Formation of a tetracyclic isoquinoline derivative by rearrangement of a [(bromophenyl)butyryl]oxazolidinone

Article abstract of DOI:10.1002/ejoc.200600918

Treatment of 3-[4-(2-bromophenyl)-2-phenylbutyryl]-4,4-dimethyloxazolidin- 2-one with LDA in THF launched a domino rearrangement sequence ending in the assembly of a tetracyclic cyclopentaoxazolo[3,2-b]isoquinolin-6-one derivative. Two mechanisms involvin

Full text of DOI:10.1002/ejoc.200600918

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200600918
Formation of a Tetracyclic Isoquinoline Derivative by Rearrangement of a
[(Bromophenyl)butyryl]oxazolidinone
Mark D. Roydhouse[a] and John C. Walton*[a]
Keywords: Rearrangement / Cyclisation / Heterocycles / Radical ions / EPR spectroscopy
Treatment of 3-[4-(2-bromophenyl)-2-phenylbutyryl]-4,4-di-
methyloxazolidin-2-one with LDA in THF launched a domino
rearrangement sequence ending in the assembly of a tetra-
cyclic cyclopentaoxazolo[3,2-b]isoquinolin-6-one derivative.
posed. EPR spectroscopic and 13C-labelling experiments sug-
gested that both were operative.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Two mechanisms involving an SRN1-type process were pro- Germany, 2007)
its acyl chloride and treated with 4,4-dimethyloxazolidin-2-
one to afford [(bromophenyl)butyryl]oxazolidinone (1a).
Introduction
Classic radical-mediated ring closures proceed by intra-
molecular additions onto alkene acceptors and usually yield
five-membered rings.[1] Recently, we have been studying a
novel cyclisation strategy that involves fast intramolecular
coupling of a radical and an anion (SRN1 process). This
strategy has previously been used by Wolfe and co-workers
to prepare, for example, isoquinolines and oxindoles from
haloaryl amides.[2] The process works for a different range
of functionality and has potential for the formation of qua-
ternary centres and larger ring sizes.[3] Rossi et al. have ably
reviewed the limited number of such cyclo-couplings in the
context of SRN1 processes in general.[4] We previously re-
ported that precursors containing bromoaryl groups at-
tached to oxazolines by propyl and butyl linkers yielded
indane and tetralin derivatives, respectively, when treated
with LDA.[5] We decided to investigate analogous precur-
sors containing oxazolidinones in the hope that more ef-
ficient and selective methodology could be developed. Ac-
cordingly, we carried out exploratory experiments with
[(bromophenyl)butyryl]oxazolidinone derivatives 1, but
found that, in practice, LDA-promoted reactions of 1 led to
the assembly of a new fused tetracyclic cyclopentaoxazolo-
isoquinoline system in a remarkable multistage rearrange-
ment.
Good conditions for the promotion of ring closure via
RN1 processes involve the use of 3 equiv. of LDA in
S
THF.[2d,5] When the oxazolidinone 1a was treated with
LDA in this way at –78 °C, and the solution was sub-
sequently warmed to room temperature and stirred for 48 h,
a waxy-white solid was isolated after column chromatog-
raphy (75%). It was deduced that the structure was
cyclopenta-oxazolo-isoquinolin-6-one
derivative
2a
(Scheme 1) from a careful study of the COSY, HSQC and
HMBC spectra, and of model systems.[6] Two new ste-
reocentres were created during the rearrangement so that
four stereoisomers were possible. However, two of these
with cis HO and Ph groups were shown by DFT computa-
tions to be 46 kJmol–1 higher in energy. Furthermore, their
formation would be strongly sterically disfavoured during
the ring-closure steps. Thus 2a was a 50:50 mixture of only
Results and Discussion
The precursor oxazolidinone was prepared by treatment
of 2-(2-bromophenyl)ethyl iodide with the dianion derived
from phenylacetic acid. The resulting acid was converted to
[a] University of St. Andrews, School of Chemistry,
EaStChem, St. Andrews, Fife, KY16 9ST, U.K.
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
Scheme 1.
Eur. J. Org. Chem. 2007, 1059–1063
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1059

M. D. Roydhouse, J. C. Walton
SHORT COMMUNICATION
the (10R,10aR) and (10S,10aS) enantiomers. The fused
tetracyclic structure must have been assembled by a rather
novel rearrangement.
There are several possibilities for the mechanism of the
rearrangement and two of these are outlined in Scheme 2.[7]
By analogy with previous SRN1 reactions involving oxazol-
ines, we expected enolate formation from 1, accompanied
by electron transfer and fast loss of bromide, to give the
radical anion intermediate 3. Intramolecular cyclo-coupling
of this enolate with the internal aryl radical would yield a
dihydroindene radical anion 4. Path (a) depicts a 1,3-arene-
otropic migration of the oxazolidinone group, yielding reso-
nance-stabilized radical anion 5, followed by a 6-exo-trig
nucleophilic cyclisation onto the oxazolin-ring carbonyl to
afford the radical anion 6. 1,5-Hydrogen transfer from the
bis(allylic) H atom to the oxyanion in 6 will yield the tetra-
cyclic radical anion 7ai. Alternatively, 6 would be readily
deprotonated by the LDA to yield the radical dianion 7aii.
Electron transfer from 7ai or 7aii to more of the anion de-
rived from precursor 1 and LDA, i.e. 1–, will yield the tet-
racycle 2, simultaneously propagating the SRN1-type chain.
In the case of 7aii a final protonation during work-up yields
2. We are not aware of any previous reports of 1,3-acyl-type
group migrations of this type for radical anions. However,
Yoshida et al. have described related migrations following
attack on arynes by acyl anions.[8a,8b] Clearly, the first steps
in our rearrangement could be formation of an aryne and
ring closure by intramolecular nucleophilic addition to give
an aryl carbanion analogous to 4. Path (a) could then pro-
ceed more or less as shown, except that the H atom and
electron-transfer steps would not be necessary. The penulti-
mate intermediate would be an oxy anion analogous to 6
(without the cyclohexadienyl H atom) which would pick up
a proton during work-up to yield 2a.
Scheme 2.
Alternatively, product 2a could have been formed
through path (b) which shows a 6-exo-trig cyclisation
within 4 onto the oxazolidinone ring carbonyl yielding in-
termediate 8. Ring opening of the oxazolidinone, subse-
quent re-cyclisation and proton transfer lead to 7bi and
thence to the observed product 2 after electron transfer to
more of the anion derived from 1, i.e. 1–. Again, reaction
might proceed by deprotonation of 9 to yield the analogous
radical dianion 7bii. An analogous path starting from an
aryne intermediate is also possible.
were appropriate for a radical anion, but it was not obvious
which intermediate was responsible for the main spectrum.
DFT computations, with the UB3LYP functional,[10] were
implemented for the three species 4, 5 and 7i. Only the com-
puted parameters for radical anion 7i showed satisfactory
Both mechanisms of Scheme 2 involve radical anion in-
termediates and, in the hope of spectroscopically charac-
terising them, a smaller-scale reaction of 1a with LDA in
THF was carried out under identical conditions and sam-
ples were removed under nitrogen and examined by 9-GHz
EPR spectroscopy. Broad, weak spectra were observed be-
fore irradiation, but they became stronger and better re-
solved in the temperature range 240–195 K, after a few min-
utes UV photolysis (Figure 1). The spectrum persisted over
many hours. The computer simulation (Figure 1 and
Table 1) of the main component indicated two fairly large
H atom hyperfine splittings (hfs) together with five small
hfs and suggested that a broad central signal was also pres-
Figure 1. 9-GHz EPR spectrum obtained on treatment of 1a with
LDA in THF at 215 K with UV photolysis. Lower panel: computer
ent.[9] The g factor (2.0032) and the magnitudes of the hfs simulation.
1060
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1059–1063

Ring-Closure Reaction of Bromophenylbutyryl Oxazolidinone
SHORT COMMUNICATION
agreement with experiment. In 7i the unpaired electron re- bicyclo[3.2.0]heptyl-type species 11 and 13, which might of
sides mainly in the indane aromatic ring and hence H(3) course be transition states, would contain large amounts of
and H(5) of this ring have fairly large hfs.[11]
strain so the absence of this process is easily explained.
Table 1. Experimental and computed hfs (G)[a] for radical anion 7i.
H(3) H(5) H(2) H(1) H(4) H(1Ј) H(11)
7i exptl.[b] 7.7
4.5
–3.5
1.4
–0.8
1.0
0.7
1.0
0.7
1.0
0.5
1.0
–0.4
7i DFT[c] –7.3
[a] 1 G = 0.1 mT. [b] Data at 215 K in THF; g = 2.0032, tentative
assignments of hfs to specific
311++G(d,p)//UB3LYP/6-31G*.
H atoms. [c] UB3LYP/6-
The EPR data, together with the sensitivity of the reac-
tion to light and oxygen, support the SRN1-like mechanisms
of path (a) or (b), rather than aryne-mediated routes, al-
though the latter cannot be definitely ruled out. The con-
centration of the radical anion, measured by double inte-
gration of the EPR signal, was 0.54 m at 215 K and
changed little with temperature. The ratio of the radical
anion concentration to initial precursor [7]/[1a] was
0.96%.[12] This is consistent with the chain process of
Scheme 2 with a moderate chain length. Note that radical
anion 7 is formed in the penultimate H atom transfer step
of path (a) and of path (b), and hence the EPR experiment
cannot distinguish between these two pathways.
Scheme 3.
The fact that some 13C was found attached to the benzo
ring of 2b [C(6)] proves that some arenotropic migration
[route (a)] must have taken place, but obviously this was
accompanied by the cyclisation/ring opening route (b).
According to path (a), the butyryl carbonyl C atom ends
up attached to the benzo ring as C(6) in 7a, whereas, ac-
cording to path (b), the butyryl carbonyl C atom remains
where it is and becomes C(9) of 7b (Scheme 2). Potentially
the two mechanistic pathways could be distinguished by
labelling this C atom. Accordingly, precursor 1b labelled
with 13C at the butanoyl carbonyl was prepared from
PhCH213CO2H. When 1b was treated with LDA in THF
under the same conditions, tetracycle 2b was isolated. We
were able to obtain crystals of 2b, and X-ray diffraction
confirmed the tetracyclic structure (see the supporting in-
formation; for supporting information see also the footnote
on the first page of this paper).
Conclusions
We conclude that ring closure of bromophenyl oxazolidi-
none 1 takes place efficiently, even though this involves for-
mation of a quaternary centre. The presence and siting of
the two carbonyl groups facilitate further rearrangement,
which results in the one-pot assembly of an unusual iso-
quinolinone derivative with an alkyl bridge between the
benzo and heterocyclic rings. The initial steps were most
likely similar to those reported for the analogous oxazol-
ine[5] and led to the radical anion 3, which closed by SRN1-
type cyclo-coupling to produce 4. An analogous process via
initial aryne formation could not be ruled out. By analogy
with the oxazoline system, isoquinolinones with other
bridge sizes should be accessible.
The 13C label was shown to be distributed to both C(6)
and C(9) in 2b (ratio 52% to 48%). This seemed to indicate
that both pathways were operative. However, the fact that
the 13C distribution was close to 50:50 opened the possibil-
ity that a pre-equilibration of the label between the but-
anoyl and oxazolidinyl carbonyl groups might have taken
place via a symmetrical intermediate. Such a mechanism,
involving symmetrical four-membered-ring intermediate 12,
Experimental Section
3-[4-(2-Bromophenyl)-2-phenylbutyryl]-4,4-dimethyloxazolidin-2-one
(1a): To a solution of 4,4-dimethyloxazolidin-2-one[13] (0.65 g,
5.6 mmol) in dry distilled THF (17 mL) under nitrogen at –78 °C
is shown in Scheme 3. If this pre-equilibrium were opera- was added 2.5  nBuLi solution in hexanes (2.24 mL, 5.6 mmol).
tional, the 13C label should be scrambled to both the but-
After stirring for 10 min, 4-(2-bromophenyl)-2-phenylbutyryl chlo-
ride (1.9 mL, 5.6 mmol) in THF (5 mL) was added with a syringe.
The resulting nearly colourless solution was stirred for 30 min at
anoyl and oxazolidinyl carbon atoms in unreacted precursor
1b remaining at the end of the reaction. A separate reaction
–78 °C and then warmed to room temperature within 30 min. Ex-
of 1b with LDA under identical conditions was therefore
cess acyl chloride was quenched by addition of saturated ammo-
carried out but stopped after 3 h, leaving about 1/3 of the
nium chloride solution (10 mL). The THF and hexane were evapo-
precursor unreacted. However, it was found that all the 13
C
rated and the resultant slurry extracted with dichloromethane
(2ϫ15 mL). The combined organic layers were washed with NaOH
solution (10 mL) and brine (10 mL). After drying (Na2SO4) and
removal of the solvent, the crude product was purified by column
label in the unreacted 1b remained at the butanoyl carbon
and none was found in the oxazolidine ring, although prod-
uct 2b again showed ca. 50:50 scrambling. This result indi-
cated that pre-equilibration of label did not occur. The aza- chromatography (SiO2, hexanes/EtOAc, 9:1) to give pure 1a as a
Eur. J. Org. Chem. 2007, 1059–1063
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1061

M. D. Roydhouse, J. C. Walton
SHORT COMMUNICATION
12
clear oil (1.5 g, 64%); Rf = 0.1 (SiO2, hexanes/EtOAc, 9:1). IR (nu-
(%) = 417 (100) [M + H+]. Calcd. for
(MH)+: 417.0895; found 417.0893.
C
13CH2379BrNO3
20
1
jol): νmax = 1698 (C=O), 1778 (C=O). H NMR: δ = 1.39 (s, 3 H),
˜
1.58 (s, 3 H), 1.98–2.14 (m, 1 H), 2.33–2.45 (m, 1 H), 2.51–2.76 (m,
2 H), 3.76 (d, J = 8.4 Hz, 1 H), 3.91 (d, J = 8.4 Hz, 1 H), 5.03 (t,
J = 7.4 Hz, 1 H), 6.98–7.03 (m, 1 H), 7.14–7.37 (m, 1 H), 7.46–
7.49 (m, 1 H) ppm. 13C NMR: δ = 24.2 (CH3), 25.2 (CH3), 34.0
(CH2), 34.2 (CH2), 49.9 (CH), 60.6 (C), 74.9 (CH2), 124.4 (C),
127.3 (CH), 127.5 (CH), 127.7 (CH), 128.6 (CH), 128.7 (CH), 130.4
(CH), 132.8 (CH), 138.8 (C), 141.0 (C), 153.6 (C), 174.5 (C) ppm.
MS (CI): m/z (%) = 416 (100) [M + H+]. Calcd. for C21H2379BrNO3
(MH)+: 416.0861; found 416.0855.
10a-Hydroxy-8,8-dimethyl-10b-phenyl-8,9,10a,10b-tetrahydro-1H-
cyclopenta[de]oxazolo[3,2-b]isoquinolin-6(2H)-one (2a): A solution
of the oxazolidinone 1a (0.5 g, 1.2 mmol) in THF (3.5 mL) was
added to a solution of LDA in THF (7.5 mL, 0.48 ) at –78 °C
under nitrogen. After stirring for 10 min the solution was warmed
to room temperature within 30 min, at which time more THF
(21 mL) was added. The mixture was then stirred 48 h at room
temp. After this time a saturated solution of ammonium chloride
(4 mL) was added and the aqueous layer was extracted with diethyl
ether (3ϫ6 mL). The combined organic layers were washed with
water (10 mL) and dried (MgSO4) to give the crude product. Col-
umn chromatography (SiO2, hexanes/THF, 9:1) yielded pure 2a as
a white wax (301 mg, 75%); Rf = 0.2 (SiO2, hexanes/EtOAc, 9:1).
1H NMR: δ = 0.70 (s, 3 H), 1.60 (s, 3 H), 2.64–2.69 (m, 1 H), 2.76–
2.85 (m, 1 H), 2.91–2.99 (m, 2 H), 3.13 (br. s, 1 H), 3.42 (d, J =
8.2 Hz, 1 H), 3.94 (d, J = 8.2 Hz, 1 H), 7.07–7.10 (m, 2 H), 7.17–
7.21 (m, 3 H), 7.40–7.45 (m, 2 H), and 7.80–7.82, (m, 1 H) ppm.
13C NMR: δ = 22.4 (C13), 25.0 (C13), 31.3 (C1), 36.3 (C2), 58.9
(C7), 59.5 (C9b), 78.1 (C8), 112.4 (C9a), 123.8 (C5), 127.1 (C12),
127.4 (C11), 128.0 (C5a), 128.5 (C3), 128.8 (C4), 129.6 (C10), 139.1
(C9d), 142.5 (C2a), 145.8 (C9c) and 162.6 (C6) ppm. MS (CI): m/z
(%) = 336 (100) [M + H+]. Calcd. for C21H22NO3 (MH)+: 336.1600;
found 336.1604.
[1-13C]4-(2-Bromophenyl)-2-phenylbutyric Acid and Corresponding
Acyl Chloride: To a solution of [1-13C]phenylacetic acid (1 g,
7.3 mmol) in THF (30 mL), under N2 at –78 °C was added care-
fully dropwise 2 equiv. of nBuLi (5.9 mL, 14.7 mmol). After stirring
for 20 min 2-bromophenethyl iodide (2.3 g, 7.3 mmol) in THF
(5 mL) was added dropwise over 5 min. The mixture was warmed
to room temp. over 2 h and stirred overnight. The mixture was
quenched by addition of solid ammonium chloride and the solvent
removed. The residue was dissolved in 2  HCl (100 mL), and the
cloudy mixture extracted with dichloromethane (3ϫ50 mL). The
combined organic layers were dried (MgSO4) and concentrated to
give the crude acid which was purified by column chromatography
(SiO2, hexanes/EtOAc, 9:1) to give a clear oil, (502 mg, 22%); Rf
1
= 0.2 (SiO2, hexanes/EtOAc, 9:1). H NMR: δ = 2.05–2.19 (m, 1
H), 2.33–2.47 (m, 1 H), 2.62–2.78 (m, 2 H), 3.62 (apparent q, J =
7.6 Hz, 1 H), 7.04 (ddd, J = 7.8, 6.8, 2.0 Hz, 1 H), 7.12–7.37 (m, 7
H), 7.50 (dd, J = 7.8, 1.0 Hz, 1 H) ppm. 13C NMR: δ = 32.8 (CH2),
34.0 (CH2), 51.0 (d, J = 49.0 Hz, CH), 124.4 (C), 127.5 (CH), 127.7
(CH), 127.8 (CH), 128.2 (CH), 128.8 (CH), 130.5 (CH), 132.9
(CH), 137.9 (C), 140.5 (C) and 179.5 (13C, enhanced) ppm. The
acyl chloride was prepared by addition of oxalyl chloride (0.3 mL,
3.12 mmol) dropwise over 20 min to the acid (0.5 g, 1.56 mmol)
and DMF (3 drops) in DCM (5 mL) at 0 °C. The resulting mixture
was stirred at room temp. for 2 h, followed by concentration to give
the crude product which was dissolved in dry distilled THF and
used immediately.
The reaction of 1a (250 mg, 0.6 mmol) with LDA in THF was re-
peated on half the above scale in a quartz flask. After warming the
solution to room temp. it was photolysed for 3 h with light from a
400-W medium-pressure Hg lamp placed ca. 20 cm from the flask.
1
After similar work-up, the H NMR spectrum the whole product
(186 mg) showed it to consist of a 50:50 mixture of 2a and unre-
acted 1a together with moderate amounts of unidentified by-prod-
ucts. The results were consistent with faster formation of 2a but
also diversion of the reaction into other channels.
13C-Labelled 10a-Hydroxy-8,8-dimethyl-10b-phenyl-8,9,10a,10b-
tetrahydro-1H-cyclopenta-[de]oxazolo[3,2-b]isoquinolin-6(2H)-one
(2b): A solution of oxazolidinone 1b (320 mg, 0.76 mmol) in THF
[2-13C]3-[4-(2-Bromophenyl)-2-phenylbutyryl]-4,4-dimethyloxazol- (2 mL) was added to a solution of LDA in THF (0.48 , 6 mL) at
idin-2-one (1b): To a solution of 4,4-dimethyloxazolidinone (0.18 g,
1.56 mmol) in dry, distilled THF (5 mL) under nitrogen at –78 °C
was added 2.5  nBuLi solution in hexanes (0.63 mL, 1.56 mmol).
–78 °C under nitrogen. After stirring for 10 min the solution was
allowed to warm to room temp. over 30 min, at which time more
THF (15 mL) was added. The mixture was then stirred 48 h at
After stirring for 10 min, [1-13C]4-(2-bromophenyl)-2-phenylbu- room temp. After this time a saturated solution of ammonium chlo-
tyryl chloride (0.53 g, 1.56 mmol) in THF (2 mL) was added with
a syringe. The resulting nearly colourless solution was stirred for
30 min at –78 °C and then allowed to warm to room temperature
over 30 min. Excess acyl chloride was quenched by addition of sat-
urated ammonium chloride solution (5 mL). The THF and hexane
were removed in vacuo and the resultant slurry extracted with
dichloromethane (2ϫ10 mL). The combined organic layers were
washed with NaOH solution (5 mL) and brine (5 mL). After drying
(Na2SO4) and removal of the solvent, the crude product was puri-
fied by column chromatography (SiO2, hexanes/EtOAc, 9:1) to give
pure 1b as a clear oil (81%); Rf = 0.1 (SiO2, hexanes/EtOAc, 9:1).
ride (3 mL) was added and the aqueous layer was extracted with
diethyl ether (3ϫ4 mL). The combined organic layers were washed
with water (8 mL) and dried (MgSO4) to give the crude product.
Column chromatography (SiO2, hexanes/THF, 9:1) yielded pure 2b
as clear needles (38% after recrystallisation from DCM/hexane)
1
m.p. 210–211 °C; Rf = 0.2 (SiO2, hexanes/EtOAc, 9:1). H NMR:
δ = 0.70 (s, 3 H), 1.60 (s, 3 H), 2.64–2.69 (m, 1 H), 2.76–2.85 (m,
1 H), 2.91–2.99 (m, 2 H), 3.13 (br. s, 1 H), 3.42 (d, J = 8.2 Hz, 1
H, 50% of signal shows JHC = 5.8), 3.94 (d, J = 8.2 Hz, 1 H), 7.07–
7.10 (m, 2 H), 7.17–7.21 (m, 3 H), 7.40–7.45 (m, 2 H), and 7.80–
7.82, (m, 1 H) ppm. 13C NMR: δ = 22.4 (C13), 25.0 (C13), 31.3
IR (film): ν
= 1660 (C=O), 1774 (C=O) cm–1. 1H NMR: δ = (C1), 36.3 (C2), 58.9 (C7), 59.5 (C9b), 78.1 (C8), 112.4 (≈ 50%
˜
max
1.39 (s, 3 H), 1.58 (s, 3 H), 1.98–2.14 (m, 1 H), 2.33–2.45 (m, 1 H), enhanced, 13C9a), 123.8 (C5), 127.1 (C12), 127.4 (C11), 128.0
2.51–2.76 (m, 2 H), 3.76 (d, J = 8.4 Hz, 1 H), 3.91 (d, J = 8.4 Hz, (C5a), 128.5 (C3), 128.8 (C4), 129.6 (C10), 139.1 (C9d), 142.5
1 H), 5.03 (apparent q, J = 7.1 Hz, 1 H), 6.98–7.03 (m, 1 H), 7.14– (C2a), 145.8 (C9c) and 162.6 (≈ 50% enhanced 13C6) ppm. The
7.37 (m, 1 H) and 7.46–7.49 (m, 1 H) ppm. 13C NMR: δ = 24.2 proportions of the 13C label at the two sites were determined from
(CH3), 25.2 (CH3), 34.0 (CH2), 34.2 (CH2), 49.9 (d, J = 49.9 Hz,
CH), 60.6 (C), 74.9 (CH2), 124.4 (C), 127.3 (CH), 127.5 (CH), 127.7
the 13C NMR spectrum observed with inverse gated decoupling so
as to provide valid integral values [52% 13C(6) and 48% 13C(9a)]
(CH), 128.6 (CH), 128.7 (CH), 130.4 (CH), 132.8 (CH), 138.8 (C), and from integration of the two sets of multiplets at δ = 3.42 ppm
1
141.0 (C), 153.6 (C) and 174.5 (enhanced, 13C) ppm. MS (CI): m/z
in the H NMR spectrum [51% 13C(6) and 49% 13C(9a)]. One set
1062
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1059–1063

Ring-Closure Reaction of Bromophenylbutyryl Oxazolidinone
SHORT COMMUNICATION
showed coupling to the 13C nucleus and the other did not (see sup-
porting information). The tetracyclic structure was confirmed by
X-ray diffraction which showed the Ph and OH groups to be trans.
435–443; c) J.-W. Wong, K. J. Natalie Jr, G. C. Pisipati, P. T.
Flaherty, T. D. Greenwood, J. F. Wolfe, J. Org. Chem. 1997, 62,
6152–6159; d) S. A. Dandekar, S. N. Greenwood, T. D. Green-
wood, S. Mabic, J. S. Merola, J. M. Tanko, J. F. Wolfe, J. Org.
Chem. 1999, 64, 1543–1553.
CCDC-606950 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
[3] J. C. Walton, Top. Curr. Chem. 2006, 264, 163–200.
[4] R. A. Rossi, A. B. Pierini, A. B. Peñéñory, Chem. Rev. 2003,
103, 71–167.
[5] M. D. Roydhouse, J. C. Walton, Chem. Commun. 2005, 4453–
Supporting Information (see also the footnote on the first page of
this article): General experimental methods, preparative details of
1a,b and for their rearrangements, X-ray structure for 2a, NMR
spectra for 1a,b, 1D and 2D NMR spectra for 2a,b.
4455.
[6] M. Prein, P. J. Manley, A. Padwa, Tetrahedron 1997, 53, 7777–
7794.
[7] We thank a referee for important mechanistic suggestions.
[8] a) H. Yoshida, M. Watanabe, J. Ohshita, A. Kunai, Chem.
Commun. 2005, 3292–3294; b) H. Yoshida, M. Watanabe, J.
Ohshita, A. Kunai, Tetrahedron Lett. 2005, 46, 6729–6731.
[9] The broad feature could be due to radical anion 4 or 5.
[10] M. J. Frisch, et al., Gaussian 03, revision A1, Gaussian Inc.
Pittsburg PA, 2003 (see Supporting Information for full ci-
tation).
[11] The EPR spectrum of radical dianion 7aii (or its lithium salt)
is expected to be virtually identical to that of 7ai and hence the
EPR spectrum does not exclude this mechanistic possibility.
[12] Assuming the spectrum corresponds to 7i, i.e. ignoring the con-
tribution from the broad component.
Acknowledgments
We thank the Engineering and Physical Sciences Research Council
(EPSRC) for financial support and Professor A. M. Z. Slawin for
the X-ray structure.
[1] S. Z. Zard, Radical Reactions in Organic Synthesis, Oxford Uni-
versity Press, Oxford, UK, 2003 and references contained
therein.
[2] a) J. F. Wolfe, M. C. Sleevi, R. R. Goehring, J. Am. Chem. Soc.
1980, 102, 3646–3647; b) R. R. Goehring, Y. P. Sachdeva, J. S.
Pisipati, M. C. Sleevi, J. F. Wolfe, J. Am. Chem. Soc. 1985, 107,
[13] S. Jones, C. Smanmoo, Tetrahedron Lett. 2004, 45, 1585–1588.
Received: October 20, 2006
Published Online: January 12, 2007
Eur. J. Org. Chem. 2007, 1059–1063
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1063