Stereoselective synthesis of (±)-β-elemene by a doubly diastereodifferentiating internal alkylation: A remarkable difference in the rate of enolization between syn and anti esters

Article abstract of DOI:10.1016/S0040-4020(00)01130-3

A total synthesis of the sesquiterpene (±)-β-elemene (1) has been achieved starting from a simple acyclic precursor 4. Key transformations include a 'doubly diastereodifferentiating folding and allylic strain-controlled' intramolecular ester enolate alkylation. In the course of the synthesis, we encountered remarkably different reactivity between syn and anti diastereomeric esters in the enolization step.

Full text of DOI:10.1016/S0040-4020(00)01130-3

TETRAHEDRON
Pergamon
Tetrahedron 57 (2001) 1247±1252
Stereoselective synthesis of (^)-b-elemene by a doubly
diastereodifferentiating internal alkylation: a remarkable
difference in the rate of enolization between syn and anti esters
Deukjoon Kim,a,p Jongkook Lee,a Jiyoung Changa and Sanghee Kimb
aResearch Institute of Pharmaceutical Sciences, College of Pharmacy, Seoul National University, San 56-1, Shillim-Dong, Kwanak-Gu,
Seoul 151-742, South Korea
bNatural Products Research Institute, Seoul National University, 28 Yungun, Jongro, Seoul 110-460, South Korea
Received 12 October 2000; accepted 1 December 2000
AbstractÐA total synthesis of the sesquiterpene (^)-b-elemene (1) has been achieved starting from a simple acyclic precursor 4. Key
transformations include a `doubly diastereodifferentiating folding and allylic strain-controlled' intramolecular ester enolate alkylation. In the
course of the synthesis, we encountered remarkably different reactivity between syn and anti diastereomeric esters in the enolization step.
q 2001 Elsevier Science Ltd. All rights reserved.
The sesquiterpene b-elemene (1), isolated from sweet¯ag,
juniper oils and Libanotis transcaucasica,1 is a constituent
of many essential oils.2 Recently, Yuan et al. reported that
elemene induces apoptosis and regulates expression of bcl-2
protein in human leukemia K562 cells.3 Elemenum
emulsion is in clinical trials4a due to its potent antitumor
activity.4 Reported herein is a stereoselective synthesis of
(^)-b-elemene (1)5 using our `doubly diastereodifferentiat-
ing folding and allylic strain-controlled'6 intramolecular
ester enolate alkylation (IEEA) as a key step. During the
course of our investigation, we also observed a remarkable
difference in the rate of enolization between syn and anti
diastereomeric esters during the intramolecular ester enolate
alkylation step (vide infra).
of diol 5, followed by reaction of the resulting ditosylate 6
with NaCN, yielded dinitrile 7 in 77% overall yield in two
steps. Methanolysis of dinitrile 7 and reduction of the corre-
sponding diester 8 with LAH produced diol 9 in 58% yield
for the two steps. Finally, conversion of diol 9 to the desired
dibromide 4 was accomplished by treatment with Ph3P and
CBr4 at 08C in 92% yield. Ozonolysis of dibromide 4 in
CH2Cl2 followed by in situ Wittig reaction with
(carbethoxyethylidene)triphenylphosphorane
furnished
unsaturated ester 10 (E:Zˆ10:1, 92% total yield). Treatment
of a,b-unsaturated ester 10 with DIBALH led to the forma-
tion of allylic alcohol 11 (99%), which was subjected to
Johnson orthoester Claisen rearrangement conditions8
(triethyl orthopropionate, propionic acid, 1258C) to yield a
readily separable 80:20 mixture of two diastereomeric g,d-
unsaturated esters 3a and 3b in 57% total yield. The relative
stereochemistries of Claisen products 3a and 3b were
assigned by comparison with an analogous system prepared
In our retrosynthetic analysis for (^)-b-elemene (1), as
summarized in Scheme 1, we envisioned that key cyclo-
hexanecarboxylate 2a could be stereoselectively synthe-
sized from internal alkylation substrate 3 based upon our
IEEA strategy by which the two homomorphic diastereo-
topic bromoethyl groups in 3 could be distinguished, thus
establishing the relative stereochemistry of three of the
stereogenic centers of (^)-b-elemene (1) in a single step.
Further analysis indicated dibromide 4 should be an ideal
synthetic precursor for acyclic substrate 3.
The starting dibromide 4 was prepared from the known diol
57 in a straightforward ®ve-step sequence. Thus, tosylation
Keywords: rate of enolization; doubly diastereodifferentiating internal
alkylation; (^)-b-elemene.
Corresponding author. Tel.: 182-2-880-7850; fax: 182-2-888-0649;
p
Scheme 1.
e-mail: deukjoon@plaza.snu.ac.kr
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)01130-3

1248
D. Kim et al. / Tetrahedron 57 (2001) 1247±1252
under the identical reaction conditions in the recent
systematic study by Daub on the acyclic diastereoselection
of the orthoester Claisen rearrangement.8b The moderate
chemical yield of 3a/3b could be improved to 87% by the
use of phenol instead of propionic acid as an acid catalyst
(phenol, triethyl orthopropionate, toluene, 1108C,
3a:3bˆ63:37).
was employed to yield the desired cyclohexanecarboxylate
2a but in an inferior yield and with poor stereoselectivity
(47%, 3:1 ratio at C7) compared to anti isomer 3a. The
chemical yield of syn ester 3b could be improved to 85%
by using LHMDS instead of LDA as the base in THF at
room temperature overnight.
The above experimental observations can be rationalized by
assuming that the stereoselection at C7 is dependent upon
the temperature at which the cyclization of the enolate
occurs. The enolization of anti-isomer 3a takes place at
2788C at a much faster rate (2 h) than the corresponding
syn-isomer 3b and the resulting enolate undergoes the
internal alkylation at this temperature to give a good yield
(89%) of the desired product 2a with a 6:1 isomer ratio at
C7. On the other hand, the enolization of syn-ester 3b with
excess LDA in THF at 2788C proceeds at a much slower
rate and this was con®rmed by quenching the reaction
mixture after 2 h with CD3OD and showing that a-proton
of the recovered syn-ester 3b was not exchanged with
deuterium. The enolate generated from syn-ester 3b with
excess LDA at room temperature undergoes the intra-
molecular ester enolate alkylation at a much higher
temperature and this higher alkylation temperature thereby
results in a lower stereoselectivity at C7 in the case of 3b
(i.e. 3:1 vs 6:1). To our knowledge, a difference in reactivity
between syn and anti diastereomeric esters in an enolization
step has not been reported, and is usually not considered to
be an important factor in ester enolate alkylation reactions.
Initially we anticipated the relative stereochemistry of esters
3 would be inconsequential in intramolecular ester enolate
alkylation since the stereochemistry would be lost during
the enolization step. However, to our surprise, the two cycli-
zation substrates 3a and 3b showed different reactivity as
depicted in Scheme 2. Intramolecular ester enolate alkyl-
ation of anti-dibromo ester 3a with excess LDA in THF at
2788C for 2 h provided the desired cyclohexanecarboxylate
2a with 6:1 selectivity at C7 in 89% yield. On the other hand,
syn-dibromo ester 3b remained mostly unchanged under the
same reaction conditions and a small amount (15%) of the
same 6:1 mixture of cyclohexanecarboxylates 2a and 2b
was formed. The cyclization of syn-dibromo ester 3b
could be facilitated when higher temperature (rt, 10 min)
Although more systematic studies are needed to elucidate
the origin of the distinct differences in the rate of enoliza-
tion, one possible reason may be due to imposed nonbonded
interactions in Ireland's chair-like transition state model,9
where the lithium cation is coordinated to the carbonyl
oxygen and the bases as depicted in Fig. 1. Assuming that
the b-hydrogen has a gauche conformation with the
b-methyl substituent to minimize possible 1,2-interactions,
the TS I of anti-isomer 3a suffers from fewer nonbonded
interactions than the TS II of syn-isomer 3b, which has the
bulky dibromo alkyl substituent near OEt group. Therefore,
the deprotonation of anti-ester 3a by LDA in THF proceeds
at a much faster rate than the syn-isomer 3b.
With the desired cyclohexanecarboxylate 2a in hand, we
turned our attention to its conversion to (^)-b-elemene
(1). The inseparable mixture (6:1 at C7) of primary bromides
2a and 2b was treated with potassium t-butoxide (THF, 08C)
to give a mixture of ole®ns 12a and 12b, which was then
subjected to DIBALH reduction to yield alcohols 13a and
13b as an inseparable 6:1 mixture in 86% overall yield.
Chemoselective Wacker10 reaction of the terminal alkenes
Scheme 2. (a) TsCl, pyridine, CHCl3, 08C, overnight, 84%. (b) NaCN,
DMF, re¯ux, 13 h, 92%. (c) PTSAzH2O, MeOH, re¯ux, 48 h, 58%. (d)
LAH, THF, rt, overnight, 100%. (e) CBr4, Ph3P, CH2Cl2, 08C, 1.5 h,
92%. (f) O3, CH2Cl2, 2788C, 1 h; Ph3P, rt, 0.5 h. (g) Ph3PˆC(Me)CO2Et,
CH2Cl2, rt, 14 h, 92% for two steps. (h) DIBALH, toluene, 2788C, 2 h,
99%. (i) C2H5C(OEt)3, PhOH, toluene, re¯ux, 13 h, 87%; or C2H5C(OEt)3,
CH3CH2CO2H, 1258C, 57%. (j) See table.
Figure 1.

D. Kim et al. / Tetrahedron 57 (2001) 1247±1252
1249
75 MHz) d 132.6, 119.7, 116.9, 36.8, 32.2, 21.0; IR (neat)
2249, 1643 cm21 HRMS calcd for C8H10N2 (M1)
134.0844, found 134.0844.
;
1.1.3. 3-Allylpentanedioic acid dimethyl ester (8). A
mixture of dinitrile 7 (900 mg, 6.71 mmol) and p-toluene-
sulfonic acid monohydrate (3.83 g, 20.1 mmol) in MeOH
(14.0 mL) was re¯uxed for 48 h. The mixture was diluted
with EtOAc, and washed with saturated aqueous NH4Cl,
water, and brine. The organic layer was dried over anhy-
drous Na2SO4, and concentrated at reduced pressure. The
resulting residue was puri®ed by column chromatography
on silica gel (hexane/EtOAc, 7:1) to afford dimethyl ester 8
Scheme 3. (a) t-BuOK, THF, 08C, 8 h, 89%. (b) DIBALH, toluene, 2788C,
2 h, 92%. (c) CuCl, PdCl2, O2, DMF/H2O, rt, 8 h, 93%. (d) i) PCC, NaOAc,
CH2Cl2, 08C, 2 h, 92%. ii) Separation. (e) Ph3PˆCH2, THF, 2788C to rt,
1 h, 89%.
of 13a and 13b, followed by PCC oxidation, led to the
formation of a readily separable mixture keto-aldehyde
15a and its C7 isomer 15b (7:1, 92% total yield). Finally,
double Wittig methylenation of keto-aldehyde 15a provided
the desired (^)-b-elemene (1) in 89% yield, whose 1H and
13C NMR spectral data were in good agreement with those
reported by Thomas et al.11 (Scheme 3).
1
(1.50 g) in 58% yield. H NMR (CDCl3, 300 MHz) d 5.74
(ddt, Jˆ16.3, 11.0, 7.2 Hz, 1H), 5.08±5.02 (m, 2H), 3.67 (s,
6H), 2.52±2.30 (m, 5H), 2.14 (dd, Jˆ7.1, 6.1 Hz, 2H); 13C
NMR (CDCl3, 75 MHz) d 172.7, 135.2, 117.5, 51.3, 38.1,
37.7, 31.6; IR (neat) 1739 cm21; HRMS calcd for C9H13O3
(M12OCH3) 169.0865, found 169.0865.
In conclusion, we have accomplished a stereoselective total
synthesis of (^)-b-elemene (1) utilizing a `doubly diaster-
eodifferentiating folding and allylic strain-controlled' IEEA
strategy as a means of establishing the relative stereo-
chemistry of the three stereogenic centers present in the
natural product. More importantly, we encountered a
remarkable difference in the rate of enolization between
syn and anti diastereomeric esters during the internal alkyl-
ation, which should be considered as an important factor in
applications of IEEA to natural product synthesis.12
1.1.4. 3-Allylpentane-1,5-diol (9). To a suspension of LAH
(330 mg, 8.70 mmol) in THF (5.0 mL) was added dropwise
a solution of dimethyl ester 8 (435 mg, 2.17 mmol) in THF
(6.0 mL) at 08C. The mixture was stirred overnight at rt and
then cooled to 08C. To the reaction mixture were added
water (0.3 mL), aqueous 3N NaOH solution (0.3 mL), and
water (0.9 mL) in order. The mixture was stirred for 2 h, and
®ltered through a short pad of Celite. The ®ltrate was
concentrated at reduced pressure, and the resulting residue
was puri®ed by column chromatography on silica gel
(hexane/EtOAc, 2:1) to give diol 9 (322 mg) in quantitative
1
yield. H NMR (CDCl3, 300 MHz) d 5.85±5.71 (m, 1H),
5.08±5.00 (m, 2H), 3.77±3.48 (m, 4H), 2.10 (ddt, Jˆ7.2,
6.2, 1.2 Hz, 2H), 2.09 (bs, 1H), 1.80 (bs, 1H), 1.78 (tt,
Jˆ12.6, 6.3 Hz, 1H), 1.66±1.48 (m, 4H); 13C NMR
(CDCl3, 75 MHz) d 136.5, 116.4, 60.1, 38.5, 36.0, 30.6;
IR (neat) 3336 cm21; HRMS calcd for C8H12 (M122H2O)
108.0939, found 108.0926.
1. Experimental
1.1. Data for compounds
1.1.1. Ditosylate (6). To a solution of diol 5 (5.80 g,
49.9 mmol) in CHCl3 (50.0 mL) were added pyridine
(12.1 mL) and TsCl (23.80 g, 124.8 mmol). The reaction
mixture was stirred at 08C overnight, diluted with EtOAc,
and washed with brine. The organic layer was dried over
anhydrous Na2SO4, and concentrated at reduced pressure.
The resulting residue was puri®ed by column chromato-
graphy on silica gel (hexane/EtOAc, 4:1) to afford ditosylate
1.1.5. 6-Bromo-4-(2-bromoethyl)hex-1-ene (4). To a solu-
tion of diol 9 (151 mg, 1.05 mmol) in CH2Cl2 (10.1 mL)
were added Ph3P (826 mg, 3.15 mmol) and CBr4 (696 mg,
2.10 mmol). After 1.5 h at 08C, the mixture was diluted with
hexane, and ®ltered through a short silica gel column. The
®ltrate was concentrated at reduced pressure and the result-
ing residue was puri®ed by column chromatography on
silica gel (hexane) to afford dibromide 4 (260 mg) in 92%
1
6 (17.80 g) in 84% yield. H NMR (CDCl3, 300 MHz) d
7.74 (d, Jˆ8.3 Hz, 4H), 7.35 (d, Jˆ8.0 Hz, 4H), 5.61±
5.50 (m,1H), 4.99 (d, Jˆ11.0 Hz, 1H), 4.94 (dd, Jˆ17.8,
1.5 Hz, 1H), 3.99 (dd, Jˆ9.9, 4.5 Hz, 2H), 3.91 (dd, Jˆ9.9,
5.7 Hz, 2H), 2.46 (s, 6H), 2.17±2.02 (m, 3H); 13C NMR
(CDCl3, 75 MHz) d 145.0, 133.5, 132.5, 129.9, 127.9,
1
yield. H NMR (CDCl3, 300 MHz) d 5.81±5.67 (m, 1H),
5.11±5.04 (m, 2H), 3.43 (t, Jˆ6.6 Hz, 4H), 2.12±2.09 (m,
2H), 1.93±1.79 (m, 5H); 13C NMR (CDCl3, 75 MHz) d
135.2, 117.4, 36.7, 36.3, 35.1, 31.0; IR (neat) 1257 cm21
HRMS calcd for MH21 269.9619, found 269.9459.
;
118.4, 68.4, 37.7, 31.4, 21.6; IR (neat) 1362, 1174 cm21
.
1.1.2. 3-Allylpentanedinitrile (7). To a solution of ditosyl-
ate 6 (767 mg, 1.81 mmol) in DMF (9.0 mL) was added
NaCN (350 mg, 7.14 mmol). The reaction mixture was
re¯uxed for 13 h, diluted with EtOAc, washed with brine,
and dried over anhydrous Na2SO4. The organic layer was
concentrated and the resulting residue was puri®ed by
column chromatography on silica gel (hexane/EtOAc, 5:1)
to afford dinitrile 7 (223 mg) in 92% yield. 1H NMR
(CDCl3, 300 MHz) d 5.70 (ddt, Jˆ17.3, 10.0, 7.1 Hz, 1H),
5.27±5.20 (m, 2H), 2.58 (dd, Jˆ17.1, 6.1 Hz, 2H), 2.51 (dd,
Jˆ17.0, 6.3 Hz, 2H), 2.35±2.16 (m, 3H); 13C NMR (CDCl3,
1.1.6. Ethyl (E)-7-bromo-5-(2-bromoethyl)-2-methyl-2-
heptenoate (10). To a cooled (2788C) solution of dibro-
mide 4 (168 mg, 0.62 mmol) in CH2Cl2 (2.0 mL) was added
dropwise a saturated solution of ozone in CH2Cl2 (2788C)
over 1 h. Excess ozone was removed by a stream of nitrogen
and triphenylphosphine (160 mg, 0.61 mmol) was added,
and the mixture was warmed to rt. After 30 min,
(carbethoxyethylidene)triphenylphosphorane (450 mg, 1.24
mmol) was added to the reaction mixture, which was stirred
overnight. The mixture was diluted with hexane and then

1250
D. Kim et al. / Tetrahedron 57 (2001) 1247±1252
®ltered through a short silica gel column. The ®ltrate was
concentrated and the residue was puri®ed by column chro-
matography on silica gel (hexane/EtOAc, 50:1) to give
E-unsaturated ester 10 (185 mg, 83%) and Z-isomer
144.4, 114.3, 60.2, 47.5, 43.2, 37.4, 36.0, 33.3, 32.5, 30.8,
30.4, 18.9, 14.4, 14.2; IR (neat) 1731, 1179 cm21; HRMS
calcd for C14H25Br2O2 (MH2212CH3) 383.0221, found
383.0040. syn-Dibromo ester 3b: 1H NMR (CDCl3,
500 MHz) d 4.93 (dq, Jˆ1.6 Hz, 1H), 4.87 (d, Jˆ1.4 Hz,
1H), 4.20±4.10 (m, 2H), 3.46±3.40 (m, 2H), 3.36±3.29 (m,
2H), 2.39 (td, Jˆ11.0, 3.1 Hz, 1H), 2.35±2.29 (m, 1H),
2.04±1.97 (m, 1H), 1.95±1.88 (m, 1H), 1.77±1.62 (m,
3H), 1.60 (s, 3H), 1.42 (ddd, Jˆ13.9, 11.2, 2.8 Hz, 1H),
1.27 (t, Jˆ7.2 Hz, 3H), 1.05 (d, Jˆ6.7 Hz, 3H), 1.09±1.04
(m, 1H); 13C NMR (CDCl3, 75 MHz) d 176.3, 143.0, 115.7,
60.3, 47.8, 43.0, 37.2, 35.8, 34.3, 33.3, 17.4, 16.4, 14.3; IR
(neat) 1730, 1175 cm21; HRMS calcd for C15H28Br2O2
(MH221) 398.0456, found 398.0283.
1
(19 mg, 9%). E-Ester 10: H NMR (CDCl3, 500 MHz) d
6.73 (tq, Jˆ7.4, 1.5 Hz, 1H), 4.21 (q, Jˆ7.1 Hz, 2H),
3.46±3.39 (m, 4H), 2.22 (t, Jˆ6.8 Hz, 2H), 2.03 (tt,
Jˆ12.9, 6.5 Hz, 1H), 1.92±1.82 (m, 7H), 1.31 (t,
Jˆ7.1 Hz, 3H); 13C NMR (CDCl3, 75 MHz) d 167.5,
138.2, 129.7, 60.4, 36.3, 35.2, 31.4, 30.7, 14.1, 12.5; IR
(neat) 1711, 1254 cm21; HRMS calcd for C12H22Br2O2
(MH221) 355.9987, found 355.9813. Z-Ester: 1H NMR
(CDCl3, 500 MHz) d 5.88 (tq, Jˆ7.5, 1.4 Hz, 1H), 4.20
(q, Jˆ7.1 Hz, 2H), 3.44 (t, Jˆ7.1 Hz, 4H), 2.54±2.51 (m,
2H), 1.93 (q, Jˆ1.4 Hz, 3H), 1.95±1.85 (m, 5H), 1.31 (t,
Jˆ7.1 Hz, 3H); 13C NMR (CDCl3, 75 MHz) d 167.8, 139.2,
129.4, 60.2, 36.6, 35.9, 32.2, 31.0, 20.8, 14.3; IR (neat)
1713, 1227 cm21
1.1.10. Ethyl (1R*,2S*,4R*)-4-(2-bromoethyl)-2-isopro-
penyl-1-methylcyclohexane-1-carboxylate (2a) and ethyl
(1R*,2S*,4S*)-4-(2-bromoethyl)-2-isopropenyl-1-methyl-
cyclohexane-1-carboxylate (2b). Cyclization of 3a with
LDA. To a cooled (2788C) solution of anti-dibromo ester
3a (134 mg, 0.34 mmol) in dry THF (41.0 mL) was added
dropwise a 0.5 M solution of LDA in THF (6.7 mL). The
mixture was stirred for 2 h at 2788C, quenched with satu-
rated aqueous NH4Cl, diluted with EtOAc, and washed with
brine. The organic layer was dried over anhydrous Na2SO4,
concentrated in vacuo, and the residue was puri®ed by
column chromatography on silica gel (hexane/Et2O,
100:1) to give an inseparable mixture of cyclohexane-
carboxylates 2a and 2b (95 mg, 89%, 2a:2bˆ6:1 by
1.1.7. (E)-7-Bromo-5-(2-bromoethyl)-2-methyl-2-hepten-
1-ol (11). To a cooled (2788C) solution of E-unsaturated
ester 10 (1.00 g, 2.81 mmol) in dry toluene (20.0 mL) was
added dropwise DIBALH (8.5 mL, 1.0 M solution in
toluene). After 2 h at 2788C, the reaction mixture was
quenched with methanol (10.0 mL) and stirred at rt for
2 h. The precipitate was ®ltered off through a pad of Celite.
The ®ltrate was concentrated in vacuo and the resulting
residue was puri®ed by column chromatography on silica
gel (hexane/EtOAc, 3:1) to give allylic alcohol 11 (875 mg,
99%). 1H NMR (CDCl3, 500 MHz) d 5.41 (tq, Jˆ7.2,
1.4 Hz, 1H), 4.03 (s, 2H), 3.43 (td, Jˆ7.0, 1.5 Hz, 4H),
2.08 (t, Jˆ5.9 Hz, 2H), 1.90±1.84 (m, 5H), 1.68 (s, 3H);
13C NMR (CDCl3, 100 MHz) d 137.0, 122.2, 68.6, 36.5,
35.8, 31.2, 30.4, 13.9; IR (neat) 3350, 1011 cm21; HRMS
calcd for C10H20Br2O (MH221) 313.9881, found 313.9720.
1
500 MHz H NMR).
1.1.11. Cyclization of 3b with LDA at 2788C for 2 h and
quenching with CD3OD. To a stirred solution of syn
dibromo ester 3b (12.2 mg, 0.031 mmol) in dry THF
(3.8 mL) was added dropwise a 0.5 M solution of LDA in
THF (0.6 mL). After 2 h at 2788C, pre-cooled CD3OD
(0.5 mL) was slowly added to the mixture at 2788C. The
mixture was stirred for 30 min at this temperature, diluted
with EtOAc and washed with brine. The organic layer was
dried over anhydrous Na2SO4 and concentrated in vacuo.
The resulting residue was puri®ed by column chromato-
graphy on silica gel (hexane/Et2O, 100:1) to give a mixture
of cyclohexanecarboxylates 2a and 2b (1.5 mg, 15%,
2a:2bˆ6:1 by 500 MHz 1H NMR), and syn-ester 3b
(9.8 mg, 80%).
1.1.8. Ethyl (2R*,3R*)-3-[4-bromo-2-(2-bromoethyl)-
butyl]-2,4-dimethyl-4-pentenoate (3a) and ethyl (2S*,
3R*)-3-[4-bromo-2-(2-bromoethyl)butyl]-2,4-dimethyl-
4-pentenoate (3b). Claisen rearrangement with pro-
pionic acid. A mixture of dibromo allylic alcohol 11
(29 mg, 0.092 mmol), triethyl orthopropionate (2.0 mL)
and propionic acid (1 mg) was stirred for 24 h at 1258C.
Concentration in vacuo and puri®cation of the residue on
silica gel (hexane/EtOAc, 100:1) gave an easily separable
80:20 mixture of dibromo esters 3a and 3b (21 mg, 57%).
1.1.12. Cyclization of 3b with LDA at rt. To a stirred
solution of syn dibromo ester 3b (16.7 mg, 0.042 mmol)
in dry THF (5.0 mL) was added dropwise a 0.5 M solution
of LDA in THF (0.9 mL). After 10 min at rt, the mixture
was quenched with saturated aqueous NH4Cl, diluted with
EtOAc, and washed with brine. The organic layer was dried
over anhydrous Na2SO4 and concentrated in vacuo. The
resulting residue was puri®ed by column chromatography
on silica gel (hexane/Et2O, 100:1) to give a mixture of
cyclohexanecarboxylates 2a and 2b (6.3 mg, 47%,
2a:2bˆ3:1 by 300 MHz 1H NMR).
1.1.9. Claisen rearrangement with phenol. To a solution
of dibromo allylic alcohol 11 (741 mg, 2.36 mmol) in dry
toluene (6.0 mL) was added triethyl orthopropionate
(1.5 mL) and phenol (7 mg). The reaction mixture was
re¯uxed for 13 h. The solvent was removed at reduced
pressure and the residue was puri®ed by column chromato-
graphy on silica gel (hexane/EtOAc, 100:1) to give dibromo
esters 3a (517 mg, 55%) and 3b (304 mg, 32%). anti-
1
Dibromo ester 3a: H NMR (CDCl3, 500 MHz) d 4.86
(dq, Jˆ1.5 Hz, 1H), 4.77±4.76 (m, 1H), 4.14±4.03 (m,
2H), 3.46±3.30 (m, 4H), 2.50±2.44 (m, 1H), 2.40 (ddd,
Jˆ11.6, 8.6, 3.3 Hz, 1H), 2.04±1.91 (m, 2H), 1.78±1.70
(m, 2H), 1.69 (dd, Jˆ1.4, 0.7 Hz, 3H), 1.68±1.63 (m, 1H),
1.39 (ddd, Jˆ14.0, 11.2, 3.0 Hz, 1H), 1.29 (ddd, Jˆ13.8,
10.2, 3.6 Hz, 1H), 1.23 (t, Jˆ7.2 Hz, 3H), 1.13 (d,
Jˆ6.8 Hz, 3H); 13C NMR (CDCl3, 75 MHz) d 175.5,
1.1.13. Cyclization of 3b with LHMDS. To a stirred solu-
tion of syn dibromo ester 3b (65 mg, 0.16 mmol) in dry THF
(15.0 mL) was added dropwise a 1.0 M solution of LHMDS
in THF (1.6 mL). The mixture was stirred overnight at rt,
quenched with saturated aqueous NH4Cl, diluted with

D. Kim et al. / Tetrahedron 57 (2001) 1247±1252
1251
EtOAc and washed with brine. The organic layer was dried
over anhydrous Na2SO4 and concentrated in vacuo. The
residue was puri®ed by column chromatography on silica
gel (hexane/Et2O, 100:1) to give a mixture of cyclohexane-
carboxylates 2a and 2b (44 mg, 85%, 2a:2bˆ3:1 by
500 MHz 1H NMR). 2a and 2b: 1H NMR (CDCl3,
precipitate was ®ltered off through a pad of Celite. The
®ltrate was concentrated in vacuo and the resulting residue
was puri®ed by column chromatography on silica gel
(hexane/EtOAc, 10:1) to give an inseparable 6:1 mixture
1
of alcohols 13a and 13b (108 mg, 92%). 13a and 13b: H
NMR (CDCl3, 500 MHz) d [5.93 (ddd, Jˆ16.8, 11.3,
5.6 Hz) and 5.80 (ddd, Jˆ17.2, 10.6, 6.5 Hz), 1H], [5.07
(d, Jˆ1.9 Hz) and 4.99 (dt, Jˆ17.3, 1.6 Hz), 1H], [5.04
(dt, Jˆ7.6, 1.8 Hz) and 4.92 (dt, Jˆ10.2, 1.5 Hz), 1H],
4.85±4.83 (m, 1H), 4.76±4.75 (m, 1H), 3.40±3.34 (m,
2H), [2.55±2.49 (m) and 2.05±1.96 (m), 1H], [2.34 (dd,
Jˆ12.6, 3.5 Hz) and 2.24±2.20 (m), 1H], [1.91 (ddd,
Jˆ13.6, 12.6, 5.0 Hz), 1.82±1.76 (m) and 1.66±1.10 (m),
6H], 1.75 (d, Jˆ1.4 Hz, 3H), [0.94 (s) and 0.90 (s), 3H]; 13C
NMR (CDCl3, 75 MHz) d 148.9, 148.7, 143.9, 142.0, 113.9,
112.80, 112.75, 112.1, 72.47, 72.43, 49.4, 44.1, 41.9, 38.9,
38.5, 36.1, 33.6, 31.6, 31.3, 27.3, 25.2, 22.7, 22.5, 16.7,
16.4; IR (neat) 3388, 1042 cm21; HRMS calcd for
C13H22O (M1) 194.1671, found 194.1674.
500 MHz)
d
[4.83 (dq, Jˆ1.6 Hz) and 4.80 (dq,
Jˆ1.6 Hz), 1H], [4.72±4.71 (m) and 4.65±4.64 (m), 1H],
4.17±4.04 (m, 2H), [3.45 (t, Jˆ7.1 Hz) and 3.43 (t,
Jˆ7.0 Hz), 2H], [2.80 (dd, Jˆ10.4, 4.3 Hz) and 2.67 (dd,
Jˆ13.0, 3.4 Hz), 1H], [2.05±1.99 (m) and 1.97±1.91 (m),
2H], 1.83 (q, Jˆ7.0 Hz, 2H), [1.78±1.73 (m) and 1.67±1.56
(m), 4H], [1.70 (dd, Jˆ1.2, 0.6 Hz) and 1.64 (dd, Jˆ1.2,
0.7 Hz), 3H], 1.26 (t, Jˆ7.1 Hz, 3H), [1.13 (s) and 1.12 (s),
3H]; 13C NMR (CDCl3, 75 MHz) d 177.9, 177.8, 146.6,
146.5, 113.1, 112.8, 60.3, 60.2, 48.3, 46.4, 43.1, 39.9,
37.1, 36.0, 35.8, 32.7, 32.3, 31.8, 31.3, 30.9, 30.8, 27.0,
25.3, 23.8, 22.8, 17.7, 15.2, 14.12, 14.08; IR (neat) 1725,
1235 cm21; HRMS calcd for C15H28Br2O2 (M1) 316.1038,
found 316.1033.
1.1.16. 1-[(1R*,3S*,4R*)-4-(Hydroxymethyl)-3-isopro-
penyl-4-methylcyclohexyl]-1-ethanone (14a) and 1-[(1S*,
3S*,4R*)-4-(hydroxymethyl)-3-isopropenyl-4-methyl-
cyclohexyl]-1-ethanone (14b). To a stirred solution of
alcohols 13a and 13b (68 mg, 0.35 mmol, 13a:13bˆ6:1)
in DMF (7.0 mL) and water (1.0 mL) were added PdCl2
(62 mg, 0.35 mmol) and CuCl (173 mg, 1.75 mmol) under
an oxygen atmosphere. The reaction mixture was stirred for
8 h, diluted with a 2:1 mixture of hexane and EtOAc, and
washed with brine. The organic layer was dried over
anhydrous Na2SO4 and concentrated in vacuo. The residue
was puri®ed by column chromatography on silica gel
(hexane/EtOAc, 5:1) to give an inseparable 8:1 mixture of
ketones 14a and 14b (68.4mg, 93%). 14a and 14b: 1H NMR
(CDCl3, 500 MHz) d 4.88±4.87 (m, 1H), 4.77 (dd, Jˆ1.5,
0.6 Hz, 1H), 3.40 (d, Jˆ10.9 Hz, 1H), 3.33 (d, Jˆ11.0 Hz,
1H), [2.67±2.65 (m) and 2.39 (tt, Jˆ11.9, 3.9 Hz), 1H],
[2.23 (dd, Jˆ12.6, 3.8 Hz) and 2.19 (dd, Jˆ9.1, 4.7 Hz),
1H], 2.16 (s, 3H), 1.89±1.40 (m, 6H), 1.76 (s, 3H), [0.93
(s) and 0.89 (s), 3H]; 13C NMR (CDCl3, 75 MHz) d 211.7,
211.2, 147.9, 147.5, 113.2, 113.0, 71.9, 71.6, 51.5, 48.3,
47.1, 45.2, 38.5, 38.4, 35.5, 32.4, 29.2, 28.1, 27.78, 27.74,
1.1.14. Ethyl (1R*,2S*,4R*)-2-isopropenyl-1-methyl-4-
vinylcyclohexane-1-carboxylate (12a) and ethyl (1R*,
2S*,4S*)-2-isopropenyl-1-methyl-4-vinylcyclohexane-1-
carboxylate (12b). To a stirred solution of cyclohexane-
carboxylates 2a and 2b (23.7 mg, 0.075 mmol,
2a:2bˆ6:1) in dry THF (2.0 mL) was slowly added a
1.0 M solution of t-BuOK (1.5 mL) in THF. After 8 h at
08C, methyl iodide was added to the mixture, which was
then stirred overnight. The reaction mixture was diluted
with EtOAc, washed with brine and dried over anhydrous
Na2SO4. Concentration of the solution in vacuo and puri®-
cation of the residue on silica gel (hexane/EtOAc, 100:1)
gave an inseparable 6:1 mixture of diole®ns 12a and 12b
(15.6 mg, 89%). 12a and 12b: 1H NMR (CDCl3, 500 MHz)
d [5.96 (ddd, Jˆ17.9, 10.0, 6.1 Hz) and 5.80 (ddd, Jˆ17.2,
10.6, 6.5 Hz), 1H], [5.07 (dt, Jˆ6.2, 1.6 Hz), and 4.94 (dt,
Jˆ10.7, 1.5 Hz), 1H], [5.04 (d, Jˆ1.6 Hz) and 5.01 (dt,
Jˆ17.3, 1.6 Hz), 1H], [4.82 (dq, Jˆ1.6 Hz) and 4.80 (dq,
Jˆ1.6 Hz), 1H], [4.72±4.70 (m) and 4.66 (dt, Jˆ1.9,
0.9 Hz), 1H], 4.05±4.17 (m, 2H), [2.86 (dd, Jˆ10.8,
4.0 Hz) and 2.69 (dd, Jˆ13.0, 3.4 Hz), 1H], [2.54±2.48
(m) and 2.13±2.09 (m), 1H], [1.96 (td, Jˆ13.8, 4.3 Hz)
and 1.81 (ddd, Jˆ13.9, 10.8, 4.8 Hz), 1H], [1.69 (s) and
1.65 (t, Jˆ1.0 Hz), 3H], [1.75±1.70 (m) and 1.67±1.59
(m), 4H], [1.39 (q, Jˆ12.8 Hz) and 1.31±1.28 (m), 1H],
[1.26 (t, Jˆ7.2 Hz) and 1.25 (t, Jˆ7.2 Hz), 3H], [1.16 (s)
and 1.13 (s), 3H]; 13C NMR (CDCl3, 75 MHz) d 178.1,
177.9, 146.8, 146.6, 143.6, 142.0, 113.8, 112.8, 112.3,
60.2, 48.3, 46.4, 46.2, 43.2, 41.5, 37.1, 36.0, 32.6, 32.1,
31.4, 26.8, 25.8, 23.7, 22.8, 15.2, 14.1; IR (neat) 1729,
1227 cm21; HRMS calcd for C15H24O2 (M1) 236.1776,
found 236.1779.
23.5, 22.9, 22.7, 21.7, 16.3; IR (neat) 3444, 1698 cm21
;
HRMS calcd for C13H22O2 (M1) 210.1620, found 210.1621.
1.1.17. (1R*,2S*,4R*)-4-Acetyl-2-isopropenyl-1-methyl-
cyclohexane-1-carbaldehyde (15a) and (1R*,2S*,4S*)-4-
acetyl-2-isopropenyl-1-methylcyclohexane-1-carbalde-
hyde (15b). A mixture of ketones 14a and 14b (93 mg,
Ê
0.44 mmol, 14a:14bˆ8:1), 4 A molecular sieves (470
mg), sodium acetate (182 mg, 2.22 mmol) and PCC
(477 mg, 2.21 mmol) in dry CH2Cl2 (19.0 mL) was stirred
for 2 h at 08C and then diluted with ether. The mixture was
®ltered through a pad of Florisil and washed repeatedly with
ether. Concentration of the solution in vacuo and puri®ca-
tion of the residue on silica gel (hexane/EtOAc, 10:1) gave
the desired aldehyde 15a (73.8 mg, 80%) and isomer 15b
(10.8 mg, 12%) in a total 92% yield. 15a: 1H NMR (CDCl3,
500 MHz) d 9.50 (s, 1H), 4.88 (dq, Jˆ1.5 Hz, 1H), 4.69 (dt,
Jˆ1.3, 0.7 Hz, 1H), 2.48±2.42 (m, 2H), 2.18 (s, 3H), 1.91±
1.86 (m, 1H), 1.82 (dtd, Jˆ13.5, 3.6, 1.8 Hz, 1H), 1.66 (s,
3H), 1.71±1.63 (m, 2H), 1.56±1.47 (m, 1H), 1.42 (dt,
Jˆ12.8, 3.3 Hz, 1H), 1.07 (d, Jˆ0.4 Hz, 3H); 13C NMR
1.1.15. [(1R*,2S*,4R*)-2-Isopropenyl-1-methyl-4-vinyl-
cyclohexyl]methanol (13a) and [(1R*,2S*,4S*)-2-isopro-
penyl-1-methyl-4-vinylcyclohexyl]methanol (13b). To a
cooled (2788C) solution of ole®ns 12a and 12b (143 mg,
0.61 mmol, 12a:12bˆ6:1) in dry toluene (8.5 mL) was
added dropwise a 1.0 M solution of DIBALH in toluene
(3.6 mL). The reaction mixture was stirred for 2 h at
2788C before the addition of methanol (7.0 mL). The
mixture was allowed to warm to rt and stirred for 2 h. The

1252
D. Kim et al. / Tetrahedron 57 (2001) 1247±1252
(b) Hunter, G. L. K. J. Food Sci. 1964, 29, 25. (c) Sorm, F.;
Herout, V.; Sykora, V. Chem. Listy 1955, 49, 942.
2. (a) Sorm, F.; Herout, V.; Romanuk, M. Chem. Listy 1958, 52,
1969. (b) Wichramasinghe, M. Phytochemistry 1980, 19, 255.
3. Yuan, J.; Gu, Z. L.; Chou, W. H.; Kwok, C. Y. Chung Kuo Yao
Li Hsueh Pao 1999, 20, 103.
(CDCl3, 100 MHz) d 210.7, 206.0, 145.0, 113.8, 50.8, 49.5,
46.6, 32.3, 28.2, 27.7, 23.2, 22.3, 12.9; IR (neat) 1711 cm21
;
HRMS calcd for C13H20O2 (M1) 208.1463, found 208.1463.
15b: 1H NMR (CDCl3, 500 MHz) d 9.39 (s, 1H), 4.90 (dq,
Jˆ1.5 Hz, 1H), 4.71 (dt, Jˆ1.4, 0.7 Hz, 1H), 2.71 (dq,
Jˆ4.8 Hz, 1H), 2.59 (dd, Jˆ10.6, 4.1 Hz, 1H), 2.17 (s,
3H), 1.96±1.92 (m, 2H), 1.87±1.73 (m, 3H), 1.69 (s, 3H),
1.30±1.26 (m, 1H), 1.03 (s, 3H); 13C NMR (CDCl3,
100 MHz) d 210.9, 205.7, 145.5, 113.6, 49.5, 46.2, 42.4,
4. (a) Wang, J.; Zhang, H.; Sun, Y. Chung Hua Chung Liu Tsa
Chih 1996, 18, 464. (b) Choi, B. G.; Kwak, E. Y.; Chung,
B. H.; Cho, W. J.; Cheon, S. H. Arch. Pharm. Res. 1999, 22,
575. (c) Yang, H.; Wang, X.; Yu, L.; Zheng, S. Zhonghua
Zhongliu Zazhi 1996, 18, 169; Chem. Abstr. 1996, 125,
212058z. (d) Zhong, Y.; Tang, J. Hunan Yixue 1996, 13,
119; Chem. Abstr. 1996, 125, 315937q. (e) Huang, L.; Hu,
Z.; Li, Z. Hunan Yixue 1996, 13, 241; Chem. Abstr. 1996,
125, 316472w.
29.0, 28.0, 26.8, 23.9, 21.4, 14.1; IR (neat) 1724 cm21
;
HRMS calcd for C13H20O2 (M1) 208.1463, found 208.1464.
1.1.18. (^)-b-Elemene (1). To a suspension of methyltri-
phenylphosphonium iodide (1.45 g, 3.59 mmol) in dry THF
(5.0 mL) was slowly added a 1.6 M solution of n-BuLi in
hexane (1.5 mL). The suspension was stirred for 10 min at
2788C and for 30 min at rt. A solution of aldehyde 15a
(50 mg, 0.24 mmol) in THF (1.7 mL) was added to the
mixture at 2788C. The reaction mixture was slowly warmed
to rt and stirred for 1 h. After the mixture was quenched with
saturated aqueous NH4Cl solution, the mixture was ®ltered
through a short silica gel column. Concentration of the solu-
tion in vacuo and puri®cation of the residue on silica gel
(hexane) gave (^)-b-elemene (1) (43.7 mg, 89%). 1H NMR
(CDCl3, 500 MHz) d 5.82 (dd, Jˆ17.5, 10.9 Hz, 1H), 4.91
(dd, Jˆ8.7, 1.3 Hz, 1H), 4.88 (dd, Jˆ2.1, 1.5 Hz, 1H), 4.82
(dq, Jˆ1.6 Hz, 1H), 4.72±4.71 (m, 1H), 4.70 (dq, Jˆ1.5 Hz,
1H), 4.60±4.59 (m, 1H), 2.03±2.00 (m, 1H), 1.97±1.92 (m,
1H), 1.75 (s, 3H), 1.71 (dd, Jˆ1.4, 0.8 Hz, 3H), 1.63±1.59
(m, 1H), 1.58±1.40 (m, 5H), 1.00 (s, 3H); 13C NMR (CDCl3,
75 MHz) d 150.4, 150.3, 147.7, 112.1, 109.8, 108.2, 52.7,
45.7, 39.9, 39.8, 32.9, 26.8, 24.8, 21.1, 16.6; IR (neat) 1644,
887 cm21; HRMS calcd for C15H24 (M1) 204.1878, found
204.1878.
5. For previous syntheses of b-elemene, see; (a) Patil, L. J.; Rao,
A. S. Tetrahedron Lett. 1967, 2273. (b) McMurry, J. E.;
Kocovsky, P. Tetrahedron Lett. 1985, 26, 2171. (c) Corey,
E. J.; Roberts, B. E.; Dixon, B. R. J. Am. Chem. Soc. 1995,
117, 193.
6. (a) For an early example of a doubly diastereodifferentiating
process, see: Stork, G.; Leonia, N. J. U.S. patent 2,793,233,
May 21, 1957; Chem. Abstr. 1958, 52, 441. (b) Kim, D.; Kim,
I. H. Tetrahedron Lett. 1997, 38, 415. (c) For a review on
differentiation of diastereotopic groups, see: Maier, M.
Group Selective Reactions; Waldmann, H., Ed.; Organic
Synthesis Highlights II. VCH Verlagsgesellschaft: Weinheim,
1995; pp 203±222. (d) For examples of folding and allylic
strain-controlled intramolecular ester enolate alkylation, see:
(1) Kim, D.; Ahn, S. K.; Choi, W. J.; Kim, H. S. Tetrahedron
Lett. 1997, 38, 4437 and references cited therein; (2)
Tokoroyama, T.; Kusaka, H. Can. J. Chem. 1996, 74, 2487.
7. Wasson, B. K.; Gleason, C. H.; Levi, I.; Parker, J. M.;
Thompson, L. M.; Yates, C. H. Can. J. Chem. 1961, 39, 923.
8. (a) Johnson, W. S.; Werthemann, L.; Bartlett, W. R.;
Brocksom, T. J.; Li, T.; Faulkner, J.; Peterson, M. R. J. Am.
Chem. Soc. 1970, 92, 741. (b) Daub, G. W.; Edwards, J. P.;
Okada, C. R.; Allen, J. W.; Maxey, C. T.; Wells, M. S.;
Goldstein, A. S.; Dibley, M. J.; Wang, C. J.; Ostercamp,
D. P.; Chung, S.; Cunningham, P. S.; Berliner, M. A. J. Org.
Chem. 1997, 62, 1976.
1.2. Supporting information
Copies of the 1H and 13C NMR spectra of 10, 3a, 3b, 2a and
2b, 15a, 15b and 1 as well as the IR and HRMS spectra for
1.
9. Ireland, R. E.; Wipf, P.; Armstrong, J. D. J. Org. Chem. 1991,
56, 650.
10. Tsuji, J. Synthesis 1984, 369.
Acknowledgements
11. Thomas, A. F.; Brauchli, R. J. Agric. Food Chem. 1991, 39,
431.
This work was supported by 2000 BK21 Project for
Medicine, Dentistry and Pharmacy and the Basic Research
Program of the Korea Science and Engineering Foundation
(1999-2-215-001-3).
12. For instance, use of the Ireland ester enolate Claisen
rearrangement for the synthesis of syn ester 3a would lead
to better overall stereoselectivity at C7.
References
1. (a) Pigulevskii, G. V. J. Org. Chem. USSR 1962, 32, 3054.