Synthesis of chiral chromans by the Pd-catalyzed asymmetric allylic alkylation (AAA): Scope, mechanism, and applications

Article abstract of DOI:10.1021/ja048078t

The Pd-catalyzed asymmetric allylic alkylation (AAA) of phenol allyl carbonates serves as an efficient strategy to construct the allylic C-O bond allowing access to chiral chromans in up to 98% ee. The effect of pH and the influence of olefin geometry, as well as substitution pattern on the ee and the absolute configuration of the chiral chromans were explored in detail. These observations suggest a mechanism involving the cyclization of the more reactive π-allyl palladium diastereomeric intermediate as the enantiodiscriminating step (Curtin-Hammett conditions). This methodology led to the enantioselective synthesis of the vitamin E core, the first enantioselective total synthesis of (+)-clusifoliol and (-)-siccanin, and the synthesis of an advanced intermediate toward (+)-rhododaurichromanic acid A.

Full text of DOI:10.1021/ja048078t

Published on Web 09/03/2004
Synthesis of Chiral Chromans by the Pd-Catalyzed
Asymmetric Allylic Alkylation (AAA): Scope, Mechanism, and
Applications
Barry M. Trost,* Hong C. Shen, Li Dong, Jean-Philippe Surivet, and
Catherine Sylvain
Contribution from the Department of Chemistry, Stanford UniVersity,
Stanford, California 94305-5080
Received April 2, 2004; E-mail: bmtrost@stanford.edu
Abstract: The Pd-catalyzed asymmetric allylic alkylation (AAA) of phenol allyl carbonates serves as an
efficient strategy to construct the allylic C-O bond allowing access to chiral chromans in up to 98% ee.
The effect of pH and the influence of olefin geometry, as well as substitution pattern on the ee and the
absolute configuration of the chiral chromans were explored in detail. These observations suggest a
mechanism involving the cyclization of the more reactive π-allyl palladium diastereomeric intermediate as
the enantiodiscriminating step (Curtin-Hammett conditions). This methodology led to the enantioselective
synthesis of the vitamin E core, the first enantioselective total synthesis of (+)-clusifoliol and (-)-siccanin,
and the synthesis of an advanced intermediate toward (+)-rhododaurichromanic acid A.
Introduction
rhododaurichromanic acid A (8), and siccanin (10). Therefore,
it is important to develop general methodologies10 to construct
Chiral chromans or closely related chromens are a class of
compounds possessing significant biological properties (Figure
1). For example, vitamin E (1)1 and its analogues trolox (2)2
and MDL-73404 (3)3 are important lipophilic antioxidants.
MDL-73404 (3) also exhibits cardioprotective effects during a
myocardial infarction. Conocurvone (4), a naphthoquinone
trimer isolated from a Conospermum sp. in 1993,4 prevents the
cytopathic effects and replication of HIV in human T-lympho-
blastic cells (CEM-SS). Centchroman (5), an estrogen antago-
nist, is an antifertility agent.5 Clusifoliol (6) was a component
isolated from Peperomia species which have been used to treat
malignant tumors.6 Daurichromenic acid (7) and rhododau-
richromanic acid A (8) have also shown potent anti-HIV
activity.7 Nebivolol (9) is an anti-hypertensive agent.8 Siccanin
(10) is a potent antifungal agent and used clinically in Asia.9
Prior to our work, there were no asymmetric total syntheses of
centchroman (5), clusifoliol (6), rhododaurichromenic acid (7),
chiral chromans or chromens, which could potentially lead to
natural products of this general type, and/or their analogues with
more desirable pharmacological features.
The Pd-catalyzed asymmetric allylic alkylations (AAA) are
powerful tools in organic synthesis.11 Our initial studies of the
Pd-catalyzed AAA of phenols in an intermolecular fashion led
to the enantioselective formation of the allylic C-O bond, which
was followed by cyclization to provide chiral chromans (Scheme
1).12 However, this strategy has two aspects yet to improve.
First, this reaction generally gave a small amount of regioiso-
mers, resulting from the competitive attack at the primary allylic
terminus. Furthermore, the ee of this reaction is generally lower
than 90%. To address these issues, Pd-catalyzed intramolecular
asymmetric allylic alkylation of phenol allyl carbonates could
potentially afford chiral chromans in one step (Scheme 2)
(10) For approaches leading to chiral chromans or chromenes, see: (a)
Chandrasekhar, S.; Reddy, M. V. Tetrahedron, 2000, 56, 6339-6344. (b)
Kingsbury, J. S.; Harrity, J. P. A.; Bonitatebus, P J., Jr.; Hoveyda, A. H.
J. Am. Chem. Soc. 1999, 121, 791-799. (c) Cohen, N.; Lopresti, R. J.;
Saucy, G. J. Am. Chem. Soc. 1979, 101, 6710-6716. (d) Mizuguchi, E.;
Suzuki, T.; Achiwa, K. Synlett 1994, 929-930. (e) Mizuguchi, E.; Suzuki,
T.; Achiwa, K. Synlett 1996, 743-744. (f) Mizuguchi, E.; Achiwa, K.
Synlett 1995, 1255-1256. (g) Tietze, L. F.; Gorlitzer, J. Synthesis 1997,
877-878. (h) Solladie, G.; Moine, G. J. Am. Chem. Soc. 1984, 106, 6097-
6098. (i) Harada, T.; Hayashiya, T.; Wada, I.; Iwa-ake, N.; Oku, A. J. Am.
Chem. Soc. 1987, 109, 527-532. (j) Gassman, P. G.; Amick, D. R. J. Am.
Chem. Soc. 1978, 100, 7611-7619. (k) Inoue, S.; Ikeda, H.; Sato, S.; Horie,
K.; Ota, T.; Miyamoto, O.; Sato, K. J. Org. Chem. 1987, 52, 5495-5497.
(l) Hubscher, v. J.; Barner, R. HelV. Chim. Acta 1990, 73, 1068-1086.
(m) Kitamura, M.; Suga, S.; Kawai, K.; Noyori, R. J. Am. Chem. Soc.
1986, 101, 6071-6072. (n) Uozumi, Y.; Kato, K.; Hayashi, T. J. Am. Chem.
Soc. 1997, 119, 5063-5064.
(1) Vitamin E; Machlin, L. J., Ed.; Marcel Dekker: New York, 1980.
(2) Terao, K.; Niki, E. J. Free Rad. Biol. Med. 1986, 2, 193-201.
(3) Grisar, J. M.; Petty, M. A.; Bolkenius, F. N.; Dow, J.; Wagner, J.; Wagner,
E. R.; Haegele, K. D.; Jong, W. D. J. Med. Chem. 1991, 34, 257-260.
(4) (a) Decosterd, L. A.; Parsons, I. C.; Gustafson, K. R.; Cardellina II, J. H.;
McMahon, J. B.; Cragg, G. M.; Murata, Y.; Pannell, L. K.; Steiner, J. R.;
Clardy, J.; Boyd, M. R. J. Am. Chem.. Soc. 1993, 115, 6673-6679. (b)
Laatsch, H. Angew. Chem., Int. Ed. Engl. 1994, 33, 422-424.
(5) Sankaran, M. S.; Prasad, M. R. N. Contraception 1974, 9, 279-289.
(6) (a) Seeram, N. P.; Jacobs, H.; McLean, S.; Reynolds, W. F. Phytochemistry
1998, 49, 1389-1391. (b) Tanaka, T.; Asai, F.; Iinuma, M. Phytochemistry
1998, 49, 229-232.
(7) Kashiwada, Y.; Yamazaki, K.; Ikeshiro, Y.; Yamagishi, T.; Fujioka, T.;
Mihashi, K.; Mizuki, K.; Cosentino, L. M.; Fowke, K.; Morris-Natschke,
S. L.; Lee, K.-H. Tetrahedron 2001, 57, 1559-1563.
(11) (a) Trost, B. M.; Crawley, M. L. Chem. ReV. 2003, 103, 2921-2943. (b)
Trost, B. M. Chem. Pharm. Bull. 2002, 50, 1-14. (c) Helmchen, G. Pfaltz,
A. Acc. Chem. Res.2000, 33, 336-345 (d) Jacobsen, E. N.; Pfaltz, A.;
Yamamoto, H. “Comprehensive Asymmetric Catalysis”, Vol. II, pp 833-
884, Springer-Verlag: Berlin Heidelberg, 1999.
(8) (a) De Cree, J.; Geukens, H.; Leempoels, J.; Verhaegen, H. Drug DeV.
Res. 1986, 8, 109-117. (b) Van de Water, A.; Janssen, W.; Van Nueten,
J.; Xhonneux, R.; De Cree, J.; Verhaegen, H.; Reneman, R. S.; Janssen, P.
A. J. J. CardioVasc. Pharmacol. 1988, 11, 552-563.
(9) Isabashi, K. J. Antibiot. Ser. A 1962, 15, 161-167.
(12) Trost, B. M.; Toste, D. F. J. Am. Chem. Soc. 1998, 120, 9074-9075.
9
11966
J. AM. CHEM. SOC. 2004, 126, 11966-11983
10.1021/ja048078t CCC: $27.50 © 2004 American Chemical Society

Synthesis of Chiral Chromans
A R T I C L E S
Figure 1. Representative chiral chromans and chromens.
Scheme 1. Pd-Catalyzed Intermolecular Allylic Alkylation of
Phenols To Access Chiral Chromans
of the enantiotopic faces of the substrate during the ionization
event. Sinou’s group also screened a considerable number of
chiral phosphine ligands for this transformation. They obtained
up to 50% ee when they employed a chiral monophosphine
ligand (NMDPP) (eq 2).14 It should be noted that only 7% ee
of the desired chroman ent-17b was obtained employing the
Trost ligand 16c and substrate 15b (eq 3). Different from the
Achiwa’s proposal, the Sinou group presumed that the nucleo-
philic attack on the π-allyl palladium intermediates was the
enantiodiscriminating step.
without the complication of regioselectivity. In this reaction, a
chiral Pd complex catalyzes the ionization of the leaving group
such as carbonate to form π-allyl palladium intermediate 12,
which could be attacked by phenol intramolecularly to form
chiral chroman 13. A subsequent oxidation of chroman 13 will
lead to chiral chromen 14.
Prior to our investigation, two groups have examined this
approach to chiral vinylchromans. After screening a large variety
of chiral phosphine ligands and reaction conditions, Achiwa and
co-workers showed that optically active vinylchromans such as
17a could be obtained with moderate enantiomeric excess (54%
ee) using BFFPA ligand (eq 1).13 Furthermore, they proposed
that the asymmetric induction derives from the chiral recognition
Despite the low ee of the Pd-catalyzed AAA reaction
employing the Trost ligand 16c reported by Sinou’s group, our
group started investigating this reaction in the context of the
asymmetric synthesis of the core of vitamin E (1). We were
pleased to discover that trans-allyl carbonate 15a could undergo
a Pd-catalyzed intramolecular cyclization to generate chiral vinyl
chroman 17a in 96% yield and 84% ee (eq 4).15 It should be
noted that a tetrasubstituted stereogenic center is constructed
in this reaction. Since then, we have explored the scope and
studied the factors that dictate the enantioselectivity of this
reaction, such as the choice of chiral ligand and additive, the
olefin geometry of the starting material, and reaction condi-
Scheme 2. Projected Synthesis of Chiral Chromans and Chromens via Pd-Catalyzed AAA Reactions
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11967

A R T I C L E S
Trost et al.
prepared on a multigram scale.14,15 The key step in this synthe-
tic sequence is the olefination of ketone 22. Three dif-
ferent conditions have been examined (a-c). The Horner-
Wardsworth-Emmons reaction using n-butyllithium as a base
afforded allylic ester 23 with a better E to Z ratio (11:1) than
the case in which sodium hydride was used (7:1). Still’s protocol
using trifluoroethyl phosphonoacetate17 gave a 4.6 to 1 ratio of
allylic esters still favoring the E isomer 23. Fortunately, trans
allylic alcohol 24 can be crystallized, whereas the cis allylic
alcohol is an oil. The separation of the trans and cis allylic
alcohols is crucial for a good enantioselectivity because the trans
and cis allyl carbonates will give opposite enantiomers using
the same chiral ligand, as revealed in the later experiments.
To have access to either trans or cis trisubstituted allyl
carbonates without the complication of isolation of the geo-
metrical isomers, a different route was developed involving a
Negishi coupling18 of alkyl iodide 29 and vinyl iodide 3019 or
3220 as a key step (Schemes 4 and 5). The Negishi coupling
strategy is also applied to the preparation of substrates 15d,
15d′, 15e, and 15e′. It should be noted that the mono-
demethylation of 33 was achieved using lithium propylthiolate
in HMPA. To our surprise, the conditions using sodium ethyl
thiolate in DMF afforded allyl ethyl sulfide. Under these thermal
basic conditions, allyl alcohol and DMF presumably generated
allyl formate, which was displaced by ethyl thiolate to give allyl
sulfide.
tions.16 In light of these results, a mechanistic working model
is proposed to accommodate all of our observations.
The syntheses of trans-disubstituted allyl carbonate 15g and
cis-disubstituted allyl carbonate 15g′ are also straightforward
(Schemes 6 and 7).21 This strategy is utilized to prepare all trans-
and cis-disubstituted allyl carbonates shown in Table 6.
Pd-Catalyzed Asymmetric Allylic Alkylations (AAA). In
the comprehensive studies of conditions for the Pd-catalyzed
(13) Mizuguchi, E.; Achiwa, K. Chem. Pharm. Bull. 1997, 45, 1209-1211.
(14) Labrosse, J.-R.; Poncet, C.; Lhoste, P.; Sinou, D. Tetrahedron: Asymmetry
1999, 10, 1069-1078.
(15) Trost, B. M.; Asakawa, N. Synthesis 1999, 1491-1494.
(16) Trost, B. M.; Shen, H. C.; Dong, L.; Surivet, J.-P. J. Am. Chem. Soc. 2003,
125, 9276-9277.
(17) Still, W. C.; Gennart, C. Tetrahedron Lett. 1983, 24, 4405-4408.
(18) Negishi, E.; Liou, S.-Y.; Xu, C.; Huo, S. Org. Lett. 2002, 4, 261-264.
(19) Wang, Q.; Wei, H.; Schlosser, M. Eur. J. Org. Chem. 1947, 12, 3263-
3268.
(20) Queneau, Y.; Krol, W. J.; Bornmann, W. G.; Danishefsky, S. J. Bull. Soc.
Chim. Fr. 1993, 130, 358-370.
(21) Chandrasekhar, S.; Reddy, M. V. Tetrahedron 2000, 56, 6339-6344.
Synthesis of Substrates. Following the protocol depicted in
Scheme 3 , trans trisubstituted allyl carbonate 15c can be easily
9
11968 J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004

Synthesis of Chiral Chromans
A R T I C L E S
Scheme 3. First Generation Synthesis of Trans Trisubstituted Allyl Carbonate 15ca
a
Reagents and conditions: (a) i. (MeO)2SO2, K2CO3, acetone; ii. NaSC2H5, DMF; iii. TsOH, DHP, THF. (b) nC4H9Li, TMEDA, THF then DMF. (c)
Ph3PdCHC(O)CH3, PhCH3 (d) i. NaBH4, CH3OH; ii. H2 (1 atm), 5% PtO2, EtOAc, iii. TPAP, NMO, 4 Å MS, CH2Cl2. (e) (MeO)2P(O)CH2CO2Me,
nC4H9Li, THF, E/Z 11:1. (f) (CF3CH2O)2P(O) CH2CO2CH3, KN(TMS)2, THF, E:Z 4.6:1. (g) (MeO)2P(O) CH2CO2CH3, NaH, THF, E:Z 7:1. (h) i. DIBAL-
H, THF, -78°; ii. TsOH, MeOH. iii. MeOCOCl, pyridine, CH2Cl2.
Table 1. Effects of Ligands and Additives on the Pd-Catalyzed
AAA Reaction of 15c
Table 3. Influence of Solvents on the Pd-Catalyzed AAA Reaction
of 15ca
entry Pd (%)
ligand (%)
additvie (equiv)
[C] (M) time (h) yield (%) ee (%)a
ligand
entry
Pd (%)
16c (%)
solvent
yield (%)
ee (%)b
1
2
3
4
5
6
7
8
9
10
11
12
13
14
4
4
4
4
2
1
4
4
4
4
4
4
4
4
16c (6)
16c (6)
16c (6)
16c (6)
16c (3)
None
Et3N (1)
AcOH (1)
AcOH (1)
AcOH (1)
0.2
0.2
0.01
0.2
0.3
0.5
0.2
0.2
0.2
1
1
1.3
1
1
3
1
1
1
95
96
99
94
96
73
95
98
99
99
40
24
99
95
-14
-22
75
84
74
72
73
75
-5
1
2
3
4
5
6
7
4
4
4
4
4
4
4
6
6
6
6
6
6
6
DMSO
THF
>95
12
87
81
94
58
43
62
78
84
70
52
Toluene
ClCH2CH2Cl
CH2Cl2
CH3CN
HOAc
16c (1.5) AcOH (1)
16c (6)
16c (6)
16c (6)
16c (6)
16c (6)
16c (6)
16d (6)
16e (6)
AcOH (0.5)
PhCO2H (1)
H3PO4 (0.5)
nBu4NOAc (1) 0.2
nBu4NCl (1)b
0.2
nBu4NCl (0.3) 0.2
72
10
a The reaction of eq 5 using substrate 15c was performed with
(dba)3Pd2‚CHCl3, ligand 16c and 1 equiv of HOAc, at 0.2 M for 1 h in the
stated solvent. b Measured by chiral GC.
1
66
1
63
1
33
AcOH (1)
AcOH (1)
0.2
0.2
1
1
69
64
Table 4. Effect of Temperature on the Pd-Catalyzed AAA
Reaction of 15ca
a Measured by chiral GC. b With 1 equiv of HOAc.
entry
T (°
C)
time (h)
yield (%)
ee (%)b
1
2
0
rt
40
45
1
1
1
1
58
94
92
95
72
84
85
82
Table 2. Effects of Additives on the Pd-Catalyzed AAA Reactions
of Substrate 15k
3
ligand
4c
entry
Pd (%)
16c (%)
additive (equiv)
[C] (M)
yield (%)
ee (%)a
1
2
3
4
4
4
2
2
4
4
4
2
2
6
6
6
3
3
6
6
3
3
3
none
Et3N (1)
0.2
0.1
0.2
0.1
0.1
0.2
0.2
0.2
0.1
0.1
>95
>95
>95
>95
>95
>95
>95
>95
>95
>95
-16
-21
76
76
75
75
75
75
76
a The reaction was performed as in Table 3 using compound 16c as
ligand, methylene chloride as solvent at the stated temperature. b Measured
by chiral GC. c In sealed tube.
PhCO2H (1)
PhCO2H (1)
PhCO2H (1)
AcOH (1)
AcOH (0.5)
AcOH (1)
AcOH (0.5)
AcOH (1)
4
5b
6
Equation 5 and Table 1 show our studies on the influence of
ligand and additive on the yield and ee of the reaction. Without
any additive and in the presence of 2 mol % Pd2dba3 chloroform
complex and 6 mol % ligand 16c, substrate 15c gave 95% yield
but only 14% ee favoring the formation of S chroman (ent-
17c) (entry 1). The addition of 1 equiv of triethylamine slightly
improved the ee and still favored the formation of S chroman.
Interestingly, the ee was dramatically enhanced when 1 equiv
of acetic acid was added (entries 3-7). Moreover, the absolute
configuration of chroman was reversed to R. After the optimiza-
tion of substrate concentration and catalyst loading, R chroman
17c was obtained in 94% yield and 84% ee (entry 4). To explore
the influence of different acids, we have also screened benzoic
acid which gave similar results (entry 8), and phosphoric acid
7
8
9
10
75
a Measured by chiral GC. b Performed at 40 °C.
asymmetric allylic alkylation, it was found that the best reaction
conditions were identical for both di- and trisubstituted trans
allyl carbonates among all the examined cases. Taking a
trisubstituted allyl carbonate 15c as an example, the effects of
additive, concentration, catalyst loading, ligands, leaving group,
solvent, and temperature were systematically studied and the
results are summarized in Tables 1-4.
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11969

A R T I C L E S
Trost et al.
Table 5. Pd-Catalyzed AAA Reactions of Trans and Cis Trisubstituted Allyl Carbonates
a The ee’s of products were determined by chiral GC unless otherwise noted. b The ee’s of products were determined by chiral HPLC.
which gave lower ee’s but still good yield (entry 9). Entry 10
clearly demonstrates an acetate ion effect on improving the ee
compared to the additive free case (entry 1). Meanwhile, it
suggests that the acid may have a synergistic effect on the
improvement of ee. In entry 12, tetrabutylammonium chloride
was employed to speed up Pd π-σ-π equilibration. Although
the ee was low, it was higher than entry 1, and the reaction
favored the formation of the R enantiomer. We also observed
that the standard ligand 16c (R, R) was superior to 16d or 16e,
despite the similar yields obtained for all ligands in the presence
of acetic acid (entries 13 and 14).
To probe whether a better leaving group could improve the
reaction outcome, Troc group was employed as shown in eq 5
and Table 2. It is again noted that substrate 15k favored the
formation of S chroman ent-17c in the presence of triethylamine
or in the absence of any additive (entries 1 and 2). Reactions
with acetic or benzoic acid as additive gave similar yield and
ee compared to the carbonate substrate 15c.
We have also screened a number of solvents (Table 3).
Dichloromethane and 1,2-dichloroethane are most suitable for
high yields and higher ee’s. The temperature profile shown in
Table 4 shows that room temperature to 45 °C gave better ee’s
and yields than lower temperature such as 0 °C.
The optimized reaction conditions were employed for various
tri- and disubsituted allyl carbonates (Tables 5 and 6).22 The
impact of the olefin geometry of the substrates on the reaction
was also studied. Interestingly, we observed that trans- and cis
trisubstituted allyl carbonates 15c and 15c′ both led to R
chroman 17c using ligands of opposite chirality (entries 1 and
6, Table 5). Furthermore, the cis-allyl carbonate 15c′ afforded
chiral vinyl chroman 17c in substantially higher ee (97%) than
the trans-counterpart 15c (84% ee). Similar observations were
also made for substrates 15d, 15d′, 15e, and 15e′ (entries 2, 7,
3 and 8, Table 5).
Trans disubstituted allyl carbonates, 15g, 15b, 15h, and 15i
in the presence of (R, R)-ligand 16c, and their cis disubstituted
counterparts 15g′, 15b′, 15h′, and 15i′ in the presence of (S,
S)-ligand, favored the formation of the same chroman enanti-
omer, namely, 17g′, 17b′, 17h′, and 17i′ respectively (Table
6). More interestingly, the trans series gave much higher ee than
the cis counterparts, which is opposite to the trend we observed
for the trisubstituted allyl carbonates. Moreover, the para
substituents seem to have little effect over the ee’s of the reaction
for the trans series. In contrast, the substituent effect on the cis
disubstituted cases on the ee of the reaction is much more
pronounced.
Determination of Absolute Configuration of Vinyl Chro-
mans. The absolute configuration of these compounds was
established by comparing with known compounds from the
literature (such as 17a and 17b),13 or derivatization (vide infra,
such as 17c and 17f). The absolute configuration of other
chroman products was determined by comparing the sign of
the optical rotation, the retention time order on chiral GC or
chiral HPLC for both enantiomers with that of 17a, 17b, 17c,
and 17f, whose absolute configuration and chiral HPLC/GC
retention time order for both enantiomers have been established.
The method of assigning absolute configuration of an R-oxy-
R,R-disubstituted acetic acid employing phenylglycine methyl
9
11970 J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004

Synthesis of Chiral Chromans
A R T I C L E S
Table 6. Pd-Catalyzed AAA Reactions of Trans and Cis Disubstituted Allyl Carbonates
a The ee’s of products were determined by chiral HPLC.
palladium intermediates do not interconvert fast enough prior
to cyclization, i.e., the kinetically formed π-allyl palladium
complex was immediately captured by nucleophile. In this
scenario, the ionization leading to the more reactive π-allyl
palladium intermediate is the enantiodiscriminating step. The
second plausible mechanism envisions that the interconversion
of diastereomeric Pd π-allyl intermediates is fast relative to the
nucleophilic attack, and the asymmetric induction is determined
by the reaction rate of the two diastereomeric intermediates
(Curtin-Hammett conditions). The mechanistic picture is more
complex if the rate of interconversion and nucleophilic attack
is comparable.
Figure 2. ∆δ values (δ(S) - δ(R)) for 10-(S)-PGME amide 43 and 10-(R)-
PGME amide 44.
ester (PGME) was developed by Kusumi and Yabuuchi.23
Applying this method, we were able to assign the absolute
configuration of chroman 17c and 17f. For example, chiral
chroman 17c was converted to aldehyde 41 via a sequence of
dihydroxylation and oxidative cleavage reactions. The resulting
aldehyde 41 was further oxidized to carboxylic acid 42,24 which
was then coupled with (S)-PGME and (R)-PGME, respectively,
to afford two diasteromeric amides 43 and 44 (Scheme 8). The
∆δ values (δ(S) - δ(R)) were calculated by subtracting the proton
chemical shifts of 10-(S)-PGME amide 43 with the chemical
shifts of the corresponding protons of 10-(R)-PGME amide 44.
The comparison with the described model allowed us to
determine that chroman 17c has the R configuration (Figure
2).
The addition of 1.5 equiv of triethylamine led to a dramatic
drop of both ee (8%) and yield (34%) (entry 2, Table 1). If the
ionization of leaving group is the enantiodiscriminating step,
the presence of base should speed up the cyclization so that the
full π-σ-π equilibration does not occur. Therefore, the ee
should increase. The observation that ee decreases suggests that
the enantiodiscriminating step is most likely not the ionization
of carbonate, as proposed by the Achiwa group.13
Our studies summarized in Tables 5 and 6 provide further
evidence to support the second mechanism (Curtin-Hammett
conditions). Three important observations help to elucidate the
mechanistic picture. First, both trans- and the corresponding cis-
allyl carbonates lead to the same chiral chroman as the major
enantiomer using ligands of opposite chirality. Second, trisub-
stituted cis-allyl carbonates (15c′, 15d′, and 15e′) afford chiral
chromans in substantially higher ee than their trans-counterparts
(15c, 15d, and 15e). Third, disubstituted trans-allyl carbonates
(such as 15g, 15b, 15h, and 15i) afford chiral vinyl chromans
Mechanistic Discussion
There are at least two possible mechanisms for the catalytic
asymmetric allylic alkylation (AAA) of phenol allyl carbonates.11b
The first mechanism assumes that the diastereomeric π-allyl
(22) Compound 15a and 15f were prepared according to Achiwa’s procedure.
See ref 13.
(23) Yabuuchi, T.; Kusumi, T. J. Org. Chem. 2000, 65, 397-404.
(24) Bal, B. S.; Childers, W. E., Jr.; Pinnick, H. W. Tetrahedron 1981, 37, 2091-
2096.
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11971

A R T I C L E S
Trost et al.
Scheme 4. Synthesis of Cis Trisubstituted Allyl Carbonate 15c′
Scheme 5. Second Generation of Trans Trisubstituted Allyl Carbonate 15c
Scheme 6. Synthesis of Trans Disubstituted Allyl Carbonate 15g
in substantially higher ee’s than the cis-counterpart (such as
15g′, 15b′, 15h′, and 15i′).
The first observation can be rationalized by the mechanistic
proposal shown in Schemes 9 and 10. On the basis of our
working model,25 the (R, R)-ligand 16c can be exemplified by
the simplified cartoon picture. The flaps and walls of the chiral
scaffold represent the phenyl groups of the triarylphospine
moiety. The trans-allyl carbonate (15c, 15d, or 15e) generates
syn-π-allyl palladium complex 46 kinetically via a “matched”
ionization (“matched” indicates that the leaving group departs
(25) Trost, B. M.; Toste, F. D. J. Am. Chem. Soc. 1999, 121, 4545-4554.
9
11972 J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004

Synthesis of Chiral Chromans
A R T I C L E S
Scheme 7. Synthesis of Cis Disubstituted Allyl Carbonate 15g′
Scheme 8. Preparation of Diastereomeric Amides 43 and 44
under the flap of the chiral pockets (Scheme 9). To the extent
cyclization occurs at this stage, the product is S enantiomer ent-
17c, ent-17d, or ent-17e. Intermediate 46 invokes a disfavored
steric interaction between the large substituent bearing the
phenol nucleophile and the wall of the chiral scaffold, i.e., the
nucleophilic attack would occur from the more hindered left
front quadrant. Therefore, the cyclization of intermediate 46 is
a “mismatched” event. The equilibration of intermediate 46 leads
to 47 via a π-σ-π process through C1. Intermediate 47 is
considerably less encumbered because the large substituent is
under the right front flap and therefore better accommodated.
Furthermore, the nucleophilic attack from under the flap is not
blocked, in contrast to the cyclization of intermediate 46.
Therefore, the cyclization of the less sterically encumbered
intermediate 47 is a “matched” event, in which 17c, 17d, or
17e is obtained as the major enantiomer. The observed ee then
depends on the rate difference of cyclization from the two
diastereomeric complexes, 46 and 47, compared to the rate of
their interconversions. To the extent that a “mismatched”
ionization occurs then 47 forms directly and should lead to high
ee in its cyclization.
at C1 to form 50. Apparently, the nucleophilic attack involving
intermediate 49 is less favored due to the steric interaction
between the right front wall and the incoming nucleophile
(mismatched cyclization). Therefore, intermediate 50 is more
reactive toward cyclization under the left front flap to form R
chroman 17c, 17d, or 17e (matched cyclization).
To address the second observation that cis trisubstituted allyl
carbonates (15c′, 15d′, and 15e′) gave substantially higher ee’s
than the trans counterparts (15c, 15d, and 15e), we may consider
the reactivity difference of intermediate 46 and 47 vs 49 and
50. While the steric stress between the large arylethyl moiety
and the wall in 46 destabilizes it and enhances its propensity to
undergo π-σ-π isomerization to 47, that same interaction
favors a conformation of this moiety that facilitates cyclization.
Thus, cyclization versus π-σ-π becomes somewhat competi-
tive. Since cyclization via 46 vs 47 gives mirror image products,
the net result is a diminished ee. On the other hand, the arylethyl
moiety is nicely accommodated in 49 in the extended conforma-
tion which cannot cyclize. Adopting the required conformation
for cyclization requires this moiety to encounter a severe steric
interaction with a “wall” of the ligand. Isomerization to
diastereomeric complex 50 places this moiety into a conforma-
tion favorable to cyclization as well as bringing the nucleophile
toward the sterically less congested flap-therefore enhancing
the rate of cyclization. The decreased rate of cyclization for
diastereomer 49 and the increased rate for 50 then lead to a
The Curtin-Hammett situation also applies to the case of
cis-trisubstituted allyl carbonates 15c′, 15d′, and 15e′ (Scheme
10). The “matched” ionization of carbonate 15c′, 15d′, or 15e′
led to anti-π-allyl palladium intermediate 49 as the kinetic
product which could subsequently undergo a π-σ-π process
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11973

A R T I C L E S
Trost et al.
Scheme 9. Working Model for the Pd-Catalyzed Intramolecular AAA Reaction of Trans Trisubstituted Allyl Carbonates
Scheme 10. Working Model for the Pd-Catalyzed Intramolecular AAA Reaction of Cis Trisubstituted Allyl Carbonates
larger rate difference and a higher ee. In general, cis allyl
carbonates 15c′, 15d′, and 15e′ all gave higher ee’s than their
trans counterparts.
Interestingly, the Pd-catalyzed intramolecular AAA reaction
of cis disubstituted allyl carbonates gave much lower ee than
their trans counterparts (Table 6), as opposed to what we
observed for the trisubstituted substrates in Table 5. As shown
in Scheme 11, the “matched” ionization of trans disubstituted
allyl carbonate 15b kinetically generates syn-π-allyl palladium
complex 51, which can undergo π-σ-π isomerization at C1
to form 52. The nucleophile then preferentially approaches
π-allyl intermediate 52 under the flap of the front right quadrant
(matched cyclization), which is sterically more favorable than
intermediate 51 (mismatched cyclization). Consequently, the
formation of the R enantiomer 17b is favored. It is known that
syn-π-allylpalladium complexes are more stable than the cor-
9
11974 J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004

Synthesis of Chiral Chromans
A R T I C L E S
Scheme 11. Working Model for the Pd-Catalyzed AAA Reaction of Trans Disubstituted Allyl Carbonate 15b
responding anti complexes.26 Therefore, the conversion of syn-
π-allyl complexes 51 and 52 to anti-π-allyl complexes 53 and
54 at C3 is thermodynamically disfavored. Therefore, the
mechanistic picture for trans disubstituted allyl carbonates is
essentially the same as that of the trans trisubstituted allyl
carbonates, which accounts for similar ee’s for the trans tri-
and disubstituted allyl carbonate series.
In contrast to the disubstituted cases, the π-σ-π equilibration
at C3 of trisubstituted π-allyl palladium complex is disfavored.28
Thus, the syn-anti interconversion of the substituted terminus
does not occur as already outlined in Schemes 9 and 10, only
two of the four possible diastereomeric complexes need to be
considered and can nicely account for the selectivity differences.
Therefore, the enantioselectivity of the Pd-catalyzed intramo-
lecular AAA reaction of phenols is presumably governed by
the π-σ-π equilibration of diastereomeric Pd complexes
(Curtin-Hammett conditions). The kinetically generated π-al-
lylpalladium intermediate via the “matched” ionization is less
reactive, and will isomerize to the more reactive intermediate
via the π-σ-π process. For trisubstituted substrates, such an
isomerization can only occur at one allyl terminus; however,
for disubstituted substrates, the isomerization can occur at both
allyl termini.
Conversely, the “matched” ionization of cis allyl carbonate
15b′ leads to anti-π-allylpalladium complex 55, which could
isomerize to 56 through π-σ-π at C1 (Scheme 12). Steric
factors leads to preferential nucleophilic attack via intermediate
56 over 55 to give the R isomer 17b. Meanwhile, the conversion
of anti-π-allylpalladium complex 55 to the syn complex 57 is
thermodynamically favored, and 57 can undergo a “matched”
cyclization leading to the S isomer ent-17b. Such a π-σ-π
isomerization at a secondary carbon is still kinetically accessible
albeit slower than at a primary carbon. On the other hand,
although syn-π-allylpalladium complex 58 can be formed, either
via π-σ-π isomerization of anti-π-allylpalladium complex 56,
or via such an isomerization of syn-π-allylpalladium complex
57 at C1, 58 could only undergo a “mismatched” cyclization to
the R enantiomer 17b. Therefore, the likely pathway leading to
17b is 15b′ to 55 to 56 to 17b. The facile π-σ-π process
from 55 to the thermodynamically more favored syn complex
5727 followed by a “matched” cyclization leading to ent-17b
becomes a competitive pathway with respect to the pathway
leading to 17b with the net effect of diminished ee’s.
Finally, the unusual acetic acid effect to enhance both the
yield and ee of the Pd-catalyzed AAA reaction may have wide
implications for various reactions.29 The effect of pH alone does
not seem to account for all the improvement of ee, as suggested
by the fact that the use of phosphoric acid as an additive led to
low ee. As demonstrated by entry 10 of Table 1, tetraalkylam-
monium acetate improved the yield and ee of the reaction
(27) (a) Trost, B. M.; Verhoeven, T. R. J. Org. Chem. 1976, 41, 3215-3216.
(b) Ohta, T.; Hosokawa, T.; Murahashi, S. I.; Miki, K.; Kasai, N.
Organometallics 1985, 4, 2080-2085.
(28) (a) Faller, J. W.; Incorvia, M. J. Organomet. Chem. 1969, 19, P13-P14.
(b) Faller, J. W.; Mattina, M. J. Inorg. Chem. 1972, 11, 1296-1297. (c)
Trost, B. M.; Keinan, E. J. Org. Chem. 1979, 44, 3452-457. (d) Trost, B.
M.; Keinan, E. Tetrahedron Lett. 1980, 21, 2591-2594.
(26) Faller, J. W.; Thomsen, M. E.; Mattina, M. J. J. Am. Chem. Soc. 1971, 93,
2642-2643.
(29) For another example of acetic acid effect on the Pd-catalyzed asymmetric
reaction, see: Trost, B. M.; Fandrick, D. R. 2003, 125, 11 836-11 837.
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11975

A R T I C L E S
Trost et al.
compared to the additive-free case. Thus, the acetate ion seems
to play a critical role in improving the enantioselectivity. The
addition of tetrabutylammonium chloride (entry 12 of Table 1)
was employed to speed up the π-σ-π equilibration of π-allyl
palladium complex, and it did give higher ee than entry 1, albeit
still considerably lower than in the case of acetate. In light of
these results, the acetate or chloride ion may help facilitate the
π-σ-π process of π-allylpalladium intermediates as required.
The potential bidentate nature of acetate may account for its
enhanced effectiveness compared to chloride. The acidity of
acetic acid may synergistically contribute to the high enanti-
oselectivity by slowing down the cyclization, thus allow the
π-allylpalladium intermediates to reach full equilibrium prior
to cyclization. The protonation of the methoxide formed from
the ionization of methyl carbonate by acetic acid, can avoid
the deprotonation of phenol to give phenoxide, a much more
reactive nucleophile for cyclization than neutral phenol.
temperature was -35 °C, then the E:Z selectivity and yield both
dropped (eq 8).
Synthesis of Clusifoliol and Des-Carboxy Daurichromenic
Acid A MOM Ether. To demonstrate the utility of this
methodology, we chose vitamin E,15 nebivolol, (+)-clusifoliol,
(+)-rhododaurichromanic acid A, and (-)-siccanin as targets.
Prelimimary reports describe a part of our work on the first
enantioselective total syntheses of (+)-clusifoliol16 and (-)-
siccanin.30 We report the details of the synthesis of (+)-
clusifoliol and herein a synthesis of descarboxydaurichromenic
acid A MOM ether 59, a possible intermediate toward the total
synthesis of daurichromenic acid (7), a precursor of (+)-
rhododaurichromanic acid A (8) (Scheme 13).
Attempts to ortho metalate the methyl ether and alkylate
failed.35 Following a protocol developed by Dauben’s group,36
the sequential demethylation, O-prenylation,37 and prenyl group
rearrangement proceeded smoothly to yield the desired (+)-
clusifoliol 6 (Scheme 15). The specific optical rotation, chiral
HPLC eluting time, and all other spectroscopic data of our
synthetic sample matched that of the authentic sample kindly
provided by Dr. McLean and Dr. Jacobs. Since we have
established the absolute configuration of 17c, this synthesis
allows assignment of the absolute configuration of 6 as depicted.
Rhododaurichromanic acid A (8) and daurichromenic acid
(7) were both isolated from the leaves of Rhododendron
dauricum by Kashiwada’s group in 2001.38 Daurichromenic acid
demonstrated potent anti-HIV activity with an EC50 value of
0.00567 µg/mL and therapeutic index (TI) of 3710; whereas,
rhododaurichromanic acid showed relatively potent anti-HIV
activity with an EC50 value of 0.37 µg/mL and TI of 92.
The retrosynthetic analysis of rhododaurichromanic acid A
(8) and daurichromenic acid (7) can be easily envisioned to be
derived from chiral chroman 17c, as shown in Scheme 16.39
The key step involves a C-C bond formation between vinyl
iodide 63 and vinyl chroman 17c or its derivatives.
Clusifoliol 6 was isolated from Peperomia clussifolia. Its
structure has been established as 3,4-dihydro-2,7-dimethyl-6-
(3-methyl-2-butenyl)-2-(4-methyl-1,3-pentadienyl)-2H-1-ben-
zopyran-5-ol by spectroscopic methods. In one of the papers,
the authors claimed no optical rotation6b whereas, in the other
one, a significant value (+160° in EtOH) was reported.6a
As shown in Scheme 14, clusifoliol 6 can be derived from a
Julia olefination31 and a subsequent prenylation of chroman 17c,
by the Pd-catalyzed asymmetric allylic alkylation with excellent
ee and yield. Chiral chroman 17c can be easily converted to
aldehyde 41 via a dihydroxylation followed by an oxidative
cleavage (Scheme 8). The one-pot protocol of dihydroxylation/
oxidative cleavage32 only gave trace amount of desired aldehyde
accompanied with decomposition of the starting material. The
resulting aldehyde 41 then reacted with phenyl prenyl sulfone
to give the trans olefin in good yield and selectivity (9:1 to
11:1) via a Julia olefination protocol, but three steps are required
(eq 6). The modified Julia protocol using phenyltetrazole prenyl
sulfone was also examined to shorten the steps. However, a
1:1 mixture of E/Z isomers was obtained when KHMDS was
used as the base.33 As previously reported,34 a more polar
medium with lithium as the counterion favors the desired trans
diene which was obtained in 98% yield and an excellent E/Z
ratio (20:1) initiating the reaction at -78 °C. If the initial
Various types of cross-coupling methods failed to unite the
two building blocks, the chroman core 17c with the side chain
63. An alternative strategy involves the addition of side chain
to chromen aldehyde 64. Envisioning that a more easily
cleavable protecting group for the phenol would be needed, the
(35) (a) Bates, R. W.; Gabel, C. J.; Ji, J.; Rama-Devi, T. Tetrahedron, 1995,
51, 8199-8212. (b) Trost, B. M.; Saulnier, M. G. Tetrahedron Lett. 1985,
26, 123-126.
(30) Trost, B. M.; Shen, H. C.; Surivet, J. P. Angew. Chem., Int. Ed. 2003, 42,
3943-3947.
(36) Dauben, W. G.; Cogen, J. M.; Behar, V. Tetrahedron Lett. 1990, 31, 3241-
3244.
(31) (a) Julia, M.; Righini-Tapie, A.; Verpeaux, J.-N. Tetrahedron, 1983, 39,
3283-3287. (b) Trost, B. M.; Arndt, H. C.; Strege, P. E.; Verhoeven, T.
R. Tetrahedron. Lett. 1976, 17, 3477-3478.
(37) Bu, X.; Zhao, L.; Li, Y. Synthesis, 1997, 11, 1246-1248.
(38) Kashiwada, Y.; Yamazaki, K.; Ikeshiro, Y.; Yamagishi, T.; Fujioka, T.;
Mihashi, K.; Mizuki, K.; Cosentino, L. M.; Fowke, K.; Morris-Natschke,
S. L.; Lee, K.-H. Tetrahedron 2001, 57, 1559-1563.
(39) For the racemic total syntheses of rhododaurichromanic acids A and B,
see: (a) Kurdyumov, A. V.; Hsung, R. P.; Ihlen, K.; Wang, J. Org. Lett.
2003, 5, 3935-3938. (b) Kang, Y.; Mei, Y.; Du, Y.; Jin, Z. Org. Lett.
2003, 5, 4481-4484.
(32) Demuth, M.; Ritterskamp, P.; Weigt, E.; Schaffner, K. J. Am. Chem. Soc.
1986, 108, 4149-4154.
(33) Baudin, J. B.; Hareau, G.; Julia, S. H.; Lorne, R.; O. Bull. Soc. Chim. Fr.
1993, 130, 356-357 and reference therein.
(34) Liu, P.; Jacobsen, E. N. J. Am. Chem. Soc. 2001, 123, 10 772-10 773.
9
11976 J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004

Synthesis of Chiral Chromans
A R T I C L E S
Scheme 12. Working Model for the Pd-Catalyzed AAA Reaction of Cis Disubstituted Allyl Carbonate 15b′
methyl ether was exchanged for MOM as shown in Scheme 17
steps a and b. The latter, 66, was hydroborated to give a 3:1
regioisomeric mixture of alcohols, the major being the primary
alcohol 67. The MOM group may serve later as an ortho-
directing group for metalation to introduce the carboxylate group
to the aromatic ring. Interestingly, when alcohol 67 was
submitted to DDQ oxidation, the reaction cleanly gave a tricyclic
compound 71 (eq 9). Presumably the benzylic cation formed
The synthetic efforts toward the total synthesis of rhododau-
richromanic acid A demonstrated the utility of the Pd-catalyzed
AAA reaction of phenol allyl carbonates to form chiral
chromans, and the subsequent oxidation leading to chiral
chromens. A highly Pd-catalyzed regioselective allylic deoxy-
genation was developed to introduce the side chain to this natural
product.
Experimental Section
General Procedure for Chroman Formation. To a degassed
mixture of Pd2(dba)3‚CHCl3 (2 mol %) and (R, R) chiral ligand 16c (6
mol %) or ent-16c was added dichloromethane. The solution was stirred
for 10 min under argon to yield a yellow solution. To this solution
was added acetic acid (1-1.2 equiv). After 5 min, a solution of
carbonate in dichloromethane was added. The reaction mixture was
stirred at room temperature for 1-10 h. The volatiles were removed
under reduced pressure and the residue was purified directly over silica
gel eluting with 1:11 of diethyl ether in petroleum ether to afford the
chiral chroman.
in situ can be intramolecularly trapped by alcohol, which
completely ruled out the potentially competitive elimination to
afford the corresponding chromen. This observation demands
the protection of the primary alcohol with TBDPS prior to DDQ
oxidation. The resulting chroman silyl ether 68 underwent DDQ
oxidation smoothly to give chromen 69. The removal of the
silyl group to alcohol 70 followed by oxidation using TPAP
afforded chromenic aldehyde 64 (path b).
The subsequent vinyllithium addition to aldehyde 64 yielded
allylic alcohol 72 as a mixture of diastereomers, both of which
were converted to carbonate. The highly Pd-catalyzed regiose-
lective reduction (2.5 mol % of π-allyl palladium chloride dimer,
10 mol % of phosphite 74 and 2 equiv of bulky L-selectride) of
both diastereomers of carbonate 73 gave cleanly the desired
product 62, without scrambling or isomerization of the double
bond.
(R)-6-Benzyloxy-2,5,7,8-tetramethyl-2-vinyl-chroman (17a): A
colorless oil. [R]D ) +49.4 (c 1.20, CH2Cl2) (HPLC: Chiralpak OD,
heptane/2-propanol); IR (neat): 1454, 1411, 1370, 1254, 1088 cm-1
;
1H NMR (400 MHz, CDCl3): δ 7.70-7.33 (m, 5H), 5.86 (dd, J )
17.2, 10.8 Hz, 1H), 5.13 (dd, J ) 17.2, 1.2 Hz, 1H), 5.03 (dd, J )
10.8, 1.2 Hz, 1H), 4.69 (s, 2H), 2.65-2.45 (m, 2H), 2.23 (s, 3H), 2.17
(s, 3H), 2.15 (s, 3H), 1.93 (m, 1H), 1.83 (m, 1H), 1.40 (s, 1H); 13C
NMR (50 MHz, CDCl3): δ 148.4, 147.9, 141.9, 138.1, 128.5, 128.0,
127.8, 127.7, 125.9, 122.6, 117.7, 113.3, 75.5, 74.6, 31.7, 26.9, 20.8,
12.8, 11.9, 11.8; Anal. Calcd for C12H14O: C, 81.95, H, 8.13. Found:
C, 81.86; H, 8.26.
(R)-2-Vinylchromane (17b): a colorless oil. [R]D ) +80.3 (c 0.27,
CH2Cl2, 84% ee) (HPLC: Chiralpak OJ, 100% heptane); IR (neat):
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11977

A R T I C L E S
Trost et al.
Scheme 13. Targets for the Application of Pd-Catalyzed AAA Reaction Leading to Chiral Chromans
Scheme 14. Retrosynthetic Analysis of the Total Synthesis of Clusifoliol 6
Scheme 15. Completion of Synthesis of Clusifoliol
1
1581, 1488, 1458, 1233 cm-1; H NMR (300 MHz, CDCl3): δ 1.86
(m, 1 H), 2.08 (m, 1 H), 2.86-2.73 (m, 2 H), 4.55 (m, 1 H), 5.24
(ddd, J ) 10.5, 1.2, 1.2 Hz, 1 H), 5.39 (ddd, J ) 17.4, 1.5, 1.5 Hz, 1
H), 6.00 (ddd, J ) 17.4, 10.5, 5.4 Hz, 1 H), 6.84 (m, 2 H), 7.09 (m, 2
H); 13C NMR (75 MHz, CDCl3): δ 152.1, 137.5, 129.5, 127.3, 121.7,
120.1, 116.8, 116.2, 76.1, 27.4, 24.1, 20.4; HRMS Calcd for C11H12O
[M]+: 160.0888; Found: 160.0888.
(R)-5-Methoxy-2,7-dimethyl-2-vinyl-chroman (17c): [R] ) +54.0
(c 2.18, CHCl3), 97% ee; IR (neat): 2936, 1618, 1586, 1139 cm-1; 1H
NMR (300 MHz, CDCl3): δ 6.40 (s, 1H), 6.25 (s, 1H), 5.85 (dd, J )
9
11978 J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004

Synthesis of Chiral Chromans
A R T I C L E S
Scheme 16. Retrosynthetic Analysis of Rhododaurichromanic Acid A (8) and Daurichromenic Acid (7)
Scheme 17. Synthesis of Chromen Aldehyde 64a
a Reagents and conditions: (a) C2H5SH, NaH, DMF, 120 °C. (b) n-C4H9Li, MOM-Cl, THF, 0 °C to room temperature. (c) BH3‚THF, THF, 0 °C to room
temperature then H2O2, NaOH, rt to 50 °C. (d) TBDPS-Cl, DMAP, CH2Cl2, rt. (e) DDQ, PhH, reflux. (f) TBAF, THF, 0 °C. (g) TPAP, NMO, 4 Å MS,
CH2Cl2, rt.
Scheme 18. Synthesis of Chromene MOM Ether 62
10.7, 17.3 Hz, 1H), 5.20 (dd, J ) 1.2, 17.3 Hz, 1H), 5.07 (dd, J ) 1.2,
10.7 Hz, 1H), 3.81 (s, 3H), 2.68 (td, J ) 5.6, 16.8 Hz, 1H), 2.47 (ddd,
J ) 6.1, 9.8, 16.8 Hz, 1H), 2.31 (s, 3H), 1.93 (dd, J ) 4.8, 5.8, 13.4
Hz, 1H), 1.79 (ddd, J ) 5.6, 9.8, 13.4 Hz, 1H), 1.43 (s, 3H); 13C NMR
(75 MHz, CDCl3): δ 157.4, 154.3, 141.3, 136.9, 113.6, 110.0, 107.2,
102.6,76.1, 55.2, 31.0, 26.7, 21.6, 16.7. HRMS: Calcd for C14H18O2
[M+]: 218.1307. Found: 218.1303. Chromatographic separation (GC,
Cyclosil B): T oven ) 160 °C, tR (S, minor) ) 18.5 min, tR (R, major)
) 18.9 min.
(R)-2-Methyl-2-vinylchroman (17d): a colorless oil. [R]D ) +48.1
(c 0.55, CDCl3, 95% ee); IR (neat): 2932w, 2850m, 1649w, 1494s,
1434m, 1259m, 1218s, 1198m, 1139m, 1045m, 999m, 930m, 872w,
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11979

A R T I C L E S
Trost et al.
847w, 810m, 792w cm-1; 1H NMR (500 MHz, CDCl3): δ 7.10 (t, J )
7.5 Hz, 1H), 7.02 (d, J ) 7.5 Hz, 1H), 6.86 (d, J ) 7.5 Hz, 1H), 6.82
(td, J ) 1.0, 7.5 Hz, 1H), 5.85 (dd, J ) 10.5, 17.0 Hz, 1H), 5.18 (dd,
J ) 1.0, 17.5 Hz, 1H), 5.06 (dd, J ) 1.0, 11.0 Hz, 1H), 2.71 (m, 2H),
1.93 (m, 1H), 1.83 (m, 1H), 1.39 (s, 3H); 13C NMR (125 MHz,
CDCl3): δ 154.0, 141.2, 129.3, 127.3, 121.3, 119.7, 116.8, 113.9, 76.6,
31.7, 29.1, 27.1; Anal. Calcd for C12H14O: C, 82.72, H, 8.10. Found:
C, 82.50; H, 8.29. Chromatographic separation (GC, Cyclosil B): T
oven ) 120 °C, tR(S) ) 21.6 min, tR (R) ) 22.7 min.
(m, 1 H), 2.05 (m, 1 H), 2.26 (s, 3 H), 2.79-2.70 (m, 2 H), 3.84 (s, 3
H), 4.62 (m, 1 H), 5.21 (ddd, J ) 13.0, 2.0, 2.0 Hz, 1 H), 5.35 (ddd,
J ) 21.5, 1.5, 1.5 Hz, 1 H), 6.01 (m, 1 H), 6.74 (m, 2 H), 6.47 (s, 1
H), 6.53 (s, 1 H); 13C NMR (125 MHz, CDCl3): δ 148.0, 141.4, 137.3,
128.9, 122.0, 121.4, 116.4, 110.1, 76.2, 55.8, 27.0, 23.6, 20.0; HRMS
Calcd for C13H16O2 [M]+: 204.1150; Found: 204.1045.
2-(2-Iodoethyl)-1,3-dimethoxy-5-methylbenzene (29). To alcohol
28 (35 mg, 0.18 mmol) in THF (2 mL) at 0 °C were added
triphenylphosphine (56 mg, 0.21 mmol), imidazole (16 mg, 0.24 mmol)
and iodine (59 mg, 0.23 mmol). The mixture was warmed to room
temperature and stirred for 1 h. The mixture was washed with sodium
thiosulfate solution (1 mL), and the aqueous layer was extracted with
diethyl ether (2 × 15 mL). The combined organic layers were
concentrated and chromatographed eluting with 2% diethyl ether in
petroleum ether to afford iodide 29 (53 mg, 0.17 mmol, 97%).
IR (neat): 2934w, 2936w, 1608m, 1587s, 1463m, 1115m, 1303w,
1239m, 1117s, 1086m, 970w, 814m cm-1; 1H NMR (300 MHz, CDCl3)
δ 6.18 (s, 2H), 3.77 (s, 6H), 3.20 (s, 4H), 2.15 (s, 3H); 13C NMR (75
MHz, CDCl3): δ 157.9, 138.1, 114.3, 100.5, 55.5, 27.9, 22.1, 4.7. Anal.
Calcd for C11H15O2: C, 43.16; H, 4.94. Found: C, 43.32; H, 5.09.
HRMS: Calcd for C11H15O2I: 306.0117. Found: 306.0118.
(R)-6-Methoxy-2-methyl-2-vinylchroman (17e): [R]D ) +60.0 (c
1.0, CDCl3, 98% ee). IR (neat): 2926s, 2854s, 1495s, 1465w, 1429w,
;
1374w, 1228m, 1094w, 1044w, 923w cm-1 1H NMR (500 MHz,
CDCl3): δ 6.80 (d, J ) 9.0 Hz, 1H), 6.71 (dd, J ) 3.0, 9.0 Hz, 1H),
6.59 (d, J ) 3.0 Hz, 1H), 5.86 (dd, J ) 11.0, 17.5 Hz, 1H), 5.20 (dd,
J ) 1.0, 17.5 Hz, 1H), 3.76 (s, 3H), 2.71 (m, 2H), 1.93 (m, 1H), 1.84
(m, 1H), 1.59 (s, 3H); 13C NMR (100 MHz, CDCl3): δ 152.9, 148.0,
141.4, 121.8, 117.3, 113.9, 113.8, 113.4, 76.3, 55.6, 31.7, 29.7, 27.1.
Anal. Calcd for C13H16O2: C, 76.44, H, 7.90. Found: C, 76.19; H,
8.08. Chromatographic separation (GC, Cyclosil B): T oven ) 150
°C, tR (S, minor) ) 20.5 min, tR (R, major) ) 20.8 min.
(R)-5,8-Dimethoxy-2,6,7-trimethyl-2-vinylchroman (17f): a color-
less oil. IR (neat): 2935, 1463, 1414, 1329, 1259, 1132, 1088, 1027,
Z-2-(5-Hydroxy-3-methyl-pent-3-enyl)-3-methoxy-5-methyl-phe-
nol (31). To alkyl iodide 29 (75 mg, 0.25 mmol) and zinc chloride (55
mg, 0.41 mmol) was added THF (2 mL) to afford a colorless solution.
To this solution at -78 °C was added tert-butyllithium (0.33 mL, 0.50
mmol, 1.5 M in pentane). After 30 min, this solution was warmed to
RT and stirred for 3 h. In another flask containing Pd(dppf)Cl2 (7.5
mg, 0.01 mmol) and 0.5 mL of THF was added n-butyllithium (8 uL,
0.02 mmol, 2.5 M in hexane) to result a dark purple red solution. To
this solution was added vinyl iodide 30 (75 mg, 0.25 mmol) in 1 mL
of THF. This solution was then transferred to the first flask containing
alkyl zinc species via cannula at room temperature. The solution
immediately turned brownish black but resumed the reddish color in 5
min. The solution was stirred overnight. Without workup, the mixture
was purified eluting with 5% diethyl ether in petroleum ether to afford
a colorless oil, which was dissolved in THF (2 mL). To this solution
was added TBAF (0.16 mL, 0.16 mmol, 1 M in THF). After 2 h, without
workup this mixture was directly chromatographed eluting with 5% to
50% diethyl ether in petroleum ether to afford allylic alcohol (20 mg,
0.08 mmol, 32% over 2 steps) as a colorless oil. To the resulting allylic
alcohol was added lithium n-propyl thiolate (0.8 mL, 0.4 mmol, 0.5 M
in HMPA). The resulting orange mixture was heated at 120 °C
overnight. Without workup, the mixture was chromatographed eluting
with 5% to 60% diethyl ether in petroleum ether to afford allylic alcohol
31 (15 mg, 0.064 mmol, 80%, 26% over three steps) as a colorless oil.
IR (neat): 3320, 2936, 1660, 1615, 1594, 1514, 1463, 1416, 1378,
1
970, 920 cm-1; H NMR (300 MHz, CDCl3): δ 5.89 (dd, J ) 11, 17
Hz, 1H), 5.21 (d, J ) 17 Hz, 1H), 5.07 (d, J ) 11 Hz, 1H), 3.79 (s,
3H), 3.66 (s, 3H), 2.79-2.52 (m, 2H), 2.17 (s, 3H), 2.12 (s, 3H), 1.96-
1.72 (m, 2H), 1.46 (s, 3H); 13C NMR (75 MHz, CDCl3): δ 151.5,
145.2, 142.7, 141.4, 128.5, 120.1, 113.5, 113.3, 76.3, 60.1, 31.2, 26.8,
17.4, 12.2, 12.0; HRMS Calcd for C16H22O3 [M+]: 262.1569; Found:
262.1562; Chromatographic separation (GC, Cyclosil B): T oven )
170 °C, tR (S, minor) ) 32.4 min, tR (R, major) ) 31.6 min.
(R)-6-Fluoro-2-vinylchroman (17g): [R]D ) +62.4 (c 0.35, Et2O,
80% ee); IR (neat): 2926s, 2853m, 1493s, 1434w, 1258m, 1217s,
1138w, 1044w, 997w, 929m, 871w, 809m cm-1; 1H NMR (400 MHz,
CDCl3): δ 6.78 (m, 3H), 5.96 (ddd, J ) 5.6, 10.4, 17.2 Hz, 1H), 5.36
(dt, J ) 1.2, 17.2 Hz, 1H), 5.23 (dt, J ) 1.2, 10.4 Hz, 1H), 4.51 (m,
1H), 2.84 (ddd, J ) 6.4, 10.0, 16.8 Hz, 1H), 2.73 (dt, J ) 4.8, 16.4
Hz, 1H), 2.05 (m, 1H), 1.84 (m, 1H); 13C NMR (125 MHz, CDCl3):
δ 157.9, 150.7, 137.6, 123.0 (d, J ) 7.6 Hz), 117.8 (d, J ) 8.1 Hz),
116.7, 115.5 (d, J ) 9.9 Hz), 114.2 (d, J ) 22.9 Hz), 76.3, 27.4, 24.6;
HRMS Calcd for C11H11OF [M+]: 178.0794; Found: 178.0791; AD
column (Flow rate: 0.4 mL/min, S isomer: 15.80 min, R isomer: 16.68
min).
(R)-6-Methyl-2-vinylchromane (17h): a clear oil (96%). [R]D
)
+76.8 (c 0.69, CHCl3, 17h), 87% ee (HPLC: Chiralpak OJ, 99.7%
1
heptane/0.3% i-PrOH). IR (neat): 1498, 1243, 1227 cm-1; H NMR
(300 MHz, CDCl3): δ 1.84 (m, 1 H), 2.03 (m, 1 H), 2.24 (s, 3 H),
2.82-2.69 (m, 2 H), 4.51 (m, 1 H), 5.22 (ddd, J ) 10.5, 1.5, 1.5 Hz,
1 H), 5.39 (ddd, J ) 17.4, 1.5, 1.5 Hz, 1 H), 5.97 (ddd, J ) 17.4, 10.5,
5.4 Hz, 1 H), 6.74 (m, 2 H), 6.87 (m, 2 H); 13C NMR (75 MHz,
CDCl3): δ 152.2, 137.6, 129.8, 127.9, 127.3, 121.4, 116.5, 116.1, 76.0,
27.5, 24.1, 20.4; HRMS Calcd for C12H14O [M]+: 174.1044; Found:
174.1047.
1319, 1227, 1151, 1076, 1034, 978 cm-1 1H NMR (300 MHz,
;
CDCl3): δ 6.50 (s, 1H), 6.28 (s, 2H), 5.46 (t, J ) 7.5 Hz, 1H), 3.98 (t,
J ) 6.0 Hz, 2H), 3.79 (s, 3H), 2.75 (t, J ) 7.5 Hz, 2H), 2.30 (t, J )
7.8 Hz, 2H), 2.25 (s, 3H), 1.80 (s, 3H); 13C NMR (75 MHz, CDCl3):
δ 158.3, 154.7, 143.0, 137.2, 123.2, 113.0, 109.1, 103.7, 58.8, 55.5,
31.2, 23.9, 22.4, 21.5. HRMS: Calcd for C17H24O4 [M+]: 292.1674.
Found: 292.1675.
5-(2,6-Dimethoxy-4-methylphenyl)-2-methyl-pent-2-en-1-ol (33).
To alkyl iodide 29 (124 mg, 0.405 mmol) and zinc chloride (110 mg,
0.81 mmol) was added THF (3 mL) to afford a colorless solution. To
this solution at -78 °C was added tert-butyllithium (0.59 mL, 0.89
mmol, 1.5 M in pentane). After 30 min, this solution was warmed to
RT and stirred for 3 h. In another flask containing Pd(dppf)Cl2 (15
mg, 0.02 mmol) and 0.5 mL of THF was added n-butyllithium (16 uL,
0.04 mmol, 2.5 M in hexane) to result a dark purple red solution. To
this solution was added vinyl iodide 32 (151 mg, 0.49 mmol) in 1.5
mL of THF. This solution was then transferred to the first flask
containing alkyl zinc species via cannula at room temperature. The
solution immediately turned brownish black but resumed the reddish
color in 5 min. The solution was stirred overnight. Without workup,
(R)-6-Methoxy-2-vinylchromane (17i): a colorless oil. [R]D
+74.6 (c 2.19, CHCl3); 89% ee (HPLC: Chiralpak OJ, 99.7% heptane/
0.3% i-PrOH); IR (neat): 1496, 1425, 1269, 1269, 1221, 1051 cm-1
)
;
1H NMR (300 MHz, CDCl3): δ 1.82 (m, 1 H), 2.03 (m, 1 H), 2.72 (m,
1 H), 2.81 (m, 2 H), 3.74 (s, 3 H), 4.49 (m, 1 H), 5.22 (ddd, J ) 10.5,
1.5, 1.5 Hz, 1 H), 5.36 (ddd, J ) 17.4, 1.5, 1.5 Hz, 1 H), 5.98 (ddd, J
) 17.4, 10.5, 5.4 Hz, 1 H), 6.58 (d, J ) 3.0 Hz, 1 H), 6.67 (m, 1 H),
6.78 (d, J ) 9.0 Hz, 1 H); 13C NMR (75 MHz, CDCl3): δ 153.2, 148.5,
137.6, 122.3, 117.3, 116.2, 113.9, 113.3, 76.0, 55.7, 27.5, 24.5; HRMS
Calcd for C12H14O [M]+: 174.1045; Found: 174.1040.
(R)-8-Methoxy-6-methyl-2-vinylchromane (17j): a white solid
(82%); mp. 35-36 °C; [R]D ) -62.3 (c 0.21, CHCl3), 82% ee
(HPLC: Chiralpak OJ, 99.7% heptane/0.3% i-PrOH). IR (neat) 1592,
1
1554, 1493, 1275, 1228 cm-1; H NMR (500 MHz, CDCl3): δ 1.86
9
11980 J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004

Synthesis of Chiral Chromans
A R T I C L E S
the mixture was purified eluting with 5% diethyl ether in petroleum
ether to afford a colorless oil, which was taken up in 2 mL of THF. To
this solution was added TBAF (1.5 mL, 1.5 mmol, 1 M in THF). After
1 h, without workup, the mixture was purified by column eluting with
5% to 20% diethyl ether in petroleum ether to afford 33 (78 mg, 0.312
mmol, 77%) as a colorless oil.
to 25% diethyl ether in petroleum ether to afford cis allylic alcohol 40
(450 mg, 1.45 mmol, 81%) as a yellow oil.
IR (neat): 3374b, 2959s, 2933s, 2860s, 1614w, 1495s, 1423m,
1362w, 1257s, 1214s, 1145m, 1022m, 899m, 840m, 804m, 780m, 692w
cm-1; 1H NMR (500 MHz, CDCl3): δ 6.84 (dd, J ) 3.5, 9.5 Hz, 1H),
6.79 (td, J ) 3.0, 8.0 Hz, 1H), 6.71 (dd, J ) 5.0, 9.0 Hz, 1H), 4.08 (d,
J ) 6.5 Hz, 2H), 2.65 (t, J ) 7.0 Hz, 2H), 2.38 (q, J ) 7.5 Hz, 2H),
1.43 (bs, 1H), 1.04 (s, 9H), 0.24 (s, 6H); 13C NMR (100 MHz,
CDCl3): δ 157.9, 156.0, 149.4, 134.5 (d, J ) 7.1 Hz), 131.5, 129.3,
118.9 (d, J ) 8.0 Hz), 116.6 (d, J ) 22.5 Hz), 112.9 (d, J ) 22.4 Hz),
58.4, 30.5, 27.4, 25.7, 18.1, -4.28. HRMS: Calcd for C17H27O2FSi
(M+): 310.1764. Found: 310.1783.
IR (neat): 3384b, 2935s, 2850s, 1668w, 1609s, 1588s, 1464s, 1416s,
1381m, 1313s, 1241s, 1185s, 1168sa, 1124s, 1005m, 970m, 813m,
668m cm-1; 1H NMR (300 MHz, CDCl3): δ 6.36 (s, 2H), 5.38 (t, J )
6.9 Hz, 1H), 4.12 (d, J ) 7.2 Hz, 2H), 3.79 (s, 6H), 2.73 (t, J ) 8.4
Hz, 2H), 2.33 (s, 3H), 2.15 (t, J ) 8.4 Hz, 2H), 1.75 (s, 3H); 13C NMR
(75 MHz, CDCl3): δ 158.0, 140.8, 136.7, 122.8, 115.5, 59.3, 55.6,
38.9, 21.9, 21.3, 16.2. HRMS: Calcd for C15H22O3: 250.1569. Found:
250.1566.
(R)-5-Methoxy-2,7-dimethyl-chroman-2-carboxyaldehyde (41).
To a solution of (R)-5-methoxy-2.7-dimethyl-2-vinyl-chroman 17c (0.20
g, 0.92 mmol) in 3.5 mL of dichloromethane was added N-methyl-
morphonline-N-oxide (0.3 g, 2.56 mmol) and aqueous osmium tet-
raoxide (0.27 mL, 4% in water, 0.043 mmol). The solution was stirred
for 5 h at room temperature and diluted with water (10 mL) and
dichloromethane (20 mL). The organic layer was dried over magnesium
sulfate and evaporated to dryness. The residue was chromatographed
with 1:1 to 1:5 of petroleum ether in diethyl ether. The resulting brown
oil (contaminated with osmium residue) was resuspended in acetone
(4 mL) and a solution of sodium periodate (0.4 g, 1.87 mmol) in water
(1 mL) was added. After a white precipitate was formed the reaction
mixture was stirred at room temperature for additional 20 min. The
mixture was filtered through a pad of diatomaceous earth, and the filtrate
was partitioned between water (5 mL) and dichloromethane (10 mL).
The aqueous layer was extracted with dichloromethane (2 × 10 mL).
The combined organic extracts were washed with brine (20 mL) and
dried over magnesium sulfate. The residue was separated by flash
chromatography eluting with 5% to 25% diethyl ether in petroleum
ether to afford aldehyde 41 as a colorless oil (0.19 g, 0.86 mmol, 94%).
[R]D ) +16.3 (c 0.47, Et2O); IR (film): 2935, 1738, 1619, 1587,
1463, 1146, 1113 cm-1; 1H NMR (300 MHz, CDCl3): δ 9.65 (s, 1H),
6.44 (s, 1H), 6.27 (s, 1H), 3.78 (s, 3H), 2.64 (td, J ) 6.4, 17.3 Hz,
1H), 2.47 (ddd, J ) 6.6, 9.2, 16.1 Hz, 1H), 2.30 (s, 3H), 2.22 (m, 1H),
1.77 (ddd, J ) 6.6, 9.5, 15.8 Hz, 1H), 1.40 (s, 3H); 13C NMR (75
MHz, CDCl3): δ 203.5, 157.5, 153.5, 137.4, 110.0, 107.1, 103.4, 80.0,
55.3, 27.2, 21.5, 21.2, 16.1. HRMS: Calcd for C13H16O3: 220.1099.
Found: 220.1093.
5-(2-Hydroxy-6-methoxy-4-methylphenyl)-2-methyl-pent-2-en-1-
ol (34). To allylic alcohol 33 (50 mg, 0.2 mmol) was added
n-propyllithium thiolate (2 mL, 1.0 mmol, 0.5 M in HMPA). The
resulting orange solution was heated at 120 °C for overnight. Without
workup, the mixture was purified by flash chromatography eluting with
5% to 60% diethyl ether in petroleum ether to afford allylic alcohol 34
(41 mg, 0.17 mmol, 87%) as a white solid.
mp: 92-94 °C; IR (neat): 3329, 2935, 1672, 1616, 1595, 1463,
1
1418, 1109 cm-1; H NMR (500 MHz, CDCl3) δ: 6.31 (s, 1H), 6.28
(s, 1H), 5.42 (qt, J ) 1.2, 5.8 Hz, 1H), 5.37 (s, 1H), 4.16 (br t, J ) 5.8
Hz, 2H), 3.80 (s, 3H), 2.75 (dd, J ) 7.6 Hz, 2H), 2.28 (s, 3H), 2.21
(dd, J ) 8.3 Hz, 2H), 1.80 (s, 1H), 1.77 (d, J ) 1.2 Hz, 3H); 13C NMR
(125 MHz, CDCl3) δ: 158.2, 1154.2, 140.9, 136.9, 122.8, 113.4, 108.9,
104.0, 59.3, 55.5, 38.6, 21.6, 21.5, 16.4. HRMS: Calcd for C14H19O2
[M+-OH]: 219.1385. Found: 219.1343.
2-But-3-ynyl-4-fluoro-phenol (39). To a solution of aldehyde 37
(1.72 g, 6.10 mmol) in methanol (20 mL) at room temperature were
added potassium carbonate (1.68 g, 12.2 mmol) a solution of then
acetyldiazophosphonate (1.41 g, 7.32 mmol) in methanol (10 mL). The
solution was stirred at room temperature overnight. The mixture was
filtered through a pad of diatomaceous earth and washed with diethyl
ether (50 mL). The volatile fraction was removed and the residue was
chromatographed eluting with 5% to 20% diethyl ether in petroleum
ether to afford alkyne 39 (0.85 g, 5.18 mmol, 85%) as a colorless oil.
IR (neat): 3504b, 3301s, 2937m, 2117w, 1507s, 1436s, 1329w,
1269m, 1176s, 1092m, 870m, 810m, 764m, 711m, 639s cm-1; 1H NMR
(300 MHz, CDCl3): δ 6.78 (dd, J ) 3.0, 9.0 Hz, 1H), 6.72 (td, J )
3.0, 8.4 Hz, 1H), 6.60 (dd, J ) 4.8, 8.7 Hz, 1H), 5.95 (bs, 1H), 2.74
(d, J ) 7.2 Hz, 2H), 2.41 (td, J ) 2.7, 7.2 Hz, 2H), 1.92 (t, J ) 2.7
Hz, 2H); 13C NMR (75 MHz, CDCl3): δ 158.4, 155.2, 128.5 (d, J )
7.3 Hz), 116.7 (d, J ) 23.1 Hz), 116.1 (d, J ) 8.5 Hz), 113.6 (d, J )
23.1 Hz), 84.0, 69.2, 29.5, 25.5. HRMS: Calcd for C10H9FO: 164.0637.
Found: 164.0638.
(R)-5-Methoxy-2,7-dimethyl-2-(4-methyl-penta-1,3-dienyl)-chro-
man (59). Method A. To a solution of prenyl phenyl sulfone (0.38 g,
1.82 mmol) in distilled THF (4 mL) at -78 °C was added potassium
bis(trimethylsilyl)amide (1.25 mL, 2.0 mmol, 1.6 M in hexanes)
resulting in formation of an orange solution was formed. After 30 min,
to the above solution was slowly added (R)-5-methoxy-2,7-dimethyl-
chroman-2-carbaldehyde (0.40 g, 1.82 mmol) in 3 mL of THF
(including rinsing solvent). The reaction solution turned light yellow
from orange. The reaction mixture was stirred at -78 °C for 1 h and
quenched with 8 mL of water. The aqueous fraction was extracted with
diethyl ether (3 × 50 mL). The combined organic layers were dried
with magnesium sulfate and concentrated in vacuo before silica gel
chromatography eluting with 5% to 50% diethyl ether in petroleum
ether to afford the sulfone alcohol. The sulfone alcohol was then diluted
with 10 mL of dicholomethane and treated with pyridine (3.2 mL) and
acetic anhydride (1.3 mL) followed by a few crystals of DMAP at room
temperature. After being stirred overnight, the solution was diluted with
200 mL of diethyl ether and washed with 1N HCl (3 × 30 mL) and
brine. The organic fraction was then dried with magnesium sulfate and
concentrated. The residue was chromatographed eluting with 5-40%
diethyl ether in petroleum ether to afford the corresponding acetate.
The acetate was then dissolved in 20 mL of distilled methanol, and to
this solution was added sodium bibasic phosphate (2 g) and sodium
amalgam (1.5 g, amalgam in mercury) at 0 °C under argon (be careful
about the pressure build-up). The solution was stirred for 30 min and
filtered through a diatomaceous earth pad on a filtration funnel. The
Z-5-[2-(tert-Butyldimethylsilyloxy)-5-fluorophenyl]-pent-2-en-1-
ol (40). To alkyne phenol 39 (0.81 g, 4.94 mmol) in 5 mL of
dichloromethane at room temperature were added imidazole (0.89 g,
5.93 mmol) and TBSCl (673 mg, 9.88 mmol). The reaction mixture
was stirred overnight and directly chromatographed eluting with 5%
diethyl ether in petroleum ether to afford silyl phenol ether (1.14 g,
4.10 mmol, 83%). To a solution of the silyl phenol ether (1.14 g, 4.10
mmol) in THF (15 mL) at -78 °C was added n-butyllithium (3.1 mL,
4.92 mmol, 1.6 M in hexane). After 30 min at this temperature, to this
dark solution was added paraformaldehyde (185 mg, 6.15 mmol, dried
in a vacuum overnight). The mixture was slowly warmed to room
temperature over 1 h and stirred at room temperature for an additional
1 h. The mixture was concentrated and directly chromatographed eluting
with 5% to 20% diethyl ether in petroleum ether to afford propargyl
alcohol (1.06 g, 3.44 mmol, 84%). To a solution of propargyl alcohol
(554 mg, 1.80 mmol) in methanol (5 mL) was added Lindlar catalyst
(100 mg). This mixture was stirred under 1 atm of hydrogen at room
temperature for 18 h then chromatographed directly eluting with 5%
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11981

A R T I C L E S
Trost et al.
filtrate was concentrated in vacuo and the residue was chromatographed
eluting with 5% to 20% diethyl ether in petroleum ether to afford
5-methoxy-2,7-dimethyl-2-(4-methyl-penta-1,3-dienyl)-chroman 59 (243
mg, 0.89 mmol, 49%, E/Z ) 15:1 to 9:1) as a light yellow oil.
Clusifoliol (6).36 To prenyl ether 61 (12 mg, 0.037 mmol) in 1.5
mL of distilled benzene was added montmorillonite KFC (24 mg, in
two portions, Aldrich) at room temperature. The reaction mixture was
stirred at room temperature for 2 d. Without workup, the mixture was
purified by flash chromatography eluting with 5% diethyl ether in
petroleum ether to afford the desired natural product clusifoliol (3,4-
dihydro-2,7-dimethyl-6-(3-methyl-2-butenyl)-2-(4-methyl-1,3-pentadi-
enyl)-2H-1-benzo-pyran-5-ol) 6 (7 mg, 0.021 mmol, 58%) as a colorless
oil which turned pale brown over 1 d in freezer.
Method B.35 To lithium bis(trimethylsilyl)amide (22 mg, 0.13 mmol)
was added 5-prenylsulfonyl-1-phenyltetrazole (30 mg, 0.11 mmol) in
0.5 mL of a mixture of DMF and HMPA (4:1) at -78 °C resulting in
a yellow solution. Immediately to this mixture was added a solution
of(R)-5-methoxy-2,7-dimethylchroman-2-carbaldehyde (10 mg, 0.045
mmol) in 0.5 mL of a mixture of DMF and HMPA (4:1). The resulting
solution was slowly warmed to room temperature over 2 h. Without
workup, this mixture was directly chromatographed eluting with 2-5%
diethyl ether in petroleum ether to afford desired 5-methoxy-2,7-
dimethyl-2-(4-methyl-penta-1,3-dienyl)-chroman 59 (12 mg, 0.044
mmol, 98%, E/Z ) 20:1) as a light yellow oil.
[R]D ) +0.68 (c 0.9, EtOH); [R]D ) +0.62 (c 0.7, EtOH, authentic
sample); IR (film): 3420, 2925, 2855, 1660, 1619, 1592, 1446, 1417,
1
1259, 1101, 1071, 985, 961 cm-1; H NMR (500 MHz, CDCl3): δ
(Experimental value) 6.33 (dd, J ) 15.0, 11.0 Hz, 1H), 6.17 (s, 1H),
5.76 (d, J ) 11.0 Hz, 1H), 5.55 (d, J ) 15.5 Hz, 1H), 5.14 (tm, J )
7.0 Hz, 1H), 4.46 (bs, 1H), 3.36 (dd, J ) 7.0, 14.0 Hz, 1H), 3.25 (dd,
J ) 7.0, 14.5 Hz, 1H), 2.62 (ddd, J ) 5.5, 5.5, 16.5 Hz, 1H), 2.49
(ddd, J ) 6.0, 9.5, 16.5 Hz, 1H), 2.21 (s, 3H), 1.91 (ddd, J ) 6.0, 6.0,
12.5 Hz, 1H), 1.77 (s, 3H), 1.73 (m, 1H), 1.73 (s, 3H), 1.66 (s, 3H),
1.66 (s, 3H), 1.41 (s, 3H); (Literature value) 6.36 (dd, J ) 10.6, 15.4
Hz, 1H), 6.14 (s, 1H), 5.76 (d, J ) 10.6 Hz, 1H), 5.55 (d, J ) 15.5
Hz, 1H), 5.14 (t, J ) 7.1 Hz, 1H), 3.36 (dd, J ) 7.1, 14.9 Hz, 1H),
3.25 (dd, J ) 7.1, 14.9 Hz, 1H), 2.62 (ddd, J ) 5.6, 5.6, 16.3 Hz, 1H),
2.49 (ddd, J ) 6.2, 9.7, 16.3 Hz, 1H), 2.20 (s, 3H), 1.91 (ddd, J ) 5.9,
5.9, 13.3 Hz, 1H), 1.77 (s, 3H), 1.73 (m, 1H), 1.73 (s, 1H), 1.64 (s,
3H), 1.64 (s, 3H), 1.40 (s, 3H); 13C NMR (100 MHz, CDCl3): δ
(Literature value) 152.1 (152.2), 151.1 (151.2), 135.2 (135.1), 135.0
(135.0), 134.1 (134.2), 130.3 (130.3), 125.0 (125.1), 124.5 (124.5),
123.2 (123.3), 120.0 (120.0), 107.7 (107.7), 105.9 (105.9), 76.0 (75.9),
31.5 (31.4), 27.4 (27.5), 26.0 (26.0), 25.9 (25.9), 24.9 (25.0), 19.5 (19.3),
18.3 (18.2), 18.0 (17.9), 17.0 (17.2). Chiral HPLC separation (AD
column): flow rate 1 mL/min, tR (S, minor) ) 3.52 min, tR (R, major)
) 4.79 min, ee ) 97%. HRMS: Calcd for C22H30O2: 326.2246.
Found: 326.2245.
[R]D ) -5.0 (c 1.27, CHCl3); IR (film): 2929, 1618, 1585, 1352,
1
1115 cm-1; H NMR (500 MHz, CDCl3): δ 6.42 (dd, J ) 10.8, 15.1
Hz, 1H), 6.39 (s, 1H), 6.24 (s, 1H), 5.79 (d, J ) 10.8 Hz, 1H), 5.62 (d,
J ) 15.1 Hz, 1H), 3.81 (s, 3H), 2.65 (td, J ) 5.8, 16.8 Hz, 1H), 2.51
(ddd, J ) 6.1, 9.3, 15.6 Hz, 1H), 2.30 (s, 3H), 1.93 (td, J ) 5.8, 13.4
Hz, 1H), 1.82 (m, 1H), 1.76 (s, 3H), 1.71 (s, 3H), 1.44 (s, 3H); 13C
NMR (125 MHz, CDCl3): δ 157.4, 154.3, 136.8, 135.3, 134.0, 125.2,
124.5, 110.1, 107.3, 102.5, 75.9, 55.2, 31.8, 26.7, 25.9, 21.6, 18.2, 16.8.
HRMS: Calcd for C18H24O2: 272.1776. Found: 272.1776.
(R)-2,7-Dimethyl-2-(4-methylpenta-1,3-dienyl)-chroman-5-ol (60).
To a flask containing 3 mL of anhydrous DMF (from a newly opened
bottle, quality is important) at 0 °C was added sodium hydride (153
mg, 60%) followed by the addition of ethanethiol (0.26 mL). The
solution was warmed to room temperature and stirred for 30 min before
the addition of diene 59 (119 mg, 0.44 mmol) in 4 mL of anhydrous
DMF. The solution was heated at 120 °C for 18 h. To the mixture was
added 20 mL of water and extracted with diethyl ether (3 × 50 mL).
The combined organic fractions were dried with magnesium sulfate
and concentrated in vacuo. The residue was chromatographed eluting
with 5% to 20% diethyl ether to afford phenol 60 (110 mg, 0.43 mmol,
97%) as a pale yellow oil.
(R)-tert-Butyl-[2-(5-Methoxymethoxy-2,7-dimethyl-2H-chromen-
2-yl)-diphenylsilane (69). To a solution of chroman 68 (15 mg, 0.03
mmol) in 0.8 mL of benzene at room temperature was added DDQ
(17 mg, 0.75 mmol). The resulting blue solution was heated under reflux
for 2 h. Without workup, the mixture was directly purified by flash
chromatography eluting with 5% diethyl ether in petroleum ether to
afford chromen 69 (12 mg, 0.024 mmol, 80%) as a light brown oil.
[R]D ) -0.03 (c 0.68, CDCl3); IR (neat): 3072w, 2929s, 2857s,
1618m, 1571w, 1464m, 1428m, 1388w, 1325w, 1156m, 1112s, 1056s,
[R]D ) -10.1 (c 1.2, CHCl3); IR (film): 3419b, 2925s, 2857m,
1627m, 1586s, 1516w, 1436s, 1415s, 1352m, 1312m, 1274s, 1254s,
1130m, 1088s, 1051m, 996m, 872w, 822w, 739w cm-1; 1H NMR (300
MHz, CDCl3): δ 6.29 (dd, J ) 10.8, 15.3 Hz, 1H), 6.24 (s, 1H), 5.67
(d, J ) 10.8 Hz, 1H), 5.50 (d, J ) 15.6 Hz, 1H), 4.48 (s, 1H), 2.46 (m,
1H), 2.13 (s, 3H), 1.80 (m, 2H), 1.65 (s, 3H), 1.59 (s, 3H), 1.33 (s,
3H); 13C NMR (100 MHz, CDCl3): δ 154.8, 153.4, 137.3, 135.6, 133.7,
125.5, 124.4, 110.1, 107.2, 105.7, 76.0, 31.7, 29.7, 26.9, 25.9, 21.2,
18.3, 16.7. HRMS: Calcd for C17H22O2: 258.1620. Found: 258.1618.
1
1112s, 1060s, 1008m, 823m, 739m, 702s cm-1; H NMR (300 MHz,
CDCl3): δ 7.68 (m, 4H), 7.40 (m, 6H), 6.60 (d, J ) 6.0 Hz, 1H), 6.42
(s, 1H), 6.25 (s, 1H), 5.48 (d, J ) 6.0 Hz, 1H), 5.15 (s, 2H), 3.85 (bs,
2H), 3.47 (s, 3H), 2.24 (s, 3H), 2.05 (m, 1H), 1.97 (m, 1H), 1.27 (s,
3H), 1.04 (s, 9H); 13C NMR (125 MHz, CDCl3): δ 153.4, 152.7, 135.5,
135.1, 134.8, 129.6, 127.7, 127.4, 117.0, 111.0, 108.8, 107.4, 94.6,
78.6, 60.0, 56.1, 43.2, 29.7, 26.8, 29.7, 19.1. HRMS: Calcd for
C31H38O4Si [M+]: 502.2539. Found: 502.2540.
(R)-2-(5-Methoxymethoxy-2,7-dimethyl-2H-chromen-2-yl)-etha-
nol (70). To silyl ether 69 (48 mg, 0.127 mmol) in THF (3 mL) at
room temperature was added TBAF (0.25 mL, 0.25 mmol, 1 M in THF).
The solution was stirred at room temperature for 20 h and directly
chromatographed eluting with 5% to 40% diethyl ether in petroleum
ether to afford alcohol 70 (30 mg, 0.114 mmol, 90%) as a pale brown
oil.
(R)-2,7-Dimethyl-5-(3-methyl-but-2-enyloxy)-2-(4-methylpenta-
1,3-dienyl)-chroman (61). To a solution of phenol 60 (20 mg, 0.078
mmol) in acetone (2 mL) were added potassium carbonate (50 mg,
0.36 mmol) and 4-bromo-2-methyl-2-butene (26 mg, 20 µL, 0.17
mmol). The reaction mixture was heacted under reflux for 2 h. Without
workup, the solution was chromatographed eluting with 5% to 10%
diethyl ether in petroleum ether to afford ether 61 (25 mg, 0.077 mmol,
99%) as a colorless oil.
[R]D ) -15.7 (c 1.1, CDCl3, 29.4 °C) IR (film): 2964s, 2926s,
2857s, 1617m, 1585s, 1434s, 1414s, 1381s, 1352m, 1259s, 1235m,
1
1097s, 1013m, 959w, 810m cm-1; H NMR (500 MHz, CDCl3): δ
6.38 (dd, J ) 11.0, 15.0 Hz, 1H), 6.35 (s, 1H), 6.23 (s, 1H), 5.76 (dd,
J ) 1.0, 11.0 Hz, 1H), 5.60 (d, J ) 16.0 Hz, 1H), 5.48 (m, 1H), 4.47
(d, J ) 6.5 Hz, 2H), 2.63 (dt, J ) 5.5, 11.0 Hz, 1H), 2.52 (ddd, J )
6.0, 9.0, 11.0 Hz, 1H), 2.27 (s, 3H), 1.89 (m, 1H), 1.79 (m, 1H), 1.78
(s, 3H), 1.73 (s, 3H), 1.72 (s, 3H), 1.69 (s, 3H), 1.41 (s, 3H); 13C NMR
(125 MHz, CDCl3): δ 156.9, 154.4, 137.0, 136.7, 135.3, 134.2, 125.2,
124.6, 123.9, 120.3, 110.0, 103.8, 75.9, 64.9, 31.9, 26.8, 25.9, 25.8,
21.6, 18.3, 18.2, 17.0. HRMS: Calcd for C22H30O2: 326.2246. Found:
326.2247.
[R]D ) -6.3 (c 1.7, CDCl3, 20.0 °C); IR (film): 3448b, 1915s,
1855m, 1616m, 1571m, 1458m, 1386m, 1325w, 1202w, 1156s, 1108s,
1058s, 1009m, 924w, 828w, 778w cm-1 1H NMR (300 MHz,
;
CDCl3): δ 6.71 (d, J ) 9.9 Hz, 1H), 6.46 (s, 1H), 6.31 (s, 1H), 5.52
(d, J ) 9.6 Hz, 1H), 5.16 (s, 2H), 3.87 (t, J ) 5.7 Hz, 2H), 3.49 (s,
3H), 2.25 (s, 3H), 2.01 (t, J ) 6.0 Hz, 2H), 1.38 (s, 3H); 13C NMR (75
MHz, CDCl3): δ 152.7, 139.9, 126.9, 117.8, 110.9, 108.6, 107.8, 104.1,
94.6, 78.5, 59.4, 56.1, 42.9, 29.7, 26.2. HRMS: Calcd for C15H20O4
[M+]: 264.1368. Found: 264.1362.
9
11982 J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004

Synthesis of Chiral Chromans
A R T I C L E S
(R)-(5-Methoxymethoxy-2,7-dimethyl-2H-chromen-2-yl)-acetal-
dehyde (64). To a solution of alcohol 70 (60 mg, 0.23 mmol) in 2 mL
of dichloromethane was added 4Å molecular sieves and N-methyl
morpholine N-oxide (40 mg, 0.34 mmol) at 0 °C. The mixture was
then stirred at room temperature for 30 min. At 0 °C to this mixture
was added tetrapropylammonium perruthenate (4 mg, 0.011 mmol)
resulting in a dark brown solution. After 1 h at room temperature-
,without workup, the mixture was directly chromatographed eluting
with 5% to 20% diethyl ether in petroleum ether to afford aldehyde 64
as a yellow oil (58 mg, 0.22 mmol, 98%).
[R]D ) +67.5 (c 1.1, CDCl3); IR (neat): 2964s, 2925s, 2856s, 1616s,
1572m, 1453m, 1404w, 1386m, 1325w, 1299w, 1232w, 12021, 1156s,
1109s, 1059s, 1008s, 945w, 924w, 827w, 777w, 709w cm-1; 1H NMR
(500 MHz, CDCl3): δ 6.67 (d, J ) 10.0 Hz, 1H), 6.42 (s, 1H), 6.31
(d, J ) 1.0 Hz, 1H), 5.48 (d, J ) 10.0 Hz, 1H), 5.16 (s, 2H), 5.11 (t,
J ) 7.0 Hz, 1H), 5.08 (t, J ) 7.0 Hz, 1H), 3.49 (s, 3H), 2.25 (s, 3H),
2.11 (m, 4H), 1.95 (dd, J ) 6.5 Hz, 2H), 1.73 (ddd, J ) 6.0, 10.5,
14.0 Hz, 1H), 1.67 (s, 3H), 1.66 (ddd, J ) 7.0, 10.5, 13.5 Hz, 1H),
1.59 (s, 3H), 1.57 (s, 3H), 1.37 (s, 3H); 13C NMR (125 MHz, CDCl3):
δ 153.7, 152.7, 139.5, 135.2, 131.3, 127.2, 124.3, 124.1, 117.2, 110.9,
108.7, 107.2, 94.7, 78.1, 56.1, 41.0, 39.7, 29.7, 26.6, 26.3, 25.7, 22.6,
21.9, 17.7, 15.9. HRMS: Calcd for C24H34O3 [M+]: 370.2508.
Found: 370.2524.
[R]D ) +44.4 (c 0.5, CDCl3, 21.6 °C); IR (film): 2959m, 2928m,
2859w, 2741w, 1996w, 1724s, 1616s, 1571m, 1453m, 1386m, 1302w,
1233w, 1203w, 1156s, 1110s, 1059s, 1006m, 940w, 923w, 829w, 777w
1
cm-1; H NMR (300 MHz, CDCl3): δ 9.87 (t, J ) 2.4 Hz, 1H), 6.75
(d, J ) 10.2 Hz, 1H), 6.48 (s, 1H), 6.34 (s, 1H), 5.57 (d, J ) 9.9 Hz,
1H), 5.17 (s, 2H), 3.49 (s, 3H), 2.72 (d, J ) 2.4 Hz, 2H), 2.26 (s, 3H),
1.52 (s, 3H); 13C NMR (75 MHz, CDCl3): δ 201.0, 152.9, 152.8, 140.3,
125.4, 118.6, 110.9, 108.4, 108.0, 94.7, 76.1, 56.2, 53.6, 26.6, 21.9.
HRMS: Calcd for C15H18O4 [M+]: 262.1205. Found: 262.1195.
(R)-2-(4,8-Dimethyl-nona-3,7-dienyl)-5-methoxymethoxy-2,7-di-
methyl-2H-chromene (62). To a solution of vinyl iodide 63 (30 mg,
0.12 mmol) in THF (1 mL) at -78 °C was added n-butyllithium (93
µL, 0.15 mmol, 1.6 M in hexane). After 30 min, to this colorless clear
solution was added a solution of aldehyde 64 (15 mg, 0.057 mmol) in
THF (0.5 mL). The resulting mixture turned pale yellow and was
warmed to room temperature from -78 °C over 10 h. The mixture
was quenched with water and chromatographed eluting with 5% to
50% diethyl ether in petroleum ether to afford allylic alcohol 72 (20
mg, 0.515 mmol, 90%, dr ) 2:1) as a slightly brown oil. The reaction
was repeated to afford another 20 mg of allylic alchol 72. To a solution
of allylic alcohol 72 (40 mg, 0.103 mmol) in THF (2 mL) was added
n-butyllithium (78 µL, 0.124 mmol) at -78 °C. After 10 min, to this
solution was added methyl chloroformate (12 mg, 10 µL). The resulting
solution was stirred for 2 h at -78 °C and warmed to room temperature.
Without workup, the mixture was purified by flash column chroma-
tography eluting with 5% to 15% diethyl ether in petroleum ether to
afford allylic carbonate 73 (38 mg, 0.086 mmol, 83%) as a colorless
oil. To a solution of allylic carbonate 73 (26 mg, 0.059 mmol) in THF
(2 mL) were added π-allyl palladium chloride dimer (1 mg, 0.0029
mmol) and phosphite 74 (1.6 mg, 0.012 mmol). To this solution at 0
°C was added L-selectride (0.12 mL, 0.12 mmol, 1 M in THF). The
resulting dark yellow solution was warmed to room temperature and
stirred for an additional 20 min. The reaction mixture was directly
chromatographed eluting with 5% diethyl ether in petroleum ether to
afford chromen 62 (16 mg, 0.043 mmol, 74%) as a single isomer and
a colorless oil.
Conclusion
In summary, the Pd-catalyzed intramolecular asymmetric
allylic alkylation (AAA) of phenol allyl carbonates provides a
facile access to chiral chromans. The acetic acid additive
dramatically increased the yield and ee of this type of reaction.
The influence of olefin geometry and substitution pattern on
the ee, and the absolute configuration of the chiral chromans
suggest a mechanism involving the cyclization of the more
reactive π-allylpalladium diastereomer as the enantiodiscrimi-
nating step (Curtin-Hammett conditions). This methodology
led to the first enantioselective total synthesis of clusifoliol,
siccanin, and the enantioselective synthesis of intermediates
leading to the enantioselective total synthesis of dodaurichro-
manic acid A and nebivolol.
Acknowledgment. We thank the National Institutes of Health
(GM-13598) and the National Science Foundation for their
generous support of our programs. H.C.S. was a Stanford
Graduate Fellow. Mass spectra were provided by the Mass
Spectrometry Regional Center of the University of CaliforniasSan
Francisco supported by the NIH Division of Research Resources.
Supporting Information Available: Experimental procedures
for the preparation of 15b, 15c, 15j-f 19-24, 26, 33, 42-44,
65-68. This material is available free of charge via the Internet
at http://pubs.acs.org.
JA048078T
9
J. AM. CHEM. SOC. VOL. 126, NO. 38, 2004 11983