First catalytic C-acylation of enoxysilanes: An efficient route to β-diketones

Full text of DOI:10.1021/jo960090j

J . Org. Chem. 1996, 61, 3885-3887
3885
Ta ble 1. Rea ction of Cycloh exa n on e Tr im eth ylsilyl En ol
Eth er w ith Acetyl Ch lor id e Ca ta lyzed by MCln -n Na I
Lew is Acid System sa (Eq 2)
F ir st Ca ta lytic C-Acyla tion of
En oxysila n es: An Efficien t Rou te to
â-Dik eton es1
products and yieldb
entry
MCl (mol %)
2a
3a
Christophe Le Roux, Ste´phanie Mandrou, and
J acques Dubac*
1
2
3
4
5
6
ZnCl2 (5)
SbCl3 (5)
SbCl5 (5)
SnCl4 (5)
BiCl3 (5)
BiCl3c (5)
32
0
0
26
65
64
8
0
0
8
5
5
He´te´rochimie Fondamentale et Applique´e (UPRES(A)5069
CNRS), Universite´ Paul Sabatier, 118 route de Narbonne,
31062 Toulouse Ce´dex, France
Received J anuary 16, 1996
a
b
Solvent CH2Cl2/Et2O (9/1); rt, 1 h. Yields determined by GC
analysis. c BiCl3 is added alone without sodium iodide.
Acylation reactions are one of the most important
group of reactions. The acylation of ketones is one of
these. A variety of methods exist in the literature for
the synthesis of â-diketones. For example, the reaction
of ketone metal enolates with acyl chlorides2 or acyl
cyanides3 (eq 1, X ) Cl, CN), the acylation of enamines,4
and the direct BF3 acylation of ketones5 are some reac-
tions which preferentially lead to C-acylated ketones (2)
usually contaminated with variable amounts of O-acyl-
ated products (3).
systems toward condensation reactions involving enoxy-
silanes and various electrophiles such as aldehydes,
ketones, and R,â-unsaturated ketones.12,13 In search of
new applications for these heavy metal catalytic systems,
we decided to attempt the acylation of enoxysilanes using
acyl chlorides as electrophiles.
Resu lts a n d Discu ssion
Our previous studies on the dramatic increase in the
catalytic activity of bismuth(III) chloride by metallic
iodides revealed that a partial Cl/I exchange reaction
occurs with the concomitant formation of bismuth(III)
iodide and the corresponding metallic chloride.12 We
were interested to know if such an activation could be
extended to Lewis acids like those reported by Tirpak
and Rathke.9 Thus, we examined the catalytic activities
of some Lewis acid/sodium iodide systems for the acetyl-
ation of the enoxysilane derived from cyclohexanone (1a )
(eq 2) (Table 1, entries 1-5). The study revealed that
Because of their ease of preparation and stability, silyl
enol ethers (1, M ) SiR3) derived from ketones have been
acylated without any catalyst or promoter. However, the
methods reported are restricted to polyhalogenated acid
chlorides with complete C-acylation6 or to oxalyl and
malonyl dichlorides.7 Rathke et al. reported the acylation
of silyl enol ethers by acetyl tetrafluoroborate8 or by acyl
chlorides,9 the latter reaction being promoted by a
stoichiometric amount of Lewis acid, among which zinc
chloride and antimony trichloride are the most efficient.
The catalytic acylation of enoxysilanes has been ob-
served in the presence of a heavy metal catalyst such as
mercury(II) chloride10 or tris(dimethylamino)sulfonium
difluorotrimethylsilicate (“TASF”),11 but these two meth-
ods lead exclusively to O-acylation.
activation by sodium iodide is stronger in the case of
bismuth(III) chloride than in any other of the Lewis acids
tested. However, we found later that the addition of
sodium iodide to BiCl3 was not necessary for the acetyl-
ation of the enoxysilane 1a (see Table 1, entry 6). We
also found that sodium chloride and bismuth iodide,
formed by halogen exchange between BiCl3 and NaI,12
are not the catalysts for the reaction 2.
Recently, we reported the catalytic activity, in hetero-
geneous medium, of bismuth(III) chloride-metallic iodide
(1) For our previous work on bismuth catalysts, see: Labrouille`re,
M.; Le Roux, C.; Oussaid, A.; Gaspard-Iloughmane, H.; Dubac, J . Bull.
Soc. Chim. Fr. 1995, 132, 522.
In order to extend these results, we carried out other
acylations using various enoxysilanes with acetyl chloride
and 3-pentanone trimethylsilyl enol ether (1c) with
various acyl chlorides. The overall yields and the selec-
tivities in C-acylation, summarized in Tables 2 and 3,
clearly confirmed the efficiency of these catalysts com-
pared to any other metallic chloride. In particular, note
that whereas benzoyl acetone was synthesized in 5% yield
using 100 mol % of SbCl3,9 only 5 mol % of BiCl3-3NaI
raised this yield to 61% (Table 2, entry 4).
(2) (a) House, H.; Auerbach, R. A.; Gall, M.; Peet, N. P. J . Org. Chem.
1973, 38, 514. (b) Beck, A. K.; Hoekstra, M. S.; Seebach, D. Tetrahe-
dron Lett. 1977, 13, 1187.
(3) Howard, A. S.; Meerholz, C. A.; Michael, J . P. Tetrahedron Lett.
1979, 15, 1339.
(4) Stork, G.; Brizzolara, A.; Landesman, H.; Szmuszkovicz, J .;
Terrell, R. J . Am. Chem. Soc. 1963, 85, 207.
(5) (a) Mao, C. L.; Frostick, F. C.; Man, E. H.; Manyik, R. M.; Wells,
R. L.; Hauser, C. R. J . Org. Chem. 1969, 34, 1425. (b) Manyik, R. M.;
Frostick, F. C.; Sanderson, J . J .; Hauser, C. R. J . Am. Chem. Soc. 1953,
75, 5030.
(6) Murai, S.; Kuroki, Y.; Hasegawa, K.; Tsutsumi, S. J . Chem. Soc.,
Chem. Commun. 1972, 946.
(7) (a) Murai, S.; Hasegawa, K.; Sonoda, N. Angew. Chem., Int. Ed.
Engl. 1975, 14, 636. (b) Effenberger, F.; Ziegler, T.; Scho¨nwa¨lder, K.
H.; Kesmarszky, T.; Bauer, B. Chem. Ber. 1986, 119, 3394.
(8) Hopka, I.; Rathke, M. W. J . Org. Chem. 1981, 46, 3771.
(9) Tirpak, R. E.; Rathke, M. W. J . Org. Chem. 1982, 47, 5099.
(10) Kramarova, E. N.; Baukov, Y. I.; Lutsenko, I. F. Zh. Obshch.
Khim. 1973, 43, 1857.
We discovered that for less reactive acyl chlorides the
addition of zinc iodide (instead of sodium iodide) to
(12) Le Roux, C.; Gaspard-Iloughmane, H.; Dubac, J .; J aud, J .;
Vignaux, P. J . Org. Chem. 1993, 58, 1835.
(13) Le Roux, C.; Gaspard-Iloughmane, H.; Dubac, J . Bull. Soc.
Chim. Fr. 1993, 130, 832.
(11) Limat, D.; Schlosser, M. Tetrahedron 1995, 51, 5799.
S0022-3263(96)00090-4 CCC: $12.00 © 1996 American Chemical Society

3886 J . Org. Chem., Vol. 61, No. 11, 1996
Notes
Ta ble 2. Rea ction s of Tr im eth ylsilyl En ol Eth er s w ith
Acetyl Ch lor id e Ca ta lyzed by BiCl3-3Na I (5 m ol %)a (Eq
1, R ) Me, X ) Cl)
Sch em e 1
products and % yieldb
reaction time,
entry silyl enol ether
min
20
2
3
1
OSiMe3
OSiMe3
(1a )
2a , 65
3a , 5
2
(1b)
20
2b, 84
3b, 2
3
4
5
6
OSiMe3 (1c)
40
40
40
60
2c, 88
2d , 61
2e, 93
2f, 92
3c, 3
3d , 4
3e, 0
3f, 2
OSiMe3
OSiMe3
OSiMe3
(1d )
(1e)
(1f)
Ph
Ph
Ph
a
Solvent CH2Cl2/Et2O (9/1). b Yields determined by GC analysis.
Ta ble 3. Rea ction of Acid Ch lor id es w ith 3-P en ta n on e
Tr im eth ylsilyl En ol Eth er (1c) (Eq 1)a
product and yieldc (%)
entry
R
catalystb reaction time
2
3
1
2
3
4
5
6
7
8
9
Me
Me
Et
A
B
A
B
A
B
A
B
A
B
20 min
10 min
30 min
20 min
1h
30 min
10 h
1 h
2c, 66
2c, 88
2g, 65
2g, 79
2h , 54
2h , 84
2i, 50
2i, 81
2j, 58
2j, 87
3c, 10
3c, 3
3g, 3
3g, 1
3h , 5
3h , 2
3i, 3
3i, 1
3j, 17
3j, 0
Et
i-Pr
i-Pr
t-Bu
t-Bu
Ph
Sch em e 2
6 h
2 h
10
Ph
a
b
Solvent CH2Cl2/Et2O (9/1); rt. Catalyst A;BiCl3 (5 mol %).
Catalyst B:BiCl3-1.5ZnI2 (5 mol %). c Yields determined by GC
analysis.
than isobutyrophenone, deprotonation may partially
contribute to the overall process and this could explain
the presence of variable amounts of the starting ketone
in the final mixture (before workup) resulting from the
protiodesilylation of the ketone enol silane by HCl. One
cannot also preclude that such a protiodesilylation might
lead to the diketone from the enoxysilane 4. Another
plausible mechanism involves a concerted electronic
process (not shown).
Our results also indicate that C-acylation is preferred
over O-acylation. This high regioselectivity can be
rationalized by examining the stabilities of the two
possible intermediates: the carbocation intermediate J ,
which is the result of C-acylation, and the oxonium
intermediate (Scheme 2) resulting from O-acylation. We
propose that an anchimeric assistance between silicon
and the oxygen of the carbonyl group might be respon-
sible for an additional stability of the intermediate J by
leading to a six-membered ring carbocation.14 A similar
assistance has previously been postulated for the haloge-
nation of â-silyloxycarbonyl compounds.15
bismuth(III) chloride gave shorter reaction times and
higher yields (Table 3; results for BiCl3-3NaI not re-
ported). Our previous studies on these catalysts have
also shown that BiCl3-1.5ZnI2 was a more powerful
catalyst than BiCl3-3NaI.12,13 This enhancement is
directly related to the specific surface of the catalyst,12
but with the Bi/Zn system, owing to the formation of
ZnCl2, one cannot preclude that the latter plays a minor
part in the reaction.
As we do not know the nature of the interaction
between BiCl3 and the acyl chloride, this latter will be
represented by “RCOCl-BiCl3” in our discussion. Our
mechanistic proposal (Scheme 1) involves the initial
formation of a cationic intermediate J resulting from the
attack of the complex I on the enoxysilane 1. The
intermediate could decompose in a second step through
two plausible pathways. In the first route (path a) the
â-diketone 2 and the chlorotrimethylsilane are formed
after desilylation. The second path (path b) leads to, after
deprotonation, the enoxysilane 4. Our observations
before aqueous workup have shown that the â-diketone
2 and chlorotrimethylsilane are formed directly and no
enoxysilane 4 is detected by 1H NMR, and also the
acetylation of the enoxysilane derived from isobutyrophe-
none proceeds easily although no proton is available for
deprotonation (Table 2, entry 6). This set of observations
seems to preclude path b. However, for ketones other
(14) As suggested by a referee the assistance of the silicon to
stabilize the cationic intermediate would be also described in K,
although silicon is known to stabilize positive charges at the beta
position through hyperconjugation. However, the intermediate K
would explain the selective desilylation in our mechanism.

Notes
Concerning the activation of bismuth(III) chloride by
J . Org. Chem., Vol. 61, No. 11, 1996 3887
ethers and acyl chlorides were added by syringe. Silyl enol
ethers were prepared from the corresponding ketone by reaction
with trimethylchlorosilane, triethylamine, and sodium iodide in
acetonitrile according to the procedure described by Dunogue`s
and co-workers.19
metallic iodides, this activation might be due to the
formation of the very reactive mixed bismuth(III) halides,
BiCl2I or BiClI2, resulting from the disintegration of the
crystal lattice of BiCl3 by NaI or ZnI2. Other mechanisms
are also plausible, as for example the “in situ” formation
of an acyl iodide in the catalytic cycle. In effect, an acyl
chloride can be converted to an acyl iodide by a metallic
iodide,16 and such a conversion might be realized by one
of the mixed bismuth(III) halide mentioned above and/
or by NaI or ZnI2.
All product yields were determined by GC analysis with
n-alkanes as internal standards.
Typ ica l P r oced u r e. 1-Phenyl-2-methyl-1,3-butanedione20
(2e, Table 2, entry 5). Bismuth(III) chloride (158 mg, 0.5 mmol)
and sodium iodide (225 mg, 1.5 mmol) were transferred under
argon to a flame-dried 50 mL flask equipped with a septum inlet
and magnetic stir bar. The flask was removed from the glovebag
and connected to an argon line, and 10 mL of a mixture of
dichloromethane/ether (9/1) was added by a syringe. This was
followed by the addition of acetyl chloride (860 mg, 11 mmol),
and the suspension was stirred for 5 min at room temperature.
Then, propiophenone silyl enol ether (2 g, 10 mmol) in 2 mL of
the same solvent mixture was added all at once under rapid
stirring. The reaction was monitored by following the formation
of chlorotrimethylsilane (1H NMR). After 40 min at ambient
temperature, the reaction mixture was quenched with 40 mL of
a saturated sodium hydrogen carbonate solution. The layers
were separated, and the aqueous layer was washed three times
with 10 mL of dichloromethane. The organic layers were
combined, dried over sodium sulfate, and analyzed by GC. The
organic layer was concentrated under reduced pressure and
analyzed by 1H NMR and IR. Spectral analyses clearly indicated
the presence of 1-phenyl-2-methyl-1,3-butanedione as the major
component of the crude mixture: 1H NMR (CDCl3) δ 1.32 (d,
3H, J ) 7 Hz), 2.05 (s, 3H), 4.43 (q, 1H, J ) 7 Hz), 7.3-7.5 (m,
3H), 7.8-8 (m, 2H); IR (neat) 1720 and 1678 (CdO), 1596 cm-1
(CdC, enol form).
Con clu sion
Bismuth(III) chloride-metallic iodide systems are the
first catalysts for the C-acylation of enoxysilanes. In
almost all cases studied, reactions proceed rapidly at
room temperature with excellent regioselectivity and
yields. These catalysts are inexpensive and have low
toxicity.17 Moreover, it seems that the metallic iodide
activation observed might be extended to other metallic
chlorides leading to the possible modulation of Lewis acid
strength. An understanding of the mechanism of this
reaction and a search for the active species are currently
underway. Attempts to apply these catalysts to the
acylation of other organosilanes are now in progress.
Exp er im en ta l Section
All other compounds synthesized had spectra identical to
those previously reported.9
Gen er a l. Dichloromethane and ether were purified using
known procedures.18 Bismuth(III) chloride, metallic iodides, and
acyl chlorides were purchased and used without further puri-
fication. All transfer of bismuth(III) chloride and metallic
iodides were done in a glovebag under argon. Ketone silyl enol
Ack n ow led gm en t . We thank Professor Emeritus
Mario Onyszchuk (McGill University, Montreal) for his
assistance in the preparation of the manuscript. Sup-
port of this work by the Centre National de la Recherche
Scientifique is gratefully acknowledged.
(15) Le Roux, C.; Gaspard-Iloughmane, H.; Dubac, J . Org. Chem.
1994, 59, 2238.
(16) Hoffmann, H. M. R.; Haase, K. Synthesis 1981, 715.
(17) (a) Dill, K.; Mc Gown, E. L. In The Chemistry of Organic
Compounds of Arsenic, Antimony and Bismuth; Patai, S., Ed.; J .
Wiley: New York, 1994; pp 695-713. (b) Irwing Sax, N.; Bewis, R. J .
Dangerous Properties of Industrial Materials; Van Nostrand Rein-
hold: New York, 1989; pp 283-284 and 522-523.
J O960090J
(19) Cazeau, P.; Duboudin, F.; Moulines, F.; Babot, O.; Dunogue`s,
J . Tetrahedron 1987, 43, 2075.
(20) (a) J acquier, R.; Petrus, C.; Petrus, F. Bull. Soc. Chim. Fr. 1966,
2845. (b) Nakano, T.; Umano, S.; Kino, Y.; Ishii, Y.; Ogawa, M. J . Org.
Chem. 1988, 53, 3752.
(18) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals; Pergamon Press: New York, 1992.