New insights on the mechanisms of the pH-independent reactions of benzo[a]pyrene 7,8-diol 9,10-epoxides

Article abstract of DOI:10.1021/ja004359z

The rates and products of the reactions of (±)-7β,8α-dihydroxy-9β,10β-epoxy-7,8,9, 10-tetrahydrobenzo[α]pyrene alpyrene (1) and (±)-7β,8α-dihydroxy-9α,10α-epoxy-7,8,9, 10-tetrahydrobenzo[α]pyrene (2) in water and dioxane-water mixtures have been determined over a pH range wider than that of earlier studies. This study provides additional insight on the mechanisms of the pH-independent reactions of 1 and 2. The rate profile for reaction of 1 shows acid-catalyzed hydrolysis at pH <5, a rate plateau at pH 5-9.5, a negative inflection at pH 10-11.5, and a rate increase at pH > 11.5. The rate decrease between pH 10 and pH 11.5 is accompanied by a decrease in the yield of tetrols from 60% (pH 8) to 29% (pH 11.2) and is interpreted to be the result of a partial change in mechanism brought about by attack of hydroxide ion acting as a base to deprotonate a carbocation intermediate and regenerate 1 at pH < 10, thus reducing the contribution of the pathway for tetrol formation in which water attacks the carbocation. The rate profile for the reaction of 2 exhibits only a single rate plateau at intermediate pH, along with increases in rate at low and high pH because of second-order reactions of 2 with H⁺ and HO^-, respectively. The lack of a rate depression at pH < 10 and the product studies for the reaction of 2 in dilute sodium azide solutions suggest that the tetrol-forming reactions of the pH-independent reaction of 2 are concerted or near-concerted.

Full text of DOI:10.1021/ja004359z

J. Am. Chem. Soc. 2001, 123, 6785-6791
6785
New Insights on the Mechanisms of the pH-Independent Reactions of
Benzo[a]pyrene 7,8-Diol 9,10-Epoxides
Lanxuan Doan,† Bin Lin,† Haruhiko Yagi,‡ Donald M. Jerina,‡ and Dale L. Whalen*,†
Contribution from the Department of Chemistry and Biochemistry, UniVersity of Maryland, Baltimore
County, Baltimore, Maryland 21250, and the Laboratory of Bioorganic Chemistry, NIDDK, National
Institutes of Health, Bethesda, Maryland 20892
ReceiVed December 27, 2000
Abstract: The rates and products of the reactions of (()-7â,8R-dihydroxy-9â,10â-epoxy-7,8,9,10-tetrahy-
drobenzo[a]pyrene (1) and (()-7â,8R-dihydroxy-9R,10R-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (2) in water
and dioxane-water mixtures have been determined over a pH range wider than that of earlier studies. This
study provides additional insight on the mechanisms of the pH-independent reactions of 1 and 2. The rate
profile for reaction of 1 shows acid-catalyzed hydrolysis at pH <5, a rate plateau at pH 5-9.5, a negative
inflection at pH 10-11.5, and a rate increase at pH >11.5. The rate decrease between pH 10 and pH 11.5 is
accompanied by a decrease in the yield of tetrols from 60% (pH 8) to 29% (pH 11.2) and is interpreted to be
the result of a partial change in mechanism brought about by attack of hydroxide ion acting as a base to
deprotonate a carbocation intermediate and regenerate 1 at pH >10, thus reducing the contribution of the
pathway for tetrol formation in which water attacks the carbocation. The rate profile for the reaction of 2
exhibits only a single rate plateau at intermediate pH, along with increases in rate at low and high pH because
of second-order reactions of 2 with H+ and HO-, respectively. The lack of a rate depression at pH >10 and
the product studies for the reaction of 2 in dilute sodium azide solutions suggest that the tetrol-forming reactions
of the pH-independent reaction of 2 are concerted or near-concerted.
Introduction
Scheme 1
Rate profiles1 and product studies for the acid-catalyzed1-4
and pH-independent1 hydrolysis reactions of 7,8-dihydroxy-9,-
10-epoxide metabolites 1 and 2 of the carcinogen benzo[a]-
Scheme 2
pyrene have been reported. Acid-catalyzed hydrolyses of 1 and
2 proceed by rate-limiting formation of benzylic carbocations,
which undergo attack by water from both faces of the electron-
deficient benzylic carbon to yield mixtures of cis and trans
tetrols, e.g., see Scheme 1 for the reaction of 1. Whereas acid-
catalyzed hydrolysis of 1 yields 92% of cis tetrol and only 8%
of trans tetrol, acid-catalyzed hydrolysis of 2 yields 95% of trans
tetrol and only 5% of cis tetrol. The very different ratios of cis
and trans tetrols formed from the hydrolyses of 1 and 2 have
been rationalized by mechanisms involving favorable axial
attack of water on the more stable conformation of the
carbocation.5,6
The pH-independent reaction of 1 at pH 8 yields cis and trans
tetrols 4 and 5, formed in the same ratio as that formed from
the acid-catalyzed hydrolysis of 1 and a third product, ketone 6
(Scheme 2).1 An intermediate in the tetrol-forming reaction has
been detected by nucleophile trapping with azide anion and
N-acetylcysteine.7 This intermediate is assumed to be identical
to the benzylic carbocation formed in the acid-catalyzed reaction
† University of Maryland, Baltimore County.
‡ National Institutes of Health.
(1) Whalen, D. L.; Montemarano, J. A.; Thakker, D. R.; Yagi, H.; Jerina,
D. M. J. Am. Chem. Soc. 1977, 99, 5522-5524.
(2) Wood, A. W.; Wislocki, P. G.; Chang, R. L.; Levin, W.; Lu, A. Y.
H.; Yagi, H.; Hernandez, O.; Jerina, D. M.; Conney, A. H. Cancer Res.
1976, 36, 3358-3366.
(5) Gillilan, R. E.; Pohl, T. M.; Whalen, D. L. J. Am. Chem. Soc. 1982,
104, 4481-4482.
(6) Sayer, J. M.; Yagi, H.; Silverton, J. V.; Friedman, S. L.; Whalen, D.
L.; Jerina, D. M. J. Am. Chem. Soc. 1972, 104, 1972-1978.
(7) Islam, N. B.; Gupta, S. C.; Yagi, H.; Jerina, D. M.; Whalen, D. L. J.
Am. Chem. Soc. 1990, 112, 6363-6369.
(3) Yang, S. K.; McCourt, D. W.; Roller, P. P.; Gelboin, H. V. Proc.
Natl. Acad. Sci. U.S.A. 1976, 73, 2594-2598.
(4) Yagi, H.; Thakker, D. R.; Hernandez, O.; Koreeda, M.; Jerina, D.
M. J. Am. Chem. Soc. 1977, 99, 1604-1611.
10.1021/ja004359z CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/20/2001

6786 J. Am. Chem. Soc., Vol. 123, No. 28, 2001
Doan et al.
Table 1. Summary of Tetrol Yields from the Reaction of 1 in
because the cis to trans tetrol ratio in both acid-catalyzed and
pH-independent reactions is the same.
a
Water and Dioxane-Water Solutions, 0.1 M NaClO4
cis-tetrol: yield of tetrols
trans-tetrol
To ascertain the mechanisms of the pH-independent reactions
of 1 and 2, often referred to as “spontaneous” reactions, we
have carried out rate and product studies of their reactions over
a pH range greater than that for which studies were previously
reported. These new studies show that, in addition to undergoing
a change in mechanism from an acid-catalyzed process to a pH-
independent reaction at pH 5-6, the pH-independent reaction
of 1 undergoes a further change of mechanism at pH 9.5-10.5.
The pH-independent reaction of 2 does not show a similar
change in mechanism. At pH >12, the rates of reaction of both
1 and 2 increase due to second-order reactions with hydroxide
ion. In this paper, experiments that provide more insight on the
mechanisms of the pH-independent reactions of 1 and 2 are
reported and discussed.
pHb
solvent
(%)c
3.00-3.15
7.87-7.89
7.94-7.89
8.04-8.02
10:90 dioxane-water
water
5:95 dioxane-water
10:90 dioxane-water
89:11
91:9
91:9
100
60
60
56
29
29
29
16
15
91:9
11.13-11.04 water
88:12
88:12
88:12
75:25
68:32
11.21-11.14 5:95 dioxane-water
11.25-11.22 10:90 dioxane-water
12.00-11.99 water
12.07-12.07 10:90 dioxane-water
a Reactions were allowed to proceed at rt for 20-25 min. b pH values
are given for each solution before the addition of 1 and at the end of
the reaction. c The percent yields of tetrols from the reaction of 1 are
relative to the yield of tetrols formed from the acid-catalyzed hydrolysis
of 1 at pH 3.0 in 10:90 dioxane-water.
Experimental Procedures
product is detected from the reaction of 1 at pH >7 when products are
analyzed after short reaction times, e.g., <1 min for water solvent.1,11
However, it is unstable and was not detected by HPLC analyses of the
product mixtures when reaction times were 20-25 min.
Materials. Diol epoxides (()-1 and (()-2 were prepared by
published procedures.8,9 (Caution: These compounds are carcinogenic
and should be handled with caution.) Dioxane was distilled from sodium
prior to use. All other reagents were purchased from commercial
sources.
Products from the Reaction of 2. Product studies for the reaction
of 2 in water and dioxane-water solutions containing 0.1 M NaClO4
and in 10:90 dioxane-water containing 0.2 M NaClO4 have been
published.1,12 Acid-catalyzed hydrolysis of 2 produces 5-10% cis tetrols
and 90-95% trans tetrols. The pH-independent reaction of 2, however,
forms cis and trans tetrols in a 40:60 ratio. The yield of tetrols from
the pH-independent reaction of 2 is somewhat less than that from the
acid-catalyzed hydrolysis, and this loss in tetrol yield is most likely
due to formation of a minor amount of ketone 6.1,12 The pH-independent
reaction of 2 has a half-life of 29 min in water at 25 °C, and ketone 6
decomposes much faster than the pH-independent rate of 2. Therefore,
it is not detected by HPLC analysis of the product mixtures.
Products from the Reaction of 2 at pH 8.3 in 10:90 Dioxane-
Water Solutions Containing Sodium Azide. Aliquots (10.0 µL) of 2
in dioxane (2 mM) were added to vials containing 2.0 mL of 10:90
dioxane-water, 0.2 M NaClO4, 10-3 M HEPES, pH 8.3, and NaN3 in
concentrations varying from 0 to 4 mM. After being swirled, the vials
were capped and allowed to stand at rt for 18 h. An aliquot (10.0 µL)
of 2-(1-naphthyl)ethanol in dioxane (5 mM) was then added to each
vial to serve as an HPLC standard, and they were analyzed by HPLC
with 60% methanol-40% water (1.2 mL/min) as the eluting solvent.
Products were monitored by UV detection at 273 nm. The retention
times for the trans tetrol product, cis tetrol product, and 2-(1-naphthyl)-
ethanol standard are 6.5, 8.3, and 10.8 min, respectively. Under these
HPLC conditions, the cis and trans azidohydrin products13 have retention
times of 15.1 and 16.7 min, respectively. However, these azide products
are stable for only several minutes under reaction conditions and
decompose to other unidentified products other than tetrols at longer
reaction times. Relative yields of tetrol products were calculated by
comparing the area of the tetrol HPLC peaks with that of the standard
for each sodium azide concentration, and comparing this ratio with
that for reaction of 2 in the absence of azide ion. The relative yields of
tetrol products from the reaction of 2 in 1, 2, 3, and 4 mM NaN3
solutions are 56, 41, 33, and 25, respectively. The cis:trans tetrol ratio
remained constant at 40:60. The reduction in yield of tetrols from the
reaction of 2 in solutions containing sodium azide is assumed to be
equal to the yield of azidohydrin.
Analytical Procedures. All HPLC analyses were carried out with
a Waters Nova-Pak C18 reverse-phase column (8 × 100 mm).
Kinetic Procedures. For each kinetic run, approximately 5 µL of a
stock solution of 1 or 2 in dioxane (3 mM) was added to 2.0 mL of
reaction solution in the thermostated cell compartment (25.0 ( 0.2 °C)
of a UV-vis spectrophotometer. Reactions were monitored at 348 nm,
and pseudo-first-order rate constants were calculated by nonlinear
regression analysis of the absorbance vs time data. For kinetic runs at
pH 5.7-9.5, approximately 10-3 M MES (2-[N-morpholino]ethane-
sulfonic acid), HEPES (N-2-(hydroxyethyl)piperazine-N ′-2-ethane-
sulfonic acid), or CHES (2-[N-cyclohexylamino]ethanesulfonic acid)
buffer was used to maintain pH. Amine buffers are both general acid
catalysts and nucleophiles in the hydrolysis reactions of 2,10 and at the
concentrations of buffer used to maintain pH, the buffer contribution
to the rate is expected to be ∼10% of the pH-independent rate of 2
and most likely accounts for the slightly greater rate of reaction of 2 at
pH 8.5-10.0 than at pH 10-11.5, where no buffer was used for pH
maintenance.
Products from Reactions of 1 in Water Solutions, 0.1 M NaClO4.
Aliquots (10.0 µL) of 1 in dioxane (2 mM) were added to vials
containing 2.0 mL of 0.1 M NaClO4 in water, 5:95 dioxane-water, or
10:90 dioxane-water (v/v) whose pH had been adjusted with either
0.1 M HClO4 or 0.1 M NaOH. After being swirled, the vials were
capped and allowed to stand at rt for 20-25 min. An aliquot (10.0
µL) of 1-acenaphthenol in dioxane (5 mM) was then added to each
vial to serve as an HPLC standard. Buffers were not used for pH control
because they efficiently catalyze the hydrolysis of 1,10,11 potentially
resulting in changes in product ratios. The pH of each reaction solution
was adjusted to ∼5-8, and they were analyzed by HPLC with 60%
methanol-40% water as the eluting solvent, 1.5 mL/min. Products were
monitored by UV detection at 273 nm. The retention times for the trans
tetrol product, cis tetrol product, and 1-acenaphthenol standard for
product studies of 1 are 7.0, 10.4, and 8.6 min, respectively. The results
of these analyses are summarized in Table 1. Yields of tetrol products
were calculated by comparing the areas of the tetrol and standard HPLC
peaks from the reaction of 1 at each pH and at pH 3.0, where only
tetrol products are formed from the reaction of 1. The HPLC retention
time of ketone 6 under the conditions listed above is 16.2 min. This
Results and Discussion
pH-Independent Reaction of 1. The pH-rate profiles for
reaction of 1 in water, 10:90 dioxane-water, and 1:3 dioxane-
water from pH 4 to pH 9.5 have been published.1 In water at
pH <∼4.5, 1 hydrolyzes by an acid-catalyzed mechanism. At
(12) Lin, B.; Doan, L.; Yagi, H.; Jerina, D. M.; Whalen, D. L. Chem.
Res. Toxicol. 1998, 11, 630-638.
(13) Lin, B.; Doan, L.; Yagi, H.; Jerina, D. M.; Whalen, D. Polycyclic
Aromat. Compounds 1999, 16, 79-88.
(8) Yagi, H.; Thakker, D. R.; Hernandez, O.; Koreeda, M.; Jerina, D.
M. J. Am. Chem. Soc. 1977, 99, 1604-1611.
(9) Yagi, H.; Hernandez, O.; Jerina, D. M. J. Am. Chem. Soc. 1975, 97,
6881-6883.
(10) Lin, B.; Islam, N.; Friedman, S.; Yagi, H.; Jerina, D. M.; Whalen,
D. L. J. Am. Chem. Soc. 1998, 120, 4327-4333.
(11) Whalen, D. L.; Ross, A. M.; Montemarano, J. A.; Thakker, D. R.;
Yagi, H.; Jerina, D. M. J. Am. Chem. Soc. 1979, 101, 5086-5088.

Diol Epoxide pH-Independent Reactions
J. Am. Chem. Soc., Vol. 123, No. 28, 2001 6787
Scheme 3
Figure 1. Plots of log kobsd vs pH for the reaction of 1 in water, 5:95
dioxane-water, and 10:90 dioxane-water (v/v), 0.1 M NaClO4, 25.0
( 0.2 °C. The solid lines are theoretical, based on eq 3 and the rate
parameters listed in Table 2.
pH 4-5, there is a change in mechanism from acid-catalyzed
to pH-independent hydrolysis, resulting in a rate plateau that
extends from pH 5 to pH 9.5. The pH-independent reaction of
1 at pH 7-8 in 10:90 dioxane-water is reported to yield ∼7%
ketone 6, which is not stable to the reaction conditions, in
addition to cis and trans tetrols.1 Other unidentified products
were also formed, presumably from the unstable ketone. Because
of the instability of ketone 6, its yield was estimated by analysis
of the product mixture after one half-life for the reaction of 1.1
We have now determined the rates of reaction of 1 from pH 4
to pH 12.5 in water, 5:95 dioxane-water, and 10:90 dioxane-
water and carried out quantitative product studies over a wider
pH range than reported previously.
Plots of log kobsd vs pH for the reaction of 1 in dioxane-
water solutions are provided in Figure 1. These plots show, as
determined earlier, a pH-dependent rate at pH <5 (region A)
and a pH-independent rate at pH 5-9.5 (region B). These data
are consistent with a change in mechanism for the reaction of
1 from acid-catalyzed to pH-independent at pH 4-6. However,
the rate of reaction of 1 decreases again at pH 9.5-10.5 (region
C), indicating another change in mechanism. Accompanying
this change of rate for the reaction of 1 is a change in product
distribution. The rate of reaction of 1 reaches a minimun at pH
∼11.5 (labeled D) and increases at pH >12 (region E) due to
a second-order reaction of 1 with hydroxide ion.
Because of the instability of ketone 6 in water solutions, its
yields from the reaction of 1 in this study were not determined
directly. Instead, the relative yield of tetrols formed from the
reaction of 1 at pH >5 was compared with that from its reaction
at pH 3.0 in 10:90 dioxane-water, where >99% of 1 reacts by
the acid-catalyzed route and only tetrols are formed. Any
reduction in the relative yields of tetrols from the reaction of 1
at a higher pH is attributed to ketone formation. These data are
summarized in Table 1. The cis and trans tetrol ratios given in
this table are consistent with earlier studies,1-4 which show that
both the acid-catalyzed hydrolysis of 1 (pH 3) and the
pH-independent reaction of 1 (pH 8) yield ∼90% cis tetrol and
∼10% trans tetrol. The total yield of tetrol products from the
pH-independent reaction of 1 at pH 8, however, is reduced by
40% as compared to their yields from acid-catalyzed hydrolysis
of 1. It is assumed that the lowering of tetrol yields is due to
the formation of ketone 6, which is not stable at intermediate
and higher pH and decomposes to other unidentified products.
Product studies for the reaction of 1 at pH 11.2 show that
the ratio of cis and trans tetrol products is similar to that formed
at lower pH. However, the yield of tetrols is reduced from 60%
at pH 8 to 29% at pH 11.2. The reduction in tetrol yield from
the reaction of 1 at pH >10 and the inflection point in the pH-
rate profile at pH 10.5 are consistent with a change of
mechanism. A mechanism for the reaction of 1 that is consistent
with the pH-rate profiles and with the change in products as a
function of pH is given in Scheme 3.
Region A. At pH <5, the rate increases with increasing [H+],
and the principal reaction is acid-catalyzed hydrolysis of 1 via
a carbocation intermediate to yield tetrols. The acid-catalyzed
hydrolyses of a number of aryl epoxides have been shown to
proceed via intermediate carbocations, formed in rate-limiting
steps.7,11,14-16 In this mechanism, therefore, it is assumed that
the reaction of 1 with H+ to form carbocation 3 is irreversible,
i.e., ks . k-H
.
Region B. From pH 5 to pH 9.5 (region B), 1 reacts via a
pH-independent reaction to yield cis and trans tetrols (4 and 5)
and ketone (6) products. It is assumed that ketone 6 is formed
from a reaction pathway, most likely concerted, that is com-
pletely separate from the stepwise mechanism leading to
tetrols.17,18 The tetrol-forming pathway of the pH-independent
reaction of 1 involves the rate-limiting reaction of 1 with water
(k1) to yield carbocation intermediate 3, trappable by azide ion
after its rate-limiting formation.7 The magnitude of this rate
plateau is therefore equal to the sum of k1 and kc.
Region C. The reduction in rate for the reaction of 1 with an
increase in pH in region C is consistent with a mechanism in
which hydroxide ion reacts as a base with carbocation 3 to re-
form epoxide 1, thus lowering the steady-state concentration
of carbocation 3 and slowing the reaction rate for tetrol
formation. In this pH region, therefore, k-1[HO-] becomes
comparable in magnitute to ks. The rate-limiting step of the
tetrol-forming pathway therefore changes from epoxide ring
opening by water to form carbocation 3 in region B (ks .
k-1[HO-]) to trapping of carbocation 3 by water in region C
(k-1[HO-] ≈ ks). A precedent for reversal in acid-catalyzed
epoxide ring opening at pH >10 is observed in the hydrolysis
of precocene I oxide,17 and there is reversal of the general acid-
catalyzed ring opening of 2 by amine bases with pKa’s >10.10
Since the rate of ketone formation (kc) is independent of pH
and the rate of tetrol formation decreases with an increase of
pH in this region, the relative yield of ketone product increases
with pH.
Regions D and E. At pH 11.2 (region D), the rate of reaction
of 1 is at a minimum. At this pH, several reactions contribute
(14) Blumenstein, J. J.; Ukachukwu, V. C.; Mohan, R. S.; Whalen, D.
L. J. Org. Chem. 1993, 58, 924-932.
(15) Mohan, R. S.; Whalen, D. L. J. Org. Chem. 1993, 58, 2663-2669.
(16) Doan, L.; Bradley, K.; Gerdes, S.; Whalen, D. L. J. Org. Chem.
1999, 64, 6227-6234.
(17) Sayer, J. M.; Grossman, S. J.; Adusei-Poku, K. S.; Jerina, D. M. J.
Am. Chem. Soc. 1988, 110, 5068-5074.
(18) Mohan, R. S.; Gavardinas, K.; Kyere, S.; Whalen, D. L. J. Org.
Chem. 2000, 65, 1407-1413.

6788 J. Am. Chem. Soc., Vol. 123, No. 28, 2001
Doan et al.
Table 2. Summary of Rate Parametersa for the Reaction of 1 in Water and DioxaneEnDashEnDash-Water Solutions, 0.1 M NaClO4, 25 °C
solvent
rate parameter
water
5:95 dioxane-water
10:90 dioxane-water
kH (M-1 s-1
)
(6.25 ( 0.21) × 102
(1.68 ( 0.03) × 10-2
(6.9 ( 0.3) × 10-3
(5.9 ( 1.0) × 103
(2.8 ( 0.3) × 10-1
(5.59 ( 0.34) × 102
(8.1 ( 0.3) × 10-3
(3.3 ( 0.1) × 10-3
(1.8 ( 1.4) × 103
(1.7 ( 0.6) × 10-1
(5.0 ( 0.2) × 102
(4.3 ( 0.1) × 10-3
(3.1 ( 0.1) × 10-3
(9.9 ( 5.8) × 103
(1.9 ( 0.2) × 10-1
b
(kc + k1) (s-1
)
[kc + k1k2/(k-1 + k2)] (s-1
)
b
(k-1 + k2)/ks (M-1
)
b,d
kOH (M-1 s-1
)
b,d
c
kc (s-1
)
(6.9 ( 0.3) × 10-3
(9.9 ( 0.4) × 10-3
(5.9 ( 1.0) × 103
(3.3 ( 0.1) × 10-3
(4.7 ( 1.0) × 10-3
(1.8 ( 1.4) × 103
(3.1 ( 0.1) × 10-3
(1.2 ( 0.2) × 10-3
(9.9 ( 5.8) × 103
c
k1 (s-1
k-1/ks (M-1
)
c,d
)
a Rate parameters are defined in Scheme 3. b Calculated from eq 3. c Calculated from eq 4. d On basis of the assignment for Kw of 10-14 M2.
Rate data were fit by nonlinear regression analysis to eq 3,
and with the assignment of Kw ) 10-14 M2 for each solvent,
they yielded values for kH, (kc + k1), [kc + k1k2/(k-1 + k2)],
and [(k-1 + k2)/ks)]. A summary of these kinetic parameters is
provided in Table 2. Rate and product data taken together clearly
show a change of mechanism for the reaction of 1 as the pH
increases from 9 to 11. The rate profiles in Figure 1 show that
this change of mechanism is most pronounced when the solvent
is water and is not so apparent when the solvent is 10:90
dioxane-water. The pH-independent rate constant for the
reaction of 1 at pH 8 (kc + k1) is much greater than the rate
constant for the acid-catalyzed hydrolysis (kH[H+]), and all of
carbocation 3 formed from the k1 pathway at this pH goes on
to tetrol product via the ks route. At this pH, the yield of ketone
from the kc route is 40%, and the yield of tetrol from the k1
pathway is 60%. Therefore, the value of kc/(kc + k1), calculated
from product yields, is equal to 0.40. The ratio [kc + k1k2/(k-1
+ k2)]/(kc + k1), calculated from parameters listed in Table 2
for the reaction of 1 in water, is equal to 0.41. This ratio is,
within experimental error, the same as the ratio kc/(kc + k1)
calculated from the product ratios, and therefore, it can be
concluded that the rate term k1k2/(k-1 + k2) is small as compared
to kc. Because k1 ≈ kc, then k-1 must be significantly larger
than k2. Thus, attack of the hydroxide on carbocation 3 to yield
tetrol products, which is the reaction giving rise to the term
k1k2/(k-1 + k2), is at best a minor reaction pathway. The term
(k-1 + k2)/ks is therefore approximately equal to k-1/ks.
By deleting k1k2/(k-1 + k2) and k2 from eq 3, because they
are small as compared to other terms in the equation, eq 3 further
simplifies to eq 4.
to the rate. The major product is ketone formed via the kc
pathway, although tetrols are also formed from both the k1 and
the kOH pathways. At pH >12 (region E), the rate of the reaction
of 1 increases due to a second-order reaction of 1 with hydroxide
ion.
Rate Profiles. With the assumption that 3 is a steady-state
intermediate and ks . k-H, the rate expression for the mecha-
nism of Scheme 3 is given by eq 1,
(kH[H+] + k1)(ks + k2[HO-])
kobsd ) kc +
+ kOH[HO-] (1)
ks + (k-1 + k2)[HO-]
where kobsd is the pseudo-first-order rate constant for the reaction
of 1 at a given pH, kc is the first-order rate constant for the
rearrangement of 1 to ketone 6, kH is the bimolecular rate
constant for the acid-catalyzed hydrolysis of 1, k1 is the first-
order rate constant for the reaction of 1 with water to form
carbocation 3 and hydroxide ion, ks is the first-order rate constant
for the reaction of 3 with water, k-1 is the second-order rate
constant for the reaction of 3 with hydroxide ion to re-form 1,
and k2 is the second-order rate constant for the reaction of
carbocation 3 with hydroxide ion to form tetrol products.
Substitution of [HO-] ) Kw/[H+] into eq 1 and rearrangement
of the resulting equation yields eq 2.
kH[H+] + (kc + k1 + kHk2Kw/ks)
kobsd
)
+
k-1 + k2 Kw
1 +
+
(
)(
)
ks
[H ]
k1k2
kobsd
)
kc +
(
)
kH[H+] + kc + k1
kc
k-1 + k2
+ kOHKw/[H+] (2)
+
+ kOHKw/[H+] (4)
[H+]
-1
[H+]
k-1 + k2
-1
k-1 Kw
k-1
1 +
1 +
1 +
(
) (
)
ks
Kw
(
)
(
) (
)
[H+]
ks
ks
Kw
Values of kc, k1, and k-1/ks calculated by fitting the data to
eq 4 are given in a separate category in Table 2. From product
studies,7 it was estimated that the ratio of the bimolecular rate
constant for attack of azide ion on 3 as compared to the first-
order rate constant for attack of water on 3 (kaz/ks) is ∼1.4 ×
103 M-1. This rate ratio is much smaller than that observed for
activation-controlled reactions of carbocations with water and
In eq 2, the maximum value of kHk2Kw/ks in 10:90 dioxane-
water can be estimated from values of kH (5.1 × 102 M-1 s-1),1
ks (1.6 × 107 s-1),7 Kw (∼10-14 M2), and a diffusion-limiting
value for k2 (∼1010 M-1 s-1) to be 3.2 × 10-9 s-1. This rate
term is very much smaller than the magnitude of the rate plateau
(kc + k1) for each solvent and can be neglected. Equation 2
then reduces to eq 3.
azide ion (kaz/ks ∼107 M-1 19,20
)
and suggests that azide attacks
3 as a nucleophile at the diffusion-limited rate.7,21,22 Calculated
k1k2
kc +
kH[H+] + (kc + k1)
k-1 + k2 Kw
k-1 + k2
(
)
(19) Richie, C. D.; Virtanen, P. O. I. J. Am. Chem. Soc. 1972, 94, 4966-
4971.
(20) Ritchie, C. D. Acc. Chem. Res. 1972, 5, 348-354.
(21) Richard, J. P.; Rothenberg, M. E.; Jencks, W. P. J. Am. Chem. Soc.
1984, 106, 1361-1372.
(22) Richard, J. P.; Jencks, W. P. J. Am. Chem. Soc. 1982, 104, 4689-
4691.
kobsd
)
+
+
[H+]
-1
k-1 + k2
1 +
1 +
+
(
)(
)
(
) (
)
ks
ks
Kw
[H ]
kOHKw/[H+] (3)

Diol Epoxide pH-Independent Reactions
J. Am. Chem. Soc., Vol. 123, No. 28, 2001 6789
Scheme 4
Scheme 5
concerted general acid-base catalysis because the pKa of the
catalyst is intermediate between the initial and the final pKa
values of the substrate site.7 A stepwise mechanism for the
general acid-catalyzed pathway involving rate-limiting proto-
nation of zwitterion intermediate 7 by the general acid (Scheme
5) has been ruled out.7 Because the epoxide ring opening
reaction catalyzed by general acids is a concerted process, the
microscopic reverse reaction involving attack of a general base
on carbocation 3 to form epoxide 1 must also be concerted. A
special problem for a concerted mechanism arises however when
water reacts with 1 to yield carbocation 3 and hydroxide ion.
The proton donor in this case, water, has a pKa higher than that
of the product carbocation 3, and Jencks’ rule for concerted
general acid-base catalysis is violated. A concerted mechanism
for this reaction is “enforced,”24 however, if the zwitterion
structure 7 is too unstable to exist as an intermediate, i.e., it
collapses without activation to 1. The encounter complex formed
from deprotonation of 3 by hydroxide ion will not have a
significant lifetime, and the rate-limiting step for the epoxide
ring opening reaction catalyzed by water must be a physical
process such as diffusion of hydroxide ion away from the
carbocation and/or conformational isomerization of the car-
bocation.7 This physical process must also be the rate-limiting
step in the reverse reaction of hydroxide ion with 3 to form 1.
The hydroxide ion that is generated from the reaction of 1
with water must be present initially in an encounter complex,
and there must be some feature that prevents this ion from
simply collapsing with the electrophilic benzylic carbocation
to form tetrol product faster than it diffuses away. One
possibility is that the newly generated ion pair might be oriented
such that only unfavorable equatorial attack of hydroxide on
the electrophilic carbon is possible.
values of k-1/ks listed in Table 2 are also in the range of 103-
104 M-1. Since this term has a value even greater than that for
kaz/ks, then reaction of hydroxide ion as a base with 3 to re-
form 1 must also occur at the diffusion-controlled limit. From
an estimated value7 of 1.6 × 107 s-1 for ks and the value for
k-1/ks listed in Table 2 for the reaction of 1 in water, k-1 is
calculated to be 9 × 1010 M-1 s-1. Deprotonation of 3 by
hydroxide ion is estimated by pKa considerations to be a
thermodynamically favorable reaction,7 and the resulting zwit-
terion is expected to close without activation to 1. Thus, that
fraction of the pH-independent reaction of 1 at pH 8 leading to
carbocation 3 at pH >10 is reversed by thermodynamically
favorable attack of hydroxide ion as a base on 3 to re-form 1.
Therefore, there is a partial change in the mechanism for the
reaction of 1b in the pH range 9.5-11.5. At pH 5-9.5, the
general acid-catalyzed reaction of water with 1 to form
carbocation 3 (k1) is partially rate-limiting, whereas at pH >11
this step is no longer rate-limiting.
At pH >11.5, the rate of reaction of 1 increases due to the
incursion of a second-order reaction with hydroxide ion. For
the reaction of 1 in 10:90 dioxane-water at pH 12, the ratio of
trans tetrol:cis tetrol is considerably larger than that formed from
the reaction of 1 at a lower pH. This result indicates that part
of the reaction of 1 at pH 12 proceeds via a bimolecular attack
of hydroxide ion on 1, leading to trans tetrol. The total yield of
tetrol product does not increase substantially, however, and this
result suggests that other products are also formed from this
second-order reaction.
Mechanism of the Reaction of Hydroxide Ion with
Carbocation 3. A structural feature of carbocation 3 that makes
it different from other carbocations whose reactions with
hydroxide ion have been studied previously is the presence of
an R-OH group that is expected to be at least several pKa units
more acidic than other OH groups because it is located adjacent
to a carbocationic center.7 Possessing both an electrophilic
carbon center and a relatively acidic OH group, 3 can react with
hydroxide ion by two competing mechanisms, one in which
hydroxide acts as a base (k-1) and the second in which hydroxide
acts as a nucleophile (k2). The fact that k-1 is greater than k2
deserves special comment. This condition must hold because if
k2 . k-1 there would be no reversal of the carbocation-forming
reaction and consequently there would be no negative deviation
of the rate profiles for the reaction of 1 at pH >10. Hydroxide
ion must therefore react with 3 more rapidly as a base to form
epoxide 1 than as a nucleophile to form tetrols 4 and 5. The
cation-anion combination reaction of 3 with hydroxide ion to
form tetrol product must therefore occur at a rate less than the
diffusion-controlled limit, which appears somewhat surprising.
The reaction of 3 with hydroxide ion to form 1 is the
microscopic reverse of the reaction of 1 with water to form 3
and hydroxide ion (Scheme 4). In considering the mechanism
of the reaction of hydroxide ion with 3, it is therefore useful to
consider the mechanism of the epoxide ring opening reaction.
General acid-catalyzed epoxide ring opening reactions of 1 by
acids with pKa’s less than that of 3 obey Jencks’ rule23 for
The reaction of hydroxide ion with 3 to form 1 (k-1) can be
viewed as a concerted general base-catalzyed epoxide ring
closure (depicted by 8), which is expected to occur at the
diffusion-controlled limit because proton transfer from the R-OH
group to the hydroxide ion is thermodynamically favorable.7
The reactions of stable carbocations with water are also general
base-catalyzed.25,26 It has been suggested that carbocation-
hydroxide combination reactions, e.g., the reaction of hydroxide
ion as a nucleophile on a carbocation, is a concerted process
involving hydroxide ion-catalyzed addition of water to the
electrophilic carbon (depicted by 9).26 In this mechanism, an
unstable carbocation-hydroxide ion pair in which the hydroxide
(24) Jencks, W. P. Acc. Chem. Res. 1980, 13, 161-169.
(25) Ritchie, C. D. J. Am. Chem. Soc. 1975, 97, 1170-1179.
(26) Ritchie, C. D.; Wright, D. J.; Huang, D.-S.; Kamego, A. A. J. Am.
Chem. Soc. 1975, 97, 1163-1170.
(23) Jencks, W. P. J. Am. Chem. Soc. 1972, 94, 4731-4732.

6790 J. Am. Chem. Soc., Vol. 123, No. 28, 2001
Doan et al.
Scheme 6
Figure 2. Plots of log kobsd vs pH for the reaction of 2 in water (0.1
M NaClO4) and 10:90 dioxane-water (v/v) (0.2 M NaClO4), 25.0 (
0.2 °C. The solid lines are theoretical, based on kobsd ) kH[H+] + ko +
k
OHKw/[H+]. For 10:90 dioxane-water solutions, published rate data
were used.11 For water solutions, fitting of the rate data yielded values
of kH, ko, and kOH equal to (1.41 ( 0.07) × 103 M-1 s-1, (4.0 ( 0.2)
× 10-4 s-1, and (7.9 ( 1.5) × 10-3 M-1 s-1, respectively.
ion is not fully solvated is bypassed,26 and thus, the concerted
mechanism is enforced. Thus, although the reactions of hy-
droxide ion with carbocation 3 as a base and as a nucleophile
appear to be rather unrelated reactions, their mechanisms may
actually be very similar. The reaction of 3 with hydroxide ion
(k2) to form tetrols may occur at less than the diffusion-limited
rate for the same reasons that more stable carbocations react
with hydroxide ion slower than with other nucleophiles such
as azide and sulfhydryl ions: deprotonation of the water
molecule in 9 is not a thermodynamically favorable reaction
because the water molecule is not sufficiently acidic and
activation energy to generate a desolvated hydroxide ion is
required.26
of 2 with water to form this same carbocation. If this were the
case, then hydroxide ion should react as a base with this
carbocation at pH >10, reversing the carbocation-forming
reaction and resulting in a slight negative rate deviation in the
rate profile. The reaction pathway from the carbocation to the
trans and cis tetrols in a 95:5 ratio would be suppressed, and
the cis to trans tetrol ratio observed from the pH-independent
reaction of 2 at pH >10 should therefore be different from the
ratio of tetrols formed at pH <10. The fact that there is no
change in the cis to trans tetrol ratio from the reaction of 2
from pH 8.5 to pH 11.5 is evidence against the general acid-
catalyzed epoxide ring opening of 2 by water as a significant
reaction pathway in the pH-independent reaction of 2.
pH-Independent Reaction of 2. Rate-pH profiles for the
reaction of 2 in water and in 10:90 dioxane-water are provided
in Figure 2. Rate data for the reaction of 2 in 10:90 dioxane-
water containing 0.2 M NaClO4 are taken from ref 10, and rate
data for the reaction of 2 in water containing 0.1 M NaClO4
over a more extended pH range than previously reported1 are
newly acquired.
Additional evidence against general acid-catalyzed epoxide
ring opening in the pH-independent reaction of 2 is obtained
by observing the cis to trans tetrol product ratio and yields of
azidohydrins from the reaction of 2 in 10:90 dioxane-water
solutions containing sodium azide (0-4 mM) at pH 8.5. Scheme
6 outlines a possible mechanism for the reaction of 2 in solutions
containing sodium azide, where some fraction R of the pH-
independent rate constant ko proceeds by the reaction of 2 with
water to yield carbocation 10, which then reacts with solvent
via the ks pathway to yield tetrol products 11 and 12. Azide ion
efficiently captures the carbocation 10 from the reaction of 2
with H+ at pH 4.7 to yield a mixture of cis and trans
azidohydrins13 and also reacts as a nucleophile with 2 (Scheme
6) at pH 8 to produce only a trans azidohydrin.10 The yield of
The rate profiles for the reaction of 2 shown in Figure 2 show
only three distinct regions: one at pH <7 where acid-catalyzed
hydrolysis predominates, a plateau at intermediate pH where
pH-independent hydrolysis predominates, and a third region at
pH >12 where a second-order reaction of 2 with hydroxide ion
occurs. Although there appears to be a very small negative rate
deviation at pH ∼10 for the reaction of 2 in both solvents, this
apparent change in rate is only on the order of 10% and can be
attributed to the fact that 10-3 M buffer was used to maintain
the pH of the reaction solutions at pH 5.7-9.5, but no buffer
was used for the pH control of the solutions at pH >10 (see
Experimental Section). The carbocation formed by the reaction
of 2 with H+ has a lifetime comparable to that formed from the
reaction of 1 with H+.7,13 If this same carbocation were formed
to any significant extent in the pH-independent reaction of 2,
then hydroxide ion should be of a sufficient concentration at
pH >10 to reverse the carbocation-forming reaction, resulting
in a negative rate deviation in the pH-rate profiles at pH >10.
Such negative rate deviations in the rate profiles for the reaction
of 2 at pH >10 are at best on the order of experimental error.
The pH-independent reaction of 2 at pH 9-11.5 in both water
and 10:90 dioxane-water yields 60-62% of trans tetrol and
38-40% of cis tetrol. The carbocation formed from the reaction
of 2 with H+ reacts with solvent to form tetrols with a trans to
cis ratio of 92:8 in water, and some portion of the trans tetrol
formed from the pH-independent reaction of 2 could possibly
be formed from the general acid-catalyzed epoxide ring opening
azide product from the reaction of 2 in sodium azide solutions
-
is 44% when [N3-] is 1.0 mM and increases to 75% when [N3
]
is 4.0 mM. From the bimolecular rate constant of 9.3 × 10-2
M-1 s-1 for kN and the pH-independent rate constant of 1.03 ×
10-4 s-1 for ko,10 the yield of azide adduct expected from the
bimolecular route (kN) can be calculated. The yields of azide
product calculated on this assumption are slightly higher (3-
6%) than the observed yields. For example, the calculated yield
of azide product from the kN route when [N3-] ) 4.0 mM is
78%. The observed yield is 75%. This rate-product correlation
is quite good and shows that most of the azide product must be
formed from the bimolecular (kN) pathway. Thus, the fraction
of tetrol formed from the ks route must not be significant. For
the reaction of 10 in this solvent containing azide ion, the rate
ratio kaz/ks was determined to be 2.6 × 102 M-1 10
.
Therefore,
51% of 10 would be trapped by azide ion by the kaz route in 4
mM azide solution to form azide products, and the remaining
49% of 10 would react with solvent by the ks pathway to form
mostly trans tetrol 12. So, if the fraction R of the pH-independent
reaction were significant, then not only would there be a greater

Diol Epoxide pH-Independent Reactions
J. Am. Chem. Soc., Vol. 123, No. 28, 2001 6791
yield of azide adduct than that actually formed but also the cis
to trans tetrol ratio should increase significantly because
approximately half of carbocation 10 leading to mostly trans
tetrol 12 by the ks pathway would be trapped by azide ion. We
have observed that the cis to trans tetrol ratio from the reaction
of 2 both in 4 mM azide solution and in the absence of azide
ion is the same (37:63). It can be concluded that carbocation
10, with a structure identical to that formed from the reaction
of 2 with H+, cannot be formed to any significant extent in the
pH-independent reaction of 2.
independent reaction of 2, however, must involve concerted or
near-concerted reactions of water with 2.
Summary
The pH-rate profiles for the reaction of 1 exhibit a negative
deviation, at pH 10-11.5, from the rate plateau for the pH-
independent reaction. This rate decrease, along with a lowering
in the yields of tetrol products, is attributed to a partial change
in mechanism. At pH 5-10, the pH-independent reaction of 1
consists of two pathways, one pathway involving direct rear-
rangement of 1 to ketone 6 that does not involve a trappable
intermediate and a second pathway involving rate-limiting
epoxide ring opening with water to yield a carbocation 3 that
reacts with solvent in a second step to yield tetrol products. At
pH >10, attack of hydroxide ion on 3 to re-form 1 exceeds the
rate at which water attacks 3, thus reducing the steady-state
concentration of 3 and lowering the rate of tetrol formation by
attack of water on 3. The relative yield of ketone from the pH-
independent reaction of 1 therefore increases at pH >10. These
results demonstrate that attack of hydroxide on carbocation 3
as a base to re-form diol epoxide 1 is much faster than its
reaction as a nucleophile with carbocation 3 to form tetrol
products. In contrast to the pH-rate profiles for the reaction of
1, the profiles for the reaction of 2 do not exhibit significant
negative rate deviation in the rate plateaus for the pH-
independent reaction. This observation and those from rate and
product studies of the reaction of 2 in solutions containing azide
ion suggest that the reaction pathways for tetrol formation in
the pH-independent reaction of 2 occur by concerted or near-
concerted addition of water to 2.
The exact mechanism for the formation of tetrol products
from the pH-independent reaction of 2 is not clear. If water
acts as a general acid in the tetrol-forming pathway, then the
hydroxide ion generated in the encounter complex must collapse
with the carbocationic center to form tetrol faster than it diffuses
away to produce the freely solvated carbocation 10 and
hydroxide ion. Significantly more cis tetrol is formed from the
pH-independent reaction of 2 (38-40%) than from the acid-
catalyzed hydrolysis of 2 (5-8%). This observation is in contrast
to the observation that the pH-independent reactions of other
diol epoxides in the “DE-2” series yield mostly trans tetrols.27,28
A possible explanation for the greater yield of cis tetrol from
the pH-independent reaction of 2 is that it reacts to form a more
stable carbocation than other diol epoxides in the series, and
the incipient hydroxide ion from general acid-catalyzed epoxide
ring opening of 2 by water is in a position to undergo favorable
axial attack on the carbocationic center. A possible mechanism
for formation of trans tetrol product involves nucleophilic
addition of water29 to the benzylic carbon of 2. Thus, cis and
trans tetrols might be formed by completely different mecha-
nisms. The mechanism(s) of tetrol formation in the pH-
(27) Whalen, D. L.; Ross, A. M.; Yagi, H.; Karle, J. M.; Jerina, D. M.
J. Am. Chem. Soc. 1978, 100, 5218-5221.
Acknowledgment. Helpful discussions with Drs. Jane Sayer
(NIH) and Donald Creighton (UMBC) are appreciated.
(28) Sayer, J. M.; Yagi, H.; Croisy-Delcey, M.; Jerina, D. M. J. Am.
Chem. Soc. 1981, 103, 4970-4972.
(29) Ross, A. M.; Pohl, T. M.; Piazza, K.; Thomas, M.; Fox, B.; Whalen,
D. L. J. Am. Chem. Soc. 1982, 104, 1658-1665.
JA004359Z