A

ptwht©

K

1

mcfeuzC[s

sa

0d

Available online at www.sciencedirect.com

Sensors and Actuators B 131 (2008) 174–182

Novel hybrid materials for gas sensing applications madeof metal-decorated MWCNTs dispersed on

nano-particle metal oxides

R. Ionescu a, E.H. Espinosa a, R. Leghrib a, A. Felten b, J.J. Pireaux b, R. Erni c,G. Van Tendeloo c, C. Bittencourt d, N. Canellas a, E. Llobet a,∗

a MINOS, Universitat Rovira i Virgili, ETSE-DEEEA, Av. Paısos Catalans 26, E-43007 Tarragona, Spainb LISE, Falcultes Universitaires Notre Dame de la Paix, 61 rue de Bruxelles, B-5000 Namur, Belgium

c EMAT, University of Antwerp, 171 Groenenborgerlaan, B-2020 Antwerp, Belgiumd LCIA, Materia Nova, Parc Initialis, Av. Copernic, 1, B-7000 Mons, Belgium

Received 5 July 2007; received in revised form 1 November 2007; accepted 2 November 2007Available online 31 December 2007

bstract

Novel hybrid gas-sensitive materials were fabricated by means of metal-decorated multiwall carbon nanotubes (MWCNT) dispersed on nano-article metal oxides. The MWCNT were initially functionalized in an oxygen plasma for improving their dispersion and surface reactivity, and thenhey were decorated with metal nano-clusters by thermally evaporating gold or silver on the MWCNT. Active layers for gas sensing applications

ere obtained by adding a small amount of metal-decorated MWCNT to two different types of metal oxides (SnO2 and WO3). The hybrid materialsave been analyzed by means of XPS, TEM and SEM. The gas sensing potential of the fabricated hybrid materials has been tested upon exposureo different hazardous species, specifically NO2, CO, C6H6 and NH3, at low operating temperature.

2007 Elsevier B.V. All rights reserved.

NT;

tegttfwr

oln

eywords: Gas sensors; Metal-decorated MWCNT; Hybrid metal oxides/MWC

. Introduction

Carbon nanotubes (CNT) are receiving nowadays more andore attention from the gas sensor community [1–4]. A special

haracteristic that makes them interesting as active materialsor gas sensors is given by their huge surface area that can bexposed to gases [5]. Experimental reports have shown thatpon exposure to toxic gases such as NO2, NH3, CO, ben-ene or ethanol, the electrical conductance of semiconductingNT changes, even when they are operated at low temperatures

6–8], thus reducing considerably the power consumption of theensing device.

However, the agglomeration of CNT into bundles during theirynthesis appears as a drawback for forming a well-dispersedctive layer. To overcome this inconvenience, a plasma func-

∗ Corresponding author. Tel.: +34 977558502; fax: +34 977559605.E-mail address: eduard.llobet@urv.cat (E. Llobet).

iiosMwv

925-4005/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.snb.2007.11.001

Low temperature operation

ionalization process applied to the CNT has proved to befficient [9]. This treatment gives, furthermore, rise to functionalroups attached to the surface of the nanotubes, which modifieshe CNT-surface reactivity and can further improve gas detec-ion. So far, sensors fabricated with multiwall CNT (MWCNT)unctionalized with oxygen have proved to give good resultshen operated at ambient temperature, above all showing good

esponsiveness to low concentrations of NO2 [8].Anyway, in spite of the observed potential of either untreated

r functionalized MWCNT to detect gases, they show quiteow sensitivities to the hazardous species detected, that can-ot be improved even if the sensor operating temperature isncreased. In order to overcome this drawback, we consideredt worthy to investigate the approach of doping the carbon nan-tubes with metallic nanoclusters, and we recently reported a

ignificant improvement of sensitivity to NO2 when employing

WCNT decorated with Au or Ag nanoclusters as comparedith the response obtained by un-doped MWCNT [10]. A pre-ious functionalization process of the carbon nanotubes in an

Actu

otc

sstsaioo[

tlhaibgcsvr

cmigoq

sm(

2

2

Mubl

aspnnpttpi

wf

oeawdbc

2

ammimatrawdeBI

dtifgtlteot2

i

2

i(Emsp

R. Ionescu et al. / Sensors and

xygen plasma provided a more homogeneous distribution ofhe metal nanoclusters on the CNT surface as compared withluster dispersion on non-treated CNT [11].

On the other hand, metal oxides are well-known materialsuitable for detecting a wide spectrum of gases with enoughensitivity. Among these materials, SnO2 and WO3 have provedo be very suitable candidates, but when the detection of toxicpecies is devised they typically need to be operated at temper-tures ranging between 200 and 500 ◦C [12,13]. Furthermore,t is generally known that, in practice, the sensitivity of metalxide gas sensors can be enhanced by using bulk dopantsr by the addition of metal clusters to the sensing material14].

When metal oxide sensors are operated at high tempera-ures, changes in the microstructure of the gas-sensitive film areikely to occur (i.e. structural changes or coalescence [15]). Atigher temperatures, the mobility of oxygen vacancies becomesppreciable and the mechanism of conduction becomes mixedonic–electronic. The diffusion of oxygen vacancies is known toe a mechanism responsible for long-term drift in metal oxideas sensors [16,17]. Therefore, a strategy to avoid long-termhanges in their response could consist in operating the sen-ors at temperatures low enough so that appreciable structuralariation never occurs, provided that gas reactions occur at aeasonable rate.

Recently, hybrid films based on tin or tungsten oxide andarbon nanotubes have been introduced as new gas-sensitiveaterials with improved sensitivity [10,18–21]. These works

ndicated that the detection at ambient temperature of toxicases such as nitrogen dioxide, carbon monoxide and ammoniar ethanol vapors can be improved by dispersing an adequateuantity of carbon nanotubes into a metal oxide matrix.

In this paper, we study and discuss the performance in gasensing of hybrid materials consisting of Au or Ag decoratedultiwall carbon nanotubes dispersed in a metal oxide matrix

either SnO2 or WO3 is considered for this purpose).

. Experimental

.1. Carbon nanotubes functionalization and doping

The MWCNT used in the experiment were obtained fromercorp [22]. They were synthesized by arc discharge without

se of catalysts. The MWCNT powder presents 99% of car-on with 30–40% nanotube content. They have 8–30 grapheneayers, are 6–20 nm in diameter and 1–5 �m in length.

A uniform functionalization with oxygen was applied to thes-provided carbon nanotubes in order to improve their disper-ion and surface reactivity. For this activation step the MWCNTowder was placed inside a glass vessel and a magnet, exter-ally controlled from the plasma chamber, was used to stir theanotubes powder during the plasma treatment. Inductively cou-led plasma at a RF frequency of 13.56 MHz was used during

he process [23]. Once the MWCNT powder was placed insidehe plasma glow discharge, the treatment was performed at aressure of 0.1 Torr, using a power of 15 W, while the process-ng time was adjusted to 1 min. A controlled flow of oxygen

tm(e

ators B 131 (2008) 174–182 175

as introduced inside the chamber, which gave rise to oxygenunctional groups grafted at the carbon nanotube surface.

In the second processing step, the oxygen functionalized nan-tubes were decorated with metal nanoclusters by thermallyvaporating gold or silver atoms onto the MWCNT surface fromgold or silver wire, respectively. The processing parametersere adjusted to be sufficient enough in order to obtain a fairispersion of few metal nanoclusters decorating the MWCNT,ut at the same time to avoid the formation of a metallic layerovering the carbon nanotubes [11].

.2. Active layers

Sensing layers were prepared using commercially avail-ble SnO2 and WO3 nanopowders (Sigma–Aldrich). Hybridaterials were obtained by adding two different amounts ofetal-decorated MWCNT to 70 mg of metal oxide, thus obtain-

ng two different proportions of MWCNT embedded into theetal oxide matrix (1/500 and 1/250 wt%, respectively). An

dequate mixture of the components was obtained by dissolvinghem in glycerol (employed as organic vehicle), and stirring theesulting solution in an ultrasonic bath at 75 ◦C for 2 h. Themount of nanotubes to be added to the metal oxide matrixas based on a previous study [20]. The pastes obtained wereropped onto the electrode area of micro-hotplate transduc-rs (fabricated at the Centre Nacional de Microelectronica,arcelona, Spain) using a micro-injector (JBE1113 Dispenser,

&J FISNAR Inc., USA).The as-deposited sensing films were firstly dried at 140 ◦C

uring 2 h in order to burn out the organic vehicle, using a slowemperature ramp of 2.5 ◦C/min for reaching this temperaturen order to avoid the occurrence of cracks in the films. We usedor this process a lower temperature than the boiling point oflycerol (i.e., 182 ◦C [24]) in order to avoid producing cracks inhe films deposited. The drying time was however sufficientlyong for the complete evaporation of the organic vehicle. Finally,he films were annealed in situ at 450 ◦C during 3 h in ambi-nt atmosphere (this also ensures the complete removal of therganic vehicle). During the annealing process, the tempera-ure was raised from ambient to 450 ◦C using again a ramp of.5 ◦C/min.

The mean grain size of the metal oxides particles, determinedn a previous study, was near 40 nm [25].

.3. Material characterization

The chemical composition of the samples’ surfaces wasnvestigated by means of X-ray photoelectron spectroscopyXPS). XPS analyses were performed using an ESCA-300 (SCI-NTA, Sweden) photoelectron spectrometer equipped with aonochromatized Al K� = 1486.7 eV. A high-resolution hemi-

pherical electrostatic analyzer of 600 mm diameter and 75 eVass energy was used. The angle between the incident X-ray and

he photoelectron take-off direction was 45◦, with the latter nor-

al to the sample surface. The overall resolution of the systemsource + analyzer) was 0.6 eV. The background pressure duringxperiment was better than 10−9 Torr).

1 Actuators B 131 (2008) 174–182

CsCpesi

mnpovo(

2

rsioapttsafdtotf

siiAttasutir

tprrpd[

FM

3

3

r2trbft

csaptpAciessfno

bsca

76 R. Ionescu et al. / Sensors and

The size of the metal clusters and their dispersion on theNT walls were studied by means of high-resolution transmis-

ion electron microscopy (HRTEM) carried out using a PhilipsM30 FEG instrument operated at 300 kV. In order to reduceotential knock-on radiation damage caused by the 300 keVlectron beam, the electron dose was significantly decreaseduch that during the entire electron-beam exposure no changesn the nanotubes were observable.

The morphology of the hybrid films deposited onto theicrohotplate substrates was investigated by means of scan-

ing electron microscopy (SEM). SEM measurements wereerformed using a Joel JSM 6400 equipment, with a resolutionf 0.3 nm. The magnification during this study was set to valuesarying between 30,000 and 50,000. Accelerating voltages of 25r 30 kV were employed. The system allows for sample rotation360◦) and sample inclination (90◦).

.4. Gas sensing measurements

The gas sensing properties of the different hybrid active mate-ials produced were tested in the presence of different hazardouspecies, such as CO (carbon monoxide), NO2 (nitrogen diox-de), NH3 (ammonia) and C6H6 (benzene). The sensors wereperated at three different temperatures: 25 (i.e., ambient), 150nd 250 ◦C. To perform the measurements, the gas sensors werelaced inside a 5.3 dm3 test chamber, and the desired concen-rations of each contaminant under study were introduced byhe direct injection method using a gas-tight chromatographicyringe. Specifically, these concentrations were: 100, 200, 500nd 1000 ppb for NO2; 10 and 50 ppm for CO; 2, 5 and 10 ppmor NH3 and 50 and 150 ppm for C6H6. To assess the repro-ucibility of the results, each measurement was replicated 4imes. An Agilent 34970A multimeter was used for continu-usly monitoring the electrical resistance of the sensors duringhe measurement process. The data acquired were stored in a PCor further analysis.

The measurement process was as follows (identical for anypecies tested): data acquisition started 10 min before injectingnto the measurement chamber the required volume correspond-ng to the lowest concentration of the contaminant measured.fter reaching a steady state, a new amount of the same con-

aminant was injected, so that the second concentration to beested was reached. Successive injections were repeated untilll desired concentrations of the gas were measured. After eacheries of successive injections, the sensor chamber was flushedsing pure dry air for 2 h, which ensured the cleaning of bothhe chamber and the sensor surface. Finally, the airflow wasnterrupted and the sensors were left to recover their baselineesistance before performing a new set of measurements.

In addition to the previously mentioned measurements, con-rol gas sensing experiments were carried out employing bothure WO3 and pure SnO2 materials. Furthermore, we reportedecently gas sensing results obtained with Au and Ag deco-

ated MWCNT [10]. The present results are also compared withrevious works performed under the same experimental con-itions with WO3 and SnO2 films activated with Au and Ag25–27].

n“wp

ig. 1. XPS spectra recorded before and after oxygen functionalization ofWCNT.

. Results and discussion

.1. Characterization

Fig. 1 shows the XPS survey spectra of the carbon nanotubesecorded before and after the plasma treatment. The peak at84.6 eV, observed in both spectra, is generated by photoelec-rons emitted from the C 1s core level. In the XPS spectrumecorded after oxygen plasma treatment a peak near 535 eV cane observed. This peak is generated by photoelectrons emittedrom the O 1s core level. The relative atomic oxygen concentra-ion evaluated by XPS was found to be 6%.

Fig. 2 shows the XPS spectra recorded on the metal-decoratedarbon nanotubes. After the evaporation of the metals, the XPSurvey spectra showed, in addition to the mean peak situatedt 284.6 eV generated by photoelectrons emitted from the C 1seak, a doublet peak near 85.0 eV that was generated by pho-oelectrons emitted from the Au 4f core levels (Fig. 2a), or oneeak at 370.0 eV generated by photoelectrons emitted from theg 3d core level (Fig. 2b). The interaction between the metal

lusters and the CNT surface can be studied by XPS. If theres a chemical reaction at the interface, then the new chemicalnvironment of the atoms at the interface will show in the XPSpectra by the appearance of new features. In the core levelpectra recorded (Fig. 2, inset) on these samples no additionaleatures beyond the Ag 3d and Au 4f doublet were observed, thuso chemical reaction between the carbon and gold or silver atomsccurs.

The relative atomic concentration of each element, evaluatedy XPS after plasma functionalization and metal evaporation, isummarized in Table 1. The decrease in the oxygen atomic con-entration can be associated to the increase in the CNT surfacerea that is covered by gold atoms. This reduces the effective

umber of C 1s and O 1s photoelectrons passing through thegold or silver overlayer” and contributing to the XPS spectrum,hich explains the decrease in the curve for the total number ofhotoelectrons emitted.

R. Ionescu et al. / Sensors and Actuators B 131 (2008) 174–182 177

F(

thCgscf

e

TR

OptMAA

Fo

aogcloMeshfapfigood dispersion of the metal-decorated MWCNT in the glyc-erol solution affects the decoration of MWCNT, removing someof the clusters from the CNT surface. From Fig. 4 it can be seen

ig. 2. XPS spectra recorded on metal-decorated MWCNT: (a) Au-decorated;b) Ag-decorated.

Fig. 3 shows a typical TEM image recorded on oxygen plasmareated MWCNT decorated with gold clusters. The presence ofomogeneously dispersed quasi-spherical gold clusters at theNT surface can be observed. In the inset, a detailed view of aold cluster sitting on the CNT surface is shown. The preservedtructural characteristics of the graphene layer under the goldluster are a strong indication of the absence of an Au–C phase

ormation.

Similar results (not shown here) were obtained for Ag clustersvaporated on the CNT surface.

able 1elative atomic concentrations obtained by XPS

[C] (%) [O] (%) [Ag] (%) [Au] (%)

xygenlasmareated

WCNTs

94.0 6.0

u-MWCNTs 92.0 2.7 5.3g-MWCNTs 95.0 3.2 1.8 F

c

ig. 3. TEM image recorded on MWCNT decorated with gold metallic nan-clusters deposited by thermal evaporation.

Considering that the interaction between Au or Ag atomsnd the CNT surface was reported to be weak and presumablyf Van der Waals in nature [11], an inspection to evaluate if theas sensing layer preparation method employed (i.e., the soni-ation process applied for obtaining a good mixture of the activeayer components) could affect the dispersion of the metal nan-clusters at the CNT surface was performed. For this evaluation,WCNTs decorated with metal nanoclusters dispersed in glyc-

rol were analyzed. In this case no metal oxide was used and theame procedure as the one employed for the fabrication of theybrid materials (i.e., stirring the solution in an ultrasonic bathor 2 h at 75 ◦C) was implemented. Fig. 4 shows a TEM image ofmetal-decorated nanotube having undergone this process. Theresence of metal clusters laying isolated within the depositedlm reveals that the stirring process implemented to obtain a

ig. 4. TEM image recorded on Au-decorated MWCNT after a sonication pro-ess performed during the preparation of the sensing materials.

178 R. Ionescu et al. / Sensors and Actuators B 131 (2008) 174–182

-MW

tti

sd

Hi

TR

F

WSAAAAAAAAAA

F

WSAAAAAAAAAA

Fig. 5. SEM images recorded on different hybrid films: (a) Ag

hat mainly big clusters are removed from the CNT surface. Fur-her studies will be performed to estimate if the cluster removal

s size-selective and if clusters are removed intact or only in part.

It is worth noting that the clusters removed from the CNTurface during the fabrication of hybrid metal oxides/metal-ecorated CNT will be embedded in the metal oxide matrix.

mrtt

able 2esponsiveness to NO2 and CO of the different sensors studied as calculated from E

ilm Layer NO2 (250 ◦C) NO

0.1 ppm 0.5 ppm 1 ppm 0.1

O3 1.98 5.73 7.14 0.1nO2 0.81 11.04 15.72 1.0u-MWCNT/WO3 (1:500) 1.39 8.96 21.44 0.3u-MWCNT/WO3 (1:250) 0 0 0 −0.0g-MWCNT/WO3 (1:500) 1.40 8.00 24.44 0.2g-MWCNT/WO3 (1:250) 0 0 0 −0.0u-MWCNT/SnO2 (1:500) 47.36 238.56 243.92 10.6u-MWCNT/SnO2 (1:250) 73.21 471.21 485.83 30.2g-MWCNT/SnO2 (1:500) 15.24 72.02 96.69 10.3g-MWCNT/SnO2 (1:250) 19.30 98.59 135.79 9.0u-MWCNT 0 −0.01 −0.02 0g-MWCNT 0 −0.02 −0.03 −0.0

ilm Layer CO (250 ◦C)

10 ppm 50 ppm

O3 0.73 0.79nO2 0.07 0.25u-MWCNT/WO3 (1:500) 0 0u-MWCNT/WO3 (1:250) 0 0g-MWCNT/WO3 (1:500) 2.34 16.11g-MWCNT/WO3 (1:250) 0.06 0.05u-MWCNT/SnO2 (1:500) 9.32 31.10u-MWCNT/SnO2 (1:250) 25.05 74.59g-MWCNT/SnO2 (1:500) 15.03 36.76g-MWCNT/SnO2 (1:250) 14.39 45.18u-MWCNT 0 0g-MWCNT 0 0

CNT/WO3 (1/500 wt%); (b) Ag-MWCNT/SnO2 (1/250 wt%).

owever, improvements in the sensing properties (e.g., enhanc-ng their sensitivity, decreasing the operating temperature or

aking them more selective to a given target species) have beeneported to be achieved by adding small amounts of noble metalso the metal oxide active layer [28]. Thus, it can be suggestedhat besides the presence of CNTs, the presence of noble metal

q. (1)

2 (150 ◦C) NO2 (25 ◦C)

ppm 0.5 ppm 1 ppm 0.1 ppm 0.5 ppm 1 ppm

9 0.41 0.50 0 0 01 4.47 4.66 0 0.10 0.105 1.66 2.07 0 0 03 −0.06 −0.08 −0.03 −0.07 −0.089 1.51 2.78 0 0 01 −0.04 −0.05 −0.03 −0.07 −0.097 32.34 32.34 0.96 0.98 0.989 88.91 88.91 1.14 1.14 1.144 31.40 31.40 1.04 2.64 2.691 16.42 16.42 0.35 0.35 0.35

−0.04 −0.07 −0.01 −0.07 −0.091 −0.03 −0.06 −0.01 −0.04 −0.07

CO (150 ◦C) CO (25 ◦C)

10 ppm 50 ppm 10 ppm 50 ppm

0.27 0.72 0 00.14 0.34 0 00 0 0 00 0 0 00.48 2.51 0 00 0 −0.02 −0.044.37 7.50 0 0

11.29 17.52 0 010.12 16.10 0 010.96 16.44 0 00 0 0 00 0 0 0

Actuators B 131 (2008) 174–182 179

cs

csaotodtbt

3

petwiaaoi

S

waSbMflcrtiwo1dwtititSaShbmo

Fig. 6. Response to different hazardous species obtained by gas sensorseAM

to

mmTtoamtma

thIls1

R. Ionescu et al. / Sensors and

lusters inside the metal oxide matrices can act improving theensing properties.

Fig. 5 shows SEM images recorded on the different hybridarbon nanotubes/metal oxide sensing films deposited over theensors substrates. The micrographs show the presence of WO3nd SnO2 metal oxide grains, whereas carbon nanotubes arebserved only in WO3/CNT hybrids. This can be associated tohe low proportion of carbon nanotubes embedded in the metalxide matrix (1/500 or 1/250 wt%) as well as to the difference inensity between WO3 and SnO2 (7.16 and 6.95 g/cm3, respec-ively [29]); in order to obtain the desired weight proportionsetween the metal oxide and CNTs a higher amount of SnO2han WO3 is present in the hybrid materials.

.2. Gas response analysis

The gas sensing properties of the different hybrid materialsroduced were evaluated in terms of the ratio between the changexperienced by the sensor resistance after its exposure to pollu-ant species and the sensor baseline resistance in air (see Eq. (1),here S defines sensor’s responsiveness, Rair sensor’s resistance

n air and Rgas sensor’s resistance in the presence of the pollutantfter reaching a steady state). Responsiveness results for NO2nd CO are summarized in Table 2; because no response wasbtained to NH3 and C6H6, for space reasons these were notncluded in the responsiveness table.

= Rair − Rgas

Rair(1)

Sensors based on SnO2 and Au-decorated MWCNT hybridsere the most sensitive to NO2 among the different materi-

ls studied, outperforming the responsiveness of either purenO2 or pure Au-decorated MWCNT materials when operatingoth at 250 and 150 ◦C. Typical responses of metal-decoratedWCNT/SnO2 gas sensors to NO2 are shown in Fig. 6. The

uctuation in the response signal that occurs at high NO2 con-entrations is due to the increased effect of noise when sensoresistance becomes very high (i.e. comparable in magnitude tohe input impedance of the multimeter employed to acquiret). The quantity of nanotubes dispersed in the SnO2 matrixas found to play a determinant role in the responsivenessf the hybrid materials to NO2. At a concentration ratio of/250 wt%, the responsiveness of the hybrids made of metal-ecorated MWCNT (either using Au or Ag as dopants) and SnO2as significantly superior to that with the 1/500 wt% ratio when

he detection of NO2 at 250 ◦C was aimed at. When the operat-ng temperature of sensors was lowered to 150 ◦C, the particularype of metal decorating the carbon nanotubes played also anmportant role in the NO2 detection. Thus, at 150 ◦C a concen-ration ratio of 1/250 wt% of carbon nanotubes dispersed in thenO2 matrix was the most appropriate when Au was used asdopant, while 1/500 wt% of Ag-decorated MWCNT added tonO2 was the most suitable for this latter case. On the other

and, similar values of responsiveness were found at 150 ◦C foroth Au-MWCNT/SnO2 and Ag-MWCNT/SnO2 (1/500 wt%)aterials. Furthermore, it is worth mentioning that the response

f the hybrid films based on SnO2 became already saturated after

ttra

mploying different hybrid materials: (a) NO2 detection at 250 ◦C withu-MWCNT/SnO2 (1/250 wt%) sensor; (b) NO2 detection at 150 ◦C with Au-WCNT/SnO2 (1/250 wt%).

he injection of only 500 ppb of NO2 at the working temperaturef 150 ◦C (see Fig. 6b).

Regarding the responsiveness towards NO2 of the hybridaterials based on WO3 (see Fig. 7), it was at least one order ofagnitude below the one obtained by the hybrids based on SnO2.he quantity of nanotubes embedded in WO3 was of impor-

ant relevance. Thus, when the sensors were operated at 250 ◦C,nly metal-decorated MWCNT dispersed in the WO3 matrix inconcentration ratio of 1/500 wt% were able to detect NO2 (theetal used as a dopant did not change sensor performance in

his case), while at 150 ◦C the semiconducting behaviour of theetal-decorated MWCNT/WO3 hybrids changed from n-type

t a concentration ratio of 1/500 wt% to p-type at 1/250 wt%.For the measurements performed at 250 ◦C, the response time

o 100 ppb and to 500 ppb of NO2 was around 2 min both for theybrid materials based on WO3 and for the pure WO3 sensors.t rose to 3 min in the case of hybrids based on SnO2, but in thisatter case it compared very favourably to the one of pure SnO2ensors, which did not reach completely a steady state regime5 min after gas injection. When the sensor operating tempera-

ure was lowered and NO2 test measurements were performed,he sensors needed a longer time to reach the steady state. Theesponse times varied between 6 and 10 min for hybrid sensorsnd were over 15 min for pure metal oxide sensors. Recovery

180 R. Ionescu et al. / Sensors and Actu

F(

t(m

haa2a

F

Sbaetttortiibnta

rtm

pttt

3

Sdiaifims

ig. 7. (a) NO2 detection at 150 ◦C with Ag-MWCNT/WO3 (1/500 wt%) sensor;b) NO2 detection at 150 ◦C with Ag-MWCNT/WO3 (1/250 wt%) sensor.

ime varied between 10 and 20 min for Au-based hybrid sensorssee Fig. 6) and was over 30 min for Ag-based hybrids and pureetal oxide sensors.The second air pollutant tested was carbon monoxide. The

ighest responsiveness in the case of the CO tests was again

chieved by the hybrid sensors based on Au-decorated MWCNTnd SnO2 in a concentration ratio of 1/250 wt%, operated at50 ◦C (see Fig. 8). Although lower, some responsiveness waslso obtained at 150 ◦C by the hybrid sensors containing SnO2.

ig. 8. CO detection at 250 ◦C with Au-MWCNT/SnO2 (1/250 wt%) sensor.

cnfFnittst

imisisdoo

ators B 131 (2008) 174–182

imilarly to the detection of NO2, when CO was tested Au-ased hybrid sensors showed response and recovery times ofbout 5 min (operated at 250 ◦C), while the response and recov-ry times of pure metal oxides was higher than 15 min. Whenhe metal oxide employed was WO3, the only hybrid based onhis material that responded to CO was Ag-MWCNT/WO3 inhe concentration ratio of 1/500 wt% at an operating temperaturef 250 ◦C. When the Ag-MWCNT/WO3 sensor (concentrationatio 1/250 wt%) was operated at 250 ◦C, it behaved as an n-ype semiconductor in the presence of CO; operated at 150 ◦Ct did not respond at all to CO; while at room temperaturets behaviour was equivalent to an n-type semiconductor. Thisehaviour clearly suggests that not only the amount of carbonanotubes present determines the semiconducting character ofhe resulting hybrid material, but also that the operating temper-ture can play an important role in the sensing mechanism.

Unlike oxygen-functionalized MWCNT/SnO2 hybrid mate-ials, which showed good responsiveness to NO2 at roomemperature [30], metal-decorated MWCNT/metal oxide hybrid

aterials were not responsive at room temperature.Regarding the response of the gas sensors to the other two

ollutants tested (i.e., benzene and ammonia), they were not ableo detect the presence of these two contaminants at a concentra-ion level up to 10 ppm in the case of NH3 and up to 150 ppm inhe case of C6H6, for the operating temperatures investigated.

.3. Discussion

On the basis of the images of the hybrid films recorded byEM analyses, it can be derived that MWCNT are embed-ed within the metal oxide matrix. It has been reported thatn hybrid films, two different depletion layers (and associ-ted potential barriers) co-exist [18,19]: one depletion layers located at the surface of the grains of the metal oxidelm and the other one at the interface between MWCNT andetal oxide films. Since SnO2 or WO3 films behave as n-type

emiconductors and MWCNT films behave as p-type semi-onductors [8,31], it can be suggested that the hetero-structure-SnO2/p-MWCNT (n-WO3/p-MWCNT) is formed at the inter-ace between tin oxide (tungsten oxide) and carbon nanotubes.urthermore, our results indicate that the addition of metalanoclusters at the CNT surface plays a fundamental role inmproving the sensing properties. Studies are being performedo establish if the metal clusters at the CNT surface act loweringhe potential barrier of the depletion layers and/or enhancingpecific gas adsorptions or promoting specific chemical reac-ions.

Considering the sensitivity of the hybrid films (see Table 2),t can be derived that the adsorption of NO2 or CO at the

etal oxide modifies the depletion layer at the surface ofts grains and also at the n-metal oxide/p-MWCNT hetero-tructures. This combined effect may explain the improvementn responsiveness shown by tin or tungsten oxide-based hybrid

ensors as compared with either pure metal oxides, metal-ecorated CNT based gas sensors [10] or WO3 or SnO2 metalxides doped with Au or Ag noble metals [25–27]. The resultsbtained indicate also that the number of CNT added to the

Actu

mweSllib

noaas

4

qns

hhmse

citnamc

datess

A

sTasfoIEn

R

[

[

[

[

[

[

[

[

[

[

[

[

R. Ionescu et al. / Sensors and

etal oxide matrix has to be extremely small. The best resultsere obtained with the SnO2/CNT hybrids, when the pres-

nce of the carbon nanotubes was undetectable by normalEM analyses. This is in concordance with the results pub-

ished by Wei et al. [18]. We expect that keeping extremelyow the number of CNT added to the metal oxide matrix,mproved results in terms of sensitivity to NO2 and CO cane obtained.

The lack of responsiveness observed for NH3 and C6H6 by theew hybrid sensors can be associated to the non-appropriatenessf the used combination of materials for detecting such species,s suggested by Penza et al. [32] who found that Au-CNT is anppropriate material for detecting NO2 while Pt-CNT is moreuitable for detecting benzene or ammonia.

. Conclusions

In this paper we have shown that the addition of a smalluantity of metal-decorated MWCNT to metal oxides can sig-ificantly improve the detection capability of metal oxide basedensors and lower the operating temperature.

In particular, micro-sensors based on Au-MWCNT/SnO2ybrid films in a concentration ratio of 1/250 wt% showed theighest sensitivity towards NO2 and CO, among the differentaterials studied. The response mechanism is fully reversible,

ince the sensors can recover their baseline resistance after eachxposure to pollutant gases.

Our results suggest that there is an optimum amount ofarbon nanotubes to be added to each particular metal oxiden order to enhance the responsiveness. Material characteriza-ion analyses (performed by SEM and TEM) showed that theanotubes endured the process of deposition and subsequentnnealing at 450 ◦C in air, but at the same time part of theetal nanoclusters decorating the nanotube surface were deta-

hed.Based on these results, the modulation of the width of two

epletion layers existing at the surface of metal oxide grainsnd at the interface of metal oxide grains and MWCNT, respec-ively, is postulated as the mechanism that could explain thenhanced performance of hybrid metal oxide/MWCNT sen-ors in comparison with pure metal oxide or pure MWCNTensors.

cknowledgements

This work was funded in part by the Spanish Commis-ion for Science and Technology (CICYT) under grant no.IC2006-03671/MIC. E.H. Espinosa and R. Leghrib gratefullycknowledge Ph.D. studentships from Rovira i Virgili Univer-

ity. R. Ionescu holds a ‘Juan de la Cierva’ research fellowshipunded by the Spanish Ministry for Science and Education. Partsf this work are directly connected to the Belgian Program onnteruniversity Attraction Poles (PAI5/1/1) on Quantum Sizeffects in Nanostructured Materials, sponsored by the Commu-aute Francaise de Belgique.

[[

[

ators B 131 (2008) 174–182 181

eferences

[1] S. Chopra, A.M. Rao, Selective gas detection using a carbon nanotubesensor, Appl. Phys. Lett. 83 (2003) 2280–2282.

[2] C. Cantalini, L. Valentini, I. Armentano, L. Lozzi, J.M. Kenny, S. San-tucci, Sensitivity to NO2 and cross-sensitivity analysis to NH3, ethanoland humidity of carbon nanotubes thin film prepared by PECVD, Sens.Actuators B 95 (2003) 195–202.

[3] J. Li, Y. Lu, Q. Ye, M. Cinke, J. Han, M. Meyyappan, Carbon nanotubesensors for gas and organic vapor detection, Nano Lett. 3 (2003) 929–933.

[4] Y. Lu, C. Partridge, M. Meyyappan, J. Li, A carbon nanotube sensor arrayfor sensitive gas discrimination using principal component analysis, J.Electroanal. Chem. 593 (2006) 105–110.

[5] O.K. Varghese, P.D. Kichambre, D. Gong, K.G. Ong, E.C. Dickey, C.A.Grimes, Gas sensing characteristics of multi-wall carbon nanotubes, Sens.Actuators B 81 (2001) 32–41.

[6] L. Valentini, C. Cantalini, I. Armentano, J.M. Kenny, L. Lozzi, S. Santucci,Highly sensitive and selective sensors based on carbon nanotubes thin filmsfor molecular detection, Diamond Relat. Mater. 13 (2004) 1301–1305.

[7] S.G. Wang, Q. Zhang, D.J. Yang, P.J. Sellin, G.F. Zhong, Multi-walled car-bon nanotube-based gas sensor for NH3 detection, Diamond Relat. Mater.13 (2004) 1327–1332.

[8] R. Ionescu, E.H. Espinosa, E. Sotter, E. Llobet, X. Vilanova, X. Correig,A. Felten, C. Bittencourt, G. Van Lier, J.-C. Charlier, J.J. Pireaux, Oxygenfunctionalisation of MWNT and their use as gas sensitive thick-film layers,Sens. Actuators B 113 (2006) 36–46.

[9] N.O.V. Plank, R. Cheung, Functionalisation of carbon nanotubes for molec-ular electronics, Microelectron. Eng. 73/74 (2004) 578–582.

10] E.H. Espinosa, R. Ionescu, C. Bittencourt, A. Felten, R. Erni, G. Vantende-loo, J.-J. Pireaux, E. Llobet, Metal-decorated multi-wall carbon nanotubesfor low temperature gas sensing, Thin Solid Films 515 (2007) 8322–8327.

11] A. Felten, C. Bittencourt, J.-F. Colomer, G. Van Tendeloo, J.-J. Pireaux,Nucleation of metal clusters on plasma treated multi wall carbon nanotubes,Carbon 45 (2007) 110–116.

12] I. Simon, N. Barsan, M. Bauer, U. Weimar, Micromachined metal oxidegas sensors: opportunities to improve sensor performance, Sens. ActuatorsB 73 (2001) 1–26.

13] M. Graf, D. Barrettino, K.-U. Kirstein, A. Hierlemann, CMOS micro-hotplate sensor system for operating temperatures up to 500 ◦C, Sens.Actuators B 117 (2006) 346–352.

14] W. Gopel, New materials and transducers for chemical sensors, Sens. Actu-ators B 18 (1994) 1–21.

15] E. Comino, Metal oxide next term nano-crystals for gas sensing, Anal.Chim. Acta 568 (2006) 28–40.

16] P.K. Clifford, D.T. Tuma, Characteristics of semiconductor gas sensors. 1.Steady-state gas response, Sens. Actuators 3 (1983) 233–254.

17] P.K. Clifford, D.T. Tuma, Characteristics of semiconductor gas sensors.2.Transient-response to temperature-change, Sens. Actuators 3 (1983)255–281.

18] B.Y. Wei, M.C. Hsu, P.G. Su, H.M. Lin, R.J. Wu, H.J. Lai, A novel SnO2

gas sensor doped with carbon nanotubes operating at room temperature,Sens. Actuators B 101 (2004) 81–89.

19] Y. Chen, C. Zhu, T. Wang, The enhanced ethanol sensing properties ofmulti-walled carbon nanotubes/SnO2 core/shell nanostructures, Nanotech-nology 17 (2006) 3012–3017.

20] C. Bittencourt, A. Felten, E.H. Espinosa, R. Ionescu, E. Llobet, X. Correig,J.-J. Pireaux, WO3 films modified with functionalised multi-wall carbonnanotubes: Morphological, compositional and gas response studies, Sens.Actuators B 115 (2006) 33–41.

21] E.H. Espinosa, R. Ionescu, E. Llobet, A. Felten, C. Bittencourt, E. Sotter,Z. Topalian, P. Heszler, C.G. Granqvist, J.J. Pireaux, X. Correig, Highlyselective NO2 gas sensors made of MWCNTs and WO3 hybrid layers, J.Electrochem. Soc. 154 (2007) J141–J149.

22] http://www.mercorp.com.23] C. Lefebvre, J. Verbist, Characterization of acrylonitrile plasma-

polymerized on carbon fibers, Adv. Comp. Mat. Lett. 1 (1992) 34–37.24] http://www.sigmaaldrich.com/catalog/search/ProductDetail/FLUKA

/49770?LastFive.

1 Actu

[

[

[

[

[[

[

[

B

RESn

Ea

rn

Rarm

AUt

JNmc

Rmren

GBcpa

CHot

82 R. Ionescu et al. / Sensors and

25] P. Ivanov, E. Llobet, F. Blanco, A. Vergara, X. Vilanova, I. Gracia, C. Cane,X. Correig, On the effects of the materials and the noble metal additives toNO2 detection, Sens. Actuators B 118 (2006) 311–317.

26] M. Stankova, X. Vilanova, J. Calderer, E. Llobet, J. Brezmes, I. Gracia, C.Cane, X. Correig, Sensitivity and selectivity improvement of rf sputteredWO3 microhotplate gas sensors, Sens. Actuators B 113 (2006) 241–248.

27] M. Penza, G. Cassano, F. Tortorella, Gas recognition by activated WO3

thin-film sensors array, Sens. Actuators B 81 (2001) 115–121.28] S. Santucci, C. Cantalini, M. Crivellari, L. Lozzi, L. Ottaviano, M. Pas-

sacantando, X-ray photoemission spectroscopy and scanning tunnelingspectroscopy study on the thermal stability of WO3 thin films, J. Vac. Sci.Tech. A 18 (2000) 1077–1082.

29] http://www.sigmaaldrich.com.30] E.H. Espinosa, R. Ionescu, B. Chambon, G. Bedis, E. Sotter, C.

Bittencourt, A. Felten, J.-J. Pireaux, X. Correig, E. Llobet, Low tem-perature gas detection with hybrid metal oxides/MWCNTs, in: XXEurosensors Conference Aniversary, Goteborg, Sweden, September 17–20,2006.

31] L. Valentini, I. Armentano, J.M. Kenny, C. Cantalini, L. Lozzi, S. Santucci,Sensors for sub-ppm NO2 gas detection based on carbon nanotube thinfilms, Appl. Phys. Lett. 82 (2003) 961–963.

32] M. Penza, G. Cassano, R. Rossi, M. Alvisi, A. Rizzo, M.A. Signore, Th.Dikonimos, E. Serra, R. Giorgi, Enhancement of sensitivity in gas chemire-sistors based on carbon nanotube surface functionalized with noble metals(Au, Pt) nanoclusters, Appl. Phys. Lett. 90 (2007) 173–176.

iographies

adu Ionescu is a postdoctoral research fellow at the Department of Electronics,lectrical and Automatic Engineering, Rovira i Virgili University, Tarragona,

pain. His main research interests are in the field of chemical gas sensors, carbonanotubes and pattern recognition.

dwin Espinosa is a PhD student at the Department of Electronics, Electricalnd Automatic Engineering, Rovira i Virgili University, Tarragona, Spain. His

EVms

ators B 131 (2008) 174–182

esearch topic consists of gas sensors based on plasma functionalised carbonanotubes.

adouane Leghrib is a PhD student at the Department of Electronics, Electricalnd Automatic Engineering, Rovira i Virgili University, Tarragona, Spain. Hisesearch topic consists of gas sensors based on carbon nanotubes decorated withetal nanoclusters.

lexandre Felten is a postdoctoral research fellow at the LISE laboratory,niversity of Namur, Belgium. One of his research interests is in the func-

ionalisation of carbon nanotubes using cold plasmas.

ean Jacques Pireaux is the director of the LISE laboratory, University ofamur, Belgium. Professor Pireaux leads a project on the interface design ofetal nanocluster-carbon nanotube hybrids via control of structural and chemi-

al defects in a plasma discharge.

olf Erni is a researcher at the EMAT, University of Antwerp, Belgium. Hisain areas of interest are in Electron scattering and diffraction physics, aber-

ation correction in TEM and STEM, low-loss and high-resolution electronnergy-loss spectroscopy, functional materials such as optically active materials,anomaterials such as quantum dots, quantum well structures and nanoparticles.

ustav Van Tendeloo is professor of physics at the University of Antwerp,elgium. His main research interest are in superconducting and CMR materials,eramic thin films, nanotubes, nanowires, nanobelts, nanoparticles and meso-orous materials, GaN and related semiconductors, solid state phase transitionsnd modulated structures.

arla Bittencourt is a senior researcher at the LCIA, University of Mons-ainaut, Belgium. One of her research interest is in the development of metalxide and carbon nanotube hybrid materials for sensing gases at low operatingemperatures.

duard Llobet is an associate professor of electronics at the University Rovira iirgili (Tarragona, Spain). His main research interests are in the fabrication andodelling of semiconductor gas sensors and in the applications of intelligent

ystems to complex odour analysis.