Simulation studies on the responses of zno cuo cnt nanocomposite based saw sensor to various volatile organic chemicals

This novel ZnO-CuO/CNT SAW sensor combining the sensing properties of metal oxide nanostructures and CNT with improved characteristics can be used as a promising candidate for sensing im

Trang 1

Original Article

Simulation studies on the responses of ZnO-CuO/CNT nanocomposite

based SAW sensor to various volatile organic chemicals

Nelsa Abrahama,b,*, R Reshma Krishnakumara, C Unnib, Daizy Philipc

a Department of ECE, Government Engineering College Barton Hill, Thiruvananthapuram, 695035, India

b Centre for Development of Imaging Technology, Chitranjali Hills, Thiruvallom, Thiruvananthapuram, 695027, India

c Department of Physics, Mar Ivanios College, Thiruvananthapuram, 695015, India

a r t i c l e i n f o

Article history:

Received 16 October 2018

Received in revised form

19 December 2018

Accepted 19 December 2018

Available online 28 December 2018

Keywords:

ZnO

CuO

Surface acoustic waves

Sensors

Carbon nano tubes

a b s t r a c t Surface acoustic wave (SAW) sensors offer a sensitive platform for monitoring important physical entities with several advantages They can operate well in extreme conditions such as high temperature, high pressure and toxic environment This work presents a 2D model of SAW sensor with carbon nano tubes (CNT) as the adsorbent material A second model was also created by incorporating ZnO and CuO nanospheres into the sensing layer The responses of the two sensors towards various gases were ana-lysed at room temperature The design was modelled and anaana-lysed using COMSOL Multiphysics software which applies theﬁnite element analysis to solve for Eigen frequencies The shift in the resonant fre-quencies with and without the presence of gases, which is a measure of sensitivity has been estimated for all the gases The second model showed improved response This novel ZnO-CuO/CNT SAW sensor combining the sensing properties of metal oxide nanostructures and CNT with improved characteristics can be used as a promising candidate for sensing important volatile organic chemicals at room temperature

© 2018 The Authors Publishing services by Elsevier B.V on behalf of Vietnam National University, Hanoi This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/)

1 Introduction

Gas sensing plays a pivotal role in multidisciplinary areas like

in-dustrial, medical, ventilation system design, power plants and

envi-ronmental pollution monitoring Nowadays pollution is a vexing

problem in urban areas, hence air quality monitoring is becoming

vital The current trend in gas sensor development is miniaturization

which provides inexpensive, robust and safe sensors[1,2] In addition,

it helps to achieve multiplexing of sensor arrays as well[3] Catalytic,

optical or electrochemical sensing mechanisms are mostly used for

gas sensors[4e7] High sensitivity, low cost gas sensors working at

room temperature required for a real time detection of toxic gases

Semiconductor type gas sensors were developed in 1962 on the basic

of the resistive changes in the semiconducting metal oxide A wide

variety of gas sensors based on several metal oxides like ZnO, SnO2,

TiO2,Fe2O3and Bi2O3have already been studied[8e11] The general

gas sensing principle is the adsorption and the desorption of analyte

molecules on the sensing material So the sensitivity can be enhanced

by increasing the contact interfaces which are able to achieve through the use of nanomaterials Surface Acoustic Wave (SAW) resonators are a class of Micro-Electromechanical Systems (MEMS) that can also

be used in gas sensing applications[12] For chemical sensing appli-cations they are the most prominent candidates as their resonant frequency ranges from several MHz to GHz, which is much higher than that of quartz crystal microphones (QCM) This wider frequency range makes them more sensitive and opens the possibility of oper-ating in wireless mode They are able to detect analytes at ambient temperatures and can work efﬁciently even in inert atmospheric conditions Besides this, SAW renators are cheap, possess high ther-mal stability, easy to fabricate, highly selective and more efﬁcient compared to conventional gas sensors These special characteristics make them highly suitable as smart transducers which can be com-bined with a variety of sensitive coating layers including metal oxides, carbon nanotubes (CNTs), graphene layers and functional polymers

[13] Thanks to the enhanced absorption characteristics, new mate-rials such as metal organic frameworks and porous matemate-rials are increasingly utilized in theseﬁelds

SAW sensors are basically piezoelectric crystals which can sense and detect the masses of chemical vapours adsorbed on the chemically sensitive coatings In this case, inter digitated trans-ducers (IDTs) are placed on the surface of a piezoelectric substrate

to generate and receive acoustic waves The conﬁnement of the

* Corresponding author Department of ECE, Government Engineering College

Barton Hill, Thiruvananthapuram, 695035, India.

E-mail address: nelsaarun2016@gmail.com (N Abraham).

Peer review under responsibility of Vietnam National University, Hanoi.

Contents lists available atScienceDirect Journal of Science: Advanced Materials and Devices

j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l o c a t e / j s a m d

https://doi.org/10.1016/j.jsamd.2018.12.006

Journal of Science: Advanced Materials and Devices 4 (2019) 125e131

Trang 2

acoustic energy into the near surface region can increase its

sensitivity manifold The area between the IDTs (delay line) is made

sensitive by coating with chemically active species which reacts

with the target molecules Coating SAW sensors with various

nanostructured materials allow gas detection by varying in the

resonant frequency Acoustic resonators are considered as universal

transducers since they detect the mass, which is a primary feature

of any target[14] On combining with suitable molecular layers

they can be well designed for remote sensing applications Metal

oxide semiconductor based gas sensors mostly operate at very high

temperatures which require large power and special packaging

This urges the need for developing new metal oxide sensing layers

that can work safely at room temperature

CNT and its composites owing to their unique physical and

chemical properties have received considerable attention in recent

years[15,16] The properties of CNTs and their composites change

upon exposure to various gases which can be detected by various

methods The electronic properties of CNTs change on interaction

with gas molecules which can be attributed to the charge transfer

between gas molecules and nanotubes This can make them more

sensitive to gases with large binding energies

Many studies have already been conducted on hybrid metal

oxide-CNT sensors namely SnO2, WO3, TiO2 eCNT for room

tem-perature operation[17,18] Herein we report the simulation studies of

two different CNT based SAW gas sensors for the detection of

different volatile organic chemicals (VOCs) We have incorporated a

metal oxide compositae in to the CNT layer to enhance the efﬁciency

Tasaltin et al.[19]studied the 433 MHz Rayleigh wave based SAW

sensor coated with ZnO for these VOCs To the best of our knowledge

no studies have been reported on this particular ternary

nano-composite based SAW gas sensor Even though metal oxide

semi-conductor (MOX) gas sensors have several advantages, their biggest

downsides are problems related to drift and signiﬁcant energy

consumption In addition, a large amount of energy is needed to activate the interaction of the sensing layer with the gas molecules

So the sensing layer of MOX needs to be heated up for long periods Such high temperatures can even alter the structure and properties

of the sensing layer At elevated temperatures, the mobility of oxygen vacancies will become appreciable leading to the mixed ionic-electronic conduction mechanism This type of (oxygen vacancies) diffusion could produce long term drift in MOX sensors[20] Now the prime motive of the researchers is to reduce the power consumption which could prolong the battery life[21] CNTs have been deeply explored by the research world because of their ability to detect gases at ambient temperatures Thus researchers were able to bring down the power consumption to a fewmW[22] CNT based sensors also face few challenges like long response and recovery times which impede them from directly replacing metal oxides in MOX based sensors[23] So it is expected that doping metal oxide with CNT could surely enhance the sensitivity, lower preheating of the work body and also could lower the response/recovery time[24] The aim

of this paper is to model a SAW based sensor with MOX-CNT nano-composite as sensing layer to combine the advanced features of MOX and CNT and also to study the effect of adding ZnO-CuO nano-composites in to the CNT layer in sensitivity enhancement

2 Model design COMSOL Multiphysics is a platform where models can be devel-oped and analysed using Finite Element Analysis (FEA) Modern computer-aided design techniques realized in commercial softwares like ANSYS and COMSOL Multiphysics offer powerful and robust simulation tools for designs which can accurately predict the system performance avoiding the physical prototype fabrication[25]

By metal patterning SAW sensors can be conﬁgured as one port or two port resonators and delay lines Appropriate methods

Model parameters.

Variable Expression Description

C 0 100 Gas concentration ppm

C_gas_air 1e6*C 0 *P/(R- Constant*T) Gas concentration in air

K K CNT/air partition constant

rho_gas_CNT K*M*C_gas_air Mass concentration

of gas in CNT Rho_CNT 1.49 g/cm 3 Density of CNT

E_CNT 350 GPa Young's modulus of CNT

nu_CNT 0.269 Poisson's ratio of CNT

eps_CNT 8 Relative permittivity f CNT

vR 3488 m/s Rayleigh wave velocity

Width 4mm Width of unit cell

f 0 vR/width Estimated SAW frequency

t_CNT 0.5mm CNT thickness

Table 2

Air partition constant and molar mass for different gases.

Gas CNT/air partition constant Molar mass

Trichloromethane 10^4.36 119.38

Dichloromethane 10^4.18 84.93

Frequency shift of various gases for the two different sensors.

Gases Sensing layer

CNT (Df in kHz) ZnO-CuO/CNT (Df in kHz) Trichloromethane 47.82 59.8

Dichloromethane 22.45 33.42

Diethylether 76.19 89.46

Acetonitrile 7.33 10.64 2-Propanol 183.88 200.26

Fig 1 SAW gas sensor, showing the IDT electrodes (in black), the thin PIB ﬁlm (light grey), and the LiNbO 3 substrate (dark grey) A slice of geometry is removed to reveal

Trang 3

have been developed by Tsai et al.[26] to design SAW sensors

based on mass loading principle using FEM To develop a 2D

model of the sensor, the model geometry is reduced to a periodic

unit cell as shown in theFig 1 When a voltage is applied to the

input electrodes, by converse piezoelectric effect mechanical

perturbations are generated on the surface These acoustic waves

will propagate along the surface and on reaching the output

electrodes, by direct piezoelectric effect voltage gets developed

across them The wave velocity depends on many factors like

conductivity, sensing layer thickness as well as the spacer

thick-ness The different model parameters are given inTable 1

In a piezoelectric material the propagation of the wave is

gov-erned by the equation

here T represents the stress matrix, C the elasticity matrix, eTthe

piezoelectric matrix and E represents the electricﬁeld intensity

This formula serves as the basis for building the geometry Even

though there are different variants of acoustic waves (SH-SAW,

love, Lamb and leaky wave), Rayleigh mode is most popular as they

are extremely sensitive to a number of quantities If a gas-phase

analyte of certain concentration is coming in contact with its

surface, the sensing layer will adsorb these molecules until ther-modynamic equilibrium is attained Due to the increased adsorp-tion, the layer becomes denser and heavier As a result, the propagation velocity of the surface wave decreases which leads to frequency downshift (Df¼ f e f0)[25] On the assumption that the analyte forms a non-piezoelectric, isotropic and non-conducting layer with a thickness of t, partition constant K and vapour con-centration Cv, the change in frequency can be written in simpliﬁed form as

where k1and k2are coupling coefﬁcients depending on the different displacement components of the wave in the substrate and f0, the operating frequency in the absence of sensing layer[25] Air partition constants and molar masses of different gases are given inTable 2 This equation implies that the shift in frequency is proportional to the mass loaded on the surface The sensitivity of a gas sensing device (S) is given as the ratio of device response (R) to gas concentration (n) The device response for an uncoated substrate is given as

whereDm/Asrepresents the mass loaded per unit surface area[27] The piezoelectric material used in this study is YZ cut Lithium Niobate (LiNbO3) crystal It has a higher wave velocity than quartz,

Fig 2 Geometry of SAW gas sensor with ZnO-CuO nanospheres in CNT adsorbent ﬁlm.

Fig 4 Mode shape plot showing deformation for CNT based sensor.

N Abraham et al / Journal of Science: Advanced Materials and Devices 4 (2019) 125e131 127

Trang 4

another commonly used piezoelectric material The designed

sensor works in the MHz frequency range LiNbO3, CNT and

aluminium electrodes were modelled in rectangular shapes of

4mm, 500 nm and 250 nm respectively (Fig 2) Aluminium

elec-trodes are predominantly used since they have low attenuation In

theﬁrst model the adsorbing layer is only CNT while in the second

model ZnO-CuO nanocomposites were modelled as nano spheres of

25 nm diameter in the CNT adsorbent layer

After specifying the material properties, the appropriate

boundary conditions were applied to the model The ground

po-tential and ﬂoating potential were applied to the left and right

electrodes respectively This is equivalent to the open circuit

con-dition, which is ideal for gas sensing After specifying the boundary

conditions, a mesh has been created in each domain The mesh

consists of 29 boundary elements The complete mesh consists of

4438 domain elements and 505 boundary elements (Fig 3) After

completing the mesh mapping, the solution of each individual

mesh has been calculated and integrated over the entire surface

The IDTs generate harmonic frequencies in addition to the fundamental mode The presence of the aluminium IDT electrodes and the piezo electric material causes the lowest SAW mode to split

up into two Eigen solutions The lowest one represents series resonance, where propagating waves interfere constructively and the other one represents parallel (“anti-”) resonance, where they interfere destructively The sensor has been tested with 100 ppm of methanol, ethanol, trichloromethane, dichloromethane, acetoni-trile, diethyl ether, acetone, n-pentane, n-hexane, and 2-propanol The mode shape plots inFigs 4 and 5show the decay of the surface displacement with the depth since the mode of propagation

is Rayleigh mode The resonant frequency is found to be 915 MHz The shift in resonant frequency in the presence of gases lies in kHz range Both the models showed maximum response to the 2-propanol (with a shift in resonant frequency of 183.88 and 200.26 kHz for CNT and ZnO-CuO/CNT respectively) The 2-propanol

is a highly inﬂammable gas, the detection of which is very important

in manyﬁelds The enhancement in sensitivity has been observed for the second model

Most SAW devices operate in the 100e600 MHz resonant frequency range Dickert et al.[28]experimentally proved that the variation in the resonant frequency could increase the sensor response in a parabolic fashion In order to study the resonant frequencyefrequency down shift relationship, we have simulated the sensor response for the different gases (100 ppm) in the second model The obtained results is shown inFig 6 It can be seen that the frequency shift increases with the resonant frequency in agreement with equation(2) Venema et al.[27]done similar studies and found that for a given sensing layer thickness, highest sensitivity has been obtained for the sensor with high operating frequency Optimizing sensor response can surely reduce the sensing layer thickness This can further enhance the adsorption process which in turn reduces the response time The ability to decreasing sensing layer thickness, however, depends on the smallest electrode distance feasible for the process of photolithography[28]

The increase in sensitivity of the second model can be primarily attributed to the increase in surface area of the adsorbing layer The speciﬁc surface area of CNT is very large and it possesses a hollow structure, which can expose a large number of reaction sites In addition, they have much higher electrical conductivity than metal Fig 5 Mode shape plot showing deformation for ZnO-CuO/CNT based sensor.

Resonant Frequency

Trang 5

oxides The strong sp2 bonding in CNTs makes them chemically

inactive Functionalization with MOX nanocomposites can surely

improve the chemical reactivity which could further enhance the

oxygen adsorption on the compositae surface On combining these

two, there will be a reduction in the resistance of the sensing layer

The surface resistivity increases with the amount of the adsorbed

oxygen which could be removed by the reduction with the gas

molecules [29] Previous studies [30,31] on CuOeZnO

nano-composites showed their potential use as an efﬁcient photoanode

material in dye sensitized solar cells These studies also revealed

that, the tendency to form signiﬁcant agglomerates in the case of

ZnO has been reduced due to the addition of CuO, which can

signiﬁcantly increase the surface area The reduction in the

resis-tance of these hybrid nanocomposites could also increase the

overall sensitivity Studies on selective CO sensing with CuOeZnO

heterocontact [32] show that surface resistance of CuO will be

increased by the oxidative reaction of reducing gases However it

will be opposite in the case of ZnO Since the resistance of ZnO is ten

times that of CuO, the total resistance of the CuOeZnO

hetero-contact will be governed by the surface resistance of ZnO The

strong interaction between CNT and carboxyl groups present in the CuOeZnO nanocomposite would also help in the sensitivity enhancement at room temperature

Most of the available sensors operate at higher temperatures except sensors based on polymers Zheng et al.[33]studied CNT/ CuO based chemical sensor and they found that these hybrid composites can enhance the sensitivity at room temperature Ac-cording to them the enhancement in sensitivity is mainly due to the strong interaction between CNTs and carboxyl groups present in the CuO Hieu et al.[34]studied enhanced sensing properties of tin oxide doped with CNTs and metal oxide semiconductors to sense the liquid petroleum gas and ethanol Functionalization of CNT is a viable and easy method to improve the sensitivity Filling with metal oxides and noble metals is an easy method to functionalize and broaden their applications Zhang and his team[35]binded ZnO quantum dots on to CNTs by covalent coupling Comparing with single wall carbon nanotubes (SWCNT) the gas adsorption mechanism in multiwall CNT (MWNT) is more complicated How-ever they exhibit high sensitivity to certain gases

The three key parameters which play major role in the sensing mechanism of SAW based sensors are mass, conduc-tivity and elasticity of the sensing material So in these sensors the surface wave can interact with the sensing layer to cause the velocity variation in three different ways a) variation in the mass of the layer b) acoustoelectric effect c) viscoelasticity The elastic loading has been neglected in majority of SAW sensors

In the case of metal oxides, the conductivity cannot play a dominant role because when this parameter changes against various VOCs, the sensor has to be heated up to 200e300C So

in such cases the mass effect will become the prominent factor which decides the sensor response Former studies [36,37]on metal/semiconductor (SC) layered structures reveal that improved sensitivity of these structures (compared to metal or

SC layers) can be attributed to the highly active conductivity regime leading to better acoustoelectric coupling between the layers and the surface wave Earlier studies on SnO2/CNT nanocomposites show that better sensitivity is due to the exis-tence of two different depletion layers and associated potential barriers [38] Electron density studies of CNT based NO2 gas sensors reveal that charge transfer takes place from the NO gas

Fig 7 Frequency downshift for the two sensor models to different VOCs.

Trang 6

adsorbed gas molecules on the CNT produce charge transfer

which will be enhanced by the CuOeZnO nanostructures CNTs

also have special ability to sense the corresponding changes

when gas molecules get attached or detached from their surface

Chemisorbed molecules can also act as interfacial states through

which electrons and holes are captured and emitted [40]

Similar studies [41] on SnO2-CNT hetero-structure based gas

sensors reveal that the enhanced performance of these hybrid

structures is due to the nanochannels formed on the MOX

semiconductor surface which can augment the diffusion of gas

molecules in to the metal oxide surface as well as increase in

the local electric ﬁeld at the interface Two different types of

depletion layers co-exist in these types of metal oxide/CNT

hybrid structures, one at the surface of the metal oxide and

other at the interface between CNT and metal oxide [42] The

adsorption of various gases can modify these depletion layers

which has a signiﬁcant effect on the sensor response [43] The

two port SAW sensors will monitor these changes and

produce corresponding changes in the resonant frequencies The

frequency shift has been measured for the different gases with

same concentration and is shown in Fig 7 and corresponding

values are given inTable 3

The sensing layer thickness is a crucial factor which affects the

sensitivity of SAW sensors The frequency variation with sensing

layer thickness has been investigated for 2-Proponol and the

ob-tained results are shown inFig 8 The highest frequency shift is

observed at the thickness of 236 nm After this critical thickness

the frequency shift shows a downshift This downshift is due to

the change in the interaction mechanism from mass or

acousto-electric effect to elastic effect Similar results were observed for

SnO2and ZnO based SAW gas sensors[44]

4 Conclusion

The SAW based sensor technology is very promising for

sensing any analyte through the optimized integration with

sensing layers SAW gas sensors based on CNT and ZnO-CuO/CNT

nanocomposites were modelled in COMSOL Multiphysics Their

response has been tested for different VOCs at room temperature

Compared to the CNT based sensor, the ZnO-CuO/CNT sensor

showed an improved performance However, in both the cases the

response maximum is found for 2-Propanol, a highly inﬂammable

gas The enhancement in sensitivity can be attributed to the

increased surface area, the reduction in the resistance of the

sensing layer as well as the strong interaction between CNT and

the carboxyl groups present in the hybrid nanocomposite In

these types of CNT-MOX hybrid structures, the nanochannels

formed on the MOX surface can also augment the diffusion of

analyte molecules The incorporation of polymer layers in

be-tween could bring further sensitivity enhancement, which is a

future scope of the present work

Acknowledgements

The authors are pleased to acknowledge, Department of

me-chanical Engineering, Govt Engineering College Barton Hill,

Thir-uvananthapuram, Kerala, India, for providing the lab facilities

References

An electrochemical dopamine sensor Based on the ZnO/CuO nanohybrid

Structures, J Nanosci Nanotechnol 14 (2014) 6646e6652.

Z.H Ibupoto, A robust, enzyme-free glucose sensor based on lysine-assisted CuO nanostructures, Sensors 16 (2016) 1424e8220.

prop-erties of an integrated optical gas sensor based on a multilayer structure, Anal Chem (2008) 9188e9194.

mediated CuO bundle of nanowires, their structural characterization and cholesterol detection, Mater Sci Eng C 33 (2013) 3889e3898.

nanosheets and their non-enzymatic glucose sensing applications, Sensors 13 (2013) 7926e7938.

J Mater Sci Mater Electron 29 (2018) 374e381.

chemical sensors, Sov Microelectron 20 (1991) 337e355.

21e31.

nanostructures and their chemical sensing application for the determination

of cadmium ions, Electroanalysis 25 (2013) 1425e1432.

photo-detector based on coral reef like p-NiO/n-ZnO nanocomposite struc-tures, Mater Lett 108 (2013) 149e152.

modeling, J Sensors (2009) 1e11.

applica-tions of recognition layers, J Sensors 17 (2017).

a new surface acoustic wave sensor for the rapid identiﬁcation of bacteria and yeasts, Lett Appl Microbiol 42 (2006) 24e29.

of dye synthesized solar cell by few layered graphene/CuO nanocomposite as a working electrode, Digest J Nanomater Biostruct 12 (2017) 67e72.

few-layered graphene-based CdO nanocomposite-enhanced power conversion efﬁciency of dye-sensitized solar cell, J Electron Mater 47 (2018) 620e626.

A Felten, J Pireaux, X Correig, E Llobet, Hybrid metal oxide and multiwall

(2007) 137e142.

of tin oxide doped with metal oxides and carbon nanotubes: a competitive sensor for ethanol and liquid petroleum gas, Sens Actuators, B 144 (2010) 450e456.

Sen-sors 12 (2012) 1424.

Steady-state gas response, Sens Actuators, B 3 (1983) 233e254.

power gas sensing using single walled carbon nanotubes in wearable devices, IEEE Sensor J 23 (2016) 8329e8337.

operation of self-heated, suspended carbon nanotube gas sensors, Appl Phys Lett 103 (2013) 223109.

mecha-nism of carbon nanotubes: from single tubes to high-density networks, Car-bon 69 (2014) 417e423.

Lithuanian J Phys 55 (2015) 319e329.

applications, Rom J Phys 60 (2015) 502e511.

by Taguchi dynamic methods, Sensors 9 (2009) 1394e1408.

Design aspects of saw gas sensors, Sens Actuators, A 10 (1986) 47e64.

devices-sensitivity enhancement in going from 80 MHz to 1 GHz, Sens Actuators, B

46 (1998) 120e125.

conductivity of ZnO powder layers, J Chem Soc., Faraday Trans 74 (1978) 31e45.

catalytically active CuO-ZnO nanocomposite synthesized by single step method, Spectrochim Acta 200 (2018) 116e126.

active heterojunction CuO/ZnO nanocomposites for DSSC application, J Mater Sci Mater Electron 29 (2018) 21002e21013.

Trang 7

[32] Y Nakamuru, H Yoshioka, M Miyayama, H Yanagida, Selective CO gas

sensing mechanism with CuO-ZnO heterocontact, J Electrochem Soc 137

(1990) 940e943.

nanotube (CNTs)/CuO with high sensitivity to organic volatiles at room

temperature, Procedia Engineering 36 (2012) 235e245.

tin oxide doped with metal oxides and carbon nanotubes: a competitive sensor

for ethanol and liquid petroleum gas, Sens Actuators, B (2010) 450e456.

functionalized multi-walled carbon nanotubes by covalent coupling, Carbon

47 (2009) 1214e1219.

(2011) 986e993.

hydrogen detection in a surface acoustic wave sensor system, Sens Actuators,

B 82 (2002) 265e271.

(2011) 966e970.

(2000) 57e60.

Solid 14 (1960) 214e219.

3955e3960.

doped with carbon nanotubes operating at room temperature, Sens Actua-tors, B 101 (2004) 81e89.

Tendeloo, C Bittencourt, N Canellas, E Llobet, Novel hybrid materials for gas sensing applications made of metal-decorated MWCNTs dispersed on nano-particle metal oxides, Sens Actuators, B 131 (2008) 174e182.

(2013) 746e752.

Định dạng
Số trang	7
Dung lượng	1,16 MB