Selectivity in the photo-Fenton and photocatalytic hydroxylation of biphenyl-4-carboxylic acid and derivatives (viz. 4-phenylsalicylic acid and 5-phenylsalicylic acid)

Article abstract of DOI:10.1002/poc.1839

The selectivity of hydroxylation of the distal rings of 4-phenylbenzoic acid, 4-phenylsalicylic acid, and 5-phenylsalicylic acid were determined using partial TiO₂-mediated photocatalytic degradation and photo-Fenton conditions. This separation of the binding site from the phenyl group being hydroxylated allows a less-biased evaluation. The hydroxylation regiochemistry behaves as qualitatively expected for an electrophilic reaction, given the assumption that 4-carboxyphenyl is a slightly electron-withdrawing substituent. Selectivity for hydroxylation of the distal phenyl in 4- and 5-phenylsalicylic acid is reversed, due to the reversal of the electronic demand, while adsorption to the TiO₂ surface is assumed to be analogous for the two structures. Copyright

Full text of DOI:10.1002/poc.1839

Research Article
Received: 10 November 2010,
Revised: 26 December 2010,
Accepted: 3 January 2011,
Published online in Wiley Online Library: 21 February 2011
(
wileyonlinelibrary.com) DOI 10.1002/poc.1839
Selectivity in the photo-Fenton and
photocatalytic hydroxylation of
biphenyl-4-carboxylic acid and derivatives
(
5
viz. 4-phenylsalicylic acid and
-phenylsalicylic acid)
a
a
a
Timothy Hathway , Deborah Lipman Chernyshov and William S. Jenks *
The selectivity of hydroxylation of the distal rings of 4-phenylbenzoic acid, 4-phenylsalicylic acid, and
2
5-phenylsalicylic acid were determined using partial TiO -mediated photocatalytic degradation and photo-Fenton
conditions. This separation of the binding site from the phenyl group being hydroxylated allows a less-biased
evaluation. The hydroxylation regiochemistry behaves as qualitatively expected for an electrophilic reaction, given
the assumption that 4-carboxyphenyl is a slightly electron-withdrawing substituent. Selectivity for hydroxylation of
the distal phenyl in 4- and 5-phenylsalicylic acid is reversed, due to the reversal of the electronic demand, while
adsorption to the TiO
Sons, Ltd.
2
surface is assumed to be analogous for the two structures. Copyright ß 2011 John Wiley &
Supporting information may be found in the online version of this paper.
Keywords: hydroxylation; photocatalysis; photo-Fenton; selectivity; titanium dioxide
INTRODUCTION
The study of the early oxidation steps of organic molecules
under conditions of TiO -mediated photocatalytic degradation
2
has been fruitful, at first to understand the chemistry of the
mineralization process, and more recently as a probe to explore
differences among catalysts. Broadly, hydroxyl-like chemistry
and chemistry driven by single electron transfer (SET) are
the two commonly observed modes of reactivity, particularly
[
1–13]
for aromatic organics.
Compounds with multiple reactivity
modes become useful interrogators of other catalysts; for
example we and others have used 4-methoxyresorcinol,
quinoline, and 1-anisylneopentanol (1-(4-methoxyphenyl)-2,2-
4
-Biphenylcarboxylic acid (1) was chosen as a starting point
[
14–18]
for a new probe molecule because it had both the desired
functionality and some symmetry that would limit the number
of products. We hypothesized that the geometry associated
with adsorption through the carboxylate and the differing
electron demands of the proximal (di-substituted) and distal
dimethylpropan-1-ol)) as probe molecules
several modified and doped TiO photocatalysts.
to evaluate
[
19–23]
2
However, the balance of reactivity is determined not only by
functionality, but also by the ability of the molecule to bind to the
[
10,24–32]
TiO
2
surface.
We wanted to add to our bank of test
molecules, not least because we thought it important to have
molecules with different binding modes. Thus the carboxylic acid
functionality, common to organic water pollutants and their
degradation intermediates was attractive. Here, we describe the
partial degradations of biphenylcarboxylic acid derivatives 1–3
*
Correspondence to: W. S. Jenks, Department of Chemistry, Iowa State Uni-
versity, Ames, IA 50011-3111, USA.
E-mail: wsjenks@iastate.edu
2
by TiO and under photo-Fenton conditions used as a control for
a
T. Hathway, D. L. Chernyshov, W. S. Jenks
the hydroxyl-like chemistry.
Department of Chemistry, Iowa State University, Ames, Iowa, USA
J. Phys. Org. Chem. 2011, 24 1151–1156
Copyright ß 2011 John Wiley & Sons, Ltd.

T. HATHWAY, D. L. CHERNYSHOV AND W. S. JENKS
(
mono-substituted) phenyl moieties might provide a basis for
regiochemical selectivity for the SET and hydroxyl-like pathways.
We posited that the comparison of 1 to 4-phenylsalicylic acid (2)
and 5-phenylsalicylic acid (3) might prove useful, in that the
additional electron richness of the proximal ring might further
enhance SET chemistry. Moreover, to the extent that the relative
arrangements of the substituents in 2 and 3 reverse – or at least
modify – their electronic influence on the distal phenyl, it was
possible that further differentiation might be observed.
A few publications address the regiochemistry of TiO
mediated and -related hydroxylations, and they broadly suggest
that regiochemistry of TiO -mediated hydroxylation follows the
qualitative rules familiar from standard electrophilic aromatic
2
-
2
[
33–39]
substitution.
However, the comparison of substituents
necessarily combines the effects of adsorption ability and the
electronic effect of the substituent. Compounds 2 and 3 are used
to address this question without that ambiguity.
RESULTS AND DISCUSSION
2
Partial degradations of 1–3 were carried out using P25 TiO at pH 8.5
and 12, and also using photo-Fenton conditions at pH 8.5. Solubility
limited the ability to run the reactions at low pH, and previous
experience had shown that 8.5 is a near-ideal pH for the observation
[28,29]
of intermediates in degradations closely related to these.
higher pH, the TiO reactions could be done, but the photo-Fenton
reactions could not be, due to precipitation of the iron. Three classes
of initial reactions were expected under TiO degradation conditions:
i) hydroxylation of the proximal ring, e.g., conversion of 1 to 2; (ii)
At
2
2
(
hydroxylation of the distal ring, e.g., conversion of 1 to 1c (Chart 1);
and (iii) reactions related to photo-Kolbe decarboxylation through
SET at the carboxylate functionality. Previous studies on the
degradation of benzoic acid all report formation of salicylic acid,
with some reporting observation of other hydroxylated benzoic
Chart 1. Starting biphenylcarboxylic acids and their hydroxylated
derivatives.
[39–44]
Reasonable zero-order rates were obtained with conversions
up to ꢀ25% and are shown in Table 1. (Degradations to higher
conversion began to resemble first order decays, as is common
for photocatalytic degradations.) All runs were at least duplicated,
and error limits in the table represent standard deviations from
the average. Control experiments showed that no significant
degradation occurred on the timescale of the experiments when
acids and small amounts of phenol.
2
the TiO was omitted or if samples were not exposed to light.
The absolute rate constants, obtained with initial concen-
trations of 0.25 mM, depend on several experimental parameters,
including lamp intensity, sample geometry, etc. However, all
physical parameters were held constant for the TiO reactions,
2
so the relative rates are meaningful. Among them, the pH 12
Table 1. Initial rates of degradation of 1–3 using titanium
dioxide at pH 12 and 8.5 and the photo-Fenton reaction at pH
8
.5
Rate of degradation (mM/min)
Compounds
TiO
2
pH 12
TiO
2
pH 8.5
Photo-Fenton
1
2
3
15.1 ꢁ 0.9
15.9 ꢁ 1.0
17.2 ꢁ 1.6
13.1 ꢁ 0.7
10.7 ꢁ 0.9
15.6 ꢁ 0.6
1.4 ꢁ 0.5
15.3 ꢁ 1.0
14.2 ꢁ 1.4
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 1151–1156

SELECTIVITY FOR HYDROXYLATION
reactions showed no rate variations that could be firmly put
outside experimental uncertainties, and the rates at pH 8.5 were
all within a factor of about 1.5.
The rates obtained for the photo-Fenton reaction, used as a
[
3,14,45,46]
model for the hydroxyl-type reactivity in TiO
were coincidentally similar to those obtained for TiO
2
chemistry,
-mediated
2
degradations. However, they importantly showed a different
internal rate profile, in that the rate obtained for degradation of 1
was an order of magnitude slower than that for the other two
compounds. The simplest explanation for this is that the rates for
the TiO -mediated reactions are limited by interactions with the
2
2
TiO surface, while the photo-Fenton reactions revealed rates that
reflected more about the inherent reactivity of the compounds.
An alternate and perhaps more probable explanation for the
slower removal of 1 under photo-Fenton conditions is based on
the hypothesis that the o-hydroxycarboxylate functionality tends
to chelate iron ions. Thus, compounds 2 and 3 are effectively
better ligands for the iron ions than is 1. As a result, 2 and 3 are
degraded faster than 1 because the hydroxyl radicals are formed
immediately in their vicinity. Whether truly free hydroxyl radicals
are formed at all in Fenton chemistry is a matter of some
Figure 1. Product distributions and mass balance of singly hydroxylated
products for degradation of 1–3 for photo-Fenton reactions and TiO
reactions at pH 12 and 8.5 at 10% conversion of starting material. The
initial concentration of 1, 2, or 3 was 250 mm. The photo-Fenton reactions
were carried out at pH 8.5.
2
[
47]
controversy (e.g. Ref.
); however, the lack of a free hydroxyl
radical would only reinforce this explanation. (The photochemical
step of the photo-Fenton reaction is not presumed to be
to observe if the degradation rates were correlated simply to the
existence of the hydroxyl group. Its degradation rate was 0.8 mM/
min, similar to that of 1. Based on overall electron demand alone, this
result is counterintuitive. It shows that the rate of degradation is
sensitive to the relative position of the substituents and is at least
consistent with the iron chelation hypothesis to explain the relative
rates of degradation of 1–3. Regardless, it became clear that the low
overall mass balance was not a hindrance in interpreting the relative
product distributions of the distal hydroxylations. Whatever other
reactions occur in parallel with the distal hydroxylations, they do not
interfere with the relative ratios of the a–c derivatives among
themselves. Analysis at low conversion also ensures that secondary
reactions have not grossly disturbed the initial product distributions.
The regioselectivity among the distal hydroxylation products for
treatment of 1–3 can be interpreted using rules of thumb from
electrophilic reactions of benzenes. Despite the near-neutral pH,
the ‘4-carboxyphenyl’ substituent clearly acts as a slightly
electron-withdrawing group, favoring the meta-hydroxylation
position in the reaction of 1. The meta:para selectivity is a little
<3:1 after accounting for the number of hydrogens (We presume
sterics account for the lack of observation of 1a, but acknowledge
that some ortho hydroxylation is observed for the other two
compounds.) This modest selectivity is consistent with that
observed for many weak electron-withdrawing groups for
electrophilic reactions of benzene, i.e., it is not the expectation
that meta functionalization should be the exclusive result.
Given the more rapid initial degradation of molecules 2 and 3
under Fenton conditions, we presume that the qualitative o–m–p
distributions are also meaningful. Addition of the 2-OH group to
make compound 2 should have the effect of making the phenyl
group slightly more electron donating than with the carboxyl
group alone, thus somewhat negating its effect as an electro-
n-withdrawing group. Indeed, the selectivity for formation of
the meta product 2b is quite low. Particularly pleasing from an
esthetic standpoint, however, is the contrast between com-
pounds 2b and c. By reversing the alignment of the OH and
carboxyl groups with compound 3, the HO group is placed in
conjugation to the distal phenyl, presumably greatly increasing
its effective electron donation ability to the distal ring. Indeed,
[
48]
generation of hydroxyl, so the point remains. ) Indeed, 1c, an
isomer of 2 and 3 that does not have the ortho relationship
between its hydroxy and carboxylate functionality, and it was
shown (vide infra) to degrade at a rate quite similar to 1, rather
than comparable to 2 and 3.
However, our primary interest was in the product distribution.
Three monohydroxylated derivatives of each sharting material
were prepared: the products of ortho, meta, and para
hydroxylation of the distal phenyl group. They are illustrated
in Chart 1 as the ‘a ’, ‘b ’, and ‘c’ derivatives of each compound,
respectively. Compound 2 may be viewed as a hydroxylated
derivative of 1, but no other compounds with hydroxylation of 1,
2, or 3 on the ring proximal to the carboxylic acid were prepared.
After degradation of 10–20%, samples were exhaustively silylated
and examined by GC. Products were identified by their unique
retention times (and by co-injection of the authentic samples) on
GC and by GC-MS verification of the mass. No singly hydroxylated
derivatives of 2 or 3 other than compounds 2a–c and 3a–c were
observed, i.e., no derivatives of 2 or 3 with hydroxyls on the
proximal ring could be detected. (As noted below, 2 was detected
as a trace product in degradation of 1.)
Figure 1 shows the yields of products obtained at 10%
conversion of 1–3 under various conditions. The data are
interpolated from multiple runs sampled at various low
percentage conversions to be standardized to the 10%
conversion. The relative errors of the yields are ꢂ10%.
An immediately striking feature of the data in Fig. 1 is that the
mass balance accounted for by singly hydroxylated compounds
for compound 1, while still below 50%, is significantly greater
than that for the other two, under all conditions. Conspicuously
missing from all degradations of 1 at low conversion was
compound 2. Also not observed (in any reactions) were biphenyl
or 4-phenylphenol.
In the photo-Fenton reactions, at conversions of at least 10%,
HPLC analysis of the reaction mixtures revealed many smaller
peaks of higher polarity (as judged by chromatography) than any
of the compounds in Chart 1. As mentioned previously, the rate of
degradation of 1c was determined under photo-Fenton conditions
J. Phys. Org. Chem. 2011, 24 1151–1156
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

T. HATHWAY, D. L. CHERNYSHOV AND W. S. JENKS
although 3b is still observed, it is clear that the proximal phenyl
acts as a mild o:p director for hydroxylation of the distal phenyl.
These latter compounds clearly represent multiple steps of
degradation, all of which occur on the proximal phenyl, or what
remains of it. Again, no 4-phenylphenol was observed in previous
studies of the degradation of benzoic acid and its hydroxylated
TiO -MEDIATED HYDROXYLATIONS
2
2
derivatives under TiO conditions, up to three hydroxylations of
In contrast to the results with the photo-Fenton chemistry, the
rates of degradation for compounds 1–3 were quite comparable
to one another, both at pH 8.5 and 12. The simplest explanation
for this is that the rate of degradation depends in very large part
on the adsorption of the compound to the catalyst, and that this
is dominated by the carboxylate functionality, common to all
three compounds. Again, a rate check of a representative primary
product was performed: 1c was degraded at 8.4 ꢁ 0.7 mM/min
at pH 12, slightly slower than the other compounds.
benzene were observed (and extensive ring-opened products
were not identified) and phenol was a very minor product at
[
41,52]
best.
Nonetheless, these compounds clearly indicate that
reactivity can and does occur at the proximal phenyl and that
once oxidation begins there, it continues more rapidly than
reactivity at the distal phenyl. This is consistent with previous
[
28,29]
findings showing that degradative electron transfer reac-
tions typically become more facile as the ring is oxygenated.
The hydroxylation product distributions and mass balances
obtained for the TiO -mediated degradations were qualitatively
CONCLUSIONS
2
similar to those obtained with the photo-Fenton reactions, but
differed quantitatively in the following ways:
Ultimately, it must be concluded that compounds 1–3 are not
especially good probes for distinguishing hydroxyl-like chemistry
(
1) For 1, a slightly lower mass balance was obtained, but it was
still greater than for the other two compounds. The selectivity
for hydroxylation favored the meta position still, but to a
slightly smaller degree.
from single-electron transfer chemistry on TiO
2
surfaces.
Although it seems clear that there is surface involvement in
their primary chemistry (given the nearly identical rates), there is
no particular evidence for specific SET-derived products not
available through hydroxylation pathways or vice versa. The
relative dearth of interpretable intermediates low mass balances
present another difficulty, in that there is chemistry to which our
methods are blind.
Nonetheless, the results reported here surely strengthen the
assertion that selectivity in radical-based aromatic hydroxylation
under both authentic hydroxyl radical chemistry (photo-Fenton)
(
(
2) For 2, the mass balance remained essentially unchanged, but
again, somewhat more ortho and para hydroxylation is
2
observed with TiO than with photo-Fenton.
3) For 3, no meta product is observed at all and para-
hydroxylation predominates to a greater extent over ortho.
In other words, the same pattern is observed with respect to
2
and TiO -mediated conditions can be thought of in terms of a
2
ortho/para versus meta direction using TiO and photo-Fenton,
nucleophilic benzene and electrophilic oxidizing agent. Although
the biphenyls surely exist with a substantial twist between their
p planes, some electronic communication between the rings is
expected, and the carboxylic acid acts as a modest electron-
withdrawing group. The dramatically different selectivity in
hydroxylation of 2 (low selectivity, includes meta) and 3 (where
the proximal phenyl clearly acts as an ortho/para director) surely
but the overall effect is slightly more biased toward ortho/para
functionalization.
These results are consistent with the notion that the
hydroxylation is also taking place with a hydroxyl-like species,
but that the adsorption of the carboxylate to the TiO makes it an
2
effectively less electron-withdrawing substituent. Electron trans-
fer mechanisms, i.e., attack of water on, for example, adsorbed
ꢃR
demonstrates
explanation.
a subtlety in selectivity that requires this
1
, cannot be ruled out, however as water attack on the radical
[
49–51]
cation results in the same product. (e.g. Refs.
contained within, which address the oxidation of phenol.)
and references
Degradations of 1 were carried out to higher conversion
approximately 50%) to try to identify some of the HPLC peaks
EXPERIMENTAL
(
that (as in the photo-Fenton chemistry) corresponded to
downstream oxidation products. Instead of HPLC analysis, the
reactions were lyophilized and exhaustively silylated and then
subjected to GC-MS analysis. In addition to 1b and c, a trace of 2
was observed, direct evidence of chemistry on the proximal
phenyl group. Furthermore, phenylmalonic acid (phenylpropa-
nedioic acid), benzoic acid, and phenol were also observed.
Materials
Compounds 1, 1c (Scheme 1), 4-iodobenzoic acid, 4- and
5-iodosalicylic acid, and the phenylboronic acids were used as
obtained from commercial sources. Compounds 2 and 3 were
prepared via Suzuki-Miyaura coupling using Larock’s method,
as illustrated in Scheme 1 for compound 3. The hydroxylated
[
53]
products shown in Chart 1 were prepared in two steps. First, the
Scheme 1. Suzuki-Miyaura coupling to prepare 3.
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 1151–1156

SELECTIVITY FOR HYDROXYLATION
aldehyde homolog was prepared in the same manner as 2 and 3;
it was then oxidized to the phenol and formic acid under
samples were directly injected into the HPLC for identification
and quantification.
[
54]
Baeyer-Villiger conditions with m-CPBA.
Synthetic details are
given in the Supporting Information. TiO2 (Degussa P25) was
used as received.
Acknowledgements
Photolyses
Support of this work by the National Science Foundation
(
CHE0518586) is gratefully acknowledged.
The standard suspensions for photocatalytic reactions were
2
prepared to result in 100 mg of TiO per 100 ml of deionized
water. To 75 ml of water was added 100 mg of P25 TiO2.
Sonication for 5 min was used to break up larger aggregates of
REFERENCES
TiO
2
. As noted, the pH was adjusted to 12.0 ꢁ 0.5 or 8.5 ꢁ 0.5 by
adding 0.1 M NaOH at the start of the reaction and also
throughout the reaction as needed to maintain that value. The
substrate was added as a 25 ml aliquot from a 1 mM solution of
the carboxylate in water to give a final concentration of 250 mm
[
[
1] I. Izumi, F. F. Fan, A. J. Bard, J. Phys. Chem. 1981, 85, 218.
2] K. Okamoto, Y. Yamamoto, H. Tanaka, M. Tanaka, A. Itaya, Bull. Chem.
Soc. Jpn. 1985, 58, 2015.
[3] M. A. Fox, M. T. Dulay, Chem. Rev. 1993, 93, 341.
[
4] M. A. Fox, in Photocatalytic Purification and Treatment of Water and Air,
Vol. 3 (Eds. D. F. Ollis, H. Al-Ekabi), Elsevier, Amsterdam, 1993, p. 163.
5] O. Legrini, E. Oliveros, A. M. Braun, Chem. Rev. 1993, 93, 671.
starting material. The mixture was then purged with O
stirred for 30 min in the dark before the irradiation was started.
Both stirring and O purging were continued throughout the
2
and
[
2
[6] C. Mallard-Dupuy, C. Guillard, H. Courbon, P. Pichat, Environ. Sci.
Technol. 1994, 28, 2176.
[
55,56]
reaction. Ferrioxalate actinometry
was used to ensure
photon flux remained constant and, in a few instances, to
normalize rates to the rest of the data.
[7] M. R. Hoffmann, S. T. Martin, W. Choi, D. W. Bahnemann, Chem. Rev.
1995, 95, 69.
[
[
8] M. Qamar, M. Muneer, D. Bahnemann, Res. Chem. Int. 2005, 31, 807.
9] W. S. Jenks, in Environmental Catalysis (Ed.: V. H. Grassian), CRC Press,
Boca Raton, 2005, 307.
Photolyses were carried out with stirring at ambient
temperature using a modified Rayonet mini-reactor equipped
with a fan and 2 ꢄ 4-Watt broadly-emitting 350 nm ‘black light’
fluorescent tubes unless otherwise noted. Reaction times were
dependent on the degree of degradation required, although
[10] X. Li, J. W. Cubbage, W. S. Jenks, J. Photochem. Photobiol. A 2001, 143,
9.
6
[
[
[
11] A. L. Linsebigler, G. Lu, J. T. Yates, Jr., Chem. Rev. 1995, 95, 735.
12] Y. Sun, J. J. Pignatello, Environ. Sci. Technol. 1995, 29, 2065.
13] X. Li, W. S. Jenks, J. Am. Chem. Soc. 2000, 122, 11864.
25 min was used for kinetic runs.
After appropriate irradiation times, samples were removed and
[14] L. Cermenati, P. Pichat, C. Guillard, A. Albini, J. Phys. Chem. B 1997, 101,
acidified by addition of Amberlite IR-120 ion exchange resin. The
TiO and resin were separated by centrifugation, followed by
2650.
[
15] L. Cermenati, A. Albini, P. Pichat, C. Guillard, Res. Chem. Intermed.
2
2000, 26, 221.
filtration through a syringe-mounted 0.2 mm PES filter. Sample
sizes were 1 ml for kinetics or 50 ml for product studies. The latter,
larger samples were concentrated by rotary evaporation to
approximately 2 ml and the residual water was removed by
lyophilization.
[
[
16] R. Enriquez, P. Pichat, Langmuir 2001, 17, 6132.
17] M. Ranchella, C. Rol, G. V. Sebastiani, J. Chem. Soc. Perkin Trans. 2,
2000, 311.
[
18] L. Cermenati, A. Albini, P. Pichat, C. Guillard, Res. Chem. Int. 2000, 26,
221.
[
19] E. M. Rockafellow, X. Fang, B. G. Trewyn, K. Schmidt-Rohr, W. S. Jenks,
For GC-MS product studies, lyophilized 50 ml samples were
exhaustively silylated by treatment with 1 ml of anhydrous
pyridine, 0.2 ml of 1,1,1,3,3,3-hexamethyldisilazane and 0.1 ml of
Chem. Mater. 2009, 21, 1187.
[20] E. M. Rockafellow, J. M. Haywood, T. Witte, R. S. Houk, W. S. Jenks,
Langmuir 2010, 26, 19052.
[
57]
[21] E. M. Rockafellow, L. K. Stewart, W. S. Jenks, Appl. Catal. B 2009, 91,
chlorotrimethylsilane.
Samples were vigorously shaken for
5
54.
22] T. Hathway, W. S. Jenks, J. Photochem. Photobiol. A Chem. 2008, 200,
16.
1
min, and allowed to stand 5 min at ambient temperature. The
[
resulting pyridinium chloride precipitate was separated by
centrifugation prior to chromatographic analysis.
2
[23] T. Hathway, E. M. Rockafellow, Y.-C. Oh, W. S. Jenks, J. Photochem.
Photobiol. A 2009, 207, 197.
GC-MS work was done with a standard 25 m DB-5 (5% phenyl)
column for chromatography, coupled to a time-of-flight mass
spectral detector. The temperature program was 130 8C for 2 min,
followed by a ramp to 280 8C at 20 8C/min. Routine work was
done on another instrument with an FID detector.
Kinetic data were obtained using HPLC (diode array UV/Vis
detection) analysis of 1 ml aliquots that were acidified with
Amberlite and centrifuged before injection. A standard C18
reverse-phase column was used. The eluent was a 50:50 mixture
of water and acetonitrile that contained 0.1% acetic acid. The flow
was 1.0 ml/min.
[
24] Y.-C. Oh, X. Li, J. W. Cubbage, S. Jenks William, Appl. Catal. B Environ
004, 54, 105.
2
[
25] J. Cunningham, G. Al-Sayyed, S. Srijaranal, in Aquatic and Surface
Photochemistry (Ed.: G. R. Helz), Lewis, Boca Raton, FL, 1994, p. 317.
[26] J. Cunningham, P. Sedlak, J. Photochem. Photobiol. A 1994, 77, 255.
27] S. Tunesi, M. Anderson, J. Phys. Chem. 1991, 95, 3399.
28] X. Li, J. W. Cubbage, W. S. Jenks, J. Org. Chem. 1999, 64, 8525.
29] X. Li, J. W. Cubbage, T. A. Tetzlaff, W. S. Jenks, J. Org. Chem. 1999, 64,
[
[
[
8509.
[30] M. A. Fox, H. Ogawa, P. Pichat, J. Org. Chem. 1989, 54, 3847.
[
31] P. Pichat, J.-C. D’Oliveira, J.-F. Maffre, D. Mas, in Photocatalytic Puri-
fication and Treatment of Water and Air, Vol. 3 (Eds. D. F. Ollis, H.
Al-Ekabi), Elsevier, Amsterdam, 1993, 683.
Photo-Fenton reactions were set up as 100 ml solutions using
[
32] R. Enriquez, A. G. Agrios, P. Pichat, Catal. Today 2007, 120, 196.
2
5 ml of the organic probe from a 1 mM stock solution, 10 ml of a
[33] G. Palmisano, M. Addamo, V. Augugliaro, T. Caronna, P. A. Di, E. G.
3þ
solution containing 0.05 mM Fe (as 0.0277 g Fe
SO ), and 1 ml (92 mM) of 30% H . The pH was adjusted to
.5. A rayonet lamp using 2 ꢄ 350 nm lamps was employed for
photolysis. Aliquots were taken at regular time points with a
crystal of sodium thiosulfate added to quench excess H . These
2
(SO
4
)
3
in dilute
Lopez, V. Loddo, G. Marci, L. Palmisano, M. Schiavello, Catal. Today
2007, 122, 118.
H
8
2
4
2 2
O
[34] J. Poerschmann, U. Trommler, J. Chromatogr. A 2009, 1216, 5570.
[35] S. Parra, J. Olivero, L. Pacheco, C. Pulgarin, Appl. Catal. B 2003, 43, 293.
[36] H. Yoshida, H. Yuzawa, M. Aoki, K. Otake, H. Itoh, T. Hattori, Chem.
Commun. 2008, 4634.
2 2
O
J. Phys. Org. Chem. 2011, 24 1151–1156
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

T. HATHWAY, D. L. CHERNYSHOV AND W. S. JENKS
[
[
37] G. Palmisano, M. Addamo, V. Augugliaro, T. Caronna, A. Di Paola, E. G.
[47] B. Ensing, F. Buda, P. Bl o¨ chl, E. J. Baerends, Angew. Chem. Int. Ed. 2001,
40, 2893.
[48] R. G. Zepp, B. C. Faust, J. Hoign e´ , Environ. Sci. Technol. 1992, 26, 313.
[49] I. Ilisz, A. Dombi, Appl. Catal. A 1999, 180, 35.
[50] C. Minero, G. Mariella, V. Maurino, E. Pelizzetti, Langmuir 2000, 16,
2632.
ˆ
¨
LUpez, V. Loddo, G. Marcl, L. Palmisano, M. Schiavello, Catal. Today
007, 122, 118.
38] C. R. E. Jefcoate, J. R. L. Smith, R. O. C. Norman, J. Chem. Soc. B 1969,
013.
2
1
[
[
39] I. Izumi, F.-R. F. Fan, A. J. Bard, J. Phys. Chem. 1981, 85, 218.
40] A. Mills, C. E. Holland, R. H. Davies, D. Worsley, J. Photochem. Photobiol.
A 1994, 83, 257.
[51] C. Minero, G. Mariella, V. Maurino, D. Vione, E. Pelizzetti, Langmuir
2000, 16, 8964.
[
[
41] T. Velegraki, D. Mantzavinos, Chem. Eng. J. 2008, 140, 15.
42] A. H. C. Chan, C. K. Chan, J. P. Barford, J. F. Porter, Water Res. 2003, 37,
[52] F. Benoit-Marqui e´ , E. Puech-Costes, A. M. Braun, E. Oliveros, M.-T.
Maurette, J. Photochem. Photobiol. A 1997, 108, 65.
[53] J. P. Waldo, R. C. Larock, Org. Lett. 2005, 7, 5203.
[54] K. Fujishiro, S. Mitamura, Bull. Chem. Soc. Jpn. 1988, 61, 4464.
[55] C. G. Hatchard, C. A. Parker, Proc. Royal. Soc. A 1956, 235, 518.
[56] W. D. Bowman, J. N. Demas, J. Phys. Chem. 1976, 80, 2434.
[57] C. C. Sweeley, R. Bentley, M. Makita, W. W. Wells, J. Am. Chem. Soc.
1963, 85, 2497.
1
125.
[
43] Y. Deng, K. Zhang, H. Chen, T. Wu, M. Krzyaniak, A. Wellons, D. Bolla, K.
Douglas, Y. Zuo, Atmos. Environ. 2006, 40, 3665.
44] R. W. Matthews, Water Res. 1986, 20, 569.
45] J. De Laat, H. Gallard, Environ. Sci. Technol. 1999, 33, 2726.
46] J. Pignatello, Environ. Sci. Technol. 1992, 26, 944.
[
[
[
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 1151–1156