Study of the aqueous photochemistry of 4-fluorophenol, 4-bromophenol and 4-iodophenol by steady state and nanosecond laser flash photolysis

Article abstract of DOI:10.1039/a705287a

The mechanisms of the aqueous photoreactions of 4-fluorophenol, 4-bromophenol and 4-iodophenol have been studied by steady state photolysis and by nanosecond laser flash photolysis. The same photoproducts are obtained as were found in the photolysis of aqueous 4-chlorophenol although the photoreaction quantum yields vary considerably across the series of compounds. As for 4-chlorophenol the carbene 4-oxocyclohexa-2,5-dienylidene is formed from the halogenophenol by loss of HX and this species reacts efficiently with oxygen to form 1,4-benzoquinone O-oxide which subsequently yields 1,4-benzoquinone. Consideration of the reaction quantum yields together with the photophysical properties of the four halogenophenols suggests that the carbene is derived from the first excited singlet state rather than a triplet excited state.

Full text of DOI:10.1039/a705287a

View Article Online / Journal Homepage / Table of Contents for this issue
Study of the aqueous photochemistry of 4-fluorophenol,
-bromophenol and 4-iodophenol by steady state and
4
nanosecond laser flash photolysis
a
,a
b
Anne-Pascale Durand, Robert G. Brown,* David Worrall and
Francis Wilkinson
a
b
Chemistry Dept., University of Central Lancashire, Preston, UK PR1 2HE
Chemistry Dept., Loughborough University of Technology, Loughborough, UK LE11 3TU
b
The mechanisms of the aqueous photoreactions of 4-fluorophenol, 4-bromophenol and 4-iodophenol
have been studied by steady state photolysis and by nanosecond laser flash photolysis. The same
photoproducts are obtained as were found in the photolysis of aqueous 4-chlorophenol although the
photoreaction quantum yields vary considerably across the series of compounds. As for 4-chlorophenol
the carbene 4-oxocyclohexa-2,5-dienylidene is formed from the halogenophenol by loss of HX and this
species reacts efficiently with oxygen to form 1,4-benzoquinone O-oxide which subsequently yields
1
,4-benzoquinone. Consideration of the reaction quantum yields together with the photophysical
properties of the four halogenophenols suggests that the carbene is derived from the first excited singlet
state rather than a triplet excited state.
The photochemistry of 4-chlorophenol (4-CP, 1) in aqueous
solution is currently enjoying a surge of interest due to its use as
a model compound for the aqueous photochemistry of chlorin-
ated aromatics and for assessing the effectiveness of advanced
By contrast, the aqueous photochemistry of the other
4-halogenophenols has been almost completely neglected.
12
Lipczynska-Kochany
has determined the photoproducts
from aqueous 4-bromophenol (4-BP, 6) to be 1,4-benzoquinone
(major product) and 1,4-hydroquinone (minor product)
together with small amounts of 2-hydroxy-1,4-benzoquinone 7
which is a known photoproduct of 2 and almost certainly arises
from secondary photolysis. This product mix is very similar to
1–4
oxidation methods. In aqueous solution 4-CP undergoes a
range of photochemical reactions depending on the presence of
oxygen, the concentration of the chlorophenol and the pH. At
neutral or acidic pH values, where the 4-CP is not ionised, and
in the presence of oxygen, the principal photoproduct is 1,4-
5–8
that observed from 4-chlorophenol and Grabner et al. also
5–8
benzoquinone 2.
A range of minor products are also
report that aqueous 4-bromophenol gives a similar distribution
10
observed, including 1,4-hydroquinone 3, 2-hydroxy-1,4-benzo-
of photoproducts to 4-chlorophenol. Further work on 4-BP
5,7
13
quinone and several polyhydroxybiphenyls, the latter becom-
ing particularly prevalent when the solution is oxygen free.
There has been some controversy as to whether 1,4-benzo-
quinone is a photoproduct in the absence of oxygen but it is
by Lipczynska-Kochany and Kochany using EPR with a spin
trap yielded four possible radical intermediates; namely an aryl
radical (possibly a 4-hydroxyphenyl radical), hydrated elec-
trons, hydroxy radicals and the para-benzosemiquinone anion.
Again, these findings are in line with similar experiments on 4-
7,9
now clear that it is not and that earlier work which suggested
6,8
14
the opposite was in error due to insufficiently rigorous degas-
sing of the solutions.
chlorophenol. However, they are clearly at odds with the flash
10,11
photolysis results
and the similarity of the products for 1
The mechanism of 4-CP photolysis in water has very recently
been shown to involve the formation of, firstly, 4-oxocyclohexa-
and 6 would suggest that the photolysis mechanisms for the two
compounds are very similar.
2
,5-dienylidene 4, followed by 1,4-benzoquinone O-oxide 5 if
We have undertaken a fairly comprehensive study of the
10,11
8,9,11
molecular oxygen is present (Scheme 1).
These two radical
aqueous photochemistry of 4-chlorophenol.
In order to
species are readily observable by nanosecond flash photolysis.
gain further insight into the aqueous photochemistry of this
class of compounds we have also undertaken experiments on
15
the other three commonly available 4-halogenophenols. These
experiments appear to both confirm that the mechanistic details
for 1 apply across the other 4-halogenophenols and provide
insight into the reactive excited state. We report the results of
those experiments here.
OH
O
OH
O
OH
hν
Cl
O
OH
O
O
4
–CP
2
3
Experimental
4
-Chlorophenol was obtained from Aldrich Chemicals Ltd
OH
O
and was sublimed (40–45 ЊC, 11 mbar) before use. 1,4-Benzo-
quinone, 4-bromophenol, 4-iodophenol and 1,10-phenanthro-
line were obtained from Lancaster Synthesis Ltd. 1,4-Hydro-
quinone, 4-fluorophenol and the four 4-halogenoanisoles were
obtained from BDH Ltd. Water was distilled and deionised and
absolute ethanol was Burroughs AR reagent grade.
hν
O2
H2O
Cl
O
O
4
–CP
4
5
An Applied Photophysics Ltd. KR2 microsecond flash
spectrometer was used to photolyse the samples contained in
cylindrical quartz cuvettes of 10 cm length and 2 cm diameter.
Scheme 1 Photoproducts and mechanism for the photoreaction of
-chlorophenol in neutral, aerated aqueous solution at ambient
temperature
4
J. Chem. Soc., Perkin Trans. 2, 1998
365

View Article Online
a
Table 1 Absorption and emission properties of the 4-halogenophenols and 4-halogenoanisoles in water and ethanol
Absorption
Fluorescence
3
Ϫ1 Ϫ1
Compound
Solvent
λmax/nm
ε/dm mol
s
λmax/nm
φf
τf /ns
4
4
4
4
4
4
4
4
4
4
4
4
-FP
-CP
-BP
-IP
-FP
-CP
-BP
-IP
-FA
-CA
-BA
-IA
Water
Water
Water
Water
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
Ethanol
278
280
280
280
282
282
282
282
280
280
280
280
2230
1400
1370
1250
2450
1770
1600
1490
2280
1700
1500
1340
310
312
310
—
310
310
307
—
306
308
308
—
0.144
0.015
0.0004
<0.0001
0.162
0.048
0.002
<0.0001
0.166
0.023
0.0004
<0.0001
2.1
0.4
<0.1
—
1.5
0.4
<0.1
—
2.0
0.5
<0.1
—
a
The three iodo compounds were non-fluorescent under the instrumental conditions used here. Fluorescence quantum yields are ±10ꢀ and lifetimes
0.2 ns.
±
Flash energies of approximately 50 J were used with a flash
duration of approximately 25 µs FWHM. Product analysis
was carried out by HPLC with a Perkin-Elmer Series 10
LC pump, guard column and a Pecosphere-3*3C-C18 column,
a Perkin-Elmer LC-235 UV detector and a Perkin-Elmer
model 1020 integrator. The detection wavelength was 280
nm. The mobile phase used was acetonitrile–water (20:80)
acidified to pH 4.5 with orthophosphoric acid. A flow rate
3
Ϫ1
of 2 cm min was chosen. Steady state photolyses were
also carried out in a Photochemical Reactors Ltd. model
MLU18 photoreactor. Quantum yields of photoreaction
were determined using this system together with ferrioxalate
16
actinometry.
Absorption spectra were recorded on a Hewlett-Packard
model 8452A diode-array spectrometer. Fluorescence spectra
were measured using a Spex Fluoromax spectrofluorimeter.
Fluorescence decay profiles were measured by the time-
17
Fig.
1
Variation of the concentration of 4-fluorophenol (᭹),
,4-benzoquinone (᭺), 1,4-hydroquinone (᭿) and the material un-
correlated single photon counting technique at the EPSRC
Daresbury Laboratory. The equipment used at Daresbury has
1
accounted for (᭡) as a function of flash number in the photolysis of
aerated aqueous 4-fluorophenol (100 µ)
18
been described elsewhere. The decay profiles were analysed by
computer convolution and were uniformly found to be single
2
exponential decays on the basis of the calculated χ value and
rate constants (k ) and rate constants for the sum of the non-
f
the distribution of residuals.
emission processes reveals clear increases in the latter with the
increasing atomic number of the halogen for all three series of
compound/solvent. These data have to be interpreted with
some caution given that a substantial contribution to this ‘non-
radiative’ rate constant arises from ring-halogen bond scission
leading to the observed photoproducts. However, as discussed
later, even when this is taken into account a substantial heavy
atom effect remains, implicating the singlet state. On the basis
Nanosecond flash photolysis was carried out with an HY200
Nd: YAG laser (Lumonics) at an excitation wavelength of
1
1
2
66 nm. The full instrumentation is described elsewhere.
Ϫ3 Ϫ3
A concentration of approximately 1.0 × 10 mol dm was
used for the majority of the measurements. This gives an
absorbance at 266 nm of approximately 0.4. Samples were only
subjected to a single flash before they were replaced by a fresh
solution.
of the k values, we can estimate that the fluorescence lifetime
f
of 4-BP is of the order of 10 ps and for 4-IP much less than
this.
Results
Absorption and emission properties of the four 4-halogeno-
phenols in solution at ambient temperature
The absorption and emission properties of the four 4-
halogenophenols (4-FP to 4-IP) were measured in aqueous and
ethanolic solution and are presented in Table 1 alongside the
corresponding data for the four 4-halogenoanisoles (4-FA to
Photolysis of aerated aqueous 4-fluorophenol, 4-bromophenol
and 4-iodophenol
Photolysis of aerated aqueous 4-FP, 4-BP and 4-IP was carried
out under the same experimental conditions as used for the
8,9
aqueous photolysis of 4-CP. In the case of 4-FP, the first (and
major) photoproduct identified was 1,4-benzoquinone. This
appeared immediately in the photolysis (whose time course is
shown in Fig. 1) and was followed by the appearance of 1,4-
hydroquinone and 2-hydroxy-1,4-benzoquinone at later stages
of the photolysis. A fourth, as yet unidentified photoproduct,
was also obtained which is not present in the photolysis prod-
ucts of any of the other 4-halogenophenols. We are currently
trying to identify this photoproduct using LC/MS as previ-
4
-IA). It can be seen that the absorption and emission maxima
are almost completely independent of molecular structure and
solvent. In addition, the other properties given in Table 1 are
also relatively independent of the other variables (i.e. solvent
and the hydroxy/methoxy group) for a given halogen.
The extinction coefficients exhibit a significant decrease with
increasing atomic number of the halogen in both solvents, pre-
sumably reflecting the decreasing electron-withdrawing power
of the halogen as it gets heavier. The fluorescence quantum
yield and lifetime data when taken together show, as might be
expected, a clear heavy atom effect. Calculation of radiative
9
ously. For the first few flashes, the amount of 4-FP that had
photoreacted was entirely accounted for by the 1,4-benzo-
quinone formed. After further photolysis, the majority of
the consumed 4-FP is accounted for by the 1,4-benzo-
3
66
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
Fig. 2 Variation of the concentration of 4-bromophenol (᭹), 1,4-
benzoquinone (᭺), 1,4-hydroquinone (᭿) and the material un-
accounted for (᭡) as a function of flash number in the photolysis of
aerated aqueous 4-bromophenol (100 µ)
Fig. 4 Transient absorption spectra for 4-fluorophenol (1.0 m) in
aerated water at delay times of 0.21 µs (top spectrum), 0.88, 2.48, 5.59
and 9.27 µs (bottom spectrum). Absorbance range for each spectrum
0
–0.1.
determine the reaction quantum yield; the strength of the bond
between the halogen and the aromatic ring and a heavy atom
effect on the rate of non-radiative decay. The former acts to
increase the quantum yield with increasing atomic weight
whereas the latter works in the opposite direction. It appears
from the quantum yield data that the heavy atom effect is the
more dominant of the two.
Nanosecond flash photolysis of 4-fluoro-, 4-bromo- and 4-iodo-
phenol in aerated aqueous solution
Aqueous solutions of 4-FP, 4-BP and 4-IP were subjected
to nanosecond laser flash photolysis for comparison with 4-
chlorophenol. 4-Fluorophenol gives virtually identical results
to 4-chlorophenol. The transient spectra which are obtained
from a typical set of experiments are shown in Fig. 4. Com-
Fig. 3 Variation of the concentration of 4-iodophenol (᭹), 1,4-
benzoquinone (᭺), 1,4-hydroquinone (᭿) and the material un-
accounted for (᭡) as a function of flash number in the photolysis of
aerated aqueous 4-iodophenol (100 µ)
11
parison with the corresponding data for 4-CP shows that
these spectra are very similar to those for 4-CP in both their
shape and overall intensity. We therefore conclude that the same
two transient species are formed in both cases with approxi-
mately the same yields. Analysis of the individual kinetic traces
gave, as expected on the basis of this conclusion, very similar
results to those found for 4-chlorophenol. Above 400 nm the
traces were adequately fitted (see Fig. 5) by eqn. (1), with
quinone and 1,4-hydroquinone photoproducts but by the end
of the photolysis, there is a significant amount of material
unaccounted for as shown in Fig. 1.
1
,4-Benzoquinone and 1,4-hydroquinone were also identified
as photoproducts of 4-BP photolysis whose time course is
shown in Fig. 2. 2-Hydroxy-1,4-benzoquinone was not identi-
fied among the photoproducts although it was detected by
12
Lipczynska-Kochany. This is almost certainly because the
amount of 4-BP photolysed is much less than for 4-FP and
Absorbance = A exp (Ϫ k t) ϩ A exp (Ϫ k t) (1)
1
1
2
2
6
Ϫ1
4
-CP so that the concentration of the 1,4-benzoquinone, which
A = ϪA , k = 0.95 ± 0.10 × 10 s (compared with 1.0 ± 0.1 ×
1 2 1
6 Ϫ1 4 Ϫ1
is the precursor of 2-hydroxy-1,4-benzoquinone, is much lower
than for the other compounds. Photolysis of aqueous 4-IP (Fig.
10 s for 4-CP) and k = 6.7 ± 1.0 × 10 s (compared with
2
4 Ϫ1
8.5 ± 1.0 × 10 s for 4-CP). The slight difference in the values
3
) yielded 1,4-benzoquinone as the only detectable and con-
of k and k for 4-fluoro- and 4-chloro-phenol are not con-
1
2
firmed photoproduct. The photolysis rate for 4-IP was once
again greatly reduced compared even to 4-BP. As with 4-FP and
sidered to be significant.
Below 400 nm, where both transients absorb, the traces are
less easy to analyse. At some wavelengths, the observed absorp-
tion changes can be fitted with a single exponential decay plus a
constant term whereas data at other wavelengths require the
double exponential form of eqn. (1) (but with A ≠ ϪA and
4
-CP, the amount of material unaccounted for in the identified
photoproducts steadily builds up during the photolyses of both
4
-BP and 4-IP.
Based on the amount of the halogenophenol remaining at
1
2
the end of thirty flashes, the reactivity of the four compounds is
-CP > 4-FP > 4-BP > 4-IP, and this is reflected in the quan-
|A | < |A |) to achieve an acceptable fit. The relative weakness of
1 2
4
the transient signals below 400 nm means that we have less
confidence in the fitted parameters than at the higher wave-
lengths but similar rate constant values to those given above are
observed at the shorter wavelengths. This supports the sugges-
tion of an initially formed transient which absorbs below 400
nm and which decays to form a second transient, which then
itself decays i.e. there is a clear parent–daughter relationship
tum yields for the consumption of the halogenophenol which
were found to be 0.31, 0.44, 0.08 and 0.022 for 4-FP, 4-CP, 4-BP
and 4-IP, respectively. These quantum yields were determined
by ferrioxalate actinometry based on photolysis of р10ꢀ of
the initial halogenophenol concentration.
We conclude that there are two competing effects which
J. Chem. Soc., Perkin Trans. 2, 1998
367

View Article Online
Fig. 7 Transient absorption of 4-bromophenol (1.0 m) in aerated
Fig. 5 Transient absorption of 4-fluorophenol (1.0 m) in aerated
aqueous solution monitored at 470 nm, together with biexponential
fitted curve and residuals
aqueous solution monitored at 470 nm, together with biexponential
fitted curve and residuals
was flashed under identical conditions. Little or no transient
absorption was obtained and this was also the result for oxygen-
saturated and freeze–thaw degassed samples.
Discussion
All four 4-halogenophenols appear to exhibit very similar
photochemistry in aerated aqueous solution in terms of the
nature of the photoproducts obtained under steady-state
photolysis as well as the nature of the transient species observed
following nanosecond flash excitation (excluding 4-IP where
any transient absorption was too weak to detect). Comparison
of the four compounds at a similar stage of photolysis reveals
very clearly the similarities between the compounds in terms of
the photoproducts but also particularly highlights the effect of
prolonged photolysis. For 4-CP, the most reactive of the four, a
single flash is sufficient to destroy approximately 10ꢀ of the
8
compound and this loss is almost entirely accounted for by the
two main photoproducts (1,4-benzoquinone and 1,4-hydro-
quinone); there is only a small amount of material unaccounted
for.
For the less reactive compounds, a higher and higher number
of flashes is required to achieve the same level of conversion of
the starting material and it is noticeable that the yields of the
two main photoproducts decrease with increasing flash number
due to their participation in secondary photochemistry. This is
also evidenced by the increasing amount of material which is
unaccounted for by these two main photoproducts. The most
extreme case is 4-IP where only about one third of the iodo-
phenol which has been destroyed after 30 flashes is accounted
for in the main photoproducts; two thirds of the photolysed
material is present as secondary photoproducts of various
kinds. These observations are in complete agreement with the
quantum yields for the disappearance of the halogenophenol
and the yield of transient species during nanosecond laser flash
photolysis of the four compounds.
Fig. 6 Transient absorption spectra for 4-bromophenol (1.0 m) in
aerated water at delay times of 0.21 µs (top spectrum), 0.88, 2.48, 5.59
and 9.23 µs (bottom spectrum). Absorbance range for each spectrum
0
–0.05.
between the two transients. Thus below 400 nm the kinetic
traces are a combination of the decay of the first transient
species together with the rise and subsequent decay of the
second transient.
In the case of 4-bromophenol, transient spectra were
obtained which were the same shape as for the previous two
halogenophenols (Fig. 6) but which were of reduced intensity.
It appears that the same transients are being produced as for
4
-FP and 4-CP but with lower yields. Analysis of the kinetic
traces above 400 nm on the basis of eqn. (1) gave A = ϪA ,
The observation of identical initial photoproducts and trans-
ients during the flash experiments (both albeit with different
yields for the different compounds) would suggest that the
photochemistry of 4-FP, 4-BP and 4-IP in aerated, aqueous
solution follows the same path as 4-CP in that loss of HX leads
to the formation of 4-oxocyclohexa-2,5-dienylidene 4 followed
by 1,4-benzoquinone O-oxide 5 as shown in Scheme 1. These
assignments appear to be at odds with the transient species
1
2
6
4
Ϫ1
k = 0.94 ± 0.10 × 10 and k = 5.2 ± 1.0 × 10 s . Fig. 7 shows
1
2
an example of one of these analyses. Once again the traces
at wavelengths less than 400 nm are more difficult to analyse,
partly as a result of the reduced intensity of the transients and
partly because of the presence of the same two overlapping
bands as above. However, identical comments about these
kinetic traces for 4-BP may be made as for the corresponding
profiles for 4-FP.
13,14
detected by spin-trapping–EPR.
However, it is possible
that the ‘aryl radical’ detected by Lipczynska-Kochany and
Ϫ3
Ϫ3
Finally, a 10 mol dm aqueous solution of 4-iodophenol
3
68
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
Kochany in the photolysis of 4-CP and 4-BP could be the
carbene 4-oxacyclohexa-2,5-dienylidene and that 1,4-benzo-
quinone O-oxide could be confused with a semiquinone species
when spin-trapped. Alternatively, the semiquinone anion
detected in the EPR experiments could result from secondary
photochemistry of the 1,4-benzoquinone photoproduct.
yield of 0.31 and a maximum of 0.85). A similar calculation
9 Ϫ1
for 4-CP yields a range of rate constants of 1.0–2.5 × 10 s .
The lack of fluorescence lifetime data for the bromo and iodo
compounds precludes analogous calculations for them but we
note that for 4-FP and 4-CP the increase in the S → T ISC
1
1
rate constant is considerably smaller than that calculated for
1-fluoro- and 1-chloro-naphthalene (where an increase of a
factor of 100 in this ISC rate constant is found) which form
part of the 1-fluoronaphthalene to 1-iodonaphthalene series
The nature of the excited state which leads to carbene 4
which is known to have a triplet ground state ) is open to
19
(
speculation. It is possible to envisage two main routes follow-
ing excitation of the ground state halogenophenol. These
differ depending on whether loss of HX takes place from the
halogenophenol triplet state following intersystem crossing
21
analysed by Birks. In this series, the heavy atom effect is rather
greater on the S → T1 ISC process than on that for
1
T → S . If the same is true in the case of the 4-halogeno-
1
0
(
ISC) or whether it takes place from the initially excited halo-
phenols, then reaction from the excited triplet state is again
precluded on the basis of our experimental data.
We therefore conclude that the mechanism for the formation
of carbene 4 in the photolysis of the 4-halogenophenols in
neutral aqueous solution is that given in Scheme 2. We have
genophenol singlet state and the singlet carbene thus produced
then intersystem crosses to its triplet ground state.
The evidence available to help differentiate between these
two possibilities consists of the effects of changing the halogen
atom and the effect of oxygen on the yield of carbene. We have
previously found that the yield of 4 in the photolysis of aqueous
1
*
OH
OH
OH
+
O
4
-CP is approximately the same in aerated and degassed
solution and is perhaps slightly higher in oxygen-saturated
hν
+
X–
11
solution. Since oxygen is a well-known triplet quencher these
observations would tend to imply that the chlorophenol triplet
state is not involved in the photochemical pathway to 4 and we
have not observed any transient which we can assign to a triplet
state of any of the four halogenophenols. However, oxygen is
not as soluble in water as in other solvents so that this, together
with a short-lived triplet state, could account for the similar
carbene yields in aerated, oxygenated and degassed water. The
effect of oxygen is inconclusive with respect to determining the
early mechanistic details.
X
X
OH
OH
X
+
+
X–
Changing the halogen atom in the halogenophenol will have
several effects. Firstly, the C᎐X bond strength decreases from
Scheme 2 Mechanism for the formation of 4-oxocyclohexa-2,5-
dienylidene in the photoreaction of the four halogenophenols in
aerated aqueous solution at ambient temperature
20
fluorine through to iodine making bond cleavage easier
and potentially increasing the yield of 4. Secondly, the heavy
atom effect will operate and catalyse any ISC processes in the
halogenophenols. If loss of HX occurs from the excited singlet
state, ISC will compete with this and we would anticipate
decreases in the fluorescence quantum yield and lifetime
and the yield of 4 (and subsequent photoproducts) as the
lighter halogens are replaced by the heavier. The effect of
the halogen substituent on the fluorescence properties of both
the 4-halogenophenols and the 4-halogenoanisoles, where the
quantum yield and lifetime decrease considerably as the atomic
weight of the halogen increases (Table 1) is clearly in accord
with the photoreaction originating in the excited singlet state.
In addition we note that the photoreaction quantum yield and
yield of 4 both decrease as the halogen substituent in the
attempted to prove this mechanism by investigating the effect of
an external heavy atom on the fluorescence properties and
nanosecond flash photolysis of the 4-halogenophenols. We are
not able to use bromide or iodide ions to undertake such
experiments since these two ions would interfere in the reaction
mechanism (see below), so we therefore chose ethyl iodide and
xenon as possible external heavy atom quenchers. Unfortu-
nately, both of these materials are insufficiently soluble in water
to produce a significant quenching effect due to the relatively
short singlet lifetimes of all four halogenophenols. In addition,
the absorption spectrum of ethyl iodide overlaps that of the
phenols. We have therefore been unable to validate our
proposed mechanism by using such experiments.
4
-halogenophenol is changed from fluorine to iodine. This is
The reaction quantum yields reported earlier represent a
minimum yield for the formation of the carbene since
recombination of the initial hydroxyphenyl cation–halide ion
pair may occur before the cation loses a proton to form the
carbene and Grabner et al. report a quantum yield for
the formation of 4 from 4-CP of 0.75. They also found that the
decay rate of the carbene was enhanced in the presence of
added halide ions with iodide ions being the most effective and
concluded that the halide ion added directly to the carbene; this
route would also decrease the yield of photodegradation if the
halide ion adding to the carbene was the same as that in the
original halogenophenol.
entirely consistent with reaction from the halogenophenol
excited singlet state competing with ISC.
If loss of HX occurs from the halogenophenol triplet state
the heavy atom effect would be expected to operate to enhance
both S → T and T → S ISC. The first of these processes
would again decrease the fluorescence yield and lifetime but
would now enhance the yield of 4 whereas the second ISC pro-
cess would act to decrease it. We were unable to detect a trans-
ient in our flash photolysis studies which could be attributed to
a triplet state and must therefore conclude that it is formed in
low yields or is short-lived (lifetime <100 ns) or both. This
means that we are unable to quantify the effect of the halogens
10
1
1
1
0
The initial heterolysis of the aryl–halogen bond has parallels
in the photochemistry of fluoromethoxybenzenes in aqueous
on the T → S0 ISC. However, if the photoreaction does
1
22
originate from the excited triplet state, our failure to detect it
due to its short lifetime leads to the conclusion that the heavy
atom effect on its lifetime will be small. The observed variation
in the photoreaction yields is not consistent with this scenario.
In addition, the photoreaction quantum yield of 0.31 for
solution reported by Zhang and Wan. Here HF was generated
by substitution of the fluoride by water with initial loss of
fluoride ion and the generation of an aryl cation. However it is
interesting to note that a chloromethoxybenzene was felt to
exhibit behaviour best explained by initial homolysis of the
aryl–halogen bond rather than the heterolytic behaviour seen
here for 4-CP and the other halogenophenols. It would be inter-
esting to undertake faster flash experiments to try to observe
4
-FP indicates that if reaction occurs from the triplet state,
S → T ISC is already efficient and has a rate constant of ca.
1
1
1
8
Ϫ1
.5–4.0 × 10 s (based on a minimum intersystem crossing
J. Chem. Soc., Perkin Trans. 2, 1998
369

View Article Online
these earlier transients which are postulated for these various
reactions.
11 A.-P. Y. Durand, R. G. Brown, D. Worrall and F. Wilkinson,
J. Photochem. Photobiol. A, 1996, 96, 35.
2 E. Lipczynska-Kochany, Chemosphere, 1992, 24, 911.
3 E. Lipczynska-Kochany and J. Kochany, J. Photochem. Photobiol.
A, 1993, 73, 23.
4 E. Lipczynska-Kochany, J. Kochany and J. R. Bolton, J. Photochem.
Photobiol. A, 1992, 62, 229.
1
1
Acknowledgements
1
We thank the University of Central Lancashire for a Research
Studentship (A. P. D.) and for consumable support. We thank
the EPSRC for support and we also thank Dr Pierre Boule for
valuable comments on our work.
15 A.-P. Y. Durand, Ph.D. Thesis, University of Central Lancashire,
995.
1
1
1
1
6 A. M. Braun, M. T. Maurette and E. Oliveros, Photochemical
Technology, Wiley, Chichester, 1991.
7 D. V. O’Connor and D. Phillips, Time correlated single photon
counting, Academic Press, New York, 1984.
References
8 R. Sparrow, R. G. Brown, E. H. Evans and D. Shaw, J. Chem. Soc.
Faraday Trans. 2, 1986, 82, 2249.
1
2
3
A. Mills, R. H. Davies and D. Worsley, Chem. Soc., Rev., 1993, 417.
K. Rajeshwar, Chem. Ind., 1996, 454.
19 E. Wasserman and R. W. Murray, J. Am. Chem. Soc., 1964, 86, 4203;
W. Sander, W. Müller and R. Sustmann, Angew. Chem., Int. Ed.
Engl., 1988, 27, 572.
U. Stafford, K. A. Gray and P. V. Kamat, J. Phys. Chem., 1994, 98,
6
343.
4
5
6
A. Mills and R. Davies, J. Photochem. Photobiol. A, 1995, 85, 173.
P. Boule, C. Guyon and J. Lemaire, Chemosphere, 1982, 11, 1179.
E. Lipczynska-Kochany and J. R. Bolton, J. Chem. Soc., Chem.
Commun., 1990, 1596; J. Photochem. Photobiol. A, 1991, 58, 315.
K. Oudjehani and P. Boule, J. Photochem. Photobiol. A, 1992, 68,
20 Thermochemistry of Organic and Organometallic Compounds, ed.
J. D. Cox and G. Pilcher, Academic Press, London and New York,
1970.
21 J. B. Birks, Photophysics of Aromatic Molecules, Wiley-Interscience,
London, 1969.
7
8
9
3
63.
A.-P. Y. Durand, D. Brattan and R. G. Brown, Chemosphere, 1992,
5, 783.
A.-P. Y. Durand and R. G. Brown, Chemosphere, 1995, 31, 3595.
22 G. Zhang and P. Wan, J. Chem. Soc., Chem. Commun., 1994, 19.
2
Paper 7/05287A
Received 22nd July 1997
Accepted 24th October 1997
1
0 G. Grabner, C. Richard and G. Köhler, J. Am. Chem. Soc., 1994,
1
16, 11 470.
3
70
J. Chem. Soc., Perkin Trans. 2, 1998