surface of the GCE, the contact angle of the SH/MWNT/GCE increased to 71.5° (Figure 4(c)), while the contact angle of the CdSe/SH/MWNT/GCE decreased to 49.5° (Figure 4(d)) owing to immobilization of CdSe QDs onto the electrode. This result means that the GCE was modified successfully by electrochemical reduction and immobilization of CdSe QDs; moreover, hydrophilicity was introduced to the surface. In addition, the immobilization of GOD on the electrode surface led to a slight decrease in the contact angle (Figure 4(e)) because of increasing hydrophilicity. From these results, we could confirm that the GOD-immobilized biosensor was successfully fabricated.

Relatively dense decoration of the CdSe QDs on the electrode was observed in the GOD-immobilized biosensor. Energy dispersive X-ray (EDX) analysis (Figure 5) confirmed that Cd and Se were present in the GOD-immobilized biosensor. The calculated atomic ratios of Cd to Se were shown to be close to 1 : 1 (Figures 5(b) and 5(c)), which is in good agreement with the stoichiometric ratio of CdSe. These results confirm that the GOD-biosensor based on CdSe QDs was successfully prepared with phenylene polymer-grafted MWNT.

In order to confirm the introduction of the thiol group and CdSe QDs after immobilization of QDs and electrochemical reduction of 4-aminothiophenol diazonium salt, a surface analysis was performed (Figure 6) to provide high resolution XPS spectra of S2p, Cd3d, and Se3d on the SH/MWNT/GCE (a), CdSe/SH/MWNT/GCE (b), and GOD/CdSe/SH/MWNT/GCE (c). The doublet XPS spectrum of S2p on the SH/MWNT/GCE at 163.9 eV and 169.7 eV (Figure 6(a)) was confirmed, which is indicative for the signal of S in 4-thiophenyl groups, while the unknown peak originated from a disulfide or sulfide bond of conjugated aromatic polymer on the modified carbon electrode. After immobilization of the CdSe QDs, the doublet Cd3d peaks at 406 eV and 414 eV and Se3d at 55 eV appeared for the CdSe/SH/MWNT/GCE and GOD/CdSe/SH/MWNT/GCE, respectively. Also there was a decrease in the intensity of the Cd3d and Se3d peaks after immobilization of GOD onto the CdSe/SH/MWNT/GCE. From these results, we confirmed that the GOD-immobilized biosensor was successfully prepared for detection of glucose.

3.3. Characterization of an ECL Biosensor Based on CdSe QDs for the Detection of Glucose

Cyclic voltammetry and ECL intensity measurements of the fabricated ECL biosensor were carried out in 0.1 M PBS (pH = 7.0) at 100 mVs−1 to show the product mechanism of H2O2 from glucose by GOD and the ECL signal mechanism of the ECL biosensor based on CdSe QDs (Figure 7).

The electrochemical behavior of the fabricated ECL biosensor (GOD/CdSe/SH/MWNT/GCE) was confirmed as a function of glucose concentration under optimized experimental conditions (Figure 8). The redox peaks of H2O2 (below 10−3 M) could not be obtained (Figure 8(a)): however, the ECL signals were obtained to show a similar pattern (Figure 8(b)) without any change of glucose concentration. With an increasing concentration of glucose (Figure 8(c)), there was an increase in the concentration of H2O2 as a decomposition product coming from the chemical reaction between glucose and the GOD enzyme. For this reason, the ECL signal is increased because the intensity of the ECL signal is affected directly by the concentration of H2O2. The concentration of glucose enhanced the intensity of the ECL signal of the fabricated ECL biosensor based on CdSe QDs in 0.1 M pH 7.0 (see Figure 8). The calibration curve for glucose was constructed using a current ECL biosensor based on CdSe QDs. The signal intensity was linearly proportional to the concentration of glucose between 1.0 × 10−3 and 1.6 × 10−14 M (Figure 8(d)).

4. Conclusion

A rapid and sensitive ECL biosensor based on CdSe QDs was developed for the determination of glucose. The CdSe QDs were synthesized using TGA, and their morphology and electrochemical properties were confirmed by TEM, UV-vis spectrum, and PL analysis. Also, conjugated phenylene polymer was formed easily on the surface of the electrode by an electrochemical method, which was used for the immobilization of CdSe QDs and enzymes. In this luminescence mechanism, the ECL-potential signal of CdSe QDs resulted from the formation of H2O2, which arises from a chemical reaction between glucose and glucose oxidase (GOD). These results could be applied to the clinical diagnosis of diabetes. Fabricated ECL biosensors based on CdSe QDs have great potential to provide rapid, sensitive, and cost-effective approaches for the clinical diagnosis of diabetes.

Competing Interests

The authors declare that they have no competing interests.


This work was partly supported by the National Research Foundation of Korea Grant, funded by the Korean Government (NRF-2015N043).