Reactions of phenol-OH-adduct radicals. Phenoxyl radical formation by water elimination vs. oxidation by dioxygen

Article abstract of DOI:10.1039/b008434o

The reactions of OH radicals generated radiolytically in N2O-saturated aqueous solutions with phenol have been examined, focusing special attention on the addition of dioxygen and the competing dehydration reactions of the OH-adduct radicals. Using the pulse radiolysis technique, the overall rate constant of dioxygen addition of the dihydroxycyclohexadienyl radicals was determined to be k = 1.2 * 10⁹ dm³ mol^-1 s^-1. This dioxygen addition reaction was found to be practically irreversible, in contrast to other hydroxycyclohexadienyl radicals. The so-formed dihydroxycyclohexadienylperoxyl radicals subsequently eliminate HO2^./O2^.-. By using tetranitromethane (TNM) as a probe for O2^.- (formation of the nitroform anion), the overall rate constant of HO2^.-elimination (mainly of the para- and ortho-isomers, formation of hydroquinone and catechol, respectively) was determined to be 1.3 * 10⁵ s^-1. The rate constant of the dehydration of the p-OH-adduct in neutral to acidic solution was determined by monitoring the absorbance build-up of the phenoxyl radical to be k₇ = 1.8 * 10³ + 1.7 * 10⁹ * [H⁺] s^-1, and that of the o-OH-adduct to be k₆ = 1.1 * 10⁸ * [H⁺] s^-1 (the uncatalysed reaction is too slow to be measurable by this technique). The HPO4^2--catalysed dehydration rate constant of the p-OH-adduct was similarly determined to be 5.8 * 10⁷ dm³ mol^-1 s^-1. Based on the above rate constant of dioxygen addition, the rate constant of the proton-catalysed dehydration of the p-OH-adduct was determined by the competition of these two reactions on the yield of hydroquinone to be 1.0 * 10⁹ dm³ mol^-1 s^-1, and similarly that of the ortho-OH-adduct (on the yield of catechol) to be 2.1 * 10⁸ dm³ mol^-1 s^-1.

Full text of DOI:10.1039/b008434o

View Article Online / Journal Homepage / Table of Contents for this issue
Reactions of phenol-OH-adduct radicals. Phenoxyl radical
formation by water elimination vs. oxidation by dioxygen
a,b
a
a
Eino Mvula, Man Nien Schuchmann and Clemens von Sonntag*
2
a
Max-Planck-Institut für Strahlenchemie, Stiftstrasse 34-36, P.O. Box 101365, D-45413
Mülheim an der Ruhr, Germany. Tel: ϩϩ 49-208-306-3529, Fax: ϩϩ49-208-306-3951.
E-mail: vonsonntag@mpi-muelheim.mpg.de
b
Department of Chemistry, University of Namibia, Windhoek, Namibia
Received (in Cambridge, UK) 19th October 2000, Accepted 2nd January 2001
First published as an Advance Article on the web 5th February 2001
The reactions of OH radicals generated radiolytically in N O-saturated aqueous solutions with phenol have been
examined, focusing special attention on the addition of dioxygen and the competing dehydration reactions of the
OH-adduct radicals. Using the pulse radiolysis technique, the overall rate constant of dioxygen addition to the
2
9
3
Ϫ1 Ϫ1
dihydroxycyclohexadienyl radicals was determined to be k = 1.2 × 10 dm mol s . This dioxygen addition reaction
was found to be practically irreversible, in contrast to other hydroxycyclohexadienyl radicals. The so-formed dihydroxy-
ؒ
Ϫ
cyclohexadienylperoxyl radicals subsequently eliminate HO /O . By using tetranitromethane (TNM) as a probe
2
2
᝽
Ϫ
ؒ
for O2 (formation of the nitroform anion), the overall rate constant of HO -elimination (mainly of the para- and
ortho-isomers, formation of hydroquinone and catechol, respectively) was determined to be 1.3 × 10 s . The rate
constant of the dehydration of the p-OH-adduct in neutral to acidic solution was determined by monitoring the
absorbance build-up of the phenoxyl radical to be k = 1.8 × 10 ϩ 1.7 × 10 × [H ] s , and that of the o-OH-adduct
to be k = 1.1 × 10 × [H ] s (the uncatalysed reaction is too slow to be measurable by this technique). The HPO
᝽
2
5
Ϫ1
3
9
ϩ
Ϫ1
7
8
ϩ
Ϫ1
2Ϫ
-
6
4
7
3
Ϫ1 Ϫ1
catalysed dehydration rate constant of the p-OH-adduct was similarly determined to be 5.8 × 10 dm mol s .
Based on the above rate constant of dioxygen addition, the rate constant of the proton-catalysed dehydration of the
p-OH-adduct was determined by the competition of these two reactions on the yield of hydroquinone to be 1.0 ×
9
3
Ϫ1 Ϫ1
8
3
Ϫ1 Ϫ1
1
0 dm mol s , and similarly that of the ortho-OH-adduct (on the yield of catechol) to be 2.1 × 10 dm mol s .
Although the formation of phenoxyl radicals is thermo-
1
dynamically favoured, the reaction of OH radicals with phenol
mainly leads to the formation of the four isomeric dihydroxy-
cyclohexadienyl radicals (OH-adducts). Being an electrophilic
2
radical, the OH radical preferentially adds to the ortho-
3
[
reaction (1), 48%] and para-positions [reaction (2), 36%].
Addition to the meta-position [reaction (3)] amounts to only
3
8
%. Addition to the ipso-position [reaction (4)] is expected to
be quite negligible, and the resulting ipso-OH-adduct possibly
dehydrates promptly to give the phenoxyl radical [reaction (9)].
The fraction of ipso-OH-adduct is therefore not distinguishable
ؒ
from direct H-abstraction by OH [reaction (5)]. Based on the
material balance of OH radicals, their total yield [reactions (4)
3
and (5)] is attributed to the remaining 8% of OH radicals.
ϩ
The OH-adduct-radicals undergo spontaneous and H -
Ϫ
and OH -catalysed elimination of water to yield the more
4,5
resonance-stabilised phenoxyl radical [reactions (6)–(9)]. The
rate of this dehydration reaction differs among the different
isomeric OH-adduct radicals. By monitoring the kinetics of
absorbance build-up of the phenoxyl radical at 400 nm, or
the decay of the dihydroxycyclohexadienyl radicals at 330 nm
ϩ
under various pH conditions, it has been found that the H -
catalysed dehydration reaction consists of two parallel pro-
cesses, a fast and a slow component with approximately equal
4
contributions. More recently, based on the pH dependence of
the yields of hydroquinone and catechol in the presence of
an oxidant, it has been possible to assign the faster process
to the water elimination reaction of the p-OH-adduct radical
3
and the slower process to that of the o-OH-adduct radical.
waste-water is reprocessed to drinking water. Phenol is very
3 3 Ϫ1
Ozonation of drinking- and waste-water is gaining in
importance, and in arid countries this technique is now also
used in the so-called reclamation process, where municipal
reactive towards ozone [k(phenol ϩ O ) = 1.3 × 10 dm mol
3
Ϫ1 9 3 Ϫ1 Ϫ1 6,7
s ; k(phenolate ϩ O ) = 1.4 × 10 dm mol s ] and thus
3
–19
8
is readily degraded.
We have recently ascertained that the
2
64
J. Chem. Soc., Perkin Trans. 2, 2001, 264–268
This journal is © The Royal Society of Chemistry 2001
DOI: 10.1039/b008434o

View Article Online
major products, catechol and hydroquinone, have, to a large
ؒ
20
extent, OH as precursor. There are also non-radical processes
1
21
that lead to singlet dioxygen [O ( ∆ )] formation. Free radicals
2
g
that are formed during ozonation may be oxidised by dioxygen
ozonated solutions are typically rich in dioxygen, because
(
ozone is normally generated from dioxygen; for a review on
peroxyl radicals in aqueous solution see ref. 22) or by ozone
(
cf. ref. 23).
For a mechanistic description of the rather complex phenol/
O /O system, one requires reliable data on the kinetics of the
3
2
reaction of dioxygen with the phenol–OH-adduct radicals and
the competing water-elimination processes. In this context it
is of interest to learn whether dioxygen addition to these
dihydroxycyclohexadienyl radicals is reversible, as has been
24–26
found for other hydroxycyclohexadienyl radicals.
The data
to be reported in the present paper have been obtained with
the help of radiation techniques, notably pulse radiolysis.
Ϫ3
Fig. 1 Pulse radiolysis of a N O-saturated solution of phenol (10
Experimental
2
3
mol dm ) at pH 6.8. Transient absorption spectra observed 2 µs (᭿),
30 µs (᭝) and 200 µs (᭹) after the pulse (~4 Gy per pulse). Inset A:
kobs of absorbance build-up at 400 nm as function of HPO4 concen-
tration. Inset B: kobs as function of proton concentration.
Aqueous solutions of phenol (Merck, purified by zone-melting,
2
Ϫ
Ϫ3
Ϫ3
concentration typically 1 × 10 mol dm ) were prepared
in Milli-Q-filtered (Millipore) water and saturated with N O
2
or N O–O2 (4:1 v/v). γ-Radiolyses were carried out in a
2
60
Ϫ1
Monitored 200 µs after the pulse, the absorption at 400 nm
increases at the expense of that at 330 nm. The rate of absorp-
tion build-up at 400 nm, as well as the decay at 330 nm, are
dose-rate dependent for doses between 1.5 and 10 Gy per pulse,
since the bimolecular decay of the radicals is also reflected
in the kinetics. By extrapolating the reciprocal of the first half-
lives of absorption decay at 330 nm to zero dose, the rate con-3
panoramic Co-source at a dose rate of 0.1 Gy s to total
doses of up to 270 Gy (i.e. the conversion of the substrate
remained below 20%). Samples of the irradiated solution were
analysed by HPLC on a 25 cm Nucleosil 5-C18 column with a
5
cm precolumn. The eluent was an aqueous solution of 25%
3 Ϫ1
methanol flowing at 1 cm min . The retention times were:
hydroquinone, 5 min, catechol, 7.8 min, and phenol, 14.5 min.
Pulse radiolysis was carried out with a 2.8 MeV van-de-
Graaff accelerator delivering electron pulses of 0.4 µs duration.
Intermediates were monitored by optical detection. The pulse
stant of phenoxyl radical formation at neutral pH, k = 1.8 × 10
Ϫ1
s
was obtained. A similar value was obtained for the build-up
3 Ϫ1
at 400 nm. A higher value (~5 × 10 s ) has been reported for
27
the uncatalysed dehydration of the dihydroxycyclohexadienyl
radiolysis set-up has been described previously. For dosimetry,
N O-saturated 10 mol dm thiocyanate solution was used
4
Ϫ2
Ϫ3
radicals (a somewhat higher value would be obtained if the
2
Ϫ
Ϫ4
2
experimental data were not corrected for the bimolecular decay
of radicals by extrapolation to zero dose). As will be shown
below, the dehydration of the p-OH-adduct in acidic solution is
faster than that of the o-OH-adduct. The above rate constant
can thus be attributed to the uncatalysed dehydration of the
p-OH-adduct. The corresponding value for the o-OH-
adduct would be still lower, and therefore not measurable
under pulse radiolysis conditions. By monitoring the absorp-
tion decay at 370 nm at varying dose per pulse, the rate constant
of bimolecular decay of the OH-adduct radicals was found
for optical detection, taking G × ε[(SCN) ] = 4.8 × 10
m
2
᝽
Ϫ1
28
J
at 480 nm. (Although more recently a 10% higher value
29
for G × ε has been redetermined, for consistency with our
previous measurements the lower value was used.)
Results and discussion
The radical-generating system
Hydroxyl radicals are generated in the radiolysis of water
8
3
Ϫ1 Ϫ1
[
reaction (10)]. The radiation-chemical yields (G values) of the
to be 2k = 2.0 × 10 dm mol s . For a compilation of rate
constants see Table 1.
ionizing
At pH < 5, the formation of the phenoxyl radical was
observed to consist of two parallel processes, a fast build-up
followed by a slower build-up, with approximately equal yields.
The rates of these two processes increase with increasing proton
concentrations. From the linear plot of the observed first-order
rate constant of the first absorption build-up at 400 nm versus
proton concentration at pH 3.4 to 4.8 (inset B in Fig. 1), the rate
Ϫ
ؒ
ؒ
ϩ
Ϫ
H O
eaq , OH, H , H , OH , H O , H (10)
2 2 2
2
radiation
ؒ
Ϫ
Ϫ7
Ϫ1
primary radicals are G( OH) ≈ G(e ) = 2.9 × 10 mol J ,
aq
ؒ
Ϫ7
Ϫ1
Ϫ7
Ϫ1
G(H ) = 0.6 × 10 mol J , and G(H O ) ≈ 0.7 × 10 mol J .
2
2
ؒ
N O is used to convert the solvated electron into OH [reaction
11)]. The rate constant of the reaction of hydroxyl radicals
2
(
ϩ
constant of the fast H -catalysed phenoxyl radical formation
ϩ
9
3
Ϫ1
Ϫ1
Ϫ
ؒ
Ϫ
was obtained, k(H ) = 1.7 × 10 dm mol s . Similarly,
fast
ϩ 8 3 Ϫ1 Ϫ1
eaq ϩ N O ϩ H O → OH ϩ OH ϩ N
(11)
2
2
2
k(H ) = 1.1 × 10 dm mol
s
was obtained from the
slow
second and slower absorption build-up (data not shown), these
with phenol (pK = 10) [reactions (1)–(5)] has been reported to
be ~10 dm mol s .
a
3,4
10
3
Ϫ1 Ϫ1 30
values are in good agreement with the reported values.
Ϫ
The OH -catalysed dehydration reaction of the OH-adduct
radicals of phenol has been reported to be faster than the
Conversion of the dihydroxycyclohexadienyl radicals into
phenoxyl radicals
ϩ
5
Ϫ1
H -catalysed reaction (k > 1 × 10 s at pH 8.6 and 1.05 ×
1
obs
7
Ϫ1
4,5
0 s at pH ≈ 11). Efficient catalysis of water elimination by
4
The transient spectra observed in the pulse radiolysis of a
phosphate buffer has also been observed. By monitoring the
Ϫ
N O-saturated solution of phenol at pH 6.8 (Fig. 1) are in good
absorption build-up at 400 nm with increasing H PO4 concen-
tration at pH 5.8 to 4.8, we found no catalytic effect of H PO .
2
2
4
Ϫ
agreement with the reported observations. Immediately after
2
4
a 0.4 µs pulse the only prominent absorption comes from the
The catalytic effect of phosphate buffer is due solely to the
2Ϫ
phenol OH-adduct radicals at 330 nm, and very little absorp-
presence of HPO4 . The value for the rate constant of the
31
2Ϫ
tion of the phenoxyl radical at 400 nm is observed.
HPO4 -catalysed dehydration reaction has been determined
J. Chem. Soc., Perkin Trans. 2, 2001, 264–268
265

View Article Online
Table 1 Compilation of rate constants determined in the present study
Reaction
Rate constant
8
3
Ϫ1 Ϫ1
2
OH-Adduct radicals→products
2.0 × 10 dm mol
s
(2k)
3
Ϫ1
p-OH-Adduct radicals→phenoxyl ϩ H O
1.8 × 10 s
2
2
Ϫ
7
3
Ϫ1 Ϫ1
p-OH-Adduct radicals→phenoxyl ϩ H O (HPO -catal.)
5.8 × 10 dm mol
s
2
4
ϩ
9
3
Ϫ1 Ϫ1 a
p-OH-Adduct radicals→phenoxyl ϩ H O (H -catal.)
1.7 × 10 dm mol
s
2
9
3
Ϫ1 Ϫ1 b
1
.0 × 10 dm mol
s
ϩ
8
3
Ϫ1 Ϫ1 a
o-OH-Adduct radicals→phenoxyl ϩ H O (H -catal.)
1.1 × 10 dm mol
s
2
8
3
Ϫ1 Ϫ1 b
2
.1 × 10 dm mol
s
9
3
Ϫ1 Ϫ1
OH-Adduct radicals ϩ O →peroxyl radicals
Peroxyl radicals→HO2 ϩ products
1.2 × 10 dm mol
s
2
ؒ
5
Ϫ1
1.3 × 10 s
a
b
Determined by pulse radiolysis. Determined by competition kinetics.
in a weakly basic solution of phenol (pH 8.1–8.7) with varying
2
Ϫ
Ϫ
HPO4 concentration [the contribution from OH has been
Ϫ
5
10
3
Ϫ1
Ϫ1
corrected taking k(OH ) = 1 × 10 dm mol s ] to be
k(HPO4 ) = 5.8 × 10 dm mol
2Ϫ
7
3
Ϫ1 Ϫ1
s
(inset A in Fig. 1).
Reaction of the dihydroxycyclohexadienyl radicals with dioxygen
In the presence of dioxygen, the C-centered dihydroxycyclo-
hexadienyl radicals will rapidly add dioxygen to yield the corre-
sponding peroxyl radicals [e.g., reactions (12) and (13) for the
p-OH-radical adduct]. The phenoxyl radical is not known to
react with dioxygen (cf. ref. 32).
Fig. 2 Dioxygen concentration dependence of the observed first-
order rate constant of absorption decay at 330 nm in the pulse
radiolysis of a N O–O (4:1)-saturated solution of phenol at pH 6.8
2
2
(
~5 Gy per pulse).
ؒ
ؒ
H ϩ O → HO
(16)
(17)
(18)
2
2
ؒ
ϩ
Ϫ
HO2
H ϩ O2
᝽
In the pulse radiolysis of a N O–O (4:1)-saturated solution
2
2
Ϫ
Ϫ
of phenol at pH 6.8, the initial absorption of the OH-adduct
radicals at 330 nm decays rapidly, as these radicals are being
converted to the corresponding peroxyl radicals with weaker
absorbance at this wavelength. Fig. 2 shows the observed first-
order rate constant of absorbance decay at varying dioxygen
concentrations.
O
2
᝽ ϩ C(NO
)
2
4
→ C(NO
)
2
3
ϩ NO ϩ O
2 2
Ϫ
The rate constant of the second step of NF formation is
5
Ϫ1
1.3 × 10 s . The yield of this absorption build-up corresponds
Ϫ
Ϫ
᝽
Ϫ7
Ϫ1
ؒ
to G(NF ) = G(O ) = 4.6 × 10 mol J , or 80% of OH. The
2
Ϫ
yield of O2 corresponds approximately to the sum of the
o- and p-OH-adduct radicals. Since reaction (18) is not the rate-
᝽
From the slope the rate constant of dioxygen addition to the
9
3
Ϫ1
9
3
Ϫ1 Ϫ1 33
dihydroxycyclohexadienyl radicals of k = 1.2 × 10 dm mol
determining step (k18 = 1.9 × 10 dm mol s ), the rate con-
5 Ϫ1
Ϫ1
4
Ϫ1
s
is obtained. There is a small intercept (3.8 × 10 s ), which
stant determined above (1.3 × 10 s ) is thus the overall rate of
3
Ϫ1
ؒ
includes the contribution (1.8 × 10 s ) from the dehydration
HO -elimination of the peroxyl radicals of these two isomeric
2
reactions (see above) and the contribution from the bimolecular
dihydroxycyclohexadienyl radicals [e.g., reactions (14), (15) of
the p-OH-adduct-radical].
3
Ϫ1
decay (~10 s at 6 Gy per pulse), as well as that from the
ؒ
subsequent rapid HO -elimination of the dihydroxycyclohexa-
2
dienylperoxyl radicals [cf., reactions (14) or (15), see below].
These leave little room for the reverse dioxygen addition
reactions [cf. reactions (Ϫ12) or (Ϫ13)]. This is in contrast to
the other hydroxycyclohexadienylperoxyl radicals, where the
reversibility of the dioxygen addition reaction is much more
Product studies
ؒ
HO -elimination reactions of the peroxyl radicals of the o- and
2
p-OH-adducts yield catechol and hydroquinone, respectively
[
cf. reactions (14)–(16) for the formation of hydroquinone].
24–26
In the steady state γ-radiolysis of a N O–O (4:1)-saturated
pronounced.
2
2
solution of phenol at pH 6.8, we found G(catechol) =
Ϫ7
Ϫ1
Ϫ7
2
.3 × 10
mol J
and G(hydroquinone) = 1.8 × 10
mol
ؒ
HO -elimination reaction of the the dihydroxycyclohexadienyl-
2
Ϫ1
J . In acidic solution, the water elimination reactions (6)
peroxyl radicals
ϩ
and (7) become faster with increasing H concentration and
In the pulse radiolysis of a N O–O (4:1)-saturated solution of
compete with dioxygen addition. Based on this competition
between the proton-catalysed water elimination and dioxygen
addition, one can write eqn. (19) for the formation of
hydroquinone, which can be rearranged to give eqn. (20). An
analogous equation may be written for the formation of
catechol.
2
2
Ϫ4
phenol containing tetranitromethane (TNM, 3 × 10 mol
Ϫ3
Ϫ
dm ), the absorption of the nitroform anion (NF ) at 350 nm
increases rapidly following an initial jump, which is due to the
ؒ
Ϫ
reaction of HO –O
16) and (17)] with TNM [reaction (18)].
from primary hydrogens [reactions (10),
2
2
᝽
(
2
66
J. Chem. Soc., Perkin Trans. 2, 2001, 264–268

View Article Online
G(HQ) = fG(p-OH-A) ×
k12,13[O ][p-OH-A]
2
(
19)
ϩ
k12,13[O ][p-OH-A] ϩ k [H ][p-OH-A]
2
7
ϩ
G(HQ)o
G(HQ)
k [H ]
(20)
7
Ϫ 1 =
k12,13[O2]
Here G(HQ) is the G value of hydroquinone formation in
o
ϩ
the absence (i.e. lowest H concentration at neutral pH) and
ϩ
G(HQ) in the presence of H . The term f denotes the fraction
of the reactions of the p-OH-adduct radicals (abbreviated as
p-OH-A) that leads to the formation of hydroquinone.
A plot of G(HQ) /G(HQ) Ϫ 1 vs. the proton concentration
o
3
3
(
Fig. 3) yields a straight line with a slope of 3.1 × 10 dm
mol . Taking k12,13(O ϩ p-OH-A) = 1.2 × 10 dm mol
see above), k = 1.0 × 10 dm mol
Ϫ1
9
3
Ϫ1 Ϫ1
Fig. 3 Proton concentration dependence of the term G(Product)o/
G(Product) Ϫ 1 in the γ-radiolysis of a N O–O (4:1)-saturated solution
s
2
9
3
Ϫ1
Ϫ1
2
2
(
s
is obtained. The
7
of phenol, (᭹) Product = hydroquinone, (᭺), Product = catechol.
analogous plot for the formation of catechol is also shown in
8
3
Fig. 3, and based on the same assumptions k = 2.1 × 10 dm
6
Ϫ1 Ϫ1
from phenol to ozone, or whether a phenol–ozone adduct is
mol
s
is obtained. The value of k obtained in this way is
7
ϩ 9 3 Ϫ1
Ϫ
ؒ
2
formed, which releases O
᝽ /HO
to react with ozone yielding
3 Ϫ1 Ϫ1 7
2
somewhat lower than the value [k(H ) = 1.7 × 10 dm mol
fast
Ϫ
Ϫ
9
Ϫ1
O
᝽ [k(O
3
2
᝽ ϩ O
3
) = 1.6 × 10 dm mol s ].
Considering the fast elimination of O
s ] obtained above by pulse radiolysis, but is in good agree-
ment with the reported value for k(dehydration) (~1 × 10 dm
mol s ) of the p-dihydroxycyclohexadienyl radicals. On the
Ϫ
2ؒ
9
3
2
᝽ /HO from the
Ϫ1 Ϫ1
3
peroxyl radicals formed upon dioxygen addition to the OH-
adduct radicals of phenol, a chain reaction can be induced.
This makes it difficult to assess without further experiments the
primary yield of the OH-radical-forming process. According to
our scavenging experiments with tert-butyl alcohol the overall
other hand, the value of k obtained here is higher than the
6
ϩ
value of k(H )
obtained above by pulse radiolysis and the
slow
8 3 Ϫ1 Ϫ1 3
reported value (both ~1 × 10 dm mol s ) for the dehydra-
tion of the o-dihydroxycyclohexadienyl radicals.
ؒ
contribution of OH to phenol degradation is ~30%.
It is noted that the yields using dioxygen as an oxidant are
ؒ
ؒ
lower (catechol 40% of OH, hydroquinone 31% of OH) than
when p-benzoquinone was used to oxidize the dihydroxycyclo-
Acknowledgements
This work has been partly supported by the BMBF project
ؒ
hexadienyl radicals (catechol 48% of OH, hydroquinone 36%
ؒ
3
of OH). This relative difference of 13% is significant (i.e. not
0
2WT9656/6. E. M. would like to thank the German Academic
within experimental error), and we conclude that in the case
ؒ
Exchange Service (DAAD) for a stipend.
of dioxygen other reactions besides HO -elimination reactions
2
may proceed, albeit to a much lower extent than observed with
2
5
34
other aromatic compounds, e.g., benzene or terephthalate.
References
25
In the benzene study, material balance has been obtained,
and from the products that are formed under these conditions
it has been concluded that peroxyl radicals of the kind formed
in reactions (12) or (13) can also undergo intramolecular
1 M. J. Lundqvist and L. A. Eriksson, J. Phys. Chem. B, 2000, 104,
8
48.
2
3
M. Anbar, D. Meyerstein and P. Neta, J. Chem. Soc., Perkin Trans.
, 1966, 742.
2
endoperoxide formation and subsequent fragmentation. In the
N. V. Raghavan and S. Steenken, J. Am. Chem. Soc., 1980, 102,
3495.
ؒ
present system, the HO -elimination is so fast that this process
2
will contribute less to the overall decay processes.
4 E. J. Land and M. Ebert, Trans. Faraday Soc., 1967, 63, 1181.
5
M. Roder, L. Wojnarovits, G. Földiak, S. S. Emmi, G. Beggiaro and
M. D’Angelantonio, Radiat. Phys. Chem., 1999, 54, 475.
J. Hoigné and H. Bader, Water Res., 1983, 17, 185.
P. Neta, R. E. Huie and A. B. Ross, J. Phys. Chem. Ref. Data, 1988,
17, 1027.
H. R. Eisenhauer, J. Water Pollut. Control Fed., 1968, 40, 1887.
Y. Skarlatos, R. C. Barker and G. L. Haller, J. Phys. Chem., 1975,
Relevance for the reaction of ozone with phenol
6
7
In the ozonation of phenol in aqueous solutions, catechol
20
and hydroquinone are major products. At neutral pH, their
8
9
formation is suppressed when tert-butyl alcohol (which has a
7
ؒ
very low reactivity towards ozone) is added to scavenge OH.
Upon lowering the pH, the hydroquinone and catechol yields
7
9, 2587.
10 Y. Yamamoto, E. Niki, H. Shiokawa and Y. Kamiya, J. Org. Chem.,
1979, 44, 2137.
ؒ
decrease (hydroquinone more affected), as in the case of OH
1
1
1
1 P. C. Singer and M. D. Gurol, Wasser ’81 [Einundachtzig], 1981,
519.
2 J. A. Roth, W. L. Moench and K. A. Debalak, J. Water Pollut.
Control Fed., 1982, 54, 135.
3 D. Lesczynska, Environ. Prot. Eng., 1982, 8, 105.
14 P. C. Chrostowski, A. M. Dietrich and I. H. Suffet, Water Res.,
1983, 17, 1627.
5 P. C. Singer and M. D. Gurol, Water Res., 1983, 17, 1163.
6 M. D. Gurol and P. C. Singer, Water Res., 1983, 17, 1173.
7 M. D. Gurol and R. Vatistas, Water Res., 1987, 21, 895.
8 J. P. Duguet, B. Dussert, J. Mallevialle and F. Fiessinger, Water Sci.
Technol., 1987, 19, 919.
(
see above). At pH 2, their formation ceases completely. This
ؒ
strongly points to the importance of OH in the ozonation
of phenol. Similarly, but to a lesser extent, OH formation has
ؒ
also been observed in the reaction of ozone with some tertiary
35
ؒ
amines. When OH reacts with tert-butyl alcohol, formalde-
ؒ
36,37
hyde is formed in yields close to 30% with respect to OH.
In
1
1
1
1
the ozonation of phenol in the presence of tert-butyl alcohol,
we found that the formaldehyde yield (~10% with respect to
ozone reacted with phenol/phenolate) did not change notice-
ؒ
ably with pH. This is a strong indication that OH formation is
due to the reaction of ozone with phenol and, as such, does
not require the presence of phenolate in equilibrium in the
19 S. Beulker and M. Jekel, Ozone: Sci. Eng., 1993, 15, 361.
20 E. Mvula and C. von Sonntag, unpublished results.
2
2
2
1 F. Muñoz, E. Mvula, S. Braslavsky and C. von Sonntag,
unpublished results.
2 C. von Sonntag and H.-P. Schuchmann, in Peroxyl Radicals,
ed. Z. B. Alfassi, Wiley, Chichester, 1997, p. 173.
3 K. Sehested, J. Holcman, E. Bjergbakke and E. J. Hart, J. Phys.
Chem., 1987, 91, 2359.
solution [pK (phenol) = 10; note that the rate constants of
a
phenolate and phenol differ by six orders of magnitude, i.e.
phenolate reactions may dominate down to fairly low pH
values]. However, one open question remains, that is whether
Ϫ
ؒ
O3᝽
(precursor of OH) is formed directly by electron transfer
J. Chem. Soc., Perkin Trans. 2, 2001, 264–268
267

View Article Online
2
2
2
2
2
2
3
4 X.-M. Pan and C. von Sonntag, Z. Naturforsch., B: Chem. Sci.,
31 E. J. Land, G. Porter and E. Strachan, Trans. Faraday Soc., 1961, 57,
1
990, 45, 1337.
1885.
5 X.-M. Pan, M. N. Schuchmann and C. von Sonntag, J. Chem. Soc.,
Perkin Trans. 2, 1993, 289.
32 F. Jin, J. Leitich and C. von Sonntag, J. Chem. Soc., Perkin Trans. 2,
1993, 1583.
6 X. Fang, X. Pan, A. Rahmann, H.-P. Schuchmann and C. von
Sonntag, Chem. Eur. J., 1995, 1, 423.
7 C. von Sonntag and H.-P. Schuchmann, Methods Enzymol., 1994,
33 B. H. J. Bielski, D. E. Cabelli, R. L. Arudi and A. B. Ross, J. Phys.
Chem. Ref. Data, 1985, 14, 1041.
34 X. Fang, G. Mark and C. von Sonntag, Ultrason. Sonochem., 1996,
2
33, 3.
8 R. H. Schuler, L. K. Patterson and E. Janata, J. Phys. Chem., 1980,
4, 2088.
9 G. V. Buxton and C. R. Stuart, J. Chem. Soc., Faraday Trans., 1995,
3, 57.
35 F. Muñoz and C. von Sonntag, J. Chem. Soc., Perkin Trans. 2, 2000,
8
2029.
36 M. N. Schuchmann and C. von Sonntag, J. Phys. Chem., 1979, 83,
9
1, 279.
780.
0 G. V. Buxton, C. L. Greenstock, W. P. Helman and A. B. Ross,
37 R. Flyunt, A. Leitzke, G. Mark, E. Mvula and C. von Sonntag,
unpublished results.
J. Phys. Chem. Ref. Data, 1988, 17, 513.
2
68
J. Chem. Soc., Perkin Trans. 2, 2001, 264–268