Anal. Chem.
2007, 79, 2583-2587
Correspondence
A Nonoxidative Sensor Based on a Self-Doped
Polyaniline/Carbon Nanotube Composite for
Sensitive and Selective Detection of the
Neurotransmitter Dopamine
Shah R. Ali, Yufeng Ma, Rishi R. Parajuli, Yetunde Balogun, Warren Y.-C. Lai, and Huixin He*,
Department of Chemistry, Rutgers University, Newark, New Jersey 07102, and Bell Labs, Alcatel-Lucent,
Murray Hill, New Jersey 07974
Most of the current techniques for detection of dopamine
exploit its ease of oxidation. However, the oxidative
approaches suffer from a common problem. The products
of dopamine oxidation can react with ascorbic acid
present in samples and regenerate dopamine again, which
severely limits the accuracy of detection. In this paper,
we report a nonoxidative approach to electrochemically
detect dopamine with high sensitivity and selectivity. This
approach takes advantage of the high performance of our
newly developed poly(anilineboronic acid)/carbon nano-
tube composite and the excellent permselectivity of the
ion-exchange polymer Nafion. The binding of dopamine
to the boronic acid groups of the polymer with large affinity
affects the electrochemical properties of the polyaniline
backbone, which act as the transduction mechanism of
this nonoxidative dopamine sensor. The unique reduction
capability and high conductivity of single-stranded DNA
functionalized, single-walled carbon nanotubes greatly
improved the electrochemical activity of the polymer in
physiological buffer, and the large surface area of the
carbon nanotubes largely increased the density of the
boronic acid receptors. The high sensitivity along with the
improved selectivity of this sensing approach is a signifi-
cant step forward toward molecular diagnosis of Parkin-
sons disease.
Tremendous efforts have been made over the last 30 years to
detect biogenic amines, especially dopamine,1-5 as abnormal
dopamine (DA) transmission has been linked to several neurologi-
* To whom correspondence should be addressed. E-mail: huixinhe@
andromeda.rutgers.edu.
Rutgers University.
Bell Labs, Alcatel-Lucent.
(1) Venton, B. J.; Wightman, R. M. Anal. Chem. 2003, 75, 414A-421A.
(2) Wightman, R. M.; May, L. J.; Michael, A. C. Anal. Chem. 1988, 60 (13),
769A-779A.
(3) Troyer, K. P.; Heien, M. L. A. V.; Venton, B. J.; Wightman, R. M. Curr.
Opin. Chem. Biol. 2002, 6, 696-703.
(4) Heien, M. L. A. V.; Khan, A. S.; Ariansen, J. L.; Cheer, J. F.; Phillips, P. E.
M.; Wassum, K. M.; Wightman, R. M. Proc. Natl. Acad. Sci. U.S.A. 2005,
102 (29), 10023-10028.
10.1021/ac062068o CCC: $37.00 2007 American Chemical Society
Published on Web 02/08/2007
cal disorders, e.g., schizophrenia, Huntingtons disease, and
Parkinsons disease. As Parkinsons disease (PD) is characterized
by a severe depletion of the in vivo dopamine pool,2 the ability to
sensitively and selectively measure the concentration of the
neurotransmitter dopamine could potentially be used for molecular
diagnosis of PD. The ability to physiologically determine the
concentration of dopamine could also benefit the design of
therapeutics and evaluation of their therapeutic efficacy toward
PD.6
Dopamine can be easily oxidized electrochemically at conven-
tional electrodes, which have been used to detect the neurotrans-
mitter both in vitro and in vivo.1-5,7,8 However, there are a number
of problems with electrochemical methods due to the nature of
the oxidative electrode reaction of dopamine. One of the primary
problems is that the concentration of dopamine in the extracelluar
fluid of the caudate nucleus is extremely low (0.01-1 M) for a
healthy individual and in the nanomolar range for patients with
Parkinsons disease,1,9,10 while the concentrations of the main
detection interferents, e.g., ascorbic acid, are several orders of
magnitude higher and the interferents undergo oxidation within
the same potential window as dopamine.
Both sensitivity and selectivity are of equal importance for real
applications. In order to improve the selectivity of the electrodes,
efforts have been made to modify the electrode surfaces,11-15
which have shown various degrees of success by inhibiting the
(5) Phillips, P. E. M.; Stuber, G. D.; Heien, M. L. A. V.; Wightman, R. M.; Carelli,
R. M. Nature 2003, 422, 614-617.
(6) Cui, H. F.; Ye, J. S.; Chen, Y.; Chong, S. C.; Sheu, F. S. Anal. Chem. 2006,
78 (18), 6347-55.
(7) Heien, M. L. A. V.; Phillips, P. E. M.; Stuber, G. D.; Seipel, A. T.; Wightman,
R. M. Analyst 2003, 128, 1413-1419.
(8) Yoo, J.-S.; Park, S.-M. Anal. Chem. 2005, 77, 3694-3699.
(9) Justice, J. B. J. Neurosci. Methods 1993, 48, 263-276.
(10) ONeill, R. D. Analyst 1994, 119, 767-779.
(11) Wang, Z.; Liu, J.; Liang, Q.; Wang, Y.; Luo, G. Analyst 2002, 127 (5), 653-
8.
(12) Zhang, X. J.; Ogorevc, B.; Tavcar, G.; Svegl, I. G. Analyst 1996, 121, 1817-
1822.
(13) Ling, T. R.; Syu, Y. Z.; Tasi, Y. C.; Chou, T. C.; Liu, C. C. Biosens. Bioelectron.
2005, 21, 901-907.
(14) Giz, M. J.; Duong, b.; Tao, N. J. J. Electroanal. Chem. 1999, 465, 72-79.
(15) Rojas, M. T.; Kaifer, A. E. J. Am. Chem. Soc. 1995, 117, 5883.
Analytical Chemistry, Vol. 79, No. 6, March 15, 2007 2583
Figure 1. (a) Tapping mode atomic force microscopic image of the ssDNA/SWNT/PABA nanocomposite, fabricated by electropolymerization
of 3-aminophenylboronic acid in the presence of the ssDNA-wrapped, single-walled carbon nanotubes. (b) A schematic drawing of the layer-
by-layer dopamine sensor on a gold electrode. The top layer is Nafion, which electrostatically repels ascorbate away from the electrode surface.
Dopamine penetrates through this layer to the bottom layer of ssDNA/SWNTs/PABA to bind with the boronic acid groups.
interference reactions or promoting dopamine oxidation at differ-
ent potentials. Inspired by the fact that enzymes can react
selectively with their cognate substrates, efforts have also been
made to immobilize dopamine-specific enzymes, such as polyphe-
nol oxidase, onto the electrodes to increase the selectivity.16
However, the main interferent, ascorbic acid (AA), still severely
hinders the accurate detection of dopamine because the oxidized
dopamine product (dopamine-o-quinone), produced from either
direct oxidation at the electrode or by the enzymes immobilized
on the electrode, can catalytically oxidize ascorbic acid to regener-
ate dopamine that becomes available again for oxidation.17
One approach that alleviates the above-mentioned problems
is an electrochemical detection method that does not rely on
oxidation or reduction of dopamine itself. Beni et al. developed a
method to selectively detect dopamine at the interface between
two immiscible electrolyte solutions.18,19 Strawbridge et al. reported
an alternative approach to detect dopamine in the presence of
ascorbic acid based on the preferential binding of phenylboronic
acid to dopamine, and the resulting boronate ester had a different
oxidation potential than either AA or DA.20 Fabre et al. extended
this principle by using poly(anilineboronic acid) (PABA) as the
selective dopamine receptor, and in this case, the dopamine-
boronate ester complex formation greatly affected the conductivity
of the polyaniline backbone, which was used to transduce the
binding events.21 However, the detection limits of these techniques
were in the micromolar range, and therefore, these detection
methods are not sensitive enough for molecular diagnosis of
Parkinsons disease.
Due to the remarkable electrocatalytic properties and large
surface area of carbon nanotubes (CNTs), modification of elec-
trodes with CNTs has been widely applied to increase the
sensitivity of electrochemical biosensors. CNT-modified electrodes
have been reported to detect dopamine with significantly improved
sensitivity.22-24 However, the detection schemes still rely on direct
oxidation of dopamine at the electrode, despite persistence of the
associated problems, such as dopamine regeneration.
In this work, we combine the benefits of CNTs and PABA to
develop a sensitive and selective electrochemical dopamine sensor.
Recently we drew on the findings25,26 that single-stranded DNA
(ssDNA) can disperse bundled carbon nanotubes in aqueous
solution to produce PABA/ssDNA/single-walled nanotube (SWNT)
nanocomposites by in situ electrochemical polymerization of
3-aminophenylboronic acid monomers.27 The nanocomposite
2584 Analytical Chemistry, Vol. 79, No. 6, March 15, 2007
shows excellent properties through synergistic effects of the
component materials. The electrocatalytic reductive ability of the
ssDNA/SWNTs and their strong interaction with the polyaniline
backbone significantly improved the stability of PABA.28 The
conductivity and electrochemical activity in neutral solutions were
greatly enhanced. The very large surface area of the carbon nano-
tubes greatly increased the density of the boronic acid functional
groups available for sensitive detection of the target analyte.
In this paper, we demonstrate that, by depositing a thin layer
of this nanocomposite (Figure 1a) onto the electrode, dopamine
concentrations as low as 1 nM were detected with cyclic voltam-
metry and 40 pM with differential pulse voltammetry. Toward the
goal of developing a dopamine sensor for in vivo or in vitro
applications, we eliminated the interference from ascorbate while
keeping the high sensitivity afforded by the composite through
deposition of a thin layer of perfluorosulfonated ion-exchange
polymer Nafion above the composite (Figure 1b). This sensing
approach combines the permselectivity of Nafion, the high affinity
of dopamine to boronic acid, the high density of the boronic acid
groups on the electrode due to the ssDNA/SWNTs, and the
superior electrochemical activity of the composite under physi-
ological conditions. Since direct oxidation of dopamine on the
(16) Cooper, J. M.; Foreman, P. L.; Glidle, A.; Ling, T. W.; Pritchard, D. J. J.
Electroanal. Chem. 1995, 388, 143.
(17) Tse, D. C. S.; McCreery, R. L.; Adams, R. N. J. Med. Chem. 1976, 19 (1),
37-40.
(18) Arrigan, D. W. M.; Ghita, M.; Beni, V. Chem. Commun. 2004, 732-733.
(19) Beni, V.; Ghita, M.; Arrigan, D. W. M. Biosens. Bioelectron. 2005, 20, 2097-
2103.
(20) Strawbridge, S. M.; Green, S. J.; Tucker, J. H. R. Chem. Commun. 2000,
2393-2394.
(21) Fabre, B.; Taillebois, L. Chem. Commun. 2003, 2982-2983.
(22) Britto, P. J.; Santhanam, K. S. V.; Ajayan, P. M. Bioelectrochem. Bioenerg.
1996, 41, 121-125.
(23) Wang, Z.-H.; Liang, Q.-L.; Wang, Y.-M.; Lou, G.-A. J. Electroanal. Chem.
2003, 540, 129-134.
(24) Hocevar, S.; Wang, J.; Deo, R. P.; Musameh, M.; Ogorevc, B. Electroanalysis
2005, 17, 417-422.
(25) Zheng, M.; Jagota, A.; Strano, M. S.; Santos, A. P.; Barone, P.; Chou, S. G.;
Diner, B. A.; Dresselhaus, M. S.; Mclean, R. S.; Onoa, G. B.; Samsonidze,
G. G.; Semke, E. D.; Usrey, M.; Walls, D. J. Science 2003, 302, 1545-
1548.
(26) Zheng, M.; Jagota, A.; Semke, E. D.; Diner, B. A.; Mclean, R. S.; Lustig, S.
R.; Richardson, R. E.; Tassi, N. G. Nat. Mater. 2003, 2, 338-342.
(27) Ma, Y. F.; Ali, S. R.; Dodoo, A. S.; He, H. X. J. Phys. Chem. B 2006, 110,
16359-16365.
(28) Ma, Y. F.; Ali, S. R.; Wang, L.; Chiu, P. L.; Mendelsohn, R.; He, H. X. J.
Am. Chem. Soc. 2006, 128, 12064-12065.
Figure 2. (a) Cyclic voltammograms of the composite in PBS and in the presence of different concentrations of dopamine; (b) a correlation
curve of the data presented in Figure1a (n ) 5). Potential scan rate, 100 mVs-1.
Scheme 1. Complexation between Dopamine
and Poly(anilineboronic acid)
electrode was not involved in this sensing approach, its related
problems were avoided.
RESULTS AND DISCUSSION
Detection of Dopamine. After the composite film was
stabilized in 0.01 M PBS (pH 7.4) solutions, various concentrations
of dopamine were added into the electrochemical cell and the
electrochemical current of the polyaniline backbone was mea-
sured. After each addition, the potential was swept between 0.04
and 0.79 V until the cyclic voltammograms (CV) were stable.
Figure 2a shows the stable CV curves of the PABA/ssDNA/
SWNT composite before and after adding different concentrations
of dopamine in PBS solution (pH 7.4). The CV curves show two
redox couples centered at 0.25 and 0.45 V (vs Ag/AgCl),
corresponding to the transition of the polyaniline backbone from
the fully reduced leucoemeraldine state to the partially oxidized
emeraldine salt state and from the emeraldine salt state to the
fully oxidized pernigraniline state, respectively.
Upon addition of dopamine, the redox peak potentials of the
polyaniline backbone barely change. However, the faradic current
(at E ) 0.45 V) of the composite decreases (see Supporting
Information) and the current decreases with increasing concentra-
tions of dopamine (Figure 2). These data suggest that direct
oxidation of dopamine did not occur on the PABA/SWNT
composite-modified electrode. Instead, we believe that the diol
moiety of dopamine chemically binds to the boronic acid groups
along the polyaniline backbone, as shown in Scheme 1.
The faradic current of the anodic peak (IPBS) at E ) 0.45 V
was recorded from the last CV curve of the cyclic voltammetry
cycles used to stabilize the composite in PBS solution and the
faradic current in the last CV cycles of each dopamine addition
(IDA). Figure 2b shows the relative decrease of the faradic current
of the composite in PBS as a function of the concentration of
dopamine. The equation that was used to calculate each of the
data points is shown as an inset in this figure. The correlation
curve demonstrates an initial increasing linear region over which
the binding of dopamine causes a directly proportional ampero-
metric decrease, followed by a relatively horizontal regime in
which dopamine has saturated the boronic acid receptor sites
and therefore the sensor is insensitive to further neurotransmitter
addition. The linear portion, which extends from 1 to 10 nM, is
very reproducible and has a very small standard deviation
(standard error bars displayed on the curve; n ) 5). The
theoretical detection limit (defined as the concentration that
generates a signal three times larger than the noise) of this
composite toward dopamine using cyclic voltammetry is 0.6 nM.
Above the concentration of 30 nM the horizontal regime predomi-
nates: error bars for data points in this region are large because
the saturation limits were quite different for different films. We
do not know the exact reason yet at present, but we think that
the saturation limits depend on the degree of aggregation of the
ssDNA-wrapped carbon nanotubes during deposition and drying
on the electrode surfaces, which consequently influenced the
amount and the quality of the PABA deposited on the electrode.
Currently we are working on better drying methods to prevent
aggregation.
The sensitivity toward dopamine increased 104 times compared
to the previous report in which neat PABA was used to modify
the electrodes in a detection platform of microelectrochemical
transistors.21 We believe that the ssDNA/SWNTs in the composite
increased the effective electrode surface area, causing a higher
density of boronic acid groups available for dopamine binding and
thus significantly enhancing the detection sensitivity. Tapping
mode AFM images (see Supporting Information) clearly illustrate
that the PABA/ssDNA/SWNT composite film has a much greater
surface roughness than the neat PABA film. The enhanced
detection sensitivity might also be due to the improved electro-
chemical activity of the PABA in the composite: compared to neat
PABA, the nanocomposite film has enhanced stability and elec-
trochemical activity in pH 7.4 PBS due to the electrocatalytic
reductive ability of the ssDNA functionalized carbon nanotubes.27
We should mention that higher detection sensitivity could be
achieved by optimizing the detection technique. For example, a
Analytical Chemistry, Vol. 79, No. 6, March 15, 2007 2585
Figure 3. Differential pulse voltammograms of the composite in PBS Figure 4. Correlation curves for the detection of dopamine on the
and upon addition of different concentrations of dopamine. Scan rate, electrode modified with ssDNA/SWNT/PABA/Nafion composite in the
100 mVs-1; pulse amplitude, 50 mV; pulse width, 50 s. absence (b) and presence (9 (blue)) of 0.15 mM AA (n ) 3). Potential
sca rate, 100 mVs-1
microelectrochemical transistor as a detection platform normally
has higher sensitivity because the conductivity of polyaniline can
polyaniline backbone and it also binds to the boronic acid groups
change by many orders of magnitude when its redox states are
through its planar diol as dopamine does, although its binding
switched.29-33 This large change in conductance of the polymer
affinity is lower. Therefore, elimination of the interference can be
leads to amplification of the detection signal. Another technique,
easily achieved by a charge-discriminating membrane to prefer-
differential pulse voltammetry (DPV),34 could greatly decrease the
entially accumulate the positively charged dopamine (pKa 8.9) and
background charging currents and in turn also increase the
reject the negatively charged ascorbate (pKa 4.2) at the electrode
detection sensitivity. Our preliminary study demonstrates that the
surfaces in physiological pH. The widely used perfluorosulfonated
detection limit can improve by 2 orders of magnitude when DPV
polymer Nafion, a cation-exchange polymer, repels ascorbate and
instead of CV was applied to detect dopamine. Figure 3 shows
other anions and can provide a transport channel solely for cations.
that 40 pM dopamine (the smallest concentration tested) induces
Due to their biocompatibility, Nafion films have been extensively
a considerable decrease in the anodic current in DPV curves and
employed for the modification of electrode surfaces and for the
that the current decrease is proportional to the concentration of
construction of amperometric biosensors.38,39 For in vivo or in vitro
dopamine introduced into the electrochemical cell. The theoretical
detection of DA, in this work, we deposited a thin layer of Nafion
detection limit for dopamine detection using the DPV technique
on top of the PABA/SWNT composite to diminish the ascorbate
is 16 pM, a considerable improvement over recent dopamine
interference.
sensors.
Deposition of a thin Nafion layer did not alter the redox activity
Elimination of the Ascorbic Acid Interference with Nafion.
of the ssDNA/SWNT/PABA film in neutral pH solutions. The CV
Ascorbic acid is the most severe interferent in the determination
curves are very similar to the curves shown in Figure 2. There is
of DA in electrochemical sensors. It coexists with dopamine in
no indication of electrocatalytic reduction of AA occurring on the
the extracellular fluid of the central nervous systems, and its
electrode, demonstrating that the Nafion film is able to effectively
concentration is 3-4 orders of magnitude higher than that of
block the access of ascrobic acid to the ssDNA/SWNT/PABA
DA.35,36 We studied37 and found that the mechanism of interference
film. Figure 4 shows the calibration curves of dopamine in the
by ascorbic acid is very different from those approaches relying
absence and presence of 0.15 mM ascorbic acid. In the absence
on direct oxidation of dopamine at the electrode. Instead of
of ascorbic acid, it can be seen that that dopamine calibration curve
regenerating dopamine, in this nonoxidative detection approach,
follows the same shape as the binding curve described in Figure
the ascorbic acid electrocatalytically reduces the fully oxidized
2b, which was obtained on the electrodes without the Nafion layer.
(29) He, H. X.; Sheng, J. S.; Tao, N. J.; Amlani, I.; Nagahara, L. A.; Tsui, R. J. The correlation coefficient for the reproducible detection of
Am. Chem. Soc. 2001, 123, 7730. dopamine is 0.9924, and the detection limit (1.5 nM) is slightly
(30) Paul, E. W.; Ricco, A. J.; Wrighton, M. S. J. Phys. Chem. 1985, 89, 1441-
1447.
higher than that without Nafion (0.6 nM). This is perhaps due to
(31) Bartlett, P. N.; Astier, Y. Chem. Commun. 2000, 105-112. Nafion behaving as a diffusive barrier; i.e., the Nafion film
(32) Ofer, D.; Crooks, R. M.; Wrighton, M. S. J. Am. Chem. Soc. 1990, 112, decreased the amount and/or the speed of dopamine that diffused
7869.
(33) Huang, J.; Wrighton, M. S. Anal. Chem. 1993, 65, 2740.
to the ssDNA/SWNT/PABA film. With 0.15 mM ascorbic acid in
(34) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and the dopamine solutions, the data points in the calibration curve
Applications. 2nd ed.; Wiley: New York, 2001. reside in the positive error region of the calibration curve for
(35) Beni, V.; Ghita, M.; Arrigan, D. W. M. Biosens. Bioelectron. 2005, 20, 2097-
2103.
dopamine without ascorbate present. These results demonstrate
(36) Wightman, R. M.; Deakin, M. R.; Kovach, P. M.; Kuhr, W. G.; Stutts, K. J.
J. Electrochem. Soc. 1984, 131, 1578. (38) Harrison, J.; Turner, R. F. B.; Baltes, H. P. Anal. Chem. 1988, 60, 2002-
(37) Ali, S. R.; Ma, Y. F.; Parajuli, R. R.; Balogun, Y.; Lai, W. Y.-C.; He, H. X. In 2006.
preparation 2007. (39) Wang, J.; Tuzhi, P. Anal. Chem. 1986, 58, 3257-3261.

2586 Analytical Chemistry, Vol. 79, No. 6, March 15, 2007

the near-complete elimination of interference from ascorbic acid.
The strong cation exchange of Nafion leads to an uptake of the
positively changed dopamine, and the anionic ascorbate interferent
is electrostatically rejected from the surface because of the
negatively charged Nafion.
CONCLUSIONS
In this report, we have demonstrated that dopamine can be
electrochemically detected with high sensitivity and selectivity by
modifying the electrode surface with a thin layer of in situ polym-
erized poly(anilineboronic acid)/carbon nanotube composite and
a thin layer of the highly permselective Nafion film. Since direct
oxidation of dopamine is avoided in this approach, the associated
problems with direct oxidation, such as dopamine regeneration,
were prevented. The interference from ascorbic acid is eliminated
by coating a thin layer of Nafion on top of the composite.
Furthermore, the DNA-wrapped, single-walled carbon nanotubes
in the composite not only greatly improved the electrochemical
activity of the composite in physiologically relevant solutions but
they also increased the effective electrode surface area and,
therefore, the density of boronic acid groups available for dopam-
ine binding. These features significantly enhanced the sensitivity
for dopamine detection. The sensitivity of the sensor along with
its improved selectivity might allow for its potential use in the
diagnosis of Parkinsons disease.
ACKNOWLEDGMENT
Acknowledgment is made to the donors of the American
Chemical Society Petroleum Research Fund for partial support
of this research. Support from a Rutgers University Research
Council Grant is gratefully acknowledged. The fabrication of
Au substrates done at the New Jersey Nanotechnology Con-
sortium (NJNC) was made possible by support from the New
Jersey Economic Development Authority (NJEDA). S.R.A.
acknowledges an Undergraduate Research Fellowship from
Rutgers University (2004-2005). We also thank Drs. P. Huskey,
F. Jaekle, and S. Raynor for their helpful discussions and Drs. B.
Wang and G. Springsteen for assistance with the fluorescence
binding assay.
SUPPORTING INFORMATION AVAILABLE
Detailed experimental procedures and additional AFM images
of gold electrodes modified with the composite and with the pure
polymer as noted in the text. This material is available free of
charge via the Internet at http://pubs.acs.org.
Received for review November 3, 2006. Accepted January,
1, 2007.
AC062068O
Analytical Chemistry, Vol. 79, No. 6, March 15, 2007 2587