Flash photolytic generation of ortho-quinone methide in aqueous solution and study of its chemistry in that medium

Article abstract of DOI:10.1021/ja010826g

Flash photolysis of o-hydroxybenzyl alcohol, o-hydroxybenzyl p-cyanophenyl ether, and (o-hydroxybenzyl)trimethylammonium iodide in aqueous perchloric acid and sodium hydroxide solutions, and in acetic acid and biphosphate ion buffers, produced o-quinone methide as a short-lived transient species that underwent hydration back to benzyl alcohol in hydrogen-ion catalyzed (k_H⁺ = 8.4 × 10⁵ M^-1 s^-1) and hydroxideion catalyzed (k_HO^- 3.0 × 10⁴ M^-1 s^-1) reactions as well as an uncatalyzed (k_UC = 2.6 × 10² s^-1) process. The hydrogen-ion catalyzed reaction gave the solvent isotope effect k_H⁺/k_D⁺ = 0.42, whose inverse nature indicates that this process occurs by rapid and reversible equilibrium protonation of the carbonyl oxygen atom of the quinone methide, followed by rate-determining capture of the carbocation so produced by water. The magnitude of the rate constant of the uncatalyzed reaction, on the other hand, indicates that this process occurs by simple nucleophilic addition of water to the methylene group of the quinone methide. Decay of the quinone methide is also accelerated by acetic acid buffers through both acid- and base-catalyzed pathways, and quantitative analysis of the reaction products formed in these solutions shows that this acceleration is caused by nucleophilic reactions of acetate ion rather than by acetate ion assisted hydration. Bromide and thiocyanate ions also accelerate decay of the quinone methide through both hydrogen-ion catalyzed and uncatalyzed pathways, and the inverse nature of solvent isotope effects on the hydrogen-ion catalyzed reactions shows that these reactions also occur by rapid equilibrium protonation of the quinone methide carbonyl oxygen followed by rate-determining nucleophilic capture of the ensuing carbocation. Assignment of an encounter-controlled value to the rate constant for the rate-determining step of the thiocyanate reaction leads to pK_a = 1.7 for the acidity constant of the carbonyl-protonated quinone methide.

Full text of DOI:10.1021/ja010826g

J. Am. Chem. Soc. 2001, 123, 8089-8094
8089
Flash Photolytic Generation of ortho-Quinone Methide in Aqueous
Solution and Study of Its Chemistry in that Medium
Y. Chiang, A. J. Kresge,* and Y. Zhu
Contribution from the Department of Chemistry, UniVersity of Toronto,
Toronto, Ontario M5S 3H6, Canada
ReceiVed March 29, 2001. ReVised Manuscript ReceiVed June 15, 2001
Abstract: Flash photolysis of o-hydroxybenzyl alcohol, o-hydroxybenzyl p-cyanophenyl ether, and (o-
hydroxybenzyl)trimethylammonium iodide in aqueous perchloric acid and sodium hydroxide solutions, and in
acetic acid and biphosphate ion buffers, produced o-quinone methide as a short-lived transient species that
5
underwent hydration back to benzyl alcohol in hydrogen-ion catalyzed (kH ) 8.4 × 10 M-1 s-1) and hydroxide-
+
4
ion catalyzed (kHO ) 3.0 × 10 M-1 s-1) reactions as well as an uncatalyzed (kUC ) 2.6 × 102 s-1) process.
-
+
+
The hydrogen-ion catalyzed reaction gave the solvent isotope effect kH /kD ) 0.42, whose inverse nature
indicates that this process occurs by rapid and reversible equilibrium protonation of the carbonyl oxygen atom
of the quinone methide, followed by rate-determining capture of the carbocation so produced by water. The
magnitude of the rate constant of the uncatalyzed reaction, on the other hand, indicates that this process occurs
by simple nucleophilic addition of water to the methylene group of the quinone methide. Decay of the quinone
methide is also accelerated by acetic acid buffers through both acid- and base-catalyzed pathways, and
quantitative analysis of the reaction products formed in these solutions shows that this acceleration is caused
by nucleophilic reactions of acetate ion rather than by acetate ion assisted hydration. Bromide and thiocyanate
ions also accelerate decay of the quinone methide through both hydrogen-ion catalyzed and uncatalyzed
pathways, and the inverse nature of solvent isotope effects on the hydrogen-ion catalyzed reactions shows that
these reactions also occur by rapid equilibrium protonation of the quinone methide carbonyl oxygen followed
by rate-determining nucleophilic capture of the ensuing carbocation. Assignment of an encounter-controlled
value to the rate constant for the rate-determining step of the thiocyanate reaction leads to pKa ) -1.7 for the
acidity constant of the carbonyl-protonated quinone methide.
Quinone methides, such as the parent ortho isomer 1, are
molecules having a cyclohexadiene core with a carbonyl group
and a methylene unit attached. They are related to quinones,
such as 2, which have two carbonyl groups, and to quinone
dimethides, such as 3, which have two methylene units. Unlike
Scheme 1
cytotoxins responsible for the effects of such agents as antitumor
drugs, antibiotics, and DNA alkylators.2 In living systems where
water is the ubiquitous medium this biological activity must
operate against a background of wasteful quinone methide
hydration reactions. To provide information about these hydra-
tions, we have begun a detailed investigation of their kinetics
and reaction mechanisms. In this paper we describe the results
we have obtained studying the parent ortho isomer, 1, reacting
in aqueous solution.3
their relatives, however, quinone methides, with their two
different functional groups, are highly polarized and therefore
quite reactive. They may be viewed as resonance hybrids of
two principal canonical structures, one of which is dipolar
(Scheme 1). This gives these molecules both cationic and anionic
centers, and they therefore react with both nucleophiles and
electrophiles. Their reactions, moreover, often generate an
aromatic phenol ring, and this provides a considerable driving
force that adds to quinone methide reactivity. Quinone methides
have consequently become useful reaction intermediates with
wide applications in organic synthesis, and various methods for
their generation have been developed.1
We generated o-quinone methide by photoelimination of
water from o-hydroxybenzyl alcohol,1,4 4 X ) H, photoelimi-
nation of p-cyanophenol from o-hydrobenzyl p-cyanophenyl
ether, 4 X ) p-CNC6H4,5 and photoelimination of trimethy-
lammonium hydrogen iodide from (o-hydroxybenzyl)trimethy-
(2) See, e.g.: Peter, M. G. Angew. Chem., Int. Ed. Engl. 1989, 28, 555-
570. Bolton, J. L.; Pisha, E.; Zhang, F.; Qiu, S. Chem. Res. Toxicol. 1998,
11, 1113-1127. Pande, P.; Shearer, J.; Yang, J.; Greenberg, W. A.; Rokita,
S. E. J. Am. Chem. Soc. 1999, 121, 6773-6779.
(3) Part of this work has been published in preliminary form: Chiang,
Y.; Kresge, A. J.; Zhu, Y. J. Am. Chem. Soc. 2000, 122, 9854-9855.
(4) Diao, L.; Yang, C.; Wan, P. J. Am. Chem. Soc. 1995, 117, 5369-
5370.
Quinone methides also show pronounced biological activity.
They have, for example, been implicated as the ultimate
(1) For a brief review, see Wan, P.; Barker, B.; Diao, L.; Fischer, M.;
Shi, Y.; Yang, C. Can. J. Chem. 1996, 74, 465-475.
(5) McClelland, R. A. Tetrahedron 1996, 52, 6823-6858.
10.1021/ja010826g CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/28/2001

8090 J. Am. Chem. Soc., Vol. 123, No. 33, 2001
Chiang et al.
M for the alcohol and ca.. 4 × 10-5 M for the ammonium iodide. The
temperature of all reacting solutions was controlled at 25.0 ( 0.05 °C.
Reactions were monitored by following the decay of quinone methide
absorbance at λ ) 400 nm, and observed first-order rate constants were
obtained by least-squares fitting of an exponential function.
lammonium iodide, 4 OX ) (CH3)3N+I-,6 eq 1. Flash photolysis
Results
of these substrates in aqueous solution produced a transient
species with the strong absorbance at λ ) 400 nm that is
characteristic of o-quinone methide.7 Neither the lifetime of this
transient species nor the amount produced was affected by the
presence or absence of dissolved oxygen, which again is as
expected for this transient being o-quinone methide.7
Product Analysis. Analysis of spent reaction mixtures
showed that only o-hydroxybenzyl alcohol was produced from
o-quinone methide generated by flash photolysis of o-hydroxy-
benzyl alcohol in perchloric acid solution, as expected for clean
hydration of the quinone methide back to starting material. When
o-hydroxybenzyl p-cyanophenyl ether was the photolysis sub-
strate, on the other hand, a minor amount (ca. 10%) of an
additional unidentified substance was formed. In acetic acid
buffers, these products were accompanied by formation of
o-hydroxybenzyl acetate.
Because of the unknown additional product formation from
o-hydroxybenzyl p-cyanophenyl ether, this substrate was not
used for rate measurements. It was, however, employed for
quantitative product analysis, where a substrate other than
o-hydroxybenzyl alcohol was needed in order to distinguish the
hydration product from unphotolyzed starting material.
Quantitative product analyses were carried out in acetic acid
buffers using a series of solutions of fixed buffer ratio ([HOAc]/
[OAc-] ) 0.10) and constant ionic strength (0.10 M). The data
so obtained are summarized in Table S1.13
Kinetics, Perchloric Acid, Sodium Hydroxide, and Buffer
Solutions. Rates of decay of o-quinone methide were measured
in dilute perchloric acid solution over the concentration range
[HClO4] 0.001-0.1 M and also in acetic acid and biphosphate
ion buffers, using both H2O and D2O as the solvent. The ionic
strength of these solutions was kept constant at 0.10 M by
supplying sodium perchlorate as required. These results are
summarized in Tables S2 and S3.13
The rate measurements in buffers were made in series of
solutions of fixed buffer ratio, and therefore fixed hydrogen ion
concentration, but varying buffer concentration. Extrapolation
of these data to zero buffer concentration using the buffer
dilution expression of eq 2
Experimental Section
Materials. o-hydroxybenzyl acetate was prepared by treating
o-hydroxybenzyl alcohol with acetic anhydride in the presence of boron
trifluoride etherate,8 and (o-hydroxybenzyl)trimethylethylammonium
iodide was synthesized by methyl iodide methylation of (o-hydroxy-
benzyl)dimethylamine,6 itself obtained by the reaction of potassium
phenolate with N,N-dimethylmethyleneammonium iodide.9
o-Hydroxybenzyl p-cyanophenyl ether was prepared in a one-pot
process by first converting o-hydroxybenzyl alcohol to o-hydroxybenzyl
bromide, using the mild conditions of the Ph3P/CBr4 reagent,10 and
then immediately adding sodium p-cyanophenoxide. A flask containing
an argon atmosphere was charged with 1.00 g o-hydroxybenzyl alcohol,
2.33 g triphenylphosphine, 2.95 g carbon tetrabromide, and 10 mL
DMSO. This mixture was stirred at room temperature for 10 min, and
1.25 g sodium p-cyanophenoxide was then added, and that mixture
was stirred overnight. The following day, 30 mL of water was added,
the resulting mixture was extracted with two 75-mL portions of ethyl
ether, and the ether extracts were washed with saturated brine and were
dried over sodium sulfate. The residue remaining after removal of the
ether was purified by column chromatography (silical gel with hexane
eluent) to give 0.50 g (28%) of colorless crystals: mp 135-136 °C;
1H NMR (200 MHz, CDCl3): δ/ppm ) 7.59 (d, J ) 8.9 Hz, 2H),
7.32-7.25 (m, 2H), 7.07 (d, J ) 9.0 Hz, 2H), 6.99-6.86 (m, 2H),
5.81 (s, 1H), 5.22 (s, 2H); 13C NMR (50 MHz, CDCl3): δ/ppm )
162.02, 154.80, 134.49, 130.52, 129.72, 122.06, 121.43, 119.44, 116.73,
116.22, 105.17, 67.84; HRMS: m/e ) 225.0790 (calc), 225.0794 (obs).
All other materials were best-available commercial grades.
Product Analysis. Product analyses were conducted by HPLC using
a Varian Vista 5500 instrument with a NovoPak C18 reverse-phase
column and methanol-water (60/40 ) v/v) as the eluent. Reaction
solutions, containing ca.. 2 × 10-5 M substrate and ca.. 4 × 10-5
M
kobs ) kint + kcat [Buffer]
(2)
anisole, which served as an internal standard, were subjected to a single
pulse from our microsecond flash photolysis system.11 Products were
identified by comparing retention times and UV spectra with authentic
samples. The temperature of all reactions was controlled at 25.0 ( 0.1
°C, and parallel analyses of unphotolyzed reaction mixtures showed
that no thermal transformations occurred during the time of the
photolysis experiments.
gave intercepts, kint, which, together with the dilute perchloric
acid solution data, were used to construct the rate profiles shown
in Figure 1. Hydrogen-ion concentrations of the buffer solutions
needed for this purpose were obtained by calculation, using
thermodynamic acidity constants from the literature and activity
coefficients recommended by Bates.14
Kinetics. Rate measurements were made using microsecond11 and
nanosecond12 (λexc ) 248 nm) flash photolysis systems that have already
been described.11,12 o-Hydroxybenzyl alcohol was the photolysis
substrate for all rate measurements except those made in sodium
hydroxide solution, where (o-hydroxybenzyl)trimethylammonium iodide
was used instead. Initial substrate concentrations were ca.. 2 × 10-4
Figure 1 shows that o-quinone methide undergoes both acid-
catalyzed and uncatalyzed reactions. The data were therefore
analyzed using the rate law of eq 3. Least-squares fitting
provided the results kH ) (8.41 ( 0.12) × 105 M-1 s-1, kH
/
,
+
+
+
kD ) 0.418 ( 0.017, (kUC H O
)
) (2.60 ( 0.06) × 102 s-1
2
(k
)
/(k )
UC D2O
) 1.42 ( 0.06.
(6) Modica, E.; Zanaletti, R.; Freccero, M.; Mella, M. J. Org. Chem.
2001, 66, 41-52.
UC H2O
(7) Barker, B.; Diao, L.; Wan, P. J. Photochem. Photobiol. A 1997, 104,
91-96.
kobs ) kUC + kL+ [L+]
(3)
(8) Cottet, F.; Cottier, L.; Descote, G. Can. J. Chem. 1990, 68, 1251-
1257.
6
These results are consistent with kH ) 1.4 × 10 M-1 s-1 and
+
(9) Pochini, A.; Puglia, G.; Ungaro, R. Synthesis 1983, 906-907.
(10) Wagner, A.; Heitz, M.-P.; Mioskowski, C. Tetrahedron Lett. 1989,
30, 557-558. Appel, R. Angew. Chem., Int. Ed. Engl. 1975, 14, 801-811.
(11) Chiang, Y.; Hojatti, M.; Keeffe, J. R.; Kresge, A. J.; Schepp, N.P.;
Wirz, J. J. Am. Chem. Soc. 1987, 109, 4000-4009.
(12) Andraos, J.; Chiang, Y.; Huang, C. G.; Kresge, A. J.; Scaiano, J.
C. J. Am. Chem. Soc. 1993, 115, 10605-10610.
kUC ) 3.2 × 102 s-1 recently reported for these reactions.6
(13) Supporting Information; see paragraph at the end of this paper
regarding availability.
(14) Bates, R. G. Determination of pH Theory and Practice; Wiley: New
York, 1973; p 49.

Flash Photolytic Generation of o-Quinone Methide
J. Am. Chem. Soc., Vol. 123, No. 33, 2001 8091
Figure 1. Rate profiles for the hydration of o-quinone methide in H2O
(O) and D2O (∆) solution at 25 °C.
Figure 3. Relationship between observed rate constants and bromide
ion concentration for the reaction of o-quinone methide in aqueous
solution (H2O) containing 0.020 M perchloric acid at 25 °C.
these solutions no longer gave good first-order kinetics. We
believe that this was so because first-order hydration was
accompanied by an additional second-order reaction of the
quinone methide with some unphotolyzed o-hydroxybenzyl
alcohol substrate, which was present in these solutions in its
highly nucleophilic phenoxide-ion form. Such an additional
reaction would produce oligomeric products, whose formation
from o-quinone methide in basic solutions has been observed.15
We found, however, that use of (o-hydroxybenzyl)trimethy-
lammonium iodide as the photolysis substrate in sodium
hydroxide solutions gave good first-order kinetic behavior. Rates
of this reaction were therefore measured over the concentration
range [NaOH] ) 0.001-0.1 M at a constant ionic strength of
0.10 M (NaClO4). The data so obtained are summarized in Table
S5.13
Figure 2. Separation of buffer catalysis into general acid and general
base components for the reaction of o-quinone methide in aqueous
(H2O) acetic acid buffer solutions at 25 °C.
Observed first-order rate constants determined in these sodium
hydroxide solutions increased linearly with increasing sodium
hydroxide concentration, and linear least-squares analysis gave
the hydroxide-ion rate constant kHO- ) (2.95 ( 0.04) × 104
M-1 s-1. This result is consistent with kHO- ) 6.3 × 104 M-1
s-1 recently reported for this reaction.6
The decay of o-quinone methide in acetic acid buffers showed
pronounced buffer catalysis, and buffer catalytic coefficients,
kcat eq 2, were well determined. These catalytic coefficients were
Kinetics, Bromide and Thiocyanate Ion Solutions. The
decay of o-quinone methide in aqueous solution was strongly
accelerated by bromide and thiocyanate ions, in keeping with
the proclivity of quinone methides to react with nucleophiles.
These accelerations were also catalyzed by acids. As Figure 3
illustrates, the rate of the bromide ion reaction at a fixed acidity
increased linearly with bromide ion concentration, and Figure
4 shows that second-order bromide-ion rate constants, obtained
as slopes of plots such as that of Figure 3, themselves increased
linearly with hydrogen ion concentration. Similar behavior was
observed with thiocyanate ion.
For both ions, rate measurements were consequently made
in series of solutions of constant hydrogen ion concentration
but varying nucleophile concentration, using both H2O and D2O
as the solvent. Perchloric acid or biphosphate ion buffers were
employed to keep hydrogen ion concentrations constant, and
the ionic strength was maintained at 0.10 M by adding sodium
perchlorate as required. The results are summarized in Tables
S6-S9.13
separated into their general acid, kHA, and general base, kB,
components with the aid of the relationship shown in eq 4,
kcat ) kB + (kHA - kB)fA
(4)
where fA is the fraction of buffer present in the acid form. Figure
2 shows that the data obtained conform to this relationship well.
Least-squares analysis gave kHA ) (1.25 ( 0.29) × 102 M-1
s-1, kHA/kDA ) 1.35 ( 0.83, kB ) (1.33 ( 0.02) × 103 M-1
s-1, (kB)H O/(kB)D O ) 1.17 ( 0.06.
2
2
In biphosphate ion buffers, on the other hand, buffer catalysis
was weak and reliable values of kcat in these solutions were not
obtained. The zero-concentration intercepts, kint, for these
buffers, however, were well-determined.
Some rate measurements of o-quinone methide decay were
also made in concentrated perchloric acid solutions over the
concentration range [HClO4] ) 0.2-4.5 M. No additional inert
electrolyte was added to these solutions, and their ionic strength
was therefore equal to perchloric acid concentration. These data
are summarized in Table S4.13
Least-squares analysis of the data obtained in perchloric acid
solutions gave k ) (2.10 ( 0.05) × 107 M-2 s-1 and kH O/kD O
2
2
Rate measurements were also attempted in sodium hydroxide
solutions using o-hydroxybenzyl alcohol as the photolysis
substrate, but the decay of o-quinone methide absorbance in
(15) Wan, P.; Hennig, D. J. Chem. Soc., Chem. Commun. 1987, 939-
941.

8092 J. Am. Chem. Soc., Vol. 123, No. 33, 2001
Chiang et al.
effect stems from the fact that positively charged O-H bonds,
such as those in the hydronium ion, are looser than uncharged
O-H bonds, such as those in a water molecule.17 Conversion
of H3O+ into H2O in the equilibrium step of eq 5 then leads to
a tightening of the hydrogenic environment of the species
involved, and that produces an inverse isotope effect.
At sufficiently high acidities, the position of the equilibrium
preceding the rate-determining step of the mechanism of eq 5
will shift from unprotonated to protonated substrate, and acid
catalysis of the quinone methide hydration reaction will become
saturated. Figure 5 shows that such saturation was not achieved
up to perchloric acid concentrations of 4.5 M, where rates
reached the fast-reaction limit of our flash photolysis system.
In concentrated acids, in fact, hydration rates, rather than
saturating, increased more rapidly than in proportion to acid
concentration. This is often the case in concentrated acid
solutions, where reactions frequently follow acidity functions
rather than acid concentration. Although saturation of acid
catalysis was not achieved in this case, it was realized in related
studies with other more basic quinone methide substrates and
will be reported separately.
The isotope effect on the uncatalyzed o-quinone methide
hydration reaction is less mechanistically diagnostic. It does not
serve to distinguish between an analogue of the acid-catalyzed
reaction mechanism, involving rapid equilibrium substrate
protonation by proton transfer from a water molecule followed
by cation capture by the hydroxide ion so formed, eq 6, or
simple nucleophilic attack of water on the quinone methide
methylene group, with or without simultaneous proton transfer
to avoid a zwitterionic intermediate, eq 7. The first of these
Figure 4. Relationship between bromide-ion rate constants and
perchloric acid concentration for the reaction of o-quinone methide in
H2O (O) and D2O (∆) solution at 25 °C.
) 0.495 ( 0.010 for the acid-catalyzed bromide-ion reaction
and k ) (9.56 ( 0.76) × 107 M-2 s-1 and kH O/kD O ) 0.607 (
2
2
0.059 for its thiocyanate-ion counterpart.
As Figure 4 illustrates, the zero-hydrogen-ionic concentration
intercepts of the (∆kobs/∆[Nuc]) versus [L+] plots for bromide
ion were essentially zero, and a bromide-ion reaction not
catalyzed by perchloric acid could not be detected. Such an
uncatalyzed reaction could also not be seen in the much less
acidic biphosphate buffers: observed rate constants determined
in these solutions did not increase with increasing bromide ion
concentration but rather varied randomly about an average value
consistent with the rate of a reaction in biphosphate buffers with
no added bromide ion. Although an uncatalyzed bromide-ion
reaction could not be observed, an upper limit for its rate
constant could be estimated by assuming that a 10% bromide-
ion-induced increase in rate in the biphosphate-ion buffers would
have been detected. The result is kBr- < (3.25) × 102 M-1 s-1
.
In contrast to this behavior of bromide ion, the zero-hydrogen-
ion concentration intercepts of the (∆kobs/∆[Nuc]) versus [L+]
plots for thiocyanate ion were finite and well-determined, giving
k ) (1.24 ( 0.05) × 107 M-1 s-1 and kH O/kD O ) 1.20 ( 0.07
2
2
for the not acid-catalyzed portion of the thiocyanate reaction.
This uncatalyzed reaction was also apparent in biphosphate ion
buffers, where observed rate constants increased linearly with
increasing thiocyanate ion concentration. Least-squares analysis
gave k ) (1.28 ( 0.04) × 107 M-1 s-1 and kH O/kD O ) 1.08 (
uncatalyzed reaction mechanisms can be ruled out, however,
on the basis of the fact that it would require an impossibly large
value of the rate constant for its slow step, k. The rate law that
applies to the scheme of eq 6 is kUC ) k K, with K, the
equilibrium constant of the first step, equal to the ionization
constant of water, Kw, divided by the acidity constant of the
protonated substrate, KSH. The value of KSH is not known, but
the fact that the rate profile of Figure 5 shows no saturation of
acid catalysis up to the highest acid concentration used, [H+]
) 4.5 M, indicates that KSH must be greater than this, that is,
KSH > 4.5 M. The observed value of kUC is 2.6 × 102 s-1, and
combination of that with Kw ) 10-14 M-2 and KSH > 4.5 M
then leads to k > 1.2 × 1017 M-1 s-1, which exceeds the
encounter-controlled limit of k = 1 × 1010 M-1 s-1 by a
considerable margin.
2
2
0.04, in good agreement with the determination of these
quantities as the [L+] plot intercepts.
Discussion
Mechanisms of the Hydration Reactions. The rate profiles
of Figure 1 pertain to decay of o-quinone methide through
reaction with solvent-related species, where product analysis
shows that hydration to o-hydroxybenzyl alcohol is the only
process occurring. The inverse isotope effect on the acid-
+
+
catalyzed portion of this process, kH /kD ) 0.42, implies that
this reaction occurs by a preequilibrium substrate-protonation
mechanism,16 which in this case may be formulated as rapid
equilibrium protonation of the quinone methide on its carbonyl
oxygen atom followed by rate-determining capture of the cation
so formed by water, eq 5. The inverse nature of this isotope
Mechanisms of the Reactions Promoted by Acetic Acid
Buffers. Rates of decay of o-quinone methide were appreciably
accelerated by acetic acid buffers, with a somewhat stronger
acceleration by the buffer base than by the buffer acid (cf Figure
2). o-Hydroxybenzyl acetate was formed in these solutions as
(16) Keeffe, J.R.; Kresge, A. J. In InVestigation of Rates and Mechanisms
of Reactions; Bernasconi, C. F., Ed.; Wiley: New York, 1986; pp 747-
790.
(17) Kresge, A. J.; More O’Ferrall, R. A.; Powell, M. F. In Isotopes in
Organic Chemistry; Buncel, E.; Lee, C. C., Eds.; Elsevier: New York, 1987;
Chapter 4.

Flash Photolytic Generation of o-Quinone Methide
J. Am. Chem. Soc., Vol. 123, No. 33, 2001 8093
Figure 5. Comparison of rates of hydration of o-quinone methide in
concentrated (∆) and dilute (O) aqueous perchloric acid solution at 25
°C.
Figure 6. Ratio of products formed by reaction of o-quinone methide
in aqueous (H2O) acetic acid buffers, [HOAc]/[OAc-] ) 0.10, at 25
°C plotted according to eq 12.
well, and this suggests that the buffer reactions occur by
mechanisms that are analogues of those established for o-
quinone methide hydration, eqs 5 and 7, with acetate ion rather
than water playing the nucleophilic role. The acid-catalyzed
buffer reaction would then involve rapid equilibrium protonation
of the quinone methide on its carbonyl oxygen atom, followed
by rate-determining nucleophilic capture of the ensuing cation
by acetate ion, eq 8, whereas the base-catalyzed buffer reaction
would consist of simple rate-determining nucleophilic attack
of acetate ion on the methide group of unprotonated o-quinone
methide, eq 9.
where ratios of o-hydroxybenzylacetate to o-hydroxybenzyl
alcohol concentrations, [BzOAc]/[BzOH], are related to buffer
concentration through the parameter ω, which is the fraction
of the buffer catalyzed reaction that gives acetate product, plus
the term kint/kcat, which is the ratio of buffer-uncatalyzed to
buffer-catalyzed rate constants at the buffer ratio under exami-
nation (see eq 2). Figure 6 shows that the experimental data
conform to this relationship well. Least-squares analysis of two
separate series of determinations gave results whose weighted
averages are ω ) 1.00 ( 0.02 and kint/kcat ) 0.198 (0.007 M;
the latter agrees well with kint/kcat ) 0.214 ( 0.006 derived from
the buffer dilution kinetics plot at [HOAc]/[OAc-] ) 0.10.
The unit value of ω obtained in this way indicates that all of
the buffer-base-accelerated decay of o-quinone methide occurs
by the nucleophilic pathway of eq 9 and that none of this
acceleration take place by the general base-assisted hydration
reaction of eq 11. This result also makes it quite likely that the
buffer-acid-accelerated decay is a wholly nucleophilic reaction
as well and that the preequilibrium protonation followed by a
general base-assisted hydration scheme of eq 10 is not operating
at all. This is because such general base assistance is known to
become progressively less important as the species undergoing
reaction becomes more electrophilic,18 and of course the benzyl
cation intermediate of eq 10 is much more electrophilic than
the neutral quinone methide of eq 11.
These reaction mechanisms, however, might also be ac-
companied by additional mechanisms in which acetate ion,
rather than serving as a nucleophile, operates as a general base
assisting attack of water on the protonated substrate, eq 10, or
on the unprotonated substrate, eq 11.
This assignment of reaction mechanisms is supported by the
solvent isotope effects determined for these buffer-assisted
reactions. The rate-determining step of the buffer-base nucleo-
philic pathway of eq 9 does not involve proton transfer, and
the effect of changing H2O to D2O on the reorganization of the
solvent shells of the species involved is expected to be small.17
The solvent isotope effect on such a mechanism should therefore
The products of these two sets of reactions are different: those
of eqs 8 and 9 give o-hydroxybenzyl acetate, whereas those of
eqs 10 and 11 give o-hydroxybenzyl alcohol. The incursion or
absence of the additional reaction mechanisms of eqs 10 and
11 may consequently be determined by performing a quantitative
product analysis. To avoid difficulties introduced by the
simultaneous operation of both buffer acid and buffer base
catalysis, we carried out this analysis using buffers with a buffer
ratio of [HOAc]/[OAc-] ) 0.10; in these very basic buffers,
buffer acid catalysis makes a negligibly small rate contribution
and essentially all of the buffer reaction occurs by the basic
route. The product data were analyzed using eq 12,
be small, consistent with the near-unit value observed: kH O
/
2
kD O ) 1.17. The buffer-acid nucleophilic pathway of eq 8
2
does require proton transfer in its preequilibrium step, but the
solvent isotope effect here is again expected to be close to unity
because the equilibrium constant for this step consists of the
ratio of acidity constants of acetic acid and protonated quinone
methide, whose isotope effects will largely offset one another.17
Reorganization of the solvent shells in the rate-determining step
of this reaction will again contribute only a weak effect, and
the overall isotope effect will once more be expected to be
[BzOAc]
[BzOH]
ω[Buffer]
(1 - ω)[Buffer] + kint/kcat
(18) Ta-Shma, R.; Jencks, W. P. J. Am. Chem. Soc. 1986, 108, 8040-
8050. Richard, J. P.; Toteva, M. M.; Crugeiras, J. J. Am. Chem. Soc. 2000,
122, 1664-1674.
)
(12)

8094 J. Am. Chem. Soc., Vol. 123, No. 33, 2001
Chiang et al.
close to unity, consistent with the value observed, kHA/kDA
)
ion, as measured by both the Swain-Scott20 and Pearson21
parameters, the reaction of o-hydroxybenzyl cation with thio-
cyanate ion should be an encounter-controlled process as well.
On this basis, an encounter-controlled rate constant of kNuc
) 5 × 109 M-1 s-1 may be assigned to the rate-determining
step of the reaction of eq 13 for the case Nuc ) SCN-.
Combination of that with the observed rate constant kNuc/KSH
) 9.6 × 107 M-2 s-1 then leads to the estimate KSH ) 52 M,
pKSH ) -1.7, for the acidity constant of oxygen-protonated
o-quinone methide.
1.35. The buffer-assisted hydration pathways of eqs 10 and 11,
on the other hand, do involve proton transfer, and these reactions
may be expected to show solvent isotope effects significantly
greater than unity.
Mechanisms of the Bromide and Thiocyanate Ion Reac-
tions. Both bromide ion and thiocyanate ion accelerate the decay
of o-quinone methide strongly. Both accelerations, moreover,
show acid-catalysis in perchloric acid solutions, and both give
+
+
inverse hydrogen-ion isotope effects: kH /kD ) 0.49 for
+
+
bromide ion and kH /kD ) 0.61 for thiocyanate ion. The inverse
nature of these isotope effects implies that these accelerations
occur through preequilibrium substrate protonation, just like
acid-catalyzed o-quinone methide hydration (eq 5), and an
analogous reaction mechanism may be assigned, with the
stronger nucleophiles Br- and SCN- replacing water in the rate-
determining cation-capture step, eq 13.
This result may in turn be combined with the rate constant
+
for acid-catalyzed hydration of o-quinone methide, kH ) kw/
KSH ) 8.4 × 105 M-1 s-1 to give kw ) 4 × 107 s-1 for the
capture of the o-hydroxybenzyl cation by water in wholly
aqueous solution. This estimate is consistent with kw ) 2 ×
108 s-1 determined for reaction of the p-methoxybenzyl cation,
7, with water in 50:50 trifluroethanol:water.19b The value of KSH
itself, however, is considerably less than acidity constants of
oxygen-protonated simple carbonyl compounds,22 but ionization
of oxygen-protonated o-quinone methide converts an aromatic
ring into a quinone methide structure with a concomitant loss
of some aromatic stabilization, and the weaker acidity of the
quinone methide conjugate acid is understandable on that basis;
an even lower acidity constant has recently been found for an
oxygen-protonated quinone methide in a naphthoquinone sys-
tem.23
The thiocyanate ion also accelerated o-quinone methide decay
by a process that was not acid-catalyzed. This reaction gave
the near-unit solvent isotope effect kH O/kD O ) 1.14, which is
2
2
consistent with simple nucleophilic attack of SCN- on the
methylene group of unprotonated o-quinone methide in a process
analogous to the uncatalyzed hydration reaction (eq 7). A similar
bromide-ion-promoted acceleration not catalyzed by acid could
not be detected.
Acknowledgment. We are grateful to the Natural Sciences
and Engineering Research Council of Canada for financial
support of this work.
It seems likely that the reaction of thiocyanate ion with the
o-hydroxybenzyl cation, 5, by the route of eq 13 is an encounter-
controlled process. This cation should be less stable and
therefore more reactive than the p-methoxyphenylethyl cation,
6, and of similar stability and reactivity as the p-methoxybenzyl
cation, 7, both of which react with azide ion in a 50:50
Supporting Information Available: Tables of product
yields and rate data (PDF). This material is available free of
charge via the Internet at http://pubs.acs.org.
JA010826G
(19) (a) Richard, J. P.; Jencks, W. P. J. Am. Chem. Soc. 1982, 104, 4689-
4691. Richard J. P.; Rothenberg, M. E.; Jencks, W. P. J. Am. Chem. Soc.
1984, 106, 1361-1372. (b) Amyes, T. L.; Richard, J. P. J. Am. Chem. Soc.
1990, 112, 9507-9512.
(20) Swain, C. G.; Scott, C. B. J. Am. Chem. Soc. 1953, 75, 141-147.
(21) Pearson, R. G. In AdVances in Linear Free Energy Relationships;
Chapman, N. B., Shorter, J., Eds.; Plenum: New York, 1972; Chapter 6.
(22) Bagno, A.; Scorrano, G.; More O’Ferrall, R. A. ReV. Chem.
Intermed. 1987, 7, 313-352.
(23) Chiang, Y.; Kresge, A. J.; Hellrung, B.; Schu¨nemann, P.; Wirz, J.
HelV. Chim. Acta 1997, 80, 1106-1121.
trifluoroethanol-water solvent at the encounter-controlled
limit.19 Since thiocyanate ion is a better nucleophile than azide