Analytica Chimica Acta 809 (2014) 134–140

Contents lists available at ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Sensors and Bioselective Reagents

Integration of a highly ordered gold nanowires array with glucose oxidase for ultra-sensitive glucose detection Jiewu Cui a,b , Samuel B. Adeloju a,∗ , Yucheng Wu b,1 a NanoScience and Sensor Technology Research Group, School of Applied Sciences and Engineering, Monash University, Gippsland Campus, Churchill 3842, VIC Australia b Laboratory of Functional Nanomaterials and Devices, School of Materials Science and Engineering, Hefei University of Technology, Hefei 230009, Anhui, China

h i g h l i g h t s

g r a p h i c a l

a b s t r a c t

• Successfully • • • •

synthesised highlyordered gold nanowires array with an AAO template. Fabricated an ultra-sensitive glucose nanobiosensor with the gold nanowires array. Achieved sensitivity as high as 379.0 ␮A cm−2 mM−1 and detection limit as low as 50 nM. Achieved excellent anti-interference with aid of Nafion membrane towards UA and AA. Enabled successful detection and quantification of glucose in human blood serum.

a r t i c l e

i n f o

Article history: Received 15 May 2013 Received in revised form 22 October 2013 Accepted 12 November 2013 Available online 19 November 2013 Keywords: Gold nanowires array Glucose Glucose oxidase Nanobiosensor Amperometry Cross-linking immobilization

a b s t r a c t A highly sensitive amperometric nanobiosensor has been developed by integration of glucose oxidase (GOx ) with a gold nanowires array (AuNWA) by cross-linking with a mixture of glutaraldehyde (GLA) and bovine serum albumin (BSA). An initial investigation of the morphology of the synthesized AuNWA by field emission scanning electron microscopy (FESEM) and field emission transmission electron microscopy (FETEM) revealed that the nanowires array was highly ordered with rough surface, and the electrochemical features of the AuNWA with/without modification were also investigated. The integrated AuNWA–BSA–GLA–GOx nanobiosensor with Nafion membrane gave a very high sensitivity of 298.2 ␮A cm−2 mM−1 for amperometric detection of glucose, while also achieving a low detection limit of 0.1 ␮M, and a wide linear range of 5–6000 ␮M. Furthermore, the nanobiosensor exhibited excellent anti-interference ability towards uric acid (UA) and ascorbic acid (AA) with the aid of Nafion membrane, and the results obtained for the analysis of human blood serum indicated that the device is capable of glucose detection in real samples. © 2013 Elsevier B.V. All rights reserved.

1. Introduction

∗ Corresponding author. Tel.: +61 3 9902 6450; fax: +61 3 9902 6738. E-mail addresses: [email protected] (S.B. Adeloju), [email protected] (Y. Wu). 1 Tel.: +86 551 62901012; fax: +86 551 62904517. 0003-2670/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.aca.2013.11.024

The considerable development of nanomaterials in recent years has provided several novel platforms for constructing unique electrochemical biosensors that are capable of achieving large active surface area due to their small sizes. Some of the recent applications of nanomaterials in this area include the use of various nanoparticles (gold, platinum, zinc oxide, etc), nanotubes (carbon, titanium

J. Cui et al. / Analytica Chimica Acta 809 (2014) 134–140

dioxide, etc), nanowires and nanorods. However, the majority of the efforts on fabrication of biosensors with nanomaterials have mainly focused on the use of nanoparticles and dispersed one-dimensional (1D) nanomaterials. For example, Yang et al. [1] fabricated a glucose biosensor by incorporating Fe3 O4 nanoparticles to chitosan film and achieved good performance for glucose detection due to the acceleration of the electron transfer by the Fe3 O4 nanoparticles. Also, TiO2 nanotubes co-decorated with carbon nanotubes (CNTs) and Pt nanoparticles were successfully utilized to construct amperometric glucose biosensors with improved performance by Pang and collaborators [2]. Besides these examples, various nanoparticles and one-dimensional nanomaterials have been employed to fabricate glucose biosensors [3–8]. In general, glucose biosensors based on one-dimensional nanomaterials, especially ordered one-dimensional nanomaterials, are known to give better performance than those based on the use of nanoparticles [9–15]. However, a noted limitation in this regard is that most of the glucose biosensors based on the use of one-dimensional nanomaterials have focused largely on different kinds of materials such as ZnO nanowires/nanorods array [16–19], TiO2 nanotubes array [2,20] and noble metal nanowires array [21–23], while largely ignoring other factors such as the morphology and size of 1D nanomaterials array affecting the performance of the glucose biosensors. Also, various approaches such as physical adsorption [16,22] and crosslinking [23] have been utilized for constructing 1D-based glucose biosensors and achieved good results. However, 1D nanostructures array has rarely been explored and utilized in fabricating glucose biosensors. It is worth noting that 1D nanomaterials array also plays a significant role in the improvement of performance of glucose biosensors. Among various 1D nanomaterials, gold nanowires have many advantages, such as large specific surface area, chemical inertness, biocompatibility, excellent electrical conductivity and good electrochemical activity towards H2 O2 [24–26], which makes this nanomaterial an ideal platform for fabricating nanobiosensors with high performance. Furthermore, it is now well recognized that gold is one the most stable metal at the nanoscale level and this factor may be critical for fabricating highly stable nanobiosensors [27]. In this paper, highly-ordered gold nanowires array with uniform size and morphology were fabricated via well-aligned anodic aluminium oxide (AAO) template (as shown in Supplementary Fig. 1) according to previously reported method [28]. Gold nanowires array with diameter of 80 nm was selected to fabricate glucose biosensors by taking the space between nanowires and its mechanical property into account. In addition, we demonstrated the significant benefits of using well-ordered gold nanowires array (AuNWA) to fabricate a highly sensitive amperometric nanobiosensor for glucose detection by integration with GOx via cross-linking with a mixture of glutaraldehyde (GLA) and bovine serum albumin (BSA). The ability of the AuNWA to enable improved electron transfer was investigated by cyclic voltam4−/3− metry with a Fe(CN)6 redox system. Also, the morphology and feature of the AuNWA was investigated by field emission scanning electron microscopy (FESEM) and field emission transmission electron microscopy (FETEM). Furthermore, the performance of the resulting AuNWA–BSA–GLA–GOx biosensor was studied by amperometric detection of glucose and the effect of common interferants, such as uric and ascorbic acids on the performance of AuNWA–BSA–GLA–GOx biosensor and real sample analysis were also assessed.

135

received. Gold chloride trihydrate (HAuCl4 ·3H2 O), Ethylenediaminetetraacetic acid (EDTA), glutaraldehyde (GLA), bovine serum albumin (BSA), d-(+)-glucose, Nafion (5% w/v) and other chemicals were also obtained from Sigma (Australia). All chemicals were of analytical grade and used as received unless stated otherwise. MilliQ water (18.2 M cm) was used to prepare all solutions throughout the experiments. Stock solutions of GLA 25% v/v and BSA 20% w/v were prepared and stored in the refrigerator at 4 ◦ C when not in use. Human blood serum was obtained from Jackson ImmunoResearch Laboratories, Inc. (USA) and used for the detection and recovery of glucose. 2.2. Fabrication of gold nanowires array and characterization The gold nanowires array was grown, according to previously reported approaches [23,29], from a solution which contained 10 g L−1 HAuCl4 , 5 g L−1 EDTA, 20 g L−1 K2 HPO4 and 160 g L−1 Na2 SO3 , on the surface of conventional gold electrode directly with the aid of AAO template, shown in Supplementary Fig. 1. The pore diameter and the thickness of AAO template were 80 nm and 16 ␮m, respectively. Unless otherwise stated, the electrodeposition was carried out with an applied current density of 1 mA cm−2 . Prior to electrodeposition, a thin gold film was sputtered onto one side of AAO template by vacuum sputter coater to act as working electrode, and then AAO template was fixed onto gold disk electrode via a Teflon cap, similar to the approach used by Wang et al. [23]. After electrodeposition, AAO template was etched away by utilizing 3 M NaOH to expose gold nanowires array, and then the array was rinsed thoroughly with Milli-Q water and ethanol to remove the residual NaOH. All of the electrochemical depositions were carried out on a galvanostat-potentiostat with a three electrode system at room temperature. Morphology and microstructure of AuNWA was characterized by field emission scanning electron microscopy (FESEM, Hitachi S 4800) and field emission transmission electron microscopy (FETEM, JEOL JEM-2100), respectively. 2.2.1. Immobilization of glucose oxidase and measurement Glucose oxidase (GOx ) was immobilized onto the surface of AuNWA by cross-linking immobilization with glutaraldehyde (GLA) and bovine serum albumin (BSA). Effect of GLA, BSA and GOx concentrations on the performance of glucose biosensors based on gold nanowires array were optimized, respectively (As shown in Supplementary Fig. 2(a–c)). Furthermore, other parameters such as drying time, deposition time, buffer solution concentration and pH were also investigated (Supplementary Fig. 3(a–d)). Finally, a 2 ␮L aliquot of the mixture, which contained 5.5% v/v GLA, 4.8% w/v BSA, 200 U mL−1 GOx , was placed onto the surface of the AuNWA with an area of 0.0095 cm−2 and dried for 30 min at room temperature. To enable the assessment of the benefit of the AuNWA, glucose biosensors based on the use of conventional Au disc electrode were also fabricated and tested for comparison. Also to minimise the effect of interferants such as UA and AA, a Nafion layer was formed over the glucose nanobiosensor by placing a drop of 1% Nafion solution on top and left to dry. Amperometric responses were obtained with a galvanostat-potentiostat and cyclic voltammograms (CV) were recorded with a Voltalab PGZ 301 electrochemical workstation. The enzyme electrode, Ag/AgCl (3 M KCl) and Pt/Ti wire were used as working, reference and counter electrodes, respectively. 3. Results and discussion

2. Experimental 3.1. Synthesis and morphology of AuNWA 2.1. Materials and reagents Glucose oxidase (EC 1.1.3.4 181300 units/g from Aspergillus niger) was purchased from sigma (Australia) and used as

Gold nanowires array (AuNWA) was synthesized with the aid of an anodic aluminium oxide (AAO) template in combination with direct electrodeposition. The diameter of the AAO templates was

136

J. Cui et al. / Analytica Chimica Acta 809 (2014) 134–140

Fig. 1. (a) Top-view FESEM image of AuNWA and (b) TEM image of AuNW.

successfully tuned from 40 nm to 100 nm. An AAO template with diameter of ca. 80 nm was subsequently chosen to fabricate the gold nanowires array. As shown in Fig. 1(a) FESEM images revealed that the AuNWs were highly ordered and parallel to each other and, thus, indicated that the AuNWA was perpendicular to the surface of the dissolved AAO templates and conventional gold disc electrode. The high magnification FESEM image of the AuNWA in the inset of Fig. 1(a) showed that the surface of the nanowires was very rough and this observation was further highlighted by the FETEM image in Fig. 1(b). The rough surface of the AuNWA was beneficial for achieving improved cross-linking immobilization of GOx and, hence, for improving the stability of the resulting nanobiosensor. It was worth noting that the diameter and length of the AuNWs were 80 and 300 nm, respectively. 3.2. Electrochemical characterization of AuNWA and AuNWA–BSA–GLA–GOx electrodes The electroactivity of the surface area of the AuNWA was inves4−/3− tigated by cyclic voltammetry with the Fe(CN)6 redox system by comparison between conventional gold electrode and AuNWA. Fig. 2(A) showed that a much more pronounced cyclic voltammet4−/3− ric behaviour (CV) was observed for the Fe(CN)6 redox system in presence of AuNWA. The extent of the magnitude of current increase caused by the incorporation of the AuNWA was determined by expression of the anodic or cathodic current (ipa or ipc ) in terms of the Randles–Sevcik equation [30]: ip = 0.4463nFAC

 nFvD 1/2 RT

(1)

where ip is the peak current of a redox couple, n is the number of electrons transfer in the redox reaction, F is the Faraday’s constant, A is the electrode area (cm2 ), C is the concentration of analyte (mol cm−3 ), R is the universal gas constant, v is the scan rate with working potential range (V s−1 ), D is the diffusion coefficient of the analyte in the bulk solution (cm−2 s−1 ), T is the absolute temperature (K). In this equation, all other parameters are constant values, except for A, and the electroactive surface area A is proportional to peak current ip . Evidently, the substantial increase in peak current ip observed for the CV obtained in the presence of AuNWA in Fig. 2(A) was indicative of an apparent increase in the electroactive surface area. Based on the current increase, the electroactive surface area of AuNWA was at least six times larger than that of the conventional Au disc electrode. The area of the gold disc electrode is 0.0095 cm−2 and, based on the SEM images of the gold nanowires array, the total estimated geometric surface area of the nanowire array is ∼0.123 cm−2 if we take nanowire dimensions and numbers

into account. From this estimate the geometric surface area of the nanowire array is ca. 13 times higher than that of the gold disc electrode. However, the effective electrochemically active surface area may be less. Therefore, it is possible that the ratio of the geometric surface area of the AuNWA to that of the gold disc electrode will be less than 13. The increased surface area of the AuNWA enabled much improved electron transfer. In addition, the shift in the peak potential in both directions with the increasing peak current was due to the large surface area of AuNWA. Fig. 2(B) shows the effect of scan rate on the electrochemical behaviour of the AuNWA–BSA–GLA–GOx electrode. Evidently, the peak current increased with increasing scan rates. Fig. 2(C) illustrates that the anodic and cathodic currents were directly proportional to the square root of the scan rate and, thus, indicated that a diffusion-controlled process was involved at the AuNWA–BSA–GLA–GOx electrode surface. Furthermore, the peakto-peak separation increased with increasing scan rates (as shown in Fig. 2(B)), suggesting that the electrode process was quasireversible [31]. The investigation of the interfacial region of the surfaces of the conventional Au and AuNWA electrodes with and without BSA–GLA–GOx layer (biofilm) by electrochemical impedance spectroscopy (EIS) was also used to discern the effect of the incorporation of AuNWA and biolayer on the electrode behaviour. The EIS results were shown as Nyquist plots in Fig. 2(D), based on 4−/3− the utilization of the Fe(CN)6 as the redox couple, modified Randles electrical circuit in Fig. 2(D) was employed to fit the EIS data. The equivalent electrical circuit contained electrolyte solution resistance (Rs ), charge transfer resistance (Rct ), Warburg impedance and constant phase element (CPE). We used CPE here to illustrate the double layer capacitance instead of the capacitance due to the rough surface of AuNWA observed in Fig. 1. As shown in Fig. 2(D), both the conventional Au and AuNWA electrodes with the BSA–GLA–GOx layer gave semicircle regions and linear portions. The linear portion represented the diffusion controlled process, while the semicircle region revealed the electron transfer limited process and the diameter of the semicircle was proportional to the charge transfer resistance (Rct ) on the electrode surface. Evidently, the diameter of the semicircle obtained with the AuNWA–BSA–GLA–GOx electrode was smaller than that of the conventional Au–BSA–GLA–GOx electrode and, thus confirmed that the incorporation of the 3D AuNWA on conventional gold electrode favoured the electron transfer. This special 3D architecture improved the conductivity of the glucose nanobiosensor and decreased the electron transfer resistance. It was also noted that there was no obvious semicircle in the high frequency region of the Nyquist plot obtained for the AuNWA without BSA–GLA–GOx

J. Cui et al. / Analytica Chimica Acta 809 (2014) 134–140

137

Fig. 2. Electrochemical behaviour of Au and AuNWA electrodes. (A) Cyclic voltammograms of (a) conventional Au disc electrode, and (b) AuNWA in 0.1 M KCl with 10 mM K3 [Fe (CN)6 ], scan rate: 100 mV s−1 ; (B) cyclic voltammograms of glucose biosensor in 0.05 M PBS containing 0.1 M KCl and 10 mM K3 [Fe (CN)6 ] at different scan rates: (a) 20 mV s−1 , (b) 40 mV s−1 , (c) 60 mV s−1 , (d) 80 mV s−1 , (e) 100 mV s−1 , (f) 120 mV s−1 ; (C) Relationship between peak current density and square root of the scan rate from Fig. 2 (B); (D) Nyquist plot of: (a) AuNWA–BSA–GLA–GOx electrode, (b) AuNWA electrode, (c)Au–BSA–GLA–GOx electrode in 0.05 M PBS (pH 7.4) in the presence of 10 mM K3 [Fe (CN)6 ] and 0.1 M KCl.

layer, indicating the involvement of only a diffusion controlled process. Evidently, the semicircle observed in the high frequency region of the plot for AuNWA–BSA–GLA–GOx electrode illustrates the effect of the barrier to electron transfer, and it was clearly demonstrated that BSA–GLA–GOx layer was formed on the surface of gold nanowires, indicating successful immobilization of GOx on the AuNWA.

path between GOx active sites and the surface of AuNWs enabled decreased response time. The detection principle of the resulting glucose nanobiosensors was based on the widely accepted reaction, involving the consumption of oxygen and production of H2 O2 , resulting from the oxidation of ␤-d-glucose to ␤-d-gluconic acid

3.3. Amperometric detection of glucose with AuNWA–BSA–GLA–GOx nanobiosensor Glutaraldehyde (GLA) and bovine serum albumin (BSA) are used as co-cross linking agent and spacer to immobilze GOx in a GLA-BSA layer and, hence, to permit uniform distribution of the BSA–GLA–GOx layer around AuNWs. Also, the lysine-rich BSA forms Schiff base (C=N) with GLA instead of GOx so that the structure of GOx is not destroyed and GOx activity is retained. Once glucose reaches active sites of GOx , it is oxidized to gluconic acid and H2 O2 . Subsequently, part of H2 O2 transfers to AuNWs surface, where it is oxidized to give amperometric current which is proportional to the glucose concentration. Upon addition, the BSA–GLA–GOx mixture readily wetted the 3D structure of the AuNWA and penetrated into the interspace between AuNWs to form a BSA–GLA–GOx layer, resulting in uniform distribution of GOx in BSA-GLA mixture layer around AuNWs, as evident in Fig. 3. This schematic diagram illustrated the typical model of glucose nanobiosensors based on 3D AuNWA, BSA–GLA mixture layer with GOx was readily formed between the interspaces (illustrated in Fig. 1(a)) around the AuNWs and on the surface of AuNWA. This reduced H2 O2 transfer

Fig. 3. Schematic illustration of the AuNWA–GLA–BSA–GOx glucose biosensor.

138

J. Cui et al. / Analytica Chimica Acta 809 (2014) 134–140

Fig. 4. (A) Cyclic voltammograms (CVs) obtained for Au–GLA–BSA–GOx and AuNWA–GLA–BSA–GOx biosensors in presence of 2 mM glucose in PBS; (B) CVs of AuNWA–GLA–BSA–GOx biosensor in presence of increasing glucose concentration.

and H2 O2 catalyzed by the immobilized GOx , which are clearly illustrated in the following equations: Glucose + GOx (FAD) → Gluconic acid + GOx (FADH2 )

(2)

GOx (FADH2 ) + O2 → GOx (FAD) + H2 O2

(3)

H2 O2 → 2H+ + O2 + 2e− (on surface of AuNWs)

(4)

As presented in Eqs. (2)–(4), the H2 O2 produced at the active sites of GOx migrated to the surface of the AuNWs to produce the amperometric response. Therefore, the amperometric measurement relied on detecting the electron transfer (current) resulting from the oxidation of H2 O2 , as illustrated in Fig. 3 and by Eq. (4). This detection process was particularly favoured by the high electrochemical activity of AuNWA towards H2 O2 . In this respect, the 3D architecture of AuNWA provided an excellent platform for immobilization of GOx with abundant active enzyme sites, resulting in an acceleration of H2 O2 transfer between the GOx active sites and the surface of AuNWs. This improvement was attributed to three distinct factors. First, the large active surface area increased the GOx loading in the vicinity of the surface of the AuNWs, reducing the H2 O2 transfer path between GOx active sites and the AuNWA surface and, thus, accelerating electron transfer. Second, the AuNWs with high catalytic property were sensitive to the H2 O2 produced from the catalytic decomposition of glucose. Third, compared to the smooth surface of the Au disc electrode, the unique 3D structure of AuNWA impeded the diffusion of H2 O2 from the surface of the AuNWs and, thus, enabled the retention of more H2 O2 , resulting in much improved sensitivity of the nanobiosensor to glucose. High catalytic performance towards glucose was further confirmed by comparing the cyclic voltammograms obtained for two glucose biosensors constructed with a conventional Au disc electrode and the AuNWA. Fig. 4(A) shows that, compared to the Au-BSA–GLA–GOx biosensor, the anodic current obtained with the AuNWA–BSA–GLA–GOx nanobiosensor increased significantly due to increased GOx loading with the 3D structure of AuNWA and the high electrochemical activity of AuNWA towards H2 O2 . Also, it was evident in Fig. 4(B) that the anodic current obtained with the AuNWA–BSA–GLA–GOx nanobiosensor increased with increasing glucose concentration. Based on the CV responses (Fig. 4(B)), an applied potential of 0.7 V was employed for the amperometric detection of glucose with the nanobiosensor. Fig. 5(A) shows typical amperometric responses obtained for glucose with the AuNWA–BSA–GLA–GOx nanobiosensor. As expected, the current increased with the successive addition of glucose and achieved a steady state in less than 5 s. The fast response was due to the 3D structure of the AuNWA, which reduced the

H2 O2 transfer path between GOx active sites and the AuNWA. Undoubtedly, this substantial improvement was due to the much improved ability of the AuNWA–BSA–GLA–GOx biosensor to enable much better H2 O2 transfer between GOx active sites and the surface of the AuNWs and eventually improved electron transfer, which was much better than that of the Au disc glucose biosensor. As uric and ascorbic acids are often present with glucose in human blood, a permselective Nafion (1 wt%) membrane was used to cover the surface of the glucose nanobiosensors to improve its selectivity. Although the sensitivity of the glucose nanobiosensor with the Nafion membrane (298.2 ␮A cm−2 mM−1 ) was lower than that without the membrane (379.0 ␮A cm−2 mM−1 ), it was still higher than those reported in Table 1 for other glucose nanobiosensors. The detection limit of the glucose nanobiosensor was calculated as dl = 3/m, where  was the standard deviation of background current, m was the sensitivity of the biosensors [32]. The achieved detection limits were 0.05 ␮M and 0.1 ␮M for the glucose nanobiosensors without and with Nafion membrane, respectively. These are better than those reported for other glucose biosensors based on one dimensional nanomaterials [23,33,34]. Also, as shown in Fig. 5(B), a linear concentration range of 5–5000 ␮M (R2 = 0.997) was achieved for the glucose nanobiosensor in the absence of the Nafion membrane, while a slightly wider linear range of 5–6000 ␮M (R2 = 0.998) was achieved in the presence of the membrane. The slight improvement in the achieved linear range for the latter may be due to the improved retention of GOx by the Nafion membrane.

Table 1 Comparison of performance with other electrochemical glucose biosensors based on one-dimentional nanomaterials array. Matrix

Sensitivity (␮A mM−1 cm−2 )

Detection limit (␮M)

AuNWA

298.2

0.1

5–6000

TiO2 NTA ZnO NTA ZnO NRA ZnO NWA ZnO NRA Au NWA Porous Au NWA Pt NWA ZnO NRA Prussian blue NWA

0.24 30.85 23.1 35.3 N/A N/A 15.6 110 106.60 17.6

5.7 10 10 1 3 5 46 0.5 0.1 1

6–1500 10–4200 10–3450 10–1600 5–300 10–10,000 50–2000 1–30,000 10–17,000 2–10,000

Linear range (␮M)

Refs. Present work [3] [16] [17] [18] [19] [22] [23] [29] [38] [39]

Note: NWA represents nanowires array, NRA represents nanorods array, and NTA means nanotubes array.

J. Cui et al. / Analytica Chimica Acta 809 (2014) 134–140

139

Fig. 5. (A) Amperometric response of AuNWA–BSA–GLA–GOx nanobiosensor with successive addition of (a) 50 ␮M, (b) 500 ␮M glucose in 0.05 M PBS. Top-left inset showed response of glucose nanobiosensors for addition of 50 ␮M glucose and bottom-right inset presented the response of glucose nanobiosensors with successive addition of 5 ␮M, (B) Typical calibration curve of AuNWA–BSA–GLA–GOx nanobiosensor without (a) and with (b) Nafion membrane, (C) Lineweaver–Burk plot.

Furthermore, the bioactivity of the immobilized GOx was investigated by calculating the apparent Michaelis–Menten constant Kapp m. According to Lineweaver–Burk equation [35]: 1 = iss



app

Km imax

  1 C

+

1 imax

(5)

where iss is the steady state current after addition of glucose, imax is the maximum current, and C is glucose concentration in measurement solution. From the Lineweaver–Burk plot shown in Fig. 5(C), the apparent Michaelis–Menten constant Kapp mwas calculated to be 6.5 mM, which was much smaller than previously reported values [23,36,37], indicating an excellent affinity between immobilized GOx in as-prepared nanobiosensor and the glucose in solution.

in Fig. 1. The effect of the surface roughness on the stability of the nanobiosensor was investigated by comparing the performance of BSA–GLA–GOx biosensors formed on AuNWA and on Au disc electrode. The response of the glucose biosensor based on Au disc electrode decreased to 86% after two weeks, and some of the GLA–BSA–GOx layers started to peel off gradually over four weeks due to the smooth surface of the gold disc electrode. In contrast, the 3D nanostructure and the rough surface of AuNWA enabled better anchoring of GOx in the BSA-GLA layer and prevented the leakage of the enzyme. Even after four weeks, it was difficult to peel off the GLA–BSA–GOx layer from the 3D nanostructure. All of the glucose nanobiosensors were kept in 0.05 M PBS at 4 ◦ C when not in use to ensure adequate stability.

3.4. Interference and stability study Fig. 6 shows the amperometric responses of AuNWA–BSA–GLA–GOx nanobiosensor with Nafion (1% w/v) membrane with successive addition of 100 ␮M glucose, 100 ␮M uric acid (UA), 100 ␮M ascorbic acid (AA) and 100 ␮M glucose. It revealed that the high response achieved for amperometric detection of glucose with the AuNWA–BSA–GLA–GOx nanobiosensor with the Nafion film was not affected by the presence of UA and AA coexisting with glucose in blood. The influence of these substances at the levels usually present in blood was negligible. The reason for this phenomenon is the ability of the Nafion membrane to excluded anions and, thus, preventing the majority of the anionic interferences of UA and AA from reaching the surface of the AuNWs. Furthermore, 92% of the original amperometric response of the AuNWA–BSA–GLA–GOx nanobiosensor with Nafion membrane was maintained for over four weeks, indicating an excellent stability, due mainly to Nafion membrane, the 3D nanostructure and the rough surface of the AuNWs, as identified

Fig. 6. Effect of uric and ascorbic acids on the performance of the AuNWA–BSA–GLA–GOx nanobiosensor. Successive addition of 100 ␮M glucose, 100 ␮M uric acid, 100 ␮M ascorbic acid and 100 ␮M glucose.

140

J. Cui et al. / Analytica Chimica Acta 809 (2014) 134–140

Table 2 Analysis of glucose in human blood serum with the AuNWA–GLA–BSA–GOx biosensor with Nafion membrane. Samples

Added (␮M)

Found (␮M)a

R.S.D (%)

Recovery (%)

1 2 3 4 5

0 50.0 500.0 1000.0 2000.0 Average

530.5 582.2 1016.5 1546.8 2594.1 –

3.2 2.9 2.7 2.3 3.1 2.8

– 103.4 97.2 101.6 103.2 101.4

a

acknowledges the support provided by Monash University for this research. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.aca.2013.11.024. References

n = 6.

3.5. Application to real samples The suitability of the proposed nanobiosensor for the detection of glucose in diluted human blood serum was investigated by using an established calibration curve b in Fig. 5(B). The recovery efficiency of the use of the nanobiosensor with a standard calibration curve b for quantification of glucose was investigated with four standard glucose spikes between 50 and 2000 ␮M. The results presented in Table 2 reveal that relatively high recovery of the glucose spikes was achieved between 97.2% and 103.4%, with an average recovery of 101.4 ± 2.9%. Thus, confirming the excellent capability of the glucose nanobiosensors for the detection of glucose in human blood serum.

[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]

4. Conclusions Highly ordered AuNWA was successfully synthesized with AAO template in combination with direct electrodeposition, and used in conjunction with GLA and BSA to fabricate a highly sensitive and selective AuNWA–BSA–GLA–GOx nanobiosensor with an extremely high H2 O2 transfer rates. AuNWA–BSA–GLA–GOx nanobiosensor with Nafion membrane achieved a high sensitivity of 298.2 ␮A cm−2 mM−1 , a detection limit of 0.1 ␮M and a linear concentration of 5–6000 ␮M. It was obvious from these data that the AuNWA maintained an abundant active enzyme sites which accelerated the rate of H2 O2 transfer between GOx active sites and AuNWA surfaces, accelerating the electron transfer between H2 O2 and AuNWs. The very low detection limit achieved for the amperometric detection of glucose by the AuNWA–BSA–GLA–GOx nanobiosensor opens up future possibility for its detection in nonblood clinical samples, such as saliva and sweat, in diabetic patients. Further extension of this approach to other enzymes and nanowire materials for fabrication of new generation ultra-high sensitive nanobiosensors is possible and some of these are currently being investigated in our laboratories for detection of other substrates. Acknowledgments One of the authors (J.W. Cui) thanks China Scholarship Council for providing the financial support for his visit to Australia and

[16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39]

L. Yang, X. Ren, F. Tang, L. Zhang, Biosens. Bioelectron. 25 (2009) 889–895. X. Pang, D. He, S. Luo, Q. Cai, Sens. Actuators, B 137 (2009) 134–138. J. Wang, M. Musameh, Y. Lin, J. Am. Chem. Soc. 125 (2003) 2408–2409. S.H. Lim, J. Wei, J. Lin, Q. Li, J.K. You, Biosens. Bioelectron. 20 (2005) 2341–2346. J. Lu, I. Do, L.T. Drzal, R.M. Worden, I. Lee, ACS Nano 2 (2008) 1825–1832. Z. Wen, S. Ci, J. Li, J. Phys. Chem. C 113 (2009) 13482–13487. J.C. Claussen, M.M. Wickner, T.S. Fisher, D.M. Porterfield, ACS Appl. Mater. Interfaces 3 (2011) 1765–1770. B.G. Choi, J. Im, H.S. Kim, H.S. Park, Electrochim. Acta 56 (2011) 9721–9726. B.Y. Wu, S.H. Hou, F. Yin, J. Li, Z.X. Zhao, J.D. Huang, Q. Chen, Biosens. Bioelectron. 22 (2007) 838–844. Y.G. Liu, X.M. Feng, J.M. Shen, J.J. Zhu, W.H. Hou, J. Phys. Chem. B 112 (2008) 9237–9242. T. Kong, Y. Chen, Y. Ye, K. Zhang, Z. Wang, X. Wang, Sens. Actuators, B 138 (2009) 344–350. Y. Zhou, S. Yang, Q. Qian, X. Xia, Electrochem. Commun. 11 (2009) 216–219. Z. Wang, S. Liu, P. Wu, C. Cai, Anal. Chem. 81 (2009) 1638–1645. W. Yang, K.R. Ratinac, S.P. Ringer, P. Thordarson, J.J. Gooding, F. Braet, Angew. Chem. Int. Ed. 49 (2010) 2114–2138. K. Yang, G. She, H. Wang, X. Ou, X. Zhang, C. Lee, S. Lee, J. Phys. Chem. C 113 (2009) 20169–20172. A. Wei, X.W. Sun, J.X. Wang, Y. Lei, X.P. Cai, C.M. Li, Z.L. Dong, W. Huang, Appl. Phys. Lett. 89 (2006) 123902 (1-3). J. Liu, C. Guo, C.M. Li, Y. Li, Q. Chi, X. Huang, L. Liao, T. Yu, Electrochem. Commun. 11 (2009) 202–205. X. Liu, Q. Hu, Q. Wu, W. Zhang, Z. Fang, Q. Xie, Colloid Surf., B 74 (2009) 154–158. D. Pradhan, F. Niroui, K.T. Leung, ACS Appl. Mater. Interfaces 2 (2010) 2409–2412. Y. Liu, Y. Zhu, Y. Zeng, F. Xu, Nanoscale Res. Lett. 4 (2009) 210–215. P. Benvenuto, A.K.M. Kafi, A. Chen, J. Electroanal. Chem. 627 (2009) 76–81. X. Zhang, D. Li, L. Bourgeois, H. Wang, P.A. Webley, Chem. Phys. Chem. 10 (2009) 436–441. Y. Wang, Y. Zhu, J. Chen, Y. Zeng, Nanoscale 4 (2012) 6025–6031. M. Yang, F. Qu, Y. Li, Y. He, G. Shen, R. Yu, Biosens. Bioelectron. 23 (2007) (2007) 414–420. Y. Lu, M. Yang, F. Qu, G. Shen, R. Yu, Bioelectrochemistry 71 (2007) 211–216. S. Guo, D. Wen, S. Dong, E. Wang, Talanta 77 (2009) 1510–1517. N.V. Skorodumova, S.I. Simak, Phys. Rev. B: Condens. Matter 67 (2003) 121401 (1-4). H. Wang, C. Zou, B. Yang, H. Lu, C. Tian, H. Yang, M. Li, C. Liu, D. Fu, J. Liu, Electrochem. Commun. 11 (2009) 2019–2022. M. Yang, F. Qu, Y. Lu, Y. He, G. Shen, R. Yu, Biomaterials 27 (2006) 5944–5950. J.E.B. Randles, Trans. Faraday Soc. 44 (1948) 327–338. R.R. Moore, C.E. Banks, R.G. Compton, Anal. Chem. 76 (2004) 2677–2682. G.L. Long, J.D. Winefordner, Anal. Chem. 55 (1983) 712A–724A. S. Su, Y. He, S. Song, D. Li, L. Wang, C. Fan, S. Lee, Nanoscale 2 (2010) 1704–1707. M.F E., S.K. Arya, S. Bhansali, Analyst 136 (2011) 1686–1689. H. Lineweaver, D. Burk, J. Am. Chem. Soc. 56 (1934) 658–666. Z. Zhao, X. Chen, B. Tay, J. Chen, Z. Han, K. Khor, Biosens. Bioelectron. 23 (2007) 135–139. D. Patil, N.Q. Dung, H. Jung, S.Y. Ahn, D.M. Jang, D. Kim, Biosens. Bioelectron. 31 (2012) 176–181. R. Ahmad, N. Tripathy, J.H. Kim, Y. Hahn, Sens. Actuators, B 174 (2012) 195–201. F. Qu, A. Shi, M. Yang, J. Jiang, G. Shen, R. Yu, Anal. Chim. Acta 605 (2007) 28–33.

Integration of a highly ordered gold nanowires array with glucose oxidase for ultra-sensitive glucose detection.

A highly sensitive amperometric nanobiosensor has been developed by integration of glucose oxidase (GO(x)) with a gold nanowires array (AuNWA) by cros...
2MB Sizes 0 Downloads 0 Views