Highly diastereoselective synthesis of propargylic 1,2-anti-diol derivatives using alpha-alkoxypropargylstannanes.

Article abstract of DOI:10.1021/ol016536m

Propargylic 1,2-anti-diol derivatives 2 and 10 are prepared in high yield and excellent diastereoselectivity by addition of alpha-alkoxypropargylstannanes 4a and 4b to aldehydes in the presence of BuSnCl(3). We also introduce the use of KF on Celite as a

Full text of DOI:10.1021/ol016536m

ORGANIC
LETTERS
2001
Vol. 3, No. 19
3057-3060
Highly Diastereoselective Synthesis of
Propargylic 1,2-anti-Diol Derivatives
Using r-Alkoxypropargylstannanes
Brad M. Savall, Noel A. Powell, and William R. Roush*
Department of Chemistry, UniVersity of Michigan, Ann Arbor, Michigan 48109
roush@umich.edu
Received August 3, 2001
ABSTRACT
Propargylic 1,2-anti-diol derivatives 2 and 10 are prepared in high yield and excellent diastereoselectivity by addition of r-alkoxypropargyl-
stannanes 4a and 4b to aldehydes in the presence of BuSnCl3. We also introduce the use of KF on Celite as a convenient and mild reagent
for removal of the organotin waste products of these reactions.
In connection with an ongoing problem in natural products
synthesis, we needed to accomplish the anti-γ-methoxy-
propargylation of an aldehyde (Figure 1). The reactions of
aldehydes with allenylmetal reagents have been extensively
developed for the synthesis of homopropargylic alcohols.1-3
Especially noteworthy are Marshall’s elegant studies and
synthetic applications of chiral allenyltin reagents.1,2,4,5
However, a general, highly diastereoselective procedure for
synthesis of anti-propargylic diol derivatives such as 2 is
currently unavailable. Epsztein demonstrated in the 1970s
that γ-alkoxyallenyl zinc reagents give the targeted propar-
gylic anti diol monoethers with moderate selectivity (e4:
1),6 and more recently Yamamoto reported that γ-alkoxy-
allenyltitanium reagents give the anti diol derivatives with
88:12 to 95:5 selectivity.7 However, the selectivity of this
process was less than what we hoped to achieve in our
projected total synthesis. Recognizing the facility with which
propargyl stannanes isomerize to allenylstannanes under
Lewis acidic conditions,5,8 we anticipated that propargyl
stannane 4a might serve as a suitable precursor to the
γ-alkoxyallenylstannane 5a needed for synthesis of 2. Indeed,
Yamamoto has demonstrated that the R-alkoxypropargyl-
stannane 6 cyclizes upon treatment with SnCl4 to give the
anti-â-hydroxypropargyl ether 8 with excellent selectivity,
(1) Chemler, S. R.; Roush, W. R. In Modern Carbonyl Chemistry; Otera,
J., Ed.; Wiley-VCH: Weinheim, 2000; p 403.
(2) Marshall, J. A. Chem. ReV. 1996, 96, 31.
(3) Yamamoto, H. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Ed.; Pergamon Press: Oxford, 1991; Vol. 2, p 81.
(4) Marshall, J. A.; Perkins, J. F.; Wolf, M. A. J. Org. Chem. 1995, 60,
5556.
(5) Marshall, J. A.; Yu, R. H.; Perkins, J. J. Org. Chem. 1995, 60, 5550.
(6) Mercier, F.; Epsztein, R.; Holand, S. Bull. Soc. Chem. Fr. 1972, 690.
(7) Ishaguro, M.; Ikeda, N.; Yamamoto, H. J. Org. Chem. 1982, 47, 2225.
(8) Guillerm, G.; Meganem, F.; LeQuan, M.; Brower, K. B. J. Organo-
met. Chem. 1974, 67, 43.
Figure 1.
10.1021/ol016536m CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/28/2001

Table 1. Reaction of R-Methoxy Propargylstannane 4a and Isobutyraldehyde
entry
equiv 4a
Lewis acid (equiv)
methoda
temp
workup
yieldb (%)
selectivity 2a :3a c
90-94:10-6
1
2
3
4
5
6
7
8
9
1.0
1.0
1.0
1.0
1.0
1.0
1.2
1.2
1.3
1.5
2.0
SnCl4 (1.0)
SnCl4 (1.0)
SnCl4 (1.0)
A
B
C
A
B
A
A
A
A
A
A
-78 f 0 °C
-78 °C
Et3N or KF(aq)
Et3N
NA
Et3N
KF(aq)
KF/Celite
KF/Celite
KF(aq)
KF/Celite
KF/Celite
KF/Celite
25-66
0
0
36
56
71d
93-96
79
85d
83d
93e
-78 f 0 °C
-78 f 0 °C
-78 f 0 °C
-78 f 0 °C
-78 f 0 °C
-78 f 0 °C
-78 f 0 °C
-78 f 0 °C
-78 f 0 °C
BuSnCl3 (1.0)
BuSnCl3 (1.0)
BuSnCl3 (1.0)
BuSnCl3 (1.2)
BuSnCl3 (1.2)
BuSnCl3 (1.3)
BuSnCl3 (1.5)
BuSnCl3 (2.0)
nd
nd
nd
97:3
nd
nd
nd
97:3
10
11
a Method A: A solution of the Lewis acid (1.0 M in CH2Cl2) was added dropwise to a -78 °C mixture of isobutyraldehyde and 4a in CH2Cl2. Method
B: A solution of the Lewis acid (1.0 M in CH2Cl2) was added dropwise to a -78 °C solution of 4a in CH2Cl2 followed by addition of isobutyraldehyde.
Method C: A solution of aldehyde and 4a was added to a cooled solution of SnCl4. b Combined yield of products after silica gel chromatography, unless
1
noted otherwise. c Product ratios determined by H NMR analysis of the crude product. d Yield of HPLC purified 2a. e Isolated yield of 2a.
presumably by way of the allenyltrichlorostannane interme-
diate 7 (Figure 2).9 We report herein our studies of the
γ-alkoxypropargylation of aldehydes by use of the in situ
generated allenylstannane reagents 5a and 5b (X ) Bu) and
demonstrate that the targeted anti propargylic diol mono-
ethers can now be prepared consistently with outstanding
diastereoselectivity (17:1 to g50:1 ds) with a range of
aldehyde substrates.
and with only trace amounts of the allenylstannane regio-
isomer. Although 4a has been previously synthesized by
treatment of the propargyllithium intermediate with Bu3-
SnCl,11 we were not able to obtain isomerically pure
propargylstannane by using this procedure.12
The reaction of 4a and isobutyraldehyde was studied in
some detail to define conditions for the R-alkoxypropargyl-
ation reaction (Table 1). Initial experiments were performed
by addition of SnCl4 (1 equiv) to a mixture of 4a and
isobutyraldehyde at -78 °C (entry 1). Although this reaction
displayed good selectivity for the anti diastereomer 2a,13 the
isolated yields were moderate and irreproducible. The order
of addition of the reagents also proved crucial: if SnCl4 was
added to 4a followed by the aldehyde, or to a mixture of 4a
and aldehyde, 2a and 3a were not obtained (entries 2 and
3). Since the reaction mixtures discolored and a precipatate
formed upon addition of the SnCl4 to 4a, we reasoned that
propargylstannane 4a is not stable to SnCl4. However, when
we switched to BuSnCl314 as the Lewis acid, the yields and
selectivity were greatly improved (entries 4-11).
Although we were encouraged by these initial results
(entries 4 and 5), separation of the major product 2a from
the organostannane byproducts proved difficult, and the
Figure 2.
R-Alkoxypropargylstannanes 4a and 4b were prepared by
deprotonation of propargyl ethers 9a and 9b with t-BuLi
followed by transmetalation with zinc chloride according to
Zweifel’s procedure (see Figure 3).10 The intermediate
allenylzinc species were then treated with Bu3SnCl to give
the targeted propargylstananne reagents in 82-84% yield
(9) Kadota, I.; Hatakeyama, D.; Seki, K.; Yamamoto, Y. Tetrahedron
Lett. 1996, 37, 3059.
(10) Zweifel, G.; Hahn, G. J. Org. Chem. 1984, 49, 4565.
(11) Anies, C.; Lallemand, J.-Y.; Pancrazi, A. Tetrahedron Lett. 1996,
37, 5519.
Figure 3.
3058
Org. Lett., Vol. 3, No. 19, 2001

yields sometimes varied. Of the several reported methods
for removing tin wastes from reactions,4,15-20 we tried treating
the ether extracts with Et3N4 and stirring the extracts with
aqueous KF.15,16 Although these methods were generally
successful, the large amount of precipitate/solid residue
formed in both procedures was cumbersome and in some
cases the product 2a still contained alkyltin derived byprod-
ucts after chromatographic purification. Because of these
problems, we developed an alternate workup procedure for
removal of the organotin wastes.
did not occur below -40 °C, so all subsequent experiments
were performed starting at ca. -50 °C (see Table 2).
With an optimized procedure in hand for perfoming the
anti-γ-alkoxypropargylation reaction, we explored the reac-
tions of 4a with other aldehydes (Table 2). These reactions
proceeded in excellent yield and diastereoselectivity. The
least selective substrate in this exploratory study was
crotonaldehyde, which gave products of 1,2-carbonyl addition
in 96% yield with 96:4 diastereoselectivity. All other
substrates gave selectivities in the 97-98% ds level. These
reactions were slightly less selective when performed in less
polar solvents such as toluene or hexane (see footnote d in
Table 2).
It was also of interest to develop a reagent that contained
a readily removable oxygen protecting group so that we could
access 1,2-anti diol units. Initial attempts using propargyl
silyl ethers were thwarted by a retro-Brook rearrangement24
that ensued when the O-silyl propargyl ethers were treated
with strong base. However, synthesis of the MOM ether
derivative 4b was straightforward (Figure 3), and the
reactions of this reagent with a representative set of aldehydes
proceeded in high yield and with excellent selectivity (Table
3). The selectivity realized by using reagent 4b is especially
noteworthy since the additions of the corresponding allenyl-
zinc reagent showed lower diastereoselectivity (4:1 to 17:1
in the best case) and yield (77-86%).
We reasoned that a solid-supported fluoride source could
be a useful reagent for removal of tin residues, because the
solid support would increase the surface area relative to solid
KF or to biphasic extraction mixtures such as ether/aqueous
KF. Potassium fluoride on Celite has been used as a catalyst
for alkylations21 and intramolecular Michael additions22 but
to the best of our knowledge has not been used for the
removal of organotin wastes. The KF/Celite reagent was
prepared according to the literature21 and dried under vacuum.
We were delighted to find that stirring the crude ether extracts
from the reaction of 4a and isobutryaldehyde with KF/Celite
for 1 h resulted in the removal of the majority of the
organotin wastes as determined by TLC analysis. Filtration
of the solid gave the crude product with substantially less
organotin residues as compared to previous methods. Puri-
fication of the crude material by silica gel chromatography
yielded 2a free of organotin impurities.
Further optimization of the γ-alkoxypropargylation reac-
tion using the new workup procedure (Table 1, entries 7-11)
established that only a slight excess of 4a and BuSnCl3 are
necessary for 2a to be obtained in excellent yield and
diastereoselectivity. A direct comparison of the KF/Celite
workup (93% yield, entry 7) versus aqueous KF workup
(79% yield, entry 8) showed that the KF/Celite procedure
was superior in terms of yield and convenience.23 Analysis
of these reactions by TLC revealed that the carbonyl addition
Table 3. Reactions of R-Methoxymethyl Propargylstannane 4b
with Aldehydesa
RCHO
products
yieldb (%)
10:11c
isobutyraldehyde
benzaldehyded
crotonaldehyde
hydrocinnamaldehyde
pivaldehyde
10a ,11a
10b,11b
10c,11c
10d ,11d
10e,11e
95
97
90
98
91
98:2
97:3
94:6
96:4
>98:2
Table 2. Reactions of R-Methoxy Propargylstannane 4a with
Aldehydesa
a Reactions were performed as described in Table 2. b Combined product
yield after silica gel chromatography. c Product ratios determined by 1H
NMR analysis of the crude product.
Attempts to extend this methodology to R-alkoxypro-
pargylstannane 12 were unsuccessful (Figure 4). These
reactions did not proceed to any significant extent, and we
RCHO
products
yieldb (%)
2:3c
isobutyraldehyde
benzaldehyded
crotonaldehyde
hydrocinnamaldehyde
pivaldehyde
2a ,3a
2b,3b
2c,3c
2d ,3d
2e,3e
96
96
96
98
96
97:3
97:3
96:4
97:3
>98:2
(12) We have recently found that mixtures of propargyl- and allenyl-
stannanes give essentially the same results in BuSnCl3-promoted reactions
with aldehydes.
(13) The stereochemistry of compounds 2a,b and 3a were assigned by
ozonolysis of the acetylene to the carboxylic acid followed by reduction of
the acid to the diol and conversion to the acetonide. 1H NMR and NOE
analysis confirmed the stereochemistry of the acetonide derivatives. The
stereochemistry of all other compounds was assigned using Hoffman’s
analysis of 1H NMR chemical shifts in 1,2-diol systems (Landmann, B.;
Hoffmann, R. W. Chem. Ber. 1987, 120, 331). Authentic samples of the
syn disatereomers were obtained by addition of the lithiated 9a and 9b to
the corresponding aldehydes.
a A -50 °C solution of 4a (1.2 equiv) and the aldehyde in CH2Cl2 was
treated dropwise with a 1.0 M solution of BuSnCl3 (CH2Cl2) followed by
warming to 0 °C. b Combined product yield after silica gel chromatography.
c Product ratios determined by 1H NMR analysis of the crude product.
d Reaction was performed in toluene and hexane as solvent and provided
95:5 and 96:4 mixtures of diastereomers 2b and 3b, respectively.
Org. Lett., Vol. 3, No. 19, 2001
3059

Figure 4.
could not isolate or detect the desired products 12 or 13 under
several sets of reaction conditions.
Because the intermediate allenylstannanes 5a and 5b
(generated in situ from the propargylstannanes 4a and 4b)
are racemic, it was not obvious at the outset that these
reagents would function well in reactions with chiral, non-
racemic aldehydes. Recognizing that an opportunity existed
for kinetic resolution of the racemic reagent in a reaction
with a chiral aldehyde,25,26 we examined the anti-γ-meth-
oxypropargylation of aldehyde 15.27 The data summarized
in Figure 5 suggest that one enantiomer of the racemic
reagent 4a reacts with 15 at a faster rate than the other
enantiomer, as diastereoselectivity of 4:1 (33% yield) was
achieved when 1.2 equiv of 4a was employed, 8:1 (83%
yield) with 2.5 equiv of the reagent, and 20:1 with 5 equiv
of 4a. Isomerically pure 16 was obtained in 84% yield from
this experiment. Removal of the alkynyl TMS group by
treatment of 16 with K2CO3 in MeOH at 23 °C (1 h) then
provided the terminal alkyne 17 in 89% yield. Alkyne 17 is
a key intermediate in our total synthesis of bafilomycin A1,28
and this sequence thus constitutes a significant improvement
of our synthesis of this material. We anticipate that 17 will
Figure 5.
serve a similar role in our ongoing total synthesis of
formamicin.29
In summary, we have developed a highly diastereoselective
procedure for the synthesis of propargyl 1,2-anti-diol deriva-
tives using the BuSnCl3-promoted addition of R-alkoxy
propargylstannane reagents to aldehydes and have introduced
the use of KF/Celite as a convenient method for the removal
of organotin residues from the reaction mixtures. We have
also established that racemic reagent 4a undergoes a highly
diastereoselective reaction with chiral aldehyde 15, presum-
ably via a kinetic resolution process. Further applications of
this methodology will be reported in due course.
Acknowledgment. Financial support provided by the
National Institutes of Health to W.R.R. (GM 26782 and GM
38436), and a Postdoctoral Fellowship to N.A.P. (GM 20278)
is gratefully acknowledged.
(14) Marshall, J. A.; Palovich, M. R. J. Org. Chem. 1997, 62, 6001.
(15) Stille, J. K.; Milstein, D. J. Am. Chem. Soc. 1978, 100, 3636.
(16) Leibner, J. E.; Jacobus, J. J. Org. Chem. 1979, 44, 449.
(17) Berge, J. M.; Roberts, S. M. Synthesis 1979, 471.
(18) Curran, D. P.; Chang, C.-T. J. Org. Chem. 1989, 54, 3140.
(19) Crich, D.; Sun, S. J. Org. Chem. 1996, 61, 7200.
(20) Edelson, B. S.; Stoltz, B. M.; Corey, E. J. Tetrahedron Lett. 1999,
40, 6729.
(21) Ando, T.; Yamawaki, J. Chem. Lett. 1979, 45.
(22) Harwood, L. M.; Loftus, G. C.; Oxford, A.; Thomson, C. Synth.
Commun. 1990, 20, 649.
(23) KF/Celite does not cleave the alkynyl TMS group nor a secondary
TBS ether even after 24 h.
Supporting Information Available: Experimental pro-
cedures for synthesis of R-alkoxypropargylstannanes 4a,b;
representative procedure for the additions of 4a,b to alde-
hydes; tabulated spectroscopic data for anti-diol derivatives
2a-e, 10a-e, 16, and 17; and stereochemical assignments
for 2a,b and 3a. This material is available free of charge
via the Internet at http://pubs.acs.org.
(24) Moser, W. H. Tetrahedron 2001, 57, 2065.
(25) Poisson, J.-F.; Normant, F. H. J. Am. Chem. Soc. 2001, 123, 4639.
(26) Marshall, J. A.; Wang, X.-J. J. Org. Chem. 1990, 55, 6246.
(27) The details of the synthesis of aldehyde 15 will be reported
elsewhere.
OL016536M
(28) Scheidt, K. A.; Tasaka, A.; Bannister, T. D.; Wendt, M. D.; Roush,
W. R. Angew. Chem., Int. Ed. Engl. 1999, 38, 165.
(29) Powell, N. A.; Roush, W. R. Org. Lett. 2001, 3, 453.
3060
Org. Lett., Vol. 3, No. 19, 2001