Determination of serum glucose using flow injection analysis and highly selective glucose sensor based on composite films

Article abstract of DOI:10.1016/j.electacta.2012.01.013

A novel flow injection analysis (FIA) system suitable for measurements of glucose in blood serum is developed. In the proposed FIA system, a new kind of glucose sensor based on composite polymer films and well-immobilized enzyme was fabricated. An electrochemical technique of scanning electrochemical microscopy (SECM), and electrochemical impedance spectroscopy (EIS) were used for the characterization of the newly fabricated biosensor. A wide linear range of 0.1-50 mM for glucose detection was reported in virtue of the new configuration of the sensor and the developed FIA system. The reproducibility of signals was quite good with relative standard deviation (RSD) values for n = 4 injections (typically 5.7%). Animal blood serum was directly injected and assayed in this simulative physiological system. Good analytical recovery of glucose spiked into serum samples, with recoveries in the range of 96.7-105.0%, was exhibited. Under optimized conditions, detection of serum glucose for normal people and diabetics using our proposed method is possible.

Full text of DOI:10.1016/j.electacta.2012.01.013

Electrochimica Acta 65 (2012) 90–96
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Determination of serum glucose using flow injection analysis and highly
selective glucose sensor based on composite films
∗
Chuncui Huang, Qian Wang, Chuantao Gu, Hui Bo Shao
Key Laboratory of Cluster Science, Ministry of Education of China, School of Chemistry, Beijing Institute of Technology, Beijing 100081, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel flow injection analysis (FIA) system suitable for measurements of glucose in blood serum is devel-
oped. In the proposed FIA system, a new kind of glucose sensor based on composite polymer films and
well-immobilized enzyme was fabricated. An electrochemical technique of scanning electrochemical
microscopy (SECM), and electrochemical impedance spectroscopy (EIS) were used for the characteri-
zation of the newly fabricated biosensor. A wide linear range of 0.1–50 mM for glucose detection was
reported in virtue of the new configuration of the sensor and the developed FIA system. The reproducibil-
ity of signals was quite good with relative standard deviation (RSD) values for n = 4 injections (typically
Received 19 October 2011
Received in revised form 5 January 2012
Accepted 5 January 2012
Available online 20 January 2012
Keywords:
Serum glucose
Flow injection analysis
Glucose sensor
5
.7%). Animal blood serum was directly injected and assayed in this simulative physiological system. Good
analytical recovery of glucose spiked into serum samples, with recoveries in the range of 96.7–105.0%,
was exhibited. Under optimized conditions, detection of serum glucose for normal people and diabetics
using our proposed method is possible.
©
2012 Elsevier Ltd. All rights reserved.
1
. Introduction
were applied [4]. Kolev et al. successfully developed a novel FIA sys-
tem that used a polymer inclusion membrane (PIM) for the on-line
extractive separation and determination of Zn(II) in the presence
of a range of other metal ions [5]. The FIA method has been used
by some groups for glucose measurements [6,7] as well, however,
problems such as poor ability of anti-interference and narrow lin-
ear range are difficult to resolve, producing complicated process. In
the present work, a kind of glucose sensor with new configuration
was fabricated and further applied as the detector in the FIA sys-
tem, resulting in good ability of anti-interference and wide linear
range for sample detection, and the process for glucose measure-
ments was very simple. Animal serum samples containing glucose
were directly injected into the system and further carried by the
flowing streams to the position of detector where samples reacted
with oxygen and enzyme. Consequently, hydrogen peroxide gen-
erated by the enzymatic reaction would be detected and measured
by the biosensor in proportion to the amount of glucose in sam-
ples [8]. A typical recording of output has the form of a current
peak, and the height of the peak is related to the concentration of
the analyte according to a standard calibration curve obtained with
glucose standards prepared [9].
Diabetes mellitus is a group of dysmetabolic syndrome resulted
from genetic cause, metabolic diseases, autoimmune disorders,
microorganism infection, toxins, etc. Diabetic emergency such as
hyperglycemia and hypoglycemia needs to be avoided, and it is also
quite necessary to confirm the effectiveness of clinical treatment.
Therefore, testing of physiological glucose levels for normal people
and diabetics is critical. Glucose sensors, as one of the most pop-
ular electrochemical biosensors, have been extensively developed
and applied in clinical testing of diabetes, biological and chemical
analysis and food industry [1–3].
FIA technique was first introduced in 1975, presenting quan-
titative information of samples reproducibly in flowing streams,
and has been widely applied in many fields including biochem-
istry, environmental science and medical science. A certain volume
of samples was injected into a reagent stream continuously flow-
ing through a tube, forming a sample area, and further mixed with
the reagent(s) in streams, and finally reached the detector that typ-
ically quantitates the product of the chemical reaction. Meyerhoff
et al. have reported the measurement of S-nitrosothiols (RSNOs)
in animal blood plasma using FIA method, in which a differen-
tial experiment step and an amperometric nitric oxide (NO) sensor
Polymer films have been widely employed in biosensors to
increase permselectivity, to prevent the electrode surface from
fouling, to entrap or immobilize a mediator, to extend the lin-
ear range of biosensors, and to improve the biocompatibility of
biosensors [10–12]. Polymer films, categorized as conducting,
nonconducting and composite, are mostly fabricated using solvent
∗ Corresponding author. Tel.: +86 10 68912667; fax: +86 10 68912631.
E-mail address: hbs@bit.edu.cn (H.B. Shao).
0
013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2012.01.013

C. Huang et al. / Electrochimica Acta 65 (2012) 90–96
91
F
B
C
D
E
A
Fig. 1. Schematic diagram of flow injection analysis system. (A) Temperature controlled carrier stream (PBS buffer); (B) peristaltic pump; (C) six-port rotary injection valve;
D) temperature controlled flow-through cell equipped with detector (amperometric glucose sensor); (E) waste; (F) recorder.
(
casting, electropolymerization, and adsorption. Belanger et al.
electropolymerized a conducting glucose oxidase/polypyrrole film
onto the Pt disk of a rotating ring-disk electrode to fabricate a
glucose sensor [13]. Yacynych et al. evaluated the effectiveness
of various electropolymerized films and reported that poly(1,3-
phenylenediamine/resorcinol) film covered electrode was not
fouled by serum solution. In that work, glucose oxidase (GOD)
was immobilized onto the electrode surface by crosslinking with
glutaraldehyde followed by an electropolymerized film, and the
resulted sensor had a linear response (2.5–10 mm) using FIA
method [14]. In the present work, we simultaneously immobilized
GOD into the interspace of poly(1,3-phenylenediamine/resorcinol)
molecules during electropolymerization followed by modification
of Nafion perfluorinated ion exchange resin (5 wt.% solution in
2.2. Apparatus and method
Electrochemical experiments were carried out using a devel-
oped FIA system, the diagram of which is shown in Fig. 1. A
peristaltic pump (BT01-Y21515, Tianjin Xieda electron. Co., Ltd.,
Tianjin, China) was applied to pump the carrier stream (PBS),
and the thermostat (85-2, Siyue Instrument, Co., Shanghai, China)
was used to keep the required temperature. The sample loop and
polyetheretherketone (PEEK) tubing used for carrier stream were
purchased from a commercial supplier (RUSH Science & Technol-
ogy Co., Ltd., Hangzhou, China). All current measurements from the
glucose sensor were performed using a potentiostat (CHI 920C,
Chenhua Instrument Co., Ltd., Shanghai, China). Samples were
injected into the carrier stream through the six-port rotary valve
(Rhenodyne 7725i) and loaded into the fitted 200 ␮L sample loop.
Immediately, samples would be carried to the thermostatic flow-
through cell equipped with detector in the oxygen atmosphere,
and glucose contained in samples would be reacted to generate
hydrogen peroxide as a product. Glucose sensor, as the detec-
tor, would respond to this hydrogen peroxide, and a current peak
related to glucose concentration was displayed by the recorder.
Calibration curves were obtained by injecting freshly prepared
standard glucose solutions into the FIA system. All samples and
all components of the FIA system including the carrier stream,
tubing and electrochemical cell were prevented from light with
aluminum foil. Furthermore, EDTA was added to blood samples
as an anticoagulant. Construction procedure of polymer films was
analysed by means of the technique of scanning electrochemical
microscopy (SECM, Chenhua Instrument Co., Ltd., Shanghai, China)
and electrochemical impedance spectroscopy (EIS, CHI 920C, Chen-
hua Instrument Co., Ltd., Shanghai, China).
lower aliphatic alcohols/H O mix containing 45% water) film. A
2
wider linear range of 0.1–50 mM for serum glucose detection was
achieved in FIA system, and such configuration of glucose sensor
and so wide linear range for detection have rarely been reported
by now.
2
. Experimental
2.1. Materials and reagents
Glass carbon working electrode (1.5 mm in radius), platinum
wire counter electrode (0.5 mm in radius) and Ag/AgCl reference
electrode were purchased from CH Instrument (Chenhua Instru-
ment Co., Ltd., Shanghai, China). All potentials were reported
against the Ag/AgCl reference electrode.
All chemicals were of analytical grade or better and used
as received without further purification. d-(+)-Glucose and
glucose oxidase (GOD) were obtained from Sigma–Aldrich.
Potassium ferricyanide, ethylenediaminetetraacetic acid (EDTA),
2.3. Fabrication of glucose sensors
1
,3-phenylenediamine (m-PD) and resorcinol were purchased from
The glass carbon working electrode was mechanically polished
a chemical supplier (J&K Scientific Co., Ltd., Beijing, China). Alumina
polishing powder (1.0 ␮m, 0.3 ␮m, 0.05 ␮m) was obtained from
CH Instrument (Chenhua Instrument Co., Ltd., Shanghai, China).
Potassium chloroplatinate was obtained from a commercial chem-
ical supplier (Boyuan Chemicals Co., Ltd., Jinan, China). Nafion
perfluorinated ion exchange resin was purchased from Dupont
Company (DE, USA). 0.01 M phosphate buffer solution (PBS) was
prepared freshly from appropriate reagents. All aqueous solutions
were prepared with 18.2 Mꢀ cm ultrapure water using a Milli-Q
filter (Research UV, Hetai Instrument Co., Ltd., Shanghai, China).
with successively finer grades of deagglomerated alumina slur-
ries down to 0.05 ␮m in particle size. An ultrasonic cleaner was
used to remove residual alumina loosely bound to the electrode
surface. The polishing procedure was repeated until the working
electrode was electrochemically clean for use. The ensuing working
electrode was platinized in a saturated potassium hexachloroplati-
nate solution by cycling the potential from +0.70 to −0.35 V (vs.
Ag/AgCl) at a scan rate of 20 mV/s for 8 min using the potentiostat.
Poly(m-PD/resorcinol) films were electrochemically grown from
a fresh solution containing 1.5 mM m-PD and 1.5 mM resorcinol

9
2
C. Huang et al. / Electrochimica Acta 65 (2012) 90–96
1
0
.5
.5
10000
8
6
000
000
-
-
-
-
0.5
1.5
2.5
3.5
4000
2000
0
0
0.1
0.2
0.3
0.4
0.5
0.6
GCE
GCE/Polymer
GCE/Polymer/GOD
Potential/V vs.Ag/AgCl
Fig. 2. Cyclic voltammogram of ferrocenemethanol in 0.1 M KCl solution.
0
5000
10000
15000
20000
Zre(Ω)
monomers dissolved in deoxygenated buffer. Enzyme immobiliza-
tion was performed by adding 700 U/mL of GOD to the monomers
solution prior to electropolymerization. Films were grown gradu-
ally by cycling the potential from 0.00 to +0.90 V (vs. Ag/AgCl) at a
scan rate of 2 mV/s for 18–21 h. During electropolymerization, the
solution was protected from light and bubbled with nitrogen all
the time. After electropolymerization, the working electrode was
further modified with a thin layer of Nafion via dip-coating and
then dried in the nitrogen atmosphere for 10 min. The fabricated
Fig. 4. EIS for the growth of poly(m-PD/resorcinol) polymers with and without GOD
on glass carbon electrode (GCE) in PBS with 10 mM Fe(CN)6
5 mV; frequency range 60 kHz–1 Hz.
4−/3−
. Signal amplitude
Allen J. Bard group based on scanning tunneling microscopy (STM)
and the ultramicroelectrode (UME), and was suggested to give
chemical information of systems investigated [15,16]. SECM can
operate in many working modes, and the widely used one is pos-
itive/negative feedback mode [17]. Chemical information such as
conductive or nonconductive of substrate can be obtained by means
of the tip current. As shown in Fig. 2, the cyclic voltammogram of
ferrocenemethanol presents typical steady-state curves, indicating
that the tip working electrode is ideal for use. Fig. 3 exhibits the
experimental results when platinized glass carbon electrode and
platinized glass carbon electrode modified with polymers were
applied as the substrate respectively. As shown in Fig. 3a, the
current on tip electrode increased when the tip approached to
platinized glass carbon electrode substrate, demonstrating that
the substrate is conductive. This is in agreement with the posi-
tive feedback working mode of SECM. However, as illustrated in
Fig. 3b, the current decreased during the same approaching pro-
cess when the platinized glass carbon electrode modified with
polymers was applied as the substrate. This is in agreement with
the negative feedback working mode of SECM, and it demon-
strates that the nonconductive composite polymer films have been
modified onto the surface of conductive platinized glass carbon
electrode.
◦
biosensor was stored in the PBS at 4 C.
3
. Results and discussion
3.1. Analysis of construction procedure of polymer
Scanning electrochemical microscopy (SECM) is a kind of novel
technique for in situ electrochemical study. It was developed by
a 1.1
1
0
0
0
0
.9
.8
.7
.6
0
4
8
12
16
20
Electrochemical impedance spectroscopy (EIS) is frequently
used in the characterization of electrode surface during the pro-
cedure of construction [18,19]. The semicircle diameter in the
impedance spectrum equates to the charge-transfer resistance, Rct,
at the electrode surface [20]. As shown in Fig. 4, the impedance
spectrum of bare glass carbon electrode presents an approxi-
mate straight line, demonstrating that electron transfer to the
surface of glass carbon electrode is a diffusion-control step. After
electropolymerization, the charge-transfer resistance increased
to approximate 7000 ꢀ. It is suggested that the nonconductive
poly(m-PD/resorcinol) films have been modified onto the surface
of the electrode, and that this kind of composite polymer films
do not facilitate electron transfer. This result is consistent with
the analysis performed by SECM. This nonconductive polymer was
applied onto biosensors to prevent interferences, to prevent foul-
ing of the electrode surface by proteins and other substances and
to entrap biocomponent [13]. The charge-transfer resistance fur-
ther increased to about 12,000 ꢀ when GOD was in the presence
of monomers solution during electropolymerization, which mani-
fests that GOD had been well immobilized into the composite poly
L=d/a
b 1.1
1
0
0
0
0
.9
.8
.7
.6
0
4
8
12
16
20
L=d/a
Fig. 3. Experimental approach curves on platinized glass carbon electrode as sub-
strate (a) and on substrate of platinized glass carbon electrode modified with
polymers (b). I, normalized tip current; IT, current at the tip electrode; IT,∞, diffusion-
limiting steady-state current at the tip electrode; L, normalized distance between tip
and substrate; d, distance between tip and substrate; a, radius of the tip electrode.
(m-PD/resorcinol) films.

C. Huang et al. / Electrochimica Acta 65 (2012) 90–96
93
0
.75 V
b 0.8
a
0.8
0
.6 V
×
0.75 V
0.6 V
0
0
.55 V
.45 V
0
0
0
.6
.4
.2
0
0
0
0
.6
.4
.2
0
0
.55 V
0.45 V
0
.1mM
5
0
10
15
20
25
0
0.1
0.2
0.3
0.4
0.5
0.6
Time/min
Glucose Concentration/mM
c
0.12
0
0
0
.09
.06
.03
0
0.3
0.4
0.5
0.6
0.7
0.8
Potential/V vs.Ag/AgCl
Fig. 5. The amperometric response to consecutively added 0.1 mM glucose (a), the corresponding calibration curves (b), and amperometric current (with RSD for n = 5
injections) of the biosensor varies with the applied potential (c).
3
.2. Optimum applied potential for glucose measurements
0.75 V, indicating that the optimal potential for glucose detection
is 0.6 V. Herein, glucose measurements using FIA system were per-
formed at the potential of 0.6 V.
In order to obtain better performance in terms of sensitiv-
ity to samples, optimum applied potential has to be determined
prior to FIA experiments. Amperometric response to successively
added standard glucose at different applied potential and the cor-
responding calibration curves are shown in Fig. 5. Response to
standard glucose became higher when applied potential increased
from 0.45 V to 0.6 V (vs. Ag/AgCl). However, response to samples
decreased when the applied potential was further increased to
3.3. Optimum pH for glucose measurements
Affinity of GOD to substrate electrode and stability of enzyme in
immobilized state can be affected by the pH value of the reacting
solution [21]. Herein, the effect of pH value on biosensor per-
formance was investigated by measuring the current response to
a 0.8
7
7
.5
.0
b
0.8
7
7
.5
.0
0
0
0
.6
.4
.2
0
6.5
5
.6
0
0
0
.6
.4
.2
0
6.5
.6
.0
5
.0
5
5
0
.1mM
5
0
10
15
20
25
30
0
0.1
0.2
0.3
0.4
0.5
0.6
Time/min
Glucose concentration/mM
c 0.12
0
0
0
.09
.06
.03
4
5
6
7
8
9
pH Value
Fig. 6. The amperometric response to consecutively added 0.1 mM glucose (a), the corresponding calibration curves (b), and amperometric current (with RSD for n = 5
injections) of the biosensor varies with the pH value (c).

9
4
C. Huang et al. / Electrochimica Acta 65 (2012) 90–96
a
1
.8
.6
.4
.2
0
b 0.8
0
0
0
0
0
0
0
.6
.4
.2
0
0
.1mM
5
0
10
15
20
25
0
0.1
0.2
0.3
0.4
0.5
0.6
Time/min
Glucose Concentration/mM
c 0.16
0
0
0
.12
.08
.04
0
0
20
40
Temperature/
60
80
Fig. 7. The amperometric response to consecutively added 0.1 mM glucose (a), the corresponding calibration curves (b), and amperometric current (with RSD for n = 5
injections) of the biosensor varies with the temperature(c).
consecutively added 0.1 mM glucose at optimum potential. As can
be seen in Fig. 6, the biosensor presents the optimal response to
a
samples at pH 6.5, indicating that pH 6.5 is the optimum acidic
condition for enzyme activity. However, in this work, the buffer
solution was adjusted to pH 7.3 in consideration of the physiological
conditions.
1μA
50mM
2
5mM
3
.4. Optimum temperature for glucose measurements
1
0mM
5
mM
Enzyme stability and activity are susceptible to thermal con-
1
mM
ditions. In case of thermal denaturation, the effect of varying
temperature on the biosensor was examined by measuring the
response to successively added 0.1 mM glucose between 20 C and
0
.1mM
◦
◦
6
0 C. As illustrated in Fig. 7, the biosensor response to glucose
5
min
gradually increased with increasing temperature and reached its
maximum value at 50 C. This is because enzyme activity increases
◦
at higher temperature. However, response to glucose decreased
when temperature further increased to 60 C. It is likely that nat-
b 4
◦
ural thermal degradation of enzyme occurred at much higher
3
2
1
0
y = 0.0524x + 0.2473
temperature. As a result, the optimum temperature for GOD is
2
◦
R = 0.99
5
0 C corresponding to a series of temperature values. During mea-
surements of samples, obtaining good response to glucose and
minimizing possible experimental errors such as solution evap-
oration have to be considered. In addition, it is better to keep
physiological conditions when performing experiments. Therefore,
◦
3
7 C was chosen as the working temperature as a compromise in
this study.
0
10
20
30
40
50
3.5. Response to glucose using flow injection analysis system
Glucose Concentration/mM
As shown in Fig. 8a, the amperometric detector yielded a fast
Fig. 8. Response to standard glucose and the corresponding calibration curve. Flow
rate, 2 mL/min.
response to glucose and the response current returned to the base-
line within 5 min. A linear range of 0.1–50 mM using FIA system is
presented in Fig. 8b. As shown in Fig. 9, the reproducibility of the
signals is quite good, with RSD values for n = 4 injections typically
3.6. The ability of anti-interference using FIA system
5
.7%. It is not difficult to find out that this kind of glucose sensor
It is well known that some electroactive species in serum, such
as ascorbic acid (AA), l-cysteine (l-Cys) and urea may influence
the performance of a biosensor [8]. Therefore, the anti-interference
applied in FIA system possesses quite good accuracy, reproducibil-
ity and wider linear range for glucose detection.

C. Huang et al. / Electrochimica Acta 65 (2012) 90–96
95
a
72mM
a
1
μA
1 A
5
0mM
5
0mM
50mM
3
6mM
2
5mM
2
5mM
2
5mM
1
0mM
5
mM
1
0mM
1
2mM
10mM
5mM
mM
5
mM
1
mM
1mM
1
3
mM
1
mM
0
.1mM
min
2
.5mL/min
2.0mL/min
1.5mL/min
5
5min
b
4
3
2
1
0
4
3
2
1
0
b
2 mL/min
.5 mL/min
.5 mL/min
2
1
0
15
30
45
60
75
Glucose Concentration/mM
0
10
20
30
40
50
Fig. 9. Response to different glucose standard solutions (a) and calibration curve
using FIA system (b). Flow rate, 2 mL/min. Data are represented as means ± SD (n = 4).
Flow rate/mL/min
Fig. 11. Effect of the flow rate of carrying buffer on the response to different con-
centrations of glucose.
ability of the fabricated biosensor applied in FIA system was inves-
tigated. In this work, 5 mM urea (more than physiological levels),
3
.7. Optimum flow rate for glucose measurements using FIA
0.1 mM l-Cys (more than physiological levels) and 0.1 mM AA
system
(
around physiological levels) were injected into the FIA system
consecutively, followed by injection of various concentrations of
glucose as a comparison. As can be seen from Fig. 10, these inter-
ferences did not have any obvious effect on the biosensor compared
with successively injected standard glucose samples using the FIA
method. Such good ability of anti-interference can be attributed
to two factors. On the one hand, interfering species are likely
blocked by the composite polymer films and Nafion film modi-
fied on the biosensor surface. On the other hand, different from
electrochemical reaction of hydrogen peroxide, interferences can
probably not be oxidized at the low applied potential, and this
is the catalytic function of the platinized platinum. These results
indicate that the developed glucose biosensor and the proposed
FIA method exhibit good quality in anti-interference, and hence
the biosensor applied in FIA system can be used for blood serum
measurements.
Biosensor performance would be influenced by the flow rate
when glucose sensor was applied in FIA system. The key factors
considered during method development were the rate of enzy-
matic reaction and the sensor’s response time toward hydrogen
peroxide. The optimum flow rate should not only be slow enough
to allow complete oxidation of the glucose by the enzyme but also
fast enough to detect hydrogen peroxide within the response time
of sensor. Three different flow rates were chosen and examined
considering of not causing pressure problems and not producing
noisy amperometric signals. As clearly shown in Fig. 11, increas-
ing the flow rate from 1.5 mL/min to 2 mL/min, the sensitivity of
the system increased significantly. This effect is likely due to the
reduced dispersion and hence dilution of the glucose sample plug
within the FIA system when the flow rate is greater [4]. As a result,
lower amounts of glucose can be detected at the higher flow rates.
However, when further increasing the flow rate to 2.5 mL/min, the
sensitivity of the system decreased. It is likely that carrier stream
containing samples passed by the detector at such high flow rate
and there is not adequate time for enzyme to completely oxidize
glucose. As a result, 2.0 mL/min was determined to be the optimum
flow rate and was used for sample measurements.
2
0mM
glucose
1μ A
3
mM
glucose
3.8. Detection of glucose in blood serum
4
injections
4
injections
0
.1mM
of 5mM urea
of 0.1mM AA
Fresh blood serum samples (from rabbits) were obtained from
School of Life Science and Technology at Beijing Institute of Tech-
nology and centrifuged (2000 rpm) for 15 min. The supernatant
glucose
(serum) was carefully aspirated at room temperature and pooled
5
min
into a tube for assay using the FIA system. The levels of glucose
in the serum samples as determined from calibration of the FIA
system were 4.8 ± 0.09 mM (Table 1) which are within the normal
range of serum glucose of rabbits. The accuracy of this FIA method
was evaluated by performing recovery experiments with serum
4
injections
of 0.1mM L-cys
Fig. 10. Response to standard glucose solutions and interferences using FIA system.
Flow rate, 2 mL/min.

9
6
C. Huang et al. / Electrochimica Acta 65 (2012) 90–96
Table 1
Summary of glucose concentration in animal serum using the FIA system.
the method can accurately quantitate the levels of glucose added
to the serum samples.
Injections
Glucose concentration (mM)
4
. Conclusion
1
2
3
4.7
4.8
4.9
In summary, we have described a relatively simple FIA sys-
tem to detect serum glucose in a wide linear range of 0.1–50 mM
using a fabricated glucose sensor that was modified with compos-
ite films to prevent interferences and to entrap enzyme. Animal
blood serum was directly injected and assayed in the FIA system.
Good reproducibility and analytical recovery were achieved using
the proposed method. The main advantages of this new FIA sys-
tem are the wide linear range for detection and the potential to
have a high throughput assay system, which can be referred as the
simulation of the flowing physiological system when compared to
the measurement of glucose in blood serum or whole blood using
the previously reported method [22,23]. In addition, electrochem-
ical arrangement used here for FIA measurements of glucose could
also be employed as a simplified HPLC method [24], and could
be applied for detection of serum glucose for normal people and
diabetics under optimized conditions.
Average
4.8 ± 0.09
a
0
.5μA
5
0mM
2
0mM glucose
added to serum
2
5mM
Blood
serum
5
mM
0
.1mM
3
mM glucose
4mM glucose
Standard glucose calibration
added to serum added to serum
5
min
Acknowledgements
b 4
This work was supported by the National Natural Science Foun-
dation of China (21173023). We are indebted to the Project of Base
for Introducing Talents of Discipline to Universities (111 project,
No. B07012) in China. Great help from Prof. Yan Li in School of Life
Science and Technology at Beijing Institute of Technology is greatly
acknowledged.
3
2
1
0
y = 0.0527x - 0.0324
2
R
= 0.9922
0
10
20
30
40
50
References
Glucose Concentration/mM
[
[
1] D.S. Bindra, Y. Zhang, G.S. Wilson, Anal. Chem. 63 (1991) 1692.
2] C. Shan, A. Ivaska, L. Niu, et al., Anal. Chem. 81 (2009) 2378.
Fig. 12. Typical response to serum, standard glucose and serum that was spiked with
varying amounts of standard glucose (a), and the corresponding standard glucose
[3] L. Doretti, D. Ferrara, P. Gattolin, et al., Biosens. Bioelectron. 11 (1996) 365.
[4] C. Huang, E. Brisbois, M.E. Meyerhoff, Anal. Bioanal. 400 (2011) 1125.
[5] L.L. Zhang, R.W. Cattrall, S.D. Kolev, Talanta 84 (2011) 1278.
◦
calibration curve (b). Flow rate, 2.0 mL/min; temperature, 37 C; pH 7.3; applied
potential, 0.6 V (vs. Ag/AgCl).
[
6] M.J. McGrath, E.I. Lwuoha, M.R. Smyth, et al., Biosens. Bioelectron. 19 (2003)
01.
7] J. Yu, S. Liu, H. Ju, Biosens. Bioelectron. 10 (1995) 937.
4
[
Table 2
[8] T. Kong, Y. Chen, X. Wang, et al., Sens. Actuators B 138 (2009) 344.
[9] Z. Fang, Flow Injection Analysis, vol. 1, Scientific, Beijing, 1999 (Chapter 9).
10] J. Liu, M. Agarwal, K. Varahramyan, Sens. Actuators B: Chemical 135 (2008) 195.
11] E.M. Genies, M. Marchesiello, Synth. Met. 57 (1993) 3677.
12] H. Kudo, T. Sawada, K. Mitsubayashi, et al., Biosens. Bioelectron. 22 (2006) 558.
13] D. Belanger, J. Nadreau, G. Fortier, Electroanalysis 4 (1992) 940.
Summary of glucose recovery experiments in animal serum.
[
[
[
[
Amount of glucose
Added (mM)
% Recovery
Found (mM)
3
4
2.9
3.9
21
96.7%
97.5%
105.0%
[14] R.J. Geise, J.M. Adam, A.M. Yacynych, et al., Biosens. Bioelectron. 6 (1991) 151.
[
[
[
[
[
[
15] A.J. Bard, F.R. Fan, J. Kwak, et al., Anal. Chem. 61 (1989) 132.
16] J. Rodríguez-López, A.J. Bard, J. Am. Chem. Soc. 132 (2010) 5121.
17] J. Kwak, A.J. Bard, Anal. Chem. 61 (1989) 1221.
18] R. Pauliukaite, A. Maria, M.A. Brett, et al., Electrochim. Acta 52 (2006) 1.
19] Y. Feng, S. Chen, W. Guo, et al., Electrochim. Acta 53 (2007) 1743.
20] P. Salazar, R.D. O’Neill, J.L. González-Mora, et al., Sens. Actuators B 152 (2011)
137.
2
0
samples that were spiked with varying amounts of standard glu-
cose, the result of which is shown in Fig. 12 and Table 2. Note that in
Fig. 12, when the spiked samples were injected into the FIA system,
the peak height increased in proportion to the amount of glucose
added to the sample. Based on these recovery data, it is evident that
[21] A. Umar, M.M. Rahman, Y.-B. Hahn, Talanta 78 (2009) 284.
[22] P. Bhandare, Y. Mendelson, A. Peura, et al., Vib. Spectrosc. 3 (1994) 363.
[23] S.V. Dzyadevich, Y.I. Korpan, V.N. Arkhipova, et al., Biosens. Bioelectron. 14
(1999) 283.
[
24] T. Sato, K. Katayama, H. Tazaki, et al., Res. Veter. Sci. 84 (2008) 26.