Acid-catalyzed phenylcyclohexene oxide hydrolysis: Role of para-phenyl substituent on syn:anti hydration ratio

Article abstract of DOI:10.1021/jo990220z

Rate and product studies of the hydronium ion-catalyzed hydrolysis of 1- phenylcyclohexene oxide and its p-methoxy and p-methyl derivatives in 1:9 dioxane-water solutions were carried out under conditions in which the diol products are stable. It is shown that the cis:trans diol ratio from hydrolysis of this series of epoxides does not systematically increase with the electron-donating ability of the para-substituent, contrary to conclusions in the literature. The equilibrium constants and rates for acid- catalyzed approach to equilibration of cis- and trans-1-(p- methoxyphenyl)cyclohexane-1,2-diol and of cis- and trans-1-(p- methylphenyl)cyclohexane-1,2-diol were measured, and the cis-diol was determined to be the more stable isomer in each case. An intermediate in the acid-catalyzed hydrolysis of 1-(p-methoxyphenyl)cyclohexene oxide is trapped, after its rate-limiting formation, by azide ion. For a series of 1- arylcyclohexene oxides, results are interpreted in terms of a mechanism in which there is a discrete carbocation intermediate, and products are determined solely by the partitioning reactions of this intermediate, with the pathway leading to the more stable product being energetically favored.

Full text of DOI:10.1021/jo990220z

J . Org. Chem. 1999, 64, 6227-6234
6227
Acid -Ca ta lyzed P h en ylcycloh exen e Oxid e Hyd r olysis: Role of
P a r a -P h en yl Su bstitu en t on Syn :An ti Hyd r a tion Ra tio
Lanxuan Doan, Kevin Bradley, Sonya Gerdes, and Dale L. Whalen*
Department of Chemistry and Biochemistry, University of Maryland, Baltimore County,
Baltimore, Maryland 21250
Received February 5, 1999
Rate and product studies of the hydronium ion-catalyzed hydrolysis of 1-phenylcyclohexene oxide
and its p-methoxy and p-methyl derivatives in 1:9 dioxane-water solutions were carried out under
conditions in which the diol products are stable. It is shown that the cis:trans diol ratio from
hydrolysis of this series of epoxides does not systematically increase with the electron-donating ability
of the para-substituent, contrary to conclusions in the literature. The equilibrium constants and
rates for acid-catalyzed approach to equilibration of cis- and trans-1-(p-methoxyphenyl)cyclohexane-
1
,2-diol and of cis- and trans-1-(p-methylphenyl)cyclohexane-1,2-diol were measured, and the cis-
diol was determined to be the more stable isomer in each case. An intermediate in the acid-catalyzed
hydrolysis of 1-(p-methoxyphenyl)cyclohexene oxide is trapped, after its rate-limiting formation,
by azide ion. For a series of 1-arylcyclohexene oxides, results are interpreted in terms of a mechanism
in which there is a discrete carbocation intermediate, and products are determined solely by the
partitioning reactions of this intermediate, with the pathway leading to the more stable product
being energetically favored.
In tr od u ction
Sch em e 1
The acid-catalyzed hydrolyses of 1-arylcyclohexene
oxides 1a -d yield mixtures of cis- and trans-diols 2 and
3
resulting from syn and anti hydration of the epoxide
1
group, respectively (Scheme 1). The percent of cis-diol
product was reported to increase systematically with
increasing electron-donating ability of the para-substitu-
ent; reported yields of cis-diol are given in parentheses.
When the 4-substituent is the strongly electron-with-
drawing nitro group, only very little (7.5%) of cis-diol is
formed, along with the major trans-diol product. The
reported yield of cis-diol product increases to 63% for X
Acid-catalyzed hydrolysis of 1,2,3,4-tetrahydronaphtha-
)
H, 83% for X ) CH
3
3
and 95.2% for X ) OCH .
lene-1,2-oxide (10a ) yields 94% trans-diol and only 6%
The mechanism proposed1 to explain the substituent
-3
4
cis-diol. However, acid-catalyzed hydrolysis of 6-meth-
effect on the stereoselectivity of the acid-catalyzed hy-
drolysis of 1a -d is outlined in Scheme 2. In this mech-
anism, reaction of protonated epoxide 5 leads to “intimate
ion-dipole pair” 6. Intermediate 6 can either react directly
with solvent to yield anti product 8, or isomerize to a
oxy-1,2,3,4-tetrahydronaphthalene-1,2-oxide (10b) yields
5
8
1% cis-diol product and 19% trans-diol product. In the
acid-catalyzed hydrolysis of both systems 1 and 10,
therefore, electron-donating substituents in the aromatic
ring favor the reaction leading to cis-diol.
“more cationic, solvent-separated ion-dipole pair” 7.
In the acid-catalyzed hydrolyses of the series of tet-
Ion-dipole pair 7 is proposed to collapse almost entirely
to the syn product 9 under an entropically favored
process. Electron-donating groups substituted in the aryl
ring would stabilize the “more carbocationic” intermedi-
ate 7 more than they would the “intimate ion-dipole pair”
5,6
6
rahydro epoxides 11a -e
and diol epoxides 12a -d ,
derived from polycyclic aromatic hydrocarbons, the rela-
tive yield of syn hydration product also increases quite
dramatically with the ability of the aryl group to stabilize
positive charge at the benzylic carbon of the epoxide
group. The numbering after the names refers to the
location of the epoxide group, and the syn:anti hydration
ratios are provided in parentheses. The varying cis:trans
diol product ratio from the acid-catalyzed hydrolyses of
10-12 were rationalized by a mechanism outlined in
Scheme 3. Protonation of the epoxide group and ring
opening from the more stable ground state conformation
13 yields an R-hydroxycarbocation 14, in which the
hydroxy group occupies an axial position. This carboca-
6
, thereby favoring the pathway leading to syn product.
Substituent effects on syn/anti hydration ratios were
also noted by us independently in several other systems.4
-6
(
1) Battistini, C.; Balsamo, A.; Berti, G.; Crotti, P.; Macchia, B.;
Macchia, F. J . Chem. Soc., Chem. Commun. 1974, 712-713.
2) Battistine, C.; Crotti, P.; Donatella, D.; Macchia, F. J . Org. Chem.
979, 44, 1643-1647.
3) Crotti, P.; Dell’Omodarme, G.; Ferretti, M.; Macchia, F. J . Am.
Chem. Soc. 1987, 109, 1463-1469.
4) Becker, A. R.; J anusy, J . M.; Bruice, T. C. J . Am. Chem. Soc.
979, 101, 5679-5687.
5) Gillilan, R. E.; Pohl, T. M.; Whalen, D. L. J . Am. Chem. Soc.
982, 104, 4481-4482.
(
1
(
(
1
1
(
(6) Whalen, D. L.; Ross, A. M.; Yagi, H.; Karle, J . M.; J erina, D. M.
J . Am. Chem. Soc. 1978, 100, 5218-5221.
1
0.1021/jo990220z CCC: $18.00 © 1999 American Chemical Society
Published on Web 08/04/1999

6
228 J . Org. Chem., Vol. 64, No. 17, 1999
Doan et al.
Sch em e 2
Sch em e 3
Sch em e 4
tion can then partition between reaction with solvent and
isomerization to a more stable conformation 15, in which
the hydroxy group occupies an equatorial position. It was
rationalized that carbocations 14 and 15 are related in
structure to 2-cyclohexenyl carbocations, which undergo
pected to occur with inversion of configuration at the
benzyl carbon, and also yields trans-diol.
7
preferential axial attack of solvent. Axial attack of water
Hydrolysis reactions proceeding via a relatively un-
stable intermediate 14 cannot be distinguished from a
concerted mechanism proceeding via transition state 16
by product studies alone. In a study of the acid-catalyzed
hydrolysis of tetrahydronaphthalene oxide 10a in potas-
sium chloride solutions, however, it was determined that
the cis/trans diol product ratio depended on the concen-
tration of chloride ion, although the rate of reaction of
on 14 and 15 would yield the trans- and cis-diols 17 and
1
8, respectively. When the aryl group is not able to
stabilize positive charge at the benzyl position sufficiently
well, the initially formed carbocation reacts with solvent
to yield trans diol faster than it isomerizes to 15. Thus,
anti hydration is favored. However, when the aryl group
is better able to stabilize positive charge, e.g., when a
methoxy group is substituted in the aryl group or when
the aryl group itself is better able to stabilize positive
charge, then conformational isomerization of 14 to the
more stable isomer 15 competes with attack of water on
1
0a was not significantly changed.4 It was therefore
concluded that chloride ion captures an intermediate,
after its rate-limiting formation. Thus, a concerted mech-
anism for the acid-catalyzed hydrolysis of 10a is ruled
out.
1
4, and most of the product is derived from attack of
solvent on 15. For these systems, syn hydration is
favored. For systems with electron-withdrawing groups
in the aryl group, or where the carbocations 14 and 15
are too unstable to exist as intermediates in water
solution, addition of solvent to protonated epoxide is
expected to be concerted with benzyl C-O bond cleavage
via transition state 16. This concerted reaction is ex-
The acid-catalyzed hydrolyses of cis- and trans-â-
methyl-p-methoxystyrene oxides (anethole oxides) 19a
and 19b yield the same mixture of 20% erythro- and 80%
8
threo-diols (Scheme 4). Thus, the acid-catalyzed hydroly-
sis of trans-â-methyl-p-methoxystyrene oxide proceeds
with mostly syn hydration, whereas the acid-catalyzed
hydrolysis of cis-â-methyl-p-methoxystyrene oxide pro-
(
7) Goering, H. L.; J osephson, R. R. J . Am. Chem. Soc. 1962, 84,
2
779-2785.
(8) Mohan, R. S.; Whalen, D. L. J . Org. Chem. 1993, 58, 2663-2669.

Acid-Catalyzed Phenylcyclohexene Oxide Hydrolysis
J . Org. Chem., Vol. 64, No. 17, 1999 6229
Ta ble 1. Su m m a r y of Ra te Con sta n ts for Rea ction of
Ar ylcycloh exen e Oxid es in 1:9 Dioxa n e-Wa ter (v/v), 0.1
M Na ClO4, 25.0 ( 0.2 °C
kH (M- s-1
1
)
ko (s-1
)
compound
4
3
2
-5
-6
1
1
1
a (X ) OCH3)
b (X ) CH3)
c (X ) H)
(6.00 ( 0.15) × 10
(4.91 ( 0.23) × 10
(3.60 ( 0.02) × 10
(3.49 ( 0.16) × 10
(2.55 ( 0.21) × 10
constants is provided in Table 1. The increase in reactiv-
2
ity of ca. 1.7 × 10 for 1a compared to 1c toward acid-
catalyzed hydrolysis is consistent with a mechanism in
which the transition state has substantial positive charge
development at the benzyl carbon. In a Hammett plot of
+
+
log k vs σ for these three epoxides, F is calculated to
H
+
be -2.8. The absolute value of F is smaller than that
observed for the acid-catalyzed hydrolysis of para-
substituted styrene oxides (-4.2) and reflects the less-
F igu r e 1. Plots of log kobsd versus pH for reaction of 1-aryl-
cyclohexene oxides 1a -c in 1:9 dioxane-water (v/v) solutions,
9
2
4
5.0 ( 0.2 °C, µ ) 0.1 (NaClO ). The solid lines are theoretical,
ened need for stabilization of the transition state by the
para-substitutent because the benzyl carbon in 1a -c is
tertiary instead of secondary.
Sp on ta n eou s Rea ction s of 1a a n d 1b. The rates and
preliminary product studies of the spontaneous reactions
of 1a and 1b were determined, although detailed studies
to determine their reaction mechanisms were not carried
out. The yield of diols from the spontaneous reaction of
based on eq 1 and the rate constants listed in Table 1.
ceeds with mostly anti hydration. An intermediate in
these reactions, assumed to be a carbocation, is trapped
by azide ion. It was proposed that the reactions of both
epoxide isomers with hydronium ion yield the same
discrete carbocation intermediate, and that the syn/anti
product ratio is determined solely by the partitioning
reactions of this intermediate with nucleophiles.
1
a at pH 11-12 is reduced by approximately half
compared to the yield of diols from the acid-catalyzed
reaction of 1a at pH 6. Another major product with a
longer HPLC retention time, presumably due to ketone
product, is also formed from the spontaneous reaction.
The cis:trans diol product ratio changes slightly from 74:
To determine if acid-catalyzed hydrolyses of arylcyclo-
hexene oxides proceed via discrete, trappable intermedi-
ates or instead by reactions that are concerted or nearly
concerted, we have examined the rates and products of
the acid-catalyzed hydrolysis of arylcyclohexene oxides
2
6 in the acid-catalyzed reaction of 1a to 61:39 in the
1
a -c. Our results show that the diol products from
spontaneous reaction. In contrast, the spontaneous reac-
tion of 1b at pH 11 yields more trans-diol than cis-diol
hydrolysis of 1a are not stable under the acidic conditions
used for earlier product studies and that the percent of
1
(
1
cis:trans ) 33:67), whereas acid-catalyzed hydrolysis of
cis-diol product actually formed in the acid-catalyzed
hydrolysis of 1a under conditions in which the diols are
stable is considerably lower than that reported. Instead
of a relationship in which the percent of cis-diol product
from acid-catalyzed arylcyclohexene oxide hydrolysis
increases systematically with increasing electron-donat-
ing ability of the para-substituent, we find that the
electronic nature of the para-substituent within this
series does not have a large or systematic effect on the
cis/trans hydration ratio. For these epoxides, it is pro-
posed that epoxide ring opening with hydronium ion
leads to a discrete carbocation intermediate that under-
goes partitioning reactions with solvent to yield cis- and
trans-diol products, and that the pathway leading to the
cis-diol is favored because this product is thermodynami-
cally more stable than the trans-diol.
b at pH 5 yields mostly cis-diol (cis:trans ) 83:17). The
yield of diols from the spontaneous reaction of 1b is also
reduced by about half compared to that from the acid-
catalyzed reaction.
P r od u ct Stu d ies of th e Acid -Ca ta lyzed Rea ction s
of 1a -c. We have observed that reaction of 1a in 0.2 N
H SO solution at room temperature for 24 h yields 92%
2 4
of cis-diol and 8% of trans-diol products, a result that is
in reasonably good agreement with an earlier study in
which it reported that the hydrolysis of 1a under these
1
conditions yields 95.3% of cis-diol and 4.7% of trans-diol.
We have also determined the ratios of cis- and trans-diol
products from the hydrolysis of 1a -c in 1:9 dioxane-
water solutions at pH 5-6. Relative yields of cis- and
trans-diol products from acid-catalyzed reactions of 1b
and 1c 1:9 dioxane-water at pH 5.0 were found to be
1
very similar to those reported from reaction of 1b and
Resu lts
1
c in 0.2 N H
However, acid-catalyzed hydrolysis of 1a at pH 6.0 yields
4% cis-diol and 26% trans-diol. This cis:trans diol ratio
is significantly different than that found for hydrolysis
of 1a in 0.2 N H SO solution. A summary of the products
2 4
SO solution for 24 h at room temperature.
Ra te Stu d ies. The rates of reaction of 1a -c in 1:9
dioxane-water solutions over an extended pH range were
determined. Plots of log kobsd for reaction of these epoxides
versus pH are provided in Figure 1. Rate data for reaction
7
2
4
of 1a and 1b were fit to eq 1, where k
H
is the second-
from acid-catalyzed hydrolysis of these epoxides is pro-
order rate constant for the hydronium ion-catalyzed
1
vided in Table 2. The product ratios reported for reaction
reaction and k
o
is the first-order rate
of 1a -c in 0.2 N H
2 4
SO solution for 24 h, listed in Scheme
, give a rather good correlation between the electron-
1
+
kobsd ) k [H ] + k
(1)
donating ability of the para-substituent and the relative
H
o
yield of cis-diol product. However, the yields of cis-diol
constant for the spontaneous reaction. The spontaneous
reaction of phenylcyclohexene oxide 1c was too slow for
(9) Blumenstein, J . J .; Ukachukwu, V. C.; Mohan, R. S.; Whalen,
o
k to be accurately determined. A summary of the rate
D. L. J . Org. Chem. 1993, 58, 924-932.

6
230 J . Org. Chem., Vol. 64, No. 17, 1999
Doan et al.
Ta ble 2. Rela tive Yield s of Diol P r od u cts fr om
Acid -Ca ta lyzed Hyd r olysis of Ar ylcycloh exen e Oxid es
Ta ble 3. Su m m a r y of Secon d -Or d er Ra te Con sta n ts a n d
Equ ilibr iu m Con sta n ts for Acid -Ca ta lyzed Equ ilibr a tion
of cis- a n d tr a n s-Ar ylcycloh exa n ed iols 2 a n d 3 in 1:9
Dioxa n e-Wa ter
1
a -c in 1:9 Dioxa n e-Wa ter (v/v), 0.1 M Na ClO4
compound
pHa
cis-diol:trans-diol
1
k-1 (M-1 h-1
compound
k1 (M- h-1
)
)
Keq
1
1
1
a (X ) OCH3)
b (X ) CH3)
c (X ) H)
6.0
5.0
4.0
74:26
83:17
65:35
2a , 3a (X ) OCH3)a
2b, 3b (X ) CH3)b
1.04
1.22 × 10
10.4
0.149
0.10
0.082
-2
a
a
pH at which epoxide was allowed to react for >10 half-lives
before products were analyzed.
Calculated from values of kobsd for approach to equilibrium
from both 2a and 3a in 0.03 M HClO4 solution. b Calculated from
value of kobsd for approach to equilibrium from 3b in 0.1 M HClO4
solution.
the sum of the forward and reverse rate constants. Thus,
+
k
obsd ) (k
1
+ k-1)[H ]. The half-life for approach to
equilibrium of 2a and 3a in 0.03 M HClO
4
solution is ca.
+
2
h. Dividing kobsd by [H ] gives the second-order rate
-
1
-1
constant of 11.4 M
h
for acid-catalyzed approach to
equilibrium (k + k-1) from 2a or 3a .
1
Extrapolation of the percent composition versus time
data in Figure 2 for reaction of either 2a or 3a yields a
calculated equilibrium mixture containing 91% of the cis-
diol 2a and 9% of the trans-diol 3a . Thus, the equilibrium
constant Keq (k
corresponding to a ∆G° of 1.4 kcal/mol. From the equi-
librium constant and values of (k + k-1), values for k
and k-1 are calculated. A summary of these rate con-
stants is provided in Table 3.
We have also measured the rate constant with which
3
cis-diol 2b (X ) CH ) is converted to an equilibrium
mixture of cis- and trans-isomers (2b and 3b). A plot of
1
/k-1, eq 2) is calculated to be 0.10,
F igu r e 2. Plots of percent trans-diol 3a versus time in the
equilibration reactions starting from either trans-diol 3a or
1
1
cis-diol 2a in 1:9 dioxane-water (v/v) solutions, 0.03 M HClO
5.0 ( 0.2 °C, µ ) 0.1 (NaClO ). The lines are theoretical, based
on eq 4.
4
,
2
4
from reaction of 1a -c in 1:9 dioxane-water solution at
pH 4-6 (Table 2) do not correlate with the electron-
donating ability of the para-substituent. Under both
conditions, however, cis-hydration is still the major
reaction pathway.
Acid -Ca ta lyzed Isom er iza tion of Diols 2 a n d 3. To
determine if diols 2a and 3a are stable under the reaction
conditions used to study the acid-catalyzed hydrolysis of
these data is provided as Supporting Information, and
values of k
half-life for approach to equilibrium for these diols in 1:9
dioxane-water containing 0.1 M HClO is 43 h. p-
Methoxy-substituted diols 2a and 3a are approximately
1
, k-1, and Keq are provided in Table 3. The
4
7
2
0 times more reactive than p-methyl-substituted diols
b and 3b toward acid-catalyzed equilibration. However,
the equilibrium constant Keq is very similar for the two
equilibria.
1
a , their interconversions catalyzed by H
2 4
SO in water
and HClO in 1:9 dioxane-water were studied. A sample
4
Tr a p p in g of a n In ter m ed ia te in th e Hyd r olysis
of 1a by Azid e Ion . To assess whether there is a discrete
intermediate in the acid-catalyzed hydrolysis of 1a , we
have determined the rates and products of its reaction
in 1:9 dioxane-water solutions containing sodium azide,
a strong nucleophile that is very reactive with carbo-
cations. The rate of reaction of 1a in solutions containing
sodium azide in concentrations up to 0.01 M shows no
detectable change. However, the yield of diols from the
reaction decreases with increasing sodium azide concen-
tration and two new products formed in a 3:2 ratio are
detected by HPLC. These products were isolated and
determined to be azidohydrins, formed from capture of
an intermediate carbocation 21 by azide ion (Scheme 5).
The ratio of azidohydrin products is independent of the
concentration of azide ion, and therefore the incursion
of a second-order reaction of 1a with azide ion to yield a
single, trans-azidohydrin at higher concentrations of
sodium azide where rate constants could not be deter-
mined spectrophotometrically does not occur. Since the
yields of these azidohydrins increase with increasing
azide ion concentration without a corresponding increase
in rate constant for reaction of 1a , the product-forming
step must occur after rate-limiting formation of an
intermediate.
of pure trans-diol 3a was subjected to the same reaction
conditions used for the hydrolysis of epoxide 1a in 0.2 N
H SO . Workup of the reaction after 20 h yielded a mix-
2 4
ture of diols 2a and 3a in a 92:8 ratio. This diol mixture
is the same as that obtained from hydrolysis of 1a under
these conditions. Therefore, trans-diol 3a isomerizes
2 4
readily to the cis-isomer in 0.2 N H SO solution,
demonstrating that the trans-diol product 3a formed from
hydrolysis of 1a is not stable to these reaction conditions.
To establish whether diols 2a and 3a are relatively
stable at pH > 4 in 1:9 dioxane-water solutions, the
pseudo-first-order rate constants, kobsd, for conversion of
2
a or 3a in 1:9 dioxane-water containing 0.03 M HClO
4
to an equilibrium mixture of 2a and 3a (eq 2) were
determined. Figure 2 shows plots of percent composition
of the reaction mixtures as functions of time, starting
with either pure 2a or pure 3a . Values of the rate
constants kobsd for approach to equilibrium
k1
+
+
cis-diol (2) + H y z trans-diol (3) + H
(2)
k-1
-
1
from 2a and 3a were calculated to be 3.37 ( 0.03 × 10
-
1
-1
-1
h
and 3.49 ( 0.01 × 10
h
, respectively. These two
rate constants are the same within experimental error
and reflect the fact that the measured rate constant kobsd
for approach to an equilibrium from either direction is
The fraction of diol products formed from the reaction
of 1a by the mechanism of Scheme 5 is equal to k /(k +
s s

Acid-Catalyzed Phenylcyclohexene Oxide Hydrolysis
J . Org. Chem., Vol. 64, No. 17, 1999 6231
We have also carried out rate and product studies for
the reaction of 1-(p-methylphenyl)cyclohexene oxide 1b
in solutions containing sodium azide. Hydrolysis of 1b
in 0.1 M sodium azide solution resulted in a reduction
in yield of diol products by ca. 20%, and two new products
assumed to be azide adducts were detected. The much
lower amount of trapping of an intermediate in the
reaction of 1b by azide ion, compared to that for trapping
of the intermediate in the reaction of 1a , can be at-
tributed to the greater reactivity toward solvent of the
3
intermediate carbocation 21 when X ) CH . Conse-
quently, less of the intermediate in the reaction of 1b
will be trapped by azide ion, which most likely also reacts
1
0
3
with 21 (X ) CH ) at or near the diffusional limit.
Tertiary R-substituted 4-methylbenzyl carbocations are
observed to be more reactive with solvent than the
corresponding R-substituted 4-methoxybenzyl carbo-
F igu r e 3. Plot of percent diol yield versus sodium azide
concentration in the acid-catalyzed reaction of 1a at pH 6.0
in 1:9 dioxane-water (v/v) solutions, 25.0 ( 0.2 °C, µ ) 0.1
(NaClO
4
). The line is theoretical, based on eq 3 and a
11
cations.
-
1
calculated value for kaz/k
s
of 30 ( 1 M
.
Discu ssion
Sch em e 5
Diol Isom er iza tion . Values of Keq listed in Table 3
for acid-catalyzed equilibration of diols with the p-
methoxyphenyl (2a , 3a ) and p-methylphenyl groups (2b,
3
b) are very similar. Only 8-9% of the trans-isomer is
present at equilibrium, and these results demonstrate
that the cis-diol is more stable than the trans-diol for
these two sets of isomers in 1:9 dioxane-water solution.
The similarity of Keq for interconversion of the cis- and
3 3
trans-diols 2 and 3 where X ) OCH and CH suggest
that the value of Keq is not greatly influenced by the para
substituent. Thus, the cis-diols 2c and 2d should also be
more stable than the corresponding trans-diols 3c and
3
d , respectively.
From the reported heats of combustion of the cis and
k
az[N
3
-]). The percent yield of diol products from the
trans isomers of 1,2-cyclohexanediol and of 1-phenyl-1,2-
cyclohexanediol,12 the cis isomer is slightly more stable
than the trans isomer in each case (ca. 0.5-1.0 kcal/mol).
If the aryl group in both cis- and trans-1-arylcyclohexane-
reaction of 1a is therefore given by eq 3, and a plot of
the percent yield of diol products versus the concentration
1
kaz
00
1
,2-diols occupies an equatorial position, then only one
%
diol products )
(3)
hydroxy group in the cis-diol 2 would be located in an
axial position (conformation 24) whereas both hydroxy
groups in the trans-diol 3 would be forced to occupy axial
positions (conformation 25). Conformation 24 (cis-diol)
has two 1,3-diaxial H-OH interactions, one gauche
phenyl-OH interaction, and one gauche OH-OH inter-
action, whereas conformation 25 (trans-diol) has four 1,3-
diaxial H-OH interactions and one gauche phenyl-OH
interaction.
1
+
[N3-]
ks
of azide ion is given in Figure 3. By fitting these data to
eq 3, a value for kaz/k
carbocations that have kaz/k
to that for reaction of 1a , it has been proposed that their
-
1
s
is calculated to be 30 M . For
partitioning ratios similar
s
reactions with azide ion occur at the diffusional limit,
9
-1
with a bimolecular rate constant kaz of ca. 5 × 10 M
-
1 10
s
.
For this value of kaz, a value for k
s
for reaction of
carbocation 21a in 1:9 dioxane-water is calculated to be
8
-1
1
.7 × 10 s . This calculated rate constant is slightly
larger than k calculated for reaction of R-dimethyl-
s
6 4
substituted 4-methoxybenzyl carbocation (4-MeOC H C-
+
) in 1:1 trifluoroethanol-water (1.3 × 10 s-1).11
7
3 2
(CH )
The rate of reaction of a para-substituted benzyl car-
bocation with solvent generally increases as the carbo-
cation becomes less stable.10 However, for a series of
tertiary R-substituted 4-methoxybenzyl carbocations of
From the rate data listed in Table 3, the half-life for
approach to equilibrium of 2a and 3a is calculated to be
2
.0 h at 25 °C in 0.03 M HClO
solution. Therefore, the half-life for equilibration of these
diols in 0.2 N H SO solution is expected to be less than
h. In contrast, the half-life for equilibration of 2a and
4
, 1:9 dioxane-water
s
widely varying stabilities, the value of k is nearly
independent of the thermodynamic stability of the car-
bocation.11
2
4
1
3
a at pH 6 under the condition used for our product
(
10) Richard, J . P.; Rothenberg, M. E.; J encks, W. P. J . Am. Chem.
Soc. 1984, 106, 1361-1372.
11) Aymes, T. L.; Stevens, I. W.; Richard, J . R. J . Org. Chem. 1993,
8, 6057-6066.
(
(12) von Verkade, P. E.; Coops, J ., J r.; Maan, C. J .; Verkade-
5
Sandbergen, F. A. Ann. 1928, 467, 217-239.

6
232 J . Org. Chem., Vol. 64, No. 17, 1999
Doan et al.
Sch em e 6
trans-diol product, whereas axial attack of solvent on
conformation 28 yields cis-diol product. If conformations
2
7 and 28 are in rapid equilibrium, then the ratio of diol
products will be independent of the equilibrium concen-
trations of carbocation conformations, but will instead
depend only on the difference in energy between the
transition states leading to cis- and trans-diol products
(
Curtin-Hammett Principle).14 It is reasonable to assume
that those factors that contribute to the greater stability
of the cis-diol product will therefore contribute to the
lowering of the transition state for this product-forming
pathway. In the acid-catalyzed hydrolysis of phenylcy-
1
5,16
clohexene oxides, benzo[a]pyrene diol epoxides
and
8
other epoxides that react via carbocation intermediates
that are sufficiently stable to be trapped by external
nucleophiles, the energy difference between the syn- and
anti-hydration products may be the principal factor
determining whether syn or anti hydration is favored. For
3
studies (Table 2) is calculated to be 2.5 × 10 days. For
the conditions of our product studies for acid-catalyzed
hydrolysis of 1a (Experimental Section), therefore, equili-
bration of the diol products 2a and 3a is negligible.
Diols 2b and 3b are much less reactive then diols 2a
and 3a toward acid-catalyzed isomerization. Whereas the
half-life for approach to equilibrium for 2a and 3a in 0.03
+
those epoxides that react with H to yield carbocations
that are not sufficiently stable to attain conformational
equilibration, then solvent attack on the initially formed
carbocation may become the major product-forming step.
+
For other epoxides that would react with H to yield
M HClO
the half-life for approach to equilibrium for 2b and 3b
in 0.1 M HClO at 25 °C in 1:9 dioxane-water solution
is 43 h. Diols with hydrogen and nitro at the para-
position of the phenyl ring (2c,d and 3c,d ) will be even
more stable in acidic solution than those with methyl at
the para-position (2b, 3b).
4
at 25 °C in 1:9 dioxane-water solution is 2 h,
extremely unstable carbocations, concerted reactions may
predominate.
4
Mech a n ism of Hyd r olysis of 1-(p-Nitr op h en yl)-
cycloh exen e Oxid e 1d a n d Oth er P h en ylcycloh ex-
en e Oxid es w ith Electr on -With d r a w in g Gr ou p s
Su bstitu ted in th e P h en yl Rin g. We did not evaluate
the relative stabilities of p-nitrophenylcyclohexane diols
2d and 3d in sulfuric acid solution. However, they are
expected to be much more stable toward acid than other
diols in the series with p-phenyl substituent groups that
are not electron-withdrawing, and should be stable to the
The cis:trans diol mixtures formed from hydrolysis of
epoxide 1a and from reaction of trans-diol 3a in 0.2 N
H
2
SO
similar to the equilibrium mixture of 91% cis-diol and
% of trans-diol in 1:9 dioxane-water. Thus, the value
of Keq for isomerization of these diols in 0.2 N H SO is
4
at room temperature for 20 h (ca. 92:8) are very
9
conditions used for hydrolysis of 1d in 0.2 N H
2
SO
4
solution.1 Acid-catalyzed hydrolysis of 1d (X ) NO
-3
) and
2
4
2
very close to that for their interconversion in 1:9 diox-
ane-water.
other phenylcyclohexene oxides with electron-withdraw-
ing groups substituted in the phenyl ring results in
greater yields of trans-diol,1 and therefore these ep-
oxides must react at least partially by a mechanism
different than that by which 1a -c react. The presence
of the strongly withdrawing nitro group in the phenyl
ring of 21d (Scheme 5) greatly destabilizes this carbo-
cation, and therefore a concerted pathway proceeding via
transition state 29 (Scheme 6) may contribute to the
much higher yield of trans-diol product. Another pos-
-3
Mech a n ism of Hyd r olysis of Ar ylcycloh exen e
Oxid es 1a -c. From the diol product distributions sum-
marized in Table 2 for acid-catalyzed hydrolysis of 1a -
c, it is clear that there is not a good correlation within
this series between the relative yield of cis-diol product
and the electron-donating ability of the para-substituent
in the phenyl ring. Upon substitution of hydrogen at the
para-position by methyl, the relative yield of cis-diol
increases from 65% to 83%. However, further substitution
of the methyl group by a much more electron-donating
methoxy group results in a decrease of the relative yield
of cis-diol to 74%. For the acid-catalyzed hydrolysis of
epoxides 1a -c, therefore, the relative yields of cis- and
trans-diols do not change in a large or systematic way.
A mechanism that accounts for the cis:trans diol
product ratio from acid-catalyzed hydrolysis of arylcy-
clohexene oxides 1a -c is given in Scheme 6. In this
scheme, it is assumed that axial epoxide ring opening
and axial attack of solvent on the intermediate carbo-
cation are energetically favored over equatorial ring
opening and equatorial attack of solvent on the interme-
diate.13 For acid-catalyzed hydrolysis of 1a -c, axial ring
opening of the epoxide (26) yields an intermediate carbo-
cation, which may exist in two chair conformations, 27
and 28. Axial attack of solvent on conformation 27 yields
+
sibility is that reaction of 1d with H proceeds with axial
opening to form the unstable carbocation conformation
27d , which reacts with solvent by axial attack to yield
trans-diol product faster than it undergoes conforma-
tional isomerization to 28d . The exact role of solvent
molecules that are closely coordinated to the transition
states and/or intermediates is not clear.
Con clu sion s
The cis:trans diol ratio from the acid-catalyzed hy-
drolysis of 1-phenylcyclohexene oxide and its para-
phenyl-substituted methoxy and methyl derivatives do
not depend systematically on the electron-donating ability
of the para-substituent, contrary to published conclusions.
Instead, the cis:trans diol ratios from acid-catalyzed
hydrolysis of these three epoxides, under conditions in
(
14) (a) Curtin, D. Y.; Rec. Chem. Prog. 1954, 15, 111. (b) Eliel, E.
(
13) Eliel, E. L. Stereochemistry of Carbon Compounds; McGraw-
L. Stereochemistry of Carbon Compounds; McGraw-Hill: New York,
1962; pp 151-152.
Hill: New York, 1962; pp 230-231.

Acid-Catalyzed Phenylcyclohexene Oxide Hydrolysis
J . Org. Chem., Vol. 64, No. 17, 1999 6233
with pH in the range 4-10, ca.. 10- M of a buffer reagent
3
which the diol products are stable, are very similar. Acid-
catalyzed equilibration of the cis- and trans-p-methoxy-
phenylcyclohexene diols 2a and 3a occurs readily at pH
was added. Buffers used were acetic acid (pH 4.1-5.5); MES
(
2
8
2-[N-morpholino]ethanesulfonic acid), pH 5.0-6.5; HEPES (N-
-hydroxyethylpiperazine-N′-2-ethanesulfonic acid), pH 7.0-
.0; EPPS (N-2-hydroxyethyl)piperazine-N′-3-propanesulfonic
<
2 and accounts for the higher published yield of cis-
diol from the acid-catalyzed hydrolysis of 1a in 0.2 N
SO solution. For 1-phenylcyclohexanediols possessing
acid), pH 8.0-8.7; and CHES (2-[N-cyclohexylamino]ethane-
sulfonic acid), pH 8.5-10.0. For determining pH-rate profiles,
the reactions of 1a -c were monitored at 235, 240, and 225
nm, respectively. The reactions of 1a and 1b in solutions
containing sodium azide were monitored at 240-242 nm.
Pseudo-first-order rate constants were calculated by nonlinear
regression analysis of the time versus absorbance data.
An a lysis of P r od u cts fr om Acid -Ca ta lyzed Hyd r olysis
of 1a -c. A 10-µL solution of each epoxide in dioxane (2 mg/
mL) was added by syringe to 1.0 mL of 1:9 dioxane-water
4
H
2
4
p-methoxy and p-methyl substituents in the phenyl ring
and presumably for other 1-arylcyclohexanediols, the cis-
isomer is more stable than the trans-isomer. This obser-
vation may account for the greater yields of cis-diol from
acid-catalyzed hydrolysis of 1-(p-methoxyphenyl), 1-(p-
methylphenyl), and 1-phenylcyclohexene oxides. An in-
termediate in the acid-catalyzed hydrolysis of 1-(p-
methoxyphenyl)cyclohexene oxide is trapped, after its
rate-limiting formation, by azide ion. The acid-catalyzed
hydrolysis of this series of epoxides is accommodated by
a mechanism involving rate-limiting formation of a
carbocation intermediate followed by partitioning of this
intermediate to diol products, with the pathway leading
to the more stable cis-diol being energetically favored.
solution, containing 0.1 M NaClO , that had been adjusted to
pH 6.0 for reaction of 1a , to pH 5.0 for reaction of 1b, and to
pH 4.0 for reaction of 1c. At these pH values, k
+
H
[H ] > 100k
o
;
e.g., the acid-catalyzed reaction predominates. The reaction
solutions of 1a and 1b were thoroughly mixed and allowed to
stand at room temperature for ca. 2 min (ca. 10 and 8.5 half-
lives, respectively); the reaction solution of 1c was allowed to
stand for 4 min (ca. 12 half-lives). The solutions were then
analyzed directly by reverse phase HPLC on a C18 column. The
products were eluted with 3:2 (v/v) methanol-water as eluent
at a flow rate of 1.0-1.2 mL/min and monitored by UV
detection at 273 nm. Retention times for 2a and 3a were 6.7
and 4.1 min; for 2b and 3b, 11.0 and 5.8 min; for 2c and 3c,
10.7 and 6.7 min, respectively. A summary of the relative
yields of diol products from acid-catalyzed hydrolysis of 1a -c
is provided in Table 2.
Exp er im en ta l Section
Ma t er ia ls. 1-(p-Methoxyphenyl)cyclohexene, 1-(p-meth-
ylphenyl)cyclohexene, and 1-phenylcyclohexene were prepared
by addition of cyclohexanone to the appropriate Grignard
reagent, followed by p-toluenesulfonic acid-catalyzed dehydra-
1
7
tion of the alcohol from the Grignard reaction. Dioxane was
distilled from sodium prior to use. All other reagents were
purchased from commercial sources and used without further
purification.
From larger scale reactions, diol products from hydrolyses
1a -c separated by column chromatography on alumina with
pentane-ether as eluent. The identities of the diol products
from hydrolyses of 1a -c were established by comparison of
Published preparations of 1a -c involved conversion of the
their 1H NMR spectra with published H NMR data.
1
18
precursor olefin to a bromohydrin followed by potassium tert-
1
8
butoxide ring closure of the bromohydrin to form the epoxide
Acid -Ca ta lyzed Equ ilibr a tion of Diols 2a a n d 3a . A
solution containing 30 mg of 2a in 0.5 mL of dioxane was
prepared. A 30 µL portion of this solution was added by syringe
in 1:9 dioxane-water (µ ) 0.10,
NaClO ) at 25.0 °C. At different time intervals, 0.3 mL of the
reaction solution was removed, quenched with 0.3 mL of 0.03
M NaOH in 1:9 dioxane-water, and analyzed by reverse phase
HPLC on a C18 column under the same conditions used for
analysis of the diol products from hydrolysis of 1a . The results
of this experiment are graphically illustrated in Figure 2.
The percent of trans-diol in the reaction solution during the
or direct epoxidation of the olefin by peroxybenzoic acid.19 We
synthesized 1a -c by direct epoxidation of the olefin under
biphasic, buffered conditions.
to 3.0 mL of 0.03 M HClO
4
1
-(p-Meth oxyp h en yl)cycloh exen e Oxid e (1a ). A solution
of 0.155 g (0.9 mmol) of m-chloroperoxybenzoic acid (85%) in
mL of methylene chloride was added dropwise over 5 min to
4
5
a well-stirred biphasic mixture of 0.105 g (0.56 mmol) of 1-(p-
methoxyphenyl)cyclohexene in 5 mL of methylene chloride and
5
mL of 10% (w/w) sodium carbonate in an ice-water bath.
The reaction mixture was allowed to stir for an additional 5
min. The aqueous phase was separated from the organic phase
and washed with several milliliters of methylene chloride. The
methylene chloride solutions were combined, washed with 1.0
M NaOH, and dried over sodium sulfate. Removal of solvent
yielded 0.11 g of crude product, which was recrystallized from
pentane-diethyl ether solution to yield 82 mg (45%) of 1a , mp
equilibration experiments is given by eq 4, where (% trans)
t
-
kobsd t
(
% trans) ) [(% trans) - (% trans) ] e
+
t
0
eq
(% trans)eq (4)
1
8
4
6.9-47.4 °C, lit. mp 44-45 °C.
-(p -Met h ylp h en yl)cycloh exen e oxid e (1b) a n d 1-(p -
is the percent of trans-diol at a given time, (% trans) is the
0
initial concentration of trans-diol, (% trans)eq is the equilibrium
concentration of trans-diol, and kobsd is the pseudo-first-order
rate constant for the approach to equilibrium. In the equilib-
rium experiment starting from pure trans-diol 3a , fitting of
1
m eth oxyp h en yl)cycloh exen e oxid e (1c) were prepared by
procedures similar to that for preparation of 1a , except that
the reaction time for epoxidation was increased to 30 min.
Kin etics P r oced u r es. For each kinetic run, approximately
-
1
-1
(% trans)
t
to eq 4 yielded values of 3.37 ( 0.03 × 10
h
for
k
obsd and 9.44 ( 0.27% for (% trans)eq. In the equilibration
5
µL of a stock solution of 1 in dioxane (ca. 3 mg/mL) was added
experiment starting from pure cis-diol 2a , fitting of (% trans)
t
to 2.0 mL of 1:9 dioxane-water solution in the thermostated
cell compartment (25.0 ( 0.2 °C) of a UV-vis spectrophotom-
eter. The ionic strength of each solution was made to be 0.10
-
1
-1
to eq 4 yielded values of 3.49 ( 0.01 × 10
h
and 8.84 (
0
.14% for kobsd and (% trans)eq, respectively.
The procedure used to determine the rate constant for acid-
catalyzed equilibration of 3a in 0.03 M HClO to an equilib-
4
M with added NaClO . For maintenance of pH for solutions
4
rium mixture of 2a and 3a was also used to determine the
rate constant for acid-catalyzed equilibration of 3b to an
equilibrium mixture of 2b and 3b, except that the concentra-
(
15) Islam, N. B.; Gupta, S. C.; Yagi, H.; J erina, D. M.; Whalen, D.
L. J . Am. Chem. Soc. 1990, 112, 6363-6369.
16) Lin, B.; Islam, N.; Friedman, S.; Yagi, H.; J erina, D. M.; Whalen,
D. L. J . Am. Chem. Soc. 1998, 120, 4327-4333
17) Davies, M. T.; Dobson, D. F.; Hayman, D. F.; J ackman, G. B.;
Lester, M. G.; Petrow, V.; Stephenson, O.; Webb, A. A. Tetrahedron
1
1
1
(
4
tion of HClO used was 0.1 M instead of 0.03 M. Fitting of the
-
2
(
diol composition to eq 4 yielded a value of (1.61 ( 0.13) × 10
-1
h
for kobsd and 7.6 ( 2.1% for (%trans)eq.
962, 18, 751-761.
18) Balsamo, A.; Crotti, P.; Macchia, B.; Macchia, F. Tetrahedron
973, 29, 2183-2188.
19) Berti, G.; Bottari, F.; Macchia, B.; Macchia, F. Tetrahedron
965, 21, 3277-3283.
P r od u ct Stu d ies of th e Rea ction of 1a in Sod iu m Azid e
(
Solu tion s. a . An a lytica l P r oced u r e. Aliquots (30 µL) of 1a
in dioxane (4 mg/mL) were added to vials containing 1.0 mL
(
of 1:9 dioxane-water containing concentrations of NaN
3

6
234 J . Org. Chem., Vol. 64, No. 17, 1999
Doan et al.
varying between 0 and 0.1 M, with pH adjusted to 5.8-5.9 by
addition of 0.1 M HClO in 1:9 dioxane-water. Ionic strength
was kept constant at 0.1 M by addition of NaClO . After
6.95 (d, J ) 9.1 Hz, 2 H), 7.44 (d, J ) 9.1 Hz, 2 H). Anal. Calcd
4
for C13
17 2 3
H O N : C, 63.14; H, 6.93. Found: C, 63.28; H, 7.15.
4
Absorptions due to the OCH
hydrogens and the hydrogen
3
addition of the epoxide, the reaction vials were capped, shaken,
and allowed to stand at room temperature for ca. 5 min. An
aliquot (25 µL) of 3b in dioxane (6 mg/mL) was then added to
each vial to serve as an HPLC standard. Solutions were then
analyzed by HPLC on a reverse phase C18 column with 3:2
methanol-water as eluting solvent, 1.4 mL/min. The yields
of diol products were calculated by comparing the areas of their
HPLC peaks with that of the standard. The total yield of diols
was lowered as the sodium azide concentration was increased.
These results are graphically represented in Figure 3. Two new
products, formed in a 3:2 ratio and assumed to be azide
products formed from reaction of azide ion from both sides of
the electron deficient benzyl carbon of the carbocation inter-
mediate, were detected by HPLC (retention times 15 and 20
min). The yields of these products increase with increasing
azide ion concentrations.
R to the OH group are located at δ 3.8 and are not resolved.
However, a COSY experiment revealed correlations between
this signal and other absorptions at δ 1.2 and δ 1.7, consistent
with that expected for coupling of the hydrogen R to the OH
group to each of two vicinal methylene hydrogens. After sev-
eral chromatography fractions containing both azide products,
the major azide product with HPLC retention time of 15
min eluted and was recrystallized from diethyl ether-pen-
tane solution to give 35 mg of crystalline product; mp 78.5-
-
1
1
8
4 3
0.5 °C; IR (CCl ) 2110 cm ; H NMR (200 MHZ, CDCl )
δ 1.2-2.1 (9 H), 3.82 (s, 3 H), 3.87 (dd, J ) 10.2, 4.2 Hz, 1
H), 6.95 (d, J ) 9.1 Hz, 2 H), 7.41 (d, J ) 9.1 Hz, 2 H). Anal.
Calcd for C13
.00.
The cis stereochemistry was assigned to the major azide
17 2 3
H O N : C, 63.14; H, 6.93. Found: C, 63.35; H,
7
1
1
product (cis-23) on the basis of the larger vicinal H- H
coupling between the methine hydrogen R to the OH group
and the neighboring methylene hydrogens in its NMR spec-
trum, by analogy with the greater width of the absorption of
the methine hydrogen in cis-diol 2a compared to that of the
trans-diol 3a 18 and of related cis and trans hydroxy ethers.
b. P r ep a r a tive P r oced u r e. A sample of crude 1a (0.36 g)
in 5 mL of dioxane was added to 50 mL of 0.5 M NaN in 1:9
3
dioxane-water solution, pH 5.8, and the resulting solution was
stirred at room temperature for 1.5 h. The reaction solution
was then extracted with 100 mL of diethyl ether. The ether
extract was washed with water and dried over calcium sulfate.
Removal of solvent yielded 0.45 g of crude azide products. A
portion of this product (0.1 g) was chromatographed on silica
gel (240-400 mesh) with 1:1 diethyl ether-pentane as eluting
solvent.
3
Su p p or tin g In for m a tion Ava ila ble: Plot of percent
trans-diol 3b versus time for the approach to equilibrium of
2
4
b and 3b, starting from pure 3b, in 0.1 M HClO , 1:9
The minor azide product with HPLC retention time of 20
min (see Analytical Procedure above) eluted first as an oil.
dioxane-water (v/v), 25.0 ( 0.2 °C. This material is available
free of charge via the Internet at http://pubs.acs.org.
Approximately 20 mg of this azide was isolated; IR (CCl
4
) 2110
-
1 1
cm ; H NMR (200 MHZ, CDCl
3
) δ 1.4-2.3 (9 H), 3.83 (4 H),
J O990220Z