Nitrogen dioxide sensing characteristics at elevated temperature of sol-gel glass thin films containing substituted porphyrin dyes

Article abstract of DOI:10.1039/b006344o

An optical gas sensor device based on tetra-substituted porphyrin dye entrapped in sol-gel glass displays greatly improved signal strength (1400% at 295 K) on exposure to 4.4 ppm nitrogen dioxide (in N₂) gas compared to a multilayer Langmuir-Blodgett film fabricated from the same porphyrin. Temporal response was found to be similar, but differences in the two deposition methods were observed in terms of signal strength and response/recovery times. This behaviour is explained in terms of the differing nature of the two film types. A room temperature sensitivity of 176 ppb was measured and the response obeyed Langmuir adsorption kinetics.

Full text of DOI:10.1039/b006344o

View Article Online / Journal Homepage / Table of Contents for this issue
Nitrogen dioxide sensing characteristics at elevated temperature of
sol±gel glass thin ®lms containing substituted porphyrin dyes
O. Worsfold,a C. M. Dooling,a T. H. Richardson,*a M. O. Vysotsky,b R. Tregonning,b
C. A. Hunterb and C. Malinsc
aDepartment of Physics and Astronomy, University of Shef®eld, Shef®eld, UK S3 7RH.
Tel: z44-0114-2224280; Fax: z44-0114-2728079; E-mail: T.Richardson@Shef®eld.ac.uk
bDepartment of Chemistry, University of Shef®eld, Shef®eld, UK S3 7RH
cDepartment of Instrumentation and Analytical Science, UMIST, P. O. Box 88, Manchester,
UK M60 1QD
Received 3rd August 2000, Accepted 2nd November 2000
First published as an Advance Article on the web 14th December 2000
An optical gas sensor device based on tetra-substituted porphyrin dye entrapped in sol±gel glass displays
greatly improved signal strength (1400% at 295 K) on exposure to 4.4 ppm nitrogen dioxide (in N2) gas
compared to a multilayer Langmuir±Blodgett ®lm fabricated from the same porphyrin. Temporal response was
found to be similar, but differences in the two deposition methods were observed in terms of signal strength
and response/recovery times. This behaviour is explained in terms of the differing nature of the two ®lm types.
A room temperature sensitivity of 176 ppb was measured and the response obeyed Langmuir adsorption
kinetics.
volume ratio of the chemical transducer is desirable for
1. Introduction
optimum gas sensing performance. A variety of thin ®lm
deposition techniques have been utilised to this end, such as
vacuum evaporation13 and spin-coating.14 Deposition of thin
®lms using the Langmuir±Blodgett (LB) technique has led to
excellent gas sensing response for a number of organic
compounds.15 However, LB ®lms of porphyrins are well
known to have poor stability with respect to mechanical and
thermal treatment16 and entrapment of organics in a sol±gel
matrix has been shown to stabilise the ®lm for other systems.17
A further disadvantage of using LB ®lms for gas sensing is the
labour intensive nature of the ®lm fabrication, which requires
specialised equipment for the deposition of ®lms of reprodu-
cibly high quality.
The sol±gel process is an ambient temperature method that
may be used to fabricate thin glass ®lms that have well de®ned
optical and physical characteristics, such as thickness, porosity
and refractive index. Details of the in¯uence of varying process
parameters on the properties of the resultant silica ®lms are
described elsewhere,18,19 but in general, factors such as pH and
ratio of water to silicone alkoxide precursor can be used to
de®ne the nature of the glass. Analyte-sensitive reagents may be
added at the precursor stage, and will subsequently become
entrapped in the porous coating to yield dye-doped sensor
®lms. Sol±gel-derived materials have been used in the
development of a wide range of optical sensors for pH
monitoring20 and the detection of analytes such as dissolved
oxygen21 and ammonia gas.22,23 We have entrapped the
porphyrin dye within sol±gel glass ®lms fabricated from a
methyltriethoxysilane (MTEOS) precursor. The substitution of
a reactive alkoxy group in the glass precursor with an
unreactive alkyl group results in decreased polarity and
hydrophilicity of the resulting organically modi®ed silica
(ormosil) glass surface.24,25 These alkyl functions populate
the surface of the micro-porous glass, thereby reducing the
tendency of ®lms prepared under conditions of acid catalysis to
be signi®cantly acidic and hydrophilic. Such characteristics
make these ®lms ideal hosts for the type of porphyrin dye
materials we have employed during this study, which are not
readily compatible with hydrous environments.
Over a number of years there has been a large volume of
literature concerning dedicated gas sensor devices for labora-
tory and industrial applications. Areas that best highlight this
include the environmental monitoring of toxic gases, develop-
ment of emergency early-warning systems and the possibility of
in vivo monitoring of blood gases.1 With respect to nitrogen
dioxide gas sensing, a number of detection techniques have
been discussed in the literature, including conductometric2 and
micro-gravimetric methods,3 and more recently work by
ourselves and others monitoring changes in the optical
properties of thin ®lms in the presence of analyte gas.4 An
excellent review of the major principles of optical sensor design
and application and discussion of the numerous advantages of
this approach over more well established methods can be found
elsewhere,5 but principally these are the low power require-
ment, low cost and electrical isolation possible with optical
sensor systems.
Several types of material have been reported as displaying
spectral modulation in the presence of nitrogen dioxide, and
have subsequently been incorporated into optical sensor
devices. These include metal oxides,6 polymers,7 phthalocya-
nines,8 azobenzenes9 and porphyrins.10 Porphyrins are highly
conjugated organic molecules which have been used for a
number of sensor applications11 due to their inherent chemical
stability and favourable optical properties including spectral
changes in the visible region (400±900 nm) which will enable
the use of cheap LED's in a commercial device.
The ideal sensor operates over a range of temperatures with
no discernible loss of response, is highly selective and exhibits
fast temporal response and recovery characteristics.12 While
many devices have been previously detailed, operation in
environments of elevated temperature is rarely reported,
mainly due to the fact that very often sensor performance
beyond ambient laboratory conditions is disappointing. In this
paper we aim to highlight the sensing characteristics of our
laboratory prototype at elevated temperatures that are often
found in industrial environments.
When considering thin ®lm sensors, a high surface area to
DOI: 10.1039/b006344o
J. Mater. Chem., 2001, 11, 399±403
This journal is # The Royal Society of Chemistry 2001
399

View Article Online
One of the aims of this work was to show that the
entrapment of the porphyrin in a sol±gel matrix could produce
a ®lm with gas sensing characteristics comparable to the much
thinner LB ®lm analogue. The highly micro-porous nature of
sol±gel derived glass thin ®lms facilitates gaseous ¯ow, but in
general thin ®lms mechanically deposited onto glass or
polymeric substrates are an order of magnitude thicker than
a multilayer LB ®lm. Previous work has shown that entrap-
ment of an azobenzene derivative in a sol±gel matrix enabled
fast and sensitive response to NO2 gas,26,27 that showed often
superior sensor performance in comparison to the LB ®lm
analogue, and we aim to further illustrate this using our own
sensor system.
(rendered hydrophobic by immersion in 1,1,1,3,3,3-hexam-
ethyldisilazane (HMDS) for 24 h) at a rate of 1000 mm min21
and a surface pressure of 15 mN m21. The transfer rate was
much greater than for conventional LB ®lm deposition.
2.3. Deposition of sol±gel thin ®lms
Porphyrin dye (EBO or EHO, 5 mg) was completely dissolved
in MTEOS (1.60 g) by the addition of 2±3 drops of chloroform.
The resultant deep purple solution was then vigorously stirred
during the dropwise addition of aqueous HCl (pH 1, 0.42 g).
Stirring was continued for one hour, after which the solution
had become homogeneous and had developed a dark green
colour. This sol was immediately used to spin coat thin ®lms of
porphyrin doped sol±gel onto glass substrates.
A few drops of the dye-doped sol were placed onto the
surface of clean glass slides (Blue Star, Smethwick, UK),
revolving at 6000 rpm. These produced consistent ®lms of a
thickness of approximately 500 nm. The coated glass slides
were dried overnight (70 ³C), and were left to stabilise in
laboratory conditions for one week to ensure that any post-
fabrication structural evolution of the glass ®lms that may
affect the porosity and thickness had gone to completion.19
2. Experimental
2.1. Materials
The synthesis and structures of the dye materials 5,10,15,
20-tetrakis[3,4-bis(2-ethylhexyloxy)phenyl]-21H,23H-porphine
(EHO) and 5,10,15,20-tetrakis[3,4-bis(2-ethylbutoxy)phenyl]-
21H,23H-porphine (EBO) are shown in Scheme 1. The
synthesis and characterisation of EHO is described elsewhere.28
Similar procedures were used in the synthesis of EBO and
characterised as follows.
2.4. Gas sensing
1H NMR 250 MHz (CDCl3): d (ppm)~8.823 (s, 8H), 7.785
(s, 4H), 7.711 (d, 4H, 3JHH~8.2 Hz), 7.244 (d, 4H,
3JHH~8.2 Hz), 4.184 (d, 8H, 3JHH~5.8 Hz), 4.095 (d, 8H,
The NO2 sensing chamber was designed in-house29 and is
shown in Fig. 1. Absorbance changes were monitored using a
World Precision Instruments optical ®bre spectrophotometer
driven by Spectromate software. Thermostatic control of the
sensor ®lms involved ®xing the substrate in position together
with a Peltier heater/cooling system and passing the gas in a
direction perpendicular to the light interrogation. Nitrogen
dioxide (4.4 ppm in nitrogen, BOC, Guilford, UK) was ¯owed
through the gas chamber via the minimum length of Te¯on
tubing. The response temperature of each exposure was varied
and is shown on the relevant diagrams. The temperature of
recovery in nitrogen (BOC, Guildford, UK) was ®xed at 353 K
for all experiments. Elevated temperatures were used in the
recovery cycles due to the slow (several hours) recovery
observed at room temperature for these systems.
3JHH~5.8 Hz), 1.908 (q, 4H, JHH~5.8 Hz), 1.85±1.42 (m, 36
3
H), 1.076 (t, 24H, 3JHH~7.4 Hz), 0.923 (t, 24H,
3JHH~7.4 Hz), 22.738 (s, 2H). UV±VIS (CHCl3): lmax/nm
(e61023 mol21 l cm21)~427 (408), 521 (19), 557 (12), 594 (7),
650 (6). FAB (zve), m/z: 1414 (Mz) (Calc. for C92H126O8
1414). Melting point: 250±253 ³C.
2.2. Langmuir±Blodgett ®lm deposition
Thin ®lm assemblies of EHO were prepared using a constant
perimeter Joyce±Loebl mini-trough. The porphyrin solution
was spread onto a pure water subphase (pH#6.2, 20 ³C) and
transferred using a fast-deposition technique onto glass plates
The signal strength of the gas response was calculated using
eqn. (1),
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Ao{At
Sl~
|100
(1)
Ao
where Sl~the signal strength at a given wavelength; At~the
absorbance after exposure to gas at equilibrium; A0~the
absorbance before exposure to gas.
2.5 Adsorption/desorption theory
There are a number of isotherm models which describe the
adsorption/desorption of gas molecules onto a solid surface
including the BET,30 Toth,31 Freundlich32 and Hill33 models
amongst others. A review of these is provided elsewhere.34 The
Fig. 1 Schematic design of sensing apparatus. Gas ¯ow direction is
perpendicular to the substrate.
Scheme 1 Synthetic route and structure of porphyrins EHO and EBO.
400
J. Mater. Chem., 2001, 11, 399±403

View Article Online
simplest isotherm model to describe the adsorption of a gas
onto a solid surface (and the model on which a number of the
subsequent isotherms above were based) is the Langmuir
adsorption isotherm35 which has been used by a number of
workers to characterise their systems.36 The Langmuir
adsorption isotherm is usually expressed as eqn. (2),
®lms is discussed in greater detail elsewhere,28 but is similar to
that described by the dye-doped sol±gel ®lm. These ®lms
displayed fast response with high signal strength on exposure to
4.4 ppm NO2, detailed in Table 1, which also outlines the NO2
sensing characteristics of sol±gel entrapped dyes.
The response time data shown in Table 1 indicate that the
sol±gel entrapped porphyrin ®lm has faster temporal response
during the fourth exposure to 4.4 ppm NO2 compared to the
®rst exposure. This pre-gassing phenomenon is common with
diffusion-limited systems and has been attributed to the
incomplete recovery of the system where the remaining
adsorbed NO2 molecules lower the required binding energy
of NO2 at the unoccupied sites on further exposures thus
decreasing the response times.41 A fast t50 response time is
recorded for the LB ®lm with a t90 response time approximately
the same as the sol±gel ®lm. This indicates that the signi®cantly
thinner LB ®lm has a very fast initial reaction mechanism with
the analyte gas (compared with the sol±gel), followed by a
second slower mechanism obtaining full response. There is little
signi®cant change in the response kinetics of the LB ®lms on
multiple exposures.28 This is probably because the LB system
contains a very thin sensor layer and therefore diffusion related
factors are less dominant.
nads
Ns
lc
h~(
)~‰
Š
(2)
1zlc
where h~fractional surface coverage; nads~number of mole-
cules of gas adsorbed (3 DAbsorbance.); Ns~number of
adsorption sites; l~constant relating to the adsorbability of
the gas; c~concentration of the gas.
By rearranging eqn. (2) the following relationship [eqn. (3)] is
obtained.
c
c
)~( )z(
Ns
1
(
)
(3)
nads
Nsl
Therefore a plot of c/nads vs. c will be a straight line if
Langmuir kinetics are obeyed.
3. Results and discussion
A signal strength of approximately 1400% (at 470 nm) was
observed for the EHO±MTEOS ®lm, which is more than three
times that observed for the multilayer LB ®lm. However there
is signi®cant similarity in the signal strength of the two systems
at the Soret peak and 700 nm. The response times of the sol±gel
system are higher than the LB analogue implying that diffusion
of the gas through the ®lm is the limiting factor during the gas
response for the sol±gel deposited thin ®lms. A similar increase
in the recovery times of the sol±gel compared with the LB ®lm
analogue would indicate that diffusion is also dominant during
the recovery cycle. The classical numerical method for
determining the effect of diffusion is to plot the change in
absorbance as a function of the square root of time. However
this equation could not be applied to our system and more
complicated models based on the diffusion of gases at a
heterogeneous surface are being investigated.
Fig. 3 shows the effect of temperature on gas sensing
response at 470 nm for a EHO±MTEOS ®lm on exposure to
4.4 ppm NO2. A marked decrease in the response is observed as
the temperature increases probably due to a shift in the
equilibrium to desorption at elevated temperatures. As the
temperature increases, the response times (t50 and t90) decrease
substantially until a loss of response above 350 K whereas the
LB ®lm did not demonstrate this signi®cant loss of response
despite heating to 373 K as shown in Table 2.
The fundamental differences in these two types of membrane
are highlighted by this behaviour. The LB ®lm system has a
wider operating range with regard to temperature, and some
response is shown even above 350 K. However, sol±gel ®lms
fail badly under these conditions due to their hydrous nature. It
is clear that as the temperature of the system approaches the
boiling point of water there is a drastic retardation in
performance, which does not correspond to any thermal
decay of the dye material or membrane, as simply cooling the
®lm will restore its ef®cacy. We must therefore associate this
with loss of moisture at elevated temperature. Dye-doped sol±
gel ®lms have been described previously as being very effective
moisture sensors,42 although to a lesser extent when consider-
ing ormosil ®lms such as those described here. A number of
features concerning moisture dependent sensor performance
have been outlined in work by the authors in which the
moisture content of the analyte gas is controlled.43 In changing
temperature we also vary the moisture content of the vapour
phase as the vapour pressure of water in the sensor membrane
rises with temperature, and this perhaps accounts for the
temporal variations in sensor performance.
We succeeded in entrapping two similar substituted porphyrin
dyes in both LB ®lms and within a sol±gel matrix. The acidic
preparative conditions involved in sol±gel synthesis proved
non-detrimental to the dye materials despite the colour change
from violet to green in the evolving sol. The colour change is
probably related to incomplete hydrolysis and condensation
within the sol. These groups are capable of interacting with
electron rich species such as the porphyrin dye, and in doing so
will effect the above colour change.37 However these interac-
tions are not strong enough to prevent the analyte interacting
with the dye and the desired sensor response being resultant.38
The UV±Visible absorption spectrum of an EHO±MTEOS
thin ®lm, under a nitrogen atmosphere, shows the characteristic
Soret band at 430 nm with Q-bands at 480, 500 and 520 nm,
which are clearly displayed in Fig. 2. Both the spectra and
subsequent variations on exposure to analyte gas are compar-
able for both LB ®lm and sol±gel systems. On exposure to
4.4 ppm NO2 at 293 K the absorption spectrum of the doped
sol±gel ®lm undergoes a signi®cant hyperchromic shift at
470 nm and 700 nm with a simultaneous hypochromic shift of
the Soret band at 435 nm. There also appears to be a small
bathochromic shift (y5 nm) of the Soret peak which could be
attributed to a change in the J-aggregation state of the
porphyrins as observed in other organic systems.39,40 It was
found that while ®lms of EBO±MTEOS displayed good
absorption spectra, there was no observable change on
exposure to 4.4 ppm NO2. Only a small and slow response to
NO2 was observed when depositing the EBO material as a
multilayer LB ®lm. The NO2 sensing behaviour of the EHO LB
Fig. 2 UV±Visible absorption spectrum of EHO±MTEOS ®lm before
and after exposure to 4.4 ppm NO2 at 295 K.
The t50 response times of the sol±gel to NO2 reach a constant
J. Mater. Chem., 2001, 11, 399±403
401

View Article Online
Table 1 Data based on absorbance changes at 700 nm, 470 nm and 430 nm respectively on exposure to 4.4 ppm NO2 at 293 K where t50, and t90 are
the times taken for 50% and 90% of overall signal for response (resp) and recovery (rec) respectively. Recovery temperature for all experiments was
353 K
Material
t50resp/s
t90resp/s
S700 nm(%)
S470 nm(%)
S430 nm(%)
t50rec/s
t90rec/s
EHO±LB ®lm31
MTEOS±EHO Run1
MTEOS±EHO Run 4
25
112
100
190
383
225
742
750
760
337
1400
1350
60
61
58
33
66
47
73
159
134
level of 45 s (Table 2) at 325 K, whereas the t90 response times
continue to decrease above this temperature. This phenomenon
is not fully understood but a possible explanation is that at low
temperatures more than one mechanism (diffusion and surface
adsorption) is controlling the response. At elevated tempera-
tures a single mechanism (diffusion) would appear to control
the response time.
Fig. 4 shows the effect of NO2 concentration on the response
characteristics for the sol±gel±EHO system. A response to a
concentration as low as 0.176 ppm (176 ppb) was obtained at
room temperature using our present system. Using eqn. 2
(linear form of the Langmuir adsorption isotherm) a plot of c/
nads vs. c gave a straight line (Fig. 5) indicating that Langmuir
adsorption kinetics were obeyed. From Fig. 5 the number of
adsorption sites is calculated from 1/gradient~0.11 k A21
where k is a constant.
the porphyrin ring because nitrogen dioxide dissolves in water
to yield a mixture of nitrous and nitric acid and this process
occurs within the sol±gel membrane, as there will be both
chemisorbed and physisorbed moisture within the membrane
available for solvation. Subsequent acid attack of the amine
functionalities within the porphyrin backbone may offer a
signi®cant contribution to the spectral variations encountered
here, which lessens as the moisture within the sol±gel ®lm
decreases at elevated temperature. The protonation mechanism
is further emphasised by the addition of tri¯uoroacetic acid
(TFA) to
a solution of EHO in CHCl3 which, when
interrogated using UV±Visible spectroscopy, recorded similar
spectral changes (Fig. 6) to the EHO±NO2 gas system.
However, direct protonation of the porphyrin ring would
involve the formation of covalent bonds. This would require
other chemical agents (i.e. alkalis) to facilitate full recovery to
the unprotonated form. The EHO±NO2 system exhibits full
recovery (v50 s) when a small amount of energy (heating) is
applied to the substrate. Full recovery, within 3 hours, was also
observed when the gassed sample was exposed to the air at
room temperature. Therefore, the probable mechanism
involves the interaction of the electrophilic NO2 gas with the
same sites where protonation occurs to account for the similar
spectral changes. The incoming NO2 forms a charge-transfer
type bond with the active porphyrin sites and with the
application of heat full recovery occurs because of the low
energy involved in the charge-transfer bond.
One possible gas sensing mechanism involves protonation of
The sudden loss of response to NO2 for the MTEOS±EHO
system above 350 K and its greater sensitivity compared with
the LB system would indicate that moisture and the hydro-
Fig. 3 Effect of temperature on absorption variation at 470 nm of
EHO±MTEOS ®lm on exposure to 4.4 ppm NO2, recovery in N2 at
353 K.
Fig. 5 Plot for linear form of the Langmuir adsorption isotherm for
EHO±MTEOS sol±gel ®lm.
Table 2 The effect of temperature on response times for EHO±
MTEOS ®lm on cyclic exposure to 4.4 ppm NO2
T/K
t50resp/s
t90resp/s
294
303
313
325
334
343
112
84
55
44
45
45
383
285
262
212
199
167
Fig. 4 Effect of NO2 concentration on the absorbance of EHO±
MTEOS sol±gel ®lm measured at 295 K.
402
J. Mater. Chem., 2001, 11, 399±403

View Article Online
10 R. H. Tredgold, F. C. J. Young, P. Hodge and A. Hoorfar, Proc.
IEEE, 1985, 132, 151.
11 R. Purrello, S. Gurrieri and R. Lauceri, Coord. Chem. Rev., 1999,
192, 683.
12 G. Harsanyi, Polymer Films in Sensor Applications ± Technology,
Materials, Devices and their Characteristics, Technomic Publish-
ing, Lancaster, USA, 1995.
13 N. E. Agbor, M. C. Petty and A. P. Monkman, Sens. Actuators, B,
1995, 28, 173.
14 X. Li, S. Shen, Q. Zhou, H. Xu, D. Jiang and A. Lu, Thin Solid
Films, 1998, 324, 274.
15 T. H. Richardson, Functional Organic and Polymeric Materials,
John Wiley and Sons, Chichester, 2000.
16 F. Feng, M. Mitsuishi, T. Miyashita, I. Okura, K. Asai and
Y. Amao, Langmuir, 1999, 15, 8673.
17 S. A. Wallington, T. Labayen, A. Poppe, N. A. J. M. Sommerdijk
and J. D. Wright, Sens. Actuators, B, 1997, 38±39, 48.
18 C. J. Brinker and G. W. Scherer, Sol±Gel Science, Academic Press,
San Diego, 1990.
19 C. McDonagh, F. Sheridan, T. Butler and B. D. MacCraith,
J. Non-Cryst. Solids, 1996, 194, 72.
Fig. 6 UV±Visible absorbance spectrum detailing the effect of
tri¯uoroacetic acid (TEA) on solution of EHO (unknown concentra-
tion) in CHCl3.
philic nature of the sol±gel matrix are important for good gas
sensing characteristics. However the exact effect of moisture on
the gas sensing mechanism is not yet fully understood and is the
subject of further work.
20 T. M. Butler, B. D. MacCraith and C. M. McDonagh, Proc.
SPIEÐInt. Soc. Opt. Eng., 1995, 2508, 168.
21 A. K. McEvoy, C. M. McDonagh and B. D. MacCraith, Analyst,
1996, 121, 785.
22 R. Klein and E. Voges, Sens. Actuators, B, 1993, 11, 221.
23 V. Chernyak, R. Reisfeld, R. Gvishi and D. Venezky, Sens.
Mater., 1990, 2, 117.
24 P. Innocenzi, M. O. Abdirashid and M. Guglielmi, J. Sol±Gel Sci.
Technol., 1994, 3, 217.
25 W. Schwertfeger, U. Glaubitt and J. Schubert, J. Non-Cryst.
Solids, 1992, 145, 85.
26 O. Worsfold, C. Malins, M. G. Forkan, I. R. Peterson,
B. D. MacCraith and D. J. Walton, Sens. Actuators, B, 1999,
56, 15.
27 D. J. Walton, L. S. Miller, I. R. Peterson, A. Bradford,
O. Worsfold, J. Scheerder, D. A. Parry, M. G. Forkan,
C. Malins and B. D. MacCraith, Synth. Met., 2000, 109, 41.
28 C. M. Dooling, T. H. Richardson, O. Worsfold, R. Tregonning,
M. O. Vysotsky, C. A Hunter, K. Kato, K. Shinbo and F. Kaneko,
J. Mater. Chem., in the press, DOI 10.1039/b006342h.
29 T. Richardson, V. C. Smith, R. A. W. Johnstone, A. J. F. N. Sobral
and A. M. d'A. Rocha-Gonsalves, Thin Solid Films, 1998, 329,
315.
30 S. Brunauer and P. H. Emmett, J. Am. Chem. Soc., 1937, 10, 2682.
31 J. Toth, Acta Chimica, Acad. Sci. Hung., 1971, 69, 311.
32 H. Freundlich, Kapillarchemie, vol. 1, Akademische Verlags-
gellschaft mbH, Leipzig, 1909.
4. Conclusions
Sol±gel entrapped porphyrins were deposited as thin ®lms and
exposed to 4.4 ppm NO2. Greater signal strength to the analyte
gas compared with the LB ®lm analogue indicates the
possibility of using the more easily fabricated sol±gel thin
®lm deposition method for producing organic gas sensors. A
response to a concentration of NO2 as low as 176 ppb was
recorded at room temperature. A number of similarities are
evident between the hydrous sol±gel ®lms and the organic LB
®lms, and there is evidence of strong moisture dependence of
the sensor mechanism in dye-doped sol±gel ®lms, which is not
encountered for LB ®lms.
Acknowledgement
CMD gratefully acknowledges the ESPRC for providing
funding for some of the research described in this report.
33 A. V. Hill, J. Physiol. (London), 1910, 40, proceedings iv.
34 O. Altin, O. Ozbelge and T. Dogu, J.Colloid Interface Sci., 1998,
198, 130.
References
35 I. Langmuir, J. Am. Chem. Soc., 1918, 40, 1361.
36 B. Chague, J. P. Germain, C. Maleysson and H. Robert, Sens.
Actuators, 1985, 7, 199; S. Goldberg, Plant Soil, 1997, 193, 35.
37 T. Fujii and K. Toriumi, J. Chem. Soc., Faraday Trans., 1993, 89,
3437.
1
J. Janata, Principles of Chemical Sensors, Plenum Press, New
York, 1989.
JX. Li, Y. Chen and H. Xu, J. Mater. Chem., 1999, 9, 1415.
Langmuir±Blodgett Films, ed. G. G. Roberts, Plenum Press, New
York, 1990.
2
3
38 Y. Dong, S. V. Pappu and Z. Wu, Anal. Chem., 1998, 70, 4730.
39 J. Heesemann, J. Am. Chem. Soc., 1980, 102, 2167.
40 T. L. Pernier, J. Chim. Phys., 1988, 85, 1081.
41 P. B. M. Archer, A. V. Chadwick, J. J. Miasik, M. Tamizi and
J. D. Wright, Sens. Actuators, 1989, 16, 379.
42 P. J. Skrdla, S. S. Saavedra, N. R. Armstrong, S. B. Mendes and
N. Peyghambarian, Anal. Chem., 1999, 71, 1332.
43 C. Malins, A. Doyle, B. D. MacCraith, F. Kvasnik, M. Landl,
4
5
C. D. George, T. Richardson, M. E. Hofton, C. M. Vale,
M. G. M. Neves and J. A. S. Cavaleiro, Mater. Sci. Eng. C, 1999,
8±9, 559.
Fibre Optical Chemical Sensors and Biosensors Vol I & II, ed.
O. S. Wolfbeis, CRC Press, Boca Raton, FL, 1991.
D. E. Williams, Sens. Actuators, B, 1999, 57(1-3), 1.
G. A. Reineccius, ACS Symp. Ser., 1996, 631, 241.
B. Bott and T. A. Jones, Sens. Actuators, 1984, 5, 43.
L. S. Miller, D. J. Walton, D. A. Parry, A. L. Newton and
C. G. D. Sykesud, UK Patent Application GB9412280, 1996.
6
7
8
9
Ï
P. Simon, L. Kalvoda, R. LukasÏ, K. Pu¯er and I. BabusÂõk,
J. Environ. Monit., 1999, 1, 417.
J. Mater. Chem., 2001, 11, 399±403
403