Lili Bie,
Xue Luo,
Qingqing He,
Daiping He,
Yan Liu and
Ping Jiang*
Key Laboratory of Green Synthesis and Applications, College of Chemistry, Chongqing Normal University, Chongqing 401331, China. E-mail: jphdp868@126.com
First published on 20th September 2016
Hierarchical Cu/Cu(OH)2 nanorod arrays grown on commercially available Cu foam (Cu/Cu(OH)2 NRA/CF) was prepared via a three-step strategy involving the wet chemical synthesis of Cu(OH)2 NRA/CF, a chemical reduction reaction for conversion from Cu(OH)2 NRA/CF to Cu NRA/CF, and finally galvanostatic anodization to grow Cu(OH)2 nanoparticles on Cu NRA/CF. Hierarchical Cu/Cu(OH)2 NRA/CF shows high catalytic activity towards glucose oxidation in an alkaline solution and can serve as a promising electrode material for enzyme-free glucose sensing. At an applied potential of +0.5 V, the sensor showed a broad detection range of 0.001–1.0 mM, a high sensitivity of 9.18 mA mM−1 cm−2, and a low detection limit of 0.45 μM (S/N = 3). Furthermore, the sensor exhibits excellent selectivity against common interferents and good reliability for glucose detection in human serum samples.
Hierarchical nanostructured architectures18–20 self-assembled from low-dimensional nanostructured building blocks have become a class of significant functional materials. Due to their larger surface, good conductivity and permeability, hierarchical nanomaterials arouse much interest in many emerging research fields, especially environmental remediation, sensor, and energy conversion. Various hierarchical nanostructures such as hyper-branched Cu@Cu2O coaxial nanowires,21 hierarchical Ni/NiO core–shells nanostructure22 and 3D hierarchical cobalt oxide architecture,23 have been demonstrated as novel electrode materials for glucose sensing with high sensitivity and good selectivity. For electrochemical tests, however, it requires the effective immobilization of these hierarchical nanomaterials on electrode surfaces with the use of polymer binder (Nafion or PTFE) as a film-forming agent. The whole process is time-consuming and such polymer binder may block the active sites of the catalyst, inhibit the diffusion of analyte, and increase the series resistance of the device, leading to reduced catalytic activity.24,25 As a result, it is of significant importance to prepare self-supported electrode by directly growing hierarchical nanostructures on conductive substrate.26–29
Metal foam such as Cu foam30,31 and Ni foam,26,32 as a commercial available material with 3D porous structure and excellent electrical conductivity, has been used extensively as an efficient conductive substrate for constructing various self-supported electrodes. Due to the unique pore structure and large surface area, metal foam not only facilitates sufficient transport of reactants and products in electrochemical process, but also provides more catalyst loading and more active sites exposure, leading to an improved catalytic activity.
In this paper, we demonstrated the successful synthesis of hierarchical Cu/Cu(OH)2 nanorod arrays supported on Cu foam (Cu/Cu(OH)2 NRA/CF) via a three-step strategy. Firstly, Cu(OH)2 nanorod arrays grown on Cu foam (Cu(OH)2 NRA/CF) were fabricated by a simple wet chemical reaction. The Cu(OH)2 NRA/CF were subsequently converted to Cu nanorod arrays on Cu foam (Cu NRA/CF) by a NaBH4 reduction reaction. Finally, galvanostatic anodization of corresponding Cu NRA/CF produces hierarchical Cu/Cu(OH)2 NRA/CF. As a novel self-supported electrode for enzyme-free glucose sensing, the presented Cu/Cu(OH)2 NRA/CF exhibits superior catalytic activity than other sensors including Cu(OH)2 NRA grown on conductive substrate, commercial Cu foam, Cu/Cu(OH)2 NRA grown on Cu foil (Cu/Cu(OH)2 NRA/CFL), and various other hierarchical nanostructured glucose sensors. The high performance of Cu/Cu(OH)2 NRA/CF can be attributed to the following reasons: (1) the 3D CF used as conductive substrate with a large surface area leads to a significant increase in the catalyst loading and more active site exposure. (2) Hierarchical structure of Cu/Cu(OH)2 NRA owns large specific surface area, facilitating more active site exposure. (3) The direct growth of Cu/Cu(OH)2 NRA on CF enables intimate contact, good adhesion and excellent electrical connection between them.33 Moreover, the absence of a polymer binder avoids the blocking of active site24 and the inhibition of diffusion.25
Scheme 1 Three-step synthesis process for hierarchical Cu/Cu(OH)2 NRA/CF and its application for glucose sensing. |
Fig. 1 shows the X-ray diffraction (XRD) patterns for Cu(OH)2 NRA/CF (curve a), Cu NRA/CF (curve b) and Cu/Cu(OH)2 NRA/CF (curve c), respectively. Curve a shows diffraction peaks indexed to orthorhombic phase Cu(OH)2 (JCPDS no. 72-0140)34 and that of metallic copper (JCPDS no. 04-0836)35 originated from CF, suggesting successfully synthesis of Cu(OH)2 NRA grown on Cu foam by wet chemical method. After chemical reduction with NaBH4 solution, only the diffraction peaks (curve b) of the copper are observed at 43.3°, 50.4° and 74.1°, which are well indexed to (111), (200) and (220) planes of metallic Cu (JCPDS no. 04-0836), implying the complete reduction of Cu(OH)2 NRA. Following galvanostatic anodization, the diffraction peaks (curve c) indexed to Cu(OH)2 (JCPDS no. 72-0140) appears again, indicating that galvanostatic anodization cause the growth of Cu(OH)2 on Cu NRA/CF.
Fig. 1 XRD patterns of Cu(OH)2 NRA/CF (curve a), Cu NRA/CF (curve b) and Cu/Cu(OH)2 NRA/CF (curve c). |
Fig. 2a shows SEM images of Cu(OH)2 NRA/CF in low-magnification, indicating they are 1D structures about several micrometers in length. The high-magnification SEM image (see Fig. 2b) further reveals that they have diameters ranging from 100 nm to 200 nm with smooth surface. After chemical reduction with NaBH4 solution, such Cu(OH)2 NRA were found converted to Cu NRA with diameter less than 100 nm, rough surface and decreased length (Fig. 2c and d). The development of such morphology is associated with hydrogen release and conversion from Cu(OH)2 to metallic Cu. Galvanostatic anodizing Cu NRA in an alkaline solution for 200 s produces a large quantity of uniform nanoparticles with diameter around 50 nm grown on its surface (Fig. 2e and f), constructing a hierarchical structure consisted by one-dimensional nanorods and zero-dimensional nanoparticles. XRD patterns in Fig. 1 confirmed the composition of these nanostructures. All results suggests the successful fabrication of hierarchical structure of Cu/Cu(OH)2 NRA/CF.
Fig. 2 Low and high-magnification SEM images of Cu(OH)2 NRA/CF (a and b), Cu NRA/CF (c and d), and Cu/Cu(OH)2 NRA/CF (e and f). |
Nitrogen adsorption–desorption isotherm measurement (Fig. S1†) was conducted to investigate the surface areas of the as prepared Cu/Cu(OH)2 NRA scratched down from the Cu foil. The Brunauer–Emmett–Teller (BET) specific surface area (SSA) of Cu/Cu(OH)2 NRA was calculated to be 27.60 m2 g−1, much larger than that of Cu NRA (12.25 m2 g−1), which can be attributed to its hierarchical structure.
Electrochemical impedance spectrum (EIS) is an effective way to measure the electron transfer resistance. Fig. 3a shows the typical EIS of the CF, Cu(OH)2 NRA/CF, Cu NRA/CF, and Cu/Cu(OH)2 NRA/CF. As for bare Cu foam, a small semicircular portion at high frequency is observed, illustrating low transfer resistance and good conductivity of the CF. After being covered with Cu(OH)2 NRA, the semicircle diameter was observed increases from a value close to 30 ohm up to 110 ohm, implying inferior conductivity of Cu(OH)2. However, it returns to a value about 25 ohm after the complete conversion from Cu(OH)2 NRA/CF to Cu NRA/CF. Compared with the bare CF, Cu NRA/CF shows lower transfer resistance which can be attributed to the superior conductivity of Cu nanostructure. After the growth of Cu(OH)2 nanoparticles on the surface of Cu NRA, the semicircle diameter has a slight increase but still keeps smaller than that of Cu(OH)2 NRA/CF, implying superior conductivity and faster electron transfer of the hierarchical Cu/Cu(OH)2 NRA/CF.
The electrochemical behavior of Cu/Cu(OH)2 NRA/CF was examined in 0.1 M NaOH solution (Fig. 3b). For comparison, CF and Cu/Cu(OH)2 NRA/CFL were also examined. Like any other Cu based nanomaterials,31,36,37 the oxidative current of Cu(II) to Cu(III) in the potential range +0.4 V to +0.7 V was not clear. However, the cathodic peak at about +0.57 V, corresponding to the reduction of Cu(III) to Cu(II), was observed on reversing the scan, clearly indicating the presence of Cu(III).38 Here, we compared the cyclic voltammograms (CVs) behaviors of three kinds of Cu based electrodes. As is shown, bare CF exhibits lower current response while Cu/Cu(OH)2 NRA grown on both CF and CFL show higher amperometric responses, which should be attributed to the larger specific surface area of the Cu/Cu(OH)2 NRA originated from its hierarchical nanostructure. Due to the pore structure, CF owns larger specific surface and benefits more Cu/Cu(OH)2 NRA loading and more active sites exposure. Consequently, Cu/Cu(OH)2 NRA/CF exhibits larger current responses than that of Cu/Cu(OH)2 NRA/CFL in the same conditions.
The electrocatalytic activity of Cu/Cu(OH)2 NRA/CF towards glucose oxidation in an alkaline solution was also investigated. Fig. 4a shows the CVs of Cu/Cu(OH)2 NRA/CF electrode in the absence (black curve) and presence (red curve) of 1.0 mM glucose in 0.1 M NaOH solution. No obvious oxidation peak was observed in absence of glucose. However, a gradual increase in anodic current was observed at +0.5 V when the glucose was added, followed a slight decrease in cathodic current, indicative of an irreversible electro-oxidation reaction that occurred at approximately +0.5 V during scanning.31 According to the previous research, the enhanced anodic current is attributed to the electrochemical oxidation of glucose with the participation of Cu(III).39,40 The possible mechanism of the electrocatalytic process of glucose can be expressed as followings:38
Cu(II) + OH− → Cu(III) + e− |
Cu(III) + glucose → Cu(II) + gluconolactone |
Fig. 4b shows the CVs curves of Cu/Cu(OH)2 NRA/CF in 0.1 M NaOH solution including 0.1 mM glucose at different scan rates. The anodic peak current of oxidation of glucose (the corresponding voltage is at +0.5 V) was found proportional to the scan rate. As indicated in the inset of Fig. 4b, a good linearity between scan rate and perk current within the range of 1–50 mV s−1 was observed, indicating a surface-controlled electrochemical kinetic process of Cu/Cu(OH)2 NRA/CF in alkaline solution.
The optimal operating potential for electrochemical detecting glucose with Cu/Cu(OH)2 NRA/CF was further studied. Fig. 5 shows the amperometric responses towards dropwise addition of 0.01 mM glucose into 0.1 M NaOH solution at different potentials. A well-defined, stable and rapid change in current can be observed with the successive additions of 0.01 mM glucose under detection potentials from 0.40 V to 0.55 V in 0.05 V increments. The Cu/Cu(OH)2 NRA/CF exhibits sensitive and rapid current response to the glucose addition, achieving steady state current in less than 5 s. The sensitivity is enhanced steadily in the range from 0.40 to 0.50 V. However, a slight decrease is observed when the potential beyond 0.55 V. Therefore, the followed amperometric detection of glucose was performed under the potential of 0.50 V.
Fig. 5 Amperometric response with successive addition of 0.01 mM glucose at different detection potentials. |
Fig. 6a shows an amperometric response of Cu/Cu(OH)2 NRA/CF with successive addition of the glucose solution with different concentrations into 0.1 M NaOH solution, at an applied potential of 0.50 V. The addition of glucose was incremented at the rate of 1, 5, 10, 50, 100 and 500 μM, respectively. As shown in Fig. 6b, the Cu/Cu(OH)2 NRA/CF gives a linear dependence range from 1 μM to 1 mM of glucose with a correlation coefficient (R2) of 0.991 and a remarkably high sensitivity of 9.18 mA mM−1 cm−2. Such sensitivity is better than those of 418 μA mM−1 cm−2 at Cu(OH)2 nanotube arrays38 on copper foils, and 5.85 mA mM−1 cm−2 at commercial Cu foam.41 Moreover, compared with other copper-based enzyme-free glucose sensors with hierarchical nanostructure (Table S2†), the sensitivity is still better than those of 1.321 mA mM−1 cm−2 at CuO/TiO2 nanocomposites,42 2.657 mA mM−1 cm−2 at hierarchical CuO nanoflowers,43 1.42 mA mM−1 cm−2 at hyper-branched Cu@Cu2O coaxial nanowires,21 0.408 mA mM−1 cm−2 at CuO/ZnO hybrid hierarchical nanostructures,44 3.252 mA mM−1 cm−2 at chrysanthemum-like CuO hierarchical film,36 4.078 mA mM−1 cm−2 at candock-like CuO hierarchical film,36 5.368 mA mM−1 cm−2 at dandelion-like CuO hierarchical film,36 and 3.36 mA mM−1 cm−2 at Cu(OH)2 nanorods decorated porous graphene foam (Cu(OH)2/PGF).45 The superior sensitivity for Cu/Cu(OH)2 NRA/CF can be attributed to the following three reasons: (1) the porous feature for CF not only favors more catalyst loading and more active sites exposure, but allows more easy diffusion of electrolyte and faster electron transfer in process of electrochemistry. (2) The hierarchical structure of Cu/Cu(OH)2 NRA also facilitates the exposure of more active sites during the electrocatalytic process. (3) The direct growth of Cu/Cu(OH)2 NRA on CF avoids the use of polymer binder, thus effectively decreases the contact electrical resistance between them and benefits more active site exposure.
The response time was found less than 5 s to reach 95% of steady-state current from the enlarged segment of the amperometric response curve (inset in Fig. 6a) and the limit of detection (LOD) was found to be 0.45 μM based on signal-to-noise ratio of 3. The detection limit of the as-prepared Cu/Cu(OH)2 NRA/CF is relatively low among other hierarchical Cu based nanostructures (Table S1†), implying high performance towards glucose sensing.
It is well known that some easily oxidative species such as dopamine (DA), ascorbic acid (AA), and uric acid (UA), usually co-exist with glucose in biological sample and consequently interfere with the detection of glucose. Therefore, anti-interference property is another important parameter for the practical use of glucose sensors. Herein, the interference test was also investigated using Cu/Cu(OH)2 NRA/CF to evaluate the possible application in glucose detection in human serum. Since the normal physiological level of glucose (3–8 mM) is more than 10 times higher than those of the other three interfering species,46 the molar ratio of 10:1 was used for AA, UA, and DA to estimate their interference effects in the glucose sensing. Fig. 7a shows the amperometric response of the Cu/Cu(OH)2 NRA/CF for 0.1 mM glucose at 0.5 V, followed by the additions of 0.01 mM UA, 0.01 mM DA and 0.01 mM AA in a 0.1 M NaOH solution. Obviously, almost no current response for such electroactive interfering species was detected in comparison to that of glucose, indicating the high selectivity of our present Cu/Cu(OH)2 NRA/CF based sensor for glucose detection in comparison to the normally co-existing electroactive interfering species.
We further evaluated the reliability of the Cu/Cu(OH)2 NRA/CF sensor by detecting the concentration of glucose in human serum samples. All serum samples were supplied by Chongqing Emergency Medical Central. Fig. 7b shows the amperometric responses of the Cu/Cu(OH)2 NRA/CF sensor upon addition of 5 μM glucose and three human serum samples at an applied potential of 0.5 V in 0.1 M NaOH solution. The determination of each sample was done three times to verify the reproducibility of the values. The concentration of glucose was determined by comparing the amperometric response of 5 μM glucose and that of human serum sample. Table 1 lists the glucose concentrations in serum samples obtained by the Cu/Cu(OH)2 NRA/CF sensor and those measured by the hospital. The relative standard deviations (RSD) and bias are less than 3% and 0.15 mM, respectively. Obviously, the Cu/Cu(OH)2 NRA/CF is suitable for glucose detection in human serum.
Sample | Glucose concentration obtained from the hospital (mM) | Glucose concentration obtained from the present sensor (mM) | RSD (%) | Bias (mM) |
---|---|---|---|---|
1 | 6.26 | 6.19 | 1.06 | −0.07 |
2 | 5.75 | 5.87 | 2.29 | 0.12 |
3 | 4.73 | 4.83 | 2.45 | 0.10 |
For practical applications, the sensors should have the advantage of good reproducibility and long-term stability. Therefore, we further probed the reproducibility and stability of the Cu/Cu(OH)2 NRA/CF sensor in our experiments with the amperometric method. The reproducibility was evaluated by the measurements of the response currents towards 0.2 mM glucose in 0.1 M NaOH solution at the Cu/Cu(OH)2 NRA/CF sensor. The results of five parallel experiments show that the Cu/Cu(OH)2 NRA/CF sensor has good reproducibility (RSD = 2.72%, n = 5). Moreover, the above-mentioned results obtained from human samples (Table 1) also confirm the good reproducibility of present sensor.
The long-term stability of the Cu/Cu(OH)2 NRA/CF sensor was examined by measuring the amperometric response currents towards 0.01 mM glucose in 0.1 M NaOH solution. As shown in Fig. S2,† the current response was found maintained 80% of the original value after a month conservation at room temperature, suggesting the good long-term stability of the Cu/Cu(OH)2 NRA/CF sensor.
Footnote |
† Electronic supplementary information (ESI) available: Experimental section and figures. See DOI: 10.1039/c6ra19576h |
This journal is © The Royal Society of Chemistry 2016 |