Enantioselective biomimetic cyclization of isoprenoids using Lewis acid-assisted chiral Bronsted acids: Abnormal claisen rearrangements and successive cyclizations

Article abstract of DOI:10.1021/ja001165a

Syntheses of polycyclic isoprenoids have been achieved by several groups; however, no general "biomimetic" method has yet been reported. In this paper we describe the biomimetic cyclization of simple isoprenoids to polycyclic isoprenoids using Lewis acid-assisted chiral Bronsted acids, "chiral LBAs". This is the first example of a proton-induced enantioselective ene cyclization in synthetic chemistry. Geranyl phenyl ethers, o-geranylphenols, and geranylacetone derivatives were directly cyclized at -78 °C in the presence of (R)-binaphthol derivatives and tin tetrachloride. During the cyclization, [1,3] abnormal Claisen rearrangement often took place. The enantioselectivities were up to 90% ee. Compounds bearing a farnesyl group could also be cyclized under the same conditions to give the natural products (-)-Ambrox and (-)-chromazonarol. These chiral LBAs recognize a trisubstituted terminal olefin enantiotopically and generate site-selective carbocations on the substrates. The absolute stereochemistry of the cyclization is discussed with model studies using DFT calculations on the B3LYP/LANL2DZ level.

Full text of DOI:10.1021/ja001165a

J. Am. Chem. Soc. 2000, 122, 8131-8140
8131
Enantioselective Biomimetic Cyclization of Isoprenoids Using Lewis
Acid-Assisted Chiral Brønsted Acids: Abnormal Claisen
Rearrangements and Successive Cyclizations
Shingo Nakamura,† Kazuaki Ishihara,‡ and Hisashi Yamamoto*,†
Contribution from the Graduate School of Engineering, CREST, Japan Science and Technology
Corporation (JST), and Research Center for AdVanced Waste and Emission Management (ResCWE),
Nagoya UniVersity, Furo-cho, Chikusa, Nagoya 464-8603, Japan
ReceiVed April 3, 2000
Abstract: Syntheses of polycyclic isoprenoids have been achieved by several groups; however, no general
“biomimetic” method has yet been reported. In this paper we describe the biomimetic cyclization of simple
isoprenoids to polycyclic isoprenoids using Lewis acid-assisted chiral Brønsted acids, “chiral LBAs”. This is
the first example of a proton-induced enantioselective ene cyclization in synthetic chemistry. Geranyl phenyl
ethers, o-geranylphenols, and geranylacetone derivatives were directly cyclized at -78 °C in the presence of
(R)-binaphthol derivatives and tin tetrachloride. During the cyclization, [1,3] abnormal Claisen rearrangement
often took place. The enantioselectivities were up to 90% ee. Compounds bearing a farnesyl group could also
be cyclized under the same conditions to give the natural products (-)-Ambrox and (-)-chromazonarol. These
chiral LBAs recognize a trisubstituted terminal olefin enantiotopically and generate site-selective carbocations
on the substrates. The absolute stereochemistry of the cyclization is discussed with model studies using DFT
calculations on the B3LYP/LANL2DZ level.
Introduction
Scheme 1. A Hypothetical Simple Cyclization Catalyzed by
a “Cyclase”
Many isoprenoids (over 30 000 compounds) have been
characterized, identified, and reported. They play important roles
in stabilizing membranes, and in the construction of signal
transduction networks, visual pigments, antibiotics, etc.1,2 Iso-
prenoids have extraordinarily diverse structures, which are
created by enzymes called “cyclases”. Various polycyclic
isoprenoids are generated from simple linear polyene substrates
such as geranyl pyrophosphate, farnesyl pyrophosphate, gera-
nylgeranyl pyrophosphate, and squalene. The fundamental
skeletons of polycyclic isoprenoids have many chiral centers
including quaternary carbons and are mainly constructed by such
cyclases in a single cyclization reaction,3 recognized as an
enzyme-controlled asymmetric ene reaction (Scheme 1). Diverse
families of isoprenoid structures, often formed from the same
substrate in an enzyme-specific manner, are thought to be based
on these four steps:4 (1) generation of the carbocation, (2) control
over the conformation of the substrate, (3) stabilization of
intermediates, and (4) quenching of the final carbocation.
Carbocations generated on substrates are stabilized during the
successive cyclization. Shifts of hydrogens and/or alkyl groups
often take place, and the final carbocations are quenched to give
the product. Generation of the carbocation is the most important
step, since this is truly the first step in achieving complete
absolute and relative stereoselectivity. There are three primary
routes by which carbocations are generated: pyrophosphate
elimination, olefin protonation, and epoxide ring opening.1 In
mono- and sesquiterpene biosyntheses, most initial carbocations
are generated by pyrophosphate elimination. Some of the
triterpenes such as sterols are biosynthesized by diastereose-
lective cyclization of enantiopure 2,3-oxidosqualene through
epoxide opening. The protonation of olefin is the most important
route especially in the biosyntheses of longer isoprenoids.
No catalyst with a cavity analogous to that of a natural
enzyme has yet been designed in synthetic chemistry. The most
important feature required for an artificial cyclase is that
asymmetric induction in the protonation of the terminal isoprenyl
group of isoprenoids, and the successive cyclization without
* To whom correspondence should be addressed.
† The Graduate School of Engineering.
‡ ResCWE.
(1) ComprehensiVe Natural Products Chemistry. Vol. 2 Isoprenoids
Including Cartenoids and Steroids; Sir Derek, D., Nakanishi, K., Meth-
Cohen, O., Cane, D. E., Eds.; Elsevier Science Ltd.: Oxford, 1999.
(2) Ourisson, G.; Nakatani, Y. Chem. Biol. 1994, 1, 11.
(3) Abe, I.; Rohmer, M.; Prestwich, G. D. Chem. ReV. 1993, 93, 2189.
(4) (a) Tarshis, L. C.; Yan, M.; Poulter, C. D.; Sacchettini, J. C.
Biochemistry 1994, 33, 10871. (b) Lesburg, C. A.; Zhai, G.; Cane, D. E.;
Christianson, D. W. Science 1997, 277, 1811. (c) Starks, C. M.; Back. K.;
Chappell, J.; Noel, J. P. Science 1997, 277, 1815. (d) Wendt, K. U.; Poralla,
G. E.; Schulz, G. E. Science 1997, 277, 1820.
10.1021/ja001165a CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/10/2000

8132 J. Am. Chem. Soc., Vol. 122, No. 34, 2000
Nakamura et al.
Scheme 2. In Situ Preparation of (R)-1-4‚SnCl4
an enzymatic cavity must be done at low temperature to control
the substrate conformation. The stereochemical implications of
polyene cyclizations initiated by protonation at the terminal
C-C double bond can be explained by the Stork-Eschenmoser
hypothesis,5 which postulates synchronous internal anti additions
via chairlike conformations of nascent cyclohexane rings.
Computational6 and preparative studies7 on the cyclization that
begins with epoxide opening have been done to reveal that the
first step must be the concerted step; however, the concertedness
of the overall ring-forming process is a matter of debate.8
Despite extensive studies on acid-catalyzed diastereoselective
polyene cyclizations using chiral auxiliaries9 and antibody-
catalyzed diastereoselective cyclization,10 enantioselective pro-
cesses using synthetic chiral catalysts have not yet been re-
ported.
We have found that the Lewis acid-assisted chiral Brønsted
acids (chiral LBAs)11,12 serve as chiral Brønsted acid catalysts
and reagents. The coordination of a Lewis acid to a Brønsted
acid restricts directional access to the proton and increases the
Brønsted acidity. And the chiral LBAs, which are generated
from chiral Brønsted acids, provide us the useful protons
surrounded by chiral environments. We have previously reported
enantioselective protonations11a,b of silyl enol ethers using an
optically active (R)-2,2′-dihydroxy-1,1′-binaphthyl ((R)-BINOL
or (R)-1)‚SnCl4 complex, as well as its catalytic version11c using
(R)-2-hydroxy-2′-methoxy-1,1′-binaphthyl ((R)-BINOL-Me or
(R)-2)‚SnCl4 complex as a catalyst and 2,6-dimethylphenol as
an achiral proton source.12 Recent research has revealed the
isomerization11g of TBDMS enol ethers under the same condi-
tions and indicates the existence of carbocation intermediates.
The aliphatic substrates such as geranyltrimethyltin and meso-
enediol disilyl ether were also enantiomerically protonated by
chiral LBA.11e,f We report here that (R)-LBAs, (R)-1-4‚SnCl4
(Scheme 2), are useful as artificial geranyl and farnesyl cyclases.
Our chiral LBAs gave their protons to the terminal trisubstituted
olefin of simple isoprenoids enantiotopically to initiate ene
cyclizations.
Results and Discussion
Cyclization of Polyolefinic Alcohols. First, the enantiose-
lective cyclization of geranyl derivatives was examined using
(R)-LBAs, (R)-1-4‚SnCl4, in dichloromethane at -78 °C. For
example, the reaction of dienol 5, prepared by methylation of
geranyl acetone, with 1 equiv of (R)-2‚SnCl4 gave a diastere-
omeric mixture of decalins 6 and 7 in 58% yield (trans-6:cis-7
) 88:12), and the enantiomeric excess of the major isomer 6
was 56% (eq 1). The use of (R)-2‚SnCl4 gave slightly better
(5) (a) Woodward, R. B.; Bloch, K. J. Am. Chem. Soc. 1953, 75, 2023.
(b) Eschenmoser, A.; Ruzicka, L.; Jeger, O.; Arigoni, D. HelV. Chim. Acta
1955, 38, 1890. (c) Stork, G.; Burgstahler, A. W. J. Am. Chem. Soc. 1955,
77, 5068. (d) van Tamelen, E. E. Pure Appl. Chem. 1981, 53, 1259. (e)
Johnson, W. S. Tetrahedron 1991, 47 (41), xi-xxiii.
(6) (a) Gao, D.; Pan, Y-. K.; Byun, K.; Gao, J. J. Am. Chem. Soc. 1998,
120, 4045. (b) Corey, E. J.; Staas, D. D. J. Am. Chem. Soc. 1998, 120,
3526.
enantioselectivity than the use of (R)-1‚SnCl4. This tendency is
similar to that in the enantioselective protonation reported
earlier.12 The Lewis acids such as hafnium tetrachloride and
zirconium tetrachloride were usable in the enantioselective
cyclization, but were not completely dissolved in the reaction
solution. The cyclization of homogeraniol (8) with (R)-2‚SnCl4
successfully proceeded to give trans-fused bicyclic compound
9 as a major isomer with 49% ee (eq 2). Compound 9 is known
to be a natural product.13
(7) (a) Corey, E. J.; Virgil, S. C. J. Am. Chem. Soc. 1991, 113, 4025.
(b) Corey, E. J.; Virgil, S. C.; Sarshar, S. J. J. Am. Chem. Soc. 1991, 113,
4025. (c) Corey, E. J.; Virgil, S. C.; Liu, D. R.; Sarshar, S. J. J. Am. Chem.
Soc. 1992, 114, 1524. (d) Corey, E. J.; Virgil, S. C.; Cheng, H.; Baker, C.
H.; Matsuda S. P. T.; Singh, V.; Sarshar, S. J. Am. Chem. Soc. 1995, 117,
11819. (e) Corey, E. J.; Cheng, H.; Baker, C. H.; Matsuda S. P. T.; Li, D.;
Song, X. J. Am. Chem. Soc. 1997, 119, 1277. (f) Corey, E. J.; Cheng, H.;
Baker, C. H.; Matsuda S. P. T.; Li, D.; Song, X. J. Am. Chem. Soc. 1997,
119, 1289. (g) Pale-Grosdemange, C.; Feil, C.; Rohmer, M.; Poralla, K.
Angew. Chem., Int. Ed. Engl. 1998, 37, 2237.
(8) Snowden, R. L.; Eichenberger, J.-C.; Linder, S. M.; Sonnay, P.; Vial,
C.; Schulte-Elte, K. H. J. Org. Chem. 1992, 57, 955.
(9) (a) Bartlett, P. A. In Asymmetric Synthesis; Morrison, J. D., Ed.;
Academic Press: New York, 1982; vol. 3, part B, p 390. (b) Corey, E. J.;
Lee, J. J. Am. Chem. Soc. 1993, 115, 8873. (c) Heinemann, C.; Demuth,
M. J. Am. Chem. Soc. 1999, 121, 4894.
(10) (a)Hasserodt, J.; Janda, K. D.; Lerner, R. A. J. Am. Chem. Soc.
1997, 119, 5993. (b) Hasserodt, J.; Janda, K. D.; Lerner, R. A. J. Am. Chem.
Soc. 2000, 122, 40.
(-)-Ambrox (12), which is found in Givenchy’s Extravagance
d’Amarige, for example, is the most important commercial
substitute for ambergris,14 due to its unique olfactive and fixative
(11) (a) Ishihara, K.; Kaneeda, M.; Yamamoto, H. J. Am. Chem. Soc.
1994, 116, 11179. (b) Ishihara, K.; Nakamura, S.; Yamamoto, H. Croat.
Chem. Acta 1996, 69, 513. (c) Ishihara, K.; Nakamura, S.; Kaneeda, M.;
Yamamoto, H. J. Am. Chem. Soc. 1996, 118, 12854. (d) Yanagisawa, A.;
Ishihara, K.; Yamamoto, H. Synlett 1997, 411. (e) Taniguchi, T.; Ogasawara,
K. Tetrahedron Lett. 1997, 38, 6429. (f) Ishihara, K.; Ishida, Y.; Nakamura,
S.; Yamamoto, H. Synlett 1997, 758. (g) Ishihara, K.; Nakamura, H.;
Nakamura, S.; Yamamoto, H. J. Org. Chem. 1998, 63, 6444. (h) Ishihara,
K.; Nakamura, S.; Yamamoto, H. J. Am. Chem. Soc. 1999, 121, 4906.
(12) Nakamura, S.; Kaneeda, M.; Ishihara, K.; Yamamoto, H. J. Am.
Chem. Soc. 2000, 122, 8120-8130.
(13) (a) Neumann, S.; Simon, H. Biol. Chem. Hoppe-Seyler 1986, 367,
723. (b) Mookherjee, B. D.; Patel, R. R. Int. Congr. Essent. Oils [Pap.],
7th, 1977; Japan Flavor Fragrance Manufactures’ Assoc.: Tokyo, Japan,
1979; Vol. 7, pp 179-82.
(14) Ohloff, G.; Winter, B.; Fehr, C. In Perfumes, Art, Science and
Technology; Muller, P. M., Lamparsky, D., Eds.; Elsevier Applied Science:
New York, 1991; p 289.

Cyclization of Isoprenoids Using Chiral LBAs
J. Am. Chem. Soc., Vol. 122, No. 34, 2000 8133
properties. Its scarcity has been a stimulus for chemical
synthesis.8,15 The successful preparation of (-)-12 was achieved
by the enantioselective cyclization of homofarnesol (11) pro-
moted with (R)-2‚SnCl4, although the enantioselectivity and
diastereoselectivity were moderate (eq 3). Minor products 13-
of (R)-BINOL and tin tetrachloride was the most effective for
controlling the absolute and relative stereochemistries in this
cyclization (54% ee, 95% ds). It seems that the stereoselectivity
depends on the reactivity of the chiral LBA as a promoter; the
reactivities decreased in the order (R)-1‚SnCl4, (R)-2‚SnCl4, and
(R)-4‚SnCl4. The cyclization was accelerated by the coordination
of tin tetrachloride to several Brønsted acids, although the
reaction proceeded slowly even in the presence of tin tetrachlo-
ride alone. Unfortunately, (R)-4‚SnCl4 was not effective for the
cyclization of aliphatic substrates 5, 8, and 11, probably because
its Brønsted acidity is too low.
On the other hand, the cis-fused bicyclic compound 18 was
observed as the major product under the same conditions using
nerylphenol (19) (eq 5). However, the conversion and enantio-
15 obtained were also identified by comparison with authentic
samples.15g,h Homofarnesol (11) was produced from com-
mercially available nerolidol in two steps according to the known
procedure.16 This three-step total synthesis of (-)-Ambrox is
shorter than any previously reported enantioselective total
synthesis.8,15
Cyclization of Polyolefinic Phenol Derivatives. The catalytic
activity of chiral LBAs is inhibited to some degree by a hydroxy
group which serves as an internal nucleophilic terminator in
polyolefinic alcohols. To investigate the present cyclization
system in detail, we chose more reactive o-geranylphenol (16).
The cyclization of 16 with (R)-1‚SnCl4 in dichloromethane at
-78 °C was completed within a day, and the trans-fused tricyclic
compound 17 was obtained as a major diastereomer (84% ds)
(eq 4).17 However, the optical yield of 17 was only 36% ee.
selectivity were slightly lower than in the previous cases, and
undesired monocyclic products were also produced, probably
because cyclization was unfavorable.
In further studies, we found that 21 was obtained with much
better selectivity from the reaction of geranyl phenyl ether 20
(R1 ) R2 ) H) with (R)-4‚SnCl4 (Table 1, entry 1). The proton
in (R)-4‚SnCl4 is strong enough to promote the reaction and
yet weak enough to control the reaction to achieve a higher ee.
Surprisingly, the reaction proceeded smoothly even in the
presence of 20 mol % (R)-4‚SnCl4 to give 21 with 77% ee and
98% ds (entry 2). Other examples are summarized in Table 1.
Table 1. Enantioselective Cyclization of Geranyl Ary Ether 20
20
21
(day) yield (%)a ee (%)b 21:22
(R)-LBA 4 time
(equiv)
ratioc
entry R1
R2
The enantioselectivity was improved to 50% ee by using (R)-
2‚SnCl4.11c Among various (R)-BINOL derivatives screened, we
found that (R)-4‚SnCl4 prepared from the monobenzoyl ester
1
2
3
4
5
6
7
8
9
H
H
F
F
Cl
Cl
Br
Br
Br
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
Me
1.1
0.2
1.1
0.2
1.1
0.2
1.1
0.2
0.15
1.1
0.2
1.1
0.2
1.1
0.2
1
4
1
4
1
4
1
4
6
1
4
1
4
1
4
98 (79)
98 (76)
98
72
99
97
87
85 (71)
94
92
94
84
92
80
69
77
63
79
65
82
63
87
90
62
67
70
42
62
46
98:2
98:2
94:6
70:30
98:2
97:3
94:6
89:11
95:5
95:5
97:3
95:5
94:6
89:11
91:9
(15) (a) Stoll, M.; Hinder, M. HelV. Chim. Acta 1950, 33, 1251. (b)
Hinder, M.; Stoll, M. HelV. Chim. Acta 1950, 33, 1308. (c) Tanimoto, H.;
Oritani, T. Tetrahedron: Asymmetry 1996, 7, 1695. (d) Kutney, J. P.; Cirera,
C. Can. J. Chem. 1997, 75, 1136. (e) Akita, H.; Nozawa, M.; Shimizu, H.
Tetrahedron: Asymmetry 1998, 9, 1789. (f) Barrero, A. F.; Altarejos, J.;
Alvarez-Manzaneda, E. J.; Ramos, J. M.; Salido, S. J. Org. Chem. 1996,
61, 2215. (g) Ohloff, G.; Giersch, W.; Pickenhagen, W.; Furrer, A.; Frei,
B. HelV. Chim. Acta 1985, 68, 2022. (h) Escher, S.; Giersch, W.; Niclass,
Y.; Bernardinelli, G.; Ohloff, G. HelV. Chim. Acta 1990, 73, 1935. (i)
Maleczka, R. E., Jr.; Paquette, L. A. J. Org. Chem. 1991, 56, 6538. (j)
Kutney, J. P.; Chen, Y.-H. Can. J. Chem. 1994, 72, 1570. (k) Barrero, A.
F.; Altarejos, J.; Alvarez-Manzaneda, E. J.; Ramos, J. M.; Salido, S.
Tetrahedron 1993, 49, 6251.
10 Me
11 Me
12 OMe
13 OMe
14
15
H
H
82
a Unless otherwise noted, GC yields are indicated. Isolated yields
are indicated in parentheses. b Ee values were determined by GC or
HPLC analysis of isolated pure product. c Ratios were determined by
GC or HPLC analysis of crude products in which other minor products
were included.
(16) Barrero, A. F.; Altarejos, J.; Alvarez-Manzaneda, E. J.; Ramos, J.
M.; Salido, S. J. Org. Chem. 1996, 61, 2215.
(17) (a) Yamada, S.; Katagiri, T.; Tanaka, J. Yuki Gosei Kagaku Kyokai
Shi 1971, 29, 81. (b) Shigemasa, Y.; Kuwamoto, H.; Sakazawa, C.;
Matsumura, T. Nippon Kagaku Kaishi 1973, 2423.

8134 J. Am. Chem. Soc., Vol. 122, No. 34, 2000
Nakamura et al.
Scheme 3. Possible Pathways from Geranyl Phenyl Ethers
20 to the Tricyclic Products 21 or 17
Figure 1. Percent ee value observed in the cyclization of 20 with (R)-
4‚SnCl4 at -78 °C in CH2Cl2 as a function of the σp parameter for R1
in 20. (O) and (b) refer to the use of a stoichiometric amount of (R)-
4Sn‚Cl4 and the use of a catalytic amount (20 mol %) of (R)-4‚SnCl4,
respectively.
The use of (R)-4‚SnCl4 without exception resulted in the highest
enantioselectivity and diastereoselectivity; however, the catalytic
use of (R)-4‚SnCl4 for substrates 20 with electron-donating
groups (R1 ) OMe, R2 ) H) or bulky groups (R1 ) H, R2 )
Me) on their phenols reduced the enantioselectivity (entries 12-
15). The reaction on substrates 20 with electron-withdrawing
groups on their phenols (R1 ) Br, R2 ) H), in contrast, gave
21 and increased the enantioselectivity to 87% ee (entry 8) and
90% ee (entry 9). These results indicate the importance of
performing the reaction under mild conditions appropriate for
each substrate to achieve maximal enantioselectivity and dia-
stereoselectivity. The relative stereochemistry of the major
product 21 was determined to be trans on the basis of the
similarity of the 1H and 13C NMR spectra to those of 21 (R1 )
H, R2 ) Me), which had been assigned as trans by an X-ray
analysis (see the Supporting Information). The absolute stere-
ochemistry of 21 produced by (R)-LBAs is speculated to be
(4aS,9aS) on the basis of the absolute stereopreference on the
enantioselective cyclization of 11.
A plot of the ee value observed in the cyclization of 20 with
(R)-4‚SnCl4 as a function of the σp parameter18 for the
substituent R1 on the aromatic ring in 20 is shown in Figure 1.
Interestingly, the ee value observed in the reaction with catalytic
amounts of the (R)-LBA was increased in proportion to the σp
parameter. The ee value observed in the reaction with a
stoichiometric amount of the (R)-LBA, however, was about 60-
70% ee and was independent of the σp parameter. These
phenomena can be understood by assuming that the catalytic
use of LBA and weak Lewis basicity of the oxygen atom in 20
relatively suppress the undesired coordination of LBA or tin
tetrachloride with the oxygen atom during the cyclization.
Nevertheless, it is not clear why there is a difference between
stoichiometric and catalytic amounts of LBA.
rearrangement before subsequent cyclization (path A). While
we have not yet found any direct evidence of intermediates,
we have a suggestive finding for path A: Although it is known
that abnormal Claisen rearrangements generally do not occur
smoothly,19 the use of (R)-2‚FeCl3 for the reaction of 20
(R1 ) R2 ) H) gave 16 in 76% yield, and trace amounts of 21
and 22 were also observed. On the contrary, it seems that
phenoxy migration after proton-induced cyclization (path B) is
unfavorable because the C-O bond cleavage with the genera-
tion of unstable primary carbocation formation is quite un-
likely.
Control experiments were performed with the metal halides
tin tetrachloride, hafnium tetrachloride, and gallium trichloride
under the same conditions as in Table 1, and the results are
shown in Table 2. Although all three metal halides could
promote the ene cyclization as Lewis acid (entries 1, 8, and
13), high enantioselectivity was not observed without a hydroxy
group on the ligands (entries 6, 7, 10, 11, 15, and 16). The
presence of OH on the ligand is not quite essential as stated
since ligands 29 and 30 gave modest enantiomeric excesses
(11-19%). No reactivity was observed using hafnium tetra-
chloride or gallium trichloride with aprotic bidentate ligand (R)-
28 (entries 11 and 16). Interestingly, the product ratio in the
ene reaction is mostly dependent on the metal. The use of tin
tetrachloride and its coordinate complexes produced 17 selec-
tively, and the use of (R)-2‚SnCl4 and (R)-4‚SnCl4 produced
17 with good enantiomeric excesses (entries 2 and 5). In
contrast, the use of hafnium tetrachloride, zirconium tetrachlo-
ride, and their coordinate complexes were realtively less
selective for producing 17, and their use as chiral LBA was
also less effective. For example, (R)-2‚HfCl4 and (R)-2‚ZrCl4
produced monocyclization product 25 in 23% and 27% yields,
respectively, together with 17 (entries 9 and 17). In both cases,
the cis isomer of 26, which was generated via a stepwise
cyclization from 25,21 was observed along with other isomers.
Interestingly, (R)-2‚GaCl3 generated unrearranged cyclic com-
pound 26 in high yield but with low enantiomeric excess (entry
Geranyl phenyl ether 20 is more reactive than 16, probably
because of the lack of a hydroxy group in the former. The
existence of the hydroxy group in the substrate reduces the
catalytic activity of LBA by its coordination with tin tetrachlo-
ride. Although it is surmised that the reaction of geranyl aryl
ether 20 takes place via [1,3]-rearrangement (abnormal Claisen
rearrangement)19 and cyclization, it is not clear which of these
two steps occurs first (Scheme 3). Nevertheless, it is reasonable
to suppose that the reaction takes place via abnormal Claisen
(20) (a) Imamura, K.; Yoshikawa, E.; Gevorgyan, V.; Yamamoto, Y. J.
Am. Chem. Soc. 1998. 120, 5339. (b) Kido, Y.; Yoshimura, S.; Yamaguchi,
M.; Uchimaru, T. Bull. Chem. Soc. Jpn. 1999, 72, 1445. (c) Yamaguchi,
M.; Arisawa, M.; Omata, K.; Kabuto, K.; Hirama, M.; Uchimura, T. J.
Org. Chem. 1998, 63, 7298.
(21) Ghosh, S.; Banik, B. K.; Ghatak, U. R. J. Chem. Soc., Perkin Trans.
1 1991, 3189.
(18) (a) Jaffe, H. H. Chem. ReV. 1953, 53, 191. (b) Yukawa, Y.; Tsuno
Y.; Sawada, M. Bull. Chem. Soc. Jpn. 1972, 45, 1198.
(19) (a) Grieco, P. A.; Clark, J. D.; Jagoe, C. T. J. Am. Chem. Soc. 1991,
113, 5488. (b) Palani, N.; Balasubramanian, K. K. Tetrahedron Lett. 1993,
34, 5001.

Cyclization of Isoprenoids Using Chiral LBAs
J. Am. Chem. Soc., Vol. 122, No. 34, 2000 8135
Table 2. Enantioselective Cyclization of Geranyl Phenyl Ether 20
(R1dR2dH) Using Various Combinations of Chiral Broønsted
Acidsa and Lewis Acids
ratiob
chiral Brønsted
entry acid‚Lewis acid
17
18
25
26
othersc 20
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
SnCl4
(R)-2‚SnCl4
89
3
0
0
0
0
0
0
0
0
0
0
0
0
0
57
8
3
12
33
0
0
0
0
0
0
0
0
0
0
0
83 (46) 10
4
0
0
0
0
0
5
We simultaneously control four chiral centers including two
quaternary carbons with the enantioselective ene cyclization.
Syntheses of (-)-Ambrox and (-)-chromazonarol acetate
indicate that these (R)-LBAs approach the re face of a trisub-
stituted terminal olefin uniformly. The number of cyclizations
and the diastereoselectivity are dependent on the spontaneous
conformation of substrates in media at low temperature.
Enzymes control this with cavities made by aromatic residues,
which also stabilize carbocations on intermediates. The design
and construction of a cavity is a possible next step to achieve
a higher yield and enantioselectivity.
(R)-27‚SnCl4 83 (0)
(R)-28‚SnCl4 65 (0)
5
2
2
6
4
(R)-4‚SnCl4
98 (79)
(R)-29‚SnCl4 94 (19)d
(R)-30‚SnCl4 96 (11)d
HfCl4
53
37
5
(R)-2‚HfCl4
(R)-27‚HfCl4 26 (0)
(R)-28‚HfCl4
(R)-4‚HfCl4
GaCl3
23 (46) 11 23 (50)
19 13 (0)
0
69 (0)
16
4
43e
42e
0
0
0
>99
0
0
8
19
2
4
0
0
0
2
2
0
0
23
6
0
0
0
(R)-2‚GaCl3
82 (11) 10
81 (0)
0
0
(R)-27‚GaCl3 11 (0)
4
0
DFT Calculations. In the previous report,12 the properties
of (R)-1‚SnCl4 and (R)-2‚SnCl4 were discussed and estimated
using DFT calculations. (R)-4‚SnCl4 was the most efficient in
the enantioselective ene reaction of geranyl aryl ethers 20.
However, less reactive substrates such as 5, 8, and 11 were not
cyclized under the same conditions, probably because (R)-4‚
SnCl4 was not sufficiently acidic to cyclize them. In contrast,
the enantioselective protonation of “thermodynamic” trimeth-
ylsilyl enol ether 34 derived from 2-phenylcyclohexanone did
not proceed using (R)-4‚SnCl4 (eq 7), although silyl enol ethers
(R)-28‚GaCl3
(R)-2‚ZrCl4
0
>99
35 (38) 19 27
19e
0
a The compounds 27-30 are not actually Brønsted acids, but we
refer to them as “Brønsted acids” by analogy. b Unless otherwise noted,
GC yields are indicated. Ee values are indicated in parentheses and
were determined by GC or HPLC analysis of isolated pure product.
c Trace amounts of chlorinated products, which were observed on GC-
MS, are included in “others”. d Enantioselectivity was reversed to give
the (+)-enantiomer. e The cis isomer of 26 was included.
14). Hafnium tetrachloride, gallium trichloride, and zirconium
tetrachloride may have the limited ability to promote abnormal
Claisen rearrangement prior to the cyclization, and produce
unrearranged cyclic compounds as a result. As mentioned above,
(R)-2‚FeCl3 under the same conditions gave only abnormal
Claisen-rearranged 16. Taken together, these results suggest that
whether the rearrangement takes place is largely based on the
properties of Lewis acids themselves.19
should be more reactive than simple olefins for accepting
protons. What is responsible for this difference? We estimated
it using Becke’s three-parameter hybrid method and the Lee-
Yang-Parr correlation functional (B3LYP), and all charges
shown in this paper were evaluated by the natural population
analysis (NPA).23
The optimized structure of a monobenzoate of biphenol
(biphenol-Bz)‚SnCl4 complex as a model structure of (R)-4‚
SnCl4 was determined at the B3LYP/LANL2DZ24 level to
understand this difference (Figure 2). The calculations predict
that the chelation of biphenol-Bz with tin tetrachloride occurs
at equatorial-equatorial sites. The bipyramidal coordination
includes four chlorines and two oxygens, one each from the
We now have several candidates to use in making a variety
of chiral LBA reagents with various activities, although those
containing tin tetrachloride are still the first choice.
To demonstrate the effectiveness of chiral LBA-promoted
enantioselective cyclization, we synthesized acetate 32 of (-)-
chromazonarol (31), a minor constituent of the brown Pacific
seaweed, Dictyopteris undulata,22 biomimetically from the
corresponding farnesyl derivatives. The cyclization of 4-ben-
11e
zyloxyphenyl farnesyl ether (33) with (S)-3‚SnCl4 gave the
desired tetracyclic compound 32 as a major diastereomer in 44%
ee after debenzylation and acylation (eq 6).
(22) (a) Fenical, W.; McConnell, O. Experientia 1975, 31, 1004. (b)
Cimino, G.; De Stefeano, S.; Minale, L. Experientia 1975, 31, 1117. (c)
Djura, P.; Stierle, D. B.; Sullivan, B.; Faulkner, J.; Arnold, E.; Clardy, J. J.
Org. Chem. 1980, 45, 1435. (d) Rodr´ıguez, J.; Quinoa´, E.; Riguera, R.;
Peters, B. M.; Abrell, L. M.; Crews, P. Tetrahedron 1992, 48, 6667.
(23) NBO 4.0: Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.;
Carpenter, J. E.; Weinhold, F., Theoretical Chemistry Institute, University
of Wisconsin, Madison, 1996.
(24) (a) Dunning, T. H., Jr. J. Chem. Phys. 1970, 53, 2823. (b) Wadt,
W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284.

8136 J. Am. Chem. Soc., Vol. 122, No. 34, 2000
Nakamura et al.
Figure 2. Optimized structures of the biphenol-Bz‚SnCl4 complex (“e”
) equatorial, “a” ) apical). Calculated on the B3LYP/LANL2DZ level
(total energy -1 012.474 715 52 au). Selected distances (Å): H1-O2
) 0.981, H1-Cla4 ) 2.440, O2-Sn3 ) 2.419, Sn3-Cla4 ) 2.402,
Sn3-Cla5 ) 2.376, Sn3-Cle6 )2.361, Sn3-Cle7 ) 2.365, Sn3-O8-
(carbonyl) ) 2.331. Torsion angles (deg): H1-O2-Sn3-Cla4 )
-11.2, C10-C11-C12-C13 ) -97.8.
carbonyl and the hydroxy groups. The acidic proton is likely to
be located at pseudoaxial sites parallel to an apical axis of the
tin atom, and electrostatic interaction between the acidic protons
and the apical chlorides is expected. Surprisingly, the axial angle
(the torsion angle C10-C11-C12-C13) of the biphenol moiety
is -97.8°, which is almost 40° larger than those of biphenol‚
SnCl4 (-52.9°)12 and monomethyl ether of biphenol (biphenol-
Me)‚SnCl4 (-53.3°),12 and the activated proton is in the center
of the chiral environment as a result. The acidity of the protons
was truly enhanced by coordinated tin tetrachloride (charge on
the proton 0.541) compared to the proton on a phenol (charge
on the proton 0.489). As expected, the extent of the enhancement
was a little less than with biphenol‚SnCl4 and biphenol-Me‚
SnCl4 (charge on each proton 0.546 and 0.549, respectively).12
This sterically crowded chiral LBA may provide a proton to
a trisubstituted olefin such as the terminal olefin of linear
isoprenoid groups in the enantioselective ene cyclization,
however, not to a four-substituted olefin such as silyl enol ethers
in the enantioselective protonation.12 Similar results were
observed using (R)-2,2′-dihydroxy-3,3′-dimethyl-1,1′-binaphthyl
(35) (eq 7). (R)-35‚SnCl4 complex, which has bulky groups near
the activated protons, promoted the enantioselective ene reaction
on 20 (R1 ) H, R2 ) Me) under the same conditions to produce
21 (R1 ) H, R2 ) Me) in 59% yield (47% ee), while the chiral
LBA showed no reactivity in the enantioselective protonation
with trimethylsilyl enol ether 34. Additionally, the reagent (R)-
35‚SnCl4 exhibited 44% ee selectivity and regular reactivity on
protonation of a trimethylsilyl enol ether, 36, derived from
2-methylcyclohexanone (42% ee using (R)-1‚SnCl4),25 which
is a sterically less bulky substrate than the silyl enol ether 34
(eq 7). The relatively small silyl enol ether 36 was also
accessible to the proton on (R)-4‚SnCl4 to produce 2-methyl-
cyclohexanone (37% ee) (eq 7). All these results confirm our
explanations.
Figure 3. Optimized structures A and B of the biphenol‚SnCl4 complex
with 2-methyl-2-butene, which respectively produce the major enan-
tiomer and the minor enantiomer (“e” ) equatorial, “a” ) apical).
Calculated on the B3LYP/LANL2DZ level. (A) Total energy
-873.637 155 338 au). Selected distances (Å): H1-C6 ) 2.153, H1-
C7 ) 2.361. Torsion angles (deg): H1-O2-Sn3-Cla4 ) 39.2, C8-
C9-C10-C11 ) -51.5. (B) Total energy -873.636 010 472 au).
Selected distances (Å): H1-C6 ) 2.193, H1-C7 ) 2.405. Torsion
angles (deg): H1-O2-Sn3-Cla4 ) 38.5, C8-C9-C10-C11 )
-52.0.
complex with 2-methyl-2-butene, exhibited a coordinated
complex, A, to produce major enantiomers and a coordinated
complex, B, to produce minor enantiomers (Figure 3). In the
optimized structure A, the methyl group, C5H3, of 2-methyl-
2-butene is located above the space which is surrounded by the
phenyl ring at the left in the figure and its adjacent apical Cl4.
In the optimized structure B, the C5H3 of the olefin which is
located on the phenyl ring at the right would cause severe steric
repulsion. Torsion angles H1-O2-Sn3-Cl4, 39.2° and 38.5°,
shown in Figure 3, are much larger than those of biphenol‚
SnCl4 and biphenol-Me‚SnCl4 (-5.9° and -2.2°, respectively).12
This phenomenon can be attributed to interaction between the
proton and π electrons of the olefin. Total energies were also
calculated to understand the outcome of the enantioselective
reaction. ∆E was appraised by comparing the two total energies,
and the value would represent ∆∆Gq between ∆Gq producing
the major enantiomer and ∆Gq producing the minor enantiomer
of the reaction model. The estimated ∆∆Gq was -0.72 kcal/
mol, and the evaluated ee value based on the Curtin-Hammett
rule at -78 °C was 73% ee.26 The result in the cyclization of
20 (R1 ) H, R2 ) Me) using (R)-1‚SnCl4 was 58% ee in toluene
at -78 °C. Thus, the actual reaction seemed to be adequately
reflected by the calculation models.
This cascade reaction can produce up to four chiral centers
at once; however, the enantioselectivity must be determined at
the initial protonation step on the terminal trisubstituted olefin
of the substrates. Geometry optimization at the B3LYP/
LANL2DZ level of the reaction models, such as biphenol‚SnCl4
(25) Unpublished results.
(26) Enders, D. Chemtech 1981, 504.

Cyclization of Isoprenoids Using Chiral LBAs
J. Am. Chem. Soc., Vol. 122, No. 34, 2000 8137
1
7 was effected by comparison of their H NMR data with those of
The optimized structure A leads to the predominant approach
of (R)-LBAs to the si face of a terminal linear isoprenyl group.
authentic samples.29 GC (â-TA, 100 kPa, column temperature 50 °C
for 3 min and then warm to 150 °C (+1 °C/min)) tR ) 36.9 (7), 38.4
((+)-6), 39.0 ((-)-6) min.
Conclusion
trans-Octahydro-2,2,5,5,8a-pentamethyl-2H-1-benzopyran (6):28,29
1
In summary, an enzyme-like reagent-controlled ene reaction
was achieved with chiral BINOL derivative‚SnCl4 complexes.
These reagents, which we call artificial cyclases, catalyzed the
cyclization of simple nonactivated isoprenoids with 40-90%
ee. The number of cyclizations and diastereoselectivities
primarily depends on the spontaneous conformation of the
substrates in media at low temperature. Asymmetric recognition
of chiral LBAs for terminal olefin to generate the initial
carbocation can control the absolute stereochemistry for the
successive cyclization. On the other hand, enzymes control this
by cavities made of aromatic residues, which also stabilize
carbocations on the substrate.27
Nonenzymatic enantioselective polyene cyclizations are very
attractive alternatives to the multistep synthesis from naturally
occurring chiral synthons. Further studies on the rational design
of “chiral proton catalysts” based on the concept of chiral LBA
are expected to provide practical artificial cyclases for the
asymmetric synthesis of a wide range of polycyclic isoprenoids.
This chiral proton-initiated asymmetric ene reaction may also
be found to be an efficient general synthetic method soon.
56% ee; TLC (hexane:CH2Cl2 ) 4:1) Rf ) 0.11; H NMR (CDCl3,
300 MHz) δ 0.78 (s, 3H), 0.90 (s, 3H), 1.18 (s, 3H), 1.27 (s, 3H), 1.29
(s, 3H), 1.20-1.80 (m, 11H); [R]25.1 ) -5.32° (c ) 0.37, CHCl3).
D
cis-Octahydro-2,2,5,5,8a-pentamethyl-2H-1-benzopyran (7): TLC
(hexane:CH2Cl2 ) 4:1) Rf ) 0.18; 1H NMR (CDCl3, 300 MHz) δ 0.96
(s, 3H), 1.06 (s, 3H), 1.20 (s, 3H), 1.22 (s, 3H), 1.33 (s, 3H), 1.10-
1.75 (m, 10H), 1.85-1.95 (m, 1H).
Enantioselective Cyclization of (E)-Homogeraniol (8)30 Promoted
by (R)-2‚SnCl4. The cyclization of 8 was carried out according to the
above representative procedure under the conditions shown in eq 2.
The crude products were purified by column chromatography on silica
gel (eluent, hexane:CH2Cl2 ) 4:1 f 0:1) to give 9 (49% ee) and 10
(44% ee), which were separable, as a 62:38 diastereomeric mixture
(46% yield). Identification of compounds 9 and 10 was effected by
1
comparison of their H NMR data with those of 6, 7, 12, 13, 14, and
15.15,16 GC (γ-TA, 30 kPa, column temperature 85 °C) tR ) 28.8 (minor
isomer of 9), 30.2 (major isomer of 9), 31.7 ((+)-10), 33.7 ((-)-10)
min.
trans-Octahydro-4,4,7a-trimethylbenzofuran (9): 49% ee; TLC
(CH2Cl2) Rf ) 0.27; 1H NMR (CDCl3, 300 MHz) δ 0.85 (s, 3H), 0.96
(s, 3H), 1.07 (s, 3H), other resonances could be not discerned; 13C NMR
(75 MHz, CDCl3) δ 21.19, 23.51, 32.76, 33.12, 40.97, 56.23, other
resonances could be not discerned; LR FAB+-MS m/z 168 ([M]+,
C11H20O1 requires 168.3).
Experimental Section
cis-Octahydro-4,4,7a-trimethylbenzofuran (10): 44% ee; TLC
(CH2Cl2) Rf ) 0.28; IR (film) 3856, 3651, 2928, 1509, 1491, 1221,
General Procedures. Infrared (IR) spectra were recorded on a
Shimadzu FT-IR 8100 spectrometer. H NMR spectra were measured
1
1
773 cm-1; H NMR (CDCl3, 300 MHz) δ 0.92 (s, 3H), 1.02 (s, 3H),
on a Varian Gemini-300 or VXR 500 spectrometer. Tetramethylsilane
1.33 (s, 3H), 1.12-1.63 (m, 6H), 1.90-1.99 (m, 2H), 3.72-3.86 (m,
2H); 13C NMR (75 MHz, CDCl3) δ 19.96, 27.09, 28.62, 29.39, 30.43,
33.59, 34.39, 53.93, 63.90, other resonances could be not discerned;
1
was used as internal standard for H NMR (δ 0.00 ppm), CDCl3 for
13C NMR (δ 77.00 ppm), and CF3C6H5 for 19F NMR (δ -63.90 ppm).
High-performance liquid chromatography (HPLC) was done with
Shimadzu 10A instruments using 4.6 mm × 25 cm Daicel CHIRALCEL
OJ, OD-H, AD, and AS. GC analysis was done with Shimadzu 17A
instruments using γ-TA (0.25 mm × 20 m), â-DA (0.25 mm × 20 m),
â-DM (0.25 mm × 20 m), and PEG (0.25 mm × 25 m). Optical
rotations were measured on a JASCO DIP-1000 digital polarimeter.
Melting points were determined using a Yanaco MP-J3. All experiments
were carried out under an atmosphere of dry argon. For thin-layer
chromatography (TLC) analysis throughout this work, Merck precoated
TLC plates (silica gel 60 GF254, 0.25 mm) were used. The products
were purified by preparative column chromatography on silica gel E.
Merck 9385. Microanalyses were accomplished at the School of
Agriculture, Nagoya University. FAB mass spectral analyses were
accomplished at the Graduate School of Engineering, Nagoya Univer-
sity. In experiments requiring dry solvents, ether and tetrahydrofuran
(THF) were purchased from Aldrich Chemical Co. as “anhydrous” and
stored over 4A molecular sieves. Benzene, hexane, toluene, and
dichloromethane were freshly distilled from calcium hydride. Tin
tetrachloride was distilled under argon. Other simple chemicals were
purchased and used.
LR FAB+-MS m/z 168 ([M]+, C11H20O1 requires 168.3); [R]25.1
-14.5° (c ) 0.10, CHCl3).
)
D
Enantioselective Cyclization of (E,E)-4,8,12-Trimethyl-3,7,11-
tridecatrien-1-ol (11)16 Promoted by (R)-2‚SnCl4. The cyclization of
1116 was carried out according to the above representative procedure
under the conditions shown in Scheme 3. The crude products were
purified by column chromatography on silica gel (eluent, hexane:CH2-
Cl2 ) 4:1 to hexanes:EtOAc ) 10:1) to give four diastereomeric
products of 12-15 as a mixture (54% yield, [R]25.1 ) -8.83° (c )
D
0.25, CHCl3)). Compounds 12-15 were not isolated, and their
1
identification was effected by comparison of their H and 13C NMR
data with those of authentic samples (Tables 3 and 4). The product
distribution presented in eq 3 was determined by GC (PEG). The
absolute stereochemistry of 12 was ascertained by comparison of GC
(γ-TA) data with those of (-)-Ambrox (Aldrich). GC (PEG, 100 kPa,
column temperature 120 °C for 5 min and then warm to 220 °C (+1
°C/min)) tR ) 16.83 (13 or 15), 17.09 (15 or 13), 19.24 (12), 20.75
(14) min; GC (γ-TA, 75 kPa, colum temperature 120 °C for 3 min and
then warm to 150 °C (+1 °C/min)) tR ) 60.55 (13 and 15), 64.14
((-)-12), 65.75 ((+)-12), 70.46 (minor enantiomer of 14), 71.82 (major
enantiomer of 14) min.
Preparation of o-Geranylphenol (16)31 and o-Nerylphenol (19).31
Both compounds were produced with simple methods reported previ-
ously. See ref 31.
General Procedure for the Preparation of Geranyl Aryl Ether
20.32 To a stirred suspension of sodium hydride (60% in oil, 176 mg,
4.4 mmol) in THF (20 mL) at room temperature under argon
Enantioselective Cyclization of (E)-2,6,10-Trimethyl-5,9-undeca-
dien-2-ol (5)28 Promoted by (R)-2‚SnCl4 (Representative Proce-
11c
dure). To a solution of 2 (66.0 mg, 0.22 mmol) in distilled dichlo-
romethane (4 mL) was added a 1.0 M solution of tin tetrachloride in
dichloromethane (20 µL, 0.2 mmol) at -78 °C under argon. After the
mixture was stirred for several minutes at the same temperature, 528
(23.6 mg, 0.1 mmol) was added dropwise at -78 °C. After the resulting
mixture was stirred for 3 days at -78 °C, pyridine (16 µL, 0.2 mmol)
was added. Then the mixture was poured onto a saturated NaHCO3
solution and extracted with ether. The combined organic layers were
dried over anhydrous MgSO4 and concentrated. The crude product was
purified by silica gel column chromatography (eluent, hexane:CH2Cl2
) 4:1) to give 6 (56% ee) and 7, which were separable, as a 88:12
diastereomeric mixture (58% yield). Identification of compounds 6 and
(29) Hoye, T. R.; Caruso, A. J.; Kurth, M. J. J. Org. Chem. 1981, 46,
3550.
(30) Kocienski, P.; Wadman, S.; Cooper, K. J. Org. Chem. 1989, 54,
1215. (b) Nagano, H.; Nagasawam T.; Sakuma, M. Bull. Chem. Soc. Jpn.
1997, 70, 1969.
(31) (a) Tanaka, S.; Ono, F.; Katagiri, T.; Tanaka, J. Bull. Chem. Soc.
Jpn. 1977, 50, 750. (b) Bigi, F.; Casiraghi, G.; Casnati, G.; Sartori, G.
Synthesis 1981, 310.
(27) Dougherty, D. A. Science 1996, 271, 163.
(28) Towne, T. B.; McDonald, F. E. J. Am. Chem. Soc. 1997, 119, 6022.
(32) Sakane, S.; Fujiwara, J.; Maruoka, K.; Yamamoto, H. Tetrahedron
1986, 42, 2193.

8138 J. Am. Chem. Soc., Vol. 122, No. 34, 2000
Nakamura et al.
Table 3. 13C NMR Assignments (δ [ppm], CDCl3, 126 MHz) for
12-15a
Geranyl p-Fluorophenyl Ether (20 (R1 ) F, R2 ) H)):33c TLC
(hexane:CH2Cl2 ) 4:1) Rf ) 0.3; IR (film) 2918, 1507, 1379, 1294,
1244, 1221, 1210, 1096, 1005, 828 cm-1; 1H NMR (300 MHz, CDCl3)
δ 1.60 (s, 3H), 1.68 (s, 3H), 1.72 (s, 3H), 2.02-2.19 (m, 4H), 4.49 (d,
J ) 6.6 Hz, 2H), 5.04-5.13 (m, 1H), 5.47 (tq, J ) 6.6, 1.2 Hz, 1H),
6.84 (dd, J ) 9.0, 4.5 Hz, 2H), 6.96 (t, J ) 9.0 Hz, 2H); 13C NMR (75
MHz, CDCl3) δ 16.64, 17.69, 25.67, 26.25, 39.51, 65.41, 115.54 (J )
2.3 Hz), 115.66, 115.83, 119.34, 123.73, 131.82, 141.32, 154.90, 155.55,
158.71; 19F NMR (282 MHz, CDCl3) δ -125.17.
12b
13c
14d
15d
Geranyl p-Chlorophenyl Ether (20 (R1 ) Cl, R2 ) H)):33b,d TLC
(hexane:CH2Cl2 ) 4:1) Rf ) 0.3; IR (film) 2924, 1491, 1287, 1238,
C(n)
C(1) 39.768 (39.95)* 38.676 (38.7)
C(2) 18.417 (18.39) 18.514 (18.5)
C(3) 42.451 (42.43) 42.317 (42.3)
C(4) 33.080 (33.06) 33.590 (33.6)
C(5) 57.272 (57.25) 46.724 (46.7)
C(6) 20.669 (20.64) 20.420 (20.4)
27.066 (27.1) 41.122 (41.2)
18.994 (19.0) 18.233 (18.3)
35.520 (35.5) 42.554 (42.6)
34.057 (34.1) 33.280 (33.2)
52.635 (52.5) 48.611 (48.7)
23.400 (23.4) 21.391 (21.4)
1
1169, 1094, 1005, 824 cm-1; H NMR (300 MHz, CDCl3) δ 1.60 (s,
3H), 1.68 (s, 3H), 1.73 (s, 3H), 2.04-2.18 (m, 4H), 4.51 (d, J ) 6.6
Hz, 2H), 5.04-5.13 (m, 1H), 5.46 (tq, J ) 6.6, 1.2 Hz, 1H), 6.84 (d,
J ) 9.0 Hz, 2H), 7.22 (d, J ) 9.0 Hz, 2H); 13C NMR (75 MHz, CDCl3)
δ 16.68, 17.70, 25.69, 26.24, 39.51, 65.13, 115.95, 119.11, 123.70,
125.37, 129.22, 131.87, 141.54, 157.42.
C(7) 39.969 (39.73)* 35.781 (35.8)* 38.700 (38.7) 36.279 (36.4)
Geranyl p-Bromophenyl Ether (20 (R1 ) Br, R2 ) H)):33d TLC
(hexane:CH2Cl2 ) 4:1) Rf ) 0.3; IR (film) 2926, 1489, 1237, 1001,
C(8) 79.916 (79.91) 80.820 (80.8)
C(9) 60.136 (60.11) 59.014 (59.0)
78.866 (78.9) 80.104 (80.2)
58.589 (58.6) 49.060 (49.2)
C(10) 36.206 (36.18) 36.030 (36.0)* 37.656 (37.7) 36.103 (36.2)
1
822 cm-1; H NMR (300 MHz, CDCl3) δ 1.60 (s, 3H), 1.68 (s, 3H),
C(11) 22.647 (22.62) 28.850 (28.9)
C(12) 64.986 (64.97) 64.081 (64.1)
25.057 (25.1) 23.491 (23.6)
64.409 (64.4) 64.081 (64.1)
1.73 (s, 3H), 2.02-2.18 (m, 4H), 4.50 (d, J ) 6.6 Hz, 2H), 5.04-5.12
(m, 1H), 5.46 (tq, J ) 6.6, 1.5 Hz, 1H), 6.79 (d, J ) 9.2 Hz, 2H), 7.36
(d, J ) 9.2 Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 16.67, 17.70, 25.69,
26.22, 39.50, 65.04, 112.66, 116.47, 119.03, 123.68, 131.87, 132.15,
141.58, 157.90.
C(17) 21.154 (21.13) 21.761 (21.8)** 29.687 (29.7) 25.870 (25.9)
C(18) 33.590 (33.58) 32.928 (32.9) 33.280 (33.3) 27.873 (28.0)
C(19) 21.154 (21.23) 28.829 (22.8)** 31.411 (31.4) 34.458 (34.5)
C(20) 15.049 (15.03) 27.709 (27.7) 20.614 (20.6) 22.046 (22.1)
Geranyl p-Tolyl Ether (20 (R1 ) Me, R2 ) H)):33b TLC (hexane:
CH2Cl2 ) 4:1) Rf ) 0.3; IR (film) 2969, 2923, 2859, 1510, 1238, 1013,
a The reference data are indicated in parentheses. Entries marked
with a single asterisk are interchangeable, and those marked with double
asterisks are interchangeable. b Reference 15k. c Reference 15i. d Ref-
erence 15h.
1
818 cm-1; H NMR (300 MHz, CDCl3) δ 1.60 (s, 3H), 1.68 (s, 3H),
1.72 (s, 3H), 2.02-2.18 (m, 4H), 2.28 (s, 3H), 4.51 (d, J ) 5.7 Hz,
2H), 5.09 (t, J ) 6.0 Hz, 1H), 5.49 (t, J ) 6.0 Hz, 1H), 6.82 (d, J )
8.7 Hz, 2H), 7.08 (d, J ) 8.7 Hz, 2H); 13C NMR (75 MHz, CDCl3) δ
16.63, 17.69, 20.46, 25.68, 26.29, 39.54, 64.86, 114.49, 119.66, 123.83,
129.73, 129.80, 131.76, 140.95, 156.70.
Table 4. 1H NMR Assignments (δ [ppm], CDCl3, 500 MHz) for
12-15a
C(n)
12b
13c
Geranyl p-Methoxyphenyl Ether (20 (R1 ) OMe, R2 ) H)):33b,e
TLC (hexane:CH2Cl2 ) 1:1) Rf ) 0.4; mp 36.0 °C; IR (KBr) 2913,
CH3
CH3
CH3
C(20)H3
C(12)HH
0.831, s (0.83, s)
0.838, s (0.84, s)
0.876, s (0.88, s)
1.085, s (1.09, s)
3.824, q, J ) 8.5 Hz
(3.83, q, J ) 8 Hz)
3.915, dt, J ) 4.0,
8.5 Hz (3.92, m)
0.816, s (0.83, s)
0.890, s (0.90, s)
1.099, s (1.10, s)
1.373, s (1.38, s)
3.77, q, J ) 8.5 Hz
(3.77, q, J ) 8.5 Hz)
3.864, m (3.86, m)
1
1514, 1240, 1034, 1009, 826 cm-1; H NMR (300 MHz, CDCl3) δ
1.61 (s, 3H), 1.68 (s, 3H), 1.72 (s, 3H), 2.02-2.18 (m, 4H), 3.77 (s,
3H), 4.49 (d, J ) 6.3 Hz, 2H), 5.10 (t, J ) 5.7 Hz, 1H), 5.49 (t, J )
5.7 Hz, 1H), 6.81-6.87 (m, 4H); 13C NMR (126 MHz, CDCl3) δ 16.62,
17.68, 25.67, 26.29, 39.54, 55.70, 65.46, 114.55, 115.62, 119.75, 123.83,
131.74, 140.92, 153.00, 153.71.
C(12)HH
C(n)
Geranyl 3,5-Xylyl Ether (20 (R1 ) H, R2 ) Me)): TLC (hexane:
CH2Cl2 ) 4:1) Rf ) 0.3; IR (film) 2919, 1613, 1595, 1323, 1293, 1167,
14d
15d
1
1154, 1057, 828 cm-1; H NMR (300 MHz, CDCl3) δ 1.61 (s, 3H),
CH3
CH3
CH3
C(20)H3
0.890, s (0.89)
1.143, s (1.14)
1.155, s (1.155s)
1.177, s (1.18)
0.924, s (0.92, s)
0.960, s (0.96, s)
1.056, s (1.06, s)
1.143, s (1.14, s)
3.801, q, J ) 8.0 Hz
(3.80, q, J ) 7.9 Hz)
3.864, m (3.86, dt,
J ) 3.5, 7.9 Hz)
1.68 (s, 3H), 1.73 (s, 3H), 2.02-2.18 (m, 4H), 2.28 (s, 6H), 4.50 (d, J
) 6.6 Hz, 2H), 5.06-5.14 (m, 1H), 5.49 (t, J ) 5.7 Hz, 1H), 6.55 (s,
2H), 6.59 (s, 1H); 13C NMR (75 MHz, CDCl3) δ 16.61, 17.68, 21.45,
25.67, 26.29, 39.54, 64.58, 112.36, 119.63, 122.31, 123.83, 131.75,
139.07, 140.91, 158.88. Anal. Calcd for C18H26O: C, 83.67; H, 10.14.
Found: C, 83.68; H, 10.12.
C(12)HH 3.812, q, J ) 8.0 Hz
(3.81, q, J ) 8.6 Hz)
C(12)HH 3.885, dt, J ) 3.0, 8.0 Hz
(3.88, dt, J ) 3.6, 8.6 Hz)
Typical Procedure for the Enantioselective Cyclization of o-
Geranylphenol (16),31 o-Nerylphenol (19),31 or Geranyl Aryl Ether
20 Promoted by (R)-4‚SnCl4. The cyclization of 16,31 19,31 or 2032
was carried out according to the above representative procedure under
the conditions shown in eqs 4 and 5 and Table 1. The crude product
was purified by silica gel column chromatography (eluent, hexane:
CH2Cl2 ) 1:0 f 10:1 f 5:1) to give 17 (21) and 18 (22) as a
diastereomeric mixture. Identification of compounds 17 (21) and 18
a The reference data are indicated in parentheses. b Reference 15j.
c Reference 15g. d Reference 15h.
atmosphere was added aryl alcohol (4.0 mmol) portionwise followed
by a catalytic amount of hydroquinone. The mixture was stirred for
0.5 h. Hexamethylphosphoramide (HMPA, 2 mL) and geranyl chloride
(0.74 mL, 4.0 mmol) were successively added. The whole mixture was
stirred for 1 day. After decomposition of excess sodium hydride with
methanol (0.5 mL), the mixture was poured onto ice-water and
extracted with ether. The combined organic layers were dried,
concentrated, and purified by column chromatography on silica gel
(hexane-CH2Cl2 as eluent) to give geranyl aryl ether (ca. 50%).
Physical properties and analytical data of the ethers are given below.
Geranyl Phenyl Ether (20 (R1 ) R2 ) H)):33a,b TLC (hexane:CH2-
1
(22) was effected by comparison of their H NMR data with those of
authentic samples (Table 5).17 The product mixture distribution is
presented in eqs 4 and 5 and Table 1. Retention times (min) are (GC
(γ-TA, 100 kPa, column temperature 130 °C)) tR ) 11.1 (18), 14.9
(25), 15.7 (cis isomer of 26), 23.8 (17), 33.7 ((-)-26), 34.9 ((+)-26)
min.
(33) (a) Goux, C.; Massacret, M.; Lhoste, P.; Sinou, D. Organometallics
1995, 14, 4585. (b) Krimer, M. Z.; Krivoshchekova, O. E.; Lavrinenko, E.
S.; Spektor, V. I.; Simonova, L. L.; Shamshurin, A. A. Zh. Vses. Khim.
O-Va. 1976, 21, 356. (c) Vig, O. P.; Trehan, I. R.; Kad, G. L.; Kumar, A.;
Kumari, S. Indian J. Chem., Sect. B 1983, 22B, 1169. (d) Arnold, Z.;
Kahovcova, J.; Pankova, M.; Svoboda, M.; Tichy, M.; Frantisek, S. Collect.
Czech. Chem. Commun. 1973, 38, 261. (e) Goux, C.; Lhoste, P.; Sinou, D.
Synlett 1992, 725.
Cl2 ) 4:1) Rf ) 0.3; IR (film) 2924, 1601, 1497, 1238, 752, 691 cm-1
;
1H NMR (300 MHz, CDCl3) δ 1.61 (s, 3H), 1.68 (s, 3H), 1.74 (s, 3H),
2.02-2.20 (m, 4H), 4.54 (d, J ) 6.6 Hz, 2H), 5.06-5.14 (m, 1H),
5.50 (tq, J ) 6.6, 1.5 Hz, 1H), 6.89-6.97 (m, 3H), 7.25-7.31 (m,
2H); 13C NMR (75 MHz, CDCl3) δ 16.64, 17.69, 25.69, 26.26, 39.53,
64.67, 114.61, 119.48, 120.52, 123.78, 129.36, 131.79, 141.11, 158.80.

Cyclization of Isoprenoids Using Chiral LBAs
J. Am. Chem. Soc., Vol. 122, No. 34, 2000 8139
Table 5. 1H NMR Assignments (δ [ppm], CDCl3, 300 MHz) for
(CH3)2C and CH3C of 21, 22, 26, and the Cis Isomer of 26
33.37, 39.84, 41.38, 47.75, 118.30, 124.22, 127.07, 129.16, 151.90,
other resonances could not be discerned; LR FAB+-MS m/z 264 ([M]+,
C16H21OCl requires 264.8); [R]27.4D ) -25.4° (c ) 0.33, CHCl3) for a
sample of 65% ee.
cis-2,3,4,4a,9,9a-Hexahydro-7-chloro-1,1,4a-trimethyl-1H-xan-
thene (22 (R1 ) Cl, R2 ) H)): TLC (hexane:CH2Cl2 ) 4:1) Rf )
0.26; GC (â-DM, 50 kPa, column temperature 160 °C) tR ) 26.5 ((()-
enantiomer) min; 1H NMR (300 MHz, CDCl3) δ 0.65 (s, 3H), 0.95 (s,
3H), 1.18 (s, 3H), other resonances could not be discerned.
trans-2,3,4,4a,9,9a-Hexahydro-7-bromo-1,1,4a-trimethyl-1H-xan-
thene (21 (R1 ) Br, R2 ) H)): TLC (hexane:CH2Cl2 ) 4:1) Rf )
0.25; GC (â-DM, 75 kPa, column temperature 160 °C for 5 min and
then warm to 170 °C (+0.1 °C/min)) tR ) 54.4 ((-)-enantiomer), 57.0
R1
R2
21
22
H
H
0.92, 1.01, 1.23
0.89, 1.00, 1.19
0.89, 0.99, 1.19
0.89, 0.99, 1.19
0.90, 0.99, 1.20
0.90, 0.99, 1.20
0.92, 1.01, 1.18
0.65, 0.97, 1.21
0.65, 0.95, 1.26
0.65, 0.95, 1.18
0.63, 0.95, 1.18
0.65, 0.95, 1.19
0.66, 0.95, 1.18
0.63, 0.96, 1.17
F
H
Cl
Br
Me
OMe
H
H
H
H
H
Me
26
cis isomer of 26
0.94, 1.02, 1.28
0.65, 100, 1.27
((+)-enantiomer) min; IR (film) 2936, 1576, 1482, 1248, 812, 656 cm-1
;
1H NMR (300 MHz, CDCl3) δ 0.89 (s, 3H), 0.99 (s, 3H), 1.19 (s, 3H),
1.22-1.70 (m, 5H), 1.92-1.98 (m, 1H), 2.57 (dd, J ) 12.6, 16.2 Hz,
1H), 2.67 (dd, J ) 5.7, 16.2 Hz, 1H), 6.63 (d, J ) 8.4 Hz, 1H), 7.12-
7.20 (m, 2H); 13C NMR (75 MHz, CDCl3) δ 19.70, 19.73, 20.61, 23.15,
32.02, 33.37, 39.82, 41.35, 47.71, 77.63, 111.55, 118.81, 124.84, 129.96,
The physical properties and analytical data of other tricyclic ethers
21 and 22 thus obtained are listed below. Stereochemistries of
1
diastereomers 21 and 22 were assigned by comparison with H NMR
spectra for (CH3)2C(1) and CH3C(4a) of 21 (R1 ) H, R2 ) Me) and
22 (R1 ) H, R2 ) Me) (see Table 3).
132.10, 152.43; [R]24.4 ) -35.2° (c ) 0.6, CHCl3) for a sample of
D
trans-2,3,4,4a,9,9a-Hexahydro-1,1,4a-trimethyl-1H-xanthene (17,
21 (R1 ) R2 ) H)):17 TLC (hexane:CH2Cl2 ) 4:1) Rf ) 0.25; HPLC
(OD-H, hexane, 1.0 mL/min) tR ) 9.3 ((+)-17), 11.5 ((-)-17) min;
GC (â-DM, 50 kPa, column temperature 150 °C) tR ) 28.1 ((-)-17),
29.9 ((+)-17); IR (film) 2934, 1584, 1487, 1456, 1246 cm-1; 1H NMR
(300 MHz, CDCl3) δ 0.92 (s, 3H), 1.01, (s, 3H), 1.23 (s, 3H), 1.25-
1.76 (m, 6H), 1.93-2.01 (m, 1H), 2.61 (dd, J ) 13.0, 16.1 Hz, 1H),
2.72 (dd, J ) 5.6, 16.1 Hz, 1H), 6.75-6.86 (m, 2H), 7.07 (d, J ) 7.2
Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 19.78, 19.84, 20.68, 23.25,
32.10, 33.38, 40.00, 41.49, 48.06, 117.01, 119.60, 122.64, 127.10,
63% ee. Anal. Calcd for C16H21OBr: C, 62.14; H, 6.84; Found: C,
62.21; H, 6.85.
cis-2,3,4,4a,9,9a-Hexahydro-7-bromo-1,1,4a-trimethyl-1H-xan-
thene (22 (R1 ) Br, R2 ) H)): TLC (hexane:CH2Cl2 ) 4:1) Rf )
0.26; GC (â-DM, 75 kPa, column temperature 160 °C for 5 min and
then warm to 170 °C (+0.1 °C/min)) tR ) 27.4 ((()-enantiomer) min;
1H NMR (300 MHz, CDCl3) δ 0.63 (s, 3H), 0.95 (s, 3H), 1.18 (s, 1H),
2.73 (d, J ) 18.0 Hz, 1H), 3.00 (dd, J ) 7.8, 18.0 Hz, 1H), other
resonances could not be discerned.
129.63, 153.25, the resonance of C(4a) could not be discerned; [R]25.2
) -26.4° (c ) 0.25, CHCl3) for sample 17 of 50% ee.
trans-2,3,4,4a,9,9a-Hexahydro-1,1,4a,7-tetramethyl-1H-xan-
thene (21 (R1 ) Me, R2 ) H)): TLC (hexane:CH2Cl2 ) 4:1) Rf )
0.25; GC (γ-TA, 100 kPa, column temp. 140 °C) tR ) 24.9 ((()-enan-
tiomer) min; GC (â-DM, 75 kPa, column temp. 150 °C for 5 min and
then warm to 170 °C (+1 °C/min)) tR ) 22.1 ((-)-enantiomer), 22.8
((+)-enantiomer) min; IR (film) 2932, 1409, 1240, 812 cm-1; 1H NMR
(300 MHz, CDCl3) δ 0.90 (s, 3H), 0.99 (s, 3H), 1.20 (s, 3H), 1.25-
1.37 (m, 1H), 1.44-1.68 (m, 4H), 1.68 (dd, J ) 5.4, 12.9 Hz, 1H),
1.91-1.98 (m, 1H), 2.25 (s, 3H), 2.56 (dd, J ) 12.0, 15.0 Hz, 1H),
2.66 (dd, J ) 5.7, 15.0 Hz, 1H), 6.66 (d, J ) 9.0 Hz, 2H), 6.87 (s,
1H), 6.88 (d, J ) 9.0 Hz, 1H); 13C NMR (75 MHz, CDCl3) δ 19.74,
19.78, 20.49, 20.66, 23.19, 32.09, 33.35, 39.98, 41.48, 48.11, 116.72,
122.27, 127.74, 128.69, 129.99, 150.92, the resonance of C(4a) could
D
cis-2,3,4,4a,9,9a-Hexahydro-1,1,4a-trimethyl-1H-xanthene (18, 22
(R1 ) R2 ) H)):17b TLC (hexane:CH2Cl2 ) 4:1) Rf ) 0.26; HPLC
(OD-H, hexane, 1.0 mL/min) tR ) 5.6 (major enantiomer), 6.4 (minor
enantiomer) min; 1H NMR (300 MHz, CDCl3) δ 0.65 (s, 3H), 0.97 (s,
3H), 1.21 (s, 3H), 1.20-2.20 (m, 7H), 2.73-2.80 (m, 1H), 3.04 (dd, J
) 8.0, 18.0 Hz, 1H), 6.72-6.85 (m, 2H), 7.05 (d, J ) 6.9 Hz, 2H);
13C NMR (75 MHz, CDCl3) δ 18.06, 21.37, 23.56, 26.99, 32.24, 33.94,
39.53, 41.60, 44.37, 75.20, 117.05, 119.76, 122.02, 126.63, 28.89,
154.45.
trans-2,3,4,4a,9,9a-Hexahydro-7-fluoro-1,1,4a-trimethyl-1H-xan-
thene (21 (R1 ) F, R2 ) H)): TLC (hexane:CH2Cl2 ) 4:1) Rf ) 0.25;
GC (γ-TA, 100 kPa, column temperature 130 °C) tR ) 26.8 ((()-
enantiomer) min; GC (â-DM, 75 kPa, column temperature 120 °C for
5 min and then warm to 150 °C (+0.5 °C/min)) tR ) 52.0 ((-)-
enantiomer), 55.8 ((+)-enantiomer) min; IR (film) 2936, 1439, 1242,
not be discerned; [R]25.3 ) -33.9° (c ) 0.5, CHCl3) for a sample of
D
62% ee. Anal. Calcd for C17H24O: C, 83.55; H, 9.90; Found: C, 83.53;
H, 9.93.
cis-2,3,4,4a,9,9a-Hexahydro-1,1,4a,7-tetramethyl-1H-xanthene (22
(R1 ) Me, R2 ) H)): TLC (hexane:CH2Cl2 ) 4:1) Rf ) 0.26; GC
(γ-TA, 100 kPa, column temperature 140 °C) tR ) 11.4 ((()-
enantiomer) min; 1H NMR (300 MHz, CDCl3) δ 0.65 (s, 3H), 0.95 (s,
3H), 1.19 (s, 1H), 2.24 (s, 3H), 2.71 (d, J ) 18.0 Hz, 1H), 2.99 (dd, J
) 9.0, 18.0 Hz, 1H), 6.63 (d, J ) 8.7 Hz, 1H), other resonances could
not be discerned.
1
1221, 1102, 810 cm-1; H NMR (300 MHz, CDCl3) δ 0.89 (s, 3H),
1.00 (s, 3H), 1.19 (s, 3H), 1.22-1.70 (m, 5H), 1.92-1.98 (m, 1H),
1.92-1.98 (m, 1H), 2.58 (dd, J ) 22.8, 12.9 Hz, 1H), 2.68 (dd, J )
16.8, 5.7 Hz, 1H), 6.64-6.71 (m, 1H), 6.73-6.79 (m, 2H); 13C NMR
(75 MHz, CDCl3) δ 19.66, 19.73, 20.60, 23.49, 32.03, 33.34, 39.88,
41.42, 47.79, 113.76 (J ) 22.7 Hz), 115.33 (J ) 21.2 Hz), 117.69 (J
) 8.0 Hz), 123.73, 149.21, 156.44 (J ) 235.5 Hz), the resonance of
C(4a) could not be discerned; 19F NMR (282 MHz, CDCl3) δ -124.55;
trans-2,3,4,4a,9,9a-Hexahydro-7-methoxy-1,1,4a-trimethyl-1H-
xanthene (21 (R1 ) MeO, R2 ) H)): TLC (hexane:CH2Cl2 ) 1:1) Rf
) 0.37; GC (γ-TA, 100 kPa, column temperature 160 °C for 5 min
and then warm to 180 °C (+1 °C/min)) tR ) 18.7 ((()-enantiomer)
min; GC (â-DM, 70 kPa, column temperature 150 °C) tR ) 72.8
((-)-enantiomer), 76.0 ((+)-enantiomer) min; IR (film) 2934, 1497,
1227 cm-1; 1H NMR (300 MHz, CDCl3) δ 0.90 (s, 3H), 0.99 (s, 3H),
1.20 (s, 3H), 12.4-1.38 (m, 1H), 1.44-1.68 (m, 4H), 1.70 (dd, J )
5.9, 12.8 Hz, 1H), 1.91-1.97 (m, 1H), 2.58 (dd, J ) 12.9, 16.2 Hz,
1H), 2.69 (dd, J ) 5.7, 16.2 Hz, 1H), 3.75 (s, 3H), 6.60-6.66 (m,
2H), 6.68 (s, 1H); 13C NMR (75 MHz, CDCl3) δ 19.67, 19.76, 20.61,
23.60, 32.07, 33.32, 39.93, 41.47, 48.05, 55.66, 113.11, 114.12, 117.46,
123.18, 147.19, 152.83, the resonance of C(4a) could not be discerned;
LR FAB+-MS m/z 248 ([M]+, C16H21OF requires 248.3); [R]24.4
-22.44° (c ) 0.27, CHCl3) for a sample of 58% ee.
)
D
cis-2,3,4,4a,9,9a-Hexahydro-7-fluoro-1,1,4a-trimethyl-1H-xan-
thene (22 (R1 ) F, R2 ) H)): TLC (hexane:CH2Cl2 ) 4:1) Rf ) 0.26;
GC (γ-TA, 100 kPa, column temperature 130 °C) tR ) 13.1 ((()-
enantiomer) min; 1H NMR (300 MHz, CDCl3) δ 0.65 (s, 3H), 0.95 (s,
3H), 1.26 (s, 1H), other resonances could not be discerned; 19F NMR
(282 MHz, CDCl3) δ -123.42.
trans-2,3,4,4a,9,9a-Hexahydro-7-chloro-1,1,4a-trimethyl-1H-xan-
thene (21 (R1 ) Cl, R2 ) H)): TLC (hexane:CH2Cl2 ) 4:1) Rf )
0.25; GC (â-DM, 50 kPa, column temperature 160 °C) tR ) 55.0 ((-
)-enantiomer), 58.8 ((+)-enantiomer) min; IR (film) 2936, 1482, 1458,
[R]23.3 ) -33.2° (c ) 0.38, CHCl3) for a sample of 70% ee. Anal.
D
1
1250, 1102, 1042, 812 cm-1; H NMR (300 MHz, CDCl3) δ 0.89 (s,
Calcd for C17H24O2: C, 78.42; H, 9.29; Found: C, 78.45; H, 9.30.
cis-2,3,4,4a,9,9a-Hexahydro-7-methoxy-1,1,4a-trimethyl-1H-xan-
thene (22 (R1 ) MeO, R2 ) H)): TLC (hexane:CH2Cl2 ) 1:1) Rf )
0.39; GC (γ-TA, 100 kPa, column temperature 160 °C for 5 min and
then warm to 180 °C (+1 °C/min)) tR ) 10.8 ((()-enantiomer) min;
3H), 0.99 (s, 3H), 1.19 (s, 3H), 1.22-1.70 (m, 5H), 1.66 (dd, J ) 5.4,
12.9 Hz, 1H), 1.92-1.98 (m, 1H), 2.56 (dd, J ) 12.9, 16.5 Hz, 1H),
2.67 (dd, J ) 5.4, 16.5 Hz, 1H), 6.68 (d, J ) 8.1 Hz, 1H), 6.91-7.05
(m, 2H); 13C NMR (75 MHz, CDCl3) δ 19.72, 20.60, 23.21, 32.02,

8140 J. Am. Chem. Soc., Vol. 122, No. 34, 2000
Nakamura et al.
1H NMR (300 MHz, CDCl3) δ 0.66 (s, 3H), 0.95 (s, 3H), 1.18 (s, 3H),
3.00 (dd, J ) 8.0, 17.9 Hz, 1H), 3.75 (s, 3H), other resonances could
not be discerned; 13C NMR (75 MHz, CDCl3) δ 18.07, 20.64, 23.96,
26.84, 32.25, 33.94, 39.56, 41.61, 44.40, 55.67, 112.50, 113.48, 117.46,
other resonances could not be discerned.
preparative HPLC (LC-908 (Japan Analytical Industry Co., Ltd.):
column, JAIGEL-1H+JAIGEL-2H; eluent, CHCl3): mp 65.2-65.8 °C;
1H NMR (300 MHz, CDCl3) δ 1.61 (s, 6H), 1.69 (s, 3H), 1.73 (s, 3H),
1.95-2.20 (m, 8H), 4.49 (d, J ) 6.6 Hz, 2H), 5.02 (s, 2H), 5.06-5.16
(m, 2H), 5.50 (dt, J ) 1.2, 6.5 Hz, 1H), 6.86 (d, J ) 9.4 Hz, 2H), 6.91
(d, J ) 9.4 Hz, 2H), 7.28-7.48 (m, 5H). Anal. Calcd for C28H36O2:
C, 83.12; H, 8.97. Found: C, 83.14; H, 8.93.
trans-2,3,4,4a,9,9a-Hexahydro-1,1,4a,6,8-pentamethyl-1H-xan-
thene (21 (R1 ) H, R2 ) Me)): TLC (hexane:CH2Cl2 ) 4:1) Rf )
0.25; GC (γ-TA, 100 kPa, column temperature 150 °C) tR ) 23.6 ((()-
enantiomer) min; GC (â-DA, 80 kPa, column temperature 150 °C for
3 min and then warm to 170 °C (+0.2 °C/min)) tR ) 43.9 ((-)-
enantiomer), 45.5 ((+)-enantiomer) min; IR (film) 2932, 1580, 1321,
1304, 1103 cm-1; 1H NMR (300 MHz, CDCl3) δ 0.92 (s, 3H), 1.01 (s,
3H), 1.18 (s, 3H), 1.25-1.38 (m, 1H), 1.45-1.65 (m, 4H), 1.67 (dd, J
) 5.3, 13.1 Hz, 1H), 1.88-2.02 (m, 1H), 2.20 (s, 3H), 2.23 (s, 3H),
2.29 (dd, J ) 13.5, 16.4 Hz, 1H), 2.57 (dd, J ) 5.0, 16.4 Hz, 1H),
6.47 (s, 1H), 6.55 (s, 1H); 13C NMR (75 MHz, CDCl3) δ 19.10, 19.62,
19.76, 20.62, 20.79, 20.97, 32.08, 33.39, 39.93, 41.58, 48.13, 76.31,
115.16, 118.24, 122.116, 136.36, 137.04, 152.91; [R]23.2D ) -20.6° (c
) 0.36, CHCl3) for a sample of 41% ee. Anal. Calcd for C18H26O: C,
83.67; H, 10.14. Found: C, 83.65; H, 10.17.
cis-2,3,4,4a,9,9a-Hexahydro-1,1,4a,6,8-pentamethyl-1H-xan-
thene (22 (R1 ) H, R2 ) Me)): TLC (hexane:CH2Cl2 ) 4:1) Rf )
0.26; GC (γ-TA, 100 kPa, column temperature 150 °C) tR ) 10.6 ((()-
enantiomer) min; 1H NMR (CDCl3) δ 0.63 (s, 3H), 0.96 (s, 3H), 1.17
(s, 3H), 2.19 (s, 3H), 2.23 (s, 3H), 2.60 (d, J ) 16.5 Hz, 1H), 2.72 (dd,
J ) 6.0, 16.5 Hz, 1H), 6.45 (s, 1H), 6.54 (s, 1H), other resonances
could not be discerned; 13C NMR (65 MHz, CDCl3) δ 18.09, 21.42,
23.57, 26.94, 32.27, 33.96, 39.55, 41.67, 44.49, 114.53, 116.74, 127.33,
129.82, 152.20, other resonances could not be discerned.
Enantioselective Cyclization of 33 Promoted by (S)-3‚SnCl4.11e
The cyclization of 33 was carried out according to the above repre-
sentative procedure under the conditions shown in eq 6. The crude
product produced in the reaction of 33 was partially purified by silica
gel column chromatography (eluent, hexanes:EtOAc ) 20:1) to remove
highly polar compounds and a monoisopropyl ether of (S)-BINOL.11e
Then, the benzyl group of the product was cleaved by stirring for 12
h at room temperature in the presence of a catalytic amount of Pd/C in
ethanol under a hydrogen atmosphere. The crude product of 31 was
transformed to the corresponding acetate 32 by using acetic anhydride
and pyridine in dichloromethane. Finally, acetate 32 was purified by
silica gel chromatography (eluent, hexanes:EtOAc ) 30:1 to 20:1).
Identification of compounds 31 and 32 was effected by comparison of
their 1H NMR spectra with those of authentic samples.17b,34 The enan-
tiomeric excess of 32 was determined by HPLC (OD-H, hexane:i-PrOH
) 200:1, 1.0 mL/min): tR ) 16.9 ((+)-32) and 26.5 ((-)-32) min.
(-)-Chromazonarol acetate (32):22 1H NMR (500 MHz, CDCl3)
δ 0.85-1.80 (m, 11H), 0.87 (s, 3H), 0.90 (s, 3H), 0.93 (s, 3H), 1.22
(s, 3H), 2.08 (dt, J ) 13.0, 3.5 Hz, 1H), 2.29 (s, 3H), 2.628 (d, J )
10.5 Hz, 1H), 2.631 (d, J ) 7.0 Hz, 1H), 6.72-6.81 (m, 3H); [R]30.4
D
) -45.55° (c ) 0.06, CHCl3) for >99% ee of (-)-32 [(-)-32 was
isolated by recycling preparative HPLC (Tosoh Co., Ltd.; column, OD
(Daicel); eluent, hexane:i-PrOH ) 200:1)].
3-Phenoxymethyl-2,4,4-trimethylcyclohexene (25): TLC (hexane:
CH2Cl2 ) 4:1) Rf ) 0.25; GC (â-DM, 50 kPa, column temperature
150 °C) tR ) 19.9 ((-)-25), 20.7 ((+)-25) min; IR (film) 2917, 1601,
Computational Methods
The geometries of all the structures have been optimized using the
hybrid B3LYP24 as implemented in the Gaussian9835 package on SGI
Indigo2 Impact R10000. The optimized structure of biphenol-Bz‚SnCl4
was verified by vibrational frequency analysis at the same level. We
used a LANL2DZ basis set for our studies because heavy metal atoms
are well expressed using the ECP-type basis sets. Diffuse functions
and polarization functions on O and Cl (RO,diffuse ) 0.0483, RO,polarization
) 0.85, RCl,diffuse ) 0.0845, RCl,polarization ) 0.60) and polarization
functions on Sn (RSn,polarization ) 0.183) were used.36 What conditions
are suited to the tasks was considered carefully in the case of
enantioselective protonation using chiral LBAs.11h And all charges were
calculated using the NBO program.23
1
1586, 1497, 1474, 1456, 1242, 752, 691 cm-1; H NMR (300 MHz,
CDCl3) δ 0.92 (s, 3H), 1.01, (s, 3H), 1.23 (s, 3H), 1.25-1.76 (m, 6H),
1.93-2.01 (m, 1H), 2.61 (dd, J ) 13.0, 16.1 Hz, 1H), 2.72 (dd, J )
5.6, 16.1 Hz, 1H), 6.77 (d, J ) 8.6 Hz, 1H), 6.75-6.86 (m, 2H), 7.07
(d, J ) 7.2 Hz, 2H); 13C NMR (75 MHz, CDCl3) δ 22.99, 23.13, 27.07,
27.44, 31.88, 32.29, 49.35, 67.85, 114.52, 120.41, 123.00, 129.35,
132.77, 158.89; LR FAB+-MS m/z 230 ([M]+, C16H22O requires 230.3);
[R]25.2 ) -20.1° (c ) 0.30, CHCl3) for sample 25 of 53% ee.
D
trans-6a,7,8,9,10,10a-Hexahydro-7,7,10a-trimethyl-6H-Dibenzo-
[b,d]pyran (26): TLC (hexane:CH2Cl2 ) 4:1) Rf ) 0.28; IR (film)
2928, 1732, 1489, 1449, 1293, 1285, 1221, 1042, 750 cm-1; 1H NMR
(300 MHz, CDCl3) δ 0.94 (s, 3H), 1.02, (s, 3H), 1.28 (s, 3H), 1.23-
1.85 (m, 6H), 2.20-2.30 (m, 1H), 2.61 (dd, J ) 13.0, 16.1 Hz, 1H),
2.72 (dd, J ) 5.6, 16.1 Hz, 1H), 6.72-6.88 (m, 2H), 7.03-7.17 (m,
2H); 13C NMR (75 MHz, CDCl3) δ 18.93, 21.85, 24.51, 32.95, 37.22,
41.86, 47.81, 64.05, 116.28, 119.81, 124.27, 126.97, 155.58, other
resonances could not be discerned; LR FAB+-MS m/z 230 ([M]+,
Acknowledgment. We thank Mr. Shoichi Kondo for deter-
mining the crystal structure of compound 21 (R1 ) H, R2 )
Me). We also acknowledge Dr. Hiroaki Wasada and Dr. Yuko
Wasada for helpful discussion on the DFT calculations.
Supporting Information Available: X-ray data for 21 (R1
) H, R2 ) Me) and optimized geometry in Z-matrix form
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
C16H22O requires 230.3); [R]22.8 ) -20.1° (c ) 0.10, CHCl3) for
D
sample 26 of 43% ee.
cis-6a,7,8,9,10,10a-Hexahydro-7,7,10a-trimethyl-6H-Dibenzo[b,d]-
pyran (cis isomer of 26): TLC (hexane:CH2Cl2 ) 4:1) Rf ) 0.28; 1H
NMR (300 MHz, CDCl3) δ 0.65 (s, 3H), 1.02, (s, 3H), 1.27 (s, 3H);
13C NMR (75 MHz, CDCl3) δ 18.94, 22.57, 32.69, 33.43, 35.96, 37.33,
42.75, 48.34, 63.00, 116.02, 119.76, 124.33, 126.91, other resonances
could not be discerned.
JA001165A
(34) Woodside, A. B.; Huang, Z.; Poulter, C. D. Organic Syntheses;
Wiley: New York, 1993; Collect. Vol. VIII, p 616.
(35) Gaussian 98 (Revision A.6): Frisch, M. J.; Trucks, G. W.; Schlegel,
H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, J. A.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam,
J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith,
T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.;
Challacombe, M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wong, M.
W.; Andres, J. L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A., Gaussian,
Inc., Pittsburgh, PA, 1998.
Preparation of 1-Benzyloxy-4-((E,E)-farnesyloxy)benzene (33).
33 was prepared from farnesyl chloride and monobenzylhydroquinone
according to a slight modification of the literature procedure.31 To a
stirred suspension of sodium hydride (60% in oil, 7 mmol) in THF (20
mL) at room temperature under argon was added 4-benzyloxyphenol
(6 mmol). The mixture was stirred for 30 min. HMPA (2 mL) and
farnesyl chloride (7 mmol) were successively added. After being stirred
at room temperature for 1 h, the whole mixture was warmed to reflux
for 6 h. After decomposition of excess sodium hydride with methanol
(1 mL), the mixture was poured onto ice-water and extracted with
ether. The combined organic layers were dried, concentrated, and
purified by column chromatography on silica gel (eluent, hexane:CH2-
Cl2 ) 5:1 to hexane:EtOAc ) 20:1) to give 33 as a 9:1 2E-2Z mixture
of white solid (ca. 50%). Finally, 33 was isolated by recycling
(36) (a) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284. (b)
Huzinaga, S.; Andzelm, J.; Klobukowski, M.; Radzio-Andzelm, E.; Sakai,
Y.; Tatewaki, H. Physical Sciences Data 16: Gaussian Basis Sets for
Molecular Calculations; Elsevier: Amsterdam, 1984; p 23.