Direct electrochemistry of glucose oxidase immobilized on nanostructured gold thin films and its application to bioelectrochemical glucose sensor

Article abstract of DOI:10.1016/j.electacta.2012.02.011

Glucose oxidase (GOx) was stably immobilized via a simple physical adsorption method onto the nanostructured Au thin films fabricated by using electrodeposition and galvanic replacement technology, which provides a facile method to prepare morphology-controllable Au films and also facilitates the preparation and application of enzyme modified electrodes. An obvious advantage of the as-prepared enzyme electrode (denoted as GOx/Au/GCE) is that the nano-Au films provide a favorable microenvironment for GOx and facilitate the electron transfer between the active center of GOx and electrodes. Cyclic voltammetry (CV) results indicate that the immobilized GOx displayed a direct, reversible and surface-confined redox reaction in the phosphate buffer solution. Furthermore, the enzyme modified electrode was used as a glucose bioelectrochemical sensor, exhibiting a linear relationship in the concentration ranges of 2.5-32.5 μmol L^-1 and 60-130 μmol L^-1 with a detection limit of 0.32 μmol L^-1 (S/N = 3) at an applied potential of -0.55 V. Due to the excellent stability, sensitivity and anti-interference ability, the Au thin films are hopeful in the construction of glucose biosensors.

Full text of DOI:10.1016/j.electacta.2012.02.011

Electrochimica Acta 67 (2012) 140–146
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Direct electrochemistry of glucose oxidase immobilized on nanostructured gold
thin films and its application to bioelectrochemical glucose sensor
Cuicui Qiua, Xia Wangb, Xueying Liub, Shifeng Houc, Houyi Maa,∗
a Key Laboratory for Colloid and Interface Chemistry of State Education Ministry, School of Chemistry and Chemical Engineering, Shandong University, Jinan 250100, China
b State Key Laboratory of Microbial Technology, Shandong University, Jinan 250100, China
c Department of Chemistry and Biochemistry, Montclair State University, Montclair, NJ 07043, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Glucose oxidase (GOx) was stably immobilized via a simple physical adsorption method onto the nanos-
tructured Au thin films fabricated by using electrodeposition and galvanic replacement technology, which
provides a facile method to prepare morphology-controllable Au films and also facilitates the prepara-
tion and application of enzyme modified electrodes. An obvious advantage of the as-prepared enzyme
electrode (denoted as GOx/Au/GCE) is that the nano-Au films provide a favorable microenvironment for
GOx and facilitate the electron transfer between the active center of GOx and electrodes. Cyclic voltam-
metry (CV) results indicate that the immobilized GOx displayed a direct, reversible and surface-confined
redox reaction in the phosphate buffer solution. Furthermore, the enzyme modified electrode was used
as a glucose bioelectrochemical sensor, exhibiting a linear relationship in the concentration ranges of
2.5–32.5 ␮mol L−1 and 60–130 ␮mol L−1 with a detection limit of 0.32 ␮mol L−1 (S/N = 3) at an applied
potential of −0.55 V. Due to the excellent stability, sensitivity and anti-interference ability, the Au thin
films are hopeful in the construction of glucose biosensors.
Received 9 November 2011
Received in revised form 30 January 2012
Accepted 5 February 2012
Available online 13 February 2012
Keywords:
Glucose oxidase (GOx)
Electron transfer
Bioelectrochemical sensor
Gold film
Electrochemical detection
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
and electrodes. Some technological problems, such as denaturation
and leaching of immobilized enzyme, and especially long-distance
With its intrinsic advantage in high sensitivity and selectivity,
fast response, easy operation, and continuous on-line detection,
bioelectrochemical sensors have received increasing attention in
clinical medicine, biochemistry and environmental monitoring.
Because of the high catalytic activity for the oxidation of glucose to
gluconolactone, glucose oxidase (GOx) has been extensively used in
various fields including food and beverage additives, biofuel cells,
textile industry, etc. [1,2]. In particular, glucose biosensors, widely
used in the direct determination of glucose in biological systems
such as blood [3] and brain [4], are becoming an indispensable
means for the diagnosis and therapy of diabetes mellitus.
In view of excellent selectivity and simple devices, much atten-
tion has been focused on the construction of the third generation
sensor. One of the key technical problems in designing this kind of
electrochemical sensors is how to achieve the fast direct electron
transfer between the electrode surface and the redox active sites of
the enzyme [5]. At the present time, various nanomaterials (e.g. Au
[6], Pt [7], ZnO [8,9] and various carbon materials [10–14]) are used
to investigate the direct electron transfer between redox enzymes
electron transfer between the active sites of enzyme and the sup-
port electrodes [15–17], need to be solved urgently in developing
the third generation sensor. Nowadays, the use of nanomaterials
has significantly improved the performance of the biosensors and
broadened their application range, which in turn greatly accelerates
the development of new microminiaturized sensors. The metal-
based nano-materials like Au [18,19], Pt [20,21] and Pd [22] have
been widely used in building biological functional interfaces. It
should be noted that, the excellent biocompatibility and interface-
dominated properties of gold nanostructures have stimulated an
increasing interest in designing novel and high-performance elec-
trochemical biosensing devices [23].
In order to prepare morphology-controllable Au thin films, the
nano-structured Co thin films deposited by cyclic voltammetry (CV)
were used as sacrificial templates. The advantage of the present
fabrication method is that the as-prepared nano-Au films were
directly deposited onto the glassy carbon electrode (GCE) surface,
which provides a better and more convenient boundary for enzyme
immobilization than Au nanoparticles. CV and electrochemical
impedance spectroscopy (EIS) were used to study the direct redox
reactions of the immobilized GOx and its inhibitory effect on non-
enzyme-mediated redox reactions. Moreover, the as-prepared GOx
modified electrodes displayed excellent selectivity, high catalytic
∗
Corresponding author. Tel.: +86 531 88364959; fax: +86 531 88564464.
E-mail address: hyma@sdu.edu.cn (H. Ma).
0013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2012.02.011

C. Qiu et al. / Electrochimica Acta 67 (2012) 140–146
141
activity and good stability for the glucose oxidation. The obtained
results are helpful for the development of high-performance glu-
cose biosensors.
deaerated by N2 bubbling for 10 min prior to electrochemical mea-
surements and a blanket of N2 was maintained throughout each
experiment.
EIS was used to study the redox behavior of the p-
benzoquinone/hydroquinone (BQ/HQ) couple at Au/GCE and
GOx/Au/GCE electrodes. EIS measurements were carried out using
EG&G potentiostat/galvanostat (Model PARC 273) and the data
acquisition was performed with EG&G PowersineTM. All impedance
measurements were performed at 0 V (vs. open circuit potentials)
in the frequency range from 10 kHz to 20 mHz using a 5 mV alter-
nating voltage superimposed on DC potentials.
2. Experimental
2.1. Reagents and materials
Glucose oxidase (GOx, EC 1.1.3.4, 250 KU g−1) was purchased
from Sigma–Aldrich Co. The chemical reagents were analytical
grade and the aqueous solutions were prepared with ultrapure
water (>18 Mꢀ cm). The phosphate buffer solution (PBS, pH 6.8)
was prepared by mixing NaH2PO4 and Na2HPO4 solutions. The glu-
cose solution was left at 4 ◦C overnight to allow the equilibration
of the optical isomerism.
3. Results and discussion
3.1. Preparation and characterization of Au thin films
2.2. Preparation of Au film modified electrode
The nanostructured Co thin film was directly deposited onto a
clean GCE with continuous potential sweep, and its surface mor-
phology was shown in the inset of Fig. 1(a). The EDS spectrum
shown in Fig. 1(a) gave a direct evidence for the existence of Co on
the GCE substrate. The formation of Co thin films has been described
in detail in the previous reports [24,25]. The nano/submicron struc-
tured Au thin films can be fabricated on the GCE substrate with ease
by means of galvanic replacement of the Co films by Au(III) ions. As
can be seen in Fig. 1(b), a number of Au “islands” with sizes ranging
from nanometer to submicron seem to float on a loose and porous
bottom layer constructed by a large quantity of Au nanoparticles
(Fig. 1(c)). The high conductive, porous and biocompatible microen-
vironment of Au films are very suitable for enzyme immobilization,
which acts as a bridge to link the active site of enzyme and the
underlying electrode, not only favoring the direct electron trans-
fer at the interface but also enhancing the stability of the enzyme
modified electrode. EDS analysis of the Au films (Fig. 1(d)) only
gives the signals of Au element, proving that all Co particles of
different sizes were completely etched away during the galvanic
replacement reaction.
Glassy carbon electrodes (GCEs) were ground with emery papers
of decreasing particle sizes to no. 3500 finish at first, and then pol-
ished with 0.05 ␮m alumina slurry. Before use, the electrodes were
cleaned ultrasonically in HNO3 (1:1), washed with ultrapure water,
and finally rinsed with absolute ethanol.
The Co film modified GCE (Co/GCE) was prepared following the
previous procedure [24]. In a typical case, the nanostructured Co
thin films were directly formed on a clean GCE (˚ = 5 mm) with con-
tinuous potential scanning at a rate of 50 mV s−1 between −0.95 and
−0.60 V (vs. SCE) for 20 cycles in a mixture solution of 0.1 mol L−1
Na2SO4 + 10 mmol L−1 Co(CH3COO)2. The GCE covered with Co thin
films was then immersed into 25 mmol L−1 HAuCl4 solution for
30 min, allowing complete replacement of Co by Au. After thor-
oughly rinsing with ultrapure water, the Au film modified GCE
(Au/GCE) was immersed in 0.1 mol L−1 PBS for carrying out the con-
tinuous potential cycling in the designated potential range until
it gave steady CV curves. The current density in this paper was
normalized by the apparent area of the electrode.
2.3. GOx immobilization
3.2. Electrochemical behavior of the GOx modified electrode
The GOx solution (2000 U mL−1) was prepared with ultrapure
water and stored in refrigerator at 4 ◦C. The freshly prepared
Au/GCE was rinsed and then immersed immediately in 2000,
500, 300 U mL−1 GOx solutions, followed by CV cycles in PBS.
After 24 h immobilization, the GOx modified Au tin film electrode
(GOx/Au/GCE) was ready, then rinsed and stored in refrigerator at
4 ◦C for use.
3.2.1. Inhibitory effect of the immobilized GOx on non-enzymatic
redox reactions
The effect of enzyme immobilization on the electron transfer
in non-enzymatic redox reactions was investigated by means of
CV and EIS methods. Fig. 2(a) shows the cyclic voltammograms
(CVs) for Au/GCE and GOx/Au/GCE in the presence of BQ. The great
difference between the two CV curves indicates that the GOx immo-
bilization on the Au thin films has a strong inhibitory influence on
the redox behavior of the BQ/HQ couple. Theoretically, the redox
reaction of the BQ/HQ couple is a non-enzymatic redox reaction
and can only occur at uncovered sites on the enzyme-modified
electrode. The less the potential difference between oxidation and
reduction peaks (ꢁEp = Epa − Epc), the better the reversibility of
redox reaction. It is observed from Fig. 2(a) that ꢁEp of the unmodi-
fied and modified electrodes were 64 mV and 320 mV, respectively.
The greatly diminished current response and obviously enlarged
ꢁEp after the GOx immobilization indicated that the rate of electron
transfer was limited at the electrode/electrolyte interface, which
resulted from the retarding of interfacial electron transfer kinetics
[26,27]. In other words, the immobilized GOx behaves as a barrier
layer to block the electron transfer involved in the redox process of
the BQ/HQ couple.
2.4. Apparatus and measurements
Surface morphologies of metal thin films were observed by field
emission scanning electron microscope (FESEM, Hitachi S-4800),
and EDS analyses were performed with an INCAX-sight energy dis-
persive X-ray spectrometer equipped on the FESEM.
All electrochemical measurements were performed in a con-
ventional three-electrode cell at room temperature (∼15 ◦C). The
modified GCE was selected as the working electrode. A bright Pt
plate (1.0 cm × 1.0 cm) and a saturated calomel electrode (SCE)
served as the counter electrode and the reference electrode, respec-
tively. In this paper, all the potentials were referred to the SCE.
The reference electrode was led to the surface of the working
electrode through a Luggin capillary. CV, differential pulse voltam-
metry (DPV) and amperometry were conducted using a CHI 650A
electrochemical workstation. The amperometric response of the
GOx/Au/GCE for successive addition of glucose solution to the
stirred PBS was recorded at −0.55 V. The electrolyte solutions were
Considering that EIS is an effective tool for probing the interfa-
cial charge transfer on surface modified electrodes [28], the effect
of enzyme concentrations on electron transfer at the enzyme mod-
ified electrodes was also investigated by means of EIS method.

142
C. Qiu et al. / Electrochimica Acta 67 (2012) 140–146
Fig. 1. EDS of Co thin film (a) and Au film (d), and SEM images of Co thin film (inset in (a)) and Au films (b, c).
Fig. 2(b) shows a set of Nyquist impedance spectra obtained with
Au/GCE and GOx/Au/GCE electrodes. For the bare Au/GCE, the corre-
sponding Nyquist spectrum displays a small depressed capacitive
loop in the high frequency range (Rct = 586 ꢀ cm2) followed by a
straight line in the low frequency region. The capacitive loop is
associated with the electron transfer at the electrode/electrolyte
interface, whereas the straight line in the low frequency is due
to the diffusion of BQ (or HQ) from the bulk solution to the
electrode surface. The diameter of the capacitive loop gradually
increased but the straight line in the low frequency gradually
became insignificant and finally disappeared with increasing the
enzyme concentration.
the barrier layers were so dense that the diffusion process of
electrochemical active species was completely blocked, which is
exemplified by the disappearance of the Warburg impedance.
3.2.2. Direct electrochemistry of GOx/Au/GCE
As mentioned above, the immobilization of GOx will retard the
electron transfer of non-enzyme mediated redox reactions at the
electrode/solution interface; however, the direct electrochemical
behavior is one of the most important features of enzyme elec-
trodes. The cyclic voltammograms CVs of Au/GCE, GCE modified
with GOx (GOx/GCE) and GOx/Au/GCE in 0.1 mol L−1 deaerated
PBS are shown in Fig. 3(a). No obvious redox peak is observed
from the CVs of the Au/GCE and only small redox peaks are
found from that of the GOx/GCE. In contrast, a pair of well-
defined and quasi-reversible redox peaks appear in the CVs of
the GOx/Au/GCE, which should be attributed to the direct elec-
tron transfer involved in the redox process of GOx coenzyme
FAD/FADH2 (GOx-FAD + 2e− + 2H+ ↔ GOx-FADH2) [11,32]. The CVs
of the GOx/Au/GCE exhibit an anodic peak at −0.434 V and a
cathodic peak at −0.454 V with a small peak potential separation
(ꢁEp = 20 mV) at scan rate of 50 mV s−1. It is obvious that the as-
prepared Au thin films provide an excellent microenvironment for
GOx and significantly enhance the rate of electron transfer. It is
determined that the formal potential (defined as (Epa + Epc)/2) is
about −0.444 V, which is close to the standard electrode potential
of FAD/FADH2 at pH 7.0 (25 ◦C) [33].
The impedance spectra were fitted with the equivalent cir-
cuit shown in Table 1, and the values of circuit elements were
determined and also listed in Table 1. Herein we only focus on
the changes of the high frequency capacitive loops. The capaci-
tive loop is usually related to the relaxation time constant of the
charge-transfer resistance (Rct) and the double-layer capacitance
(CPEdl) at the electrode/solution interface. The smaller the Rct, the
faster the rate of an electrochemical reaction [29–31]. Based on the
impedance spectra measured under different conditions, there is
no doubt that the immobilization of GOx on the Au thin films made
the electron transfer become slow. The higher the concentrations
of GOx, the more the GOx immobilized at the Au/GCE, which in
turn retards more effectively the redox reaction of the BQ/HQ cou-
ple. When the concentrations of GOx increased to 2000 U mL−1
,

C. Qiu et al. / Electrochimica Acta 67 (2012) 140–146
143
Table 1
Equivalent circuita used to fit Nyquist spectra shown in Fig. 2b and the obtained values of equivalent circuit elements.
Rs (ꢀ cm2)
Rct (ꢀ cm2)
CPEdl (ꢀ−1 cm−2 sn)
CPEW (ꢀ−1 cm−2 sn)
0 U mL−1
13.6
9.29
8.06
13.1
5.86 × 102
1.07 × 103
1.50 × 103
2.33 × 103
5.37 × 10−5(n = 0.90)
1.17 × 10−4(n = 0.85)
8.37 × 10−5(n = 0.88)
2.02 × 10−4(n = 0.74)
1.00 × 10−2(n = 0.45)
300 U mL−1
500 U mL−1
2000 U mL−1
–
–
–
a
In the equivalent circuit, Rs, Rct, CPEdl, CPEW are the solution resistance, the charge transfer resistance, the double-layer capacitance and Warburg impedance, respectively.
value is larger than the literature value (1.27 × 10−10 mol cm−2
)
Fig. 3(b) shows a group of CVs for the GOx/Au/GCE at different
scan rates in 0.1 mol L−1 deaerated PBS. It is clear that the redox
peak currents are dependent on the scan rate. Moreover, there is a
linear relationship between the peak current and the scan rate from
5 mV s−1 to 1000 mV s−1. Besides, the ratio of the cathodic peak cur-
rent and the anodic one (Ipc/Ipa) approximates to 1 and the formal
potential keeps almost unchanged. These typical characteristics
confirm that the redox reaction taking place on the GOx/Au/GCE
is a surface-controlled reaction process [9–11].
[35], thereby suggesting that the Au films provide a large area for
the enzyme immobilization.
3.3. Electrochemical detection of glucose
3.3.1. Electrochemical behavior of GOx/Au/GCE in the presence of
glucose
The CVs in Fig. 4(a) show the effect of glucose concentrations
on the electrochemical behavior of GOx/Au/GCE. As the glucose
concentration increased, the redox peak current of FAD/FADH2
was observed to decrease obviously, which has been reported
previously [36,37]. In order to have a better understanding of
the electrochemical behavior of the GOx/Au/GCE in the presence
of glucose, the voltammetric behavior of three different elec-
trodes, Au/GCE, GOx/GCE and Au/GCE modified with inactivated
GOx (inactivated-GOx/Au/GCE), were investigated under identi-
cal conditions and the CVs recorded were shown in Fig. 4(b) for
According to Laviron’s study [34], the adsorbance of GOx on the
Au thin films can also be investigated by voltammetric method. The
peak current density can be expressed as follows:
n2F2ꢂAꢃ ∗
nFQꢂ
Ip
=
=
(1)
4RT
4RT
where n is electron transfer number, ꢂ is scan rate, A is surface
area of the GCE, ꢃ * is the average surface concentration of GOx,
Q is integral charge of the cathodic peak, T, R and F are tempera-
ture, gas constant and Faraday constant, respectively. On the basis
of Eq. (1), n is determined to be 1.44, implying that two electrons
are involved the redox reaction. The average surface concentration
of the electroactive GOx adsorbed on the Au/GCE is calculated to
be 2.58 × 10−10 mol cm−2 from the slope of the Ip–ꢂ curve. This
Fig. 2. (a) CVs of the Au/GCE and GOx/Au/GCE in 0.1 mol L−1 PBS +5 mmol L−1
BQ mixed solutions. Scan rate: 50 mV s−1. GOx concentration: 2000 U mL−1. (b)
Nyquist impedance spectra for GOx/Au/GCE with different concentrations of GOx in
0.1 mol L−1 PBS + 5 mmol L−1 BQ mixed solutions at 0 V with respect to open circuit
potentials.
Fig. 3. (a) CVs of Au/GCE, GOx/GCE and GOx/Au/GCE in 0.1 mol L−1 PBS (scan rate:
50 mV s−1); (b) CVs of GOx/Au/GCE measured at scan rates of 5, 10, 20, 50, 75,
100, 150, 200, 250, 300, 350, 400, 450, 500, 550, 600, 700, 800, 900, 1000 mV s−1
in 0.1 mol L−1 PBS.

144
C. Qiu et al. / Electrochimica Acta 67 (2012) 140–146
reduction peak almost disappeared during the negative-going
potential scan (see Fig. 4(a)). Furthermore, the electrooxidation of
glucose on Au sites was also inhibited on the GOx/Au/GCE. Fig. 4(b)
has clearly shown that no obvious electrooxidation of glucose took
place on both the GOx/GCE and inactivated-GOx/Au/GCE. Never-
theless, the redox peaks of FAD/FADH2 decreased for GOx/GCE
(compared with the CV curve in Fig. 3(a) in the absence of glucose)
and disappeared for inactivated-GOx/Au/GCE (compared with the
CV curve of GOx/Au/GCE).
The decrease of redox peak current of FAD/FADH2 was also
demonstrated by the results of DPV (shown in Fig. 4(c)). Each DPV
curve displays three well-defined oxidation peaks in the potential
range between −0.8 and 0.8 V. The peaks at ∼−0.15 V and ∼0.66 V
are derived from the redox reaction of glassy carbon and Au, respec-
tively. The peak at ∼−0.6 V is attributed to the redox process of
FAD/FADH2, which significantly decreased with the increase of glu-
cose concentration. The results indicate that the addition of glucose
produces a significant influence on the direct electron transfer of
GOx, which in turn provides a feasible path to detect the changes
of glucose concentration.
3.4. Amperometric detection of glucose
It is well-known that GOx possesses a high-efficiency catalytic
activity toward glucose oxidation. The reaction between glucose
and GOx in the presence of oxygen can be described by Eqs. (2) and
(3) [38,39]:
GOxox + ␤-d(+)-glucose → GOxred + d-gluconicacid
GOxred + O2 → GOxox + H2O2
(2)
(3)
where the dissolved molecular oxygen acts as an electron acceptor.
Fig. 5(a) shows a typical current-time curve of the glucose
biosensor based on the GOx/Au/GCE at an applied constant poten-
tial of −0.55 V. A well-defined and fast amperometric response
was observed at the potential. The detection range of glucose was
from 2.5 ␮mol L−1 to 157.5 ␮mol L−1. The inset shown in the upper
right of Fig. 5(a) presents calibration curves for steady state current
versus glucose concentrations. Two linear ranges were obtained
and the fitting regression equations were given as follows:
I = −2.11 − 0.074C(2.5 − 32.5 ␮mol L−1, R = 0.998, n = 14) (4)
I = −4.44 − 0.011C(60 − 130 ␮mol L−1, R = 0.997, n = 29)
(5)
The limit of detection is 0.32 ␮mol L−1 (S/N = 3). When the con-
centration of glucose further increased to more than 150 ␮mol L−1
,
the current response tended to become a plateau, which is the
typical characteristic of the Michaelis–Menten kinetics. The appar-
ent Michaelis–Menten constant (km), which gives an indication of
the enzyme-substrate kinetics for the biosensor, can be calculated
using the Lineweaver–Burk equation [40]:
Fig. 4. (a) CVs of GOx/Au/GCE in 0.1 mol L−1 PBS with different concentrations
of glucose (scan rate: 50 mV s−1); (b) CVs of Au/GCE, GOx/GCE and inactivated-
GOx/Au/GCE in mixed solutions of 0.1 mol L−1 PBS and 1 mmol L−1 glucose at a scan
rate of 50 mV s−1; (c) difference pulse voltammograms of GOx/Au/GCE in 0.1 mol L−1
PBS with different concentrations of glucose.
1
km
1
1
=
+
(6)
is
imax
C
imax
where is is the steady-state current, C is the concentration of glu-
cose, km is the apparent Michaelis–Menten constant, and imax is
the maximum current. From the curve of the is−1 vs C−1, the km
is estimated to be 15.6 ␮mol L−1 and the imax is 6.5 ␮A. In order
to identify the role of GOx in the determination of glucose, the
amperometric responses of Au/GCE, GOx/GCE and inactivated-
GOx/Au/GCE caused by successive addition of 2.5 ␮mol L−1 glucose
to the stirred 0.1 mol L−1 PBS are shown in Fig. 5(b). It was
observed that, though a current step was caused by the addition
of glucose, no steady current produced and no linear relation-
ship between glucose concentration and current response were
obtained on Au/GCE. For the inactivated-GOx/Au/GCE, no distinct
comparison. It was found from Fig. 4(b) that, in the absence of glu-
cose, the Au/GCE electrode started to be oxidized at ∼0.65 V in the
positive-going potential scan, leading to a sharp increase in current
density; the reduction of Au oxides took place at ∼0.53 V dur-
ing the negative-going potential scan, producing a broad cathodic
peak. However, a notable increase in current appeared at ∼0.25 V
in the presence of glucose, as can be seen from the dashed curve.
By comparing the CVs of Au/GCE and GOx/Au/GCE under identi-
cal conditions, it is concluded that that the immobilization of GOx
significantly inhibited the oxidation of Au films since gold oxide

C. Qiu et al. / Electrochimica Acta 67 (2012) 140–146
145
from the electrode surface [41]. In the next few weeks, the rate
of decay for enzyme response was significantly reduced. 35 days
after being stored, the current response of GOx was 87.5% of its ini-
tial value. The long-time stability test shows that 85.3% of enzyme
activity was still retained after 70 days. This suggests that the
GOx/Au/GCE is quite efficient for retaining the activity of GOx. The
good stability may be attributed to the excellent biocompatibility
of Au nanostructures, which provided a favorable microenviron-
ment for GOx to retain its bioactivity and strong adsorption on the
Au film.
The repeatability of GOx/Au/GCE was also investigated by
detecting the response to 2.5 ␮mol L−1 glucose for five consecu-
tive measurements. The relative standard deviation (RSD) was 3.5%,
indicating that the biosensor had good repeatability.
3.6. Specificity and Interferences
It is well-known that enzyme is highly specific for substrate. The
selectivity of the enzyme electrode is closely related to the property
of enzyme. The GOx possesses the high specific catalytic activ-
ity toward the glucose oxidation even though in the presence of
other carbohydrates. The main interferences of enzyme sensors are
caused by the electroactive species diffusing to the sensor surface.
In this work, 2.5 ␮mol L−1 ascorbic acid (AA) and cysteine (l-Cys)
were consecutively added into the mixed solution of 0.1 mol L−1
PBS and 2.5 ␮mol L−1 glucose to study the anti-interference perfor-
mance of GOx/Au/GCE in the amperometric detection process. As
can be seen from the inset in the lower left of Fig. 5(a), the addition
of AA caused no obvious current response. However, 9.7% current
increment was observed when l-Cys was added. Considering that
the concentration of l-Cys in human serum is much smaller than
that of glucose (normal range: 4–7 mmol L−1), in fact the presence
of l-Cys hardly affected the detection of glucose in serum samples.
Besides, the anti-interference of the enzyme modified electrode
was further tested by DPV. The addition of 5 mmol L−1 glucose
causes a decrease of the peak current at ∼−0.6 V. However, the cur-
rent peak almost remained unchanged when 5 mmol L−1 ␣-sucrose,
d-fructose, d-mannose and d-xylose were added to the mixed solu-
tion of PBS and glucose. All the results imply that the as-prepared
biosensor has excellent anti-interference ability. This is the reason
why we immobilized GOx on electrode surface in the detection of
glucose.
Fig. 5. (a) Typical steady-state response of the as-prepared GOx/Au/GCE for succes-
sive concentration increments of 2.5 ␮mol L−1 glucose in stirred 0.1 mol L−1 PBS; (b)
current response of Au/GCE, GOx/GCE and inactivated-GOx/Au/GCE for successive
addition of 2.5 ␮mol L−1 glucose to the stirred 0.1 mol L−1 PBS. Applied potential:
−0.55 V. Insets in the upper right and the lower left of (a) present the calibration
curve and anti-interference performance of GOx/Au/GCE, respectively.
current response was observed. In contrast, regular amperomet-
ric responses were achieved on the GOx/GCE although the current
step was smaller than that acquired using the GOx/Au/GCE. This
fact indicates that the Au thin films have relatively larger specific
surfaces for the immobilization of GOx and the immobilized GOx
therefore possesses a high enzymatic activity and a good affinity
toward glucose.
3.7. Determination of glucose in human serum sample
3.5. Stability and repeatability of enzyme modified electrode
The GOx/Au/GCE was applied to the determination of glucose
concentration in human serum. A rapid and stable amperomet-
ric response was acquired at −0.55 V with the direct addition of
20 ␮L of samples into 20 mL of 0.1 mol L−1 PBS. The content of glu-
cose in the samples was calculated from the calibration curve and
the obtained results were shown in Table 2. Herein the concen-
tration of glucose in the sample was also measured using a blood
sugar analyzer (SBA-40C). The values given by the analyzer were in
good agreement with our results, thereby demonstrating that the
Stability is a basic requirement for fabrication of glucose sen-
sors. The leaching of the immobilized GOx during the usage was
investigated by comparing the changes of redox peak of FAD/FADH2
from their CVs measured in mixed solutions of 0.1 mol L−1 PBS and
0.1 mmol L−1 glucose. In the initial period, the oxidation peak cur-
rent was observed to gradually increase with the number of CV
cycles and reach the maximum in the 50th cycle (20% larger than
the initial current). As the voltammetric cycling went on, the peak
current slowly decreased. For example, the peak current in the
300th cycle was 107% of the initial value and 89.3% of the max-
imum. These results clearly indicated that the immobilized GOx
was difficult to leach from the Au thin film, which demonstrated
that the GOx modified electrode exhibited the excellent stability
and reusability.
Table 2
The comparison between the concentration values of glucose measured with
GOx/Au/GCE biosensor and SBA-40C blood sugar analyzer (average values of 4
measurements).
Sample no. Concentrations of glucose
(mmol L−1
Deviation
(mmol L−1
)
)
At the same time, the storage stability of the GOx modified
electrode was also investigated. The as-prepared enzyme electrode
was preserved in ultra-pure water at 4 ◦C in refrigerator. The slight
decrease in the sensitivity during the first 7 days (about 94.6% of
remains) was caused primarily by the leaching of the bound enzyme
Determined by GOx
modified electrode
Determined by blood
sugar–glucose analyzer
1
2
7.04
6.70
7.105
6.55
−0.065
+0.15

146
C. Qiu et al. / Electrochimica Acta 67 (2012) 140–146
as-prepared biosensor was effective and sensitive for the determi-
nation of glucose in actual samples.
[9] F. Hu, S. Chen, C. Wang, R. Yuan, Y. Chai, Y. Xiang, C. Wang, J. Mol. Catal. B:
Enzym. 72 (2011) 298.
[10] M. Li, S. Xu, M. Tang, L. Liu, F. Gao, Y. Wang, Electrochim. Acta 56 (2011) 1144.
[11] Q. Liu, X. Lu, J. Li, X. Yao, J. Li, Biosens. Bioelectron. 22 (2007) 3203.
[12] Y. Shao, J. Wang, H. Wu, J. Liu, I.A. Aksay, Y. Lin, Electroanalysis 22 (2010) 1027.
[13] C. Deng, J. Chen, X. Chen, C. Xiao, L. Nie, S. Yao, Biosens. Bioelectron. 23 (2008)
1272.
4. Conclusions
Nano/submicron structured Au films were fabricated onto GCEs
using nanostructured Co thin films as the sacrificial templates
to react with HAuCl4. The as-prepared Au thin films provide an
excellent microenvironment for the adsorption of GOx. The immo-
bilized GOx displays strong inhibition effect on the electron transfer
of non-enzyme-mediated redox reactions but presents good cat-
alytic activity and selectivity for the glucose oxidation. Because
of the excellent sensing performance with high sensitivity, anti-
interference ability and good stability, the GOx/Au/GCE is expected
to act as premising biosensors for the selective detection of glucose.
[14] X. Shangguan, H. Zhang, J. Zheng, Electrochem. Commun. 10 (2008) 1140.
[15] Z. Cao, Y. Zou, C. Xiang, L.-X. Sun, F. Xu, Anal. Lett. 40 (2007) 2116.
[16] Y. Huang, W. Zhang, H. Xiao, G. Li, Biosens. Bioelectron. 21 (2005) 817.
[17] Z.H. Dai, J. Ni, X.H. Huang, G.F. Lu, J.C. Bao, Bioelectrochemistry 70 (2007) 250.
[18] D. Feng, F. Wang, Z. Chen, Sens. Actuators B 138 (2009) 539.
[19] O. Yehezkeli, Y.-M. Yan, I. Baravik, R. Tel-Vered, I. Willner, Chem. Eur. J. 15
(2009) 2674.
[20] X. Chu, D. Duan, G. Shen, R. Yu, Talanta 71 (2007) 2040.
[21] Z. Wen, S. Ci, J. Li, J. Phys. Chem. C 113 (2009) 13482.
[22] C.-H. Lee, S.-C. Wang, C.-J. Yuan, M.-F. Wen, K.-S. Chang, Biosens. Bioelectron.
22 (2007) 877.
[23] Y. Li, H.J. Schluesener, S. Xu, Gold Bull. 43 (2010) 29.
[24] J.T. Zhang, C.C. Qiu, H.Y. Ma, X.Y. Liu, J. Phys. Chem. C 112 (2008) 13970.
[25] Q.S. Chen, S.G. Sun, J.W. Yan, J.T. Li, Z.Y. Zhou, Langmuir 22 (2006) 10575.
[26] S. Zhang, N. Wang, Y. Niu, C. Sun, Sens. Actuators B 109 (2005) 367.
[27] R.K. Shervedani, A.H. Mehrjardi, N. Zamiri, Bioelectrochemistry 69 (2006) 201.
[28] F. Patolsky, M. Zayats, E. Katz, I. Willner, Anal. Chem. 71 (1999) 3171.
[29] H.Y. Ma, C. Yang, S.H. Chen, Y.L. Jiao, S.X. Huang, D.G. Li, J.L. Luo, Electrochim.
Acta 48 (2003) 4277.
Acknowledgment
This work was supported by National Natural Science Founda-
tion of China (21073111, 21175059) and National Basic Research
Program of China (2009CB930103).
[30] H.Y. Ma, L.-Y. Ou Yang, N. Pan, S.-L. Yau, J.Z. Jiang, K. Itaya, Langmuir 22 (2006)
2105.
[31] X.J. Wu, H.Y. Ma, S.H. Chen, Z.Y. Xu, A.F. Sui, J. Electrochem. Soc. 146 (1999)
1847.
[32] G. Wang, N.M. Thai, S.-T. Yau, Biosens. Bioelectron. 22 (2007) 2158.
[33] I. Tinoco, K. Kauer, J.C. Wang, Physical Chemistry-Principles and Applications
in Biological Sciences, Prentice-Hall, Englewood Cliffs, NJ, 1978, p. 606.
[34] E. Laviron, J. Electroanal. Chem. 100 (1979) 263.
[35] R.F. Gao, J.B. Zheng, Electrochem. Commun. 11 (2009) 608.
[36] F.H. Li, J.X. Song, F. Li, X.D. Wang, Q.X. Zhang, D.X. Han, A. Ivaska, L. Niu, Biosens.
Bioelectron. 25 (2009) 883.
[37] Y.H. Wu, S.S. Hu, Bioelectrochemistry 70 (2007) 335.
[38] Q.J. Chi, S.J. Dong, Anal. Chim. Acta 278 (1993) 17.
References
[1] C.M. Wong, K.H. Wong, X.D. Chen, Appl. Microbiol. Biotechnol. 78 (2008) 927.
[2] S.B. Bankar, M.V. Bule, R.S. Singhal, L. Ananthanarayan, Biotechnol. Adv. 27
(2009) 489.
[3] M.E. Sangill, B. Pedersen, Am. J. Kidney Dis. 47 (2006) 636.
[4] F. Ahmad, A.P. Md Yusof, M. Bainbridge, S. Ab Ghani, Biosens. Bioelectron. 23
(2008) 1862.
[5] S. Park, H. Boo, T.D. Chung, Anal. Chim. Acta 556 (2006) 46.
[6] J. Li, J. Yu, F. Zhao, B. Zeng, Anal. Chim. Acta 587 (2007) 33.
[7] Y. Wang, R. Yuan, Y. Chai, W. Li, Y. Zhuo, Y. Yuan, J. Li, J. Mol. Catal. B: Enzym.
71 (2011) 146.
[8] Y. Zhai, S. Zhai, G. Chen, K. Zhang, Q. Yue, L. Wang, J. Liu, J. Jia, J. Electroanal.
Chem. 656 (2011) 198.
[39] C.X. Lei, H. Wang, G.L. Shen, R.Q. Yu, Electroanalysis 16 (2004) 736.
[40] R.A. Kamin, G.S. Wilson, Anal. Chem. 52 (1980) 1198.
[41] J.H. Pei, X.-Y. Li, Electroanalysis 11 (1999) 1266.