Dynamic Article LinksC<Analyst

Cite this: Analyst, 2011, 136, 1846

www.rsc.org/analyst PAPER

Publ

ishe

d on

16

Mar

ch 2

011.

Dow

nloa

ded

by U

nive

rsité

Lav

al o

n 30

/06/

2014

14:

33:0

3.

View Article Online / Journal Homepage / Table of Contents for this issue

Microneedle array-based carbon paste amperometric sensors and biosensors

Joshua Ray Windmiller,a Nandi Zhou,a Min-Chieh Chuang,a Gabriela Vald�es-Ram�ırez,a

Padmanabhan Santhosh,a Philip R. Miller,b Roger Narayan*b and Joseph Wang*a

Received 5th January 2011, Accepted 24th February 2011

DOI: 10.1039/c1an00012h

The design and characterization of a microneedle array-based carbon paste electrode towards

minimally invasive electrochemical sensing are described. Arrays consisting of 3 � 3 pyramidal

microneedle structures, each with an opening of 425 mm, were loaded with a metallized carbon paste

transducer. The renewable nature of carbon paste electrodes enables the convenient packing of hollow

non-planar microneedles with pastes that contain assorted catalysts and biocatalysts. Smoothing the

surface results in good microelectrode-to-microelectrode uniformity. Optical and scanning electron

micrographs shed useful insights into the surface morphology at the microneedle apertures. The

attractive performance of the novel microneedle electrode arrays is illustrated in vitro for the low-

potential detection of hydrogen peroxide at rhodium-dispersed carbon paste microneedles and for

lactate biosensing by the inclusion of lactate oxidase in the metallized carbon paste matrix. Highly

repeatable sensing is observed following consecutive cycles of packing/unpacking the carbon paste. The

operational stability of the array is demonstrated as well as the interference-free detection of lactate in

the presence of physiologically relevant levels of ascorbic acid, uric acid, and acetaminophen. Upon

addressing the biofouling effects associated with on-body sensing, the microneedle carbon paste

platform would be attractive for the subcutaneous electrochemical monitoring of a number of

physiologically relevant analytes.

Introduction

The ability to continuously extract useful physiological infor-

mation from transdermal fluids in a minimally invasive fashion

has remained a major goal of the biomedical devices commu-

nity.1 Such a capability would prove valuable for a number of

physiological monitoring applications in the fitness,1 healthcare,2

and combat3 domains. Electrochemical sensors have played

a dominant role in the field of minimally invasive biosensors,

with extensive development activity driven primarily by the

challenge of continuous glucose monitoring.4–7

Microneedle arrays have been identified as a viable route to

minimally invasive therapeutic delivery of vaccines,8 insulin,9

hormones,10 and other pharmacological agents.11,12 Accordingly,

research on these devices has primarily focused on microneedle-

mediated drug delivery rather than on employing microneedles

for analytical sensing operations. In this regard, few studies have

applied hollow microneedles for transdermal glucose sensing

applications,13,14 although these systems involve the integration

aDepartment of NanoEngineering, University of California, San Diego,9500 Gilman Drive, La Jolla, CA, 92093-0448, USA. E-mail:josephwang@ucsd.edu; Fax: +1 858 534-9553; Tel: +1 858 246-0128bJoint Department of Biomedical Engineering, University of NorthCarolina and North Carolina State University, Campus Box 7115,Raleigh, NC, 27695-7115, USA. E-mail: roger_narayan@unc.edu; Fax:+1 509 696-8481; Tel: +1 919 696-8488

1846 | Analyst, 2011, 136, 1846–1851

of flow-microchannels and the concomitant uptake of biological

fluids, which complicate the process of on-body sensing.

The realization of minimally invasive transdermal sensing of

biochemical analytes without the uptake of biological fluids

demands that the execution of the sensing procedure be per-

formed at the microneedle–transdermal fluid interface. This, in

turn, implies that the electrode transducer is directly employed at

this interface. A key challenge materializes when this approach is

taken: common solid electrode materials lack the plasticity

required to conform with the micrometre-scale geometry and

non-planar features that are the hallmark of microneedle array

devices. Moreover, the electrode material must be amenable to

the co-immobilization of enzymes, catalysts, mediators and

stabilizers. Such co-immobilization imparts selective recognition

and transduction, in conjunction with the high stability essential

for practical minimally invasive detection.

This manuscript reports on the development of a carbon paste-

loaded microneedle array designed for minimally invasive bio-

sensing applications. Of the plethora of electrode materials

available, carbon paste is characterized by a high degree of

plasticity that is essential for optimal packing. Carbon paste

electrodes (CPEs) have been widely employed in electroanal-

ysis.15 Such electrodes couple the advantages of low background

current, low cost, as well as convenient surface renewal and

modification (via the inclusion of modifiers within the paste).16 In

the following sections, we demonstrate the integration of

This journal is ª The Royal Society of Chemistry 2011

Publ

ishe

d on

16

Mar

ch 2

011.

Dow

nloa

ded

by U

nive

rsité

Lav

al o

n 30

/06/

2014

14:

33:0

3.

View Article Online

a modified CPE within microneedle arrays. The microneedle

arrays utilized in this study consist of 9-element arrays of pyra-

midal-shaped hollow microneedles, which possess a 425 mm

diameter aperture through which the modified carbon paste is

extruded. In particular, rhodium-dispersed carbon paste,

known for its extremely low potential detection of hydrogen

peroxide,16–19 was packed within the microneedles to minimize

the contribution of co-existing electroactive interferents. The

resulting needle array CPE sensor design obviates the need for

integrated microchannels and extraction of the interstitial fluid.

A uniform response was achieved following successive repacking

operations. Enzyme-dispersed metallized carbon paste-loaded

microneedle arrays are also shown to be useful for the selective

detection of lactate at a low potential of �0.15 V vs. Ag/AgCl.

This results in negligible contributions from common physio-

logical interferents (e.g., ascorbic acid, uric acid, and acetamin-

ophen) upon the lactate response and simplifies the preparation

of the sensor by precluding the need for an additional permse-

lective layer. The microneedle carbon paste sensor system thus

represents an attractive platform to realize the continuous on-

body monitoring of a multitude of relevant bioanalytes in

a minimally invasive manner.

Materials and experimental methods

Preparation of reagents

Lactate oxidase from Pediococcus sp. (LOx, E.C. 1.13.12.4),

rhodium on carbon (5% Rh w/w), polyethyleneimine (PEI),

mineral oil (d ¼ 0.838 g mL�1), L-lactic acid, hydrogen peroxide

(H2O2), L-ascorbic acid (AA), uric acid (UA), acetaminophen

(AC), ethyl alcohol, potassium phosphate monobasic, and

potassium phosphate dibasic were obtained from Sigma-Aldrich

(St Louis, MO) and were used without further purification or

modification. All experiments were performed with a 0.1 M

phosphate buffer (pH 7.0). Ultrapure water (18.2 MU cm) was

employed in all of the investigations.

Fabrication of the hollow microneedle array

Hollow microneedle arrays were fabricated at the UNC/NCSU

Department of Biomedical Engineering with the aid of Solid-

works (Dassault Systemes S.A., Velizy, France) computer

models. Substrate structures were designed with Magics RP 13

(Materialise NV, Leuven, Belgium). The needles were pyramidal

in shape with a triangular base. The dimensions of each micro-

needle were as follows: an edge length of 1250 mm, a height of

1500 mm, and a vertical cylindrical bore of 425 mm in diameter on

one of the faces of the pyramid structure. The needles were

arranged into 3 � 3 square arrays with 2 mm periodicity.

Substrates for the microneedle arrays were 10 mm � 10 mm in

extent and possessed a thickness of 500 mm. The three-dimen-

sional computer models were transferred to a Perfactory�SXGA Standard UV rapid prototyping system (EnvisionTEC

GmbH, Gladbeck, Germany) for production. This system uses

these computer models to precisely guide light from a 150 W

halogen bulb over a photocurable material, resulting in the

selective polymerization of the exposed material. Eshell 200

acrylate-based polymer (EnvisionTEC GmbH, Gladbeck, Ger-

many) was utilized as the constituent material to fabricate the

This journal is ª The Royal Society of Chemistry 2011

microneedle arrays since the resin selectively polymerizes under

visible light and exhibits a Young’s modulus of elasticity of

3050 � 90 MPa.20 The polymer also offers Class-IIa bio-

compatability per ISO 10993. A 550 mW output power beam

(step size ¼ 50 mm) with a zero-degree tilt was employed for the

polymerization of the resin. Following fabrication, the arrays

were rinsed with isopropanol for the removal of the unpoly-

merized material and subsequently placed in an Otoflash post-

curing system (EnvisionTEC GmbH, Gladbeck, Germany) for

post-build curing.

Preparation of the enzyme-functionalized rhodium-dispersed

carbon paste microelectrode array

100 mg of rhodium on carbon and 10 mg of LOx were thor-

oughly homogenized via 10 alternating 5 min cycles of vortexing

and ultrasonication. The mixture was then vortexed for an

additional 1 h. Following the homogenization process, 125 mg of

the mineral oil pasting liquid and 15 mg of the PEI enzyme

stabilizer were added to the solid mixture. Homogenization of

the resulting paste mixture was accomplished by grinding the

mixture with a mortar and pestle for an additional 1 h.

Packing of the microneedle array

A 3 mL syringe (BD Biosciences, Franklin Lakes, NJ) was

utilized as the support to extrude the metallized carbon paste

through the microneedle array. The nozzle portion of the syringe

was removed to facilitate the attachment of the microneedle

array, which was affixed (using adhesive epoxy) to this cleaved

end for durability. A copper wire was subsequently inserted into

the back end of the syringe barrel in order to create an electrical

contact to the microneedle transducer. Following this procedure,

the carbon paste mixture was loaded into the syringe from the

rear and then extruded with a plunger until the paste began to

expel through the microneedle microholes. Excess paste was

removed from the openings; the surface was later smoothed using

a wax paper. In order to investigate the repeatability of the

response after repacking the microneedles with new paste, the

array was carefully removed from the syringe and subsequently

immersed in ethanol under ultrasonication in order to remove

the extraneous carbon paste residue. A 0.15 mm diameter iridium

wire was used to facilitate removal of the paste from the micro-

hole. The aforementioned assembly and packing protocols were

then followed in order to generate a new electrode from the

cleaned microneedle array.

Instrumentation

A CH Instruments (Austin, TX) model 1232A electrochemical

analyzer was employed for all of the electrochemical measure-

ments. An external Ag/AgCl reference electrode (CH Instru-

ments CHI111) and a 0.5 mm diameter platinum wire counter

electrode were used to establish a three-electrode electrochemical

system. The electrochemical experiments were performed in

a 7 mL cell at room temperature (22 �C). Voltammetric and

chronoamperometric studies were used to evaluate the electro-

chemical behavior of the carbon paste microneedle array elec-

trode. In these electrochemical investigations, either H2O2 or

lactate was added into 5 mL of potassium phosphate buffer

Analyst, 2011, 136, 1846–1851 | 1847

Fig. 2 Scanning electron micrographs of the unpacked (A) and Rh-

carbon paste packed (B) microneedle constituent of the array.

Publ

ishe

d on

16

Mar

ch 2

011.

Dow

nloa

ded

by U

nive

rsité

Lav

al o

n 30

/06/

2014

14:

33:0

3.

View Article Online

solution in order to obtain the desired concentration. Chro-

noamperometric currents were sampled at 15 s following the

potential step. In order to obtain hydrodynamic voltammo-

grams, fixed potential amperograms were recorded in a stirred

phosphate buffer solution containing the desired H2O2 concen-

tration by varying the potential between �0.20 and +0.60 V vs.

Ag/AgCl (in 0.05 V increments). The solution was continuously

stirred using a magnetic stirrer at a rate of 100 rpm. The

morphology of the carbon paste microneedle array was exam-

ined using a field emission scanning electron microscope (Philips

XL30, Amsterdam, The Netherlands). All of the specimens were

coated with chromium prior to analysis using a sputtering

instrument (Energy Beam Sciences Emitech K575X, East

Granby, CT). A deposition current of 130 mA was applied for

30 s to deposit �15 nm of chromium onto the sample surface.

Results and discussion

Characterization of the surface morphology of the carbon paste

microelectrode array

Unmodified and modified carbon pastes can readily conform

with the non-planar features of microneedle array devices. Initial

studies were aimed at characterizing the morphology of the

carbon paste-loaded microneedle array and initiated with a close

examination of the microelectrode surface. An optical micro-

graph of the microneedle array is given in Fig. 1A. This image

shows uniform pyramidal microneedle structures (with trian-

gular bases) possessing a height of 1500 mm as well as the

cylindrical openings (425 mm diameter). Fig. 1B depicts

a microneedle array that has been packed with carbon paste and

subsequently polished. It indicates that the surface has been

smoothly polished to obtain a highly reproducible exposed area,

thereby facilitating reliable electrochemical sensing. An excellent

microelectrode-to-microelectrode uniformity is also observed,

although the surface smoothing and paste removal protocols

greatly differ from those of conventional CPEs. This is attributed

to the fact that the electrode openings are located on the side of

the pyramidal microstructure, thereby presenting additional

challenges when the surface is smoothed with the wax paper.

Pursuant to the characterization of the surface morphology,

a closer inspection of the microneedle was performed using

scanning electron microscopy (SEM). Fig. 2A depicts an electron

micrograph of a single microneedle. The structure of the

Fig. 1 Optical micrographs of the unpacked (A) and Rh-carbon paste

packed (B) microneedle array.

1848 | Analyst, 2011, 136, 1846–1851

microneedle can clearly be observed, namely, the bored cylin-

drical vacancy and the ribbed structure created by the rastering

of the light source over the polymer resin. Fig. 2B illustrates the

surface details of a single microneedle packed with the carbon

paste. A well-formed surface, a relatively smooth morphology,

and defined edges are observed, reflecting the effective filling of

the cylindrical microhole. Such a surface quality is achieved by

extruding the excess paste and later polishing the surface. It

should be noted that the microneedle and the opening appear to

be elongated due to the oblique angle at which the SEM image

was acquired.

Electrochemical characterization of the carbon paste

microelectrode array towards peroxide-based amperometric

sensing

Following the morphology investigation, the initial electro-

chemical experiments were performed to characterize the

response of the carbon paste microneedle array towards H2O2. A

hydrodynamic voltammogram (HDV) was recorded over the

�0.20 to +0.60 V range in order to deduce a suitable operating

potential and to demonstrate the strong catalytic ability of the

Rh-carbon paste microneedle array towards the redox processes

of H2O2. The results, illustrated in Fig. 3A, elucidate that the Rh-

carbon paste microneedle array offers convenient detection of

H2O2 over the entire range evaluated, with a crossover point

occurring around 0.22 V (vs. Ag/AgCl). Such lowering of the

overvoltage enables the selection of a low operating potential of

�0.15 V vs. Ag/AgCl for subsequent sensor investigations. At

this potential, a reduction current of 5.95 mA could be achieved

Fig. 3 (A) Hydrodynamic voltammogram of 0.1 M potassium phos-

phate buffer (a) and 10 mM H2O2 (b) at the rhodium-dispersed carbon

paste microneedle electrode. (B) Chronoamperograms obtained using the

rhodium-dispersed carbon paste microneedle electrode (0–500 mM H2O2

in 50 mM increments; a / k; EAPP ¼ �0.15 V vs. Ag/AgCl). The cali-

bration curve is shown in the inset.

This journal is ª The Royal Society of Chemistry 2011

Publ

ishe

d on

16

Mar

ch 2

011.

Dow

nloa

ded

by U

nive

rsité

Lav

al o

n 30

/06/

2014

14:

33:0

3.

View Article Online

for 10 mM H2O2; the contributions imparted by common elec-

troactive interferences are negligible (see data and discussion

below).

The microneedle CPE displays a wide dynamic range for H2O2

detection. Fig. 3B displays chronoamperograms for increasing

levels of H2O2 to 500 mM (in 50 mM increments). Well-defined

currents, proportional to the H2O2 concentration, are observed.

The resulting calibration curve, based on sampling the current at

15 s following the potential step, displays high linearity

(R2 ¼ 0.999; see the inset). The favorable response for 50 mM

H2O2 (curve b) indicates a limit of detection (LOD) of �20 mM

(S/N ¼ 3), which is in agreement with the low-mM LOD values

reported in the literature for bulk metallized carbon paste elec-

trodes.16,21–23 The ability to detect H2O2 at low potentials is an

attractive feature of the new Rh-carbon paste microneedle array

when positioned for use in minimally invasive oxidase-based

biosensors.

Effect of reconstitution of the carbon paste matrix within the

microelectrode array

A key advantage of carbon paste-based electrodes is their

renewable surface, which can be readily regenerated. Such

regeneration should facilitate the re-use of the microneedle array.

Accordingly, the effect of repetitive packing of the array upon

the resulting response was investigated.

As such, 5 calibration experiments were executed for H2O2

over the 50 to 500 mM H2O2 range, which involved successively

reconstituted carbon paste surfaces. Between each experiment,

the electrode was thoroughly disassembled, cleaned, reas-

sembled, and repacked; its electrochemical response was then

characterized. The results, illustrated in Fig. 4, are indicative of

a highly repeatable calibration. The response of successive

packings deviated by no more than 5.4% from the average

current at each level over the examined concentration range.

Highly linear results are observed over the concentration range

(R2 ¼ 0.997), along with a very low standard deviation

(s < 10 nA). These data demonstrate that repeated packing/

unpacking of the carbon paste constituent in the microneedle

array resulted in a reproducible electrochemical response.

Fig. 4 Calibration curve obtained for H2O2 concentrations from 0 to

500 mM in 50 mM increments (EAPP¼�0.15 V vs. Ag/AgCl, t¼ 15 s). The

effect of reconstitution of the Rh-dispersed carbon paste microneedle

array is illustrated for five subsequent reconstitution operations.

This journal is ª The Royal Society of Chemistry 2011

Biosensing of lactate at the microneedle CPE arrays

Following the optimization of the paste loading and H2O2

detection, a microneedle array CPE biosensor for lactate was

developed. Accordingly, lactate oxidase (LOx)-dispersed metal-

lized carbon paste was prepared using PEI for the electrostatic

entrapment of the enzyme within the matrix. Chronoampero-

metric calibration experiments were performed using the LOx-

Rh-carbon paste microneedle array at �0.15 V vs. Ag/AgCl for

increasing levels of lactate (0 to 8 mM in 1 mM increments).

Typical chronoamperograms are displayed in Fig. 5A; the cor-

responding calibration curve (for current sampling at t ¼ 15 s) is

shown in Fig. 5B. High linearity (R2 ¼ 0.990) and low deviation

(s < 10 nA) are observed. Although the estimated detection limit

of 0.42 mM lactate (S/N ¼ 3) is somewhat higher than the values

reported in the literature for Rh-CPE biosensors,18,23,24 it is still

well below normal physiological levels and is therefore more than

sufficient for relevant applications. It should be noted that the

linear concentration range encompasses the entire physiological

and pathological range of lactate in transdermal fluids,25,26

indicating the potential diagnostic value of the microneedle-

based lactate biosensor.

Interference study with common electroactive compounds

In order to ascertain that the lactate biosensor could function as

intended in the presence of common electroactive substances found

in transdermal fluids, an interference investigation was conducted

using physiological levels of these compounds. Fig. 6 illustrates the

results of the chronoamperometric experiments involving a poten-

tial step to�0.15 V vs. Ag/AgCl and measurements of 1 mM lactate

in the presence of 60 mM ascorbic acid (AA), 500 mM uric acid

(UA), and 200 mM acetaminophen (AC). As evident, the addition

of any of these common electroactive interferents resulted in

a negligible effect on the lactate response. A maximum current

deviation of only 1.5% from the 1 mM lactate level was observed for

the addition of AC. Such interference-free lactate detection reflects

the strong, yet preferential electrocatalytic activity of the Rh-

carbon paste microneedle array towards H2O217,18 and further

supports the potential of the microneedle paste biosensor for

lactate monitoring in transdermal fluids.

Stability of the lactate response

The stability of the microneedle array-based biosensor was

examined from repetitive chronoamperograms for 2 mM lactate

Fig. 5 (A) Chronoamperograms obtained for lactate concentrations

from 0 to 8 mM in 1 mM increments (EAPP ¼ �0.15 V vs. Ag/AgCl). (B)

Calibration curve corresponding to the chronoamperometric current at

t ¼ 15 s.

Analyst, 2011, 136, 1846–1851 | 1849

Fig. 6 Chronoamperograms illustrating the effect of physiologically

relevant electroactive interferents upon the detection of lactate

(EAPP ¼ �0.15 V vs. Ag/AgCl).

Publ

ishe

d on

16

Mar

ch 2

011.

Dow

nloa

ded

by U

nive

rsité

Lav

al o

n 30

/06/

2014

14:

33:0

3.

View Article Online

over a 2 hour period. An initial short preconditioning step was

necessary. This process involved the immersion of the carbon

paste microneedle array in a 0.1 M potassium phosphate buffer

(pH 7.0) and the concomitant recording of 6 chronoampero-

grams, followed by the immersion of the array in a 2 mM lactate

solution for 10 min while recording 2 chronoamperograms. After

such preconditioning, the current was sampled every 10 min over

the entire 2 hour stability test period. Fig. 7 illustrates the time-

course profile of the resulting current response (with the initial

reading at t ¼ 0 min normalized to 100%). A stable current was

achieved almost immediately following the initialization of the

experiment, with only a slight increase (9.7%) over the entire 2

hour time-course. The stable response reflects the integrity of the

carbon paste microneedle array biosensor. Tight packing of the

carbon paste, which prevents the potential accumulation of the

enzymatic product within the microneedle openings, is essential

for the stable response. Longer stability experiments with rele-

vant clinical samples and appropriate protective coatings are

planned in follow-up studies.

Fig. 7 Stability of the electrochemical response of the microneedle array

for 2 mM lactate (EAPP ¼ �0.15 V vs. Ag/AgCl) over a 2 hour duration.

1850 | Analyst, 2011, 136, 1846–1851

Conclusions

We have developed, evaluated, and demonstrated the attractive

sensing performance of carbon paste-containing microneedle

arrays. The coupling of CPE transducers with microneedle hosts

addresses the challenges associated with the integration of solid

electrodes with the non-planar features of microneedle arrays.

Furthermore, it obviates the need for integrated microchannels

as well as the extraction of the interstitial fluid. The low-potential

detection of H2O2 was illustrated using this microneedle system

and the effect of paste reconstitution within the microneedle

array was examined. It was demonstrated that a reproducible

amperometric response could be achieved following successive

reconstitution of the carbon paste matrix. Highly linear lactate

detection was achieved over the entire physiological range, along

with the high selectivity imparted by the very low cathodic

detection potential. The high selectivity, sensitivity, and stability

of the carbon paste microneedle array holds promise for diverse

on-body sensing applications. The microneedle carbon paste

sensor platform can be further miniaturized to serve in a multi-

tude of biosensing applications with the selection of higher-

strength polymeric resins. Moreover, the successful realization of

a patch-type microneedle-based on-body sensor paradigm would

require proper attention to the key challenge of biofouling at the

tissue–device interface in connection with the selection of

appropriate surface coatings.

Acknowledgements

This work was supported by the Office of Naval Research

(Award #N00014-08-1-1202). G.V.R. acknowledges post-

doctoral fellowship support provided by CONACyT Mexico.

References

1 B. C. Nindl, A. P. Tuckow, J. A. Alemany, E. A. Harman,K. R. Rarick, J. S. Staab, M. L. Faupel and M. J. Khosravi,Diabetes Technol. Ther., 2006, 8, 244.

2 C. Henry, Anal. Chem., 1998, 70, 594A.3 B. Nindl, Diabetes Technol. Ther., 2009, 3, 371.4 J. Wang, Chem. Rev., 2008, 108, 814.5 G. S. Wilson and Y. Hu, Chem. Rev., 2000, 100, 2693.6 E. S. Wilkins, J. Biomed. Eng., 1989, 11, 354.7 M. M. Rhemrev-Boom, J. Korf, K. Venema, G. Urban and

P. Vadgama, Biosens. Bioelectron., 2001, 16, 839.8 A. P. Raphael, T. W. Prow, M. L. Crichton, X. Chen,

G. J. P. Fernando and M. A. F. Kendall, Small, 2010, 6, 1785.9 W. Martanto, S. P. Davis, N. R. Holiday, J. Wang, H. S. Gill and

M. R. Prausnitz, Pharm. Res., 2004, 21, 947.10 M. Cormier, B. Johnson, M. Ameri, K. Nyam, L. Libiran,

D. D. Zhang and P. Daddona, J. Controlled Release, 2004, 97, 503.11 M. R. Prausnitz and R. Langer, Nat. Biotechnol., 2008, 26, 1261.12 S. Henry, D. V. McAllister, M. G. Allen and M. R. Prausnitz,

J. Pharm. Sci., 1998, 87, 922.13 E. V. Mukerjee, S. D. Collins, R. R. Isseroff and R. L. Smith, Sens.

Actuators, A, 2004, 114, 267.14 K. Tsuchiya, N. Nakanishi, Y. Uetsuji and E. Nakamachi, Biomed.

Microdevices, 2005, 7, 347.15 I. �Svancara, K. Vytras, K. Kalcher, A. Walcarius and J. Wang,

Electroanalysis, 2009, 21, 7.16 J. Wang, N. Naser, L. Angnes, H. Wu and L. Chen, Anal. Chem.,

1992, 64, 1285.17 J. Wang, J. Liu, L. Chen and F. Lu, Anal. Chem., 1994, 66, 3600.18 J. Wang, Q. Chen and M. Pedrero, Anal. Chim. Acta, 1995, 304, 41.19 J. Wang, F. Lu, L. Angnes, J. Liu, H. Sakslund, Q. Chen, M. Pedrero,

L. Chen and O. Hammerich, Anal. Chim. Acta, 1995, 305, 3.

This journal is ª The Royal Society of Chemistry 2011

Publ

ishe

d on

16

Mar

ch 2

011.

Dow

nloa

ded

by U

nive

rsité

Lav

al o

n 30

/06/

2014

14:

33:0

3.

View Article Online

20 S. D. Gittard, R. J. Narayan, J. Lusk, P. Morel, F. Stockmans,M. Ramsey, C. Laverde, J. Phillips, N. A. Monteiro-Riviere,A. Ovsianikov and B. N. Chichkov, Biotechnol. J., 2009, 4,129.

21 X. Cai, K. Kalcher, G. Kolbl, C. Neuhold, W. Diewald andB. Ogorevc, Electroanalysis, 1995, 7, 340.

22 K. Schachl, H. Alemu, K. Kalcher, J. Jezkova, I. Svancara andK. Vytras, Analyst, 1997, 122, 985.

This journal is ª The Royal Society of Chemistry 2011

23 U. Spohn, D. Narasaiah, L. Gorton and D. Pfeiffer, Anal. Chim.Acta, 1996, 319, 79.

24 R. Garjonyte, Y. Yigzaw, R. Meskys, A. Malinauskas and L. Gorton,Sens. Actuators, B, 2001, 79, 33.

25 A. Kratz, M. Ferraro, P. M. Sluss and K. B. Lewandrowski, N. Engl.J. Med., 2004, 351, 1548.

26 I. Hara, Y. Nakano, H. Okada, S. Arakawa and S. Kamidono, Int. J.Urol., 1997, 4, 202.

Analyst, 2011, 136, 1846–1851 | 1851