Intramolecular bromonium ion assisted epoxide ring-opening: Capture of the oxonium ion with an added external nucleophile

Article abstract of DOI:10.1021/jo101617h

9-Oxabicyclo[6.1.0]non-4-ene (1) undergoes intramolecular bromonium ion-assisted epoxide ring-opening using N-bromosuccinimide via a presumed oxonium ion that is subject to stereospecific, nonregioselective capture with added external nucleophiles produci

Full text of DOI:10.1021/jo101617h

pubs.acs.org/joc
Intramolecular Bromonium Ion Assisted Epoxide Ring-Opening: Capture
of the Oxonium Ion with an Added External Nucleophile
Karl J. Bonney, D. Christopher Braddock,* Andrew J. P. White, and Muhammad Yaqoob
Department of Chemistry, Imperial College London, London, United Kingdom, SW7 2AZ
c.braddock@imperial.ac.uk
Received August 17, 2010
9-Oxabicyclo[6.1.0]non-4-ene (1) undergoes intramolecular bromonium ion-assisted epoxide ring-
opening using N-bromosuccinimide via a presumed oxonium ion that is subject to stereospecific,
nonregioselective capture with added external nucleophiles producing novel bicyclo[4.2.1] and bicyclo-
[
3.3.1] ethers. Carboxylic acids (as catalyzed by tetramethylguanidine), alcohols, water, and halides
can all function as effective nucleophiles. Stereospecific direct opening of the bromonium ion with
carboxylic acids was found to be a competing process where high dilution disfavors this pathway.
1
3
Halogen-induced isotopic C NMR shifts (Δδ CBr 1.3-1.9 ppb; Δδ CCl 8.6-8.7 ppb) were found to
be most useful in unambiguously identifying halogen-bearing carbons, and correlation of these C NMR
13
shifts allowed ready assignment of diastereomeric structures. The structure of adducts 6b, 6c, 7b, 7c,
7
d, and 8a-d were all elucidated by X-ray crystallography.
Introduction
of the bromonium ion by the epoxide oxygen occurred, and a
presumed resultant oxonium ion was captured by the neces-
1
-8
The formation of a bromonium ion
by the action of
10a
sarily present bromide anion (Scheme 1, case a). Although
electrophilic bromine on an alkene followed by stereospecific
trapping by a nucleophile to give a 1,2-addition product is a
9
fundamental reaction in organic chemistry. In 1985, a further
reactivity mode was reported whereby in the reaction of an
epoxyalkene with molecular bromine, intramolecular trapping
mixtures of ethereal dibromide products were produced by
nonregioselective ring-openings, only those diastereomers of
stereospecific backside opening of the bromonium ion by the
epoxide oxygen and subsequent stereospecific backside at-
tack on the resultant oxonium ion were formed. The same
authors later reported that the intermolecular reaction of
alkenes and epoxides with molecular bromine could proceed
along the same pathway with the same stereochemical out-
(
1) For a review see: Ruasse, M.-F. Acc. Chem. Res. 1990, 23, 87–93. and
references cited therein.
2) The 3-membered cyclic bromonium ion was first suggested by Roberts
and Kimball: Roberts, I.; Kimball, G. E. J. Am. Chem. Soc. 1937, 59, 947–
48.
3) For evidence that the bromonium ion is cyclic and symmetrical see:
Winstein, S.; Lucas, H. J. J. Am. Chem. Soc. 1939, 61, 2845–2848.
4) For the observation of bromonium ions by NMR see: (a) Olah, G. A.;
(
10b
come (Scheme 1, case a).
11
12
9
More recently, both McDonald and Jamison have demon-
strated the use of electrophilic bromine sources (bromonium
(
(
Bollinger, J. M.; Brinich, J. J. Am. Chem. Soc. 1968, 90, 2587–2594. (b) Olah,
G. A.; Bollinger, J. M. J. Am. Chem. Soc. 1968, 90, 6082–6086.
(7) For the first generation and trapping of enantiomerically pure bro-
monium ions see: Braddock, D. C.; Hermitage, S. A.; Kwok, L.; Pouwer, R.;
Redmond, J. M.; White, A. J. P. Chem. Commun. 2009, 1082–1084.
(8) On the absolute configurational stability of bromonium ions in the
presence of alkenes see: Denmark, S. E.; Burk, M. T.; Hoover, A. J. J. Am.
Chem. Soc. 2010, 132, 1232–1233.
(
5) For the first isolable bromonium ion see: Strating, J.; Wieringa, J. H.;
Wynberg, H. J. Chem. Soc., Chem. Commun. 1969, 907–908.
6) For X-ray characterisation of stable bromonium ions see: (a) Slebocka-Tilk,
(
H.; Ball, R. G.; Brown, R. S. J. Am. Chem. Soc. 1985, 107, 4504–4508. (b) Bennet,
A. J.; Brown, R. S.; McClung, R. E. D.; Klobukowski, M.; Aarts, G. H. M.;
Santarsiero, B. D.; Bellucci, G.; Bianchini, R. J. Am. Chem. Soc. 1991, 113, 8532–
(9) Olah, G. A.; Laali, K. K.; Wang, Q.; Prakash, G. K. S. Onium Ions;
John Wiley & Sons: New York, 1998.
8
535. (c) Brown, R. S.; Nagorski, R. W.; Bennet, A. J.; McClung, R. E. D.; Aarts,
G. H. M.; Klobukowski, M.; McDonald, R.; Santarsiero, B. D. J. Am. Chem. Soc.
994, 116, 2448–2456. (d) Brown, R. S. Acc. Chem. Res. 1997, 30, 131–137.
(10) (a) Davies, S. G.; Polywka, M. E. C.; Thomas, S. E. Tetrahedron Lett.
1985, 26, 1461–1464. (b) Davies, S. G.; Polywka, M. E. C.; Thomas, S. E.
J. Chem. Soc., Perkin Trans. 1 1986, 1277–1282.
1
DOI: 10.1021/jo101617h
r 2010 American Chemical Society
Published on Web 11/29/2010
J. Org. Chem. 2011, 76, 97–104 97

Bonney et al.
JOCArticle
SCHEME 1. Generalized Modes of Intramolecular Bromonium
Ion-Assisted Epoxide Ring-Openings
SCHEME 2. Bromoether Formation by Capture with an Added
Nucleophile
di-sym-collidine perchlorate and NBS, respectively) to effect
spectacular intramolecular bromonium ion assisted epoxide
ring-openings, to construct complex (poly)cyclic ether com-
pounds, where the final cyclization is terminated by an intra-
molecularly positioned nucleophile (Scheme 1, case b). In
these systems, the regioselectivity of ring-closure was effec-
tively controlled by the substitution pattern of the substrate,
and stereospecific ring-openings occurred, although the bromo-
nium ion formation was not face selective. The latter work
allowed for the total synthesis of ent-dioxepandehydrothyr-
siferol and demonstrated the feasibility of this route as a
biogenetic pathway in Laurencia species. To date, however,
there have been no reports on intramolecular bromonium
ion-assisted epoxide ring-opening with an added external
Results and Discussion
10a
We selected 9-oxabicyclo[6.1.0]non-4-ene (1) as a sui-
table alkene epoxide substrate and N-bromosuccinimide
(NBS) as the source of electrophilic bromine. A carboxylic
acid was selected as a suitable nucleophile which can provide
a proton to produce succinimide as the neutral byproduct.
However, initial experiments with alkene epoxide 1 and
NBS, with acetic acid as the added nucleophile, in dichlor-
omethane, failed to give any reaction. To our delight, the
addition of catalytic quantities (1 mol %) of tetramethylgua-
1
3
nucleophile (Scheme 1, case c). Such a method should
provide additional flexibility in building (poly)cyclic ether com-
pounds with different functional groups. However, in such
systems there is an additional inherent selectivity question;
whether the added external nucleophile will attack directly
the bromonium ion of the original olefin or whether the for-
mation of an oxonium ion followed by its trapping will pre-
dominate. Herein we report the first examples of intramolec-
ular bromonium ion assisted epoxide ring-expansions with
16
nidine (TMG) led to rapid reaction giving the previously
unreported [4.2.1]bicycloether 2 and [3.3.1]bicycloether 3 (as
an inseparable 1:1 mixture, 43%) and novel bromoacetates 4
(13%) and 5 (26%) after purification by column chroma-
tography (Scheme 2). The identities and stereochemistries of
these diastereomeric compounds with identical molecular for-
mulas were unambiguously proven by a combination of X-ray
crystallography, NMR methods, and correlation by inter-
conversion (vide infra). In the reaction itself, a bromonium
ion A can evidently be trapped intramolecularly by the epoxide
to give an oxonium ion B. This can only occur when the
bromonium ion and epoxide are on opposite faces (where
presumably the bromonium ion can be formed reversibly on
either face of the alkene). The oxonium ion B can in turn be
trapped with an added external nucleophile (acetate) giving 2
and 3. A control experiment without NBS showed that the
epoxide is not ring-opened by acetic acid under these condi-
tions. A further possible isomer could possibly be envisioned
by opening of the oxonium ion B R-to the carbon bearing
bromine. This would lead to a syn-bromocarboxylate (not
shown) by double inversion at this center. Such a compound
was not detected in this or any other series. This implies that
the bromine is an effective steric shield for ring-opening of
the oxonium ion at this position. Alternatively, the bromo-
nium ion A can be directly trapped by acetate giving bro-
moacetates 4 and 5 with the epoxide left intact. In this
1
4
trapping of the presumed resultant oxonium ion by added
external nucleophiles. In particular, we report that the use of
1
5
water as the solvent (and nucleophile) gives rise to exclusive
trapping of the oxonium ion in preference to the original
bromonium ion.
(
11) Bravo, F.; McDonald, F. E.; Neiwert, W. A.; Hardcastle, K. I. Org.
Lett. 2004, 6, 4487–4489.
12) Tanuwidjaja, J.; Ng, S.-S.; Jamison, T. F. J. Am. Chem. Soc. 2009,
31, 12084–12085.
13) (a) For the intramolecular capture of a bromonium ion with the
(
1
(
oxygen of a THF and opening of the oxonium ion with an external
nucleophile see: Braddock, D. C.; Millan, S.; Perez-Fuertes, Y.; Pouwer,
R. H.; Sheppard, R. N.; Solanki, S.; White, A. J. P. J. Org. Chem. 2009, 74,
1
835–1841 and references cited therein. An electrophilic aminoalkoxylation
of olefins featuring the intermolecular capture of a bromonium ion by an
epoxide followed by capture of the oxonium intermolecularly by an amine
has been reported: (b) Zhou, L.; Tan, C. K.; Zhou, J.; Yeung, Y.-Y. J. Am.
Chem. Soc. 2010, 132, 10245–10247.
(
14) For a report of an extraordinarily stable oxonium ion see: Mascal,
M.; Hafezi, N.; Meher, N. K.; Fettinger, J. C. J. Am. Chem. Soc. 2008, 130,
3532–13533.
15) For the use of water as solvent for promoting endoselective epoxide-
opening cascades see: (a) Morten, C. J.; Jamison, T. F. J. Am. Chem. Soc. 2009,
1
(
(16) For the use of TMG as a catalyst for bromolactonisation and
intermolecular bromoacetoxylation see: Ahmad, S. M.; Braddock, D. C.;
Cansell, G.; Hermitage, S. A. Tetrahedron Lett. 2007, 48, 915–918.
1
6
31, 6678–6679. (b) Byers, J. A.; Jamison, T. F. J. Am. Chem. Soc. 2009, 131,
383–6385. (c) Vilotijevic, I.; Jamison, T. F. Science 2007, 317, 1189–1192.
9
8
J. Org. Chem. Vol. 76, No. 1, 2011

Bonney et al.
JOCArticle
TABLE 1. Concentration Dependent Product Distribution
utilized as nucleophiles by the action of NBS in the alcohol as
solvent (entries 8,9) (the use of added methanol in dichlor-
omethane gave only poor conversions and gave rise to dibro-
mide side-products) to give bicycloethers 6e and 7e, and 6f
and 7f respectively, both as inseparable mixtures. Trace quan-
tities of epoxides from direct opening of the bromonium ion
by methanol were apparent, but in insufficient quantities to
be isolated and fully characterized. In these cases there is a
slight preference (2-3:1) for the [3.3.1]bicycloether as the
major diastereoisomer. The use of water as solvent produced
a
-3
b
b
entry
[1] mol dm
% conversion
2 þ 3: 4 þ 5
1
2
3
4
a
0.25
0.025
0.0025
0.00025
99
95
63
30:70
55:45
80:20
85:15
c
72
All reactions performed with 1.0 equivalent of NBS, 15 μL of TMG
18
ether products 6g and 7g exclusively in excellent yields,
and 1.0 equivalent of AcOH for 2.5 h at rt (see Supporting Information).
As determined by inspection of the H NMR spectrum of the crude
b
1
regardless of the concentration (entries 10,11,12), again with
a slight preference (ca. 3:1) for the [3.3.1]bicycloether as the
major diastereoisomer. Finally, we prepared the known
dibromides 6h and 7h as an inseparable mixture by the action
of molecular bromine on epoxyalkene 1 in carbon tetrachlo
c
reaction mixture (see Supporting Information). After a reaction time of
72 h.
eventuality, the external nucleophile can trap the bromonium
ion regardless of which face it has been formed giving rise to
two bromoacetate epoxide diastereoisomers. The relative
stereochemistries of the products show that both bromo-
nium and oxonium ion trapping are stereospecific (albeit not
regioselective) giving a total of only four possible products.
In an effort to further favor the formation of bromoethers
10a
ride (entry 13). These experiments show that the intramole-
cular bromonium ion assisted epoxide ring-opening is a favor-
able and high-yielding process that can occur with a variety
of added external nucleophiles (carboxylic acids, alcohols,
water and halides). In polar solvents (which also function as
the nucleophile) this process is highly favored over direct
attack of the nucleophile on the bromonium ion. Under these
conditions the [3.3.1]bicycloether is the slightly preferred
diastereoisomer. In chlorinated solvents the intramolecular
bromonium ion assisted epoxide ring-opening with carboxylic
acids as added nucleophiles canbe rendered favorable in dilute
solution.
2
and 3 over bromoacetates 4 and 5, we explored the effect of
concentration on the reaction. Inspection of the results (Table 1)
show a marked concentration dependence on the relative ratios
of bicyclo ethers 2 and 3 versus bromoacetates 4 and 5, with
1
7
the former being favored at high dilution.
A series of other carboxylic acids as added external nucleo-
philes were then utilized (Table 2). Thus benzoic acid, p-
nitrobenzoic acid and 2,3,4,5-tetrafluorobenzoic acid as
added nucleophiles under the optimized conditions shown
gave bicycloethers 6a-c and 7a-c along with the bromoa-
cetates 8a-c and 9a-c (entries 1-5) respectively. A con-
centration effect is again in operation (entry 1 vs entry 2
and entry 4 vs entry 5) and the use of more dilute solutions
Both bicycloethers in the p-nitro- (6b, 7b) and tetrafluoro-
6c, 7c) benzoate series were crystalline and their structures
(
were solved by X-ray crystallography (Figure 1, 6b and 7b.
For 6c and 7c see the Supporting Information). Chloride 7d
could be obtained by HPLC separation, and was also crystal-
line, and was also confirmed as the [3.3.1]bicycloether by
19
X-ray crystallography (see the Supporting Information).
(
entries 2 and 5) provided the bicycloethers as the major
These structures confirm the expected anti relationship between
both the bromide and ether oxygen and the anti relationship
between the ether oxygen and the benzoate group, showing
that bromonium ion opening by the epoxide, and subsequent
oxonium ion opening are completely stereospecific. The tetra-
fluorobenzoate ethers 6c and 7c were hydrolyzed under basic
conditions to the alcohols 6g and 7g respectively (thus con-
firming the latter’s structures). Individual samples of authen-
tic acetates 2 and 3, benzoates 6a and 7a and methyl ethers 6e
and 7e (thus confirming their structures), were then prepared
by standard methods from the free alcohols 6g and 7g
by acetylation, benzoylation and methylation respectively
products of these reactions in high yields. The relative ratio
of all the bicycloethers was approximately unity in these
cases, indicating indiscriminate attack at either position of
oxonium ion B by the nucleophile. The benzoate ethers 6a
and 7a were formed as an inseparable mixture, while the p-
nitrobenzoate ethers 6b and 7b, and the tetrafluorobenzo-
ate ethers 6c and 7c could be separated by column chro-
matography and by HPLC respectively. Chloride could be
incorporated to give 6d and 7d in excellent yield by the use
of TMSCl and catalytic TMG in CH Cl (entry 6). Alter-
2
natively the use of LiCl in wet CH CN gave the same
3
2
bicyclic ethers along with small quantities of bromochlor-
ide epoxides 8d and 9d (entry 7). In both cases the yield of
the bicyclic ethers 6d and 7d is good even in relatively con-
centrated solution, but there is no marked preference for one
ring system over the other. Alcohols (MeOH, EtOH) could be
(see Supporting Information). With the stereochemistry all
secured for all the novel bicyclic ethers bar 6d, 6f, and 7f, inspec-
tion of all the C NMR spectra reveals a strong correlation
for the [4.2.1] ethers where the carbon bearing the bromine
1
3
(18) These alcohols have been reported previously as a 7:1 inseparable
(17) This effect isnoteasytorationalize. Assumingthattheattackofacetate
is the rate-determining step (and irreversible) then the product distribution
mixture, by bromination of an epoxydiol: Takahashi, A.; Aso, M.; Tanaka,
M.; Suemune, H. Tetrahedron 2000, 56, 1999–2006.
should remain unchanged with changing concentration. Oxonium ion B or
bromonium ion A could not be observed by H NMR monitoring of the
(19) The X-ray data is unable to distinguish between the mixed halogen
species and a mixture of the dibromo and dichloro species for 7d (see
Supporting Information for a full discussion). However, there is no doubt
1
reaction mixtures indicating that neither intermediate accumulates. The posi-
tion of the expected rapid equilibrium between A and B is expected also to be
concentration independent. One referee helpfully suggested that “If one makes
the alternative assumption that bromonium ion A and oxonium ion B are not in
rapid equilibrium, then the product distribution is the result of the relative rates
offormation ofB versusexternalnucleophilic attackonA. Higher dilution then
favors the intramolecular attack to form B, which may then react with the
added nucleophile at its leisure.”
1
3
that this is the mixed bicyclo[3.3.1]bromochloride: the C NMR spectrum
has 8 different C NMR resonances with the resonance at 50.3 ppm showing
1
3
7
9
81
a distinctive bromine isotope effect (Δδ [C Br,C Br] = 1.8 ppb) and the
resonance at 57.6 ppm showing a distinctive chlorine isotope effect (Δδ
[C Cl,C Cl] = 8.7 ppb). See also ref 20. The mass spectrum also shows the
3
5
37
distinctive isotope pattern for a monochloromonobromide [m/z=242 (24%),
240 (100%), 238 (76%): M ].
þ.
J. Org. Chem. Vol. 76, No. 1, 2011 99

Bonney et al.
JOCArticle
TABLE 2. Bromonium Ion-Assisted Epoxide Ring-Opening of 1 with Added External Nucleophile
a
a
entry
b
nucleophile (equiv)
solvent
additive
T (h)
[C]
6
7
8
9
total yield
c
6a and 7a (50%) 0.9: 1
d
1
PhCO
PhCO
2
H (1.1)
H (1.1)
CH
CH
CH
2
Cl
2
Cl
2
Cl
2
Cl
2
Cl
2
Cl
2
2
2
2
2
2
1 mol % TMG
1 mol % TMG
1 mol % TMG
1 mol % TMG
1 mol % TMG
1 mol % TMG
6
6
3
6
3
4
1
0.22 M
0.0025 M
0.0026 M 6b (41%)
8a (22%)
8a (2%)
8b (11%)
8c (4%)
8c (10%)
8d (0%)
9a (8%)
9a (2%)
9b (4%)
9c (2%)
9c (4%)
9d (0%)
80%
85%
98%
39%
58%
90%
94%
58%
51%
84%
96%
86%
48%
b
c
6a and 7a (81%) 1: 0.9
d
2
2
b
3
4-NO
2
PhCO H (1.1)
2
7b (42%)
b
c
d
4
2,3,4,5-F
2,3,4,5-F
TMSCl (1.1)
LiCl (15)
MeOH
4
-PhCO
2
H (1.1) CH
H (1.1) CH
CH
0.25 M
0.0042 M 6c (22%)
6c and 7c (33%) 0.6: 1
e e
b
5
4
-PhCO
2
7c (22%)
6d and 7d (90%) 1.4: 1
f
c
d
6
0.08 M
0.22 M
0.25 M
0.25 M
0.25 M
0.025 M
0.0025 M
0.2 M
f
7
c
6d and 7d (86%) 1.2: 1
d
b
MeCN
MeOH
EtOH
-
-
-
-
-
-
-
8d and 9d (8%) 9: 1
f
g
c
6e and 7e (52%) 1: 2.1
d,h
8
6.5
-
-
-
-
-
-
-
-
-
-
-
-
f
c
6f and 7f (51%) 1: 3.2
d
9
EtOH
9.5
2.5
2.5
2.5
3
f
d
6g and 7g (84%) 1: 3.2
d
1
1
1
1
0
1
2
3
H
H
H
2
2
2
O
O
O
H
H
H
2
O
2
O
2
O
f
f
f
c
6g and 7g (96%) 1: 3.0
d
d
c
6g and 7g (86%) 1: 3.2
c
6h and 7h (48%) 1: 1
d
Br
2
CCl
4
a
Compounds 6a-h and 7a-h were found to be indefinitely stable at room temperature in chlorinated solvents. Thus under the conditions of the
reaction no isomerization between the two is occurring, and the product ratios reflect the kinetic product distribution and not an equilibrium position.
Two equivalents of NBS. Products were inseparable by flash column chromatography. Ratio determined by H NMR spectroscopy. Isolated yield
b
c
d
1
e
f
g
after HPLC separation. NBS (1.0-1.3 equiv). The use of catalytic quantities of TMG resulted in the reaction going to completion in 4 h, but the isolated
yield of 6e and 7e was reduced at 43% (1: 2.8). Average of two runs after chromatography.
h
and convenient method for identification of the different ring
systems (Table 3). Chloride 6d and ethyl ethers 6f and 7f were
then assigned on the basis of their bromine-bearing carbon
resonances at 53.4, 54.7, and 51.4 ppm respectively. The
1
0a
known dibromides 6h (53.4 ppm) and 7h (50.4 ppm) were
also found to conform to this model. Moreover, bromide-
induced isotopic NMR shifts were most useful in unambigu-
ously identifying these carbons as the bromide bearing ones
2
0
(
Figure 2, see also Supporting Information).
2
The bromoepoxides 8a, 8b, 8c and 8d were also unam-
1
biguously identified by X-ray crystallography (Figure 3a-d).
Evidently, bromonium ions have been trapped directly by the
nucleophile to give an anti arrangement of functional groups by
stereospecific ring-opening. The diastereomeric epoxide pro-
ducts 9a-9d must come from ring-opening of the bromonium
ion when formed on the other face of the olefin. Interestingly,
bromotetrafluorobenzoate epoxide 9c proved to be unstable
on standing, and produced 6c, 7c, 8c (and 9c) in ratios com-
parable to those observed in the reaction detailed in Table 2,
entry 5. This can be rationalized by invoking bromonium ion
formation in 9c by neighboring group participation of the
(20) (a) For bromine-induced isotopic shifts see: Raynes, W. T.; Sergeyev,
N. M.; Sandor, P.; Grayson, M. Magn. Reson. Chem. 1997, 35, 141–143. (b) For
chlorine-induced isotopic shifts see: Sergeyev, N. M.; Sergeyeva, N. D.; Raynes,
W. T. J. Magn. Reson, Ser. A 1995, 115, 174–182. (c) For the use of these
methods to distinguish bromochloride regioisomers after opening of a bromo-
nium ion with chloride see ref 7. (d) For the use of these methods in natural
product structure elucidation see ref 13 and references cited therein.
FIGURE 1. (a) Molecular structure of 6b. (b) Molecular structure
of 7b.
(
21) Modeling of the disorder in the X-ray crystal structure of 8d means
atom resonates invariantly in the region 54 ( 1 ppm. The
same resonance is seen at 50.5 ( 1 ppm for the [3.3.1]bicyclic
compounds. For a given substituent, the difference between
these chemical shifts is Δδ g 3 ppm, thus providing a quick
that for 88% of the molecules in the crystal the bromide is on the same side of
the ring as the epoxide oxygen, whilst in the remaining 12% it is the chloride
that is the same side as the oxygen (i.e., the minor diastereoisomer 9d). This is
consistent with the ratio of the diastereoisomers (9:1) observed as an
1
inseparable mixture by H NMR and implies cocrystallisation.
1
00 J. Org. Chem. Vol. 76, No. 1, 2011

Bonney et al.
JOCArticle
13
TABLE 3. Correlation of C NMR Chemical Shifts for the Carbon-
Bearing Bromine for 2, 3, 6a-h, and 7a-h
a
a
3
All spectra were recorded in CDCl .
FIGURE 2. Bromide-induced isotopic shift for the CBr resonance
of 3 at 50.6 ppm (125 MHz, CDCl ).
3
FIGURE 3. (a) Molecular structure of one (8a-I) of the two crystal-
lographically independent C -symmetric molecules present in the
S
bromine and loss of tetrafluorobenzoate. The resultant bromo-
nium ion A (c.f. Scheme 2) can now be trapped by the
epoxide and the oxonium ion B trapped by the earlier
expelled tetrafluorobenzoate anion. The formation of 8c also
implies concomitant reversible loss of electrophilic bromine
crystals of 8a (see Supporting Information). (b) Molecular structure
of 8b. (c) Molecular structure of 8c. (d) Molecular structure of the
C -symmetric species 8d (see Supporting Information).
S
8
from the bromonium ion to the bulk medium and return to
13
TABLE 4. Correlation of C NMR Chemical Shifts for the Carbon-
a
the other face of the alkene and recapture by tetrafluoro-
benzoate. To the best of our knowledge the spontaneous for-
mation of a bromonium ion by NGP of bromide on a suitably
positioned leaving group is unprecedented, and this observa-
tion may be useful in designing further systems for the gener-
ation of enantiopure bromonium ions starting from enantio-
Bearing Bromine for 4, 5, 8a-c, and 9a-c
7
pure bromohydrins. It also implicitly confirms the expected
anti relationship between the bromine and tetrafluorobenzo-
ate group in 9c. Bromoacetate epoxides 4 and 5 were assigned
1
3
by comparison of the C NMR spectral data for the
bromine-bearing carbons (Table 4) with those of esters 8a-c
and 9a-c respectively. Here again, the bromide-induced
isotopic NMR shifts were most useful in unambiguously
identifying these carbons as the bromide bearing ones (see
a
3
All spectra were recorded in CDCl .
unambiguously confirms the assigned stereochemistry for
9
2
0
c, and by analogy the stereochemistries of 5 and 9a-b.
Finally, we recognized that a peracid epoxidation of
Supporting Information).
Bromotetrafluorobenzoate epoxide 9c was saponified
2
2,23
inexpensive 1,5-cyclooctadiene (11) would deliver epoxyalk-
ene 1 and produce a stoichiometric quantity of carboxylic
acid as side-product. This carboxylic acid could subsequently
also function as an in situ external added nucleophile in the
presence of a suitable electrophilic bromine source, to provide
bicyclo[4.2.1] and bicyclo[3.3.1]ethers, thereby constructing
products with four stereocenters with complete control of
relative stereochemistry in one-pot and with good atom
to give the known syn-diepoxide 10 (Scheme 3).
This
(
22) Rudolph, J.; Reddy, K. L.; Chiang, J. P.; Sharpless, K. B. J. Am.
Chem. Soc. 1997, 119, 6189–6190.
23) Saponification of bromobenzoate 8c under the same conditions led
(
to the formation of the expected diastereomeric epoxide and competitive
formation of alcohols 6g and 7g. The pre-existing epoxide evidently suffers
intramolecular nucleophilic attack by the alkoxide generated from collapse
of the tetrahedral intermediate during BAc2 ester hydrolysis.
J. Org. Chem. Vol. 76, No. 1, 2011 101

Bonney et al.
JOCArticle
SCHEME 3. Saponification to Epoxide 10
Experimental Section
Representative Procedure for Bromonium Ion-Assisted Epox-
ide Ring-Opening with Acetic Acid as the Added External Nu-
cleophile. To a solution of epoxyalkene 1 (100 mg, 0.80 mmol)
2 2
in CH Cl (3.6 mL) was added TMG (0.92 mg, 0.008 mmol,
1
1
mol %), acetic acid (0.05 mL, 1.6 mmol) and NBS (286 mg,
.6 mmol). The mixture was stirred for 1 h at room temperature,
diluted with CH
solution of Na SO
over MgSO and the solvent was evaporated in vacuo to afford
2
Cl
2
(20 mL) and quenched with an aqueous
(10%, 20 mL). The organic layer was dried
SCHEME 4. One-Pot Epoxidation-Bromination
2
3
4
the crude material. Column chromatography eluting with pe-
troleum ether/EtOAc (9:1 to 8:2) afforded first a 1:1 mixture
of the two cyclic ethers 2 and 3 (90 mg, 43%), second, epoxide
5
samples of 2 and 3 were prepared from alcohols 6g and 7g
(27 mg, 13%) and third, epoxide 4 (54 mg, 26%). Authentic
2
4
respectively (see Supporting Information).
1R*,2R*,5S*,6S*)-5-Bromo-9-oxabicyclo[4.2.1]nonan-2-yl
Acetate (2). Colorless oil: R (PE/EtOAc 8:2) 0.36; IR (neat)
(
f
-
1 1
1
4
1
2
3
734 cm ; H NMR (400 MHz, CDCl ) δ 5.15 (ddd, J = 3.8,
.9, 7.4, 1H), 4.74-4.65 (m, 1H), 4.62 (ddd, J = 2.9, 4.6, 7.5,
H), 4.20 (ddd, J = 4.9, 7.6, 9.7, 1H), 2.32-2.23 (m, 1H),
.20-2.09 (m, 3H), 2.08 (s, 3H), 2.07-1.89 (m, 3H), 1.83-1.74
efficiency. We recognized that we could run this reaction in a
one-pot stepwise fashion, with dilution after complete epox-
idation, followed by addition of NBS and TMG. Alter-
natively, all the reagents could be added at t=0, where the
epoxide would be expected to be consumed as it was pro-
duced. In the event, both approaches (see Supporting In-
formation for full information) were successively realized
using mCPBA, NBS and TMG (Scheme 4) to give meta-
chlorobenzoate ethers 12 (δC [CBr] 53.5 ppm) and 13 (δC
1
3
(
m, 1H); C NMR (100 MHz, CDCl
3
) δ 170.2, 82.7, 78.3, 73.3,
þ
5
3.7, 29.8, 29.3, 26.1, 25.9, 21.1; MS (CI , NH
3
) m/z 282, 280
Br
þ
þ
79
(
(
M þ NH
4
) ; HRMS (CI , NH
3
) m/z calcd for C10
H
19NO
3
þ
M þ NH
4
)
280.0548, found 280.0552.
1R*,2R*,5R*,6R*)-6-Bromo-9-oxabicyclo[3.3.1]nonan-2-yl
(
Acetate (3). Colorless oil: R (PE/EtOAc 8:2) 0.36; IR (neat)
f
-
1 1
1735 cm ; H NMR (400 MHz, CDCl ) δ 5.12 (dt, J = 6.0,
3
11.8, 1H), 4.45 (dt, J = 5.6, 12.7, 1H), 3.96 (dt, J = 5.5, 10.7,
2H), 2.54-2.44 (m, 1H), 2.38 (ddd, J = 6.4, 12.9, 19.6, 1H),
1
3
[CBr] 50.4 ppm).
2.29-2.20 (m, 1H), 2.07 (s, 3H), 2.10-1.85 (m, 5H); C NMR
(
100 MHz, CDCl
3
) δ 170.1, 70.3, 69.7, 66.5, 50.6, 31.1, 25.6,
þ
þ
2
5.1, 23.1, 21.2; MS (CI , NH
3
) m/z 282, 280 (M þ NH
4
) ;
Conclusion
þ
79
þ
HRMS (CI , NH
280.0548, found 280.0552.
1S*,4S*,5S*,8R*)-5-Bromo-9-oxabicyclo[6.1.0]nonan-4-yl
Acetate (4). Colorless oil: R (PE/EtOAc 9:1) 0.14; IR (neat)
3
) m/z calcd for C10
H19NO
3
Br (M þ NH
4
)
In conclusion, we have extended the intramolecular bromo-
nium ion-assisted epoxide ring-opening method first reported
by Davies and Thomas, and more recently by Jamison and
McDonald, to allow the incorporation of an added external
nucleophile. In exploring such systems there is an additional
inherent selectivity question;whether the added external
nucleophile will attack directly the bromonium ion of the
original olefin or whether the formation of an oxonium ion
followed by its trapping will predominate. Using NBS and a
catalytic quantity of TMG allowed the use of aliphatic and
aromatic carboxylic acids to function as nucleophiles as
desired, with the trapping of the presumed oxonium ion favored
at high dilution. Alcohols and water could also function as the
nucleophile when used directly as solvents (in conjunction with
NBS) to trap the presumed oxonium in high yield regardless
(
f
-
1 1
1
1
2
737 cm ; H NMR (400 MHz, CDCl ) δ 5.24 (dt, J = 3.8, 8.4,
3
H), 4.61 (ddd, J = 3.4, 6.8, 8.4, 1H), 3.04-2.95 (m, 2H),
.32-2.14 (m, 5H), 2.08 (s, 3H), 1.96-1.79 (m, 1H), 1.76-1.58
13
(
m, 2H); C NMR (100 MHz, CDCl ) δ 169.6, 73.1, 55.4, 55.2,
3
þ
5
1.0, 29.6, 28.6, 24.1, 23.5, 21.1; MS (CI , NH ) m/z 282, 280 (M
3
þ
þ
79
3
þ NH
4
) ; HRMS (CI , NH
3
) m/z calcd for C10
H
16
O
Br (M þ
þ
H) 263.0283, found 263.0284.
(1S*,4R*,5R*,8R*)-5-Bromo-9-oxabicyclo[6.1.0]nonan-4-yl
Acetate (5). Colorless oil: R (PE/EtOAc 9:1) 0.18; IR (neat)
737 cm ; H NMR (400 MHz, CDCl ) δ 5.24 (td, J = 3.4, 9.1,
f
-
1 1
1
1
3
3
3
H), 4.42 (ddd, J = 4.0, 4.9, 8.9, 1H), 3.17 (dt, J = 4.3, 7.3, 1H),
.10 (dt, J = 4.5, 9.6, 1H), 2.50-2.34 (m, 2H), 2.29-2.14 (m,
13
H), 2.11 (s, 3H), 1.91-1.73 (m, 2H), 1.62-1.46 (m, 1H); CNMR
) δ 169.9, 73.2, 55.5, 54.6, 53.7, 31.8, 27.9,
(
100 MHz, CDCl
3
of the concentration. Halides (using TMSCl, LiCl or Br ) could
2
þ
þ
þ
2
5.2, 23.3, 21.0; MS (CI , NH ) m/z 282, 280 (M þ NH ) ;
3
4
also be incorporated in high yields. These results demon-
strate that the intramolecular bromonium ion-assisted ep-
oxide ring-opening is a favorable and high-yielding process
that can occur with a variety of added external nucleophiles.
This pathway can be favored over direct attack of the exter-
nal nucleophile on the bromonium ion. However, the regio-
chemistry of the nucleophilic capture on the oxonium ion is
not highly selective and remains essentially unperturbed by
the nature of the solvent or of the nucleophile. Finally, we
have also demonstrated herein the value of halide-induced
isotopic NMR shifts in unambiguously identifying chloride
and bromide-bearing carbons, thus aiding structure elucida-
tion in general.
þ
79
3
HRMS (CI , NH ) m/z calcd for C H O Br (M þ H)
3
10 16
2
63.0283, found 263.0285.
Benzoic Acid. Following the representative procedure (above)
using benzoic acid as the added external nucleophile according
to the conditions recorded in Table 2, entry 2. The crude mixture
was purified by column chromatography eluting with petroleum
ether/EtOAc (9:1 to 8:2) to give first, an inseparable mixture of the
cyclic ether benzoates 6a and 7a, second epoxide 8a, and third,
epoxide 9a. Authentic samples of 6a and 7a were prepared from
24
alcohols 6g and 7g respectively (see Supporting Information).
(24) Tetrafluorobenzoates 6c and 7c could be separated by HPLC and
individually saponified to give pure samples of each of alcohols 6g and 7g
(see Supporting Information).
1
02 J. Org. Chem. Vol. 76, No. 1, 2011

Bonney et al.
JOCArticle
þ
þ
(
Benzoate (6a). Colorless oil: R
1R*,2R*,5S*,6S*)-5-Bromo-9-oxabicyclo[4.2.1]nonan-2-yl
(PE/EtOAc 9:1) 0.72; IR (neat)
) δ 8.08-8.03 (m, 2H),
31.1, 25.8, 25.1, 23.1; MS (CI , NH ) m/z 389, 387 (M þ NH ) ;
3
4
þ
79
5
þ
f
HRMS (CI , NH
3
) m/z calcd for C15
H
20
N
2
O
Br (M þ NH
4
)
-
1 1
713 cm ; H NMR (400 MHz, CDCl
1
7
4
1
3
387.0569, found 387.0566.
(1S*,4S*,5S*,8R*)-5-Bromo-9-oxabicyclo[6.1.0]nonan-4-yl
4-Nitrobenzoate (8b). White solid: mp 150 °C; R (PE/EtOAc
8:2) 0.33; IR (neat) 1708 cm ; H NMR (400 MHz, CDCl ) δ
.63-7.59 (m, 1H), 7.50-7.47 (m, 2H), 5.46-5.42 (m, 1H),
.85-4.80 (m, 1H), 4.69-4.65 (m, 1H), 4.26 (dt, J = 5.0, 10.0,
H), 2.37-1.91 (m, 8H); C NMR (100 MHz, CDCl
f
1
3
-1 1
3
) δ 165.6,
3
1
33.2, 130.0, 129.6, 128.5, 82.9, 78.4, 73.8, 53.6, 29.8, 29.5, 26.5,
5.8; MS (CI , NH ) m/z 344, 342 (M þ NH ) ; HRMS (CI ,
8.37-8.31 (m, 2H), 8.25-8.20 (m, 2H), 5.56 (dt, J = 3.9, 8.2,
1H), 4.83-4.70 (m, 1H), 3.18 (dt, J = 4.2, 9.1, 1H), 3.12 (dt, J =
4.2, 10.0, 1H), 2.43 (ddd, J = 4.3, 8.9, 15.7, 1H), 2.34-2.14 (m,
þ
þ
þ
2
NH
3
4
7
3
9
þ
3
) m/z calcd for C15
H
21NO
Br (M þ NH
4
) 342.0705,
1
3
found 342.0706.
4H), 2.10-1.98 (m, 1H), 1.88-1.69 (m, 2H); C NMR (100
(
Benzoate (7a). Colorless oil: R (PE/EtOAc 9:1) 0.73; IR (neat)
1R*,2R*,5R*,6R*)-6-Bromo-9-oxabicyclo[3.3.1]nonan-2-yl
MHz, CDCl ) δ 163.4, 150.8, 135.0, 130.7, 123.8, 75.0, 55.2,
3
þ
55.1, 50.5, 29.9, 28.8, 24.2, 23.5; MS (CI , NH ) m/z 389, 387
f
3
-
1
712 cm ; H NMR (400 MHz, CDCl
1
þ
þ
79
1
3
) δ 8.12-8.00 (m, 2H),
(M þ NH
4
) ; HRMS (CI , NH
3
) m/z calcd for C15
(M þ H) 387.0556, found 387.0570.
(1S*,4R*,5R*,8R*)-5-Bromo-9-oxabicyclo[6.1.0]nonan-4-yl
4-Nitrobenzoate (9b). White solid: mp 105 °C; R (PE/EtOAc
8:2) 0.25; IR (neat) 1710 cm ; H NMR (400 MHz, CDCl ) δ
20 2 5
H N O Br
þ
7
.60 (app t, J = 7.4, 1H), 7.48 (app t, J = 7.7, 2H), 5.48-5.35
(
m, 1H), 4.51 (dt, J = 5.6, 12.7, 1H), 4.17-4.12 (m, 1H), 4.02 (app t,
J = 4.9, 1H), 2.63-2.44 (m, 2H), 2.38-2.27 (m, 1H), 2.25-2.02
f
13
-1 1
(
m, 5H); C NMR (100 MHz, CDCl
3
) δ 165.6, 133.2, 130.0,
3
1
29.6, 128.5, 70.7, 69.8, 66.7, 50.6, 31.2, 25.9, 25.2, 23.2; MS
8.37-8.30 (m, 2H), 8.27-8.21 (m, 2H), 5.51 (td, J = 3.4, 9.3,
1H), 4.55 (ddd, J = 3.8, 5.2, 9.0, 1H), 3.23 (dt, J = 4.3, 8.5,
1H), 3.17 (dt, J = 4.6, 9.8, 1H), 2.57-2.43 (m, 2H), 2.42-2.21
(m, 3H), 2.10-1.95 (m, 1H), 1.93-1.82 (m, 1H), 1.73-1.59 (m,
þ
þ
þ
(
CI , NH ) m/z 344, 342 (M þ NH ) ; HRMS (CI , NH ) m/z
þ
18 3
3
4
3
7
9
calcd for C15
H
O
Br (M þ H) 325.0439, found 325.0439.
(
1S*,4S*,5S*,8R*)-5-Bromo-9-oxabicyclo[6.1.0]nonan-4-yl
1
3
Benzoate (8a). White solid: mp 113-115 °C; R (PE/EtOAc 9:1)
1H); C NMR (100 MHz, CDCl ) δ 163.6, 150.7, 135.2, 130.9,
f
3
-
1 1
þ
0
8
.20; IR (neat) 1704 cm ; H NMR (400 MHz, CDCl ) δ 8.06-
123.7, 74.9, 55.3, 54.6, 53.5, 32.1, 28.2, 25.3, 23.4; MS (CI ,
NH
3
þ
þ
.00 (m, 2H), 7.65-7.57 (m, 1H), 7.51-7.45 (m, 2H), 5.59 (ddd,
3
) m/z 389, 387 (M þ NH
4
) ; HRMS (CI , NH
3
)
7
9
þ
J = 3.3, 4.2, 7.9, 1H), 4.82 (ddd, J = 3.2, 6.1, 8.9, 1H), 3.21 (dt,
J = 4.0, 9.8, 1H), 3.15 (dt, J = 4.3, 9.9, 1H), 2.43 (ddd, J = 4.4,
m/z calcd for C15
387.0558.
H
20
2
N O
5
Br (M þ H) 387.0556, found
8
.7, 15.8, 1H), 2.32-2.17 (m, 3H), 2.16-1.96 (m, 2H), 1.85-1.68
2,3,4,5-Tetrafluorobenzoic Acid. Following the representative
procedure (above) using 2,3,4,5-tetrafluorobenzoic acid as the
added external nucleophile according to the conditions recorded
in Table 2, entry 5. The crude mixture was purified by column chro-
matography eluting with petroleum ether/EtOAc (9:1 to 8:2) to
give first, a 1:1 mixture of bicycloethers 6c and 7c, second,
epoxide 8c and third, epoxide 9c. The tetrafluorobenzoates 6c
and 7c were separated using HPLC on a 25 cm Supelcosil LC-Si
column (ID 21.2 mm), eluting with n-Hexane/EtOAc (90:10),
13
(
m, 2H); C NMR (100 MHz, CDCl
3
) δ 165.2, 133.5, 129.6,
þ
1
28.7, 73.5, 55.6, 55.5, 50.6, 29.0, 28.6, 23.8, 23.6; MS (CI ,
þ
þ
NH ) m/z 344, 342 (M þ NH ) ; HRMS (CI , NH ) m/z calcd
3
4
3
79
3
þ
for C15
H
18
O
Br (M þ H) 325.0439, found 325.0445.
(
1S*,4R*,5R*,8R*)-5-Bromo-9-oxabicyclo[6.1.0]nonan-4-yl
Benzoate (9a). White solid: mp 73-74 °C; R (PE/EtOAc 9:1)
.16; IR (neat) 1711 cm ; H NMR (400 MHz, CDCl ) δ 8.06
f
-1 1
0
3
(
dd, J = 1.2, 8.3, 2H), 7.64-7.58 (m, 1H), 7.50-7.45 (m, 2H),
.52 (td, J = 3.2, 8.6, 1H), 4.64 (dt, J = 4.4, 8.7, 1H), 3.23 (dt,
J = 4.2, 8.0, 1H), 3.17 (dt, J = 4.6, 9.4, 1H), 2.56-2.35 (m, 3H),
-
1
5
t t
flow rate 9 mL min , detecting at 254 nm. R 7c 14.8 min, R 6c
16.8 min.
2
1
1
.34-2.22 (m, 2H), 2.07-1.96 (m, 1H), 1.91-1.80 (m, 1H),
.74-1.63 (m, 1H); C NMR (100 MHz, CDCl
29.7, 128.5, 73.7, 55.7, 54.8, 52.5, 31.6, 27.7, 25.2, 23.2; MS
(1R*,2R*,5S*,6S*)-5-Bromo-9-oxabicyclo[4.2.1]nonan-2-yl
2,3,4,5-Tetrafluorobenzoate (6c). White solid: mp 80 °C; R (PE/
EtOAc 9:1) 0.33; IR (neat) 1716 cm ; H NMR (400 MHz,
CDCl ) δ 7.66-7.58 (m, 1H), 5.45 (dt, J = 4.2, 7.4, 1H), 4.78
(td, J = 2.9, 7.8, 1H), 4.67 (ddd, J = 7.6, 4.8, 2.6, 1H), 4.25-4.20
13
3
) δ 165.4, 133.5,
f
-
1 1
þ
þ
þ
(
CI , NH ) m/z 344, 342 (M þ NH ) ; HRMS (CI , NH ) m/z
3
3
4
3
7
9
þ
calcd for C H O Br (M þ H) 325.0439, found 325.0439.
1
5
18 3
1
3
p-Nitrobenzoic Acid. Following the representative procedure
above) using p-nitrobenzoic acid as the added external nucleo-
3
(m, 1H), 2.42-1.89 (m, 8H); C NMR (125 MHz, CDCl )
(
161.4, 147.9 (dd, J = 264, 13), 146.5 (dd, J = 238, 11), 143.7
(dt, J = 261, 12), 141.5 (dt, J = 259, 14), 114.7 (br m), 113.2 (dd,
J = 21, 3), 83.0, 78.0, 75.6, 53.2, 29.7, 29.4, 26.6, 25.8; MS (CI ,
phile according to the conditions recorded in Table 2, entry 3.
The crude mixture was purified by column chromatography
eluting with petroleum ether/EtOAc (9:1 to 8:2) to give first
þ
þ
þ
NH
for C15
3
) m/z 416, 414 (M þ NH
4
) ; HRMS (CI , NH
)
3
) m/z calcd
414.0328, found 414.0328.
(1R*,2R*,5R*,6R*)-6-Bromo-9-oxabicyclo[3.3.1]nonan-2-yl
2,3,4,5-Tetrafluorobenzoate (7c). White solid: mp 38-40 °C; R
(PE/EtOAc 9:1) 0.33; IR (neat) 1715 cm ; H NMR (400 MHz,
7
4
9
þ
[3.3.1]bicycloether 4-nitrobenzoate 7b, second, [4.2.1]bicycloether
H
17NO
3
F
Br (M þ NH
4
4-nitrobenzoate 6b, third epoxide 8b, and fourth, epoxide 9b.
(
1R*,2R*,5S*,6S*)-5-Bromo-9-oxabicyclo[4.2.1]nonan-2-yl
-Nitrobenzoate (6b). White solid: mp 85 °C; R (PE/EtOAc 8:2)
.43; IR (neat) 1720 cm ; H NMR (400 MHz, CDCl ) δ
.38-8.30 (m, 2H), 8.26-8.18 (m, 2H), 5.47 (ddd, J = 3.9, 4.5,
.8, 1H), 4.84 (td, J = 2.8, 8.5, 1H), 4.69 (ddd, J = 2.5, 4.8, 7.5,
H), 4.31-4.22 (m, 1H), 2.41-2.32 (m, 1H), 2.29-2.22 (m, 3H),
.48-2.03 (m, 3H), 2.02-1.93 (m, 1H); C NMR (125 MHz,
CDCl ) δ 163.7, 150.6, 135.3, 130.7, 123.6, 82.8, 78.1, 75.0, 53.2,
f
-
1 1
4
0
8
7
1
2
f
-1 1
3
3
CDCl ) δ 7.66-7.56 (m, 1H), 5.45-5.35 (m, 1H), 4.49 (dt, J =
5.5, 12.7, 1H), 4.13-4.08 (m, 1H), 4.00 (app t, J = 5.1,
1H), 2.62-2.53 (m, 1H), 2.50-2.40 (m, 1H), 2.36-2.28 (m,
1
3
1H), 2.25-2.15 (m, 1H), 2.15-2.02 (m, 4H); C NMR
(125 MHz, CDCl ) δ 161.3, 147.6 (dd, J = 260, 10), 146.5 (dd,
J = 249, 12), 143.5 (dt, J = 274, 16), 141.5 (dt, J = 255, 14),
1
3
3
3
þ
þ
29.8, 29.3, 26.4, 25.9; MS (CI , NH
HRMS (CI , NH
3
) m/z 389, 387 (M þ NH
4
) ;
114.7 (br m), 113.2 (dd, J = 21, 3), 72.4, 71.6, 69.8, 66.4, 50.2,
) m/z 416, 414 (M þ NH
þ
79
þ
þ
þ
4
3
) m/z calcd for C15
H
20
N
2
O
5
Br (M þ H)
31.0, 25.7, 25.1, 23.1; MS (CI , NH
3
) ;
Br (M þ
þ
79
4
3
87.0556, found 387.0558.
1R*,2R*,5R*,6R*)-6-Bromo-9-oxabicyclo[3.3.1]nonan-2-yl
-Nitrobenzoate (7b). White solid: mp 116 °C; R (PE/EtOAc
:2) 0.47; IR (neat) 1720 cm ; H NMR (400 MHz, CDCl ) δ
HRMS (CI , NH ) m/z calcd for C15
NH4) 414.0328, found 414.0331.
3
H17NO
3
F
þ
(
4
8
8
1
f
(1S*,4S*,5S*,8R*)-5-Bromo-9-oxabicyclo[6.1.0]nonan-4-yl
(PE/
EtOAc 9:1) 0.21; IR (neat) 1716 cm ; H NMR (400 MHz,
-
1 1
3
2,3,4,5-Tetrafluorobenzoate (8c). White solid: mp 45 °C; R
f
-
1 1
.37-8.29 (m, 2H), 8.26-8.18 (m, 2H), 5.44 (dt, J = 5.9, 10.5,
H), 4.50 (dt, J = 5.6, 11.8, 1H), 4.16 (app t, J = 5.7, 1H), 4.03
CDCl ) δ 7.64 (tdd, J = 2.5, 6.0, 10.4, 1H), 5.57 (dt, J = 3.7, 8.4,
3
(
2
app t, J = 5.0, 1H), 2.66-2.57 (m, 1H), 2.55-2.43 (m, 1H),
1H), 4.81-4.63 (m, 1H), 3.13-3.04 (m, 2H), 2.44-2.36
1
3
.39-2.30 (m, 1H), 2.29-2.05 (m, 5H); C NMR (125 MHz,
(m, 1H), 2.33-2.08 (m, 4H), 2.08-1.92 (m, 1H), 1.85-1.63
1
3
CDCl ) δ 163.7, 150.7, 135.3, 130.7, 123.6, 72.0, 69.8, 66.4, 50.1,
(m, 2H); C NMR (100 MHz, CDCl ) δ 147.7 (dd, J = 263, 12),
3
3
J. Org. Chem. Vol. 76, No. 1, 2011 103

Bonney et al.
JOCArticle
1
2
2
46.5 (dd, J = 250, 10), 143.7 (dt, J = 262, 13), 141.3 (dt, J =
56, 13), 114.2 (br m), 113.4 (dd, J = 21,1), 75.5, 55.4, 55.0, 50.3,
prepared from alcohols 6g and 7g respectively (See Supporting
Information).
2
4
þ
þ
9.5, 28.6, 24.1, 23.4; MS (CI , NH
3
) m/z 416, 414 (M þ NH
4
) ;
(1S*,2S*,5R*,6R*)-2-Bromo-5-methoxy-9-oxabicyclo[4.2.1]-
nonane (6e). Colorless oil: R (PE/EtOAc 4:1) 0.60; IR (neat)
þ
79
þ
HRMS (CI , NH ) m/z calcd for C H O F Br (M þ H)
3
15 14
3
4
f
-1 1
3
97.0062, found 397.0063.
3
1098, 1066 cm ; H NMR (500 MHz, CDCl ) δ 4.61 (ddd, J =
(
1S*,4R*,5R*,8R*)-5-Bromo-9-oxabicyclo[6.1.0]nonan-4-yl
,3,4,5-Tetrafluorobenzoate (9c). Colorless oil: R (PE/EtOAc
2.9, 7.1, 9.1, 1H), 4.58-4.54 (m, 1H), 4.24-4.13 (m, 1H), 3.65
2
9
7
8
2
1
f
(ddd, J = 3.6, 5.0, 7.1, 1H), 3.30 (s, 3H), 2.34-2.21 (m, 2H),
-
1
1
13
:1) 0.18; IR (neat) 1728 cm ; H NMR (400 MHz, CDCl ) δ
2.12-2.03 (m, 2H), 2.03-1.89 (m, 2H), 1.86-1.71 (m, 2H); C
3
.71-7.59 (m, 1H), 5.52 (td, J = 3.2, 8.8, 1H), 4.56 (dt, J = 4.4,
.8, 1H), 3.21 (dt, J = 4.3, 7.4, 1H), 3.15 (dt, J = 4.7, 9.8, 1H),
.56-2.21 (m, 5H), 2.07-1.98 (m, 1H), 1.91-1.79 (m, 1H),
.72-1.58 (m, 1H); C NMR (100 MHz, CDCl ) δ 160.9, 148.0
dd, J = 261, 9), 146.4 (dd, J = 226, 9), 144.0 (dt, J = 227, 24),
NMR (125 MHz, CDCl
28.4, 26.1, 25.1; MS (CI , NH
3
) δ 82.6, 80.4, 79.4, 57.1, 54.6, 29.7,
þ
þ
3
) m/z 254, 252 (M þ NH
4
) ;
þ
79
þ
HRMS (CI , NH ) m/z calcd for C H NO Br (M þ NH )
3
9
19
2
4
13
252.0599, found 252.0590.
(1R*,2R*,5R*,6R*)-2-Bromo-6-methoxy-9-oxabicyclo[3.3.1]-
nonane (7e). Colorless oil: R (PE/EtOAc 9:1) 0.22; IR (neat)
3
(
1
5
3
F
41.3 (dt, J = 237, 14), 114.4 (m), 113.3 (dd, J = 21, 2), 75.3,
5.5, 54.6, 52.5, 31.7, 29.7, 27.7, 25.0, 23.2; MS (CI , NH ) m/z
f
þ
-1 1
1127, 1103, 1082, 1044 cm ; H NMR (400 MHz, CDCl ) δ
3
3
þ
þ
þ
99, 397 (M þ H) ; HRMS (CI , NH
3
) m/z calcd for C15
Br (M þ H) 397.0062, found 397.0070.
Trimethylsilylchloride. Following the representative proce-
H
14
O
3
-
4.46 (dt, J = 5.5, 12.7, 1H), 4.01 (app t, J = 5.6, 1H), 3.93 (app t,
J = 5.5, 1H), 3.61 (dt, J = 5.9, 11.7, 1H), 3.38 (s, 3H), 2.49 (dd,
J = 5.9, 14.3, 1H), 2.41-2.28 (m, 1H), 2.21 (dt, J = 6.2, 13.7, 1H),
7
9
4
1
3
dure (above) using (trimethysilyl)chloride as the added external
nucleophile according to the conditions recorded in Table 2,
entry 6. The crude mixture was purified by column chromatog-
raphy eluting with petroleum ether/EtOAc (19:1) to afford an
inseparable mixture of bicyclic ethers 6d and 7d as a white solid.
The chlorides 6d and 7d were separated using HPLC on a 25 cm
Supelcosil LC-Si column (ID 21.2 mm), eluting with n-Hexane/
iPrOH (95:5), flow rate 10 mL min , detecting at 230 nm. R 6d
t
1.0 min; R
(
nonane (6d). Colorless oil: R
2.12-1.90 (m, 4H), 1.85-1.67 (m, 1H); C NMR (125 MHz,
CDCl ) δ 77.6, 69.9, 67.3, 56.3, 51.2, 31.3, 25.7, 25.1, 23.2; MS
3
þ
þ þ
4 3
) ; HRMS (CI , NH ) m/z
(CI , NH
3
) m/z 254, 252 (M þ NH
þ
4
7
2
9
calcd for C H NO Br (M þ NH ) 252.0599, found 252.0596.
9
19
Water. Following the representative procedure (above) using
water as the added nucleophile according to the conditions
recorded in Table 2, entries 10-12. The crude mixture was
purified by column chromatography to give an inseparable
-
1
2
4
4
t
7d 31.0 min.
1S*,2S*,5R*,6R*)-2-Bromo-5-chloro-9-oxabicyclo[4.2.1]-
mixture of the cyclic ethers 6g and 7g.
(1R*,2R*,5S*,6S*)-5-Bromo-9-oxabicyclo[4.2.1]nonan-2-ol (6g).
Colorless oil: R (PE/EtOAc 1:1) 0.36; IR (neat) 3379 cm ;
f
1
(PE/EtOAc 19:1) 0.48; H NMR
-1
f
1
(
500 MHz, CDCl ) δ 4.62-4.58 (m, 2H), 4.31-4.25 (m, 2H),
H NMR (400 MHz, CDCl ) δ 4.60-4.50 (m, 2H), 4.22 (dt, J =
3
3
1
3
2
NMR (125 MHz, CDCl
.44-2.35 (m, 2H), 2.25-2.18 (m, 2H), 2.18-2.00 (m, 4H);
C
5.4, 8.3, 1H), 4.17 (dt, J = 4.9, 6.8, 1H), 2.38-2.22 (m, 2H),
1
3
3
) δ 82.0, 81.5, 60.4, 53.4, 32.1, 31.2,
2.19-1.85 (m, 5H), 1.81 (s, 1H), 1.76-1.67 (m, 1H); C NMR
þ
þ.
2
7.4, 26.7; GC-MS (EI ) R
t
15.2 min: m/z 242, 240, 238 (M );
3
(100 MHz, CDCl ) δ 82.1, 80.9, 71.1, 54.3, 32.2, 29.7, 26.7, 24.9; MS
þ
35 79
þ
þ
þ
þ
HRMS (EI ) m/z calcd for C H O Cl Br (M þ H) 237.9760,
(CI , NH ) m/z 240, 238 (M þ NH ) ; HRMS (CI , NH ) m/z
8
12
3
4
3
79
2
þ
found 237.9755.
1R*,2R*,5R*,6R*)-2-Bromo-6-chloro-9-oxabicyclo[3.3.1]-
calcd for C
8
H17NO
Br (M þ NH
4
) 238.0443, found 238.0439.
(
(1R*,2R*,5R*,6R*)-6-Bromo-9-oxabicyclo[3.3.1]nonan-2-ol
(7g). Colorless oil: R (PE/EtOAc 1:1) 0.28; IR (neat) 3368 cm ;
f
H NMR (400 MHz, CDCl ) δ 4.44 (dt, J = 5.6, 12.7, 1H), 4.06
-
1
nonane (7d). White solid: mp 43-44 °C; R (PE/EtOAc 19:1)
.48; H NMR (500 MHz, CDCl ) δ 4.49-4.41 (m, 1H),
.35-4.25 (m, 1H), 3.95 (app t, J = 5.5, 2H), 2.52-2.46 (m,
H), 2.46-2.30 (m, 2H), 2.29-2.11 (m, 3H), 2.10-1.98 (m, 2H);
C NMR (125 MHz, CDCl ) δ 69.6, 69.6, 57.6, 50.3, 31.1, 30.3,
25.1, 24.9; GC-MS (EI ) R
HRMS (EI ) m/z calcd for C
found 237.9755.
Lithium Chloride. Following the representative procedure (above)
using (lithium) chloride as the added external nucleophile according
to the conditions recorded in Table 2, entry 7. The crude mixture was
purified by column chromatography eluting with petroleum ether/
EtOAc (19:1) to afford an inseparable mixture of bicyclic ethers 6d
and 7d as a white solid and epoxides 8d and 9d as an inseparable
mixture.
f
1
1
0
4
1
3
3
(dt, J = 5.8, 11.6, 1H), 3.91 (app t, J = 5.3, 1H), 3.85 (app t, J = 5.8,
1H), 2.51-2.43 (m, 1H), 2.40-2.28 (m, 1H), 2.26-2.16 (m, 2H),
1
3
13
2.15-2.08 (m, 1H), 2.05-1.93 (m, 3H), 1.90-1.81 (m, 1H);
C
3
þ
þ.
t
14.6 min: m/z 242, 240, 238 (M );
H
3
NMR (100 MHz, CDCl ) δ 69.7, 69.6, 68.4, 51.0, 31.3, 28.6, 24.6,
þ
35 79
12O Cl Br (M ) 237.9760,
þ.
þ
þ
þ
8
23.3; MS (CI ,NH
3
) m/z240, 238 (M þ NH
4
) ;HRMS(CI ,NH
) 238.0443, found 238.0434.
3
)
79
þ
4
m/zcalcd for C
8
H
17NO
2
Br (M þ NH
Acknowledgment. We thank the EPSRC (Grant no. EP/
F034253/1 to D.C.B.) and for a DTA (to K.B.). We thank P. R.
Haycock and R. N. Sheppard for the halogen-induced iso-
1
3
topic shift C NMR experiments.
(
1R*,4S*,5S*,8S*)-4-Bromo-5-chloro-9-oxabicyclo[6.1.0]nonane
8d) and (1R*,4R*,5R*,8S*)-4-bromo-5-chloro-9-oxabicyclo[6.1.0]-
nonane (9d). White solid: mp 112-113 °C; R (PE/EtOAc 19:1)
) resonances for 8d only
Supporting Information Available: General experimental,
procedure for dilution studies shown in Table 1 with H NMR
1
(
spectra, procedures for the saponification of tetrafluorobenzo-
ates 6c and 7c to provide alcohols 6g and 7g, procedures for the
preparation of authentic samples of 2, 3, 6a, 7a, 6e and 7e from
alcohols 6g and 7g, procedure and characterizing data for
bicycloethyl ethers 6f and 7f, procedure for saponification of
tetrafluorobenzoate 9c to bisepoxide 10, procedures and char-
acterizing data for the one-pot epoxidation-bromination of 1,5-
f
1
0
.32; H NMR (400 MHz, CDCl
3
δ 4.73-4.66 (m, 2H), 3.17 (dt, J = 3.9, 10.5, 1H), 3.10 (dt,
J = 4.6, 9.5, 1H), 2.68-2.55 (m, 2H), 2.27-2.14 (m, 3H),
1
3
2
.14-2.03 (m, 1H), 1.76-1.59 (m, 2H); C NMR (125 MHz,
CDCl ) resonances for 8d only: δ 60.5, 55.9, 54.9, 53.9, 31.8,
9.1, 24.0, 23.5; MS (CI ) m/z 243, 241, 239 (M þ H) ; HRMS
3
þ
þ
2
(
þ
35 79
þ
1
13
CI ) m/z calcd for C H
Cl BrO (M þ H) 238.9838, found
cyclooctadiene 11 to give 12 and 13, copies of H and C NMR
spectra for 2-5, 6a, 7a, 8a, 9a, 6b, 7b, 8b, 9b, 6c, 7c, 8c, 9c, 6d, 7d,
8
13
2
38.9838.
Methanol. Following the representative procedure (above)
1
3
8d, 6e, 7e, 6f and 7f, 6g, 7g, 10, 12, 13, copies of C NMR
resonances exhibiting bromine (2, 3, 4, 6a, 7a, 6b, 8b, 6c, 7c, 8c,
6d, 7d, 8d, 6e, 7e, 8e, 6g, 12, 13) and chlorine (6d, 7d)-induced
isotopic shifts, and X-ray crystallographic details for 6b, 6c, 7b,
7c, 7d and 8a-d. This material is available free of charge via the
Internet at http://pubs.acs.org.
using methanol as the added nucleophile according to the con-
ditions recorded in Table 2, entry 8. The crude mixture was
purified by column chromatography eluting with petroleum
ether/EtOAc (9:1 to 8:2) to give an inseparable mixture of the
cyclic ethers 6e and 7e. Authentic samples of 6e and 7e were
1
04 J. Org. Chem. Vol. 76, No. 1, 2011