Diastereoselective and Enantioselective Synthesis of Homopropargyl and Allenylcarbinols from Nonracemic Propargyl Mesylates via the Derived Allenyl and Propargyl Trichlorosilanes

Full text of DOI:10.1021/jo971853l

8976
J . Org. Chem. 1997, 62, 8976-8977
Ta ble 1. Ad d ition s of Ra cem ic P r op a r gyl/Allen yl
Tr ich lor osila n es to Ben za ld eh yd e
Dia ster eoselective a n d En a n tioselective
Syn th esis of Hom op r op a r gyl a n d
Allen ylca r bin ols fr om Non r a cem ic
P r op a r gyl Mesyla tes via th e Der ived
Allen yl a n d P r op a r gyl Tr ich lor osila n es
J ames A. Marshall* and Nicholas D. Adams
Department of Chemistry, University of Virginia,
Charlottesville, Virginia 22901
Received October 8, 1997
halide
X
R1
R2
yield, %
2:4a,b
In connection with projects involving the total synthe-
sis of bioactive polypropionate and polyether natural
products, we have been exploring SE2′ additions of chiral
allenyl and propargyl organometallic reagents to alde-
hydes leading to homopropargylic and allenylcarbinols,
respectively. Initial efforts in these investigations showed
that chiral allenylstannanes II, obtained by SN2′ dis-
placements of enantioenriched propargylic mesylates I,
afford either the homopropargylic alcohols III or alle-
nylcarbinols V of high enantiomeric purity under ap-
propriate reaction conditions (eq 1).1,2 The former rep-
resent possible intermediates for polypropionate synthesis,3
and the latter serve as precursors of tetrahydrofuran
subunits of polyethers.1
1a
1b
1c
3a
3b
Br
I
Br
Br
Br
H
H
TMS
TMS
C7H15
C5H11
Me
Me
Me
Me
86
77
73
60
76
94:6 (a )c
86:14 (b)d
0:100 (c)
0:100 (c)
40:60 (d )e
a
b
Racemic. The relative stereochemistry of 4 was not deter-
mined. ∼85:15 anti:syn. ∼80:20 anti:sin. ∼70:30 anti:syn.
c
d
e
Ta ble 2. Ad d u cts fr om Ch lor osilyla tion of Mesyla te 5
a n d in Situ Ad d ition to Ald eh yd es
R
yield, %
6:7
anti:syn (6)
ee, %
c-C6H11 (a )
C6H13 (b)
(E)-BuCHdCH (c)
DPSOCH2CH2 (d )
72
79
80
92
96:4
95:5
>99:1
86:14
982 (a )
a
89:11 (b)
70:30 (c)
92:8 (d )
96b
94b
a
a
b
Not determined. Determined by GC analysis and corrected
for the ee of the starting material.
ducts 2 were shown to be mainly anti by comparison of
the 1H NMR spectra with those of the authentic syn
isomers.1
Recent findings by Kobayashi and co-workers on ad-
ditions of allylic and propargylic trichlorosilanes to
aldehydes led us to consider an alternative, and possibly
more direct, approach to adducts related to III and V.4,5
The appropriate extension of this methodology would
involve the use of substituted chiral allenyl or propargyl
halides 1 or 3 as precursors to propargyl or allenyl
trichlorosilane intermediates that would add to aldehydes
to afford the homopropargyl or allenyl adducts 2 or 4
(Table 1). To test the feasibility of the silylation protocol
and the regio- and diastereoselectivity of the ensuing
additions, we conducted preliminary experiments on the
racemic allenyl and propargylic halides 1 and 3 with
benzaldehyde. The results of these experiments are
summarized in Table 1.
A second series of experiments was conducted with
nonracemic propargyl and allenyl reagents. As we were
concerned that the preparation of highly enantioenriched
propargylic halides would be problematic,6 we explored
the use of mesylate 5, derived from (R)-(+)-3-butyn-2-
ol,7 as the precursor of the chlorosilane intermediate.
Reaction with HSiCl3 and catalytic CuCl in the presence
of Hunig’s base, followed by addition of representative
aldehydes in DMF, afforded mainly the anti adducts
6a -d with high regio- and diastereoselectivity (Table 2).
The ee of adducts 6b and 6c was determined by GC
analysis. The carbinyl stereochemistry was assigned on
1
the basis of the H NMR spectra of the O-methylman-
It was found that the ratio of homopropargylic to
allenic adduct (2:4) is dependent upon the R1 and R2
substituents in the starting halide. Both the allenyl and
the propargyl TMS bromides 1c and 3a led to the
exclusive formation of the allenylcarbinol adduct 4c,
suggestive of a common propargylsilane precursor. The
stereochemistry of the allenyl adduct 4c was not deter-
delates.8 The stereochemistry of adducts 6a and 6d is
assigned by analogy.
The TMS propargylic mesylate 8 afforded the allenyl-
carbinols 9a -d as the major products (Table 3). The
configuration of the carbinyl center was surmised from
1
the H NMR spectrum of the O-methylmandelates.8
1
mined, but the H and 13C NMR spectra are consistent
with a single diastereomer. The homopropargylic ad-
(6) Corey, E. J .; Boaz, N. W. Tetrahedron Lett. 1984, 25, 3055. In
fact, when nonracemic bromide 3 (R1 ) TMS, R2 ) Me, X ) Br; [R]D
)
(1) Marshall, J . A.; Yu, R. H.; Perkins, J . F. J . Org. Chem. 1995,
60, 5550.
(2) Marshall, J . A.; Perkins, J . F.; Wolf, M. A. J . Org. Chem. 1995,
60, 5556.
(3) Marshall, J . A.; Xie, S. J . Org. Chem. 1995, 60, 7230.
(4) Kobayashi, S.; Nishio, K. J . Org. Chem. 1994, 59, 6620. Koba-
yashi, S.; Yasuda, M.; Nishio, K. Synlett 1996, 153.
(5) Kobayashi, S.; Nisho, K. J . Am. Chem. Soc. 1995, 117, 6392.
+5.1), prepared from the alcohol precursor of mesylate 8 (CuBr, LiBr,
THF; 81% yield), was treated with HSiCl3, CuCl (cat.), i-PrNEt, and
then c-C6H11CHO, adduct 9b was secured (74% yield) in racemic form.
(7) Available from DSM Fine Chemicals, Inc., Saddlebrook, NJ , in
∼97% ee.
(8) Trost, B. M.; Belletire, J . L.; Godleski, S.; McDougal, P. G.;
Balkovec, J . M.; Baldwin, J . J .; Christy, M. E.; Ponticello, G. S.; Varga,
S. L.; Springer, J . D. J . Org. Chem. 1986, 51, 2370.
S0022-3263(97)01853-7 CCC: $14.00 © 1997 American Chemical Society

Communications
J . Org. Chem., Vol. 62, No. 26, 1997 8977
Ta ble 3. Ad d u cts fr om Ch lor osilyla tion of Mesyla te 8
a n d in Situ Ad d ition to Ald eh yd es
R
yield, %
ee, % (9)
9:10
[R]D
C6H5 (a )
c-C6H11 (b)
C6H13 (c)
92
55
67
67
a
89:11 (a )
98:2 (b)
95:5 (c)
98:2 (d )
-108
-42
-18
-12
99b
95b
a
PhCH2CH2 (d )
a
b
Not determined. From analysis of the 1H NMR spectrum of
the O-methylmandelates.
The allene configuration of 9a -d can be tentatively
assigned by consideration of the Lowe-Brewster rules.9
All show a negative rotation in accord with the M
configuration.
F igu r e 1. Possible reaction pathways for SE2′ additions of
allenyl/propargyl trichlorosilanes to aldehydes.
Additional evidence for the configuration of 9a was
accomplished through AgNO3-catalyzed cyclization to the
2,5-dihydrofuran 111 and subsequent hydrogenolysis of
the benzylic ether with Na/NH3 (eq 2). The resulting
allylic alcohol 12 was found to have the (S) configuration
tion then occurs with retention of configuration. Chlo-
rosilanes VIII and IX and the chlorosilylcuprate precur-
sors (by analogy to allylic cuprates) may exist as an
equilibrating mixture.1,2,5,11 Addition to the aldehyde can
be envisioned to take place through transition states A
or B, analogous to those proposed for the propargyl/
allenylstannane counterparts.1,2
1
(98% ee) through H NMR analysis of the O-methylman-
delates.8
The foregoing studies establish the potential of non-
racemic propargylic and allenic chlorosilanes as reagents
for the production of homopropargylic alcohols and alle-
nylcarbinols with high diastereoselectivity and chirality
transfer. These reagents offer a useful alternative to
their stannane counterparts for the synthesis of potential
precursors to polypropionate and polyether natural prod-
ucts. Our findings also establish a possible reaction
pathway for these transformations, which should be of
predictive value for regio- and stereoselective synthesis.
A possible reaction pathway for these transformations
is depicted in Figure 1. Accordingly, the starting prop-
argylic mesylate would undergo SN2 or SN2′ displacement
by a Cl3Si cuprate (formally i-Pr2NEt‚HCl‚CuSiCl3) to
afford the intermediate chlorocuprates VI or VII.11
Conversion to the trichlorosilanes by reductive elimina-
Ack n ow led gm en t. Support for this work through
a research grant (CHE 9220166) from the National
Science Foundation is gratefully acknowledged.
(9) Lowe, G. J . Chem. Soc., Chem. Commun. 1965, 411. Brewster,
J . H. Topics Stereochem. 1967, 2, 1. The assumption is made that TMS
is more polarizable than CH(OH)R and Me is more polarizable than
H.
(10) Corey, E. J .; Boaz, N. W. Tetrahedron Lett. 1984, 25, 3063.
(11) Cf. Marshall, J . A. Chem. Rev. (Washington, D.C.) 1989, 89,
1503. Goering, H. L.; Kanter, S. S. J . Org. Chem. 1983, 48, 721.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR spectra
and experimental procedures for key intermediates, the 13C
NMR spectrum of allenylcabinol 4c, and GC traces for 6b,c
(25 pages).
J O971853L