Talanta 78 (2009) 199–206
Contents lists available at ScienceDirect
Talanta
journal homepage: www.elsevier.com/locate/talanta
Ammonia gas sensor based on electrosynthesized polypyrrole films
Stéphanie Carquigny
a,b
, Jean-Baptiste Sanchez
a
, Franck Berger
a
,
Boris Lakard
b,∗
, Fabrice Lallemand
b
a
LCPR-AC, UMR CEA E4, Université de Franche-Comté, Bâtiment Propédeutique, 16 route de Gray, 25030 Besanc¸ on Cedex, France
b
Institut UTINAM, UMR CNRS 6213, Université de Franche-Comté, Bâtiment Propédeutique, 16 route de Gray, 25030 Besanc¸ on Cedex, France
article info
Article history:
Received 28 July 2008
Received in revised form 24 October 2008
Accepted 31 October 2008
Available online 11 November 2008
Keywords:
Gas sensor
Ammonia
Polypyrrole
Electrochemistry

) and organic vapors (alcohols, ethers, halocarbons) cause
a change of electrical conductivity in the polymer matrix of organic
metals [28–33]. In comparison with most of thecommercially avail-
able sensors, based usually on metal oxides and operating at high
temperatures, the sensors made of conducting polymers have many
improved characteristics. They have high sensitivities and short
response time; especially, these feathers are ensured at room tem-
perature.
∗
Corresponding author. Tel.: +33 3 63 08 25 78.
E-mail address: [email protected] (B. Lakard).
Thus, in this paper, an original ammonia gas sensor based on
micropatterned microelectrodes functionalized by electropolymer-
ization of polypyrrole films is studied. Electrochemical deposition
has been chosen since it is the most convenient method to deposit
conducting polymer films [34–36]. Indeed, the thickness of the film
can be controlled by the total charge passed through the elec-
trochemical cell during the film growing process. More, such a
deposition also allows the preparation of films at a well-defined
redox potential in the presence of a given counter-ion, which then
also defines the level and characteristics of the doping reaction
[37]. Thus, electropolymerization is used in this study to fabricate
a gas sensor consisting in PPy films deposited on microstructured
electrode arrays and also across the insulating gap separating the
microstructured electrodes of the sensor. Indeed, if the insulat-
ing gap between the neighboring electrodes is close enough (a
few micrometers), the growing film can cover the insulated gap
and connect electrodes [38,39]. This is important in fabricating
chemiresistors for gas sensing. A microstructured interdigitated
electrode array was chosen since it represents the most suit-

1200
◦
C in water vapor, in order to produce a 1.3 ␮m thickness SiO
2
layer. Next, a 1.4-␮m thickness layer of negative photoresist (AZ
5214, from Clariant), suitable for lift-off, was deposited by spin
coating. Then, the wafer was first exposed with the mask to a 36-
mJ/cm
2
UV radiation flux delivered by an EVG 620 apparatus, and
then without any mask to a 210-mJ/cm
2
UV radiation flux. Thus,
the pattern was transferred to the resist, which was then devel-
oped, using AZ 726 developer, to dissolve the resist where the metal
was deposited. Then, a magnetron sputtering (Alcatel SCM 441
apparatus) was used to coat microsystems with titanium (30 nm,
used to improve platinum layer), then platinum (150 nm). The fab-
rication parameters for Pt and Ti films were the following ones:
base pressure: 4.6 × 10
−7
mbar, pressure (Ar) during sputtering:
5 × 10
−3
mbar, power: 150 W, target material purity: 99.99%, film
thickness: 150 nm for Pt films and 30 nm for Ti films. The remaining
resist layer was then dissolved using acetone. After the gas sensors
have been fabricated, thepattern and the dimensions are controlled
using an optical microscope. More details about the microsystem
fabrication can be found in a previous paper [27]. A microstruc-

spectroscopy (XPS, SSX-100 spectrometer). XPS was used to control
the elemental composition and to determine the oxidation state of
elements. All recorded spectra were recorded at a 35
◦
take-off angle
relative to the substrate with a spectrometer using the monochro-
matized Al K␣ radiation (1486.6 eV). The binding energies of the
core-levels were calibrated against the C
1s
binding energy set at
285.0 eV, an energy characteristic of alkyl moieties. The peaks were
analyzed using mixed Gaussian–Lorentzian curves (80% ofGaussian
character).
2.3.2. SEM
Examinations of polymer morphologies were performed using
a high-resolution scanning electron microscope. Once synthesize d
and dried, polymer samples were examined in a LEO microscope
(SEM LEO stereoscan 440, manufactured by Zeiss–Leica, Köln, Ger-
many) with an electron beam energy of 15 keV.
2.3.3. IRTF-ATR
All spectra were recorded using a Shimadzu spectrometer
(IR-Prestige 21) in ATR reflexion mode. The specific accessory
used for these analyses is the ATR Miracle Diamond/KRS5 which
allowed us to record spectra between 4000 cm
−1
and 700 cm
−1
.
Resolution was fixed at 4 cm
−1

concentrations.
The effect of gases on the sensor’s electrical properties was
recorded using a basic divisor voltage bridge (Fig. 1). With these
experimental conditions, the relationship between the variation
of the sensor’s conductance and the variation of the voltage U
R
is
defined as:
G
C
=
1
R((E/U
R
) − 1)
Any decrease (or increase) of the sensor’s conductance was
recorded as a decrease (or increase) of the electrical signal.
Each new sensor was exposed to a constant nitrogen flow for
12 h before conducing each experiment. This process allowed for
the desorption of pollutant chemical compounds adsorbed onto
the sensitive layer during the storage.
3. Results and discussion
3.1. Electrochemical synthesis of polymer films
Electrochemical synthesis of polypyrrole was performed by
cyclic voltammetry, from an aqueous solution containing 0.1 M pyr-
role and 0.1 MLiClO
4
, on the platinum microelectrodes of the sensor
using a potential sweep rate of 0.1 V s
−1

−
doping agents, on the platinum surfaces. Indeed, XPS spec-
tra of polymer samples reveal the presence of C, N, O, Cl, Pt for
all polymers. Thus, C
1s
signal (Fig. 3b) can be fitted by five differ-
ent carbon species at 284.0, 284.8, 286.1, 287.8 and 289.8 eV. The
two components at the lowest binding energy relevant to ␤ and
␣ carbon atoms, respectively, revealed the first interesting finding.
In fact, the comparison of these two carbon atoms areas showed
that, following overoxidation, the ␤ carbons in the film were less
abundant than the ␣ ones. This indicates, that the ␤ positions were
the ones involved in the polymer functionalization. The third peak
at 286.1 eV is attributed to carbons of the polymer C
NorC N
+
;
the fourth one at 287.8 eV to C
N
+
carbons and the peak much
weaker at 289.8 eV to carbonyl C
O groups. The appearance of a
C
O component may be associated with the overoxidation of PPy
at the ␤ carbon site in the pyrrole rings. The N
1s
spectra (Fig. 3c)
indicate the presence of four peaks in the case of PPy. It contains a
main signal at 399.6eV which is characteristic of pyrrolylium nitro-

across the insulating gap on the microstructured electrode arrays.
Thus, Fig. 4 shows that the whole surface of the platinum micro-
electrodes is coated by a homogeneous and very compact film of
polypyrrole composed of many nodules (1–2 ␮m long). The mean
Fig. 4. SEM image of PPy film grown on the sensor surface.
film thickness of this polypyrrole film (x) was estimated to 2.25 ␮m
from the electrical charge (q), associated with pyrrole oxidation by
application of Faraday’s law and assuming 100% current efficiency
for polypyrrole formation: x = qM/AzF, where M is the molar mass
of the polymer, F is the Faraday constant,  is the density of the
polymer and z is the number of electrons involved. The nominal
density of the polypyrrole films ()wastakenas1.5gcm
−3
and an
electron loss z of 2.25 was considered.
More, Fig. 4 shows that nodules of PPy are also present in the
insulating gap between the microelectrodes indicating that the
growing film covers the insulated gap and connect microelectrodes.
This point is important since the PPy layer must connect each pair
of interdigitated electrodes in order to obtain the sensitive layer of
the gas sensor.
3.4. Evaluation of the sensor’s electrical signal under NH
3
flow
Firstly, the sensor was exposed to a NH
3
flow at a concentration
equal to 500 ppm with a temperature of the sensitive layer near to
room temperature. The signal’s electrical variation was recorded
versus time. Fig. 5a represents the evolution of the PPy’s conduc-

variation is linear with time. In this way, the calculation of the slope
value gives us information about the sensitivity of the gas sensor.
For a concentration of ammonia equal to 500 ppm the value of the
slope equals to 63.20 nS s
−1
.
In order to evaluate a possible reproductibility of the gas sensor
under ammonia flow at room temperature, we studied two succes-
sive electrical responses of the same gas sensor under pollutant.
The purpose was to compare thesensor’s conductance between two
successive acquisitions. In Fig. 5b is represented the first electrical
response obtained under a constant NH
3
flow and nitrogen flow.
The second curve shows the successive response under NH
3
flow
and N
2
flow after a nitrogen flow exposition of the sensitive layer
during 12 h at room temperature. As shown in Fig. 5b, we notice
a superposition of the signal under ammonia flow during the first
minutes of acquisition. If we consider that the sensor’s electrical
response is measured by referring to experimental point obtained
at the beginning of the exposition of the sensor, one can say that the
electrical signal is reproductible with the same sensor at room tem-
perature. The values of the slope are nearly the same (63.20 nS s
−1
and 65.42 nS s
−1

◦
C) in order to optimize the elec-
trical response. In order to evaluate the impact of temperatures on
the electrical signal of the PPy-based gas sensor, the sensing layer
was heated at temperatures ranging from 25
◦
Cto100
◦
C(Table 1).
A concentration of ammonia equals to 500 ppm was used for this
experiment. Curves show the evolution of the electrical signal when
Fig. 6. Sensor’s electrical responses to various ammonia concentrations (a). Slope of the gas sensor’s electrical response vs. NH
3
concentrations (b).
204 S. Carquigny et al. / Talanta 78 (2009) 199–206
Fig. 7. Infrared spectra of polypyrrole powder before and after an exposition to ammonia flow for 1 h.
Table 1
Values of the slope under ammonia flow at different temperatures.
PPy’s temperature (
◦
C) Slope (nS s
−1
)
25 116.72
50 72.29
75 74.04
100 78.57
the sensor is first stabilised under N
2
flow (until 400 s), second

−1
), C C aromatic stretching (1400 cm
−1
)etC N stretch-
ing (1050cm
−1
).
Then, in order to understand the mechanism of ammonia
adsorption onto PPy surfaces, polypyrrole samples were exposed
during 1 h to NH
3
1000 ppm flow. The spectra obtained for this
sample, compared to PPy under N
2
, is represented in Fig. 7b. Com-
pared to the PPy powder spectrum (Fig. 7a), two broad absorption
bands appeared when PPy was exposed to NH
3
flow. The first
one at 3260 cm
−1
may be attributed to the stretching vibration
of N
H binding in NH
3
+◦
radical group. The second one seems
to be superposed to the C
C aromatic stretching band centered
at 1400 cm

tra. This mechanism is completing the various works realised on the
ammonia detection studies using PPy-based gas sensors [40,41].
3.7. Comparison with other works
Before this study, other authors used conducting polymer
films to develop gas sensors. These polymer films were obtained
using different techniques. The most often used technique was
the chemical deposition by dip-coating [42–45], and the oth-
ers were: spin-coating from soluble conducting polymers [46,47],
thermal evaporation by heating and deposition of the conduct-
ing polymer on a substrate [48], vapor deposition polymerization
[49], drop-coating of a dried polymer solution [50,51], UV-
photopolymerization [52], deposition of Langmuir–Blodgett film
[53] and electrochemical deposition [54,55]. We decide to use this
latter technique since the thickness of the film can be controlled
by the total charge passed through the electrochemical cell dur-
ing film growing process. Moreover, the film can be deposited
on patterned microelectrode arrays [38]. However, if the insulat-
ing gap between the neighboring electrodes is close enough, the
growing film can cover the insulated gap and connect electrodes
[39].
Amongst the various polymer films, polypyrrole is one of the
most studied andinteresting in particularthanks to its high conduc-
tivity. Consequently, many papers have already used this polymer
as active layer of gas sensors. Thus, PPy obtained by chemical oxi-
dation was used for the detection of CO [56],CO
2
[57], xylene
[58], alcohols [43,59,60] or acetone [61]. PPy obtained by vapor
deposition polymerization was also used for the detection of
methanol, ethanol, CCl

50 ppm to 150 ppm. Consequently, the results obtained in our study
are competitive with all these results since the ammonia gas sen-
sors developed in this paper showed a detectable limit of 8ppm
of NH
3
in N
2
. More, the best responses were obtained at room
temperature and were reproductible.
4. Conclusion
The aim of this work was to validate the use of polypyrrole-based
gas sensor for the detection of ammonia at concentrations lower
than 10 ppm. From this study we first electrosynthesized PPy films
doped with small anions ClO
−
4
on metallic electrodes to develop
a chemical resistor gas sensor. A homogeneous polymer deposited
film with a thickness close to the micrometer was obtained. The
various tests conducted under ammonia flow showed an interest-
ing sensitivity (lower than 10 ppm) and a good reproductibility. By
comparison withmost of chemiresistors gas sensors, our PPy-based
sensor presents best sensitivity at room temperature.
References
[1] J. Albuquerque, L. Mattoso, D. Balogh, R. Faria, J. Masters, A.G. MacDiarmid,
Synth. Met. 113 (2000) 19.
[2] H.K. Song, B. Toste, K. Ahmann, D. Hoffman-Kim, G.T.R. Palmore, Biomaterials
27 (2006) 473.
[3] W.R. Stauffer, X.T. Cui, Biomaterials 27 (2006) 2405.
[4] C.D. Dimitrakopoulos, P.R.L. Malenfant, Adv. Mater. 14 (2002) 99.

[25] B. Lakard, O. Segut, S. Lakard, G. Herlem, T. Gharbi, Sens. Actuators B 122 (2007)
101.
[26] S. Lakard, G. Herlem, N. Valles-Villareal, G.Michel, A. Propper, T. Gharbi, B. Fahys,
Biosens. Bioelectron. 20 (2005) 1946.
[27] O. Segut, B. Lakard, G. Herlem, J.Y. Rauch, J.C. Jeannot, L. Robert, B. Fahys, Anal.
Chim. Acta 597 (2007) 313.
[28] L. Torsi, M. Pezzuto, P. Siciliano, R. Rella, L. Sabbatini, L. Valli, P.G. Zambonin,
Sens. Actuators B 48 (1998) 362.
[29] R. Gangopadhyay, A. De, Sens. Actuators B 77 (2001) 326.
[30] J. Janata, M. Josowicz, Nat. Mater. 2 (2003) 19.
[31] A.C. Partridge, P. Harris, M.K. Andrews, Analyst 121 (1996) 1349.
[32] M. Brie, R. Turcu, C. Neamtu, S. Pruneanu, Sens. Actuators B 37 (1996) 119.
[33] M. Matsuguchi, J. Io, G. Sugiyama, Y. Sakai, Synth. Met. 128 (2002) 15.
[34] J.M. Slater, E.J. Watt, N.J. Freeman, I.P. May, D.J. Weir, Analyst 117 (1992) 1265.
[35] J.W. Gardner, M. Vidic, P. Ingleby, A .C. Pike, J.E. Brignell, P. Scivier, P.N. Bartlett,
A.J. Duke, J.M. Elliott, Sens. Actuators B 48 (1998) 289.
[36] P. Ingleby, J.W. Gardner, P.N. Bartlett, Sens. Actuators B 57 (1999) 17.
[37] M. Gross, D. Müller, H.G. Nothofer, U. Scherf, D. Neher, C. Braüchle, K. Meerholz,
Nature 405 (2000) 661.
[38] G.W. Lu, L.T. Qu, G.Q. Shi, Electrochim. Acta 51 (2005) 340.
[39] J. Reemts, J. Parisi, D. Schlettwein, Thin Solid Films 466 (20 04) 320.
[40] H. Bai, G. Shi, Sensors 7 (2007) 267.
[41] I. Lähdesmäki, W.K. Kubiak, A. Lewenstam, A. Ivaska, Talanta 52 (2000)
269.
[42] S.T. McGovern, G.M. Spinks, G.G. Wallace, Sens. Actuators B 107 (2005) 657.
206 S. Carquigny et al. / Talanta 78 (2009) 199–206
[43] J.H. Cho, J.B. Yu, J.S. Kim, S.O. Sohn, D.D. Lee, J.S. Huh, Sens. Actuators B 108
(2005) 389.
[44] S. Brady, K.T. Lau, W. Megill, G.G. Wallace, D. Diamond, Synth. Met. 154 (2005)
25.

[64] N. Guernion, R.J. Ewen, K. Pihlainen, N.M. R atcliffe, G.C. Teare, Synth. Met. 126
(2002) 301.
[65] C.W. Lin, B.J. Hwang, C.R. Lee, Mater. Chem. Phys. 58 (1999) 114.