Diastereoselectivity in asymmetric allylations: The role of vicinal chirality in the allyl nucleophile for SE2′ reactions with aldehydes

Article abstract of DOI:10.1139/v03-168

A series of nonracemic homoallylic alcohols have been prepared by asymmetric allylation using the (R,R)-and (S,S)-1,2-diamino-1,2-diphenylethane bis-sulfonamide controller ligands for in situ formation of chiral B-allyl-1,3,2-diazaborolidines. Diastereofa

Full text of DOI:10.1139/v03-168

120
Diastereoselectivity in asymmetric allylations: The
role of vicinal chirality in the allyl nucleophile for
SE2′ reactions with aldehydes1
David R. Williams, Kevin G. Meyer, Khalida Shamim, and Samarjit Patnaik
Abstract: A series of nonracemic homoallylic alcohols have been prepared by asymmetric allylation using the (R,R)-
and (S,S)-1,2-diamino-1,2-diphenylethane bis-sulfonamide controller ligands for in situ formation of chiral B-allyl-1,3,2-
diazaborolidines. Diastereofacial selectivity is influenced by adjacent stereochemistry incorporated into the allyl moiety
at C-2, in addition to the expected role of the chiral auxiliary. Additional asymmetry in the aldehyde reactant intro-
duces threefold stereodifferentiation. A model is developed to identify reinforcing stereochemical relationships, and ex-
amples have ascertained the relative significance of these factors. The methodology supports the construction of
complex homoallylic alcohols in a highly convergent fashion.
Key words: asymmetric allylation, diastereofacial selectivity, 1,4-stereocontrol, homoallylic alcohols.
Résumé : On a préparé une série d’alcools homoallyliques non racémiques en procédant à l’allylation asymétrique à
l’aide des ligands de contrôle (R,R)- et (S,S)-1,2-diamino-1,2-diphényléthane bis-sulfonamide pour la formation in situ
de B-allyl-1,3,2-diazaborolidines chirales. La sélectivité diastéréofaciale est influencée par la stéréochimie adjacente qui
est incorporée dans la portion allyle au niveau C-2, en plus du rôle attendu de l’auxiliaire chiral. Une asymétrie addi-
tionnelle dans l’aldéhyde utilisé comme réactif introduit une triple stéréodifférenciation. On a développé un modèle
pour identifier les relations stéréochimiques qui se renforcent et divers exemples ont permis de confirmer la significa-
tion relative de ces facteurs. La méthodologie se prête à la construction d’alcools homoallyliques complexes d’une fa-
çon extrêmement convergente.
Mots clés : allylation asymétrique, sélectivité diastéréofaciale, stéréocontrôle 1,4, alcools homoallyliques.
[Traduit par la Rédaction] Williams et al. 130
Introduction
effective chiral auxiliaries to enable enantioselective Diels–
Alder, aldol, and allylation reactions. We recognized that the
mild conditions needed for transmetalation (eq. [1]) of allyl-
stannanes 1 to nonracemic B-allyl-1,3,2-diazaborolidines 3
with bromoborane 2 would support molecular complexity
and functionalization in the allyl segment with substitution
at C-2. Thus, asymmetric allylation could prove widely use-
ful for strategic bond formation in a convergent synthesis
approach. Recent successes along these lines in our pro-
grams, towards hennoxazole A (6), amphidinolide K (7), and
phorboxazole A (8), have been encouraging. This study will
examine, in detail, the role of vicinal chirality in the allyl
component as a result of substitution at the adjacent methy-
lene (R1 ≠ R2 in 1) of the nascent nucleophile. We will ex-
plore the impact of this additional element of asymmetry for
stereochemically reinforcing and nonreinforcing relation-
ships with the auxiliary of 3 in the production of diastere-
omeric alcohols.
Asymmetric allylation processes are generally recognized
as important tools for stereocontrolled synthesis (1). Impres-
sive levels of stereoselectivity have been achieved for the re-
actions of allyl and crotyl nucleophiles with aldehydes,
leading to the formation of nonracemic homoallylic alcohols
(2). Iterations of this scheme are propagated by oxidative
cleavage of the terminal olefin and the generation of an alde-
hyde for sequentially increasing complexity. As often uti-
lized in combination with aldol reactions, asymmetric
allylation has provided, in large measure, a significant ad-
vance for syntheses of natural products of polyacetate and
polypropionate origin (3).
In 1989, Corey and co-workers (4, 5) reported (R,R)- and
(S,S)-1,2-diamino-1,2-diphenylethane N,N-sulfonamides as
Received 1 August 2003. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 9 January 2004.
Dedicated to Professor Ed Piers for his many contributions
to organic chemistry.
[1]
D. Williams,2 K.G. Meyer, K. Shamim, and S. Patnaik.
Department of Chemistry, Indiana University, 800 East
Kirkwood Avenue, Bloomington, IN 47405-7102, U.S.A.
1This article is part of a Special Issue dedicated to Professor
Ed Piers.
2Corresponding author (e-mail: williamd@indiana.edu).
Can. J. Chem. 82: 120–130 (2004)
doi: 10.1139/V03-168
© 2004 NRC Canada

Williams et al.
121
[3]
Results and discussion
Our preliminary studies (9) have shown that a range of
functionality is tolerated under the ambient conditions of
transmetalation of 1 to borane 2. Examples of compatible
functional groups include isolated alkenes, dithioacetals,
para-methoxybenzyl, and benzyl ethers, esters, vinylstan-
nanes, and most silyl ethers. The presence of Lewis-acid-
sensitive protecting groups, such as methoxymethyl (MOM),
β-methoxyethoxymethyl (MEM), and tetrahydropyranyl (THP)
ethers, as well as other ketals and acetals, are incompatible
with bromoborane 2. Additionally, most allylsilanes do not
undergo a productive transmetalation with bromoborane 2.
Upon the in situ formation of allylborane 3 at room tempera-
ture, reactions are cooled to –78 °C for the introduction of
stoichiometric aldehyde, leading to complete condensations
within 2 to 3 h. A variety of functionality may also be incor-
porated in the aldehyde component. As demonstrated by the
reactions of enantiomeric stannanes 4 and 5, the introduction
of a stereogenic center that is separated by at least one meth-
ylene unit from the reactive allyl moiety plays little or no
role in determining the diastereoselectivity of the condensa-
tion. The chiral auxiliary is the major factor to consider in
the production of the new chiral alcohols 7 and 8.
With these observations in mind, we have examined the
role of adjacent stereogenicity as a control factor for our
asymmetric allylations (eq. [1]; R1 ≠ R2). Two general path-
ways were deployed for the preparation of starting stannanes
bearing a chiral carbon at the C-2 allylic position by the in-
troduction of a methyl or hydroxyl substituent.
In the former case, we have utilized our asymmetric con-
jugate addition methodology (12) for reactions of the
alkenylcopper species derived from the Grignard reagent of
2-bromo-3-trimethylsilyl-1-propene (13) with N-enoyl-4-
phenyl-1,3-oxazolidinone 12. The imide 13 (diastereomeric
ratio (dr) 25:1) was purified by flash chromatography and
reduced to yield alcohol 14, which was subsequently O-
protected as the corresponding tert-butyldiphenylsilyl ether
15. Additional complexity was introduced by direct conver-
sion of 14 to the primary iodide 16 for alkylation with 2-
lithio-1,3-dithiane, providing the allylsilane 17. The allylic
silanes 15 and 17 were transformed into the desired
stannanes 18 and 19 via quantitative, low-temperature
reactions to yield the intermediate bromides (R = Br) with
subsequent displacement with tri-n-butylstannylcopper
dimethylsulfide complex (14). In analogous fashion, we also
prepared ent-18.3 While other routes have been devised and
reduced to practice, Scheme 1 offers simplicity and general-
ity.4
[2]
For the preparation of stannanes bearing an allylic
hydroxyl substituent, the aliphatic aldehydes 20 (R = H and
R = CH2CH=CHCH3) were converted into the chiral epoxy-
alcohols 21 via Wittig olefination, reduction, and Sharpless
asymmetric epoxidation (15), as illustrated in Scheme 2.
Mesylation and immediate reduction of crude epoxy
sulfonate with zinc dust in the presence of sodium iodide
(16) gave optically active 22 in excellent yield. The second-
ary alcohol in 22 was then used to direct allylic lithiation
(17) for C-alkylation with tri-n-butylstannyl iodide.
Acylation with benzoic anhydride led to pure 23a (R = H) or
23b (R = CH2CH=CH–CH3) for our allylation experiments.5
Results of our allylation studies are assembled in Table 1.6
Data regarding the stereoselectivity of each condensation re-
action was obtained by an initial flash silica-gel column to
separate product diastereomers from other organic compo-
nents, including the chiral auxiliary, which can be recovered
and recycled. Thus, overall yields of these attempts were
readily assessed, and the ratios of diastereomeric products
In previous studies by Nishigaichi et al. (10), moderate
stereocontrol in the addition of racemic allylstannanes 9ab
to simple aldehydes was observed. Thus, the placement of
methoxy or acetoxy substituents directly adjacent to the re-
active allyl component resulted in a diastereofacial prefer-
ence, which was strongly influenced by the choice of Lewis
acid. Related chromium- and indium-mediated allylations
(11) have produced 1,4-syn-diol derivatives with diastereo-
meric excesses of 60–90%.
3 Experimental details for the preparation of allylstannanes 18, 19, and ent-18 will be published in the full account of the synthesis of
leucascandrolide A.
4 An alternative method for general introduction of the allylic silane uses methyl ketones for kinetic deprotonation, and transformation to their
corresponding vinyl triflates. Cross-coupling with (trimethylsilyl)methyl magnesium chloride in the presence of Ni(acac)2 (THF at 22 °C)
yields the C-2-substituted allylic silanes.
5 Experimental details for the preparation of stannanes 23a, 23b, and (R)-23b will be published in the full account of our total synthesis of
amphidinolide K (see also ref. 7).
6 We have used a standard set of conditions for these examples of asymmetric allylation, and further attempts to optimize yields or product
ratios have not been made in most cases.
© 2004 NRC Canada

122
Can. J. Chem. Vol. 82, 2004
Fig. 1. Diastereofacial selection in allylation reactions.
Scheme 1. (a) CH2=C(MgBr)CH2SiMe3, CuBr·DMS, THF,
–78 °C to –30 °C, 90% (>90% diastereomeric excess (de));
(b) LiBH4, MeOH, Et2O, 0 °C, 80%; (c) t-BuPh2SiCl, imid,
DMF, room temperature (rt), 95%; (d) DCC·MeI, THF, 88%;
(e) 2-lithiodithiane, THF, –78 °C to –5 °C, 90%; (f) NBS, pro-
pylene oxide, –78 °C, CH2Cl2:DMF (3:4); then add Bu3SnLi,
CuBr·DMS to crude allyl bromide, 70%–80%.
mined by Kakisawa analysis of the corresponding Mosher
esters (18). Initially, these studies examined reactions with
simple aldehydes, as summarized in entries 1 and 2 of Ta-
ble 1. Reactions were evaluated using bromoborane 2, bear-
ing the (R,R)- and the (S,S)-sulfonamides. The presence of
α-stereogenicity in the stannanes 19 and 23a produced reac-
tions consistent with stereochemically-reinforcing and
nonreinforcing relationships with the auxiliaries (19).7 Best
results were obtained using the (R,R)-2, which yielded the
1,4-syn-diastereoisomers as major products. The benzoate
substituent of 23a in entry 2 was more effective in reinforc-
ing, as well as countering, the bias of the auxiliary compared
with the methyl substitution of 19 in entry 1.
Scheme 2. For R = H or R = CH2CH=CHCH3; (a) Ph3PC-
(CH3)COOEt, CH2Cl2, rt, 82%; (b) DIBAL, CH2Cl2, –78 °C,
93%; (c) L-(+)-DET, Ti(O-i-Pr)4, t-BuOOH, CaH2, 4 Å molecu-
lar sieves, CH2Cl2, –25 °C, 89% (>97% de); (d) MsCl, Et3N,
CH2Cl2; then NaI, Zn dust, 2-butanone, 80 °C, 90%; (e) n-BuLi,
TMEDA, Et2O, rt; then Bu3SnI, 50%; (f) benzoic anhydride,
Et3N, CH2Cl2, catalytic amount DMAP, 93%.
Our rationale for the stereoselectivity of these reactions is
illustrated in the diagrams of Fig. 1. Transmetalation gives
the sp2 hybridized borane 3, which reversibly binds the alde-
hyde in the axial orientation 24 and provides a synclinal dis-
position of the Lewis acid with respect to the aldehydic
hydrogen. This complexation serves to activate the electro-
philic carbonyl, as well as to trigger nucleophilicity of the
allyl moiety. Structure 24 also presents a rehybridization of
boron to a tetrahedral geometry, and the five-membered
heterocyclic ring assumes a conformation that minimizes the
nonbonded interactions of sulfonyl and phenyl substituents.
Facial selectivity in 24 is determined by the relative energies
of diastereomeric transition states as the reactive C1 and C3
carbons approach bonding proximity. Thus, our arguments
consider bonding via the closed, chair-like arrangements of
25 and 26 as useful models to depict interactions that may
contribute to a difference in transition-state energies. Since
facial selectivity is not determined by the initial complex-
ation, we offer the following contrivance to aid the visual-
ization of 24 to 25 and 26 (Fig. 1): This process can be
visualized from 24 by a rotation (arrow a) of the carbonyl
(approximately 90°) and subsequent attack at the re face, as
depicted in the favored situation in 25 wherein the illustra-
tion of 24 is reoriented by a 90° rotation (allyl is projected
forward). On the other hand, rotation of the aldehyde in the
clockwise direction in 24 (arrow b) results in the unfavorable
nonbonded interactions of the aldehydic hydrogen with the
were obtained by the integration of selected proton signals
in the NMR (400 MHz) spectra. Further confirmations of
product ratios in most examples were available via analyti-
cal HPLC. In cases that provided a reasonable level of
stereoselectivity, the major products were isolated and puri-
fied by silica-gel flash chromatography or preparative thin-
layer chromatography for complete spectroscopic character-
izations. Trace amounts of minor diastereomers or challeng-
ing separations from reactions that afforded little selectivity
hampered our efforts for full characterization of these mate-
rials. However, the stereochemical redundance of our study
provided fully characterized major products, which were
then compared and identified as minor components in re-
lated examples. The assignment of absolute stereochemistry
of each purified homoallylic alcohol of Table 1 was deter-
7 We have utilized the terms “reinforcing” and “nonreinforcing”, as introduced by Evans et al. (19), to describe the relationships of
stereochemical elements within a reactant.
© 2004 NRC Canada

Williams et al.
123
Table 1. Asymmetric allylations.
© 2004 NRC Canada

124
Can. J. Chem. Vol. 82, 2004
Table 1 (continued).
© 2004 NRC Canada

Williams et al.
125
Table 1 (concluded).
Note: TBS = Si-t-BuMe2; TBDPS = Si-t-BuPh2; Bz = benzoate; Piv = pivaloate.
aYields are based on purified products from flash silica-gel chromatography.
bRatios were determined by HPLC separations of diastereomers on a Zorbax silica column by elution with 1% isopropanol in hexanes.
1
cRatios were calculated from H NMR (400 MHz) data in C6D6 via the integration of alkenylic hydrogen signals.
dRatios were determined by HPLC separations on a Chiralpak AD column via elution with 1% isopropanol in hexanes.
eUnreacted aldehyde (15%–25%) was recovered in these reactions.
equatorial toluenesulfonyl group of the auxiliary, as shown
in 26 (a 90° rotation of 24 with aldehyde projected forward).
Thus, our modeling suggests that this nonbonded interaction
raises the transition-state energy for si face reactions in the
case of the (R,R)-chiral controller (20).
Fig. 2. Preferred transition-state arrangements for asymmetric
allylation using (R,R)-2.
The consequences of steric interactions resulting from al-
lylic branching of the C-2 substituent of 3 introduce an addi-
tional level of complexity. As illustrated in Fig. 1, the
stereogenic center derived from stannanes such as 19 and
23ab provides a stereochemically reinforcing scenario in 25
(X = CH3 or OBz) by a minimization of the A(1,3)-strain,
which also directs the methine hydrogen into the region near
the sulfonyl group of the auxiliary (21, 22).8,9 This becomes
problematic in the nonreinforcing case, 26. Aside from steric
arguments, the allylic oxygen in the benzoate conformer of
25 (X = OBz) is also anticipated to have electronic effects,
which favor the nucleophilic character of the olefin. The
composite picture for these allylations suggests that the con-
sequences of isopropyl substitution at C-2 in 25 (X and R1 =
CH3) is less significant than the interaction of the aldehydic
hydrogen and toluenesulfonyl moiety in 26. This aspect is
addressed with greater clarity in Fig. 2, where the proximity
of reactive centers at C-1 and C-3 in 27 directs the C-2-
branched substituent away from the developing chair. Alter-
natively, the situation in 26 (Fig. 1) projects the aldehydic
hydrogen directly under the six-membered transition state
and into the neighboring sulfonyl group. Finally, our models
have also incorporated the role of α-chirality in the alde-
hyde. Diagram 28 in Fig. 2 describes a model for Felkin–
Anh addition to the carbonyl while maintaining the favor-
able relationships of asymmetry, as previously described in
25 (23).10
Since the feature of chirality in the aldehyde component
multiplies the complexity of the reaction twofold, this aspect
was first evaluated with the enantiomerically pure (R)- and
(S)-2-methyl-3-hydroxypropanal derivatives, as shown in en-
tries 3 and 4. The influence of α-stereochemistry in the alde-
hyde accentuated the reinforcing characteristics of the
stannane and the (R,R)-auxiliary through the expected mode
of Felkin–Anh addition (entry 3), yielding predominantly the
all-syn adduct 33. On the other hand, it was not sufficient to
merely coordinate the chirality of the aldehyde with the
(S,S)-auxiliary 2 in entry 4. This example led to a 1.7:1 mix-
ture of diastereomers; and this points out the importance of
allylic stereochemistry in the starting stannane. As observed
in reactions with achiral aldehydes, the allylic benzoates 23a
and 23b led to excellent diastereoselectivity in the matched
8 The energy requirements for the A(1,3)-strain in 25 (X = CH3) are estimated to be in the range of 0.4–0.7 kcal/mol, compared with the 1,3-
H/H interaction. See ref. 21.
9 Interestingly, the preferred transition state 25 (X = OBz) may present characteristics that resemble the “inside alkoxy effect” proposed by
Stork, Houk, and Jäger for allylic ethers and alcohols. See ref. 22.
10Factors for the diastereoselectivity of the aldol reaction using boron enolates derived from chiral ethyl ketones pose some related concerns.
Some examples include those found in ref. 23.
© 2004 NRC Canada

126
Can. J. Chem. Vol. 82, 2004
cases for Felkin–Anh addition (entries 5 and 7). However, it
seems sufficient to coordinate the stereochemically reinforc-
ing asymmetry of the allylic benzoate 23a with 2 (R,R) for
stereocontrolled condensations via the anti-Felkin mode (en-
try 6).
distilled from sodium benzophenone ketyl. Dimethylforma-
mide (DMF), methylene chloride (CH2Cl2), pyridine, oxalyl
chloride, triethylamine (Et3N), benzene, and toluene were
distilled from calcium hydride. Ethyl acetate (EtOAc) and
hexanes for chromatography were distilled from glass prior
to use. Column chromatography was conducted using silica
gel 60 (230–400 mesh) from E.M. Science. Precoated glass
plates 60F-254 (E. Merck, 0.25 mm and 0.50 mm thickness)
were used for analytical and preparative thin layer chroma-
tography (TLC). Reactions were carried out in flame-dried
glassware under an inert atmosphere. Elemental analyses
were performed by Galbraith Laboratories, Tenn., U.S.A.
Our experiments have also evaluated the effect of β-
chirality in the aldehyde, as precipitated by our studies to-
ward leucascandrolide A (24). Tabulated by the results of
entries 8, 9, and 10, threefold stereodifferentiation was ob-
served. As documented in previous cases, the cooperation of
allylic chirality in the stannane with the auxiliary 2 was a
dominant feature. However, the resulting stereochemistry of
the major adduct does not conform to the expected result of
the Evans polar model for reactions of β-alkoxy aldehydes
(19). A challenging situation was also presented by the
nonracemic β,γ-unsaturated aldehyde (25)11 used in entries
11, 12, and 13. Epimerization of this sensitive substrate was
avoided, and good yields of condensation products were ob-
tained in spite of considerable steric hindrance. Thus, while
the anticipated cooperation of the asymmetry of the stannane
and 2 remained intact, the nature of the α-substitution in the
aldehyde failed to deliver a clear preference for Felkin or
anti-Felkin arrangements. More extensive studies will be
needed to probe the issues raised by these entries.
Finally, the versatility of this approach is displayed by the
functional complexity of the aldehydes used for entries 14,
15, and 16.12 Our plans recognized the reinforcing character-
istics of stannane 23b and (R,R)-2, as well as the inherent
Felkin–Anh face selectivity for the nonracemic epoxy–alde-
hyde, as shown in entry 14. This situation is remarkably pre-
dictable, as demonstrated by the use of the diastereomeric
stannane (R)-23b in entry 15, which yielded a productive re-
action with the (S,S)-sulfonamide auxiliary and provided for
Felkin–Anh addition to give the homoallylic alcohol 50. The
presence of the additional stereocenter at C-4 in the alde-
hyde (OTIPS of entries 15 and 16) did not impact the
stereoselectivity in a major way and demonstrates the flexi-
bility of this methodology for the synthesis of a family of
complex diastereoisomers.
General allylation procedure
A 35 mL Schlenk flask was charged with (S,S) or (R,R)-
N,N′-bis-para-toluenesulfonyl-1,2-diamino-1,2-diphenylethane
(0.22 g, 0.42 mmol) and heated to ~90 °C under vacuum
(30 Pa). After 12 h, the resulting white solid was cooled to
room temperature and CH2Cl2 (4 mL) was added under an ar-
gon atmosphere. The resultant solution was cooled to 0 °C
and treated with fresh boron tribromide (0.42 mL of a 1.0
mol L–1 solution in CH2Cl2 (Aldrich®), 0.42 mmol). The re-
sulting orange solution was stirred at 0 °C for 10 min,
warmed to room temperature, and stirred for 1 h. The sol-
vent and HBr were removed carefully under reduced pres-
sure (0.2 mmHg). The resulting solid was dissolved in
CH2Cl2 (4 mL), stirred for 10 min, and concentrated under
high vacuum. The yellow solid was diluted again with
CH2Cl2 (4 mL) and cooled to 0 °C, and a solution of the
allylstannane (0.47 mmol) in CH2Cl2 (~1 mL) was added.
The yellow solution was stirred for 16 h at room tempera-
ture. After this time the reaction was cooled to –78 °C, and a
solution of aldehyde (0.31 mmol) in CH2Cl2 (~1 mL) was
added. The reaction mixture was stirred at –78 °C for 2 h
and then quenched with pH 7 buffer (1 mL). The mixture
was allowed to warm to room temperature and diluted with
CH2Cl2 (15 mL). The organic phase was washed with satu-
rated aqueous NaHCO3 (15 mL). The aqueous phase was ex-
tracted with CH2Cl2 (15 mL). The combined organic phases
were dried (MgSO4), filtered, and concentrated in vacuo.
The resulting solid was dissolved in Et2O and filtered to re-
cover the sparingly soluble bis-sulfonamide chiral auxiliary.
The filtrate was concentrated in vacuo, and the residue was
purified by flash silica-gel chromatography to yield the mix-
ture of secondary homoallylic alcohols. Diastereomeric ra-
tios were initially determined by the integration of selected
proton signals in NMR (400 MHz) spectra. Product ratios
were also obtained by analytical HPLC (see notes of Table 1
for conditions). Major diastereomers were purified by flash
column chromatography or preparative TLC (silica gel) for
full characterization. Solvents for chromatography and data
for full characterization of major adducts are included below.
Experimental section
General
Infrared spectra were recorded on a Mattson Galaxy 4020
FT-IR instrument with use of an internal standard. Proton
and carbon spectra were recorded on a 400 MHz spectrome-
ter in FT mode. Proton chemical shifts are reported in δ, us-
ing as a reference the appropriate signal for residual solvent
protons, as follows: 7.26 for d-chloroform and 7.20 for d6-
benzene. Carbon chemical shifts are reported in δ, using the
center peak of the solvent signal as a reference, as follows:
77.0 for CDCl3. Mass spectra were obtained on a Kratos
MS80 RFAQQ instrument. Optical rotations were measured
using a PerkinElmer 241 polarimeter. Concentration (c) is
reported as g/100 mL, and temperature is room temperature.
Tetrahydrofuran (THF) and diethyl ether (Et2O) were freshly
(4S,7S)-1-(tert-Butyldiphenylsilanyloxy)-9-([1,3]dithian-2-
yl)-7-methyl-6-methylene-nonan-4-ol (29)
Application of the allylation procedure above led to the
11The route for preparation of the nonracemic β,γ-unsaturated aldehyde (entries 11, 12, and 13 of Table 1) has been previously described; see
ref. 25 and also ref. 8.
12The preparation of optically active aldehydes for entries 14, 15, and 16 has been described in the course of our total synthesis of
amphidinolide K. Experimental details are found in the supporting information of ref. 7.
© 2004 NRC Canada

Williams et al.
127
isolation of 29 (entry 1, Table 1) as the major adduct, which
was characterized as follows: Rf = 0.52 in 10% EtOAc–hex-
anes. IR (neat) (cm–1): 3453, 3070, 3043, 2952, 2926, 2850,
135.5, 134.7, 133.8, 133.1, 130.1, 129.6, 128.4, 128.3,
127.7, 127.6, 125.7, 113.4, 75.9, 67.9, 63.3, 42.5, 38.6, 32.4,
30.2, 28.6, 25.8, 19.2. HR-MS m/z calcd. for C39H46O4Si
([M]+): 606.3165; found: 606.3169.
1
1109. H NMR (400 MHz, CDCl3) δ: 7.70–7.67 (m, 4H),
7.45–7.36 (m, 6H), 4.91 (s, 1H), 4.87 (s, 1H), 4.02 (t, J =
6.4 Hz, 1H), 3.78–3.69 (m, 3H), 2.95–2.78 (m, 4H), 2.22
(dd, J = 3.8, 14.4 Hz, 1H), 2.18 (br s, 1H), 2.14–2.04 (m,
3H), 1.91–1.78 (m, 1H), 1.77–1.46 (m, 8H), 1.06 (s, 9H),
1.04 (d, J = 7.2 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ:
150.7, 135.5, 133.7, 129.5, 127.6, 110.9, 69.1, 64.0, 47.7,
42.5, 39.4, 33.7, 33.2, 32.0, 30.4, 28.8, 26.8, 25.9, 20.1,
19.1. HR-MS m/e calcd. for C31H44OS2Si ([M – H2O]+):
524.2603; found: 524.2607.
(2R,3R,6S)-1-(tert-Butyldiphenylsilanyloxy)-8-([1,3]dithian-
2-yl)-2,6-dimethyl-5-methylene-octan-3-ol (33)
Application of the allylation procedure above led to the
isolation of 33 (entry 3, Table 1) as the major adduct, which
was characterized as follows: Rf = 0.55 in 25% EtOAc–hex-
anes. IR (neat) (cm–1): 3504, 3071, 3046, 2955, 2932, 2857,
1
1639, 1428, 1113. H NMR (400 MHz, CDCl3) δ: 7.70–7.65
(m, 4H), 7.46–7.38 (m, 6H), 4.91–4.86 (m, 2H), 4.07–3.99
(m, 2H), 3.74–3.65 (m, 2H), 2.90–2.76 (m, 4H), 2.47 (br s,
1H), 2.20–2.06 (m, 4H), 1.91–1.61 (m, 5H), 1.61–1.51 (m,
1H), 1.07 (s, 9H), 1.04 (d, J = 7.2 Hz, 3H), 0.94 (d, J =
6.8 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ: 150.8, 135.6,
135.5, 133.3, 133.1, 129.7, 129.7, 127.6, 110.6, 70.9, 67.9,
47.7, 39.5, 39.2, 33.2, 32.2, 30.4, 26.8, 26.0, 20.1, 19.1,
10.3. HR-MS m/e calcd. for C31H44O1S2Si ([M – H2O]+):
524.2603; found: 524.2600.
(4R,7S)-1-(tert-Butyldiphenylsilanyloxy)-9-([1,3]dithian-
2-yl)-7-methyl-6-methylene-nonan-4-ol (30)
Application of the allylation procedure above led to the
isolation of 30 (entry 1, Table 1) as the major adduct, which
was characterized as follows: Rf = 0.54 in 10% EtOAc–hex-
anes. IR (neat) (cm–1): 3448, 3071, 3045, 2930, 2856, 2850,
1
1111. H NMR (400 MHz, CDCl3) δ: 7.70–7.65 (m, 4H),
7.45–7.36 (m, 6H), 4.92 (s, 1H), 4.88 (s, 1H), 4.02 (t, J =
6.4 Hz, 1H), 3.80–3.68 (m, 3H), 2.90–2.80 (m, 4H), 2.22
(dd, J = 3.8, 14.4 Hz, 1H), 2.17 (br s, 1H), 2.15–2.04 (m,
3H), 1.91–1.80 (m, 1H), 1.76–1.48 (m, 8H), 1.08–1.02 (m,
12H). 13C NMR (101 MHz, CDCl3) δ: 150.8, 135.5, 133.8,
129.6, 127.5, 111.1, 69.0, 64.1, 47.7, 42.8, 38.9, 33.8, 33.3,
32.4, 30.4, 28.8, 26.8, 26.0, 19.7, 19.1. HR-MS m/e calcd.
for C31H44OS2Si ([M – H2O]+): 524.2603; found: 524.2598.
(2S,3R,6S)-1-(tert-Butyldiphenylsilanyloxy)-8-([1,3]dithian-
2-yl)-2,6-dimethyl-5-methylene-octan-3-ol (35)
Application of the allylation procedure above led to the
isolation of 35 (entry 4, Table 1) as the major adduct, which
was characterized as follows: Rf = 0.55 in 25% EtOAc–hex-
anes. IR (neat) (cm–1): 3506, 3071, 3048, 2958, 2931, 2857,
1
1639, 1427, 1112. H NMR (400 MHz, CDCl3) δ: 7.71–7.66
(m, 4H), 7.46–7.36 (m, 6H), 4.91 (s, 1H), 4.89 (s, 1H), 4.02
(t, J = 6.4 Hz, 1H), 3.84–3.76 (m, 1H), 3.79 (dd, J = 4.8,
10.0 Hz, 1H), 3.68 (dd, J = 6.2, 10.0 Hz, 1H), 2.90–2.76 (m,
4H), 2.31 (dd, J = 2.4, 14.8 Hz, 1H), 2.20–2.05 (m, 3H),
1.90–1.62 (m, 5 H), 1.61–1.51 (m, 1H), 1.07 (s, 9H), 1.04
(d, J = 6.8 Hz, 3H), 0.96 (d, J = 7.2 Hz, 3H). 13C NMR
(101 MHz, CDCl3) δ: 151.0, 135.6, 133.2, 130.0, 127.7,
110.6, 72.3, 67.1, 47.7, 40.2, 39.7, 39.3, 33.3, 32.1, 30.4,
26.8, 26.0, 20.2, 19.2, 13.6. HR-MS m/e calcd. for
C31H44O1S2Si ([M – H2O]+): 524.2603; found: 524.2609.
(3S,6S)-6-Benzoyloxy-9-(tert-butyldiphenylsiloxy)-5-
methylene-1-phenyl-nonan-3-ol (31)
The allylation procedure above led to alcohol 31 (entry 2,
Table 1) as a single diastereoisomer, characterized as fol-
lows: Rf = 0.40 in 30% EtOAc–hexanes. [α]D +17.6 (c 3.95,
CHCl3). IR (neat) (cm–1): 3493, 3070, 2930, 2856, 1711,
1
1502, 1273, 1072. H NMR (400 MHz, CDCl3) δ: 8.04 (m,
2H), 7.75–7.15 (m, 18H), 5.32 (dd, J = 8.0, 4.5 Hz, 1H),
5.14 (s, 1H), 4.98 (s, 1H), X of ABX (δ: 3.81, m, 1H), 3.71
(m, 2H), 3.24 (br s, 1H), 2.85 (m, 1H), 2.70 (m, 1H), AB of
ABX (δA: 2.31, δB: 2.03, JAB = 13.8 Hz, JAX = 5.7 Hz, JBX
=
(2S,3S,6R,9S)-6-Benzoyloxy-9-(tert-butyldimethylsilyloxy)-
1-(tert-butyldiphenylsilyloxy)-2-methyl-5-methylene-
tridec-11-(E)-en-3-ol (37)
6.1 Hz, 2H), 2.00–1.60 (m, 6H), 1.06 (s, 9H). 13C NMR
(101 MHz, CDCl3) δ: 166.5, 146.1, 142.3, 135.5, 134.7,
133.9, 133.1, 130.1, 129.6, 128.4, 128.3, 127.7, 127.6,
125.7, 113.2, 77.9, 71.6, 63.4, 41.5, 39.0, 32.1, 29.9, 28.7,
26.8, 19.2. HR-MS m/z calcd. for C39H46O4Si ([M]+):
606.3165; found: 606.3119.
The general allylation procedure described above led to
the isolation of 37 (entry 5; Table 1) as a single
diastereoisomer, as characterized by the following data: Rf =
0.20 in 5% EtOAc–hexanes. [α]D –24.1 (c 1.10, CHCl3). IR
(neat) (cm–1): 3503 (br), 3067, 2933, 2860, 1714, 1464,
1
(3R,6S)-6-Benzoyloxy-9-(tert-butyldiphenylsiloxy)-5-
methylene-1-phenyl-nonan-3-ol (32)
1273, 1105, 706. H NMR (400 MHz, CDCl3) δ: 8.15 (m,
2H), 7.68 (m, 4H), 7.56 (m, 1H), 7.41 (m, 8H), 5.42 (m,
2H), 5.33 (m, 1H), 5.15 (s, 1H), 5.01 (s, 1H), 4.10 (m, 1H),
3.70 (m, 3H), 3.36 (d, J = 2.4 Hz, 1H), 2.29 (dd, A of ABX,
JAB = 14.2 Hz, JAX = 4.1 Hz, 1H), 2.22 (dd, B of ABX,
JBA = 14.2 Hz, JBX = 9.2 Hz, 1H), 2.14 (app t, J = 5.7 Hz,
2H), 1.90 (m, 1H), 1.79 (m, 1H), 1.63 (d, J = 5.1 Hz, 3H),
1.62 (m, 1H), 1.48 (m, 1H), 1.06 (s, 9H), 0.96 (d, J =
7.0 Hz, 3H), 0.88 (s, 9H), 0.04 (s, 3H), 0.03 (s, 3H). 13C
NMR (101 MHz, CDCl3) δ: 166.1, 146.0, 135.7, 135.6,
133.0, 130.3, 129.7, 129.5, 128.3, 127.7, 127.5, 127.4,
112.8, 77.7, 73.0, 71.9, 68.0, 40.4, 39.5, 38.2, 32.5, 29.1,
26.9, 25.9, 19.2, 18.0, 10.5, –4.4, –4.6. HR-MS (FAB, NBA)
Application of the allylation procedure above led to com-
pound 32 (entry 2, Table 1), which was characterized as fol-
lows: Rf = 0.40 in 30% EtOAc–hexanes. [α]D +17.2 (c 3.67,
CHCl3). IR (neat) (cm–1): 3491 (br), 3070, 2937, 1961,
1
1888, 1822, 1712, 1602, 1471. H NMR (400 MHz, CDCl3)
δ: 8.05 (m, 2H), 7.75–7.15 (m, 18H), 5.23 (dd, J = 8.0,
3.9 Hz, 1H), 5.13 (s, 1H), 4.98 (s, 1H), X of ABX (δ:
4.00, m, 1H), 3.71 (m, 2H), 2.88 (m, 1H), 2.83 (br s, 1H),
2.70 (m, 1H), AB of ABX (δA: 2.30, δB: 2.22, JAB = 13.8 Hz,
JAX = 5.7 Hz, JBX = 6.1 Hz, 2H), 2.00–1.60 (m, 6H), 1.06 (s,
9H). 13C NMR (101 MHz, CDCl3) δ: 166.4, 144.9, 142.3,
© 2004 NRC Canada

128
Can. J. Chem. Vol. 82, 2004
m/z calcd. for C44H65O5Si2 ([M + H]+): 729.4371; found:
729.4363.
72.7, 70.0, 66.5, 62.1, 55.2, 42.5, 42.3, 41.1, 40.6, 38.4,
36.3, 36.1, 26.9, 19.8, 19.2. HR-MS m/e calcd. for
C37H47O5Si ([M – t-Bu]+): 599.3193; found: 599.3193.
(2S,3R,6S)-6-Benzoyloxy-1,9-bis-(tert-butyldiphenyl-
silyloxy)-2-methyl-5-methylene-nonan-3-ol (38)
(2S,5R)-7-(tert-Butyldiphenylsilanyloxy)-1-{(2S,6R)-6-[2-
(4-methoxy-benzyloxy)-ethyl]-4-methylene-tetrahydro-
pyran-2-yl}-5-methyl-4-methylene-heptan-2-ol (43)
Application of the allylation procedure above led to the
isolation of 43 (entry 9, Table 1) as the major adduct, which
was characterized as follows: Rf = 0.47 in 25% EtOAc–hex-
anes. IR (neat) (cm–1): 3477, 3071, 2932, 2857, 1646, 1603,
The allylation procedure above led to the isolation of 38
(entry 6, Table 1) as a single diastereoisomer, characterized
as follows: Rf = 0.45 in 50% EtOAc–hexanes. [α]D +22.3 (c
2.80, CHCl3). IR (neat) (cm–1): 3504 (br), 3070, 2957, 2930,
1
2856, 1709, 1427, 1273, 1111, 702. H NMR (400 MHz,
CDCl3) δ: 8.05 (d, J = 7.2 Hz, 2H), 7.74–7.35 (m, 23H),
5.41 (dd, J = 7.5, 4.8 Hz, 1H), 5.18 (9s, 1H), 5.06 (s, 1H),
3.86 (m, 1H), 3.72 (m, 4H), 3.63 (br s, 1H), 2.44 (dd, A of
ABX, JAB = 14.2 Hz, JAX = 1.9 Hz, 1H), 2.14 (dd, B of
ABX, JBA = 14.2 Hz, JBX = 10.0 Hz, 1H), 1.90 (m, 3H),
1.70 (m, 2H), 1.07 (s, 9H), 0.98 (d, J = 7.0 Hz, 3H). 13C
NMR (101 MHz, CDCl3) δ: 166.2, 146.0, 135.6, 135.5,
133.9, 133.8, 133.4, 133.0, 130.3, 129.6, 129.5, 129.5′,
128.4, 127.7, 127.6, 113.2, 77.7, 74.4, 66.9, 63.5, 40.5, 38.3,
29.8, 28.7, 26.8, 19.1, 13.7. HR-MS (FAB, NBA, Na+) m/z
calcd. for C50H62O5Si2Na ([M + Na]+): 821.4034; found:
821.4039.
1
1513, 1425, 1248, 1111. H NMR (400 MHz, CDCl3) δ:
7.70–7.68 (m, 4H), 7.45–7.35 (m, 6H), 7.26 (d, J = 8.8 Hz,
2H), 6.89 (d, J = 8.8 Hz, 2H), 4.85 (s, 1H), 4.81 (s, 1H),
4.72 (app s, 2H), 4.44–4.37 (m, 2H), 4.06 (m 1H), 3.80 (s,
3H), 3.72–3.65 (m, 2H), 3.63–3.45 (m, 4H), 2.80 (br s, 1H),
2.36 (ddt, J = 6.8, 6.8, 6.8 Hz, 1H), 2.25–1.52 (m, 12H),
1.07 (s, 9H), 1.00 (d, J = 6.8 Hz, 3H). 13C NMR (101 MHz,
CDCl3) δ: 159.1, 151.2, 144.3, 135.5, 134.0, 130.6, 129.5,
129.2, 127.6, 113.8, 110.3, 108.7, 76.0, 75.8, 72.6, 66.6,
66.4, 62.1, 55.2, 42.5, 42.1, 40.7, 40.5, 38.2, 36.4, 36.1,
26.9, 20.0, 19.2. HR-MS m/e calcd. for C37H47O5Si ([M – t-
Bu]+): 599.3193; found: 599.3201.
(2S,3R,6S)-6-Benzoyloxy-2,9-bis-(tert-butyldiphenylsilyl-
oxy)-5-methylene-nonan-3-ol (39)
(2R,5S)-7-([1,3]Dithian-2-yl)-1-{(2S,6R)-6-[2-(4-methoxy-
benzyloxy)-ethyl]-4-methylene-tetrahydro-pyran-2-yl}-5-
methyl-4-methylene-heptan-2-ol (44)
The allylation procedure above led to the isolation of 39
(entry 7, Table 1) as a single diastereoisomer, characterized
as follows: Rf = 0.45 in 10% EtOAc–hexanes. [α]D +9.5 (c
1.84, CHCl3). IR (neat) (cm–1): 3489 (br), 3070, 2932, 2858,
1718, 1427, 1271, 1109, 702. 1H NMR (400 MHz, CDCl3) δ:
8.04 (d, J = 7.2 Hz, 2H), 7.68 (m, 8H), 7.56 (m, 1H), 7.40
(m, 14H), 5.39 (t, J = 6.4 Hz, 1H), 5.13 (s, 1H), 4.94 (s,
1H), 3.86 (m, 1H), 3.78 (m, 1H), 3.69 (app t, J = 6.1 Hz,
2H), 2.77 (br s, 1H), 2.34 (dd, A of ABX, JAB = 14.6 Hz,
JAX = 3.0 Hz, 1H), 2.08 (dd, B of ABX, JBA = 14.6 Hz,
JBX = 9.6 Hz, 1H), 1.86 (m, 2H), 1.64 (m, 2H), 1.08 (s, 9H),
1.06 (s, 9H). 13C NMR (101 MHz, CDCl3) δ: 166.0, 145.1,
135.8, 135.5, 134.3, 133.9, 133.6, 132.9, 130.3, 129.7,
129.6, 128.4, 127.7, 127.5, 113.4, 77.6, 74.7, 72.4, 63.5,
35.5, 29.6, 28.6, 27.0, 26.9, 19.3, 19.2, 17.7. MS (CI, NH3)
m/z (relative intensity) 606 (25), 349 (42),302 (47), 235 (56).
HR-MS m/z calcd. for C45H51O5Si2 ([M – t-Bu]+): 727.3295;
found: 727.3252.
The same procedure for allylation was applied here and
led to the isolation of 44 (entry 10, Table 1): Rf = 0.32 in
25% EtOAc–hexanes. IR (neat) (cm–1): 3480, 3070, 2935,
1
1650, 1607, 1512, 1247, 1095. H NMR (400 MHz, CDCl3)
δ: 7.25 (d, J = 8.6 Hz, 2H), 6.88 (d, J = 8.6 Hz, 2H), 4.86 (s,
1H), 4.83 (s, 1H), 4.71 (app s, 2H), 4.45–4.39 (m, 2H),
4.04–3.98 (m, 2H), 3.80 (s, 3H), 3.58–3.50 (m, 4H), 2.90–
2.78 (m, 4H), 2.27–2.05 (m, 6H), 2.02–1.92 (m, 2H), 1.89–
1.50 (m, 10H), 1.03 (d, J = 6.8 Hz, 3H). 13C NMR
(101 MHz, CDCl3) δ: 159.1, 150.5, 143.7, 130.5, 129.3,
113.7, 110.7, 109.0, 79.4, 76.0, 72.7, 70.1, 66.4, 55.2, 47.8,
42.3, 42.1, 41.0, 40.5, 39.4, 36.2, 33.1, 32.3, 30.4, 26.0,
20.0. HR-MS m/e calcd. for C29H44O4S2Na ([M + Na]+):
543.2591; found: 543.2581.
2,2-Dimethylpropionic acid (4R,5R,8S)-(E)-4,10-bis-(tert-
butyldiphenylsilanyloxy)-5-hydroxy-2,8-dimethyl-7-
methylene-dec-2-enyl ester (46)
(2R,5S)-7-(tert-Butyldiphenylsilanyloxy)-1-{(2S,6R)-6-[2-
(4-methoxy-benzyloxy)-ethyl]-4-methylene-tetrahydro-
pyran-2-yl}-5-methyl-4-methylene-heptan-2-ol (40)
Application of the allylation procedure above led to the
isolation of 40 (entry 8, Table 1) as the major adduct, which
was characterized as follows: Rf = 0.45 in 25% EtOAc–hex-
anes. IR (neat) (cm–1): 3496, 3070, 2932, 2857, 1649, 1610,
Application of the allylation procedure above led to the
isolation of 46 (entry 11, Table 1) as the major adduct,
which was characterized as follows: Rf = 0.68 in 33%
EtOAc–hexanes. IR (neat) (cm–1): 3445, 3071, 3049, 2959,
1
2931, 2857, 1731, 1428, 1282, 1112. H NMR (400 MHz,
1
1513, 1427, 1248, 1111. H NMR (400 MHz, CDCl3) δ:
CDCl3) δ: 7.73–7.63 (m, 8H), 7.45–7.30 (m, 12H), 5.38 (d,
J = 9.6 Hz, 1H), 4.81 (s, 1H), 4.80 (s, 1H), 4.27 (dd, J = 6.4,
9.2 Hz, 1H), 4.14 (s, 2H), 3.73–3.60 (m, 3H), 2.54 (d, J =
3.6 Hz, 1H), 2.33 (ddt, J = 6.8, 6.8, 6.8 Hz, 1H), 2.16 (m,
1H), 1.94 (dd, J = 9.6, 14.8 Hz, 1H), 1.79–1.70 (m, 1H),
1.56–1.46 (m, 1H), 1.15 (s, 9H), 1.07–1.02 (m, 21 H), 0.94
(d, J = 6.8 Hz, 3H). 13C NMR (101 MHz, CDCl3) δ: 178.0,
150.7, 135.9, 135.8, 135.5, 134.0, 133.6, 129.8, 129.6,
129.5, 127.7, 127.6, 127.4, 126.0, 72.9, 72.6, 68.8, 62.1,
38.8, 38.1, 36.7, 35.8, 27.2, 27.0, 26.9, 19.9, 19.3, 19.2,
7.70–7.65 (m, 4H), 7.44–7.35 (m, 6H), 7.25 (d, J = 8.6 Hz,
2H), 6.87 (d, J = 8.6 Hz, 2H), 4.83 (s, 1H), 4.80 (s, 1H),
4.71 (app s, 2H), 4.45–4.39 (m, 2H), 4.00 (m, 1H), 3.80 (s,
3H), 3.71–3.63 (m, 2H), 3.67 (br s, 1H), 3.55–3.48 (m, 4H),
2.35 (ddt, J = 6.8, 6.8, 6.8 Hz, 1H), 2.27–2.14 (m, 3H), 2.08
(dd, J = 6.4, 14.8 Hz, 1H), 2.00–1.93 (m, 2H), 1.88–1.48 (m,
6H), 1.05 (s, 9H), 0.99 (d, J = 6.8 Hz, 3H). 13C NMR
(101 MHz, CDCl3) δ: 159.1, 151.3, 143.8, 135.6, 134.7,
130.6, 129.5, 129.3, 127.6, 113.8, 110.1, 108.9, 79.2, 76.1,
© 2004 NRC Canada

Williams et al.
129
14.2. HR-MS m/e calcd. for C46H59O5Si ([M – t-Bu]+):
747.3901; found: 747.3909.
(101 MHz, CDCl3) δ: 166.2, 149.9, 144.8, 133.0, 130.2,
129.6, 128.3, 127.5, 127.4, 113.7, 77.7, 71.8, 69.5, 67.9,
57.9, 56.2, 48.5, 40.5, 37.6, 32.4, 29.0, 27.4, 25.9, 18.1,
18.0, 13.6, 12.5, 9.6, –4.4, –4.5. HR-MS (FAB, NBA, Na+)
m/z calcd. for C52H94O6Si2Sn120Na ([M + Na]+): 1013.5509;
found: 1013.5549. Anal. calcd. for C52H94O6Si2Sn (%): C
63.08, H 9.57; found: C 62.93, H 9.42.
2,2-Dimethylpropionic acid (4R,5S,8R)-(E)-4,10-bis-(tert-
butyldiphenylsilanyloxy)-5-hydroxy-2,8-dimethyl-7-
methylene-dec-2-enyl ester (47)
Application of the allylation procedure above led to the
isolation of 47 (entry 12, Table 1) as the major adduct,
which was characterized as follows: Rf = 0.68 in 33%
EtOAc–hexanes. IR (neat) (cm–1): 3513, 3071, 3047, 2959,
2930, 2857, 1731, 1427, 1282, 1149, 1112. 1H NMR
(400 MHz, CDCl3) δ: 7.70–7.60 (m, 8H), 7.45–7.34 (m,
12H), 5.52 (d, J = 9.2 Hz, 1H), 4.74 (s, 1H), 4.66 (s, 1H),
4.32 (dd, J = 4.0, 9.6 Hz, 1H), 4.26 (s, 2H), 3.8 (m, 1H),
3.70–3.60 (m, 2H), 2.28 (br s, 1H), 2.28–2.10 (m, 2H), 2.02
(dd, J = 8.8, 15.2 Hz, 1H), 1.74–1.66 (m, 1H), 1.52–1.43 (m,
1H), 1.17 (s, 9H), 1.05–1.02 (m, 21 H), 0.90 (d, J = 6.8 Hz,
3H). 13C NMR (101 MHz, CDCl3) δ: 178.0, 151.1, 135.9,
135.9, 135.5, 133.4, 134.0, 133.4, 129.7, 129.6, 129.5,
127.6, 127.6, 127.4, 126.8, 73.6, 73.5, 68.5, 62.1, 38.8, 38.1,
37.3, 35.5, 27.2, 27.0, 26.9, 20.0, 19.4, 19.2, 14.1. HR-MS
m/e calcd. for C46H59O5Si ([M – t-Bu]+): 747.3901; found:
747.3897.
(4R,5S,6R,7S,10R,13S)-10-Benzoyloxy-13-(tert-butyldi-
methylsiloxy)-5,6-epoxy-9-methylene-2-tributylstannyl-4-
triisopropylsiloxy-heptadeca-1,15-(E)-dien-7-ol (50)
The allylation procedure above led to the isolation of 50
(entry 15, Table 1) as a single diastereoisomer, characterized
as follows: Rf = 0.15 in 5% EtOAc–hexanes. [α]D –25.5 (c
2.25, CHCl3). IR (neat) (cm–1): 3483 (br), 2928, 2666, 1720,
1
1462, 1271, 1113. H NMR (400 MHz, CDCl3) δ: 8.03 (m,
2H), 7.56 (m, 1H), 7.44 (m, 2H), 5.81 (m, 1H), 5.41 (m,
1H), 5.34 (dd, J = 7.8, 4.9 Hz, 1H), 5.25 (d, J = 2.7 Hz, 1H),
5.17 (s, 1H), 5.05 (s, 1H), 3.86 (m, 1H), 3.75–3.66 (m, 2H),
3.01 (dd, J = 3.9, 2.1 Hz, 1H), 2.94 (dd, J = 3.9, 2.1 Hz,
1H), 2.89 (d, J = 1.7 Hz, 1H), 2.59 (m, 2H), 2.39 (dd, J =
14.5, 2.2 Hz, 1H), 2.16–2.10 (m, 3H), 1.89 (m, 1H), 1.75
(m, 1H), 1.63 (d, J = 5.3 Hz, 3H), 1.61 (m, 1H), 1.52–1.40
(m, 7H), 1.31 (m, 6H), 1.12–1.02 (m, 21H), 0.98–0.84 (m,
2H), 0.04 (s, 3H), 0.03 (s, 3H). 13C NMR (101 MHz,
CDCl3) δ: 166.2, 149.8, 144.9, 133.0, 130.2, 129.6, 129.1,
128.3, 127.4, 127.3, 113.6, 77.8, 71.9, 71.7, 70.0, 59.4, 58.8,
47.3, 40.4, 37.1, 32.5, 29.1, 27.4, 25.9, 18.2, 17.9, 13.7,
12.6, 9.6, –4.4, –4.6.
2,2-Dimethylpropionic acid (4R,5R,8S)-(E)-4-(tert-
butyldiphenylsilanyloxy)-5-hydroxy-10-([1,3]dithian-2-
yl)-2,8-dimethyl-7-methylene-dec-2-enyl ester (48)
The same procedure for allylation was applied here and
led to the isolation of 48 (entry 13, Table 1): Rf = 0.59 in
33% EtOAc–hexanes. IR (neat) (cm–1): 3551, 3071, 2952,
2929, 2852, 1728, 1432, 1275, 1145, 1110. 1H NMR
(400 MHz, CDCl3) δ: 7.69–7.62 (m, 4H), 7.43–7.32 (m, 6H),
5.39 (d, J = 9.6 Hz, 1H), 4.85 (s, 1H), 4.84 (s, 1H), 4.29 (dd,
J = 6.2, 9.4 Hz, 1H), 4.16 (s, 2H), 4.00 (t, J = 6.8 Hz, 1H),
3.67 (m, 1H), 2.90–2.76 (m. 4H), 2.53 (d, J = 4 Hz, 1H),
2.21–2.05 (m, 3H), 1.95 (dd, J = 9.6, 15.2 Hz, 1H), 1.90–
1.80 (m, 1H), 1.76–1.60 (m, 2H), 1.55–1.46 (m, 1H), 1.16
(s, 9H), 1.07 (s, 3H), 1.05 (s, 9H), 0.98 (d, J = 6.8 Hz, 3H).
13C NMR (101 MHz, CDCl3) δ: 178.0, 150.5, 136.0, 136.0,
135.9, 134.1, 133.7, 133.5, 129.8, 129.7, 127.7, 127.4,
126.7, 73.8, 73.4, 68.5, 47.8, 39.2, 38.8, 37.0, 33.2, 32.1,
30.4, 27.1, 27.0, 26.0, 20.0, 19.4, 14.1. HR-MS m/e calcd.
for C34H47O4S2Si ([M – t-Bu]+): 611.2685; found: 611.2690.
(4S,5S,6R,7S,10R,13S)-10-Benzoyloxy-13-(tert-butyldi-
methylsiloxy)-5,6-epoxy-9-methylene-2-tributylstannyl-4-
triisopropylsiloxy-heptadeca-1,15-(E)-dien-7-ol (51)
The allylation procedure above led to the isolation of 51
(entry 16, Table 1) as a single diastereoisomer, characterized
as follows: Rf = 0.20 in 5% EtOAc–hexanes. [α]D –11.2 (c
1.35, CHCl3). IR (neat) (cm–1): 3475 (br), 2955, 2928, 2866,
1722, 1462, 1271, 1116. 1H NMR (400 MHz, CDCl3) δ: 8.04
(m, 2H), 7.57 (m, 1H), 7.44 (m, 2H), 5.81 (s, 1H), 5.42 (m,
2H), 5.35 (m, 1H), 5.24 (d, J = 2.2 Hz, 1H), 5.18 (s, 1H),
5.06 (s, 1H), 4.12 (m, 1H), 3.90 (m, 1H), 3.69 (dddd, J =
5.9, 5.9, 5.7, 5.3 Hz, 1H), 3.15 (dd, J = 4.0, 2.1 Hz, 1H),
3.05 (dd, J = 2.4, 2.2 Hz, 1H), 2.87 (d, J = 1.9 Hz, 1H), 2.71
(dd, J = 13.8, 4.3 Hz, 1H), 2.46 (dd, J = 19.0, 9.8 Hz, 1H),
2.44 (dd, J = 14.2, 3.1 Hz, 1H), 2.18 (dd, J = 14.2, 9.5 Hz,
1H), 2.14 (dd, J = 5.7, 5.7 Hz, 2H), 1.90 (m, 1H), 1.76 (m,
1H), 1.63 (d, J = 5.2 Hz, 3H), 1.61 (m, 1H), 1.50 (m, 7H),
1.30 (m, 6H), 1.24 (m, 21H), 0.90 (m, 24H), 0.30 (s, 6H).
13C NMR (101 MHz, CDCl3) δ: 166.2, 149.9, 144.8, 133.0,
130.2, 129.6, 128.3, 127.4, 127.3, 113.7, 77.7, 77.2, 71.9,
69.5, 67.9, 57.9, 56.2, 48.5, 40.4, 37.5, 32.4, 29.0, 27.4,
25.8, 18.0, 17.9, 13.6, 12.5, 9.6, –4.4, –4.5. MS (FAB, NBA,
Na+) m/z ([M – t-Bu]+): 933.
(4R,5R,6S,7R,10S,13S)-10-Benzoyloxy-13-(tert-butyl-
dimethylsiloxy)-5,6-epoxy-9-methylene-2-tributylstannyl-
4-triisopropylsiloxy-heptadeca-1,15-(E)-dien-7-ol (49)
The allylation procedure above led to the isolation of 49
(entry 14, Table 1) as a single diastereoisomer, characterized
as follows: Rf = 0.20 in 5% EtOAc–hexanes. [α]D +6.69 (c
1.44, CHCl3). IR (neat) (cm–1): 3489 (br), 2955, 2928, 2854,
1
1720, 1462, 1271, 1113, 1068. H NMR (400 MHz, CDCl3)
δ: 8.04 (d, J = 7.2 Hz, 2H), 7.57 (m, 1H), 7.44 (m, 2H), 5.81
(s, 1H), 5.50–5.30 (m, 3H), 5.23 (d, J = 2.2 Hz, 1H), 5.18 (s,
1H), 5.06 (s, 1H), 4.12 (m, 1H), 3.90 (m, 1H), 3.69 (dddd,
J = 5.9, 5.9, 5.7, 5.3 Hz, 1H), 3.15 (dd, J = 4.0, 2.0 Hz, 1H),
3.05 (dd, J = 2.2, 2.1 Hz, 1H), 2.85 (d, J = 2.0 Hz, 1H), 2.71
(dd, J = 13.7, 4.2 Hz, 1H), 2.50–2.40 (m, 2H), 2.18 (dd, J =
14.2, 9.4 Hz, 1H), 2.14 (m, 2H), 1.84 (m, 2H), 1.62 (d, J =
5.9 Hz, 3H), 1.59–1.44 (m, 8H), 1.32 (m, 6H), 1.06 (m,
21H), 0.94–0.86 (m, 24H), 0.04 (s, 6H). 13C NMR
Conclusion
In summary, a powerfully convergent strategy for the
diastereoselective synthesis of homoallylic alcohols has been
described. Our asymmetric allylation reaction incorporates
three elements of stereodifferentiation. Optimal results are
provided when adjacent asymmetry in the starting stannane
© 2004 NRC Canada

130
Can. J. Chem. Vol. 82, 2004
9. D.R. Williams, D.A. Brooks, K.G. Meyer, and M.P. Clark. Tet-
rahedron Lett. 39, 7251 (1998).
10. Y. Nishigaichi, H. Kuramoto, and A. Takuwa. Tetrahedron
is stereochemically reinforcing with respect to chirality of
the bis-sulfonamide auxiliary. Products featuring 1,4-syn-
stereochemistry are favored. The incorporation of α- or β-
stereogenecity in the starting aldehyde has been docu-
mented. The diastereofacial selectivity of reactions with α-
substituted aldehydes is predicted by the Felkin–Anh model.
Overall, this methodology permits the construction of com-
plex arrangements of stereochemistry and functionality as a
useful tool for the synthesis of complex natural products.
Lett. 36, 3353 (1995).
11. R.J. Maguire, J. Mulzer, and J.W. Bats. J. Org. Chem. 61, 6936
(1996); (b) R.J. Maguire, J. Mulzer, and J.W. Bats. Tetrahe-
dron Lett. 37, 5487 (1996).
12. D.R. Williams, W.S. Kissel, and J.J. Li. Tetrahedron Lett. 39,
8593 (1998).
13. B.M. Trost, T.A. Grese, and D.M.T. Chan. J. Am. Chem. Soc.
113, 7350 (1991).
14. E. Piers, J.M. Chong, K. Gustafson, and R.J. Andersen. Can. J.
Chem. 62, 1 (1984).
15. Sharpless asymmetric epoxidations proceeded with >97% de;
see: Y. Gao, R.M. Hanson, J.M. Klunder, S.Y. Ko, H.
Masamune, and K.B. Sharpless. J. Am. Chem. Soc. 109, 5765
(1987).
Acknowledgements
We gratefully acknowledge the National Institutes of
Health (GM-41560 and GM-42897) for generous support of
this work.
16. S.D. Burke, J.L. Buchanan, and J.D. Rovin. Tetrahedron Lett.
32, 3961 (1991).
References
1. For a general review: Y. Yamamoto and N. Asao. Chem. Rev.
93, 2207 (1993).
17. B.M. Trost and S.A. King. J. Am. Chem. Soc. 112, 408 (1990).
18. (a) I. Ohtani, T. Kusumi, Y. Kashman, and H. Kakisawa. J.
Am. Chem. Soc. 113, 4092 (1991); (b) J.A. Dale and H.S.
Mosher. J. Am. Chem. Soc. 95, 512 (1973).
2. (a) W.R. Roush. Comprehensive organic synthesis. Vol. 2.
Edited by B.A. Trost and I. Fleming. Pergamon Press, Oxford.
1991; (b) P.V. Ramachandran. Aldrichimica Acta, 35, 23 (2002).
3. S.R. Chemler and W.R. Roush. Recent applications of the
allylation reaction to the synthesis of natural products. In
Modern carbonyl chemistry. Edited by J. Otera. Wiley-VCH,
Weinheim, Germany. 2000.
19. D.A. Evans, M.J. Dart, J.L. Duffy, and M.G. Yang. J. Am.
Chem. Soc. 118, 4322 (1996).
20. For a related discussion: W.R. Roush and L. Banfi. J. Am.
Chem. Soc. 110, 3979 (1988).
21. R. W. Hoffmann. Chem. Rev. 89, 1841 (1989).
22. For a leading reference: J. Haller, T. Strassner, and K.N.
Houk. J. Am. Chem. Soc. 119, 8031 (1997).
4. E.J. Corey, C.-M. Yu, and S.S. Kim. J. Am. Chem. Soc. 111,
5495 (1989).
5. E.J. Corey, C.-M. Yu, and D.-H. Lee. J. Am. Chem. Soc. 112,
878 (1990).
6. D.R. Williams, D.A. Brooks, and M.A. Berliner. J. Am. Chem.
Soc. 121, 4924 (1999).
7. D.R. Williams and K.G. Meyer. J. Am. Chem. Soc. 123, 765
(2001).
8. D.R. Williams, A.A. Kiryanov, U. Emde, M.P. Clark, M.A.
Berliner, and J.T. Reeves. Angew. Chem. Int. Ed. 42, 1258
(2003).
23. (a) D.A. Evans, H.P. Ng, J.S. Clark, and D.L. Rieger. Tetrahe-
dron, 48, 2127 (1992); (b) I. Paterson. Pure Appl. Chem. 64,
1821 (1992); (c) D.A. Evans and M.A. Calter. Tetrahedron
Lett. 34, 6871 (1993); (d) I. Paterson, D.J. Wallace, and S.M.
Velázquez. Tetrahedron Lett. 35, 9083 (1994).
24. D.R. Williams, S.V. Plummer, and S. Patnaik. Angew. Chem.
Int. Ed. 42, 3934 (2003).
25. D. R. Williams, M. P. Clark, U. Emde, and M. A. Berliner.
Org. Lett. 2, 3023 (2000).
© 2004 NRC Canada