A novel boron trifluoride etherate mediated deep-seated rearrangement of an α,β-epoxyketone

Article abstract of DOI:10.1016/j.tetasy.2005.07.030

Acid catalysed reaction of carvone epoxide 2 resulted in dimeric products 3 and 4, in contrast to the expected ring contraction product. Reaction of β-methylcarvone epoxides 8 and 11 with acids furnished 2-acetyl-4- isopropenylcyclopentanones 9 and 14 containing a stereodefined quaternary carbon atom. On the other hand, the reaction of epoxides 8 and 11 with boron trifluoride etherate lacks the stereoselectivity and in addition, anti-epoxide 8 furnished lactone 18 via an unusual deep seated rearrangement.

Full text of DOI:10.1016/j.tetasy.2005.07.030

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 2973–2979
A novel boron trifluoride etherate mediated
deep-seated rearrangement of an a,b-epoxyketone
Adusumilli Srikrishna* and Sripada S. V. Ramasastry
Department of Organic Chemistry, Indian Institute of Science, Bangalore 560 012, India
Received 21 July 2005; accepted 25 July 2005
Available online 19August 2005
Abstract—Acid catalysed reaction of carvone epoxide 2 resulted in dimeric products 3 and 4, in contrast to the expected ring con-
traction product. Reaction of b-methylcarvone epoxides 8 and 11 with acids furnished 2-acetyl-4-isopropenylcyclopentanones 9 and
14 containing a stereodefined quaternary carbon atom. On the other hand, the reaction of epoxides 8 and 11 with boron trifluoride
etherate lacks the stereoselectivity and in addition, anti-epoxide 8 furnished lactone 18 via an unusual deep seated rearrangement.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
This includes the nature of the substituents on the epox-
ide, the stereochemistry at the epoxide oxygen, and the
nature of the Lewis acid employed.1,2 For the develop-
ment of a convenient procedure for the enantioselective
generation of 2-acetylcyclopentanones, we have investi-
gated the acid catalysed rearrangement of the epoxides
derived from carvone. These investigations revealed
some unexpected rearrangements depending on the acid
employed and the temperature of the reaction. Herein,
we report the details of these investigations.
Epoxides are versatile intermediates in organic synthe-
sis. The inherent polarity and strain of the three-mem-
bered ring makes epoxides susceptible to reactions
with a large number of reagents. The synthetically useful
reactions of epoxides include inter- and intramolecular
nucleophilic ring opening, reduction to alcohols, deoxy-
genation to alkenes and more importantly, rearrange-
ments to carbonyl compounds and allylic alcohols.
Among the variety of functionalised epoxides, a,b-epoxy
ketones have proven to be interesting and useful sub-
strates in organic synthesis. The presence of a carbonyl
group on the epoxide carbon provides the possibility
for regio- and stereoselective transformations. In partic-
ular, acid catalysed rearrangement of appropriately
substituted a,b-epoxy ketones to 1,3-dicarbonyl com-
pounds has attracted the attention of various research
groups (Eq. 1). When a,b-epoxy ketones are treated with
Lewis acids, exclusive 1,2-carbonyl migration occurs in
many cases if the carbonyl p-bond can achieve an orbital
alignment that can result in delocalisation. When geo-
metric constraints prevent the involvement of the car-
bonyl p-bond, participation of the Lewis acid catalyst
can result in halohydrin formation. It is generally
accepted that mono- and 1,1-disubstituted epoxides,
give aldehydes on rearrangement with acids, but with
1,2-disubstituted, tri- and tetrasubstituted epoxides a
variety of factors determine the outcome of the reaction.
O
O
R’
H+
R
O
O
ð1Þ
R
R’
2. Results and discussion
It was expected that the Lewis acid mediated rearrange-
ment of carvone epoxide 2 would generate optically
active 2-acetyl-4-isopropenylcyclopentanone, a useful
chiron in natural product synthesis. However, contrary
to our expectation, reaction of carvone epoxide 2 with
boron trifluoride etherate failed to generate the ring con-
tracted product and furnished, exclusively, the dimeric
compounds 3 and 4, whose structures were deduced
from their spectral data.3 The reaction was found to
be much more efficient with Amberlyst-15 in refluxing
benzene and generated the dimers 3 and 4 in 77 and
20% yield, respectively. Formation of the dimers can
be readily explained as depicted in Scheme 1. Acid
*
Corresponding author. Tel.: +91 80 22932215; fax: +91 80 23600683/
23600529; e-mail: ask@orgchem.iisc.ernet.in
0957-4166/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2005.07.030

2974
A.Srikrishna, S.S.V.Ramasastry / Tetrahedron:
Asymmetry 16 (2005) 2973–2979
O
O
O
Y
O
O
OH
b or c
a
X
Not Formed
1
3 X = OH; Y =H
4 X = H; Y = OH
2
Scheme 1. Reagents and conditions: (a) 30% H2O2, 6 M NaOH, MeOH, 0 ꢁC, 2 h, 70%; (b) BF3ÆEt2O (excess), 0.1 M CH2Cl2, rt, 10 min, 3, 51% and
4, 24%; (c) Amberlyst-15 (1:1 weight equivalent), 0.1 M C6H6, reflux, 2.5 h, 3, 77% and 4, 20%.
mediated rearrangement of epoxide 2 generates a ter-
tiary allyl alcohol, which on dehydration generates sty-
rene 5. Dimerisation of the styrene 5 via cyclisation (at
the para- or ortho-positions to the hydroxy group) of
carbonium ion 6 generates the regioisomeric aryl inda-
nols 3 and 4 (Scheme 2). As expected, dimers 3 and 4
were found to be racemic.
structure was established from its spectral data. The
reaction was found to be very slow with sub-stoichio-
metric amounts of p-TSA. The reaction was also found
to proceed smoothly with camphorsulfonic acid (CSA)
and Amberlyst-15 (Table 1). Stereochemistry at the qua-
ternary carbon of dione 9 was assigned on the basis of
the mechanism and was supported by the NOESY spec-
trum. To confirm the stereostructure, diketone 9 was
transformed into a crystalline derivative. Thus, the reac-
tion of diketone 9 with p-toluenesulfonyl hydrazide fur-
nished the hydroxypyrazole derivative 10 (Scheme 4),
whose single crystal X-ray diffraction analysis, Figure
1, unambiguously established the stereostructure of
diketone 9.
H
O
HO
O
OH
O
2
H
H
5
In order to obtain the other diastereoisomer of 9, the
reaction was carried out with the syn-epoxide 11, which
was obtained by hydroxy directed epoxidation of allyl
alcohol 12. Thus, the stereoselective reduction6 of
b-methylcarvone 7 with lithium aluminium hydride
(LAH) at low temperature furnished the syn-allyl
alcohol 12. Reaction of allyl alcohol 12 with m-chloro-
perbenzoic acid (m-CPBA) in methylene chloride
furnished the epoxyalcohol 13 in a regio- and stereo-
selective manner, which on oxidation with pyridinium
chlorochromate (PCC) and sodium acetate in methylene
chloride generated the syn-epoxide 11. As anticipated,
the reaction of syn-epoxide 11 with an excess of p-TSA
in methylene chloride at room temperature for 8 h fur-
nished b-diketone 14 in 65% yield along with a trace
amount of epimeric b-diketone 9. Conversely, the reac-
tion of the epoxide 11 with Amberlyst-15 in methylene
chloride was found to be very clean and furnished b-
diketone 14 in 78% yield in a highly stereoselective
manner. The results are summarised in Table 1.
OH
OH
5
3 + 4
HO
6
Scheme 2.
It was reasoned that the presence of a substituent at the
b-position of carvone 1 would alter the course of the
reaction, as the intermediate tertiary carbonium ion
would be more favoured, and in addition also lead to
acetylcyclopentanes containing a stereodefined quater-
nary carbon atom. Reaction of b-methylcarvone4 7 with
30% hydrogen peroxide and sodium hydroxide in meth-
anol furnished the anti-epoxide 8 (Scheme 3).5 As anti-
cipated, reaction of anti-epoxide 8 with an excess of
p-toluenesulfonic acid (p-TSA) in methylene chloride
at room temperature for 4.5 h furnished b-diketone 9
in 74% yield in a highly stereoselective manner, whose
O
O
O
O
O
O
O
O
b
a
c
c
93%
78%
d
1
7
8
9
92%
O
O
O
HO
HO
f
e
72%
75%
12
13
11
14
Scheme 3. Reagents and conditions: (a) (i) MeMgI, Et2O, 0 ꢁC ! rt, 1 h; (ii) PCC, silica gel, CH2Cl2, rt 4 h; (b) 30% H2O2, 6 N NaOH, MeOH, 6 h;
(c) see Table 1; (d) LAH, Et2O, À70 ꢁC, 1 h; (e) m-CPBA, NaHCO3, CH2Cl2, 0 ꢁC ! rt, 2 h; (f) PCC, NaOAc, CH2Cl2, rt, 3 h.

A.Srikrishna, S.S.V.Ramasastry / Tetrahedron:
Table 1. Reactions of the ketoepoxides 8 and 11
Asymmetry 16 (2005) 2973–2979
2975
Entry
Epoxide
Reaction conditions
Products (yield, %)
a
b
c
d
e
f
g
h
i
8
8
p-TSA, rt, CH2Cl2, 4.5 h
Amberlyst-15, CH2Cl2, rt, 10 h
CSA, CH2Cl2, rt, 24 h
9 (74)
9 (65)
9 (73)
14 (65)a
14 (78)
15 (35), 16 (34)
17 (78)
8
11
11
8
p-TSA, rt, CH2Cl2, 8 h
Amberlyst-15, CH2Cl2, rt, 13 h
TiCl4, CH2Cl2, À70 ! 0 ꢁC, 3 h
TiCl4, CH2Cl2, À70 ! 0 ꢁC, 3 h
BF3ÆEt2O, CH2Cl2, À70 ꢁC, 4 h
BF3ÆEt2O, CH2Cl2, À70 ꢁC ! rt, 7 h
BF3ÆEt2O, CH2Cl2, 0 ꢁC ! rt, 14 h
BF3ÆEt2O, CH2Cl2, 0 ꢁC ! rt, 3 h
11
8
9 (60), 14 (29)
8
9 (20), 14 (10), 18 (35)
9 (51), 14 (22), 18 (12)
9 (24), 14 (70)
j
k
8
11
a Minor amount (ꢀ5%) of the dione 9 was also obtained.
tal X-ray diffraction analysis.7 On the other hand, in
contrast to the literature reports,1,8 the reaction of
anti-epoxide 8 with boron trifluoride etherate was found
to be nonstereoselective and temperature dependent.
For example, treatment of epoxide 8 with 1 equiv of
boron trifluoride etherate in methylene chloride at
À70 ꢁC for 4 h furnished a 2:1 mixture of diketones 9
and 14. On the other hand, the addition of boron triflu-
oride etherate at À70 ꢁC and then allowing the reaction
to reach room temperature over a period of 7 h fur-
nished an unusual product, lactone 18 in 35% yield,
along with 30% yield of a 2:1 mixture of the diketones
9 and 14. However, the addition of 30 mol % of boron
trifluoride etherate at 0 ꢁC and then carrying out the
reaction at room temperature for 14 h furnished dike-
tones 9 and 14, and lactone 18 in 51, 22 and 12% yields,
respectively. In a similar manner, the reaction of syn-
epoxide 11 with 30 mol % of boron trifluoride etherate
in methylene chloride furnished diketones 9 and 14 in
94% yield in 1:3 ratio. Results are summarised in Table
1. Formation of both diketones 9 and 14 from the epox-
ides 8 and 11 with boron trifluoride etherate indicates
that the reaction proceeds through a carbonium ion,
and the anchimeric assistance of the isopropenyl group
also plays a crucial role in the product distribution.
O
O
SO2
N
OH
N
TsNHNH2
9
10
Scheme 4.
Figure 1. ORTEP of the sulfonamide 10.
Next, reactions were explored with Lewis acids. Reac-
tion of epoxides 8 and 11 with titanium tetrachloride
failed to generate the b-diketones 9 and 14 (Scheme 5).
Treatment of anti-epoxide 8 with 30 mol % of titanium
tetrachloride furnished a 1:1 mixture of chlorohydrin
15 and allyl alcohol 16 in 70% yield. Conversely, the
reaction of syn-epoxide 11 under the same conditions
furnished cleanly chlorohydrin 17, mp 77–79 ꢁC, in
78% yield, whose structure was confirmed by single crys-
The structure of lactone 18 was deduced from its spec-
tral data, in particular, the presence of a carbonyl
absorption band at 1773 cmÀ1 in the IR spectrum, the
lactone carbon resonance at d 175.3 in the 13C NMR
spectrum and analysis of the 1H NMR spectrum. In
order to confirm the structure of lactone 18, it was trea-
ted with 1 equiv of bromine to furnish dibromide 19
O
O
O
OH
OH
O
O
O
Cl
O
BF3.Et2O
TiCl4
O
8
+
+
+
H
9
14
18
16
15
O
O
O
O
OH
O
Cl
TiCl4
BF3.Et2O
11
+
14
9
17
Scheme 5.

2976
A.Srikrishna, S.S.V.Ramasastry / Tetrahedron:
Asymmetry 16 (2005) 2973–2979
(Scheme 6). Single crystal X-ray diffraction analysis
(Fig. 2) of dibromide 19 unambiguously established
the structure of lactone 18.
410 spectrophotometer. 1H (300 MHz) and 13C
(75 MHz) spectra were recorded on JNM k-300 spec-
trometer. Samples were prepared using a 1:1 mixture of
CDCl3 and CCl4 as solvent for recording the NMR spec-
tra. The chemical shifts (d ppm) and coupling constants
(Hz) are reported in the standard fashion with reference
to either internal tetramethylsilane (for 1H) or the central
line (77.1 ppm) of CDCl3 (for 13C). In the 13C NMR
spectra, the nature of the carbons (C, CH, CH2 or
CH3) were determined by recording the DEPT-135 and
are given in parentheses. Low-resolution mass spectra
were recorded using a Shimadzu QP-5050A GC–MS
instrument using a direct inlet (EI) mode. Relative inten-
sities are given in parentheses. High resolution mass spec-
tra were recorded on a Micromass Q-TOF micro mass
spectrometer using electron spray ionization mode. Opti-
cal rotations were measured using a Jasco DIP-370
digital polarimeter and [a]D values are given in units of
10À1 deg cm2 gÀ1. Acmeꢀs silica gel (100–200 mesh) was
used for column chromatography (approximately 15–
20 g per 1 g of the crude product). Dry ether was
obtained by distillation over sodium and stored over
sodium wire. Dry methylene chloride was prepared by
distilling over calcium hydride. All the commercial
reagents were used as such without further purification.
Br
O
O
Br2, CCl4
O
O
H
18
Br
H
19
Scheme 6.
Figure 2. ORTEP diagram of 19.
The formation of lactone 18 from anti-epoxide 8, obvi-
ously, involves a deep-seated rearrangement. A probable
mechanism is depicted in Scheme 7. Intramolecular
nucleophilic attack of the isopropenyl olefin on to boron
trifluoride coordinated epoxide in 8 from the anti face
furnishes the bicyclo[2.2.2]octyl carbonium ion 20.
Intramolecular addition of oxygen to ketone group fol-
lowed by ring cleavage and trapping of the carbonium
ion by the resultant carboxylate 21 furnishes lactone
18. The mechanism is partially supported by the absence
of a similar product under identical conditions from
epoxide 11, which contains the epoxy oxygen and the
isopropenyl group on the same side.
4.1. 3-(3-Hydroxy-4-methylphenyl)-1,1,3,5-tetramethyl-
indan-4-ol 4 and 1-(3-hydroxy-4-methylphenyl)-1,3,3,6-
tetramethylindan-5-ol 4
To a magnetically stirred solution of carvone epoxide 2
(40 mg, 0.24 mmol) in dry benzene (2.4 mL, 0.1 M) was
added Amberlyst-15 (40 mg) and refluxed for 2.5 h. The
reaction mixture was cooled and the resin filtered off
using a sintered funnel. Evaporation of the solvent and
purification of the residue on a silica gel column using
ethyl acetate–hexane (1:30) as eluent furnished, first
the minor diol 4 (7 mg, 20% as oil. IR (neat):
m
max/cmÀ1 3521, 1664, 1587. 1H NMR (300 MHz,
CDCl3 + CCl4): d 7.04 (1H, d, J 7.8 Hz), 7.00 (1H, d,
J 7.2 Hz), 6.80 (1H, dd, J 7.8 and 1.8 Hz), 6.64 (1H, d,
J 7.2 Hz), 6.63 (1H, d, J 1.8 Hz), 4.63 (1H, br s), 4.20
(1H, br s), 2.25 and 2.13 (2H, 2 · d, J 13.2 Hz, H-2),
2.20 (3H, s), 2.14 (3H, s), 1.72 (3H, s), 1.34 (3H, s),
1.26 (3H, s). 13C NMR (75 MHz, CDCl3 + CCl4): d
154.4 (C), 151.2 (C), 150.1 (C), 147.9(C), 134.0 (C),
131.5 (CH), 130.9(CH), 123.1 (C), 122.3 (C), 118.2
(CH), 114.4 (CH), 113.0 (CH), 61.1 (CH2), 49.6 (C),
43.4 (C), 31.73 (CH3), 31.68 (CH3), 27.0 (CH3), 15.5
(2C, CH3). Mass: m/z 296 (M+, 47), 281 (100, MÀ15),
189(12), 173 (94), 158 (10), 133 (10), 121 (16). HRMS:
m/z Calcd for C20H24O2Na (M+Na): 319.1674. Found:
319.1692.
3. Conclusion
In conclusion, we have developed a convenient enantio-
selective route to 2-acetylcyclopentanones containing a
stereodefined quaternary carbon atom starting from
the readily and abundantly available monoterpene (R)-
carvone, while also observing a novel rearrangement.
4. Experimental
Melting points are recorded using Tempo and Mettler
FP1 melting point apparatus in capillary tubes and are
uncorrected. IR spectra were recorded on a Jasco FTIR
BF3
O
O
O
F3B--O
O
O--BF3
8
18
21
20
Scheme 7.

A.Srikrishna, S.S.V.Ramasastry / Tetrahedron:
Asymmetry 16 (2005) 2973–2979
2977
Further elution of the column using ethyl acetate–hex-
ane (1:20) as eluent furnished the major diol 3 (28 mg,
78%) as a solid, which was recrystallised from ethanol.
20.8 (CH3). Mass: m/z 180 (M+, 5), 162 (10), 137 (48),
123 (45), 109 (25), 96 (85), 95 (100). HRMS: m/z Calcd
for C11H16O2Na (M+Na): 203.1048. Found: 203.1049.
Mp: 177–178 ꢁC [lit.3 mp: 179–180 ꢁC]. IR (neat): mmax
/
cmÀ1 3403. 1H NMR (300 MHz, CDCl3 + CCl4): d
6.93 (1H, d, J 7.8 Hz), 6.78 (1H, s), 6.64 (1H, dd, J 7.8
and 1.8 Hz), 6.51 (1H, s), 6.47 (1H, d, J 1.8 Hz), 4.80
(1H, br s), 2.32 and 2.10 (2H, 2 · d, J 12.9Hz), 2.21
(3H, s), 2.17 (3H, s), 1.58 (3H, s), 1.26 (3H, s), 1.00
(3H, s), 1.43 (1H, s). 13C NMR (75 MHz, CDCl3 +
CCl4): d 153.5 (C), 153.4 (C), 151.4 (C), 150.8 (C),
140.8 (C), 130.6 (CH), 127.0 (CH), 122.1 (C), 120.6
(C), 119.0 (CH), 113.5 (CH), 108.7 (CH), 59.7 (CH2),
50.0 (C), 42.7 (C), 31.1 (CH3), 30.8 (CH3), 30.5 (CH3),
16.2 (CH3), 15.4 (CH3). Mass: m/z 296 (M+, 19), 281
(100, M–15), 189()9, 173 (40), 133 (10), 121 (10).
HRMS: m/z Calcd for C20H24O2Na (M+Na):
319.1674. Found: 319.1670. Anal. Calcd for C20H24O2:
C, 81.04; H, 8.16. Found: C, 80.81; H, 8.27.
4.4. (+)-(1R,5S,7S)-7-Isopropenyl-4,5-dimethyl-2-
[(4-methylphenyl)sulfonyl]-2,3-diazabicyclo[3.3.0]-
oct-3-en-1-ol 10
To a magnetically stirred solution of diketone 9 (18 mg,
0.1 mmol) in dry MeOH (1 mL) was added tosyl hydra-
zide (47 mg, 0.25 mmol) and refluxed for 2 h. Evapora-
tion of the solvent under reduced pressure followed by
purification of the residue on a silica gel column using
ethyl acetate–hexane (1:10–1:8) as eluent furnished pyr-
azole 10 (17 mg, 50%) as a colourless solid, which was
recrystallised from a mixture of CH2Cl2–hexanes. Mp:
24
154–155 ꢁC. ½aŠ ¼ þ21.9( c 1.51, CHCl3). IR (neat):
D
1
m
max/cmÀ1 3482, 1645, 1598, 892. H NMR (300 MHz,
CDCl3 + CCl4): d 7.83 (2H, d, J 8.4 Hz), 7.28 (2H, d,
J 8.4 Hz), 4.68 (1H, s), 4.60 (1H, s), 3.71 (1H, br s),
2.85–2.70 (1H, m), 2.43 (3H, s), 2.39(1H, ddd, J 12.3,
6.0 and 2.4 Hz), 2.04 (1H, t, J 12.9Hz), 1.88 (3H, s),
1.78 (1H, ddd, J 12.3, 6.3 and 2.1 Hz), 1.65 (3H, s),
1.54 (1H, t, J 12.3 Hz), 1.15 (3H, s). 13C NMR
(75 MHz, CDCl3 + CCl4): d 161.4 (C), 145.0 (C), 143.5
(C), 136.6 (C), 129.3 (2C, CH), 128.2 (2C, CH), 110.1
4.2. (À)-(1S,4S,6S)-4-Isopropenyl-1,6-dimethyl-7-oxabi-
cyclo[4.1.0]heptan-2-one 8
To an ice cold, magnetically stirred solution of (S)-3-
methylcarvone 7 (1 g, 6.1 mmol) in methanol (10 mL)
was added 30% aq H2O2 (8 mL), followed by a freshly
prepared ice cold solution of 6 M aq NaOH (2 mL) drop-
wise over a period of 20 min. The reaction mixture was
stirred for 2 h at the same temperature. It was then diluted
with water and extracted with ether (3 · 10 mL). The
combined organic extract was washed with brine and
dried over Na2SO4. Evaporation of the solvent and puri-
fication of the residue on a silica gel column using
CH2Cl2–hexane (1:5) as eluent furnished epoxide 8
(856 mg, 78%) as oil.5 IR (neat): mmax/cmÀ1 1706, 1646,
(CH2), 103.2 (C), 62.0 (C), 46.9(CH ), 43.8 (CH), 42.5
2
(CH2), 21.7 (CH3), 21.2 (CH3), 17.0 (CH3), 12.8
(CH3). HRMS: m/z Calcd for C18H23N2O2S (MÀOH):
331.1474. Found: 331.1474. Anal. Calcd for C18H24-
N2O3S: C, 62.07; H, 6.89; N, 8.04; S, 9.19. Found: C,
62.18; H, 6.83; N, 7.99; S, 9.14.
4.5. Crystal data for 10
1
893; H NMR (300 MHz, CDCl3 + CCl4): d 4.68 (1H,
X-ray data were collected at 293 K on a SMART
CCD-BRUKER diffractometer with graphite-mono-
s), 4.63 (1H, s), 2.72–2.55 (1H, m), 2.48 (1H, ddd, J
18.0, 5.1 and 1.5 Hz), 2.13 (1H, dd, J 14.4 and 4.2 Hz),
1.90 (1H, dd, J 18.0 and 12.0 Hz), 1.78 (1H, dd, J 17.4
and 11.7 Hz), 1.65 (3H, s), 1.40 (3H, s), 1.32 (3H, s). 13C
NMR (75 MHz, CDCl3 + CCl4): d 204.8 (C), 146.4 (C),
110.4 (CH2), 64.0 (C), 63.4 (C), 41.5 (CH2), 35.3 (CH2),
35.0 (CH, C-4), 20.6 (CH3), 19.6 (CH3), 11.7 (CH3).
˚
chromated Mo Ka radiation (k = 0.71073 A). The struc-
ture was solved by direct methods (SIR92). Refinement
was done by full-matrix least-squares procedures on F2
using SHELXL-97. The nonhydrogen atoms were refined
anisotropically whereas hydrogen atoms were refined
isotropically. C18H24N2O3S; MW = 348.45; colourless
crystal; crystal system: orthorhombic; space group
˚
4.3. (À)-(2S,4S)-2-Acetyl-4-isopropenyl-2-methylcyclo-
P2(1)2(1)2(1); cell parameters, a = 8.5749(23) A, b =
3
˚
˚
˚
pentanone 9
11.8643(32) A, c = 18.1951(49) A; V = 1851.08 A , Z =
4, q = 1.25 g cmÀ3, F(000) = 743.9, l = 0.192 mmÀ1
.
To a magnetically stirred solution of keto-epoxide 8
(95 mg, 0.53 mmol) in dry CH2Cl2 (5.3 mL, 0.1 M) was
added Amberlyst-15 (95 mg) and stirred for 10 h at rt.
The reaction mixture was then filtered through a small
Celite pad. Evaporation of the solvent and purification
of the residue on a silica gel column using ethyl ace-
Total number of l.s. parameters = 222, R1 = 0.046 for
3432 Fo > 4r(Fo) and 0.050 for all 3692 data.
wR2 = 0.107, GOF = 1.138, restrained GOF = 1.138
for all data. An ORTEP diagram of 10 with 50% ellip-
soidal probability has been shown in Figure 1. Crystal-
lographic data has been deposited with Cambridge
Crystallographic Data Centre (CCDC 260156).
tate–hexane (1:50) as eluent furnished diketone 9
23
(62 mg, 65%) as oil. ½aŠ ¼ À192.9 (c 1.82, CHCl3). IR
D
(neat): mmax/cmÀ1 1741, 1704, 1648, 894. 1H NMR
(400 MHz, CDCl3 + CCl4): d 4.84 (1H, s), 4.79(1H,
s), 2.77 (1H, tt, J 12.0 and 5.7 Hz), 2.56–2.45 (2H, m),
2.32 (1H, dd, J 17.6 and 12.0 Hz), 2.24 (3H, s), 1.92
(1H, ddd, J 13.0, 6.2 and 2.0 Hz), 1.79(3H, s), 1.37
(3H, s). 13C NMR (75 MHz, CDCl3 + CCl4): d 214.4
(C), 205.0 (C), 145.3 (C), 110.6 (CH2), 63.8 (C), 43.6
(CH2), 39.2 (CH), 37.8 (CH2), 26.3 (CH3), 21.1 (CH3),
4.6. (+)-(1R,4S,6R)-4-Isopropenyl-1,6-dimethyl-7-oxabi-
cyclo[4.1.0]heptan-2-one 11
To an ice cold, magnetically stirred solution of allyl
alcohol 12 (460 mg, 2.80 mmol) in dry CH2Cl2 (4 mL)
were added solid NaHCO3 (ꢀ50 mg) and 70% MCPBA
(70% assay, 820 mg, 3.32 mmol). The reaction mixture
was slowly warmed to rt and stirred for 2 h protected

2978
A.Srikrishna, S.S.V.Ramasastry / Tetrahedron:
Asymmetry 16 (2005) 2973–2979
from light. It was then washed with saturated aq
Na2SO3 solution (5 mL) and extracted with CH2Cl2
(3 · 3 mL). The combined organic extract was washed
with brine and dried over Na2SO4. Evaporation of the
solvent and purification of the residue on a silica gel col-
umn using ethyl acetate–hexane (1:20–1:10) as eluent
(3.3 mL, 0.1 M) was added a solution of TiCl4 (0.1 mL,
1 M in CH2Cl2, 0.1 mmol). The reaction mixture was
then allowed to warm to 0 ꢁC over a period of 3 h and
quenched with saturated aq NaHCO3 solution (5 mL).
The organic layer was separated, washed with brine
and dried over Na2SO4. Evaporation of the solvent and
purification of the residue on a silica gel column using
furnished the epoxy alcohol 13 (375 mg, 75%) as oil.
26
½aŠ ¼ þ24.5 (c 0.98, CHCl3). IR (neat): mmax/cmÀ1
CH2Cl2–hexane (1:4) as eluent furnished chlorohydrin
D
26
1
3435, 1645, 890. H NMR (300 MHz, CDCl3 + CCl4):
d 4.64 (2H, s), 3.74–3.64 (1H, m), 2.72–2.62 (1H, m),
2.08–1.90 (1H, m), 1.80–1.60 (3H, m), 1.65 (3H, s),
1.36 (3H, s), 1.32 (3H, s). 13C NMR (75 MHz,
CDCl3 + CCl4): d 147.3 (C), 109.7 (CH2), 72.4 (CH),
64.9(C), 64.7 (C), 39.8 (CH), 35.2 (CH 2), 33.6 (CH2),
21.5 (CH3), 20.0 (CH3), 15.1 (CH3). HRMS: m/z Calcd
for C11H16O2Na (M+Na): 203.1048. Found: 203.1047.
15 (25 mg, 35%) as oil. ½aŠ ¼ þ49.5 (c 3.11, CHCl3).
D
1
IR (neat): mmax/cmÀ1 3480, 1718, 1646, 896. H NMR
(300 MHz, CDCl3 + CCl4): d 4.85 (1H, s), 4.80 (1H,
s), 3.92 (1H, br s), 2.71–2.13 (5H, m), 1.78 (3H, s),
1.54 (3H, s), 1.50 (3H, s). 13C NMR (75 MHz,
CDCl3 + CCl4): d 210.1 (C), 145.0 (C), 111.5 (CH2),
81.7 (C), 75.3 (C), 44.9(CH ), 40.6 (CH2), 40.3 (CH),
2
25.6 (CH3), 24.2 (CH3), 20.5 (CH3). Mass: 216 (M+, 3),
173 (14), 139(23), 137 (25), 130 (12), 123 (20), 121 (14),
To a magnetically stirred solution of epoxy-alcohol 13
(120 mg, 0.66 mmol) in dry CH2Cl2 (2 mL) was added
finely powdered NaOAc (426 mg, 2.5 equiv to PCC) fol-
lowed by PCC (426 mg, 1.97 mmol) in portions. The
reaction mixture was stirred for 5 h at rt. It was then
filtered through a small silica gel column using excess
ether. Evaporation of the solvent and rapid purification
of the residue on a silica gel column using ethyl acetate–
109(18), 59 (47), 85 (100). HRMS:
C11H17ClO2Na (M+Na): 239.0815. Found: 239.0810.
m/z Calcd for
Further elution of the column using CH2Cl2–hexane
(1:1) furnished allyl alcohol 16 (20 mg, 34%) as oil.
26
½aŠ ¼ À34.5 (c 1.68, CHCl3). IR (neat): mmax/cmÀ1
D
3486, 1720, 1645, 896. 1H NMR (300 MHz,
CDCl3 + CCl4): d 5.30 (1H, br s), 4.81 (1H, s), 4.80
(1H, s), 3.57 (1H, s), 3.22–3.13 (1H, m), 2.74 (1H, dd,
J 12.3 and 10.2 Hz), 2.59(1H, ddd, J 12.3, 6.3 and
0.9Hz), 1.85 (3H, t, J 1.8 Hz), 1.75 (3H, s), 1.46 (3H,
s). 13C NMR (75 MHz, CDCl3 + CCl4): d 211.0 (C),
146.3 (C), 140.4 (C, C-3), 124.6 (CH), 111.8 (CH2),
74.8 (C), 46.5 (CH), 39.5 (CH2), 27.4 (CH3), 19.7
(CH3), 17.1 (CH3). Mass: 181 (M+ + 1, 4), 151 (28),
137 (43), 123 (100), 109(54), 95 (71). HRMS: m/z Calcd
for C11H16O2Na (M+Na): 203.1048. Found: 203.1040.
hexane (1:5) as eluent furnished the epoxy ketone 11
26
(85 mg, 72%) as oil. ½aŠ ¼ þ21.5 (c 1.21, CHCl3). IR
D
(neat):
m
max/cmÀ1 1714, 1646, 895. 1H NMR
(300 MHz, CDCl3 + CCl4): d 4.71 (2H, s), 2.83 (1H,
dd, J 14.1 and 12.3 Hz), 2.60–2.45 (1H, m), 2.12–1.90
(3H, m), 1.70 (3H, s), 1.42 (3H, s), 1.36 (3H, s). 13C
NMR (75 MHz, CDCl3 + CCl4): d 207.6 (C), 146.0
(C), 110.8 (CH2), 68.1 (C), 63.8 (C), 43.5 (CH), 39.5
(CH2), 34.8 (CH2), 21.3 (CH3), 19.5 (CH3), 11.3 (CH3).
4.7. (À)-(2R,4S)-2-Acetyl-4-isopropenyl-2-methylcyclo-
4.9. (À)-(2S,3S,5R)-3-Chloro-2-hydroxy-5-isopropenyl-
pentanone 14
2,3-dimethylcyclohexanone 17
Reaction of keto-epoxide 11 (70 mg, 0.4 mmol) in dry
CH2Cl2 (4 mL, 0.1 M) with PTSA (380 mg, 2 mmol),
as described for diketone 9, and purification of the prod-
uct on a silica gel column using ethyl acetate–hexane
Reaction of keto-epoxide 11 (80 mg, 0.44 mmol) in dry
CH2Cl2 (4.4 mL, 0.1 M) with TiCl4 (0.13 mL, 1 M solu-
tion in CH2Cl2, 0.13 mmol) for 3 h as described above,
and purification on a silica gel column using ethyl ace-
tate–hexane (1:50) as eluent furnished chlorohydrin 17
(1:50) as eluent furnished diketone 14 (45 mg, 65%) as
23
D
an oil. ½aŠ ¼ À107.9( c 1.14, CHCl3). IR (neat): mmax
/
(75 mg, 78%) as a solid, which was crystallised from
24
cmÀ1 1742, 1705, 1648, 890. 1H NMR (400 MHz,
CDCl3 + CCl4): d 4.80 (1H, s), 4.75 (1H, s), 2.84 (1H,
ddd, J 12.3, 6.2 and 2.4 Hz), 2.71 (1H, tt, J 18.3 and
7.5 Hz), 2.50 (1H, ddd, J 18.3, 7.5 and 2.3 Hz), 2.19
(1H, dd, J 18.3 and 12.0 Hz), 2.20 (3H, s), 1.77 (3H,
s), 1.48 (1H, t, J 12.0 Hz), 1.39(3H, s). 13C NMR
(75 MHz, CDCl3 + CCl4): d 214.2 (C), 203.9(C), 145.9
(C), 109.9 (CH2), 66.0 (C), 42.8 (CH2), 40.0 (CH), 39.0
hexanes. Mp: 77–79 ꢁC. ½aŠ ¼ À13.8 (c 3.9, CHCl3).
D
1
IR (neat): mmax/cmÀ1 3473, 1715, 1646, 895. H NMR
(300 MHz, CDCl3 + CCl4): d 4.79(2H, s), 2.94 (1H, d,
J 12.9Hz), 2.90–2.75 (2H, m), 2.60 (1H, br s), 2.41 (1H,
dd, J 14.0 and 11.7 Hz), 2.27 (1H, m of d, J 11.7 Hz),
1.88 (1H, m of d, J 14.0 Hz), 1.77 (3H, s), 1.68 (3H, s),
1.45 (3H, s). 13C NMR (75 MHz, CDCl3 + CCl4): d
208.3 (C), 146.3 (C), 110.8 (CH2), 78.5 (C), 74.9(C),
(CH2), 25.6 (CH3), 21.9(CH ), 21.1 (CH3). Mass: m/z
40.9(CH ), 40.44 (CH), 40.41 (CH2), 26.0 (CH3), 20.7
3
2
180 (M+, 3), 138 (15), 137 (30), 123 (27), 109(12), 96
(50), 95 (60). HRMS: m/z Calcd for C11H17O2 (M+1):
181.1228. Found: 181.1228.
(CH3), 18.1 (CH3). Mass: m/z 183 (M+ÀH2OÀCH3, 3),
149(18), 139(20), 137 (27), 123 (26), 109(28), 95 (46).
HRMS: m/z Calcd for C11H17ClO2Na (M+Na):
239.0815. Found: 239.0816. Anal. Calcd for C11H17ClO2:
C, 60.97; H, 7.91. Found: C, 60.71; H, 8.02.
4.8. (+)-(2R,3R,5R)-3-Chloro-2-hydroxy-5-isopropenyl-
2,3-dimethylcyclohexanone 15 and (À)-(2S,5R)-2-
hydroxy-5-isopropenyl-2,3-dimethylcyclohex-3-enone 16
4.10. Rearrangement of the epoxide 8 with BF3ÆEt2O
To a cold (À70 ꢁC), magnetically stirred solution of the
keto-epoxide 8 (60 mg, 0.33 mmol) in dry CH2Cl2
To a cold (À70 ꢁC), magnetically stirred solution of
keto-epoxide 8 (90 mg, 0.5 mmol) in dry CH2Cl2

A.Srikrishna, S.S.V.Ramasastry / Tetrahedron:
Asymmetry 16 (2005) 2973–2979
2979
(5 mL, 0.1 M) was added a solution of freshly distilled
BF3ÆEt2O (0.5 mL, 1 M in CH2Cl2, 0.5 mmol). The reac-
tion mixture was then allowed to warm to rt over a per-
iod of 7 h and the reaction quenched with saturated aq
NaHCO3 solution (5 mL). The organic layer was sepa-
rated, washed with brine and dried over Na2SO4. Evap-
oration of the solvent and purification of the residue on
a silica gel column using CH2Cl2–hexane (1:10) as eluent
was by full-matrix least-squares procedures on F2 using
SHELXL-97. The nonhydrogen atoms were refined
anisotropically whereas hydrogen atoms were refined
isotropically. C11H16Br2O2; MW = 340.06; colourless
crystal; crystal system: orthorhombic; space group
˚
P2(1)2(1)2(1); cell parameters, a = 7.5789(20) A, b =
3
˚
˚
˚
9.2141(24) A, c = 18.4287(47) A; V = 1286.93(6) A ,
Z = 4, q = 1.75 g cmÀ3
mmÀ1
Total number of l.s. parameters = 139,
,
F(000) = 671.9, l = 6.279
furnished, first diketones 14 (19mg, 21%) and diketone
9 (28 mg, 31%) as oils. Further elution of the column
.
R1 = 0.034 for 2187 Fo > 4r(Fo) and 0.046 for all 2604
data. wR2 = 0.080, GOF = 1.017, restrained GOF =
1.017 for all data. An ORTEP diagram of 19with
50% ellipsoidal probability has been shown in Figure
2. Crystallographic data has been deposited with Cam-
bridge Crystallographic Data Centre (CCDC 260154).
using CH2Cl2–hexane (1:2) furnished lactone 18 (15
26
mg, 17%) as oil. ½aŠ ¼ þ58.8 (c 0.97, CHCl3). IR (neat):
D
m
max/cmÀ1 1771. 1H NMR (300 MHz, CDCl3 + CCl4): d
2.62 (1H, dd, J 16.8 and 8.4 Hz), 2.35–2.14 (4H, m), 2.19
(1H, dd, J 16.8 and 8.1 Hz), 1.94 (1H, d, J 15.3 Hz), 1.66
(6H, s), 1.39(3H, s).
13C NMR (75 MHz,
CDCl3 + CCl4): d 175.3 (C), 124.9(C), 124.2 (C), 85.5
(C), 41.9(CH 2), 39.8 (CH), 35.9 (CH2), 34.0 (CH2),
27.1 (CH3), 19.4 (CH3), 19.3 (CH3). Mass: m/z 180
(M+, 31), 165 (29), 162 (25), 137 (29), 123 (30), 121
(51), 120 (30). HRMS: m/z Calcd for C11H16O2Na
(M+Na): 203.1048. Found: 203.1058.
4.13. Reaction of the keto-epoxide 11 with BF3ÆEt2O
Reaction of keto-epoxide 11 (90 mg, 0.5 mmol) in dry
CH2Cl2 (5 mL, 0.1 M) with freshly distilled BF3ÆEt2O
(0.5 mL, 1 M in CH2Cl2, 0.5 mmol) and purification
on a silica gel column using ethyl acetate–hexane
(1:10) as eluent furnished, first diketone 14 (63 mg,
70%) as oil. Further elution of the column furnished
diketone 9 (21 mg, 24%) as oil.
4.11. (À)-(1R,3R,4R,6S)-3,4-Dibromo-3,4,6-trimethyl-
7-oxabicyclo[4.3.0]nonane 19
To a magnetically stirred ice cold solution of lactone 18
(20 mg, 0.11 mmol) in dry CCl4 (3 mL) was added bro-
mine (0.1 mL, excess). The reaction mixture was stirred
for 30 min at the same temperature. Excess bromine and
the solvent were removed under reduced pressure. Puri-
fication of the residue on a silica gel column using ethyl
acetate–hexane (1:20) as eluent furnished the dibromide
19 (15 mg, 40%) as a crystalline solid, which was recrys-
Acknowledgements
We thank the CCD facility of IISc. for providing the
X-ray diffraction data and the CSIR, New Delhi, for
the award of a research fellowship to S.S.V.R.
tallised from a mixture of CH2Cl2–hexanes. Mp: 119–
References
25
122 ꢁC. ½aŠ ¼ À15.6 (c 1.22, CHCl3). IR (neat): mmax
/
D
cmÀ1 1773. 1H NMR (300 MHz, CDCl3 + CCl4): d
2.92 (1H, dd, J 17.1 and 7.2 Hz), 2.80–2.65 (1H, m),
2.72 and 2.47 (2H, 2 · d, J 16.5 Hz), 2.36–2.25 (1H,
m), 2.19(1H, d, J 17.1 Hz), 2.11 (1H, dd, J 15.0 and
6.6 Hz), 2.03 (3H, s), 1.99 (3H, s), 1.39 (3H, s). 13C
NMR (75 MHz, CDCl3 + CCl4): d 159.0 (C), 82.4 (C),
73.3 (C), 65.7 (C), 47.6 (CH2), 43.1 (CH2), 38.6 (CH),
35.7 (CH2), 31.5 (CH3), 30.5 (CH3), 27.8 (CH3). Mass:
1. House, H. O.; Ryerson, G. D. J.Am.Chem.Soc. 1961, 83,
979; Bach, R. D.; Tubergen, M. W.; Klix, R. C. Tetra-
hedron Lett. 1986, 27, 3565–3568; Klix, R. C.; Bach, R. D.
J.Org.Chem.
therein.
1987, 52, 580–586, and references cited
2. Yamano, Y.; Tode, C.; Ito, M. J.Chem.Soc., Perkin Trans.
1 1998, 2569–2581; Sankararaman, S.; Nesakumar, J. E.
J.Chem.Soc,. Perkin Trans.1
references cited therein.
3. Carman, R. M.; Van Dongen, J. M. A. M. Aust.J.Chem.
1986, 39, 817–820.
4. Srikrishna, A.; Hemamalini, P. Indian J.Chem. 1990, 29B,
152–153.
5. Santos, R. B. D.; Brocksom, T. J.; Brocksom, U. Tetra-
hedron Lett. 1997, 38, 745–748.
6. Garver, L.; van Eikeren, P.; Byrd, J. E. J.Org.Chem. 1976,
41, 2773–2774.
7. Crystallographic data has been deposited with the Cam-
bridge Crystallographic Data Centre, CCDC 260155.
8. Bach, R. D.; Klix, R. C. Tetrahedron Lett. 1985, 26, 985–
988.
1999, 3173–3175, and
+
m/z 261 and 259(M ÀBr, 5%), 179(46), 178 (28), 133
(100), 119(17), 109(12), 107 (11), 105 (10). HRMS:
m/z Calcd for C11H16Br2O2K (M+K): 376.9154. Found:
376.9155.
4.12. Crystal data for 19
X-ray data were collected at 293 K on a SMART
CCD-BRUKER diffractometer with graphite-mono-
˚
chromated Mo Ka radiation (k = 0.71073 A). The struc-
ture was solved by direct methods (SIR92). Refinement