A succinct method for preparing the Stork-Jung vinylsilane Robinson annulation reagent

Article abstract of DOI:10.1021/jo0480803

(Chemical Equation Presented) The Stork-Jung vinylsilane reagent (1) is prepared in two steps and in good overall yield. This provides rapid and efficient access to a useful methyl vinyl ketone surrogate.

Full text of DOI:10.1021/jo0480803

SCHEME 1. Stork-Jung Robinson Annulation
A Succinct Method for Preparing the
Stork-Jung Vinylsilane Robinson
Annulation Reagent
Jeananne A. Singletary, Hubert Lam, and
Gregory B. Dudley*
Department of Chemistry and Biochemistry, Florida State
University, Tallahassee, Florida 32306-4390
gdudley@chem.fsu.edu
Received October 28, 2004
SCHEME 2. Original Synthesis of the Stork-Jung
Vinylsilane (1)
The Stork-Jung vinylsilane reagent (1) is prepared in two
steps and in good overall yield. This provides rapid and
efficient access to a useful methyl vinyl ketone surrogate.
(with minor modifications) remains the commonly em-
ployed procedure.7
We envisioned that titanium-catalyzed hydromagne-
siation of 2-butynol (7) according to Sato’s protocol,8
followed by trapping with trimethylsilyl chloride
(TMSCl), would provide efficient access to vinylsilane 5.
Two key obstacles had to be overcome. First, 2-butynol
was not proven to be a reliable hydromagnesiation
substrate.9 Second, the viability of TMSCl as the elec-
trophilic partner was unclear.10 Most examples of the
Sato reaction have involved propargyl alcohols with
larger hydrocarbon appendages (e.g., 2-octynol) and
reactive electrophiles such as aldehydes, iodine, or (after
solvent exchange) iodomethane.8 Nonetheless, the appeal
of a simple, one-pot synthesis of alcohol 5 encouraged us
to investigate this further.11 We report herein the results
of our investigations.
Table 1 presents our initial screening of hydromagne-
siation conditions. Indeed, 2-butynol (7) appears to be less
reactive than more lipophilic propargyl alcohols. Under
the original Sato conditions (entry 1),8 significant amounts
of recovered 7 were observed. We hypothesized that this
apparent decrease in reactivity is related to poor solubil-
The Robinson annulation has been a cornerstone
method in organic synthesis since its establishment in
the early part of the 20th century.1 In its simplest form,
it comprises three steps: Michael addition of an enolate
nucleophile to an R,â-unsaturated ketone, aldol cycliza-
tion, and finally dehydration to afford the cyclohexenone
product.2
As part of an ongoing synthetic project in our Labora-
tory, we wished to prepare the Michael adduct between
an unstabilized cyclohexanone enolate and methyl vinyl
ketone (MVK), the prototypical electrophilic partner for
the Robinson annulation. Although polymerization of
MVK competes with direct addition, modified reagents
(“MVK equivalents”)2 have been developed for effecting
such transformations. Of these, the Stork-Jung vinyl-
silane (1) quickly emerged as the optimal reagent for our
purposes.3,4
Vinylsilane 1 is particularly well suited to the alkyla-
tion of regiospecifically generated, unstabilized enolates.
After alkylation, the vinylsilane may be treated with
mCPBA to afford the ketone, which is poised for cycliza-
tion and dehydration.
Unfortunately, vinylsilane 1 is not as simple to prepare
as other MVK equivalents. Scheme 2 illustrates the
original route to 1.2a,5 Alternative approaches have been
described (via alcohol 5),6 but the Stork-Jung synthesis
(7) (a) Denmark, S. E.; Habermas, K. L.; Hite, G. A. Helv. Chim.
Acta 1988, 71, 168. (b) Ukaji, Y.; Sada, K.; Inomata, K. Chem. Lett.
1993, 1227.
(8) (a) Sato, F.; Ishikawa, H.; Watanabe, H.; Miyake, T.; Sato, M.
J. Chem. Soc., Chem. Commun. 1981, 718. (b) Sato, F.; Kobayashi, Y.
In Organic Syntheses; Wiley and Sons: New York, 1993; Collect. Vol.
VIII, p 507.
(1) Rapson, W. S.; Robinson, R. J. Chem. Soc. 1935, 1285.
(2) Reviews: (a) Gawley, R. E. Synthesis 1976, 777. (b) Jung, M. E.
Tetrahedron 1976, 32, 3.
(3) Stork, G.; Jung, M. E. J. Am. Chem. Soc. 1974, 96, 3682.
(4) For a recent application of this and a related reagent, see:
Snider, B. B.; Shi, B. Tetrahedron Lett. 2001, 42, 9123.
(5) Stork, G.; Jung, M. E.; Colvin, E.; Noel, Y. J. Am. Chem. Soc.
1974, 96, 3684.
(9) A single, unoptimized example has been reported: (a) Lautens,
M.; Huboux, A. H. Tetrahedron Lett. 1990, 31, 3105. (b) Lautens, M.;
Zhang, C. H.; Goh, B. J.; Crudden, C. M.; Johnson, M. J. A. J. Org.
Chem. 1994, 59, 6208.
(10) (a) Two examples are described in a footnote and in undisclosed
yield: Kang, J.; Cho, W.; Lee, W. K. J. Org. Chem. 1984, 49, 1838. (b)
For trapping with stannyl chlorides, which requires solvent exchange,
see ref 9.
(6) (a) Altnau, G.; Ro¨sch, L.; Bohlmann, F.; Lonitz, M. Tetrahedron
Lett. 1980, 21, 4069. (b) Sato, F.; Watanabe, H.; Tanaka, Y.; Sato, M.
J. Chem. Soc., Chem. Commun. 1982, 1126. (c) Audia, J. E.; Marshall,
J. A. Synth. Commun. 1983, 13, 531. (d) Miller, R. B.; Al-Hassan, M.
I. Tetrahedron Lett. 1983, 24, 2055. (e) Wang, K. K.; Liu, C.; Gu, Y.
G.; Burnett, F. N.; Sattsangi, P. D. J. Org. Chem. 1991, 56, 1914.
(11) In fact, Sato reported that hydromagnesiation of 3 and isomer-
ization, followed by solvent exchange and methylation, afforded 5,6b
although this protocol has not supplanted two-step conversions of 3 f
5. The method reported herein is advantageous in that silylation occurs
without requiring the solvent exchange (vide infra), and 7 is more
readily available than 3.
10.1021/jo0480803 CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/24/2004
J. Org. Chem. 2005, 70, 739-741
739

TABLE 1. Hydromagnesiation of 2-Butynol
ment for HMPA. No desired product was obtained in the
presence of bromide or iodide ions (entries 6 and 7),
perhaps due to decomposition promoted by these harsher
reagents. Crotyl alcohol (9, most likely from protonolysis
of 8) was identified in some instances in the crude
product mixture (it is easily removed by silica gel
chromatography). However, under carefully controlled
conditions with freshly distilled reagents and base-
washed glassware, vinylsilane 5 is produced as the only
identifiable reaction product.
Entry 1 outlines the optimal conditions with respect
to yield, although the yields in entry 3 are only slightly
and perhaps insignificantly lower. This one-pot procedure
provides vinylsilane 5 in only one step from commercially
available, inexpensive reagents. Omission of HMPA
(entry 2) provides the target vinylsilane in moderate yield
while avoiding the health and safety concerns associated
with this toxic additive. Furthermore, after a brief
screening of standard conditions, the allylic alcohol was
converted to the corresponding iodide with triphenylphos-
phine and N-iodosuccinimide (Ph3P/NIS)12 to afford the
title compound (1, Scheme 3). The allyl iodide is some-
what unstable to storage, and we have found it most
convenient to generate 1 from alcohol 5 immediately prior
to use.
time bath temp yield
entry
X
solvent
(h)
(°C)
of 5 (%) byproducts
1
2
3
4
5
6
7
Cl Et2O
Cl Et2O
Cl Et2O
4b
rt
-
7
4b
36-38
36-38
36-38
68-70
80-82
36-38
-
7, 9
9
16c
28-35
0
Cl Et2O-HMPAa 12c
7
Cl hexane
Cl Bu2O
Br Et2O
16c
18c
16c
27
7, 9
-
ndd,e
ndd
7, 9
a 2.0 equiv with respect to Mg salts. b Aqueous quench. c Fol-
lowed by heating for 12-14 h with 3.0 equiv of Me3SiCl. d Not
determined. e Product contaminated with Bu2O.
TABLE 2. Reaction of 8 with TMSCl
In conclusion, we describe a convenient, two-step
procedure for preparing vinylsilane 1. Ready access to
the Stork-Jung vinylsilane Robinson annulation reagent
(1) will be especially valuable for initial screening of the
various surrogate reagents for methyl vinyl ketone in
chemical synthesis.
bath temp
(°C)
silane 5
entry
Me3SiX
additive
yield (%)c
1
2
3
4
5
6
7
Me3SiCl
Me3SiCl
Me3SiCl
Me3SiCl
Me3SiCl
Me3SiBr
Me3SiI
HMPAa
36-38
58-60
60-62
60-62
36-38
36-38
36-38
70-76
50-56
65-73
45
THFb
THFb-HMPAa
THFb-DMPUa
DMPUa
42
Experimental Section
d
THFb-HMPAa
THFb-HMPAa
-
-
d
(E)-3-Trimethylsilanylbut-2-en-1-ol (5). To an oven-dried,
three-necked, round-bottomed flask equipped with a pressure-
equalizing dropping funnel, reflux condenser, and septum was
added isobutylmagnesium chloride (1.8 M in ether, 5.1 mL, 9.2
mmol, 2.3 equiv) under argon. To the Grignard solution was
added bis(cyclopentadienyl)titanium dichloride (0.049 g, 0.20
mmol, 0.050 equiv) and the mixture was stirred for 20 min in
an ice bath. At this temperature, 2-butyn-1-ol (0.30 mL, 4.0
mmol, 1.0 equiv) in 10 mL of ether was added dropwise in
intervals over 10 min. Caution was taken with the addition as
a beige-colored heterogeneous mixture formed and the mixture
had a tendency to splash. Once the addition was complete, all
septa were exchanged for glass stoppers and every joint was
greased with silicone lubricant and wrapped with Teflon tape
and Parafilm. The milky-brown reaction mixture was allowed
to warm to room temperature and then placed into an oil bath
and heated at reflux (36-38 °C bath temperature) for 16 h.
(Note: A refrigerated circulator was used to chill the water for
the condenser.) To the refluxing mixture was added hexameth-
ylphosphoramide (3.2 mL, 18 mmol, 4.6 equiv) in 15 mL of ether
dropwise over 10 min via the addition funnel, followed by a
dropwise addition of chlorotrimethylsilane (1.5 mL, 12 mmol,
3.0 equiv). The reaction was maintained at reflux for 13 h. The
dark brown mixture was cooled in an ice bath and then quenched
with the addition of 25 mL of half-saturated ammonium chloride.
The biphasic red-orange solution was washed with two 30-mL
portions of saturated NaHCO3 solution and the aqueous phase
was extracted with two 50-mL portions of ethyl acetate. The
combined organic layers were dried over magnesium sulfate and
concentrated under reduced pressure to leave a deep red oil that
was purified by column chromatography on 16 g of silica gel
a 2.0 equiv with respect to Mg salts. b Volume of THF equal to
the initial volume of Et2O added (15 mL). c Varying amounts of
crotyl alcohol (9) observed in suboptimal cases. d No desired
product (5) was observed in the crude product mixture.
SCHEME 3. Two-Step Synthesis of Vinylsilane 1
ity of butynyloxy-magnesium chloride salts. The reaction
with 2-butynol is heterogeneous, whereas the salts of
larger propargyl alcohols are more soluble in organic
media. However, switching to the magnesium bromide
counterion (entry 7) and addition of hexamethylphos-
phoramide (HMPA, entry 4) were both deleterious.
Fortunately, increases in time and temperature resulted
in complete consumption of the 2-butynol (entry 3),
although simple trapping with TMSCl provided the
desired vinylsilane (5) in low to moderate yield (unopti-
mized).
We then focused on the silylation of hydromagnesiation
intermediate 8. Varying the reaction concentration did
not have a significant effect. In contrast to the first stage
of the reaction (formation of 8), inclusion of coordinating
cosolvents such as HMPA and THF at this point was
beneficial (Table 2). DMPU was not a suitable replace-
(12) We also found that displacement of the alcohol with chloride
(5 f 6) occurs more cleanly with Ph3P/NCS than with Ph3P/CCl4,
although at a higher cost. See the Experimental Section.
740 J. Org. Chem., Vol. 70, No. 2, 2005

(gradient elution with 5-20% ethyl acetate in hexanes) to afford
0.43 g (76%) of 5 as a thick, pale yellow oil: 1H NMR (400 MHz,
CDCl3) δ 0.073 (s, 9H), 1.32 (br s, 1H), 1.71 (dt, 3H, J ) 1.5, 0.7
Hz), 4.27 (dq, 2H, J ) 5.9, 0.6 Hz), 5.88 (tq, 1H, J ) 1.7, 5.9
Hz). 13C NMR (75 MHz, CDCl3) δ -2.4, 14.6, 59.5, 137.5, 139.2.
IR (neat) (cm-1) 3324 (OH, br), 2954 (s), 1621 (w), 1440 (m), 1404
(m), 1359 (m), 1247 (s), 1066 (s), 1013 (s), 836 (s), 749 (s), 690
(m). MS m/z 143.1 [65, M - H]+. CH: Calcd for C7H16OSi: C,
58.27; H, 11.18. Found: C, 58.04; H, 10.88.
equipped with an argon inlet and septum. Methylene chloride
(3.5 mL) was added via syringe followed by the addition of
N-chlorosuccinimide (0.10 g, 0.76 mmol, 1.1 equiv, freshly
recrystallized from dioxane/carbon tetrachloride). The reaction
mixture was stirred for 1 h in an ice bath, and then 4 h at room
temperature. When the reaction was complete (as determined
by TLC), 2.5 mL of hexanes were added and the mixture was
immediately filtered through 2 g of silica gel with the aid of 100
mL of hexanes. The filtrate was concentrated in vacuo to give a
thick yellow oil that was filtered through 1 g of neutral alumina
with 75-125 mL of hexanes and concentrated in vacuo to afford
0.075 g (66%) of 6 as a yellow liquid: 1H NMR (300 MHz, CDCl3)
δ 0.081 (s, 9H), 1.76 (d, 3H, J ) 1.5 Hz), 4.12 (d, 2H, J ) 7.3
Hz), 5.89 (tq, 1H, J ) 7.3, 1.7 Hz). 13C NMR (75 MHz, CDCl3) δ
-2.4, 14.6, 59.5. 137.5, 139.3. IR (neat) (cm-1) 2956 (s), 1615
(w), 1442 (w), 1405 (w), 1249 (s). HRMS calcd for C7H15ClSi
162.0632, found 162.0635.
((E)-3-Iodo-1-methylpropenyl)trimethylsilane (1). (E)-3-
Trimethylsilanylbut-2-en-1-ol (5, 0.20 g, 1.4 mmol, 1.0 equiv) and
triphenylphosphine (0.40 g, 1.5 mmol, 1.1 equiv) were weighed
into an oven-dried, two-necked, round-bottomed flask equipped
with an argon inlet and septum. Methylene chloride (3.5 mL)
was added via syringe followed by the addition of N-iodosuccin-
imide (0.34 g, 1.5 mmol, 1.1 equiv, freshly recrystallized from
dioxane/carbon tetrachloride). The flask was wrapped in alumi-
num foil and the reaction mixture was stirred for 1 h in an ice
bath and then 3 h at room temperature. When the reaction was
complete (as determined by TLC), 5 mL of hexanes were added
and the mixture was immediately filtered through 4 g of silica
gel with the aid of 200 mL of hexanes. The filtrate was
concentrated in vacuo give a thick yellow oil. (Note: Material
Acknowledgment. We thank the FSU Department
of Chemistry and Biochemistry for support of this work,
the NMR Facility for spectroscopic support, the Krafft
lab for the use of their IR spectrometer, and Dr. Umesh
Goli for assistance with mass spectrometry. We thank
Pfizer for a summer undergraduate research fellowship
(J.A.S.) and the MDS Research Foundation for a post-
doctoral fellowship (H.L.). We thank Prof. Marie E.
Krafft for helpful guidance and discussion.
1
at this stage was judged to be >95% pure by H NMR and was
suitable for use.) Subsequent filtration through 2 g of neutral
alumina with 100-150 mL of hexanes and concentration in
vacuo afforded 0.29 g (82%) of 1 as a pale yellow liquid: 1H NMR
(300 MHz, CDCl3) δ 0.06 (s, 9H), 1.71 (d, 3H, J ) 1.7 Hz), 3.93
(d, 2H, J ) 8.4 Hz), 6.02 (tq, 1H, J ) 6.4, 1.7 Hz). 13C NMR (75
MHz, CDCl3) δ -2.5, 1.2, 13.7, 134.2, 143.2. IR (neat) (cm-1
)
2954 (s), 1604 (w), 1437 (w), 1247 (s), 1143 (m), 835 (s). MS m/z
Supporting Information Available: General experimen-
tal procedures and NMR spectra for compounds 1, 5, and 6.
This material is available free of charge via the Internet at
http://pubs.acs.org.
254.2 [12, M]+.
((E)-3-Chloro-1-methylpropenyl)trimethylsilane (6). (E)-
3-Trimethylsilanylbut-2-en-1-ol (5, 0.10 g, 0.69 mmol, 1.0 equiv)
and triphenylphosphine (0.20 g, 0.76 mmol, 1.1 equiv) were
weighed into an oven-dried, two-necked, round-bottomed flask
JO0480803
J. Org. Chem, Vol. 70, No. 2, 2005 741