Detection of volatile organic compounds using porphyrin derivatives

Article abstract of DOI:10.1021/jp102755h

Seven different porphyrin compounds have been investigated as colorimetric gas sensors for a wide range of volatile organic compounds. The porphyrins examined were the free base and Mg, Sn, Zn, Au, Co, and Mn derivatives of 5,10,15,20-tetrakis[3,4-bis(2-e

Full text of DOI:10.1021/jp102755h

J. Phys. Chem. B 2010, 114, 11697–11702
11697
Detection of Volatile Organic Compounds Using Porphyrin Derivatives
A. D. F. Dunbar,*,† S. Brittle,‡ T. H. Richardson,*,‡ J. Hutchinson,§ and C. A. Hunter§
Department of Chemical and Biological Engineering, UniVersity of Sheffield, Sir Robert Hadfield Building,
Mappin Street, Sheffield S1 3JD, U.K., Department of Physics & Astronomy, UniVersity of Sheffield, Hicks
Building, Hounsfield Road, Sheffield S3 7RH, U.K., and Department of Chemistry, UniVersity of Sheffield,
Dainton Building, Brook Hill, Sheffield S3 7HF, U.K.
ReceiVed: March 26, 2010; ReVised Manuscript ReceiVed: July 09, 2010
Seven different porphyrin compounds have been investigated as colorimetric gas sensors for a wide range of
volatile organic compounds. The porphyrins examined were the free base and Mg, Sn, Zn, Au, Co, and Mn
derivatives of 5,10,15,20-tetrakis[3,4-bis(2-ethylhexyloxy)phenyl]-21H,23H-porphine. Chloroform solutions
of these materials were prepared and changes in their absorption spectra induced by exposure to various
organic compounds measured. The porphyrins that showed strong responses in solution were selected, and
Langmuir-Blodgett films were prepared and exposed to the corresponding analytes. This was done to determine
whether they are useful materials for solid state thin film colorimetric vapor sensors. Porphyrins that readily
coordinate extra ligands are shown to be suitable materials for colorimetric volatile organic compound detectors.
However, porphyrins that already have bound axial ligands when synthesized only show a sensor response to
those analytes that can substitute these axial ligands. The Co porphyrin displays a considerably larger response
than the other porphyrins investigated which is attributed to a switch between Co(II) and Co(III) resulting in
a large spectral change.
1. Introduction
conjugated π-electrons in the porphyrin are shifted, and hence
the absorption spectrum changes. Considerable work has already
There are increasing concerns about the impact of gaseous
toxins released either maliciously or inadvertently as pollution
from industry or combustion engines. Accurate information
about the concentration of these toxins is required, so it is
possible to determine their impact upon the environment and
the health and safety of those living and working close to their
sources. However, collecting such information is hindered
because affordable, accurate, and reliable sensors capable of
quantifying the presence of such toxins have not as yet been
developed. Therefore, there is a strong motivation to develop
more economical gas sensors for the monitoring of low
concentrations of volatile organic toxic gases.
At present, readily available gas sensors include those that
use semiconducting metal oxides such as tin, tungsten, or
chromium titanium oxides.1,2 These materials change conductiv-
ity when they interact with an oxidizing or reducing analyte.
However, they do not respond well to humid conditions, and
they require high operating temperatures of ∼150-600 °C.
Other well-known sensor devices include liquid electrolyte fuel
cells and infrared sensors.3 Conducting polymers4 have achieved
limited success as “electronic nose” devices and phthalocyanines
have attracted attention as semiconductometric sensors.5 How-
ever, the majority of these organic sensors suffers from slow
response times and incomplete recovery in the absence of the
analyte vapor.
been performed to develop a porphyrin-based sensor system for
NO2,7-10 and porphyrin-based sensors are also being incorpo-
rated into single use colorimetric sensor arrays to detect volatile
organic compounds.11,12
This work is a study of seven different porphyrins (see Figure
1): EHO in its free base form and also when metalated with
Mg, Sn, Zn, Au, Co, and Mn. We compare the reaction of these
porphyrins with a broad range of different volatile organic
compounds, and the aim of this work is to determine the
suitability of these porphyrins for use as selective optical-based
gas sensors for the different volatile organic analytes. Various
metallo-porphyrins are known to form complexes with organic
compounds and are therefore likely to result in strong optical
changes upon exposure to the organic analytes. Their potential
selectivity and ability to produce a visible optical change makes
these materials very attractive as a low cost sensor system
because the human eye could be used as the detection mech-
anism. Therefore, any high costs associated with fabricating a
sophisticated detector are negated by making use of the high
sensitivity of the human eye. The seven porphyrins were tested
in solution for changes in their optical absorption upon exposure
to the full range of analytes. These analytes were selected as
they cover a wide range of volatile organic compounds that are
typical of those commonly produced in industrial and biological
processes; e.g., amines are a byproduct produced during the
spoiling of seafood; organic acids are common products in spoilt
wines; and ethyl acetate is a potentially explosive solvent
common in many different industries where improved safety
would be beneficial. Those porphyrin analyte pairs that showed
a large spectral change upon exposure in the solution experi-
ments were then selected for further investigation. The selected
porphyrins were used to fabricate solid state Langmuir-Blodgett
films which were subsequently exposed to the corresponding
Porphyrins are highly stable and possess distinctive UV-visible
absorption spectra due to their highly conjugated π-electron
systems.6 If the analyte under investigation has a sufficiently
strong interaction with the porphyrin, the energies of the
* Corresponding authors. E-mail: a.dunbar@sheffield.ac.uk; t.richardson@
sheffield.ac.uk.
† Department of Chemical and Biological Engineering.
‡ Department of Physics & Astronomy.
§ Department of Chemistry.
10.1021/jp102755h  2010 American Chemical Society
Published on Web 08/25/2010

11698 J. Phys. Chem. B, Vol. 114, No. 36, 2010
Dunbar et al.
acetate in glacial acetic acid. Following a basic workup with
dilute sodium hydroxide, the Mn(II)porphyrin initially formed
is rapidly oxidized to Mn(III) EHO (see Figure 1(e)). To insert
tin, the free base porphyrin was refluxed with tin(II) chloride
in pyridine followed by stirring with sodium hydroxide solution
(see Figure 1(f)). Zinc was inserted into the free base EHO by
using excess zinc acetate in dichloromethane at room temper-
ature (see Figure 1(g)). All six metalloporphyrins upon reaction
completion were purified using column chromatography. The
analytes used are 1-octanol, hexylamine, octanal, acetic acid,
1-hexanethiol, 2-butanone, ethyl acetate, trimethyl phosphite,
octylamine, and triethylamine. They were all provided by Sigma
Aldrich and used as supplied.
2.2. Solution Absorption Measurement. Solutions of the
free base and metalated EHOs were prepared at concentrations
of 4.0 × 10-6, 3.2 × 10-6, 2.6 × 10-6, 4.0 × 10-6, 4.0 × 10-6
,
4.0 × 10-6, and 4.0 × 10-6 M for the free base, Mg, Sn, Zn,
Au, Co, and Mn EHO, respectively, using chloroform as the
solvent (used as supplied by Sigma-Aldrich). Quantities of 2
mL of each of these EHO solutions were placed in quartz
cuvettes, and initial absorption spectra of the solutions were
recorded using an Ocean Optics USB2000 spectrometer with a
Mikropack Mini D2 UV-vis-IR light source. The absorbance
spectra for each porphyrin solution were measured across the
range 350-850 nm with respect to a reference taken of
chloroform. Then 20 µL of the analyte to be investigated was
added and the mixture stirred before repeating the absorption
measurement. More analyte was added 20 µL at a time and the
absorption recorded each time until no further spectral changes
were observed. This process was repeated for the seven EHOs
with each of the ten different analytes described in the materials
section 2.1. Absorbance spectra of each of the analytes mixed
with chloroform (i.e., no porphyrin present) were also recorded
with respect to a chloroform filled quartz cuvette reference, to
confirm the range of optical transparency of each analyte.
2.3. Langmuir-Blodgett Film Preparation and Absorp-
tion Measurement. Langmuir-Blodgett (LB) films of each of
the porphyrins selected on the strength of their corresponding
solution interactions, i.e., free base, Mg, Zn, Co, and Mn EHO,
were prepared from concentrated chloroform solutions (∼0.3
mg/mL). Typically, 150 µL of the EHO solution (∼0.3 mg/
mL) was spread on a clean water (Elga PURELab Option
>15MΩcm)subphasesurfaceinaNIMAmodel611Langmuir-Blodgett
trough. After evaporation of the solventm the film was
compressed to and then held at a surface pressure of 15 mN/m
while deposition onto a glass plate (rendered hydrophobic by
exposure to 1,1,1,3,3,3-hexamethyldisilazane vapors for 12 h)
was carried out. Deposition was performed by inserting and
withdrawing the glass plate vertically at a rate of 10 mm/min.
Ten excursions were performed for each plate resulting in the
deposition of 20 layers (one layer for each of the up and down
strokes). If necessary, so as to maintain the surface pressure at
15 mN/m, the deposition was halted, while the trough barrier
was retracted and more of the solution added to the subphase
before recompressing to 15 mN/m again and continuing the
deposition.
Figure 1. Molecular structure of the porphyrins used in this study.
analyte. The morphologies similar to the Langmuir-Blodgett
films after transferring to a solid substrate are presented in ref
13. The materials and experimental procedures used are
described in the next section, followed by discussion of the
results, and finally some conclusions are stated.
2. Materials and Experimental Methods
2.1. Synthesis of the Porphyrins and Analyte Details. The
synthesis of 5,10,15,20-tetrakis[3,4-bis(2-ethylhexyloxy)phenyl]-
21H,23H-porphine (EHO) is described elsewhere,8 and its
chemical structure is shown in Figure 1(a). Upon metalation,
the two hydrogen atoms in the center of the porphyrin ring are
replaced with the corresponding metal ion. The metal derivatives
of 5,10,15,20-tetrakis[3,4-bis(2-ethylhexyloxy)phenyl]-21H,23H-
porphine were prepared in accordance with literature methods.14
To prepare the Au(III) porphyrin, free base EHO was refluxed
with excess potassium tetrachloroaurate(III) and sodium acetate
in glacial acetic acid. After 48 h, the reaction mixture was cooled
and treated with 10% sodium carbonate solution before being
extracted with dichloromethane. This organic solution was
washed with 10% potassium hexafluorophosphate and then
concentrated (see Figure 1(b)). Co was inserted into the free
base EHO by adding an excess of cobalt(II) acetate to a solution
of the free base EHO dissolved in dichloromethane and methanol
(7:3) at room temperature. Mg EHO was prepared at room
temperature in dichloromethane using magnesium bromide ethyl
etherate and triethylamine (see Figure 1(d)). Mn EHO was
prepared by refluxing the free base EHO with excess manganese
A purpose built sensor testing chamber was used to expose
the solid state sensor films to the analyte vapors (see ref 12 for
more details). The delivery of clean dry nitrogen to a chamber
containing the analyte was controlled with a Tylan FC-260 mass
flow controller. The temperature of the analyte was maintained
at 0 °C by submerging the analyte chamber in iced water,
therefore mixing the N2 gas with the analyte at a known vapor
pressure. For the analytes used, this method of producing analyte

Detection of Volatile Organic Compounds
J. Phys. Chem. B, Vol. 114, No. 36, 2010 11699
Figure 2. Absorbance spectra of the seven porphyrin solutions prior to exposure. Each spectrum is offset by 0.3 au from the previous for clarity,
and the data at wavelengths longer than 500 nm have been multiplied by ten. Note the double peak observed for the Co EHO solution at 417 and
439 nm.
TABLE 1: Absorbance Peak Positions for Each of the Seven Porphyrins Prior to Exposure to the Analytes
absorbance peak positions
solution
Langmuir-Blodgett film
porphyrin
free base
Soret
Q bands
Soret
Q bands
426 nm (405 nm shoulder)
519, 556, 592, and 652 nm 439 nm (asymmetric)
521, 555, 597, and 652 nm
Mg EHO 430 nm (409 nm shoulder)
567 and 610 nm
435 nm (vague shoulder)
571, 613, and 657 nm
Zn EHO
Sn EHO
Au EHO
Co EHO
426 nm (408 nm shoulder)
440 nm
428 nm (398 nm shoulder)
417 and 439 nm
550 and 593 nm
566 and 610 nm
532 nm (570 nm shoulder)
531 nm (550 nm shoulder) 427 nm (single broad peak)
436 nm
-
-
560 and 604 nm
-
-
546 nm
Mn EHO 483 nm (broad feature, 350-450 nm) 533, 588, and 625 nm
480 nm (broad feature, 350-450 nm) none observed
vapor at a known vapor pressure results in concentrations of
approximately 2600 ppm (acetic acid), 32 000 ppm (butanone),
32 000 ppm (ethylacetate), 940 ppm (hexane thiol), 2200 ppm
(hexylamine), 670 ppm (octanal), 3 ppm (octanol), 180 ppm
(octylamine), 15 000 ppm (triethylamine), and 5900 ppm
(trimethyl phosphate) for each of the different analytes cor-
respondingly (further details are provided in Supporting Infor-
mation). The main purpose of this study was to determine the
ability of the different porphyrins to sense different analytes,
and therefore the exact concentration of each analyte used was
not crucial provided it was reasonably high, as they were.
Further studies are required to determine the sensitivity, stability,
reversibility, and dynamics of each sensor action for each
sensor-analyte pair. The cooling of the gas also helped to
reduce cross contamination problems due to condensation of
the analytes onto the walls of the delivery pipes which were at
room temperature and therefore warmer than the analyte vapors.
Light was delivered and collected from the chamber by optical
fibers connected to a World Precision Instruments Spectromate
spectrophotometer which incorporates a tungsten white light
source and a multichannel photodiode detector allowing the
absorption spectra of the film under test to be recorded in situ
every 3 s over the spectral range 350-850 nm. The exhaust
gas passes through an activated charcoal filter to absorb the toxic
vapor before being vented to the atmosphere. The presence of
analyte vapors in the test chamber was found to cause increased
scattering and therefore increased the optical absorbance at all
wavelengths. In these experiments, such effects were accounted
for by subtracting a reference absorbance (measured at 700 nm
where all the samples investigated are transparent) from the
recorded experimental data. The before and after exposure
spectra and the calculated difference spectrum for each of the
thin film analyte pairs studied are included in the Supporting
Information.
3. Results and Discussion
Porphyrins possess a delocalized π-electron system associated
with the highly conjugated porphyrin ring. Upon exposure with
light with a photon energy greater than the highest occupied
molecular orbital (HOMO) to lowest unoccupied molecular
orbital (LUMO) energy gap, photons are absorbed causing
excitation of electrons from the HOMO to the LUMO. The
energy separation between the HOMO and LUMO levels is such
that the photons absorbed are in the visible wavelength range
giving the molecules their colorful appearance. The spectra
generally consist of a strong Soret absorbance at shorter
wavelengths (between 350 and 500 nm) and a number of Q-band
absorbance features at longer wavelengths (>500 nm). The initial
absorption spectra for the seven different EHO solutions are
shown in Figure 2, and their peak positions are recorded in Table
1. There are four Q bands for the free base EHO, but for the
metalated EHOs there are only two or three Q bands. This is
because the metalated EHOs have increased symmetry, therefore
making the energy levels involved in the Q-band absorption
degenerate and thus reducing the number of visible absorption
bands.

11700 J. Phys. Chem. B, Vol. 114, No. 36, 2010
Dunbar et al.
Figure 3. Summary of the sensor response for the seven porphyrins in chloroform solution with various organic analytes.
The absorption spectra of the different analytes were mea-
sured, and it was found that all the analytes are highly
transparent from 350 to >850 nm. Short wavelength (<350 nm)
absorption features which were observed in some of the
interaction experiments (data not shown) containing octanal,
butanone, and triethylamine are simply due to absorption by
unreacted analyte in the solution rather than due to any
interaction with the porphyrin, as observed by measuring the
analyte and solvent (without any porphyrin) references.
To be useful as an optical sensor, the absorption spectrum of
the sensor material must change upon exposure. The interaction
upon exposure of 2 mL of the seven different porphyrin
solutions to the different analytes is summarized in Figure 3.
The difference between the unexposed and exposed spectra was
calculated and then squared (such that all differences are
positive). The data plotted in Figure 3 are the average of the
squared difference in the range 350-750 nm.
For the porphyrin-anaylte pairs that showed strong responses
in solution, LB films of the porphyrin were produced and tested
with the corresponding analyte. Upon transferring them to a
solid substrate to form an LB film, the optical absorption peaks
broaden considerably. In some instances, there are also signifi-
cant shifts in the peak absorption wavelength, usually to longer
wavelengths. These changes are assumed to be due to either
the removal of solvent effects or aggregation of the porphyrins
upon transferring them to a solid substrate or both. Details of
the LB film absorption peaks for the LB films studied are given
in Table 1. When using solid state LB films rather than solution,
in general the color change upon exposure to the various anayltes
is more difficult to detect since the absorption peaks are now
broader. However, to function as practical gas sensing elements,
it is advantageous to have a solid state sensor. Therefore, despite
the reduction of the spectral changes observed for the solid state
films it is still worthwhile studying their behavior as gas sensors.
Figure4showsthesensorresponsefortheselectedporphyrin-analyte
pairings that had a strong response in solution, and again the
difference between the unexposed and exposed spectra was
calculated and squared.
Figure 4. Summary of the sensor response for LB films of the
porphyrins with various organic analytes. Only the sensor-analyte pairs
which generated strong responses in the solution experiments were
conducted (in some cases the acetic acid result is not reported since it
destroyed the LB films). NB. the Co EHO responses are very strong
and therefore have been reduced by an order of magnitude so all the
data can all be presented on the same chart. The Co EHO response to
trimethyl phosphate has also been truncated.
broader, lower intensity peak at longer wavelength (∼465 nm),
and the four Q bands initially present are replaced with a single
broader but more intense absorbance band at much longer
wavelength (∼700 nm). A similar reaction for free base EHO
upon exposure to NO2 is reported in ref 15. No other significant
responses were observed for the free base EHO in solution.
When the free base EHO was fabricated into a solid LB film,
its response to trimethylphosphite is considerably weaker than
in solution. Therefore, gas sensors made from the free base EHO
are highly selective, only responding significantly to NO2 as
reported elsewhere.15
In the case for the free base EHO when exposed to acetic
acid and trimethyl phosphite, the Soret peak is replaced by a
Magnesium(II) porphyrins have the tendency to axially bind
a fifth and sixth ligand sequentially. It is clear in the literature

Detection of Volatile Organic Compounds
J. Phys. Chem. B, Vol. 114, No. 36, 2010 11701
that hard ligands (such as oxygen and nitrogen) are greatly
favored over soft ligands.19 This may explain the occurrence of
responses for Mg EHO exposed to hexylamine and octanal. It
was noted that for Mg EHO exposed to acetic acid there was a
slow interaction during which the absorbance spectrum changed
from that which is characteristic of Mg EHO to that which is
typical for the free base EHO before then partially undergoing
a further change similar to that observed for free base EHO
upon exposure to acetic acid. It has been well documented that
Mg (II) porphyrins are demetalated by acids to form the
corresponding free base porphyrin.18 It is therefore likely that
exposure to acetic acid causes the conversion of Mg EHO back
to the free base EHO. When transferred to a solid LB film, Mg
EHO has strong responses to hexylamine and octanal implying
that Mg EHO has promise as a sensor material; however,
demetalation of the Mg EHO upon exposure to acetic acid as
occurs in solution is observed, therefore the chemical stability
of Mg EHO films may not be sufficient to make robust sensors.
This is also thought to be the reason why the LB film of Mg
EHO displays 3 Q bands (Table 1); i.e., the as-formed film may
actually be a blend of free base EHO and Mg EHO and as such
has a broad Soret, and 3 Q bands can be distinguished.
analytes. For this reason, LB films of Au EHO were not
fabricated and tested.
The absorption spectrum for the Co EHO solution is more
complicated than the other EHOs. It has two Soret peaks which
are attributed to the presence of both Co(II) EHO and Co(III)
EHO in the solution. Stirring the solution tends to cause a shift
toward the longer wavelength peak. The absorption spectrum
of the Co EHO in solution has two Soret peaks as shown in
Figure 2, and upon exposure one of these two peaks becomes
dominant. Which peak becomes dominant depends upon whether
the analyte used is oxidizing or reducing. Therefore, Co EHO
is a very promising material for further development of gas
sensing materials. Upon exposure to ethyl acetate no change is
observed, and both Soret peaks remain. For butanone, hexy-
lamine, octanol, octanal, octylamine, and triethylamine, the Soret
peak at 417 nm reduces or disappears, and the Soret peak at
439 nm increases in intensity. However, for hexanethiol and
acetic acid, the opposite occurs; i.e., the 417 nm peak increases
in intensity, and the 439 nm peak disappears. The different
groups of analytes cause a shift in the balance of the mixed
Co(II)/Co(III) to become either Co(II) or Co(III). Upon exposure
to trimethylphosphite, the 417 nm peak remains unchanged, but
the 439 nm peak shifts to 453 nm, presumably due to a complex
occurring for only one of the two oxidation states of Co EHO,
presumably Co(III) as the +3 oxidation state will not be
completely satisfied by complexing to the porphyrin. When
transferred to an LB film and exposed to the various analytes,
the response of Co EHO is considerably larger than for any of
the other porphyrins (the scale for the Co data in Figure 4 is
reduced by an order of magnitude, and the trimethylphosphite
column is truncated). The strength of absorbance change for
Co EHO upon exposure is thought to be due to the mixed Co(II)/
Co(III) system switching to become either Co(II) or Co(III) upon
exposure. This allows the Co EHO to coordinate different
numbers of ligands depending upon its oxidation state which
results in a significant change in the symmetry of the porphyrin
ring and hence a large spectral change. Therefore, utilizing this
change of oxidation state and the corresponding changes in the
absorption spectra holds great promise for the development of
highly sensitive toxic gas sensors.
Sn(IV) porphyrins are known to bind strongly to oxygen-
containing ligands (such as alcohols) forming 6-coordinate
complexes.20 Therefore, a response to octanol was expected.
However, in our experiments no response is observed for Sn
EHO upon exposure to any of the analytes. The Sn EHO when
synthesized already has two axial hydroxyl ligands to satisfy
the +4 oxidation state of the Sn metal atom. The presence of
these ligands may explain the longer wavelength of the Soret
peak observed in the unexposed Sn EHO absorbance spectrum.
Upon exposure, either the analytes do not exchange with these
hydroxyl ligands or if they do this may not affect the absorption
spectrum significantly and hence no appreciable optical response
is observed. Since our experiments showed no interesting
responses for the Sn EHO in solution, Sn EHO LB films were
not produced and tested.
Zinc(II) porphyrins are well-known to coordinate nitrogen
ligands to form 5-coordinate complexes16 with an associated
change in the Soret absorption spectrum because the metal is
pulled slightly out of the plane of the porphyrin in the complex
that is formed. The trend for Zn EHO upon exposure to amines
is for the Soret absorbance band to both shift slightly to longer
wavelength (from 426 to ∼433 nm) and and increase in
intensity. Meanwhile, the two Q bands shift to longer wave-
lengths, and often a third Q bands appears. The magnitude of
the Soret band shift and the intensity change appear to be
dependent upon the amine used during the exposure. Zn EHO
initially is thought to have no ligands but upon exposure forms
coordinates with the amines, and therefore the particular amine
involved will determine the change in symmetry that the
porphyrin experiences. Upon transferring to an LB film, the Zn
EHO continues to respond to the amines, albeit weakly to the
presence of amines. LB films of Zn EHO as an amine detector
were the focus of further investigations.17,21
The Au EHO when synthesized is a salt with PF6- counterion.
This may explain the broad features observed in the absorbance
spectrum for unexposed Au EHO. The Au porphyrin was
expected to exhibit an interaction with hexanethiol since it is
known that Au readily forms complexes with sulfur-containing
compounds. However, no interactions were observed with any
of the analytes. Therefore it is assumed that the Au EHO salt is
similar to the Sn EHO in that it does not readily exchange
ligands and therefore forms new complexes with any of the
Presumably, other metal ions which can exist in more than
one oxidation state should also hold great promise as gas sensors
when incorporated into the porphyrin system. Mn can also adopt
more than one valence state. When it is incorporated into the
porphyrin ring, it rapidly oxidizes to Mn(III) and takes on two
different ligands, namely, a water molecule and an OH- ion.
These asymmetric ligands mean that the Mn atom is slightly
pulled out of the plane of the porphyrin ring before interaction
with any analytes. It is this asymmetry that is thought to result
in Mn EHO’s atypical absorption spectrum (see Figure 2). In
solution, when exposed to amines a small spectral change is
observed. The Soret band and the Q-bands shift to slightly
shorter wavelengths. It is assumed in this case that the Mn EHO
exchanges ligands upon exposure to amines, and this exchange
causes a small perturbation of the Mn ion position resulting in
an optical change. For Mn EHO, it is thought that the Mn
remains in the Mn(III) state unlike the case of Co EHO where
a transition from Co(II) to Co(III) is thought to occur. When
transferred to an LB film, the Mn EHO continues to show small
but significant absorption changes when exposed to amine
analytes, and as such Mn EHO holds considerable potential to
be used as an amine sensor.22
The response of all the selected porphyrin-analyte pairs
shown in Figure 4 are comparable to, or stronger than, that of

11702 J. Phys. Chem. B, Vol. 114, No. 36, 2010
Dunbar et al.
the Zn EHO-amine response. Previous studies of Zn EHO have
shown that it can be successfully used to fabricate an alkylamine
detector.17,21 Therefore, all these porphyrin-analyte pairs high-
lighted in Figure 4 have sufficiently strong responses to be used
to develop toxic gas sensors.
water bath was estimated by extrapolating the data found in
the CRC handbook of chemistry and physics: a ready reference
book of chemical and physical data, 58th ed.; editor Robert C.
Weast; Cleveland (Ohio): CRC Press, 1977. The results are
shown in table S1. This material is available free of charge via
the Internet at http://pubs.acs.org.
4. Conclusions
The responses of several different metallo-porphyrins to a
broad range of analytes have been investigated to determine
their suitability as optical-based thin film sensors. Those
porphyrins that showed changes in their absorbance spectra upon
exposure hold considerable potential to be used as gas sensors
for the corresponding analyte. It is the process of gaining a
ligand or ligand exchange which results in a change of the
porphyrin symmetry that results in the spectral changes observed
and therefore is a necessary condition to make an effective
optical-based gas sensor.
UpontransferringtheporphyrinstosolidstateLangmuir-Blodgett
films, spectral changes similar to those observed in solution
occurred most of the time. Therefore, porphyrins that readily
exchange or coordinate extra ligands in solution are shown to
be suitable materials for solid state colorimetric detectors for
volatile organic gases. However, porphyrins that already have
strongly attached axial ligands when synthesized only show a
sensor response to those analytes that can substitute these axial
ligands, and as such Au EHO and Sn EHO showed no response
to any of the anayltes.
In particular, the response of Co EHO was very strong upon
exposure to acetic acid, butanone, hexylamine, octylamine,
triethylamine, octanol, octanal, and trimethyl phosphate. This
is attributed to switching between the Co(II) and Co(III),
resulting in the formation of either fewer or more axial ligands,
respectively, with large associated spectral changes. However,
not all metallo-pophyrins that can exist in more than one
oxidation state showed such strong responses; for example, it
is proposed that Mn EHO did not switch oxidation state but
merely exchanged ligands therefore displaying considerably
smaller spectral changes. While the spectral changes displayed
by Mn EHO were considerably smaller than those for Co EHO,
they are comparable to those observed for Zn EHO as an amine
sensor.17,21 Therefore,itisconcludedthatalloftheporphyrin-analyte
pairs reported in Figure 4 (apart from Mg EHO which was
readily demetalated and therefore not sufficiently stable) are
sufficiently strong to be used as effective toxic gas sensors.
References and Notes
(1) Mandelis; Christofides, C. Physics, Chemistry, and Technology of
Solid State Gas Sensor DeVices; Wiley-Interscience Publications: New York,
1993.
(2) Albert, K. J.; Lewis, N. S.; Schauer, C. L.; Sotzing, G. A.; Stitzel,
S. E.; Vaid, T. P.; Walt, D. R. Chem. ReV. 2000, 100, 2595.
(3) Moseley, P. T.; Norris, J. O. W.; Williams, D. E. Techniques and
Mechanisms in Gas Sensing, Adam Hilger Series; Taylor and Francis:
Bristol, 1991.
(4) Lonergan, M. C.; Severin, E. J.; Doleman, B. J.; Beaber, S. A.;
Grubb, R. H.; Lewis, N. S. Chem. Mater. 1996, 8 (9), 2298.
(5) Capone, S.; Mongelli, S.; Rella, R.; Siciliano, P.; Valli, L. Langmuir
1999, 15 (5)), 1748.
(6) George, C. D.; Richardson, T.; Hofton, M. E.; Vale, C. M.; Neves,
M.G. M.; Cavaleiro, J. A. S. Mater. Sci. Eng. C 1999, 8-9, 559.
(7) Richardson, T.; Smith, V. C.; Johnstone, R. A. W.; Sobral,
A. J. F. N.; Rocha-Gonsalves, A.M.d’A. Thin Solid Films 1998, 327-329,
3151.
(8) Dooling, C. M.; Worsfold, O.; Richardson, T. H.; Tregonning, R.;
Vysotsky, M. O.; Hunter, C. A.; Kato, K.; Shinbo, K.; Kaneko, F. J. Mater.
Chem. 2001, 11, 392.
(9) Worsfold, O.; Dooling, C. M.; Richardson, T. H.; Vysotsky, M. O.;
Tregonning, R.; Hunter, C. A.; Malins, C. J. Mater. Chem. 2001, 11, 399.
(10) Pedrosa, J. M.; Dooling, C. M.; Richardson, T. H.; Hyde, R. K.;
Hunter, C. A.; Mart´ın, M. T.; Camacho, L. J. Mater. Chem. 2002, 12, 2659.
(11) Suslick, K. S.; Rakow, N. A.; Sen, A. Tetrahedron 2004, 60, 11133.
(12) Dunbar, A. D. F.; Richardson, T. H.; McNaughton, A. J.; Hutch-
inson, J.; Hunter, C. A. J. Phys. Chem. B 2006, 110 (33), p16646–16651.
(13) Dunbar, A. D. F.; Richardson, T. H.; Hutchinson, J.; Hunter, C. A.
Sens. Actuators, B 2008, 128 (2), 468–481.
(14) Sanders, J. K. M.; Bampos, N.; Clyde-Watson, Z.; Darling, S. L.;
Hawley, J. C.; Kim, H.-J.; Mak, C. C.; Webb, S. J. The Porphyrin Handbook
(Vol. 3, Chapter 15: The axial coordination chemistry of metalloporphyrins);
Academic Press: New York, 1999.
(15) Dunbar, A.; Richardson, T. H.; McNaughton, A. J.; Barford, W.;
Hutchinson, J.; Hunter, C. A. Colloids Surf., A 2006, 284-285, 339–
344http://www.sciencedirect.com/science/article/B6TFR-4HX47CK-4/2/
fa0d2648ca551e71b1a2f0ba76e29b18.
(16) Taylor, P. N.; Anderson, H. L. J. Am. Chem. Soc. 1999, 121, 11538–
11545.
(17) Dunbar, A. D. F.; Richardson, T. H.; McNaughton, A. J.; Cadby,
A.; Hutchinson, J.; Hunter, C. A. J. Porphyrins Phthalocyanines 2006, 10
(07), p978.
(18) Dolphin, D. The Porphyrins: Structure and Synthesis, Part A;
Academic Press: London, 1978; Vol. 1.
(19) Kadish, K. M.; Shuie, L. R. Inorg. Chem. 1982, 21, 1112.
(20) Holloway, C. E.; Melnik, M. Main Group Met. Chem. 1998, 21,
371.
(21) Brittle, S.; Richardson, T. H.; Dunbar, A. D. F.; Turega, S.; Hunter,
C. A. J. Phys. Chem. B 2008, 112 (36), 11278–11283.
(22) Brittle, S.; Richardson, T. H.; Hutchinson, J.; Hunter, C. A. Colloids
Surf., A 2008, 321 (1-3), 29–33.
Acknowledgment. Dr W. Barford is acknowledged for his
helpful discussions. EPSRC (UK) Grant No GR/596845/01
provided the financial support for this work.
Supporting Information Available: The analyte vapor
concentration for each of the analytes held at 0 °C in the iced
JP102755H