Communications
(
5 mol%) efficiently promoted the desired cyclization–frag-
[
16]
mentation reaction to afford the desired aldehyde 3 in high
yield with excellent chemoselectivity and face selectivity (3/4
9
[17]
[8]
8:2). The use of AgNO (10 mol%) and KNO (1 equiv)
3
3
also led to a selective reaction (3/4 99:1); however, the
reaction was slower (87% conversion after 8 h) and the yield
lower (64%).
The ee value of 3 could be increased to 97% through one
low-temperature crystallization (pentane). Finally, site-selec-
tive hydrogenation with Pd/CaCO , followed by a Wittig
3
reaction under the modified conditions described by Corey
and Yamamoto with hydroxyalkylation of the intermediate
[
18]
betaine
(
afforded the target (ꢀ)-1 (Z/E 97:3; 43%
[
19]
2 steps)), the sandalwood odor of which was excellent.
In conclusion, we have developed an efficient enantiose-
lective synthesis of (ꢀ)-b-santalol. The key step in the
synthesis consists of a highly selective copper-catalyzed
cyclization–fragmentation of an enynol. The scope and
generality of this novel methodology will be reported in due
course.
Experimental Section
6
and 7: A mixture of catalyst 5 (314 mg, 0.45 mmol), water (3.78 mL,
210 mmol), toluene (0.24 mL; internal standard), and crotonaldehyde
(95% trans; distilled and stored in the freezer; 2.10 g (2.46 mL),
30.0 mmol) was stirred at 208C for 5 min and then treated with freshly
prepared cyclopentadiene monomer (stored in a dewar containing dry
ice; 1.98 g (2.44 mL), 30.0 mmol). The two-phase system was stirred at
238C for 6 h, then cooled to 108C, treated with saturated aqueous
NaHCO solution (2.5 mL) and solid NaCl (100 mg), and stirred for
3
15 min. The two-phase system was separated, and the aqueous phase
was washed with pentane (3 mL). Concentration (08C, 50 mbar) and
bulb-to-bulb distillation (oven temperature 1008C, 4 mbar) afforded
a mixture of 6 and 7 (70:30; 2.07 g, 61%) with 88% purity. This
material, which contained traces of toluene, dicyclopentadiene, and
two unidentified products that may have originated from cis-
crotonaldehyde, was used for the ensuing hydrogenation.
Scheme 3. Reagents and conditions: a) Raney Ni (2%), 20 bar, 808C,
2
4 h (100%); b) ClCO Me (2 equiv), pyridine, toluene, 08C!RT, 1 h
2
(
90%); c) pyrolysis (4158C; quartz tube, 3 m; N stream; 90%);
2
d) TMSCCCHO (1.0 equiv with respect to 8), Me AlCl (1.0 equiv),
2
CH Cl , ꢀ208C (72%); e) K CO (10 equiv), MeOH (89%); f) H , 5%
(ꢀ)-3: [Cu(CH CN) ]BF (0.294 g, 0.93 mmol) was added to a
2
2
2
3
2
3
4
4
Pd/CaCO
3
(5%), MeOH/H
2
O (96:4), 4 h (85%); g) EtPPh
3
I
solution of 2 (3.24 g, 18.4 mmol) in 1,2-dichloroethane (100 mL) at
room temperature under nitrogen, and the resulting mixture was
stirred at 508C for 140 min. The dark-gray mixture was then cooled to
room temperature, filtered through a short pad of silica gel, and
concentrated under vacuum to give an orange oil (3.87 g). Bulb-to-
bulb distillation of this oil (oven temperature 1008C, 0.08 mbar)
afforded (ꢀ)-3 (3.12 g, 94%, 92% ee) with 98% purity (the impurity
corresponding to the remaining 2% was 4). Crystallization once from
(
1.1 equiv), BuLi (1.1 equiv), THF, ꢀ788C; then BuLi (1.2 equiv),
788C!08C; then paraformaldehyde (6 equiv), 08C!RT, 1 h (50%).
ꢀ
[
1
A] toluene, 708C, 1 h; [B] CH Cl , ꢀ108C, 2 h; [C] ClCH CH Cl, 508C,
2
2
2
2
40 min; [D] THF, H O, 708C, 8 h (87% conversion). TMS=trimethyl-
2
silyl.
9, which has a less accessible a hydrogen atom, underwent a
2
D
0
pentane at ꢀ788C afforded (ꢀ)-3 (2.79 g) with 97% ee. ½aꢁ
=
“
homoene reaction” exclusively to afford 10 after desilyla-
3
ꢀ1
ꢀ1
ꢀ2
ꢀ3
ꢀ
267.4 degcm g dm (CHCl , c = 1.06 ꢀ 10 gcm ).
[
15]
3
tion. The formation of 10 could be prevented completely by
1
H NMR (400 MHz, CDCl ): d = 9.52 (d, J = 7.8 Hz, 1H), 6.77 (d,
3
adapting the amount of trimethylsilylpropynal used to that of
J = 15.7 Hz, 1H), 6.09 (dd, J = 15.7, 7.8 Hz, 1H), 5.01 (s, 1H), 4.58 (s,
1H), 2.76 (br s, 1H), 2.18 (br s, 1H), 1.69–1.80 (m, 2H), 1.52–1.58 (m,
2H), 1.30–1.37 (m, 1H), 1.24–1.27 (m, 1H), 1.23 ppm (s, 3H).
8.
We tested Pt, Au, and Cu catalysis for the key cyclo-
isomerization reaction of enynol 2, which existed as one major
1
3
C NMR (100 MHz, CDCl ): d = 194.5 (d), 165.4 (d), 160.4 (s),
3
1
2
30.1 (d), 104.1 (t), 49.7 (s), 46.5 (d), 46.3 (d), 37.1 (t), 29.8 (t), 23.0 (t),
2.7 ppm (q).
and one minor diastereomer. PtCl catalysis led exclusively to
2
the undesired cyclopropane 4 in 88% yield. In the presence of
+
MS: m/z (%): 176 [M ] (23), 161 (71), 147 (63), 133 (75), 119 (62),
catalytic amounts of [PPh AuCl] and AgSbF , the cyclo-
3
6
105 (100), 91 (99), 79 (58), 77 (58), 39 (34).
isomerization leading to 4 was again favored (4/3 91:9).
Evidently, the AgCl formed in situ does not influence the
reaction course. In sharp contrast, [Cu(CH CN) ]BF
4
Received: June 25, 2009
Published online: August 24, 2009
3
4
7222
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 7221 –7223