ELSEVIER Senso rs and Actuators B 28 (1995) 173-179 Polyaniline thin films for gas sensing N.E. Agbor a*b, M.C. Petty a, A.P. Monkman b ’ School of Engineering and Centre for Molecular ElectroGcs, University f Durham, Lkham DHl 3LE, UK bPhpics Department and Centre for Molecular Elecbunics, Universl?, of Dwfwn, Durham DHl 3LE, UK Received 13 May 1994; in revised form 22 December 1994, accepted 9 Januar y 1995 Abstract Thin films of polyaniline have been deposited by spinning, evaporation and by the Langmuir-Blodgett technique. The Nms are shown to possess slightly different in-plane electrical mnductivities, reflecting differences in their chemical structure and layer morphology. The conductivity is found to depend on the gas ambient. All t ypes of polyaniline films are sensitive to H2S and NO, at concentrations down to 4 ppm. However, only spun and evaporated films are responsive to SOz. Key~or& Gas sensors; Polyaniline; Thin films 1. Introduction The importance of environmental gas monitoring is well understood and much research has focused on the development of suitable gas-sensitive materials. Re- cently, there has been considerable interest in exploiting organic substanc es such as porphyrins [l], phthalocy- anines [2,3] and doped conductive polymers [4]. For maximum gas sensitivity, these compounds are usually studied as thin fihns. Among the doped conductive polymers that have been investigated are polypyrrole [5] and polythiophene [6]. Unfortunately, these materials are not readily pro- cessible. In contrast, polyaniline (PANi) is soluble in organic solvents [7] from which free-standing films can be cast [8]. In this work, polyaniline was processed into thin-film form using three different methods: spin- ning, vacuum ‘sublimation and he Langmuir-Blodgett (LB) technique. The gas sensitivities of the different films are compared. 2. Experimental 2.1. Substrate Fig. 1 shows a schematic diagram of the interdigitated electrode structure used in this work. It consi sts of gold electrodes patterned onto the surface of a quartz substrate; the overlap electrode length was 15 mm and the electrode gap was 0.38 mm. Chemiresistors were 0925-4005/95/$09.50 B 1995 Elsevier Science S.A. All rights reserved SSDI 0925-4005(95)01725-B Fig. 1. An interdigitated electrode structure on a quartz substrate: l-15 mm, d=O.38 mm and h=75 mm. fabricated by coating these electrodes with the poly- aniline films. 2.2. spun films Polyaniline powder (synthes &d in-house) [8] in the emeraldine base form was dissolved in N-methylpyr- rolidinone (NMP), in a polymer:solvent weight ratio of l:lOO, and sonicated for 30 min. Th e startin g material had a purity of 99.8%, as determined by NMR spec- troscopy [9]. The resulting solution appeared blue in reflected light. This was spun onto the interdigitated electrode structure shown in Fig. 1. Spinning was un- dertaken using a Dynapert PRS 14E model spinner, at a fixed speed of 3000 rpm for 30 s. The spun !ihns were transferred to a vacuum oven and heated to a temperature of 120 “C, at lo-” mbar for 10 min. A typical film-thickness value, obtained from a n Alpha Tenco surface profiling Talystep, was 2.0 f 0.1 pm. Full
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
ELSEVIER Sensors and Actuators B 28 (1995) 173-179
Polyaniline thin films for gas sensing
N.E. Agbor a*b,M.C. Petty a, A.P. Monkman b’ School of Engineering and Centre for Molecular ElectroGcs, University f Durham, Lk ham DHl 3LE, UK
bPhpics Department and Centre for MolecularElecbunics, Universl?, of Dwfwn, Durham DHl 3LE, UK
Received 13 May 1994; in revised form 22 December 1994, accepted 9 January 1995
Abstract
Thin films of polyaniline have been deposited by spinning, evaporation and by the Langmuir-Blodgett technique. The Nms
are shown to possess slightly different in-plane electrical mnductivities, reflecting differences in their chemical structure and
layer morphology. The conductivity is found to depend on the gas ambient. All types of polyaniline films are sensitive to H2S
and NO, at concentrations down to 4 ppm. However, only spun and evaporated films are responsive to SOz.
Key~or& Gas sensors; Polyaniline; Thin films
1. Introduction
The importance of environmental gas monitoring iswell understood and much research has focused on thedevelopment of suitable gas-sensitive materials. Re-cently, there has been considerable interest in exploiting
organic substances such as porphyrins [l], phthalocy-anines [2,3] and doped conductive polymers [4]. Formaximum gas sensitivity, these compounds are usuallystudied as thin fihns.
Among the doped conductive polymers that havebeen investigated are polypyrrole [5] and polythiophene[6]. Unfortunately, these materials are not readily pro-cessible. In contrast, polyaniline (PANi) is soluble inorganic solvents [7] from which free-standing films canbe cast [8]. In this work, polyaniline was processedinto thin-film form using three different methods: spin-
ning, vacuum ‘sublimation and the Langmuir-Blodgett(LB) technique. The gas sensitivities of the differentfilms are compared.
2. Experimental
2.1. Substrate
Fig. 1 shows a schematic diagram of the interdigitatedelectrode structure used in this work. It consists ofgold electrodes patterned onto the surface of a quartzsubstrate; the overlap electrode length was 15 mm andthe electrode gap was 0.38 mm. Chemiresistors were
0925-4005/95/$09.50 B 1995 Elsevier Science S.A. All rights reserved
SSDI 0925-4005(95)01725-B
Fig. 1. An interdigitated electrode structure on a quartz substrate:
l-15 mm, d=O.38 mm and h=75 mm.
fabricated by coating these electrodes with the poly-aniline films.
2.2. spun films
Polyaniline powder (synthes&d in-house) [8] in theemeraldine base form was dissolved in N-methylpyr-rolidinone (NMP), in a polymer:solvent weight ratioof l:lOO,and sonicated for 30 min. The starting materialhad a purity of 99.8%, as determined by NMR spec-troscopy [9]. The resulting solution appeared blue inreflected light. This was spun onto the interdigitatedelectrode structure shown in Fig. 1. Spinning was un-dertaken using a Dynapert PRS 14E model spinner,at a fixed speed of 3000 rpm for 30 s. The spun !ihnswere transferred to a vacuum oven and heated to atemperature of 120 “C, at lo-” mbar for 10 min. Atypical film-thickness value, obtained from an AlphaTenco surface profiling Talystep, was 2.0 f 0.1 pm. Full
N.E. A&r et al. I Sensors and Actuators B 28 (1995) 173-179 175
The current versus voltage characteristics of spin-coated emeraldine base polyaniline are shown in Fig.3. The measurements were undertaken in an atmosphereof nitrogen, at room temperature and after the currenthad stabilized (see next section). Data for the uncoatedelectrode (under nitrogen a current of 1.2 f 0.2x lo-l2A was measured with 10 V applied) confirmed that
the current was flowing through the polyaniline filmrather than through the substrate. The change in re-sistance for different film thicknesses (1.0, 2.0, 4.0 pm)indicates that Ohmic contacts have been establishedbetween the gold electrodes and the polymer. Usingthe thickness values from the surface profiler, theaverage room-temperature in-plane d.c. conductivitywas 4.4~hO.9XlO-‘~ S cm-‘, which is comparable tothe literature value of 1.0X 10-l’ S cm-’ for the baseform of emeraldine 171.The current versus voltagecharacteristics for both the evaporated and LB films
of polyaniline were qualitatively similar to those ofshown in Fig. 3 (including the linearity with film thick-ness).
The average room-temperature d.c. conductivity offreshly evaporated polyaniline film, in nitrogen, was1.0*0.2x lo-” S cm-‘. This compares with a valueof 2.0~10~~ S cm-’ reported in the literature forsimilar material [15]. The conductivity is slightly higherthan that of our spin-coated films (4.4&0.9x lo-” Scm-‘). This can be explained by the absence of quinoidrings to disrupt rr-r mixing between adjacent benzoidrings in the polymer chain [16].
Electrical measurements on polyaniline LB films havebeen reported previously [ll]. The film has a room-temperature conductivity in nitrogen of 10W8 cm-‘.This is significantly higher than that of the emeraldinebase form of polyaniline, suggesting that a degree of
protonation, possibly by the acetic acid, had occurred.
In general, the agreement between the conductivity
values reported here and those in the literature is not
unreasonable considering that (a) polyaniline exists in
I I I-500 500 1500
Supply voltage [mVl
Fig. 3. The room-temperature current vs. voltage characteristics forspun emeraldine base polyaniline on gold-plated interdigitated copper
electrodes for different film thicknesses: (a) 1.0 pm; (b) 2.0 pm: (c)
4.0 p.m.
different oxidation states and (b) external influences(impurities) may result in doping of the material.
3.3. Gas sensitivity
3.3-l. Nitrogen
Fig. 4(a) shows the effect of dry N2 on the d.c.conductivity of a spun polyaniline chemiresistor. Theconductivity decreased very rapidly upon the intro-duction of N2 and became stable after approximately60 min. This can be associated with the removal ofsurface/bulk trapped water molecules. A similar re-sponse was also obtained for polyaniline in LB filmform. Fig. 4(b) shows the effect of dry nitrogen on thed.c. conductivity of an evaporated polyaniline layer. Inthis case, the shorter time to achieve a stable conductivityvalue can be attributed to the lower level of water and
the evaporated film. The effects of water on the con-ductivity of polyaniline are well documented [17,18].
No evidence for oxidation (see previous section) was
noted for the evaporated film in the nitrogen envi-
N.E. Agbor et ai. f Sen.wn and Acmatm B 28 (1995) 173-179 I77
of the same thickness. Complete recovery for 10 ppmof the gas was achieved after a period of about 60
min. A similar effect was observed with an 18 LB layerpolyaniline chemiresistor. The change in conductivity,after a fixed exposure time, for both spun and LB
polyaniline chemiresistors is shown in Fig. 7. The thresh-
old for detection is about 4 ppm H,S for both films,H2S is a known reducing gas. Thus, we would expect
to observe a decrease in the conductivity of the polyaniline chemiresistors. The observed increase in con-ductivity indicates that either more than one type ofreaction site is available or that a number of differentreactions are possible. At room temperature and pres-sure, H,S dissociates in water into H+ and HS- [19-221
as illustrated in Fig. 8.The H’ ion may subsequently protonate the polymer,
i.e.,
FANi]+[H]* c== [PANiH]+ (0
where the equ~jb~um is shifted to the right duringexposure and to the left after exposure. The protonationagain produces charge carriers (semiquinone radicals)resulting in an increase in the d.c. conductivity. Thisreaction is likely to involve different sites in the polymerthan for the NO, response. As a result, the sensit~itiesof the LB and spun films are similar (compare the
poor sensitivity of the LB film to NO, in Fig. 6).Fig. 9(a) shows the effect of 10 ppm H2S on a 210
nm freshly evaporated polyaniline chemiresistor. Thisreveals an irreversible decrease in conductivity at room
oL-----J0 4 8 12
H2S cone. [vpml
vE$ cone. @ml
Fig. 7. The response of PANi films to different HsS concentrations:
(a) 1.0 pm spun film; (b) 110 nm LB film (2 V supply and temperature
20f2 “c).
aqueousphaseH++HS-
I
Fig. 8. An illustration of the state of HsS in different environments
[21]. In this work, the vapour phase is equivalent to H&surface
bound water molecules and the aqueous phase is equivalent to HsS/
water molecules trapped in the bulk of the film.
nine ruins]
0 4 8 12 16 20 24 28 32 36 40 44 48
Gas off
Fig. 9. (a) The effect of 10 ppm HaS on an evaporated polyaniline
chemiresistor at room temperature. (b)The msponse of an evaporated
polyaniline chemiresistor to different concentrations of H&3 at room
temperature (2 V supply, film thickness 210 nm and temperature
205~2 “C in both cases).
temperature. The device response to different H,S gasconcentrations is shown in Fig. 9(b). A threshold de-tection value of 10 ppm is evident. Note that spun andLB films are likely to possess more H,O than theevaporated material (Fig. 4), thus increasing the like-lihood of the reaction given by Eq. (1).
3.3.4. suZfir dioxide
SO 2 produced an increase in conductivity of spunpolyaniline as well as complete reversibility at roomtemperature. The effect of different SO, gas concen-
trations is shown in Fig. 10(a), revealing that the deviceis capable of measuring changes down to 2 ppm.Fig. 10(b) shows the response of an evaporated
polyaniline chemiresistor to different concentrations ofSOz. Here, the detection threshold is about 10 ppm,
N.E. A&r et al. I Senrors and Actuators B 28 (1995) 173479 119
[19] JA. Barbero, KG. McCurdy and P.R. Tremaine, Apparent
molar heat capacities and volumes of aqueous hydrogen sulphide
and sodium hydrogen sulphide near 25 “C: the temperature
dependencies of H2S ionization, Can. J. Chem, 60 (1982)
1873-1880.
[ZO] J.J. Carroll and A.E. Mather, The solubility of hydrogen sulphide
in water from 0 to 90 “C and pressure to 1 MPa, Gwchim
Cosmochim., 53 (1989) 1163-1170.
[21] E.C.W. Clarke and D.N. Glew, Aqueous nonelectrolyte so-lutions. Part VIII. Deuterium and hydrogen sulphide solubilities
in deuterium oxide and water, Can. J. Chem., 49 (1971) 691-698.
[22] W. Geaard, Solubility of hydrogen sulphide, dimethyl ether,
methyl chloride and sulphur dioxide in liquids. The prediction
of solubility of all gases, J. Appl. Chem. Biokzhnol., 22 (1972)
623450.
(231 P.S. Barker, J.R. Chen, N.E. Agbor, A.P. Monkman, P. Mars
and M.C. Petty, Vapour recognition using organic films and
artificial neural networks, Sensors and Actudors B, I7 (1994)
143-147.
Biographies
NE. Agbor was awarded a B.Sc. from Keele Univer-sity in 1988 and obtained an M.Sc. from the University
of Manchester in 1990. He subsequently received hisPh.D. from the University of Durham for work on gassensing using organic films.
Andy Monkman obtained his BSc. and Ph.D. degreesfrom Queen Mary College, University of London. Cur-rently he heads the Organic Electroactive Materials
Group in the Department of Physics, University ofDurham. His research activities include the charac-terization and applications of conductive polymers, es-
pecially polyaniline, and laser spectroscopy, includingfemtosecond time-resolved measurements.
Michael Petty is a professor of electronics in the
School of Engineering at the University of Durham.He is also co-director of the Durham Centre for Mo-lecular Electronics. He gained his B.Sc. and D.Sc. fromthe University of Sussex and his Ph.D. from ImperialCollege, London. His research interests include thedevelopment of organic materials, particularly Lang-muir-Blodgett films, and their incorporation in novelelectronic and optoelectronic devices.