Inter- and intramolecular stereoselective protonation of enols

Article abstract of DOI:10.1021/jo026187p

Kinetic control of the stereoselectivity of protonation of enolates and other delocalized species commonly affords the less stable diastereomer as a consequence of the considerable exothermicity and resulting sp² transition-state hybridization of the α-carbon. Protonation then is from the less hindered face of the enolate. The present study is aimed at reversing this phenomenon by intramolecular delivery of the proton. The approach employed required the synthesis of two enolate precursors, one with a 2-pyridyl group strategically close to the α-carbon and the other with a phenyl group in the same location. The synthesis required 15 steps and involved new methodology. Intramolecular proton transfer, reversing the usual stereoselectivity, was successful. The selectivity proved to depend on several factors including the exo versus endo configuration of the diastereomer reacting, the proton donor employed, and the concentration of the proton donor. A kinetic analysis permitted the determination of the relative reaction orders of the protonation on the two faces of the enolate.

Full text of DOI:10.1021/jo026187p

In ter - a n d In tr a m olecu la r Ster eoselective P r oton a tion of En ols1,2
Howard E. Zimmerman* and Pengfei Wang
Chemistry Department, University of Wisconsin, 1101 University Avenue, Madison, Wisconsin 53706
zimmerman@chem.wisc.edu
Received J uly 11, 2002
Kinetic control of the stereoselectivity of protonation of enolates and other delocalized species
commonly affords the less stable diastereomer as a consequence of the considerable exothermicity
and resulting sp2 transition-state hybridization of the R-carbon. Protonation then is from the less
hindered face of the enolate. The present study is aimed at reversing this phenomenon by
intramolecular delivery of the proton. The approach employed required the synthesis of two enolate
precursors, one with a 2-pyridyl group strategically close to the R-carbon and the other with a
phenyl group in the same location. The synthesis required 15 steps and involved new methodology.
Intramolecular proton transfer, reversing the usual stereoselectivity, was successful. The selectivity
proved to depend on several factors including the exo versus endo configuration of the diastereomer
reacting, the proton donor employed, and the concentration of the proton donor. A kinetic analysis
permitted the determination of the relative reaction orders of the protonation on the two faces of
the enolate.
1. In tr od u ction
lectivity might be reversed. One approach involves in-
tramolecular proton transfer by a moiety located on the
hindered face of the enolate. We have provided one
example of this in a previous study,4 and it remained to
be determined if intramolecular proton transfer really
is a general phenomenon.
This was the intent of the present investigation. The
enols of interest are the exo and endo pyridyl diastereo-
mers 4ex and 5en below. We note here that “endo” at
C-6 refers to the aryl being cisoid to the functional group,
and “syn” at the R-carbon (i.e., C-5) refers to the proximity
to the aryl group.
Ba ck gr ou n d . A very large number of organic reac-
tions involve the protonation of enol and enolate inter-
mediates. Conjugate additions to enones, decarboxylation
of â-ketoacids, malonic acids, dehalogenation of R-ha-
loketones, deprotonation-protonation of ketones and
esters, Birch-Barton-Stork lithium-ammonia reduction
of enones, and nucleophilic addition to ketenes are some
examples. Protonation of nitronates proceeds similarly.
Many of these examples have been summarized, and a
long series of our publications have provided evidence for
the generality of the phenomenon.3
This kinetic control results from the proton transfer
being highly exothermic with an early transition state.
Thus, the hybridization of the R-carbon remains close to
sp2 hybridized, and the approach of the proton donor
occurs preferentially to the less hindered face of the
delocalized species (enol, enolate, nitronate, etc.). Equa-
tion 1 provides one example.3d
2. Resu lts
Syn th etic Ap p r oa ch . The silyl enol ethers of these
enols promised to be ideal enol precursors. Of particular
interest were the tert-butyldimethylsilyl enol ethers of
the benzoyl ketones 26en and 26ex. Note the discussion
of this point below. Scheme 1 outlines the synthesis of
these precursors. A number of the steps of the synthetic
sequence are especially noteworthy. The basic five-
membered ring formation leading to cyclopentene 8 began
with cis-1,4-dichloro-2-butene. This led, as shown, to 8,
accompanied by cyclopropane 9 in a 1:3 ratio. However,
thermal conversion from 9 to 8 brought the yield to 50%.
In view of this intriguing phenomenon, it was of
considerable interest to ascertain whether the stereose-
(1) This is paper 267 of our general series.
(2) For paper 266 and a preliminary communication of the present
study, see Zimmerman, H. E.; Wang, P. Org. Lett. 2002, 4, 2593-2595.
(3) (a) Zimmerman, H. E. Acc. Chem. Res. 1987, 20, 263-268 and
references therein. (b) Zimmerman, H. E. J . Org. Chem. 1955, 20, 549-
557. (c) Zimmerman, H. E.; Linder, L. W. J . Org. Chem. 1985, 48,
1637-1646. (d) Zimmerman, H. E. J . Am. Chem. Soc. 1956, 78, 1168-
1173.
(4) Zimmerman, H. E.; Ignatchenko A. J . Org. Chem. 1999, 64,
6635-6645.
10.1021/jo026187p CCC: $22.00 © 2002 American Chemical Society
Published on Web 11/27/2002
9216
J . Org. Chem. 2002, 67, 9216-9226

Stereoselective Protonation of Enols
SCHEME 1. Syn th esis of r equ ir ed com p ou n d s
A particularly difficult step was the conversion of
dibromide 10 to the cyclopentadiene 14. With a variety
of bases, two undesired products were obtained. Both
DBU in DMF and sodium methoxide in THF afforded an
intramolecular SN2 product 28 as shown in Scheme 2.
Potassium tert-butoxide in THF and potassium hydroxide
in ethanol afforded the vinyl bromide 29; again, see
Scheme 2. However, the desired elimination product 14
resulted nicely with potassium tert-butoxide in 18-
crown-6 ether; this led to a selective elimination in 86%
yield to afford phenylpyridyl cyclopentadiene 14. The
transformation proved to be quite temperature-depend-
ent, giving the vinyl bromide as a minor product at room
temperature but the desired diene 14 exclusively at lower
temperatures.
SCHEME 2. Sid e r ea ction s of in ter est
The Diels-Alder reaction of nitroethylene with cyclo-
pentadiene 14 nicely afforded both the endo and exo
stereoisomers of norbornene 13 in a 3:2 ratio. And the
exo product 13ex was reduced to the corresponding
norbornane 16ex. Only the exo isomer proved to be of
value in the subsequent sequence employing a Nef
reaction and leading to the norbornyl ketone 15ex, as
depicted in Scheme 1. Although the reductionsNef
J . Org. Chem, Vol. 67, No. 26, 2002 9217

Zimmerman and Wang
SCHEME 3. Wolff r ea r r a n gem en t r esu lts
approachsgave low yields in the endo series, the reversed
sequence proved practical, this involving the initial
conversion to the norbornenone 11en followed by reduc-
tion to the keto norborane 15en . One limitation was that
under standard Nef conditions nitronorbornene 13en had
been found in previous work5 to give a rearrangement
and not to afford the desired ketone 11en . But conditions
using a silica gel conversion to oxime 12en circumvented
the problem.
A particularly interesting step was the Wolff rear-
rangement of the diazoketones 19ex and 19en . Strik-
ingly, the reaction in methanol proceeded with stereo-
chemistry depending on which diastereomer was used.
Thus, 19en afforded product 20a primarily with an anti
carbomethoxyl group whereas 19ex led mainly to product
22ex with a syn carbomethoxyl group. This result was
independent of whether pure methanol was used or,
instead, 5 or 10% methanol in benzene was employed.
The differing stereoselectivity is dramatic; this finding
is considered below in the context of the reaction mech-
anisms.
We required the endo and exo pyridyl benzoyl ketone
diastereomers, 26ex and 26en . The exo-pyridyl isomer
26ex was obtained relatively easily from acid 21ex, as
shown in Scheme 1. However, the phenyllithium reaction
led to poor yields when applied to the endo acid 21en .
But, the route via aldehyde 23en proved useful.
A real challenge was presented by the need for the silyl
enol ethers of the benzoyl ketones, 26en and 26ex. One
unsuccessful approach was deprotonation and silation.
LDA, LTMP, and potassium hydride with crown ether
all proved incapable of removing the R-proton of the exo-
pyridyl ketone 26ex. In contrast, both ketones, 26en and
26ex, reacted nicely with bromine and 2 equiv of HBr in
acetic acid to afford the R-bromo ketones 25en and 25ex.
The configurations at the R-carbon were assigned on a
mechanistic basis. The success of this route results from
the bromination having a rate-limiting step of carbonyl
protonation and the loss of the R-proton from the result-
ing cationic species, a process less subject to steric
inhibition. No large base must approach the proton to
be removed. Each of the two R-bromoketones was readily
converted to the tert-butylsilyl enol ethers 24en and 24ex
on reaction with tert-butyllithium and tert-butyldimeth-
ylsilyl chloride.
Som e F u r th er Syn th etic Asp ects In clu d in g a
Discu ssion of Sid e Rea ction s. Above, we have indi-
cated that the conversion of dibromide 10 to the desired
cyclopentadiene led to an intramolecular SN2 reaction to
afford a pyridinium product 28 as well as to the undesired
vinyl bromide 29 depending on reaction conditions. This
chemistry is outlined in Scheme 2. Scheme 2 also includes
the reaction of cyclopentene 8, which led to another
pyridinium product 30, whose structure was established
from X-ray analysis.
Another item is the Wolff rearrangement that has been
mentioned as a source of the endo and exo diastereomeric
methyl esters 20a (anti) and 22ex (syn). The reaction,
however, deserved further exploration in view of the
intramolecular protonation observed in the case of the
endo-pyridyl diastereomer. For reasons to be considered
in the Interpretative Discussion section, we also studied
the reaction of the endo pyridyl diastereomer 19en in
tert-butyl alcohol and in aqueous dioxane. Interestingly,
in contrast to the methanol runs that led to a stereose-
lectivity of 89:11 in favor of the more stable epimeric
(anti) ester 20a , runs using tert-butyl alcohol led to a ratio
of 4:6 in favor of the less stable syn isomer 31c. In water,
the stereoselectivity here was 92:8 in favor of the more
stable anti epimer 20b. The results of the Wolff rear-
rangement are outlined in Scheme 3.
Ch oice of Mod es of En ol Gen er a tion . What was
required for this study were methods of generating the
enols whose ketonization stereochemistry was to be
studied. It has been noted above that the silyl enol ethers
were considered and synthesized. Thus, one of the most
promising methods involves the reaction of silyl enol
ethers with fluoride anion. This is a reaction studied
primarily by Noyori,6 who nicely demonstrated that silyl
enol ethers on reaction with the fluoride anion afford
isolable enolates. Additionally, enolates generated in this
fashion have proven useful in reacting with electrophiles.
Thus, 24en and 24ex seemed ideal precursors.
Several important results were derived from our
fluoride-acetic acid enolate generation reactions. The
first point is that without fluoride it was observed that
the silyl enol ethers, 24en and 24ex, were unreactive.
The second observation was that as the acetic acid
concentration was increased the reactions slowed. This
is attributed to acetic acid protonating the fluoride anion
and diminishing the free fluoride concentration. The
important point is that free fluoride anion is required for
the desilylation. Finally, tetrabutylammonium chloride
proved an unreactive substitute for the fluoride.
The second approach employed has already been
indicated in the synthesis described above. Thus, the
diazoketones 19en and 19ex seemed likely to be useful
precursors. Here, it is well known that the reaction
affords ketenes that then, with alcohols or water, afford
enols that lead onward to ester or acid final products.
Differ en ces w ith Differ en t Don or s a n d Over a ll
Syn -An ti Ster eoselectivities. A remarkable result we
encountered was the finding for the endo enol 5en that
the overall protonation stereochemistry changed as dif-
ferent proton donors were utilized. See Scheme 4. This
was more dramatic than the stereochemistry with exo
enol 4ex, where there were only quantitative differences
with different donors but where invariably the preferred
stereochemistry was derived from the less hindered
protonation.
(6) (a) Nayori has characterized a sulfonium enolate of 1-phenyl-2-
propanone obtained by reaction of a silyl enol ether with fluoride.
Additionally, in
a less pure form, he has similarly generated a
tetrabutylammonium enolate. (b) Nayori, R.; Nishida, I.; Sakata, J . J .
Am. Chem. Soc. 1983, 105, 1598-1608. (c) Kuwajima, I.; Nakamura,
E. Acc. Chem. Res. 1985, 18, 181-187.
(5) Zimmerman, H. E.; Wang, P. Helv. Chim. Acta 2001, 84, 1342-
1346.
9218 J . Org. Chem., Vol. 67, No. 26, 2002

Stereoselective Protonation of Enols
SCHEME 4. Ma jor r ea ction ster eoch em istr y
F IGURE 1. Stereoselectivity versus acetic acid concentration.
F IGUR E 2. Stereoselectivity versus HCl concentration for
24en .
TABLE 1. Ster eoselectivitya w ith F ou r Differ en t P r oton
Don or s w ith Selected Don or Con cen tr a tion s
proton donor
diastereomer
lecular protonation process, and m is the order in the
intermolecular counterpart. The intercept provides the
log of the ratio of the rate constants for the two modes.
Figure 1 depicts the plot of the data for acetic acid runs
in acetonitrile. The detailed data are given in Supporting
Information. The slope of the plot gives us the value of
(n - m), or the difference in the proton-donor reaction
orders. Remarkably, in the case of acetic acid, the slope
was very close to unity. This indicates that one extra
molecule of proton donor is used in the intramolecular
ketonization process compared with the intermolecular
reaction. This is to be interpreted in the Discussion
section below. We note that this dependence on acetic
acid concentration is a kinetic phenomenon and does not
result from equilibration. Control runs showed that the
thermodynamically less stable epimer 32en was not
epimerized to the more stable 26en under the conditions
used. Additionally, the linearity (see Figure 1) and fit to
eq 2 provide support for kinetic control.
P r oton a tion by F u r th er Don or s. In contrast to the
acetic acid protonation stereochemistry, fluoride keton-
ization utilizing phenol led primarily to diastereomer
32en , the result of external, less hindered protonation.
Whereas discussion of the differing stereochemistry is
delayed (vide infra), we do note that phenol is insuf-
ficiently acidic to protonate the pyridyl nitrogen. Analo-
gously, use of triethylammonium chloride led to the same
diastereomer 32en . Reasoning similar to that for the
phenol protonation can account for the result. However,
in the Discussion section, we interpret the stereochem-
istry with varying proton donors as a function of their
sizes as well as their pKa’s. The detailed data for phenol
and triethylammonium, along with that for hydrochloric
acid (vide infra), are provided in Supporting Information.
A superficially remarkable result was obtained using
hydrochloric acid as the proton donor, as seen in the plot
in Figure 2 for the endo-pyridyl enol ether 24en . Here,
the log(Pint/Pext) plot is seen to have a negative slope, the
HOAc
phenol
Et3NHCl
HCl
endo phenyl
endo pyridyl
7.3:1
1:24b
19:1
5.7:1
9:1
3:1
1:0
1:5.3c
a
Ratio of products 32ex to 26ex for the enol 4ex and 32en to
26en for the enol 5en . Ratio given for 1.4 M HOAc. c Ratio given
b
for 0.022 M HCl.
Thus, endo enol 5en , at the higher acetic acid concen-
trations, afforded a preference for the more hindered
approach (i.e., anti product configuration). Conversely,
independent of the donor concentration, phenol and also
triethylammonium chloride proceeded via the less hin-
dered protonation (i.e., syn product).
As with the acetic acid runs, protonation by hydro-
chloric acid afforded both stereochemistries, depending
on acid concentration. However, the effect of acid con-
centration was reversed. At low HCl concentration, the
lower-energy product (i.e., anti) resulted in excess, whereas
at higher HCl concentrations, the higher-energy product
(i.e., syn) predominated. These major reaction preferences
are outlined in Table 1.
Qu a n t it a t ive Ob ser va t ion s of t h e Acet ic Acid
P r oton a tion . In our previous studies4 of intra- versus
intermolecular protonation, we have noted that the ratio
int/Pext of the two reaction modes may be obtained as in
eq 2.
P
log(Pint/Pext) ) (n - m) log[HA] + log(kint/kext) (2)
Pint is the extent of product formation via a reaction that
is nth order in proton donor. This is assumed to cor-
respond to intramolecular protonation (i.e., product
26en ). Pext is the extent of product formation via an
m-order reaction and is assumed to correspond to inter-
molecular protonation (i.e., product 32en ). Linearity
provides confirmation of the assumptions. Thus, n is then
the reaction order of the proton donor in the intramo-
J . Org. Chem, Vol. 67, No. 26, 2002 9219

Zimmerman and Wang
did lead to evidence for protonation from the more
hindered face of the enol.
In ter p r eta tion sMor e Hin d er ed P r oton a tion ver -
su s th e Acetic Acid Con cen tr a tion Dep en d en ce.
With 1.4 M acetic acid, the ratio of protonation from the
more hindered enol face to protonation from the less
hindered side of the enol 5en was 24:1. Runs in THF and
DMSO showed lower stereoselectivity (9:1 in THF and
19:1 in DMSO). The more hindered protonation in the
case of the endo enol 5en results from a preference for
intramolecular delivery of the proton from the 2-pyridyl
nitrogen to the R-carbon of the enol. The endo phenyl enol
4ex proceeds with the usual3 less hindered approach of
the proton donor and serves as a comparison model.
As described in the Results section, it was quickly
ascertained that the stereoselectivity for 5en was a
function of the proton donor concentration and followed
eq 2. This contrasts with the protonation of the endo
phenyl enol 4ex, whose stereochemistry proved to be
independent of proton donor concentration. Equation 2
had been derived with the assumption that the reaction
orders for intramolecular and intermolecular protonation
may differ and are n and m, respectively, and that the
slope of the plot is n - m. Indeed, for enol 4ex, the
stereoselectivity is independent of acid concentration,
n - m is zero, and a slope of zero results.
In the case of the endo 2-pyridyl enol 5en with acetic
acid, there is a positive slope of unity, which means that
the formation of 26en involves one more acid protonating
molecule than 32en . This provides further evidence for
the endo 2-pyridyl diastereomer reacting by a mechanism
that is different from that for the endo phenyl isomer.
Hence, there are several noteworthy points. First, the
endo 2-pyridyl protonation gives different stereochemis-
try than the endo phenyl isomer. Second, the reaction
kinetics differs for the two processes. Additionally, there
is the mechanistic matter that the 2-pyridyl nitrogen is
ideally positioned to deliver a proton intramolecularly to
the more hindered side of the enol π system. Here we
have the essentials of our picture of the endo-pyridyl
mechanism.
An important point is that one cannot assume that
protonation by acetic acid involves one proton in either
the endo phenyl case or the less hindered protonation in
the endo-pyridyl reaction. Thus, it is known8 that acetic
acid not only forms a dimer, but there is also evidence
for a trimer at low temperature. Additionally, there is
evidence that the trimer is capable of protonating pyri-
dine whereas the monomer tends to only hydrogen bond;
the dimer protonates partially. For simplicity, it is
convenient to assume that external protonation is by one
acetic acid molecule (i.e., m ) 1) and that the intramo-
lecular transfer thus involves two donor acetic acid
molecule since n - m is unity. Nevertheless, all we really
know is that one more acetic acid molecule is used for
intramolecular delivery compared with the number used
for external attack. Hence, it is possible that the acetic
acid trimer, being acidic enough to protonate the pyridine
nitrogen, is the preferred species for intramolecular
F IGURE 3. Solvent effects on the diastereoselectivity of acetic
acid protonation. THF (0)and DMSO (O).
converse of the acetic acid case (cf. Figure 1). Also, in
the HCl runs, the stereoselectivities and the qualitative
rates were independent of added fluoride anion. Thus,
the tert-butyldimethylsilyl enol ether is capable of directly
protonating at the R-carbon. Again, as in the acetic acid
runs above, the process is kinetically controlled. This is
seen (a) from control runs showing the less stable
diastereomer 32en to be stable to epimerization under
the reaction conditions, (b) from the linearity in Figure
2 and the fit to the kinetics predicated on nonequilibrat-
ing conditions, and (c) from the preferential formation
of the less stable ketone 32en with the higher acid
concentrations and preferential formation of the more
stable isomer 26en when lower acid concentrations were
employed. This last result is the reverse of that expected
if one were dealing with acid-catalyzed equilibration.
In the case of the exo-pyridyl enol ether 24ex, keton-
ization proved to be independent of acid concentration
and led exclusively to the less stable ketone 32ex, the
isomer expected as a result of less hindered protonation.
Thus, there is a dramatic difference in the behavior of
the endo and exo pyridyl enol ethers.
Solven t Effects. Interestingly, the eq 2 plots for the
three solvents usedsacetonitrile, THF, and DMSOshad
remarkably similar slopes, close to unity. The experi-
mental scatter is minor and less than the variation in
the measured slopes. Figure 3 summarizes the experi-
mental data for the other two solvents, and the detailed
data are in Supporting Information.
3. In ter p r eta tive Discu ssion
Gen er a l. The first observation to be made is that the
phenomenon of intramolecular proton transfer is a real-
ity, as it has now been demonstrated in two systemss
the present one and the bicyclo[3.1.0]hexane system we
studied earlier.4 At the outset, before beginning the
rather arduous synthesis, there was no promise of
anything but steric effects being controlling.7 Indeed, with
phenol and triethylammonium chloride only the well-
established3 and ubiquitous less hindered protonation
stereochemistry resulted. Fortunately, acetic acid in
acetonitrile was used in our initial experiments and this
(7) (a) Success was had despite not following the sage advice of
Professor R. B. Woodward, who noted that it is wise to plan research
with maximum flexibility as the end rather than a single positive or
negative result at the end with no alternatives. (b) Private communica-
tion to H.E.Z.
(8) (a) Smirnov, S. N.; Golubev, N. S.; Denisov, G. S.; Bendict, H.;
Schah-Mohammdei, P.; Limbach, H.-H. J . Am. Chem. Soc. 1996, 118,
4094-4101. (b) Golubev, N. S.; Smirnov, S. N.; Gindin, V. A.; Denisov,
G. S.; Benedict, H.; Limbach, H.-H. J . Am. Chem. Soc. 1994, 116,
12055-12056.
9220 J . Org. Chem., Vol. 67, No. 26, 2002

Stereoselective Protonation of Enols
proton transfer whereas protonation of the enolic R-car-
bon does not require this enhanced acidity and the dimer
suffices.
differ in the event that the chloride anion assisted the
loss of the silyl moiety.
A final aspect of the hydrochloric ketonizations con-
cerns the endo-phenyl diastereomer, where the stereo-
selectivity is independent of acid concentration and
proceeds exclusively by a less hindered attack of the
proton donor to afford the less stable benzoyl diastere-
omer 32ex.
In ter p r eta tion sSter eoch em istr y of th e Wolff Re-
a r r a n gem en t. One item that seemed to be unexpected
was the stereochemistry of the photochemical Wolff
rearrangement of diazoketone 19en in methanol. Despite
methanol being considerably less acidic than phenol or
the triethylammonium cation, the preferred kinetic ster-
eochemistry afforded the anti epimer of diastereomer
20a , as shown in Scheme 3. The result was clearly kinetic
since runs at different times were invariant. However,
that tert-butyl alcohol runs led to an inverted ratio with
the less stable diastereomeric ester being preferred
suggests that the small size of methanol is a factor that
permits hydrogen bonding to the pyridine nitrogen and
thus intramolecular transfer.
Discu ssion sP h en ol a n d Am m on iu m Resu lts. The
observation of the common less hindered protonation by
phenol and the ammonium ion for both the endo phenyl
and the endo pyridyl diastereomers is not only striking
in view of its contrast with the acetic acid protonation
but also important in providing confirmation of the acetic
acid mechanism. Thus, phenol and triethylammonium
proton donors differ from acetic acid in being weaker
acids by 5 orders of magnitude if one uses the acidity in
water. Whereas the estimates for the solvents used are
approximate, it is clear that these acids are incapable of
completely protonating a pyridyl nitrogen although hy-
drogen bonding is possible. There is an additional factor,
namely, the greater steric bulk of these two proton
donors, that adds to diminished hydrogen bonding to the
somewhat inaccessible pyridine nitrogen in a conforma-
tion for intramolecular transfer by phenol and triethyl-
ammonium. Hence, we can understand a preference for
the unhindered protonation, perhaps modified slightly by
some intramolecular transfer involving a hydrogen-
bonded species.
Discu ssion sHyd r och lor ic Acid Rever sa l. Several
of the results with hydrochloric acid protonation are
unique and of special interest. The first is the unusual
plot in Figure 2, with the slope being the inverse of the
usual one as in Figure 1. HCl differs from the other
proton donors used in being a much stronger acid. Thus,
1 equiv is certain to be tied up irreversibly in protonating
the pyridyl group. Still, this proton is available for
intramolecular protonation. Once protonated, the pyridyl
ring no longer has the ability or is required to utilize
additional protons, and intramolecular attack becomes
independent of the HCl concentration beyond that first
proton. At this point, intramolecular protonation is zeroth
order in HCl. However, the external protonation process
does utilize protons in the medium and will be first order
in HCl. This means that the external attack is enhanced
with increasing HCl concentration whereas the intramo-
lecular process is not. This is the reverse of what occurs
with acetic acid. To paraphrase, eq 2 predicts a negative
slope. Added to this reversal of proton donor dependence
is the observation that in the more concentrated range
of runs, pyridinium exhibits behavior similar to that of
an endo phenyl group in providing steric hindrance to
protonation. Thus, the pyridyl group has two possible
rolessit may act as a site of intramolecular transfer to
the more hindered face of the enolic system at low HCl
concentrations, or fully protonated, it may act as a bulky
moiety preventing protonation from this face at these
higher HCl concentrations.
In ter p r eta tion sGeom etr y of P r oton Deliver y by
P yr id yl. A final matter is the geometry of proton delivery
by the endo-pyridyl moiety. It seems probable that
protonation initially occurs in a pyridyl conformation in
which the nitrogen is most accessible. Protonation then
is followed by a twist and intramolecular delivery as
depicted in eq 3.
4. Con clu sion s
In considering the overall significance of the present
study, we need to comment on the importance of proton
transfer in organic chemistry, both synthetic and mecha-
nistic. Thus, an understanding of the phenomenon is
critical to an understanding of the myriad reactions10
involving enolic intermediates. The present study and one
previous case4 we have presented represent a dramatic
departure from the ubiquitous phenomenon of less hin-
dered protonation to give the less stable of alternative
diastereomeric carbonyl compounds. In our earlier re-
view,3 we listed just 25 examples of kinetic protonation
via the less hindered route.11 However, a rather interest-
ing conjecture now seems possible. For those examples
where (a) less than total stereoselectivity was observed
and where (b) the molecule had non-hydrocarbon groups,
one may inquire whether some of the more hindered
protonation derives from contributions from an intramo-
lecular effect of the type described in the present study.
Another item is suggested by the present study, namely,
A second item of special interest is the observation that
fluoride is not required for ketonization with hydrochloric
acid. Protonation of the tert-butyldimethylsilyl enol ether
exhibits behavior parallel to that of the enol itself.
Ketonization of both enols and enolates, being highly
exothermic and thus having early transition states,
utilizes relatively planar transition states. In the present
case, the release of a silyl cation seems quite parallel to
the release of a proton in the simple enol examples. Silyl
cations are, indeed, known to be reasonably stable
species.9 The basic stereochemical conclusions would not
(9) Lambert, J . B.; Zhang, S.; Ciro, S. M. Organometallics 1994, 13,
2430-2443, report of a series of stable silyl cations as tetraaryl borate
salts.
(10) (a) Also, an understanding of proton transfer in carbonyl
compounds is relevant to certain biochemical phenomena (note ref 10b).
(b) Harris, T. K.; Cole, R. N.; Comer, F. I.; Mildvan, A. S. Biochemistry
1998, 37, 16828-16838.
J . Org. Chem, Vol. 67, No. 26, 2002 9221

Zimmerman and Wang
5-P h en yl-5-(p yr id in -2-yl)cyclop en t-1,3-d ien e (14). To
4-phenyl-4-pyridylcyclopentene (4.4 g, 20 mmol) dissolved in
100 mL of ether at room temperature was added 2.3 mL of
HBr (47-49%). A white precipitate formed upon the addition
of HBr. The suspension was stirred for 10 min, 1.1 mL of
bromine (21 mmol) was then added, and the reaction was
stirred at room temperature for 45 min. Ether was removed
in vacuo, and the residue was dissolved in 70 mL of CH2Cl2.
The solution was washed with NaHCO3 solution, with Na2S2O3,
dried over MgSO4, filtered, and concentrated to yield a brown
oil. The brown oil and 18-crown-6 (15.8 g, 60 mmol) dissolved
in 300 mL of THF were stirred under N2 at -78 °C, and
t-BuOK (6.85 g, 60 mmol) was added. The reaction mixture
darkened slowly and was stirred overnight with a gradual rise
of temperature to ambient. The solvent was removed in vacuo,
and the residue was redissolved in CH2Cl2. The solution was
washed with brine, dried over MgSO4, and concentrated to
yield a brown oil. The oil was subjected to chromatography on
a 5 cm × 40 cm silica gel column and eluted with CH2Cl2 to
yield 3.7 g (86%) of 5-phenyl-5-pyridyl-cyclopent-1,3-diene (14)
what other intramolecular proton transfer moieties may
successfully operate. This is a matter that we are
currently pursuing.
Exp er im en ta l Section
Gen er a l P r oced u r es. 1H NMR spectra were recorded at
250, 300, or 500 MHz. 13C NMR spectra were recorded at 75
or 125 MHz. 1H and 13C chemical shifts were measured relative
to internal Me4Si and CDCl3, respectively. Mass spectra were
obtained in EI and ESI modes. X-ray diffraction data were
collected with the use of an area detector. Each structure was
solved under the appropriate space-group symmetry by a direct
method using SHELXTL. Column chromatography (CC) was
performed on silica gel (35-75 µm or 60-200 mesh) mixed
with 1% (v/v) of a fluorescent indicator (phosphor green,
UV254), and the slurry was packed into quartz columns to
allow monitoring with a hand-held UV lamp. Column chro-
matography was also performed on alumina absorption (80-
200 mesh). Preparative thick-layer chromatography was car-
ried out with silica gel coated on glass (10-40 µm, mixed with
a fluorescent indicator (1%, v/v) and gypsum binder (10%), 50
g of silica gel, 105 mL of water, 20 cm × 20 cm plate). Melting
points (uncorrected) were determined in open capillaries with
a heating block. Exploratory photolysis was carried out with
a 450-W medium-pressure mercury lamp. Solvents were dried
following standard methods. Workup as usual (solvent) con-
sisted of extraction with the indicated solvent and drying the
organic layer over Na2SO4.
as a pale-yellow oil: IR 3058, 1583, 1489, 1464, 1435 cm-1
;
1H NMR (CDCl3, δ) 8.54 (m, 1H), 7.54 (td, J ) 7.8, 1.8 Hz,
1H), 7.31 (dt, J ) 8.7, 0.9 Hz, 1H), 7.23-7.15 (m, 5H), 7.10-
7.05 (ddd, J ) 7.5, 4.8, 1.2 Hz, 1H), 6.96 (m, 2H), 6.45 (m,
2H); 13C NMR (CDCl3, δ) 161.64, 149.13, 143.49, 141.16,
136.22, 130.15, 128.25, 127.05, 126.56, 122.00, 121.59, 70.87;
HRMS-EI m/z (M+) calcd for C16H13N: 219.1048, obsd:
219.1039.
5-Nit r o-7-p h en yl-7-(p yr id in -2-yl)b icyclo[2.2.1]h ep t -2-
en e (13). Nitroethylene was prepared by using published
procedures.12 2-Nitroethanol13 (28.0 g, 0.31 mol) and phthalic
anhydride (68.8 g, 0.46 mol) were heated at 150 °C under 80
mm of Hg until the phthalic anhydride melted. The temper-
ature was then raised to 180 °C, and 20.2 g (89%) of nitroeth-
ylene was collected as a light-green liquid that was stable for
months in the freezer or as a chloroform solution at ambient
temperature.
4-P h en yl-4-(p yr id in -2-yl)cyclop en ten e (8). Benzyl-2-pyr-
idine (38.0 g, 0.22 mol) dissolved in 1.2 L of freshly distilled
THF was stirred under nitrogen at -78 °C, and n-BuLi (70
mL, 3.5 M) was added through a long-needle syringe. The
resulting dark-red mixture was stirred for 3 h at -78 °C and
then added to 800 mL of a THF solution of 1,4-dichloro-2-
butene (37.0 g, 0.30 mol) with mechanical stirring under
nitrogen at -78 °C through a cannula over 30 min. After 2.5
h of stirring at -78 °C, the reaction mixture was treated with
another equiv of n-BuLi (70 mL, 3.5 M) and stirred for 6 h
more and allowed to warm slowly to room temperature.
Workup as usual (ether) gave a dark oil. The oil was refluxed
in 1.5 L of decalin for 36 h. Decalin was then removed by
vacuum distillation. The residue was subjected to chromatog-
raphy on a 10 cm × 50 cm silica gel column and was eluted
with hexanes-ether (3:1). The collected fraction was decol-
orized with charcoal-silica gel (1:1) and concentrated to yield
a light-yellow solid, which was recrystallized from its hexane
solution to yield 24 g (50%) of 4-phenyl-4-(2′-pyridyl)cyclopen-
tene (8) as colorless crystals: mp 53 °C; Rf 0.5 (hexanes-ether
To 5-phenyl-5-(2′-pyridyl)cyclopentadiene (14) (4.0 g, 18.3
mmol) dissolved in 100 mL of ether was added 20 mL of HCl
(1.0 M in ether) at room temperature. A white precipitate
formed upon the addition of HCl. Ether was removed in vacuo,
and the residue was washed with ether (2 × 20 mL). The
residue that dissolved in 1.0 mL of dichloromethane was
stirred under nitrogen at room temperature, and nitroethylene
(2.7 g, 37.4 mmol) was added through a syringe. The reaction
flask was then sealed and stirred for 72 h. The thick brown
oil that formed was diluted with dichloromethane. The organic
layer was washed with NaHCO3 solution, dried, and con-
centrated in vacuo to yield a dark oil that was chroma-
tographed on a 5 cm × 40 cm silica gel column and eluted
with dichloromethane. The first fraction was 2.4 g (45%) of
5-nitro-7-exo-phenyl-7-endo-(pyridin-2-yl) bicyclo[2.2.1]hept-2-
ene (13en ): Rf 0.27 (CH2Cl2); mp 125 °C; IR 3063, 3003, 1385,
1359 (strong), 1373 cm-1; 1H NMR (CDCl3, δ) 8.52 (d, J ) 5.0
Hz, 1H), 7.55 (td, J ) 8.0, 2.0 Hz, 1H), 7.30-7.00 (m, 7H),
6.40 (dd, J ) 5.7, 3.3 Hz, 1H), 5.95 (dd, J ) 5.7, 2.7 Hz, 1H),
5.12 (dt, J ) 8.5, 3.5 Hz, 1H), 4.61 (m, 1H), 3.90 (m, 1H), 2.20
(ddd, J ) 14.0, 8.5, 3.5 Hz, 1H), 2.05(dd, J ) 14.0, 3.5 Hz,
1H); 13C NMR (CDCl3, δ) 162.1, 149.4, 142.3, 138.5, 136.9,
130.3, 128.2, 127.6, 126.1, 121.6, 121.3, 84.3, 53.1, 48.2, 29.9;
HRMS-EI m/z (M+) calcd for C18H15N2O2: 292.1212, obsd:
292.1217. The second fraction was exo-pyridyl adduct 13ex
(2.0 g, 38%): Rf 0.20 (CH2Cl2); mp 109.5-110.5 °C; ΙR
3:1); IR 3055, 2924, 2850, 1587, 1567, 1492, 1468, 1427 cm-1
;
1H NMR (CDCl3, δ) 8.54 (m, 1H), 7.53 (m, 1H), 7.25 (m, 4H),
7.16 (m, 2H), 7.08 (m, 1H), 5.81 (s, 2H), 3.43 (d, J ) 14.0 Hz,
2H), 3.04 (d, J ) 14.0 Hz, 2H); 13C NMR (CDCl3, δ) 167.3,
149.1, 148.1, 135.9, 129.2, 128.0, 126.8, 125.6, 122.4, 120.7,
57.6, 45.1; HRMS-EI m/z (M+) calcd for C16H15N: 221.1204,
obsd: 221.1197.
(11) (a) This seems to be increasingly of interest. Perhaps it is not
remarkable that our 1955 concept has been rediscovered twice (refs
11b-h) and that the rediscoveries are often cited and the original work
overlooked (e.g., note ref 11i). (b) Gerlach, U.; Haubenreich, T.; Hu¨nig,
S.; Keita, Y. Chem. Ber. 1993, 126, 1205-1215. (c) Takano, S.; Kudo,
J .; Takahashi, N.; Ogasawara, K. Tetrahedron Lett. 1986, 2405-2408.
(d) Takano, S.; Uchida, W.; Hatakeyama, S.; Ogaswara, K. Chem. Lett.
1982, 733-737. (e) Takano, S.; Goto, E.; Ogasawara, K. Tetrahedron
Lett. 1982, 5567-5569. (f) Takano, S.; Yamada, S.; Numata, H;
Hatakeyama, S.; Ogasawara, K. Heterocycles 1983, 20, 2159-2162. (g)
Takano, S.; Tananka, M.; Seo, M.; Hirama, M.; Ogasawara, K. J . Org.
Chem. 1985, 50, 931-936. (h) Note also Hu¨nig, S.; Houben-Weyl,
Protonation of Carbanions and Polar Double Bonds. In Methods of
Organic Chemistry; Helmchen, G., Ed.; Stuttgart, 1996; Vol. E21D 7,
pp 3851-3911, where the origin of the concept is not given while being
the main subject of the review. (i) Krause, N.; Ebert, S.; Haubrich, A.
Liebigs Ann/Recl 1997, 2409-2418.
3061,3004,1587, 1540 (strong), 1373 cm-1 1H NMR (CDCl3,
;
δ) 8.39 (d, J ) 5.0 Hz, 1H), 7.50 (m, 3H), 7.30 (t, J ) 7.2 Hz,
2H), 7.15 (m, 1H), 6.38 (dd, J ) 6.0, 3.5 Hz, 1H), 5.93 (dd, J )
6.0, 3.0 Hz, 1H), 5.00 (dt, J ) 8.5, 3.5 Hz, 1H), 4.65 (s, 1H),
4.05 (s, 1H), 2.35 (ddd, J ) 13.5, 8.5,3.5 Hz, 1H), 2.04 (dd, J )
13.5, 3.5 Hz, 1H); 13C NMR (CDCl3, δ) 162.8, 148.7, 141.5,
(12) Ranganathan, D.; Rao, C. B.; Ranganenthan, S.; Mehrotra, A.
K.; Iyengar, R. J . Org. Chem. 1980, 45, 1185-1189.
(13) Noland, W. E. Org. Synth. 1961, 41, 67-72.
9222 J . Org. Chem., Vol. 67, No. 26, 2002

Stereoselective Protonation of Enols
138.9, 136.0, 130.5, 129.0, 127.0, 126.4, 121.2, 120.8, 84.2, 52.9,
48.0, 29.8; HRMS-EI m/z (M+) calcd for C18H16N2O2: 292.1212,
obsd: 292.1208. The stereochemistry was assigned by an NOE
experiment.
concentrated. The residue was subjected to chromatography
on a 4 cm × 30 cm silica gel column and eluted with hexane-
dichloromethane-diethyl ether (2:2:1) to give 0.53 g (53%) of
15en .
7-exo-P h en yl-7-en d o-(p yr id in -2-yl)bicyclo[2.2.1]h ep t-5-
en -2-on e (11en ). Meth od A. 5-Nitro-7-exo-phenyl-7-endo-
(pyridin-2-yl)bicyclo[2.2.1]hept-2-ene (13en ) (1.0 g, 3.4 mmol)
mixed with 50.0 g of activated silica gel14 (8.5 mol NaOMe/kg
silica gel) was heated at 150 °C for 2.5 h, and the solid mixture
turned brown. The mixture was cooled to room temperature
and washed with dichloromethane. The dichloromethane solu-
tion was concentrated in vacuo, and the residue was subjected
to chromatography on a 2 cm × 30 cm silica gel column and
eluted with dichloromethane-diethyl ether (9:1) to yield 0.42
g (47%) of 7-exo-phenyl-7-endo-(pyridin-2-yl)bicyclo[2.2.1]hept-
5-en-2-one (11en ).
Meth od B. To 7-exo-phenyl-7-endo-(pyridin-2-yl)bicyclo-
[2.2.1]hept-5-en-2-one (11en ) (1.49 g, 5.69 mmol) in 50 mL of
MeOH cooled in an ice-water bath was added potassium
azodiformate (3.30 g, 17.07 mmol). Ten milliliters of a MeOH
solution of acetic acid (2.3 mL, 40.0 mmol) was added over 5
h. The reaction mixture was stirred for an additional 2 h, more
potassium azodiformate (2.0 g) was then added, and 5.0 mL
of a MeOH solution of acetic acid (1.5 mL) was added over 4.5
h. The reaction mixture was poured into 60 mL of water and
worked up as usual (dichloromethane). The residue was
subjected to chromatography on a 4 cm × 30 cm silica gel
column, and 0.70 g of 15en was obtained along with 0.30 g of
recovered unreacted starting material (11en ). The column was
washed with MeOH-diethyl ether (1:1) to give ca. 0.50 g of a
complex mixture that was heated in 50 mL of 0.8 M HCl at
80 °C for 3 h, and 15en was obtained as a single product. The
combined 15en fractions weighed 1.17 g (78%). Rf 0.32
(dichloromethane-diethyl ether 9:1); mp 157-159 °C; IR 1744
cm-1; 1H NMR (CDCl3, δ) 8.45 (d, J ) 5.0 Hz, 1 H), 7.48-7.12
(m, 7H), 6.94 (ddd, J ) 7.5, 5.0, 1.0 Hz, 1H), 3.64 (d, J ) 4.0
Hz, 1H), 3.58 (m, 1H), 2.12-1.80 (m, 4H), 1.62-1.46 (m, 2H);
13C NMR (CDCl3, δ) 215.9, 162.2, 149.0, 141.3, 136.33, 128.5,
126.9, 126.4, 122.4, 121.2, 65.2, 56.2, 44.2, 41.9, 26.4, 22.3;
HRMS-EI m/z (M+) calcd for C18H17NO: 263.1310, obsd:
263.1311.
7-en d o-P h en yl-7-exo-(p yr id in -2-yl)bicyclo[2.2.1]h ep t-2-
on e (15ex). 5-Nitro-7-endo-phenyl-7-exo-(pyridin-2-yl)bicyclo-
[2.2.1]hept-2-ene (13ex) (2.0 g, 6.8 mmol) and 4.0 g of (20.6
mmol) potassium azodiformate were stirred in 70 mL of MeOH
in an ice bath. Five milliliters of a of MeOH solution of acetic
acid (2.74 mL, 48 mmol) was added with a syringe pump over
3.0 h and then stirred for an additional 3 h. The reaction
mixture was added to 60 mL of diluted NaHCO3, and workup
as usual (dichloromethane) gave 1.98 g (98.5%) of 5-nitro-7-
endo-phenyl-7-exo-(pyridin-2-yl)bicyclo[2.2.1]heptane (16ex):
mp 144 °C; IR 1542 cm-1; 1H NMR (CDCl3, δ) 8.50 (d, J ) 5.0
Hz, 1H), 7.60-7.50 (m, 3H), 7.35 (m, 3H), 7.15 (m, 1H), 7.00
(ddd, J ) 7.5, 5.0, 1.0 Hz, 1H), 4.80 (m, 1H), 3.95 (s, 1H), 3.35
(s, 1H), 2.35-2.10 (m, 2H), 1.80-1.30 (m, 4H); 13C NMR
(CDCl3, δ) 162.3, 149.2, 141.6, 136.5, 128.9, 127.2, 126.7, 121.4,
121.0, 86.3, 67.8, 48.1, 41.6, 31.5, 27.2, 21.3; HRMS-EI m/z
(M+) calcd for C18H15N2O2: 294.1368, obsd: 294.1378.
16ex (1.98 g) was treated with 100 mL of 10% KOH(aq),
stirred for 36 h, and then treated with 140 mL of 4.0 M HCl.
The solution became blue immediately and was stirred for 1
h at room temperature and then at 90 °C for 2.5 h. The reaction
mixture was cooled, neutralized with NaHCO3 to pH of ca.
7-8, and worked up as usual (dichloromethane). The residue
was subjected to chromatography on a 4 cm × 30 cm silica gel
column and eluted with hexane-dichloromethane-ether (2:
2:1) to yield 15ex, which was crystallized from ether-hexane
solution to give 1.40 g (78%) of 15ex as colorless crystals: Rf
0.33 (hexane-dichloromethane-diethyl ether 2:2:1); mp 196
°C; IR 1732 cm-1; 1H NMR (CDCl3, δ) 8.54 (d, J ) 5.0 Hz, 1H),
7.55 (td, J ) 8.0, 1.8 Hz, 1H), 7.49-7.01 (m, 7H), 3.64 (m, 2H),
2.20-1.50 (m, 6H); 13C NMR (CDCl3, δ) 216.2, 162.0, 149.4,
141.9, 136.5, 128.7, 127.7, 126.7, 121.63, 121.2, 65.0, 56.4, 44.2,
42.0, 26.5, 22.6; HRMS-EI m/z (M+) calcd for C18H17NO:
263.1310, obsd: 263.1310.
Meth od B. 5-Nitro-7-exo-phenyl-7-endo-(pyridin-2-yl)bicyclo-
[2.2.1]hept-2-ene (13en ) (96 mg, 0.33 mmol) dissolved in 50
mL of MeOH was poured onto a 2 cm × 30 cm silica gel
column. The column stood for 3 weeks, and then the material
on the column was washed out with MeOH and concentrated
in vacuo to yield the corresponding oxime 12en : IR 3230.2
cm-1 1H NMR (CDCl3, δ) 10.05 (s, 1H), 8.45 (d, 1H), 7.45-
;
6.80 (m, 8H), 6.35 (m, 1H), 6.00 (m, 1H), 4.30 (s, 1H), 4.10 (s,
1H), 2.35-2.10 (m, 2H); 13C NMR (CDCl3, δ) 163.2, 161.7,
149.1, 142.1, 138.7, 136.5, 130.9, 127.8, 125.8, 122.1, 121.1,
75.8, 65.6, 53.7, 47.3, 30.0, 15.0; HRMS-ESI m/z (MH+) calcd
for C18H17N2O: 277.1341, obsd: 277.1329.
Oxime 12en dissolved in 10 mL of HCl (0.8 M) was heated
at 90 °C for 1.5 h. The aqueous solution was then cooled to
room temperature and neutralized with NaHCO3 powder.
Workup as usual (dichloromethane) gave a brown oil that was
chromatographed on a 2 cm × 30 cm silica gel column. Elution
with dichloromethane-diethyl ether (9:1) gave 75 mg (88%)
of 11en : Rf 0.43; mp 163-164 °C; IR 1741 cm-1 1H NMR
;
(CDCl3, δ) 8.53 (m, 1H), 7.52 (td, J ) 7.8, 1.8 Hz, 1H), 7.35-
7.00 (m, 7H), 6.60 (dd, J ) 5.4, 2.7 Hz, 1H), 5.98 (t, J ) 2.7
Hz, 1H), 4.08 (s, 1H), 4.04 (m, 1H), 1.95 (d, J ) 16.5 Hz, 1H),
1.87 (dd, J ) 16.5, 3.0 Hz, 1H); 13C NMR (CDCl3, δ) 213.1,
162.0, 149.4, 142.8, 141.6, 136.7, 128.4, 128.2, 128.0, 126.4,
122.2, 121.7, 63.1, 47.0, 36.2; HRMS-EI m/z (M+) calcd for
C
18H15NO: 261.1154, obsd: 261.1148.
7-exo-P h en yl-7-en d o-(p yr id in -2-yl)bicyclo[2.2.1]h ep t-2-
on e (15en ). Meth od A. To 5-nitro-7-exo-phenyl-7-endo-(pyr-
idin-2-yl)bicyclo[2.2.1]hept-2-ene (13en ) (1.13 g, 3.87 mmol)
and potassium azodiformate (2.0 g, 10.0 mmol) stirred in 40
mL of MeOH in an ice bath was added 5.0 mL of a MeOH
solution of acetic acid (1.4 mL, 24.0 mmol) over 2 h with a
syringe pump. The reaction mixture was stirred for 90 min
more and then poured into 50 mL of water and neutralized
with NaHCO3. Workup as usual (dichloromethane) gave 1.16
g of 5-nitro-7-exo-phenyl-7-endo-(pyridin-2-yl)bicyclo[2.2.1]-
heptane 16en : mp 115 °C; IR 1542 cm-1; 1H NMR (CDCl3, δ)
8.51 (ddd, J ) 5.0, 1.5, 1.0 Hz, 1H), 7.54 (td, J ) 7.5, 2.0 Hz,
1H), 7.47-7.41 (m, 2H), 7.35 (dt, J ) 8.0, 1.0 Hz, 1H), 7.26 (t,
J ) 8.0 Hz, 2H), 7.14 (tt, J ) 8.0, 1.0 Hz, 1H), 7.03 (ddd, J )
7.5, 5.0, 1.5 Hz, 1H), 4.94 (dtd, J ) 8.5,4.0,2.0 Hz, 1H), 3.97
(t, J ) 4.0 Hz, 1H), 3.31 (t, J ) 4.0 Hz, 1H), 2.32 (dd, J ) 9.0,
4.0 Hz, 1H), 2.01 (qt, J ) 11.0, 4.0 Hz, 1H), 1.83 (m, 1H), 1.71-
1.50 (m, 2H), 1.41-1.30 (m, 1H); 13C NMR (CDCl3, δ) 162.4,
149.4, 142.7, 128.6, 127.0, 126.4, 122.0, 121.3, 86.4, 68.1, 48.1,
41.9, 31.9, 27.1, 21.3; HRMS-EI m/z (M+) calcd for C18H15N2-
O2: 294.1368, obsd: 294.1358.
16en (1.16 g) was stirred with 70 mL of 10% KOH(aq) for
48 h. The solution turned blue upon the addition of 98 mL of
HCl (4.0 M). The solution was heated in an oil bath (100-105
°C) for 50 h, and the color changed slowly from blue to green
and finally to golden green. The solution was cooled, neutral-
ized with NaHCO3 to a pH of ca. 7-8, and extracted with
dichloromethane. The combined organic layers were dried and
3-Dia zo-7-en d o-p h en yl-7-exo-p yr id in -2-yl-bicyclo[2.2.1]-
h ep ta n -2-on e (19ex). To a solution of diisopropylamine (0.84
mL, 5.5 mmol) in 2.5 mL of freshly distilled THF under
nitrogen at -78 °C was added 3.6 mL of n-BuLi (1.45 M, 5.2
mmol). The reaction was stirred for 45 min at -78 °C. 15ex
(1.24 g, 4.7 mmol) dissolved in 15 mL of THF was cooled to
-78 °C under nitrogen, and LDA was added. The reaction
mixture was allowed to warm to room temperature slowly, and
5.0 mL of solvent was removed in a nitrogen stream. HCOOMe
(14) Keinan, E.; Mazur, Y. J . Am. Chem. Soc. 1977, 99, 11-12.
J . Org. Chem, Vol. 67, No. 26, 2002 9223

Zimmerman and Wang
(2.0 mL, 12.0 mmol) was then added, and the reaction mixture
was stirred for 48 h. A yellow cake formed, and the reaction
mixture was concentrated in vacuo. Dichloromethane was
added to the residue, and the solution was extracted with 10%
KOH. The organic layer was dried and concentrated to yield
0.64 g of starting material 15ex. The combined aqueous layers
were then neutralized with 4.0 M HCl to a pH of ca. 6, and
workup as usual (dichloromethane) gave a mixture of isomers
of formyl ketones (0.67 g) as a brown solid. The brown solid
was dissolved in 15 mL of dichloromethane to give a dark
purple solution that was cooled in an ice bath. Tosyl azide (0.47
g, 2.4 mmol) was added, followed by triethylamine (0.48 g, 4.8
mmol). The reaction mixture was stirred overnight in dark-
ness, the solvent was then removed, and the residue was
subjected to chromatography on a 2 cm × 30 cm neutral
alumina column and eluted with dichloromethane-hexane (2:
1) to give 0.44 g of 19ex (67% based on consumed 15ex) as
bright-yellow crystals: Rf (silica gel TLC) 0.30 (dichloro-
methane-ether 9:1); IR 2076, 1684 cm-1; 1H NMR (CDCl3, δ)
8.55(d, J ) 4.5 Hz, 1H), 7.57 (td, J ) 8.0, 2.0 Hz, 1H), 7.50-
7.45 (m, 2H), 7.28-7.10 (m, 4H), 7.07 (ddd, J ) 7.5, 5.0, 1.0
Hz, 1H), 4.46 (t, J ) 2.5 Hz, 1H), 3.68 (dd, J ) 4.0, 2.0 Hz,
1H), 2.10-1.54 (m, 4H); 13C NMR (CDCl3, δ) 198.9, 160.9,
149.4, 141.6, 136.7, 127.4, 126.8, 126.5, 121.5, 121.3, 66.7, 55.9,
7.49 (td, J ) 7.5, 2.0 Hz, 1H), 7.21 (t, J ) 7.5 Hz, 2H), 7.09 (t,
J ) 7.5 Hz, 2H), 6.94 (ddd, J ) 7.5, 5.0, 1.0 Hz, 1H), 3.78 (s,
2 H), 3.06 (s, 3H), 2.31 (s, 1H), 1.81-1.65 (m, 4H); 13C NMR
(CDCl3, δ) 172.8, 164.2, 149.2, 141.4, 136.1, 129.5, 127.6, 126.5,
121.2, 120.4, 64.4, 52.5, 50.8, 48.0, 25.4; HRMS-EI (M+) (m/z)
for C19H19NO2 calcd 293.1416; obsd 293.1415.
Meth yl 6-exo-P h en yl-6-en d o-p yr id in -2-yl-bicyclo[2.1.1]-
h exa n e-5-ca r boxyla te (20en ). 3-Diazo-7-exo-phenyl-7-endo-
pyridin-2-yl-bicyclo[2.2.1]heptan-2-one (19en ) (84 mg, 0.29
mmol) dissolved in 200 mL of MeOH was irradiated for 40 min
in a Hanovia box with a CuSO4 filter solution. The reaction
mixture was concentrated in vacuo to give a light-brown solid
that was subjected to chromatography on a 2 cm × 30 cm silica
gel column and eluted with diethyl ether to afford 66 mg (78%)
of 20en : Rf 0.73 (hexane-dichloromethane-diethyl ether 2:2:
1); mp 103-104 °C; IR 1732 cm-1 1H NMR (CDCl3, δ) 8.51
;
(m, 1H), 7.53 (td, J ) 7.5, 2.0 Hz, 1H), 7.44 (dt, J ) 7.5, 1.0
Hz, 1H), 7.28-7.19 (m, 4H), 7.10 (m, 1H), 7.01 (ddd, J ) 7.5,
5, 1.5 Hz, 1H), 3.68 (d, J ) 2.5 Hz, 2H), 3.61 (s, 3H), 2.63 (m,
1H), 1.78 (m, 2H), 1.62 (m, 2H); 13C NMR (δ) 172.4, 163.3,
149.3, 141.8, 136.3, 128.2, 127.2, 125.9, 122.4, 121.0, 61.5, 51.2,
48.1, 46.3, 26.4; HRMS-EI (M+) (m/z) for C19H19NO2 calcd
293.1416; obsd 293.1418.
6-P h en yl-6-p yr id in -2-yl-bicyclo[2.1.1]h exa n e-5-ca r ba l-
d eh yd e (23en ). Methyl 6-exo-phenyl-6-endo-pyridin-2-yl-
bicyclo[2.1.1]hexane-5-carboxylate (20en ) (199 mg, 0.68 mmol)
dissolved in 5.0 mL of dry toluene at -78 °C under nitrogen
was slowly added to 1.0 mL of DIBAL-H (1.0 M in hexanes).
The reaction mixture was stirred for 2 h and quenched with 4
mL of aqueous MeOH. Workup as usual (dichloromethane)
gave an oil. The oil was subjected to chromatography on a 2
cm × 30 cm silica gel column and eluted with diethyl ether to
yield 129 mg (80%) of 23en as a clear oil: Rf 0.63 (diethyl
ether); IR 1714 cm-1; 1H NMR (CDCl3, δ) 9.62 (d, J ) 1.0 Hz,
1H), 8.50 (ddd, J ) 5.0, 2.0, 1.0 Hz, 1H), 7.53 (td, J ) 8.0, 2.0
Hz, 1H), 7.43 (dt, J ) 8.0, 1.0 Hz, 1H), 7.29-7.18 (m, 4H),
7.15-7.08 (m, 1H), 7.04-6.99 (ddd, J ) 7.5, 5.0, 1.0 Hz, 1H),
46.9, 27.6, 23.0; HRMS-EI m/z ([M - N2]+) calcd for C18H15
-
NO: 261.1154, obsd: 261.1151.
3-Dia zo-7-exo-p h en yl-7-en d o-p yr id in -2-yl-bicyclo[2.2.1]-
h ep ta n -2-on e (19en ). To a solution of diisopropylamine (0.82
mL, 5.3 mmol) in 1.8 mL of freshly distilled THF under
nitrogen at -78 °C was added 3.9 mL of n-BuLi (1.33 M, 5.2
mmol). The mixture was stirred at -78 °C for 1 h. 7-exo-
Phenyl-7-endo-(pyridin-2-yl) bicyclo[2.2.1]hept-2-one (15en )
(1.17 g, 4.43 mmol), dissolved in 8.0 mL of THF was cooled to
-78 °C under nitrogen, and LDA was added. The reaction was
stirred overnight and allowed to warm slowly to room tem-
perature. HCOOMe (1.4 mL, 8.4 mmol) was then added, and
a yellow suspension formed. The reaction mixture was stirred
for 48 h and concentrated in vacuo. Diethyl ether was added,
and mixture was extracted with 5% KOH. The organic layer
was dried over MgSO4 and concentrated to yield 0.60 g of
starting material 15en . The recovered starting material was
subjected to the above procedure again. The combined aqueous
layer was then neutralized with 4.0 M HCl to a pH of ca. 6,
and workup as usual (dichloromethane) gave a mixture of
isomers of formyl ketones (1.01 g) as a brown solid. The solid
was dissolved in 20 mL of dichloromethane to give a dark-
purple solution. To the solution cooled in an ice bath was added
tosyl azide (0.686 g, 3.5 mmol), followed by triethylamine
(0.703 g, 7.0 mmol). The reaction mixture was kept dark and
stirred and then concentrated in vacuo. The residue was
subjected to chromatography on a 2 cm × 30 cm neutral
alumina column and eluted with dichloromethane-hexane (2:
1) to yield 0.65 g (65%) of 19en as a bright-yellow solid: Rf
(silica gel TLC) 0.33 (dichloromethane-ether 9:1); IR 2076,
3.78 (d, J ) 2.5 Hz, 1H), 2.47 (m, 1H), 1.80-1.65 (m, 4H); 13
C
NMR (δ) 202.7, 163.0, 149.2, 141.1, 136.3, 128.3, 127.0, 126.0,
122.4, 121.1, 62.0, 52.3, 47.6, 21.8; HRMS-ESI [MH + CH3-
OH]+ (m/z) for C19H22NO2 calcd 296.1650; obsd 296.1638.
6-Phenyl-6-pyridin-2-yl-bicyclo[2.1.1]hexane-5-carboxylic Acid
(21ex). To a suspension of excess KH in THF under nitrogen
was added a THF solution of methyl 6-endo-phenyl-6-exo-
pyridin-2-yl-bicyclo[2.1.1]-hexane-5-carboxylate (22ex) (35 mg,
0.12 mmol) and 18-crown-6 (63 mg, 0.239 mmol). The reaction
mixture was stirred overnight at room temperature and was
quenched with MeOH, neutralized with AcOH, and concen-
trated in vacuo. The residue was treated with water and
dichloromethane. Workup as usual gave the crude product. A
second run was carried out with 80 mg (0.273 mmol) of 22ex
under the same conditions. The combined crude products were
subjected to chromatography on a 2 cm × 30 cm silica gel
column and eluted with diethyl ether to yield 101 mg (92%) of
21ex: mp 237-238 °C; IR 1708 cm-1; 1H NMR (CDCl3, δ) 9.90
(s, 1H broad), 8.70 (dd, J ) 5.0, 1.0 Hz, 1H), 7.59 (m, 3H),
7.27 (t, J ) 7.5 Hz, 2H), 7.15 (m, 2H), 7.07 (ddd, J ) 7.5, 5.0,
1.0 Hz, 1H), 3.70 (d, J ) 3.0 Hz, 2H), 2.91(s, 1H), 1.95 (d, J )
8.0 Hz, 2H), 1.59 (d, J ) 8.0 Hz, 2H); 13C NMR (δ) 176.2, 161.8,
148.2, 141.8, 137.4, 128.6, 128.1, 126.6, 122.5, 121.2, 60.3, 48.1,
46.2, 22.4; HRMS-ESI [M - H]+ (m/z) for C18H16NO2 calcd
278.1181; obsd 278.1189.
P h en yl-(6-en do-ph en yl-6-exo-pyr idin -2-yl-bicyclo[2.1.1]-
h ex-5-yl)-m eth a n on e (26ex). To a solution of 6-phenyl-6-
pyridin-2-yl-bicyclo[2.1.1]hexane-5-carboxylic acid (21ex) (122
mg, 0.44 mmol) in 7.0 mL of dry diethyl ether under nitrogen
at room temperature was added 0.92 mL of PhLi (1.0 M, 0.92
mmol). After 10 min, the reaction was quenched with aqueous
AcOH and extracted with chloroform. The concentrated residue
was subjected to chromatography on a 2 cm × 30 cm silica gel
column and eluted with dichloromethane-diethyl ether (9:1)
to yield 60 mg (40%) of 26ex, which was recrystallized from
1
1686 cm-1; H NMR (CDCl3, δ) 8.48 (d, J ) 5.0 Hz, 1H), 7.50
(td, J ) 8.0, 2.0 Hz, 1H), 7.44-7.15 (m, 6H), 6.99 (ddd, J )
7.5, 5.0, 1.0 Hz, 1H), 4.52 (m, 1H), 3.69 (m, 1H), 2.06-1.70
(m, 4H); 13C NMR (CDCl3, δ) 198.8, 161.9, 149.4, 140.4, 136.6,
128.8, 126.9, 126.5, 122.0, 121.5, 67.3, 55.81, 46.6, 27.2, 23.1;
HRMS-EI m/z ([M - N2]+) calcd for C18H15NO: 261.1154,
obsd: 261.1135.
Meth yl 6-en d o-P h en yl-6-exo-p yr id in -2-yl-bicyclo[2.1.1]-
h exa n e-5-ca r boxyla te (22ex). 3-Diazo-7-endo-phenyl-7-exo-
pyridin-2-yl-bicyclo[2.2.1]heptan-2-one (19ex) (98 mg, 0.31
mmol) dissolved in 200 mL of MeOH was irradiated for 40 min
in a Hanovia box with a CuSO4 filter solution. The reaction
mixture was concentrated in vacuo to give a light-brown
residue that was subjected to chromatography on a 2 cm × 30
cm silica gel column and eluted with hexane-dichloromethane-
diethyl ether (2:2:1) to yield 22ex (71%): Rf 0.15 (hexane-
dichloromethane-diethyl ether 2:2:1); mp 152-153 °C; IR
1
1723 cm-1
;
H NMR (CDCl, δ3) 8.46 (m, 1H), 7.61 (m, 2H),
9224 J . Org. Chem., Vol. 67, No. 26, 2002

Stereoselective Protonation of Enols
dichloromethane: mp 240 °C; IR 1667 cm-1; 1H NMR (CDCl3,
δ) 8.50 (d, J ) 4.5 Hz, 1H), 7.85 (m, 2H), 7.71 (m, 2H), 7.55-
7.09 (m, 8H), 6.98 (ddd, J ) 7.5, 5.0, 1.0 Hz, 1H), 3.81 (d, J )
2.5 Hz, 2H), 3.32 (s, 1H), 1.77-1.70 (dd, J ) 12.0, 5.0 Hz, 2H),
1.62-1.54 (dd, J ) 12.0, 5.0 Hz, 2H); 13C NMR (δ) 200.5, 162.4,
149.3, 142.6, 136.2, 136.1, 132.9, 128.7, 128.5, 128.0, 127.9,
126.5, 121.9, 120.6, 61.2, 50.7, 49.5, 22.1; HRMS-ESI [MNa]+
(m/z) for C24H21NONa calcd 362.1521; obsd 362.1533. The
structure was confirmed by X-ray analysis.
1H), 7.54-7.38 (m, 5H), 7.28-7.21 (m, 2H), 7.17 (td, J ) 7.0,
1.0 Hz, 1H), 7.08 (dt, J ) 7.5, 1.0 Hz, 1H), 6.98-6.81 (m, 3H),
6.48 (td, J ) 7.0, 1.0 Hz, 1H), 4.26 (dt, J ) 6.0, 1.0 Hz, 1H),
4.20 (dt, J ) 6.0, 1.0 Hz, 1H), 2.35 (m, 1H), 2.27 (m, 1H), 1.98
(m, 1H), 1.86 (m, 1H); 13C NMR (δ) 191.4, 164.1, 149.1, 139.0,
136.4, 134.7, 132.2, 130.8, 129.3, 128.9, 128.3, 127.6, 127.4,
126.8, 121.5, 120.8, 73.9, 60.8, 54.2, 54.0, 25.95, 24.7; HRMS-
ESI [MH]+ (m/z) for C24H21BrNO calcd 418.0806; obsd 418.0798.
ter t-Bu tyldim eth ylsilyloxy En ol Eth er of 26-exo (24ex).
(5-Bromo-6-endo-phenyl-6-exo-pyridin-2-yl-bicyclo-[2.1.1]hex-
5-yl)-phenyl-methanone (25ex) (78 mg, 0.19 mmol) dissolved
in 4.0 mL of THF under nitrogen was cooled at -78 °C. HMPA
(0.24 mL, 1.20 mmol) was added, followed by 0.53 mL of t-BuLi
(1.16 M, 0.61 mmol) and 1.0 mL of a THF solution of TBS
chloride (189 mg, 1.20 mmol). The reaction was stirred
overnight and allowed to warm slowly to room temperature.
The solvent was removed in vacuo, and the residue was
subjected to chromatography on a 2 cm × 30 cm silica gel
column. Elution with hexane-dichloromethane-diethyl ether
(2:2:1) gave 82 mg (97%) of 24ex as a colorless oil: Rf 0.33
(hexane-dichloromethane-diethyl ether (2:2:1); IR, no car-
P h en yl-(6-exo-ph en yl-6-en do-pyr idin -2-yl-bicyclo[2.1.1]-
h ex-5-yl)-m eth a n on e (26en ). Meth od A. 3-Diazo-7-exo-
phenyl-7-endo-pyridin-2-yl-bicyclo[2.2.1]heptan-2-one (19en )
(234 mg, 0.81 mmol) was irradiated in 250 mL of dioxane-
water (1:1) for 135 min. The reaction mixture was concentrated
in vacuo and then diluted with dichloromethane. The organic
solution was extracted with 5% KOH (5 × 10 mL), and the
combined aqueous layers were neutralized with HCl and then
acidified with AcOH. Workup as usual (chloroform) gave 173
mg (76%) of 21en : IR 1708.2 cm-1; 1H NMR (CDCl3, δ) 10.05
(s, broad, 1H), 8.58 (d, J ) 5.0 Hz, 1H), 7.55 (m, 2H), 7.30-
7.20 (m, 4H), 7.15-7.00 (m, 2H), 3.72 (d, J ) 2.5 Hz, 2H), 2.66
(s, 1H), 1.89 (d, J ) 8.0 Hz, 2H), 1.64 (d, J ) 8.0 Hz, 2H); 13
C
1
bonyl absorption; H NMR (CDCl3, δ) 8.54 (ddd, J ) 5.0, 2.0,
NMR (δ) 176.2, 162.8, 148.6, 141.3, 137.2, 128.3, 127.2, 126.0,
122.8, 121.4, 61.2, 48.0, 46.4, 22.3; HRMS-ESI [M - H]+ (m/z)
for C18H16NO2 calcd 278.1181; obsd 278.1174.
1.0 Hz, 1H), 7.53 (td, J ) 7.5, 2.0 Hz, 1H), 7.46-7.40 (m, 2H),
7.24-6.98 (m, 10H), 3.91 (d, J ) 6.0 Hz, 1H), 3.74 (d, J ) 6.0
Hz, 1H), 1.91-1.64 (m, 4H), 0.90 (s, 9H), -0.13 (s, 3H), -0.25
(s, 3H); 13C NMR (δ) 162.8, 149.3, 143.3, 138.6, 136.1, 128.3,
127.7, 127.6, 127.2, 126.7, 126.5, 125.8, 122.3, 120.7, 118.9,
62.0, 51.9, 50.8, 25.8, 25.0, 24.9, 18.2, -4.4, -4.5, -26.3;
HRMS-ESI [MNa]+ (m/z) for C30H35NOSiNa calcd 476.2386;
obsd 476.2363.
21en (123 mg, 0.44 mmol) dissolved in 6.0 mL of dry diethyl
ether under nitrogen at room temperature was treated with
4.0 mL of PhLi (0.89 M, 3.52 mmol) in ether. After 5 min, the
reaction mixture was poured into 10 mL of 4.0 M HCl, and
the pH was adjusted to ca. 10 with KOH. The two layers were
separated, and the aqueous layer was acidified with AcOH and
extracted with chloroform to give 50 mg of unreacted starting
material 21en . The organic layer was dried and concentrated.
The residue was subjected to column chromatography on a 2
cm × 30 cm silica gel column to yield 38 mg of 26en (43%
based on the consumed starting material).
Meth od B. 6-Phenyl-6-pyridin-2-yl-bicyclo[2.1.1]hexane-5-
carbaldehyde (23en ) (74 mg, 0.28 mmol) dissolved in 1.5 mL
of dry diethyl ether under nitrogen at room temperature and
0.37 mL of PhLi (0.89 M, 0.33 mmol) were combined. The
reaction mixture was stirred for 10 min and then poured into
an NH4Cl solution. Workup as usual (dichloromethane) gave
the alcohol 27en . The crude product of 27en dissolved in 3.0
mL of acetone was treated with J ones reagent (2.94 M, 0.28
mmol) for 15 min. Acetone was removed in vacuo, and the
residue was treated with 5% KOH and dichloromethane and
worked up as usual. The residue was subjected to chromatog-
raphy on a 2 cm × 30 cm silica gel column and eluted with
dichloromethane-ether (9:1) to yield 26en , which was recrys-
tallized from dichloromethane to give 80 mg (84%) of 26en as
colorless needle crystals: Rf 0.57 (dichloromethane-diethyl
ether 9:1); IR 1667.4 cm-1; 1H NMR (CDCl3, δ) 8.59 (ddd, J )
5.0, 2.0, 1.0 Hz, 1H), 7.93-7.87 (m, 2H), 7.63-7.36 (m, 5H),
7.27-7.22 (m, 4H), 7.15-7.04 (m, 2H), 3.89 (d, J ) 2.5 Hz,
2H), 3.13 (s, 1H), 1.76-1.55 (m, 4H); 13C NMR (δ) 200.2, 163.7,
149.3, 141.7, 136.5, 136.2, 132.9, 128.5, 128.3, 128.0, 127.3,
125.9, 122.6, 121.2, 61.9, 51.1, 49.5, 22.1; HRMS-ESI [MNa]+
(m/z) for C24H21NONa calcd 362.1521; obsd 362.1538. This
structure was confirmed by X-ray analysis.
ter t-Bu t yld im et h ylsilyloxy E n ol E t h er of 26-en d o
(24en ). To phenyl-(6-endo-phenyl-6-exo-pyridin-2-yl-bicyclo-
[2.1.1]hex-5-yl)-methanone (26en ) (230 mg, 0.68 mmol) sus-
pended in 3.0 mL of AcOH was added 0.25 mL of HBr (48%,
2.0 mmol), followed by 20 mL of Br2 (0.1 M in AcOH, 2.0 mmol).
After the reaction mixture was heated at 95 °C for 28 h, AcOH
was removed by vacuum distillation. The residue was treated
with 5% KOH and dichloromethane. Workup as usual (dichlo-
romethane) gave 25en , which was used directly for the next
step without further purification. Spectral data for 25en : IR
1675.1 cm-1; 1H NMR (CDCl3, δ) 7.77 (d, J ) 5.0 Hz, 1H), 7.61
(m, 2H), 7.45 (td, J ) 7.0, 1.0 Hz, 1H), 7.35-7.15 (m, 8H),
7.12-7.04 (m, 1H), 6.77 (ddd, J ) 7.0, 5.0, 1.0 Hz, 1H), 4.43
(d, J ) 6.0 Hz, 1H), 4.23 (d, J ) 6.0 Hz, 1H), 2.36-2.15 (m,
2H), 2.00-1.90 (m, 2H); 13C NMR (δ) 191.2, 160.7, 148.3, 144.1,
135.8, 134.4, 132.4, 129.1, 128.4, 128.1, 127.8, 126.8, 126.7,
126.0, 125.3, 121.2, 74.2, 61.4, 54.4, 54.1, 25.8, 25.0; HRMS-
ESI [MH]+ (m/z) for C24H21BrNO calcd 418.0806; obsd 418.0819.
25en dissolved in 15 mL of freshly distilled THF under
nitrogen was cooled at -78 °C, and 0.85 mL of HMPA (4.4
mmol) was added, followed by 1.9 mL of t-BuLi (1.16 M, 2.2
mmol). Three milliliters of a THF solution of TBS chloride
(0.685 g, 4.4 mmol) was then added. The reaction was stirred
overnight and allowed to warm slowly to room temperature.
The solvent was removed in vacuo, and the residue was
subjected to chromatography on a 2 cm × 30 cm silica gel
column. Elution with hexane-dichloromethane-ether (2:2:1)
yielded 300 mg (98%) of 24en as a colorless oil that is very
stable at ambient temperature: IR, no carbonyl absorption;
1H NMR (CDCl3, δ) 8.42 (ddd, J ) 5.0, 2.0, 1.0 Hz, 1H), 7.44
(td, J ) 7.5, 2.0 Hz, 1H), 7.34 (dt, J ) 8.0, 1.5 Hz, 1H), 7.29-
7.08 (m, 11H), 6.93 (ddd, J ) 7.5, 5.0, 1.0 Hz, 1H), 3.92 (q,
J ) 6.0 Hz, 2H), 1.86-1.66 (m, 4H), 0.89 (s, 9H), -0.17 (s,
3H), -0.28 (s, 3H); 13C NMR (δ) 164.2, 148.7, 141.9, 138.6,
135.7, 128.2, 127.7, 127.6, 127.5, 126.7, 126.7, 125.9, 122.2,
120.6, 62.8, 51.3, 50.5, 25.8, 24.8, 24.6, 18.2, -4.5; HRMS-ESI
[MNa]+ (m/z) for C30H35NOSiNa calcd 476.2386; obsd 476.2382.
5-Br om o-6-en do-ph en yl-6-exo-pyr idin -2-yl-bicyclo[2.1.1]-
h ex-5-yl)-p h en yl-m eth a n on e (25ex). Phenyl-(6-endo-phen-
yl-6-exo-pyridin-2-yl-bicyclo[2.1.1]hex-5-yl)-methanone (26ex)
(58 mg, 0.17 mmol) with 43 µL of HBr (0.34 mmol, 48%)
dissolved in 1.0 mL of AcOH was heated in an oil bath (∼100
°C), and 2.0 mL of Br2 (0.20 mmol, 0.1 M solution in AcOH)
was added. The reaction mixture was stirred in the oil bath
for 30 h and then cooled and neutralized with KOH solution.
Workup as usual (dichloromethane) gave the residue that was
chromatographed on a silica gel column and eluted with
dichloromethane-diethyl ether (9:1) to yield 57 mg (80%) of
25ex: mp 182-183 °C (decomposed to a dark-brown oil); IR
1666 cm-1; 1H NMR (CDCl3, δ) 8.42 (ddd, J ) 5.0, 2.0, 1.0 Hz,
Requ ir em en t for F lu or id e for Desilyla tion . It was
observed that with all of the reactants present except for
tetrabutylammonium fluoride no reaction occurred. Addition-
ally, in the acetic acid, phenol, and triethylammonium chloride
J . Org. Chem, Vol. 67, No. 26, 2002 9225

Zimmerman and Wang
reactions, as the proton donor concentration was increased,
the desilylation reaction slowed.
HRMS-ESI [MNa]+ (m/z) for C24H21NONa calcd 362.1521; obsd
362.1515.
1
Spectral data for 32en : IR 1679.0 cm-1; H NMR (CDCl3,
Kin etic P r oton a tion Stu d y. In a typical run with acetic
acid as the proton donor, to the CH3CN solution of tert-
butyldimethylsilyloxy enol ether 24en (0.022 M, 0.44 mL) was
added 2.2 µL of acetic acid, followed by 19.4 µL of 1.0 M
tetrabutylammonium floride in THF. The reaction mixture was
stirred at room temperature for 4 h. Then, water-dichloro-
methane quenching and extraction afforded a mixture of the
anti and syn benzoyl products (26en and 32en , respectively)
and the starting material 24en . The mixture was analyzed
by 1H NMR to give the anti and syn benzoyl isomers ratio
26en /32en .
In a typical run with HCl as the proton donor, to the CH3-
CN solution of tert-butyldimethylsilyloxy enol ether 24en
(0.022 M, 0.44 mL) was added 4.8 µL of 4.0 M HCl. The
reaction mixture was stirred at room temperature for 3 min.
Then, water-dichloromethane quenching and extraction af-
forded a mixture of the anti and syn benzoyl products. The
mixture was analyzed by 1H NMR to give the anti and syn
benzoyl isomers ratio 26en /32en . The ratio of 26en /32en did
not change with longer reaction times (e.g., 15 min under the
same reaction conditions).
δ) 8.02 (ddd, J ) 4.5, 2.0, 1.0 Hz, 1H), 7.52-7.05 (m, 12H),
6.90 (ddd, J ) 7.0, 4.5, 1.0 Hz, 1H), 4.05 (s, 2H), 2.94 (s, 1H),
2.01-1.79 (m, 4H); 13C NMR (δ) 198.7, 163.2, 148.1, 144.3,
137.1, 135.7, 132.2, 128.5, 128.3, 128.1, 127.6, 126.7, 125.7,
124.7, 121.2, 64.4, 58.8, 49.5, 25.9; HRMS-ESI [MNa]+ (m/z)
for C24H21NONa calcd 362.1521; obsd 362.1530.
Sta bility of 6-exo-P h en yl-6-en d o-p yr id in -2-yl-bicyclo-
[2.1.1]h exa n e-5-ben zok et on e (32en ). In a typical run, to
6-exo-phenyl-6-endo-pyridin-2-yl-bicyclo[2.1.1]hexane-5-ben-
zoketone (32en ) (15 mg, 0.044 mmol) dissolved in 2.0 mL of
CH3CN was added acetic acid (162.2 µL, 2.816 mmol), followed
by 88.0 µL of tetrabutylammonium floride in THF (0.088 mmol,
1.0 M). The reaction mixture was stirred at room temperature,
and small portions of this solution (0.3.0 mL) were taken with
2.0 mL of dichloromethane and 2.0 mL of water at the end of
1, 2, 4, and 50 h. The aqueous layer was extracted with
dichloromethane (3 × 2 mL), and the combined extracts were
washed with water (3 × 2 mL), dried over sodium sulfate, and
concentrated in vacuo. The residue was analyzed by 1H NMR.
There was no epimerization observed.
The material from the kinetic runs was combined and
subjected to silica gel chromatography using dichloromethane-
hexane-ether (2:2:1) to afford 26en and 32en from the endo
diastereomeric runs and 26ex and 32ex from the exo runs.
Ack n ow led gm en t. Support of this research by the
National Science Foundation is gratefully acknowledged
with special appreciation for its support of basic re-
search.
1
Spectral data for 32ex: IR 1675.0 cm-1; H NMR (CDCl3,
δ) 8.45 (ddd, J ) 5.0, 2.0, 1.0 Hz, 1H), 7.54-7.46 (td, J ) 7.5,
2.0 Hz, 1H), 7.45-7.41 (dt, J ) 7.5, 1.5 Hz, 1H), 7.40-7.25
(m, 6H), 7.10-7.05 (d, J ) 8.0 Hz, 1H), 7.03-6.91 (m, 4H),
4.01 (s, 2H), 2.83 (s, 1H), 2.02-1.75 (m, 4H); 13C NMR (δ)
198.9, 164.3, 149.2, 141.6, 137.5, 136.2, 132.0, 129.4, 127.9,
127.8, 127.5, 126.5, 121.2, 120.4, 119.5, 63.9, 58.7, 49.4, 25.8;
Su p p or tin g In for m a tion Ava ila ble: Kinetic data and
X-ray data (CIF files). This material is available free of charge
via the Internet at http://pubs.acs.org.
J O026187P
9226 J . Org. Chem., Vol. 67, No. 26, 2002