The Origin of Base Catalysis in the OH Oxidation of Phenols in Water

Article abstract of DOI:10.1021/jp031365j

Time-resolved Raman and transient absorption studies on hydroxyl ( OH) radical reactions with aqueous 4-carboxyphenol ( ^-O₂C-PhOH), the substrate in p-hydroxybenzoate hydroxylase, have led to the first direct observation of the basic form of a dihydroxycyclohexadienyl (OH-adduct of phenolate anion) radical and hydration-induced intramolecular electron transfer in this species. The base-catalyzed phenoxyl radical formation in this model system has been quantitatively described in terms of pK_a of the phenolic proton (9.05) of the OH-adduct and the rate of OH ^- elimination (2.9 × 10⁶ s^-1) from its basic form. A small fraction, 12(+2)%, of the phenoxyl radical is formed via water elimination from the OH-adduct at the ipso position of the hydroxyl group (rate > 10⁷s^-1). About 3% OH addition is seen at the carboxylic ipso position in basic solutions, which produces the p-benzosemiquinone radical anion (rate ～ 10⁶s^-1). This work provides spectroscopic and kinetic evidence of the early chemical steps in the phenoxyl radical formation by OH oxidation and establishes the precise relationship between the formation rate and pH. A relationship between the rates of OH^- elimination from the OH adducts of phenolate anions and pK _a of the corresponding phenols is given.

Full text of DOI:10.1021/jp031365j

3478
J. Phys. Chem. A 2004, 108, 3478-3484
•
The Origin of Base Catalysis in the OH Oxidation of Phenols in Water
G. N. R. Tripathi* and Yali Su
Radiation Laboratory, UniVersity of Notre Dame, Notre Dame, Indiana 46556
ReceiVed: December 31, 2003
Time-resolved Raman and transient absorption studies on hydroxyl (•OH) radical reactions with aqueous
4-carboxyphenol (-O2C-PhOH), the substrate in p-hydroxybenzoate hydroxylase, have led to the first direct
observation of the basic form of a dihydroxycyclohexadienyl (•OH-adduct of phenolate anion) radical and
hydration-induced intramolecular electron transfer in this species. The base-catalyzed phenoxyl radical formation
in this model system has been quantitatively described in terms of pKa of the phenolic proton (9.05) of the
•OH-adduct and the rate of OH- elimination (2.9 × 106 s-1) from its basic form. A small fraction, 12(+2)%,
•
of the phenoxyl radical is formed via water elimination from the OH-adduct at the ipso position of the
•
hydroxyl group (rate > 107s-1). About 3% OH addition is seen at the carboxylic ipso position in basic
solutions, which produces the p-benzosemiquinone radical anion (rate ∼ 106s-1). This work provides
•
spectroscopic and kinetic evidence of the early chemical steps in the phenoxyl radical formation by OH
oxidation and establishes the precise relationship between the formation rate and pH. A relationship between
the rates of OH- elimination from the •OH adducts of phenolate anions and pKa of the corresponding phenols
is given.
•
Introduction
solution and the OH-adduct to a phenolate anion in basic
solutions, have never been observed. Their observation and
decay properties are crucial to understanding the relationship
between the rate of phenoxyl radical formation and pH.
Phenoxyl radicals represent the intermediate state in the
oxidation of phenols and reduction of quinones. They have been
identified in a number of biochemical systems, such as proteins
and enzymes, and are implicated in cancer and aging.1-6 The
background emission observed with certain peroxyoxalate
chemiluminescence reagents has been attributed to the phenoxyl
radical.3 They are associated with anti-oxidant properties of
phenols that have found medical and industrial applications. The
mechanism of formation, modes of reaction, and structural
properties of phenoxyl radicals have been studied extensively.6-7
Synthetic models have been developed recently for understand-
ing the copper-phenoxyl radical array in galactose oxidase.4
Phenoxyl radicals display CO bonds of various strengths, which
have been studied recently by time-resolved Raman and
electronic structure computations.8-10 A wide variation is
seen in the redox potentials (E°) of phenoxyl-phenolate couples
(X-PhO•/XPhO-) in aqueous solution which have been used to
determine the O-H bond energy in phenols.11
During the course of time-resolved Raman studies of phe-
noxyl radicals derived from aqueous 4-carboxyphenol (-O2C-
PhOH), the substrate in oxygen-consuming enzymes p-hydroxy-
benzoate hydroxylase,12 the vibrational features were found to
be similar to that of the unsubstituted radical (PhO•), showing
-
minor effects of -CO2 substitution on bond properties. The
reduction potential of the -O2C-PhO•/-O2C-PhO- couple
(0.90 eV) is only slightly (∼12%) higher than that of PhO•/
PhO- (0.80 eV). A weak perturbation of charge distribution in
PhO- by the -CO2 group is also indicated by the proton
-
dissociation constant (pKa) of -O2C-PhOH (9.35) which is
comparable to that of PhOH (10).13 Thus, the structure and
chemical properties of -O2C-PhOH are close to those of PhOH.
However, the -CO2- perturbation, although small, slows down
the rate of OH- elimination from the OH-adduct of -O2C-
•
PhO-. If the rate of reaction of X-PhO- with OH can be
•
•
The oxidation of phenols by the OH radical in water has
accelerated to the extent that it becomes faster than the rate of
phenoxyl radical (X-PhO•) formation, the adduct precursor of
the phenoxyl radical (X-PhO-(OH)•) can be observed. This
condition is achievable for -O2C-PhOH due to its very high
solubility in water.
remained a topic of extensive research activity because of its
environmental, chemical, and biochemical implications. In the
commonly held reaction mechanism, the •OH radical reacts with
phenols by addition to the ring carbon sites. Phenoxyl radicals
are produced by acid/base-catalyzed loss of water from the
•OH-adducts (dihydroxycyclohexadienyl radicals).5-7 A push-
pull mechanism has been proposed to explain the catalysis.7 In
this mechanism, an aqueous proton attaches to the cyclohexa-
dienyl•-OH (•OH-adduct) in acidic solutions and pulls out a
water molecule, with consequent loss of the phenolic proton.
In basic solutions, the hydroxyl ion removes the phenolic proton,
transferring electronic charge to the cyclohexadienyl•-OH and
pushing OH- out into solution. However, the key intermediates
We have obtained the transient absorption spectrum of an
•OH-adduct of a phenolate anion for the first time, using -O2C-
PhOH as a model system. It is this intermediate that is mainly
responsible for the phenoxyl radical (-O2C-PhO•, Figure 1)
formation in neutral and basic aqueous solutions. The observa-
tion of X-PhO-(OH)• is key to understanding the early chemical
steps in the •OH oxidation of aqueous phenols in basic solutions.
The acid-base equilibrium between the •OH-adducts of phenol
and phenolate anion in this model system has been examined
and a relationship between the rate of phenoxyl radical formation
and pH has been derived. The rate of OH- elimination (ke) from
in the mechanism, the OH2+-adduct to a phenol in acidic
•
* Corresponding author. E-mail: tripathi@hertz.rad.nd.edu.
10.1021/jp031365j CCC: $27.50 © 2004 American Chemical Society
Published on Web 03/30/2004

•
Base Catalysis in the OH Oxidation of Phenols
J. Phys. Chem. A, Vol. 108, No. 16, 2004 3479
Figure 1. UB3PW91structure of -O2C-PhO• (see text).
X-PhO-(OH)• decreases with pKa of the phenol (X-PhOH). A
relationship between the pKa of the phenolic proton and the
rate of phenoxyl radical formation has been established.
The •OH oxidation of 4-carboxyphenol has been investigated
previously by transient absorption, but misassignment of the
phenoxyl absorption has occurred.14 More importantly, the
•OH adduct of the phenolate anion, the key intermediate in the
base-catalyzed phenoxyl radical formation, as shown in this
work, was not observed.
Figure 2. Transient absorption spectra obtained on electron pulse
irradiation of N2O-saturated aqueous solution containing (A) 1 mM
-O2C-PhOH and 0.1 M NaN3 at pH 11.3, 1 µs after the pulse, (B) 10
mM -O2C-PhOH at pH 11.2, 5 µs after electron pulse, (C) 10 mM
-O2C-PhOH and 0.2 M t-BuOH at pH 11, 5 µs after electron pulse.
(D) B-C.
Experimental Section
4-Carboxyphenol (-O2C-PhOH) was oxidized to its phe-
noxyl radical state in aqueous solution by pulse radiolysis. On
electron pulse irradiation of N2O-saturated water, the •OH radical
is the main reactive species present in solution on the 100-ns
time scale.15 When desired, the •OH radical was converted into
secondary oxidant N3• by reaction with 0.1 M NaN3 in solution.
Pulse radiolysis-time-resolved optical absorption and resonance
Raman techniques, described in detail in several previous
publications from this Laboratory, were used for transient
detection.8,16 Radiolysis by 8 MeV, 5 ns electron pulses from a
linear accelerator facility in the Laboratory, which typically
produces a radical concentration of 3 × 10-6 M per pulse, was
used in optical absorption measurements. In Raman experiments,
2 MeV, ∼100 ns electron pulses, delivered by a Van de Graaff
accelerator at variable dose rates, capable of producing 5 ×
10-4 M and lower radical concentrations, were applied. The
Raman scattering was probed by an excimer (100 mJ)-pumped
dye laser pulse (10 ns), tuned in resonance with the optical
absorption of the radicals. The spectra were recorded by using
an optical multichannel analyzer (OMA), accompanied with an
intensified gated diode array detector, with the gate pulse
synchronized with the Raman signal pulse. Extensive signal
averaging was performed to improve the S/N ratio in the Raman
spectra, with the accelerator and laser operated at a repetition
rate of 7.5 Hz. In both experiments, a flow system was used to
refresh the solution between consecutive electron pulses. Raman
band positions were measured with reference to the known
Raman bands of common solvents, such as ethanol, and are
accurate to within (2 cm-1 for sharp bands and (5 cm-1 for
broad and shoulder bands.
Figure 3. (A) Raman spectrum excited at 420 nm, 100 ns after electron
pulse irradiation of a N2O-saturated aqueous solution containing 5 mM
-O2C-PhOH and 0.1 M NaN3 at pH 11. The transient is identified as
-O2C-PhO•. (B) Raman spectrum before irradiation. Spectrum (A) is
displayed after subtraction of contribution by (B) from the raw 100 ns
spectrum.
•
the species produced by N3 oxidation is indeed a phenoxyl
radical was ascertained by recording its resonance Raman (RR)
spectra, at several excitation wavelengths between 390 and 435
nm, at different time intervals after the electron pulse. Because
of low extinction coefficient, the resonance enhancement (
ꢀmax2) by the radical is relatively weak. Therefore, the spectra
were recorded at the maximum permissible radiation dose rate.
This is one of the rare chemical transients with ꢀmax < 2 × 103
M-1 cm-1 for which resonance Raman spectra could be
obtained. Figure 3A shows the transient Raman spectrum in
the 800-1700 cm-1 region. The spectrum was excited at 420
nm, 100 ns after the electron pulse. A strongly enhanced 1515
Results and Discussion
Spectroscopic and Structural Characterization of the
Aqueous -O2C-PhO•. The transient absorption observed 2 µs
after electron pulse irradiation of an N2O-saturated aqueous
solution containing 1 mM -O2C-PhOH and 0.1 M NaN3 at
cm-1 Raman band dominates the spectrum. The spectral features
+• 17
pH 11.3 is shown in Figure 2A. The absorption maximum (λmax
)
are analogous to that of isoelectronic -O2C-PhNH2
.
is located at 415 nm (ꢀmax ∼ 1650 M-1 cm-1), with a prominent
shoulder band at 398 nm (ꢀmax ∼ 1450 M-1 cm-1). Oxidation
The intensity profiles in the 390-440 nm RR spectra of
phenoxyl (X-PhO•) radicals show marked dependence on the
nature of the substituent group (X).8 The Raman spectrum in
Figure 3A is typical of a phenoxyl radical that is very little
affected by the nature of X. It closely resembles the RR spectra
of X-PhO•, with X ) H, CH3, and F. The spectral assignment
is straightforward, based on comparison with p-substituted
phenoxyl radicals. The strong band at 1515 cm-1 in Figure 3A
corresponds to the 1505 cm-1 band of PhO•, and represents the
•
in this case is by N3 radical, which is believed to react by direct
electron transfer. The reaction period, as determined by temporal
evolution of optical absorption in Figure 2A, was 250((15) ns
which gave a rate constant of 4.0((0.2) × 109 M-1 s-1 for the
reaction of N3 with -O2C-PhO-.
•
The absorption spectrum in Figure 2A is different from the
spectrum attributed to -O2C-PhO• in the literature.14 Whether

3480 J. Phys. Chem. A, Vol. 108, No. 16, 2004
Tripathi and Su
of this radical and the transient oxidation products of -O2C-
PhO• in Figure 4D. However, the spectrum in Figure 4D
contains characteristic features of the radicals derived from
biphenyl systems. The inter-ring stretch is observed at a
frequency of ∼1330 cm-1 in these radicals (note the presence
of a band at 1337 cm-1 in Figure 4D).17,19,21 The ortho and
para carbon sites have high spin density in phenoxyl and its
isoelectronic radicals. In para-substituted radicals, it is the
ortho-ortho coupled product that is obtained.19 We tentatively
-•
assign the spectrum in Figure 4D to 2,2′-(-O2C-PhO)2
.
It should be mentioned that -O2C-PhNH2+• is an exception
to the radicals derived from para-substituted benzenes in which
the tail-tail coupled radical product (benzidine cation radical)
has been observed in mildly acidic solutions.17 It is a conse-
quence of slow decarboxylation of -O2C-PhNH2+• on the 100
µs time-scale. -O2C-PhNH2 is isoelectronic with -O2C-
+•
PhO•. However, no evidence of decarboxylation of -O2C-PhO•
was found.
The Mechanism of the •OH Oxidation of Aqueous -O2C-
PhOH. Yield of Phenoxyl Radical. Figure 2B depicts the
absorption spectrum obtained on pulse radiolysis of an N2O-
saturated aqueous solution containing 10 mM -O2C-PhOH at
pH 11.2, 5 µs after the electron pulse. Under the chemical
conditions used, this spectrum should mainly contain contribu-
tions from the transients produced by •OH oxidation. Compared
Figure 4. 435 nm Raman spectra observed on electron pulse irradiation
of a N2O-saturated aqueous solution containing 5 mM -O2C-PhOH
and 0.1 M NaN3 at pH 11. Time-interval (A) 100 ns, (B) 2 µs, (C) 5
µs, and (D) 100 µs. (E) No electron beam. Spectra in A-D are depicted
after subtraction of (E) from the raw spectra. (F) Spectrum of the p,p′-
biphenylsemiquinone anion radical, (OPh-PhO)-•, at pH 11.
CO stretching mode. The very weak 1400 cm-1 band corre-
sponds to a ring CC stretching vibration with a large CH bending
component. The 1172 cm-1 vibration mainly involves CH
bending motion and the 979 cm-1 vibration ring distortion. It
has been shown previously that the ∼980 cm-1 ring distortion
mode is resonance-enhanced parallel to the ∼1500 cm-1
mode.8,9 Identification of the observed species was further
supported by comparison of the experimental Raman frequencies
with calculated frequencies using UHF and DFT (UB3PW91)
computational procedures, details of which will be published
elsewhere. The UB3PW91/6-31G** structure of the radical is
furnished in Figure 1.18
Transient Products. The resonance Raman spectra (Figure
4) recorded at different time intervals after the electron pulse
show the effect of the subsequent chemistry. The 1515 cm-1
-O2C-PhO• Raman band is reduced to about half of its intensity
2 µs after the electron pulse and bands at 1594 and 1337 cm-1
appear (Figure 4B). At 100 µs after the electron pulse the 1515
cm-1 band disappears almost completely and the 1594 and 1337
cm-1 bands gain their maximum intensity (Figure 4D). Two
weaker bands at 1530 and 1509 cm-1 also become apparent.
Figure 4E shows Raman signals when no radiolysis was applied.
These background signals (Figure 4E) have been already
subtracted from the spectra in Figures 4A-D. It is known that
phenoxyl radical and anisole and aniline cation radicals decay
by tail-tail coupling.17,19-21 The coupled product (4,4′-
substituted biphenyl derivative) is oxidized by its radical
precursor to form a radical that has an overlapping absorption
and therefore is seen in resonance Raman as a transient decay
product of the primary species. The mechanism of formation
of the 4,4′-substituted biphenyl radicals in the reaction has been
examined in detail in several previous publications from this
Laboratory.17,19,21 The coupled radical products are generally
not produced when the para-position of the primary radicals is
blocked by a substituent group.17 The RR spectrum of the p,p′-
biphenylsemiquinone anion radical [(C6H4O)2]-•, produced on
oxidation of 4,4′-dihydroxybiphenyl at pH 11, is displayed in
Figure 4F. There is no equivalence between the Raman signals
•
to the spectrum in Figure 2A, produced by the reaction of N3
with -O2C-PhO-, the spectrum in Figure 2B looks different.
In particular, there is a prominent absorption peak at ∼400 nm
in Figure 2B that is not seen in Figure 2A. To determine if the
•
transient spectrum in Figure 2B is indeed due to the OH
reaction, 0.2 M tert-butyl alcohol was added to the solution to
scavenge the •OH radical. The spectrum obtained on radiolysis
of this solution, 5 µs after the electron pulse, is shown in Figure
2C. This spectrum is tentatively attributed to the H• atom adduct
(-O2C-PhO-(H•)). In N3• oxidation, the H• atom is completely
scavenged by N3- because of its high concentration in solution22
and the -O2C- PhO-(H)• species is not formed. Figure 2D is
the difference spectrum of Figures 2B and 2C. There is no
difference in the absorption spectra in Figures 2A and 2D. The
400 nm transient was attributed previously to -O2C-PhO•.14
Further identification of the transient represented by the
absorption in Figure 2D was done by time-resolved resonance
Raman. Figure 5 shows the transient Raman spectra in the
1300-1800 cm-1 region, at different time intervals after the
electron pulse. The RR spectrum in Figure 5A is identical to
•
the one obtained on N3 oxidation and given in Figure 4A, except
for an additional very weak Raman peak at 1620 cm-1. The
•
Raman signal intensity of -O2C-PhO•, produced by OH
oxidation, was comparable to that of the radical produced by
N3• oxidation of -O2C-PhOH, within experimental uncertainly
((10%),23 suggesting comparable yield of phenoxyl radical in
both reactions.
In near neutral solutions a small amount of phenoxyl radical
is formed at a fast rate (<100 ns). Figure 6A depicts the
resonance Raman spectrum obtained on pulse radiolysis of an
N2O-saturated aqueous solution containing 5 mM -O2C-PhOH
at pH 7 (5 mM phosphate buffer), 1 µs after the electron pulse.
•
The signal intensity is only 12% of the intensity seen on N3
oxidation (Figure 6B). We attribute this yield to the water loss
from the •OH adduct at the ipso position to the hydroxyl group,
although it may have some contribution from the hydrogen atom
abstraction or direct electron transfer. It should be noted that
the yield of the ipso adduct determined here by the Raman
method has considerable uncertainly due to the baseline problem

•
Base Catalysis in the OH Oxidation of Phenols
J. Phys. Chem. A, Vol. 108, No. 16, 2004 3481
Figure 7. The growth (A) and decay (B) of the p-benzosemiquinone
radical anion signal with time, on electron pulse irradiation of N2O-
saturated aqueous solution of 5 mM HO2C-PhOH at pH 11. (C) Plot
of ln(I0 - I) vs time. I is Raman signal intensity at time t and I0
maximum Raman signal intensity of p-benzosemiquinone radical anion,
monitored by the 1620 cm-1 peak.
Figure 5. Raman spectra excited at 435 nm and observed (A) 100 ns,
(B) 200 ns, (C) 500 ns, (D) 1 µs, (E) 2 µs, (F) 5 µs, and (G) 100 µs
after electron pulse irradiation of a N2O-saturated aqueous solution of
5 mM -O2C-PhOH at pH 10.7. (H) 435 nm Raman spectrum of
p-benzosemiquinone radical anion produced on electron pulse irradiation
of N2O-saturated aqueous solution of 5 mM hydroquinone at pH 11.
Figure 8. Absorption spectra obtained on electron pulse irradiation
of a N2O-saturated aqueous solution containing 10 mM -O2C-PhOH
and 5 mM phosphate buffer at (A) pH 7, 2.0 µs after the pulse, and
(B) pH 11, 100 ns after the pulse.
The Adduct Precursors. The absorption spectra of numerous
•OH-adducts of phenols are reported in the literature, with λmax
in the 360 nm region.5-7 These transients decay by reaction
with base to form the phenoxyl radical. On electron pulse
radiolysis of N2O-saturated aqueous -O2C-PhOH at pH ∼ 7
Figure 6. Raman spectrum excited at 420 nm, 100 ns after electron
pulse irradiation of a N2O-saturated solution of 5 mM -O2C-PhOH
containing (A) 5 mM phosphate buffer at pH 7 and (B). 0.1 M NaN3
at pH 10.8.
(phosphate buffer), a transient absorption at 365 nm (λmax
)
appears as the initial reaction product which we attribute mainly
to the •OH adduct of -O2C-PhOH at the ortho position (Figure
8A). From the rate of formation of this species, the rate constant
•
in Figure 6A. The remaining ∼88% of the OH addition must
•
occur at ortho, meta, and para ring positions of the hydroxyl
for the reaction of OH with -O2C-PhOH was determined as
group in -O2C-PhOH. The OH attack at the meta positions
6((1) × 109 M-1 s-1. We have estimated that about 73((10)%
of the •OH radicals add to the ortho position. There is a drop in
the yield of the 365 nm transient with pH (Figure 9), indicating
that this species is in equilibrium with another species, which
is a likely precursor of the phenoxyl radical. This precursor
species can be readily observed if its rate of formation is faster
than that of the phenoxyl radical. This condition is satisfied
with a high concentration (>5 mM) of -O2C-PhOH in solution.
A high concentration of -O2C-PhOH shortens the reaction
•
of phenols is generally small. The meta isomer of the adduct
radical is also relatively long-lived. The yield of the meta isomer
•
has been estimated as 12% of the total OH yield by ESR.24
Therefore ∼76% of the •OH radicals must add at the ortho and
para positions, combined.
Yield of p-Benzosemiquinone Radical Anion. The weak
Raman signal at 1620 cm-1 in Figure 5A, that is not present in
the spectrum of -O2C-PhO• in Figure 3A, was identified with
the p-benzosemiquinone radical anion, by comparison with the
435 nm RR spectrum obtained on oxidation of hydroquinone
by •OH in basic aqueous solution (Figure 5H).8,9,25 The
maximum yield of the radical, determined by comparison with
the latter spectrum, was estimated as ∼3%. This signal grows
at the rate of ∼106 s-1 and decays on the 100 µs time scale
(Figures 7A,B,C). This is a minor product of the reaction. The
Raman detection of the p-benzosemiquinone radical anion in
•
period with OH and catalyzes the acid-base equilibrium of
•
the OH adduct (this catalysis is briefly discussed below). In
the pH range 7-10, OH- concentration in solution is too low
to induce equilibrium on the sub-microsecond or a shorter time
scale. Considering a diffusion-controlled limit for the rate of
reaction of the •OH adduct with the parent phenol, at least a 10
mM concentration of -O2C-PhOH is required to attain acid-
base equilibrium on the 10 ns time scale. A high substrate
concentration is key to observing the •OH adduct of a phenolate
anion and following its decay into the phenoxyl radical.
The absorption spectrum of the transient produced on reaction
of •OH with 10mM -O2C-PhOH at pH 11 is shown in Figure
•
the OH oxidation of several para-substituted phenols and the
mechanism of formation have been discussed previously.8 We
•
conclude that about 73% of the OH radicals add to the ortho
position of -O2C-PhO-.

3482 J. Phys. Chem. A, Vol. 108, No. 16, 2004
Tripathi and Su
Figure 11. Time dependence of absorption at 420 nm at pH 10.74,
9.55, 9.05, and 8.75 for N2O-saturated solution of 10 mM -O2C-PhOH
and 5 mM phosphate buffer. There is no change in absorbance with
time at pH 9.05, as decay of absorption due to the •OH adduct of -O2C-
PhO- is compensated by the growth of absorption due to the phenoxy
radical (-O2C-PhO)•. The extinction coefficient of the later radical is
half of its precursor at 420 nm.
Figure 9. Time dependence of the 365 nm absorption at pH 8.21,
8.75, 9.05, 9.55, and 10.74 for N2O-saturated aqueous solution of 10
mM -O2C-PhOH with 5 mM phosphate buffer.
11, the OH-adduct of -O2C-PhO- (420 nm transient, Figure
8B) decays at a rate of 2.9 × 106 s-1 (t1/2 ) 0.24 µs), and this
rate is independent of pH. As pointed out earlier, time-resolved
Raman positively identifies the species formed on decay of the
•OH-adduct as the phenoxyl radical. The rate of OH- loss from
-O2C-PhO-(OH)• (ke ) 2.9 × 106 s-1) is an order of
magnitude faster than the rate of OH--elimination from the
•OH adduct of -O2C-PhNH2.17 However, the rate of OH-
elimination from the protonated form, -O2C-PhOH(OH)•, is
much slower (see later sections). The amine group behavior is
intermediate between that of O- and OH in this respect.
The pH-Dependence of Phenoxyl Radical Formation. The
acid-base equilibria of the parent phenols occurs in the same
•
pH range as that of the OH adducts. The rate constant for the
reaction of OH is usually identical for both forms of phenols.
Figure 10. pH dependence of the percentage initial absorbance of (A)
365 nm transient and (B) 420 nm transient, measured at different
wavelengths, after electron pulse irradiation of N2O-saturated aqueous
solution containing 10 mM -O2C-PhOH and 5 mM phosphate buffer.
•
If the rate of phenoxyl radical formation from the basic form
of phenol were faster than that from the acid form, the observed
•
•
pKa of the OH adduct determined as 9.05 ((0.05) from the acid-
rate would show pH-dependence, even if the OH did not add
to phenolate. However, the decay of -O2C-PhOH(OH)• would
not correspond with the growth of the phenoxyl radical that is
commonly observed.
base titration curve.
8B. The λmax in this spectrum is similar to that of -O2C-PhO•
(Figure 2A), but ꢀmax is about twice as large at 420 nm. The
initial (∼30 ns) absorbance of the 420 and 365 nm transients
monitored with pH is shown in Figure 10. It can be seen that
the two species interconvert with change in pH. There is little
doubt that the 365 and 420 nm transients (Figure 8) represent
the acid and basic forms of the same species. The pKa of the
species is determined as 9.05 from Figure 10. If the acid-base
equilibrium is not established, for which a high substrate
concentration in solution is required, the pH-dependence of the
365 nm absorbance will give a false pKa. Also, the rate of
With the observation of -O2C-PhO-(OH)• and its rate of
conversion into phenoxyl radical, a precise relationship between
the rate of phenoxyl radical formation and pH can be worked
•
out. Let us consider the acid-base equilibrium of the OH
adduct and phenoxyl formation, as shown below:
•
phenoxyl radical formation from the basic form of the OH
adduct cannot be correctly estimated. To our knowledge, no
•
earlier report on the acid-base equilibrium of the OH adduct
of a phenol, as presented in Figure 10, can be found in the
literature.
B is the total base concentration in solution. Equilibrium (1)
will be reached only when kr + kf . ke.26 The rate (k) of
phenoxyl (-O2C-PhO•) radical formation is given by
Intramolecular Electron Transfer and Phenoxyl Radical
Formation. The mechanism of phenoxyl radical formation
becomes quite clear as one monitors the rate of decay of the
•OH adducts and formation rate of the phenoxyl radical (Figures
9 and 11). As -O2C-PhO• and its adduct precursor have
overlapping and rather similar absorptions (Figures 2B and 8B),
it would be difficult to make a distinction between them,
particularly if they had comparable absorbance, unless resonance
Raman spectroscopy is applied as a diagnostic tool. At pH >
k ) ke kf /(kr + kf)
(2)
•
where kf /(kr + kf) is the fraction of the basic form of the OH
adduct at a particular pH of solution which can be estimated
from Figure 10. The solid line in Figure 10A represents a
theoretical expression for the equilibrium fraction {(kf /(kr +
kf)} of -O2C-PhO-(OH)•, given by 10(pH-pK )/(1 + 10(pH-pKa)),
a

•
Base Catalysis in the OH Oxidation of Phenols
J. Phys. Chem. A, Vol. 108, No. 16, 2004 3483
•
TABLE 1: Observed and Estimated Rates of Decay of OH
Adducts to Phenolate Anions Forming Phenoxyl Radical
(X-PhO-(OH)• f X-PhO•+ OH-) in Basic Aqueous
Solutions
rate of phenoxyl radical formation (ke)
phenol
(XC6H4OH)
pKaa
kcal
kobs
-
X ) CO2
9.35
2.9 × 106
7.7 × 105
2.9 × 106
(9.05)b
X ) COCH3
X ) CHO
X ) CN
X ) NO2
X ) H
8.79
7 × 105
4 × 105
3 × 105
7 × 105
c
8.3
7.05
9.9
2.5 × 105
7 × 104
108
>107
•
a pKa of phenols (XPhOH). b pKa of OH adduct (XPhOH(OH)•).
c Estimated pKa) 8.69, assuming kcal ) kobs
.
water (pKa ∼ 15).28 In other words, the added OH group is not
part of the conjugated system. Therefore, one can safely assume
minimal dependence of the ring-OH bonding in the •OH adduct
(X-PhO-(OH)•) and its transition state (X-PhO•(OH-)) on the
nature of X-PhO-. On the other hand the negative charge on O
is delocalized in X-PhO-, which amounts to structural stabiliza-
tion by a free energy (∆G° in eV) difference of 0.059 × ∆pKa,
with respect to a hypothetical reference system (∆pKa ) 0) for
which there is no delocalization of the charge.29 ∆G° is
additional energy required to attain the transition state (X-PhO--
(OH)• f X-PhO•(OH-)), and is expected to lower the rate of
OH- elimination by a factor of exp(-∆G°/RT). In other words,
ke ) k0 exp(-∆G°/RT), where k0 is a constant that relates to
the reference system (e.g., pKa ) 15).29,30 Using ke ) 2.9 ×
106 s-1, pKa ) 9.05 and ∆pKa ) 5.95 for -O2C-PhOH(OH)•,
Figure 12. Comparison between observed (kobs) and calculated (kcal
rates of the phenoxyl radical formation at different pH levels.
)
and the solid line in Figure 10B that of -O2C-PhOH(OH)• (kr /
(kr + kf)), given by 1/(1 + 10(pH-pK )). The observed and
a
calculated values of k for different pH levels are compared in
Figure 12. It can be seen that there is an excellent agreement
between the experimental and theoretical values of k. If a similar
behavior is assumed for all phenols, the only parameters that
are needed for quantitative estimation of the rate of phenoxyl
radical formation (k) are the pKa of X-PhOH(OH)• and ke.27
As pointed out earlier, the equilibrium (1) is catalyzed by
the substrate. At pH < 9 (kf , kr), k can be approximated as
(kf /kr) ke. The decay rate of the 365 nm transient, monitored at
pH 8.4, was found to increase almost linearly with the -O2C-
PhOH concentration (between 1 mM and 10 mM) in solution,
according to the relation, k (s-1) ) 7.5 × 104 + 2.8 ×
108[-O2C-PhO-]. This dependence suggests that the rate
constant for the reaction of -O2C-PhOH(OH)• with -O2C-
PhO- is slower than the rate constant for the reaction of
-O2C-PhO-(OH)• with -O2C-PhOH, which we estimate as
g109 M-1 s-1 and g 1.5 × 109 M-1 s-1, respectively. The term
7.5 × 104 s-1 in k combines the rates of OH--induced and
spontaneous loss of water from -O2C-PhOH(OH)•, which we
roughly estimate as 1.5 × 104 s-1 and 6 × 104 s-1, considering
that OH- should react at a diffusion-controlled rate.
•
we estimate k0 ) 2.5 × 1012 s-1. Since pKa for any other OH
adduct (X-PhOH(OH)•) is not yet known, it is useful to relate
ke to pKa of the parent phenol (X-PhOH), which is 0.3 eV higher
than that of the •OH adduct for X ) -O2C. Thus, we use ke )
k0′ exp(-∆G°/RT), where k0′ ) k0 exp(-0.3 × 0.059/RT) )
1.25 × 1012 s-1, and ∆G° (in eV) ) 0.059 × ∆pKa for the
phenol, for estimation of ke in different systems.
The estimated decay rates of the •OH adducts forming
phenoxyl radical ((X-PhO-(OH)• f X-PhO• + OH-) are given
in Table 1 for a few selected phenolates. Experimental deter-
mination of ke has been made for X ) CN, CHO, and COCH3
in p-X-PhOH by conductivity method.7 For X ) CN (pKa )
8.3), estimated ke ) 2.5 × 105 s-1 and observed ke ) 3 × 105
s-1. For X ) COCH3 (pKa ) 8.79), estimated ke ) 7.7 × 105
s-1 and observed ke ) 7 × 105 s-1. It can be seen that there is
an excellent agreement between the estimated ke and the
observed ke for the systems for which experimental data is
available. pKa for X ) CHO could not be found in the literature.
The decay kinetics of -O2C-PhO-(OH)• transient (420 nm)
and its conversion into the phenoxyl radical is shown in Figure
11. It can be seen that the excess absorption at 420 nm decays
at a slower rate with decrease in pH, determined by the fraction
of adduct to phenolate at ∼30 ns. At pH 9.05, there is no change
in absorbance with time, as the drop in the absorbance of the
adduct radical due to its decay into the phenoxyl radical is
compensated by the growth of absorbance of the phenoxyl
radical. At pH 8.75, there is a slight growth of absorbance at
420 nm, as the -O2C-PhO-(OH)• absorption at ∼30 ns is less
than the absorption of the phenoxyl radical into which it decays.
For X ) NO2, pKa ) 7.05 and estimated ke ) 1.4 × 104 s-1
.
This rate is quite low and the formation of phenoxyl radical is
unlikely to be observed due to interference of radical-radical
reactions, unless the experiments are performed at initial radical
concentration of ∼10-6 M or less. For unsubstituted phenoxyl
(X ) H), pKa ) 9.9 and the estimated ke ∼ 108 s-1. Therefore,
the •OH adduct to phenolate anion is potentially observable, if
experiments are performed at a substrate concentration of >100
mM. To our knowledge, the reaction of hydroxyl radical with
high concentrations of aqueous phenols has not yet been
investigated.
•
Relationship between pKa of the OH Adduct and ke.
Trends in the data available in the literature indicate that the
•
rate of phenoxyl radical formation from the OH adduct of a
phenolate anion (ke) depends on pKa of its acid form. If pKa is
too low, as in nitrophenols, formation of the phenoxyl radical
is not observed. Unfortunately, the pKa values of phenolic
protons in the •OH adducts are not known. Therefore we assume
that they are comparable to or slightly lower (∼0.3 eV) than
pKa values of the parent phenol, in view of the results obtained
in this work.
Summary
The structure and functions of phenoxyl radicals in proteins
and enzymes and DNA interactions in biochemical systems are
of current scientific interest. The reaction of hydroxyl radical
The pKa values of the hydroxycyclohexadienyl radicals are
similar to those of the hydroxyl proton in aliphatic alcohols or

3484 J. Phys. Chem. A, Vol. 108, No. 16, 2004
Tripathi and Su
(•OH) with organic substrates is of fundamental importance in
environmental chemistry, synthetic organic chemistry, and
radiation biology. This paper presents a mechanistic investiga-
tion of the hydroxyl radical reaction with phenols, using
(8) Tripathi, G. N. R. In AdVances in Spectroscopy, Vol. 18, Time-
resolVed Spectroscopy; Clark, R. J. H., Hester, R. E., Eds.; John Wiley &
Sons: New York, 1989; pp 157-218.
(9) Tripathi, G. N. R. J. Chem. Phys. 2003, 118, 1378. Tripathi, G. N.
R. J. Phys. Chem. A 1998, 74, 6044; J. Am. Chem. Soc. 1998, 120, 5134.
Tripathi, G. N. R.; Sun, Q.; Schuler, R. H. Chem. Phys. Lett. 1989, 156,
51. Tripathi, G. N. R.; Schuler, R. H. J. Phys. Chem. 1987, 91, 5881.
Tripathi, G. N. R.; Schuler, R. H. J. Phys. Chem. 1988, 92, 5129. Sun, Q.;
Tripathi, G. N. R.; Schuler, R. H. J. Phys. Chem. 1990, 94, 6273. Tripathi,
G. N. R.; Schuler, R. H. J. Chem. Soc., Faraday Trans. 1993, 89, 4177.
Mukherjee, A.; McGlashen, Z.; Spiro, T. G. J. Phys. Chem. 1995, 99, 4912.
For earlier references, see citations in ref 6.
(10) Engstro¨m FS, M.; Himo, F.; Gra¨slund, A.; Minaev, B.; Vahtras,
O.; A° gren, H. J. Phys. Chem. A 2000, 104, 5149. Wise, K. E.; Pate, J. B.;
Wheeler, R. A. J. Phys. Chem. B 1999, 4764. Chipman, D. M.; Liu, R.;
Zhou, X.; Pulay, P. J. Chem. Phys. 1994, 116, 44, and references therein.
(11) Das, T. N.; Neta, P. J. Phys. Chem. A 1998, 102, 7081. Lind, J.;
Shen, X.; Eriksen, T. E.; Mere´nyi, G. J. Am. Chem. Soc. 1990, 112, 479,
and references therein.
(12) Eppink, M. H.; Bunthol, C.; Schreuder, H. A.; van Berkel, W. J.
FEBS Lett. 1999, 443, 251. Eppink, M. H.; Schreuder, H. A.; van Berkel,
W. J. Eur. J. Biochem. 1995, 231, 157. van Berkel, W.; Westphal, A.;
Eppink, M.; de Kok, A. Eur. J. Biochem. 1992, 201, 411, and references
cited in these papers.
(13) Serjeant, E. P.; Dempsey, P. Ionization Constants of Organic Acids
in Aqueous Solution; Pergamon Press: Oxford, 1979.
(14) Anderson, R. F.; Patel, K. B.; Stratford, M. R. L. J. Chem. Soc.,
Faraday Trans. 1987, 83, 3177.
(15) Buxton, G. V.; Greenstock, C. L.; Helman, W. P.; Ross, A. B. J.
Phys. Chem. Ref. Data 1988, 17, 513.
(16) Tripathi, G. N. R. In Multichannel Image Detectors II; Talmi, Y.,
Ed.; ACS Symposium Series 236; American Chemical Society: Washington,
DC, 1983; p 171. Patterson, L. K.; Lilie, J. Int. J. Radiat. Phys. Chem.
1974, 6, 129. Janata, E.; Schuler, R. H. J. Phys. Chem. 1982, 86, 2078.
(17) Tripathi, G. N. R.; Su, Y. J. Am. Chem. Soc. 1996, 118, 2235.
(18) Tripathi, G. N. R., unpublished results. Calculated frequencies 1502,
1449, 1162, and 983 cm-1 match well with the experimental frequencies.
(19) Tripathi, G. N. R. Chem. Phys. Lett. 1992, 409, 199.
•
4-carboxyphenol as a model system. The OH adduct of a
phenolate anion, which is a key intermediate in the phenoxyl
radical formation, has been observed for the first time and its
acid-base equilibrium has been examined. The adduct radical
decays into phenoxyl radical, identified by time-resolved
resonance Raman spectroscopy.
Two important relationships that define the early chemical
•
events in the OH oxidation of phenols in water have been
developed. The first is concerned with the relationship between
the rate of phenoxyl radical formation (k) and base concentration
in solution, and the second is concerned with the dependence
of k on pKa of phenols. k ) ke × 10(pH-pK )/(1 + 10(pH-pKa)),
a
•
where ke is the spontaneous decay rate of the OH adduct of
the phenolate anion and pKa relates to its acid form. ke and pKa
in different phenolates relate by ke ) k0 exp(-0.059 × ∆pKa/
RT), where ∆pKa is estimated with respect to an aliphatic
reference system for which pKa ) 15. The value of k0 is
estimated as 1.25 × 1012 s-1, if pKa of the parent phenol is
used for determination of ke, and as 2.5 × 1012 if the pKa of the
•OH adduct is used. This relationship has been applied to phenols
for which values of ke are known and an excellent agreement
has been found.
Acknowledgment. The research described herein was sup-
ported by the Office of Basic Energy Sciences of the Department
of Energy. This is Contribution No. NDRL 4407 from the Notre
Dame Radiation Laboratory.
(20) Nonhebel, D. C.; Walton, J. C. Free-radical Chemistry: Structure
and Mechanism; Cambridge University Press: New York, 1974.
(21) Tripathi, G. N. R.; Schuler, R. H. Radiat. Phys. Chem. 1988, 32,
251. Tripathi, G. N. R.; Chipman, D. M. J. Phys. Chem. A 2002, 106, 8908.
(22) Alfassi, Z. B.; Schuler, R. H. J. Phys. Chem. 1985, 89, 3359.
References and Notes
(1) Lucarini, M.; Mugnaini, V.; Pedulli, G. F.; Guerra, M. J. J. Am.
Chem. Soc. 2003, 125, 8318.
•
•
(23) The yield of N3 is 10% higher than that of OH, which is within
the uncertainty of the yield measured by weak Raman signals.
(24) Neta, P.; Fessenden, R. W. J. Phys. Chem. 1974, 78, 523.
(25) Tripathi, G. N. R. J. Am. Chem Soc. 2003, 125, 1178, and references
therein.
(2) Barnett, N. W.; Hindson, B. J.; Lewis, S. W. Anal. Chim. Acta
1998, 362, 131. Nordlund, P.; Sjoberg, B.-M.; Eklund, H. Nature 1990,
345, 593. Bender, C. J.; Sahlin, M.; Babcock, G. T.; Barry, B. A.;
Chandrashekar, T. K.; Salowe, S. P.; Stubbe, J.; Lindstrom, B.; Petersson,
L.; Ehrenberg, A.; Sjoberg, B.-M. J. Am. Chem. Soc. 1989, 111, 2245.
Babcock, G. T. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 10893, and references
therein. Xie, C.; Lahti, P. M. Tetrahedron Lett. 1999, 40, 4305-4308.
(3) Barnett, N. W.; Bos, R.; Lewis, S. W.; Russell, R. A. Anal.
Commun. 1997, 34, 17.
(4) Halfen, J. A.; Jazdzewski, B. A.; Mahapatra, S.; Berreau, L. M.;
Wilkinson, E. C.; Que, L., Jr.; Tolman, W. B. J. Am. Chem. Soc. 1997,
119, 8217. Jazdzewski, B. A.; Tolman, W. B. Coord. Chem. ReV. 2000,
200, 633.
(5) von Sonntag, C. The Chemical Basis of Radiation Biology; Taylor
and Francis: New York, 1987.
(6) Land, E. J.; Ebert, M. Trans. Faraday Soc. 1967, 63, 1181.
Raghavan, N. V.; Steenken, S. J. Am. Chem. Soc. 1980, 102, 3495. Mvula,
E.; Schuchmann, M. N.; von Sonntag, C. J. Chem. Soc., Perkin Trans. 2
2001, 264.
(26) Bell, R. P. The Proton in Chemistry; Cornell University Press:
Ithaca, New York, 1959.
•
(27) In the case of the rate of reaction of OH with phenol , ke, the
formation of phenoxyl radical from phenolate anion would appear to occur
at a diffusion-controlled rate and no distinction can be made between the
•
base-catalyzed loss of water from the OH adduct of neutral phenol and
the loss of H+ and OH- in two successive steps from this species as
demonstrated here.
(28) Taniguchi, H.; Schuler, R. H. J. Phys. Chem. 1985, 89, 3095; J.
Am. Chem. Soc. 1984, 106, 1507.
(29) Pauling, L. The Nature of the Chemical Bond; Cornell University
Press: Ithaca, New York, 1960.
(30) The pKa of the reference system is taken as 15 for illustration, and
the value of k0 depends on this choice. However, the relationship between
•
(7) Steenken, S. Top. Curr. Chem. 1996, 177, 125. Steenken, S. J.
Chem. Soc., Faraday Trans., 1 1987, 83, 113, and references therein.
the rate of decay of the OH adduct of the phenolate anion and pKa of the
corresponding phenol is independent of this choice.