Bismuth Compounds in Organic Synthesis
TABLE 2. Bismuth Triflate Catalyzed
Three-Component Synthesis of Homoallyl Ethers Using
Trialkylorthoformatesa
to the development of three one-pot methods for the
conversion of aldehydes to a variety of homoallyl ethers
(Tables 1-3) catalyzed by bismuth(III) triflate. In the
first approach (Table 1), the acetal was generated in situ
and then converted to the corresponding homoallyl ether.
The use of a nonaqueous workup eliminates an aqueous
waste stream. The acetal was generated in situ by
reaction of the aldehyde with trialkylorthoformate and
the corresponding alcohol in the presence of 0.1 mol %
Bi(OTf)3‚xH2O. The reaction progress was monitored by
GC. After the aldehyde was consumed, the alcohol was
removed under reduced pressure, and then acetonitrile,
the corresponding allylsilane, and 1.0 mol % of the
catalyst were added. It was found that this was more
effective than having 1.0 mol % Bi(OTf)3‚xH2O in the
beginning of the reaction. If the methanol was not
removed prior to addition of the allyltrimethylsilane, no
allylation was observed and the acetal intermediate was
isolated. With a more nucleophilic silane such as meth-
allyltrimethylsilane (Table 1, entry 2), the reaction
proceeded without the addition of the second portion of
the catalyst. To the best of our knowledge, this is the first
example of such a highly catalytic one-pot method for the
generation of homoallyl ethers. This procedure works well
with a variety of aldehydes. Both dimethyl and diethyl
acetals (entry 6) could be generated easily and converted
to the corresponding homoallyl ethers.
The allylation reaction proceeded rapidly and smoothly
at room temperature in all cases. In contrast, the use of
harsher Lewis acid catalysts such as TiCl4 requires
inconveniently low temperatures (-78 °C) for the ally-
lation reaction.1 Acetals derived from ketones reacted
sluggishly with allyltrimethylsilane. In a few trial runs
with cyclohexanone and acetophenone, the expected
homoallyl ether did form, but significant amounts of
unreacted ketone remained (>30%). Although with time
(24 h) the ketone was slowly converted to the homoallyl
ether, indicating that an equilibrium exists between the
ketone and acetal, the allylation reaction never reached
completion.
a Reaction conditions: 2 equiv of trialkylorthoformate and 2
equiv of allyltrimethylsilane were used. b Refers to yield of isolated,
purified product. Yields are not optimized. The purity of all
products was determined to be 96-99% by 1H and 13C NMR
spectroscopy and GC analysis unless otherwise mentioned. Su-
perscript against yield refers to literature reference for the product.
c Reaction was carried out with 5.0 mol % of Bi(OTf)3‚xH2O.
d Reaction was carried out with 2.0 mol % of Bi(OTf)3‚xH2O.
e Product was determined to be 94% pure by GC and NMR
spectroscopy. Remainder was starting material.
During the course of developing the sequential one-
pot allylation method with aldehydes, we found that even
if acetal formation was not complete (ca. 70% complete
as determined by GC analysis), high yields of the ho-
moallyl ether were obtained upon addition of the allyl-
trimethylsilane. This observation suggested that the
reaction of the acetal with allyltrimethylsilane shifts the
aldehyde-acetal equilibrium toward acetal formation,
which subsequently undergoes rapid allylation. This
observation prompted us to investigate a three-compo-
nent, one-pot method for the conversion of aldehydes to
homoallyl ethers.23 These results are summarized in
Table 2. This method involves stirring the aldehyde,
trialkylorthoformate, and allyltrimethylsilane in the
presence of Bi(OTf)3‚xH2O in CH3CN. The reaction is
rapid and gives moderate to good yields of the homoallyl
ether (Table 2). A control experiment revealed that the
reaction of allyltrimethylsilane with aldehydes under the
reaction conditions is very slow. Although allyltrimeth-
ylsilane does react with trialkylorthoformates, the suc-
cess of the three-component method indicates that the
conversion of the aldehyde to the acetal is the fastest
reaction in the pot and the acetal in turn undergoes rapid
allylation. Although the method worked with the aro-
matic conjugated aldehyde, cinnamaldehyde (entry 5), the
attempted conversion of trans-2-hexenal (entry 6) to the
corresponding homoallyl ethyl ether gave poor yields.
(21) For recent reviews on use of bismuth(III) compounds in organic
synthesis, see: (a) Leonard, N. M.; Wieland, L. C.; Mohan, R. S.
Tetrahedron 2002, 58, 8373. (b) Gaspard-Iloughmane, H.; Le Roux, C.
Eur. J. Org. Chem. 2004, 2517. For some recent applications of
Bi(OTf)3 as a catalyst in organic synthesis, see: (c) Yadav, J. S.; Reddy,
B. V. Subba; Swamy, T.; Rao, Raghavender, K. Tetrahedron Lett. 2004,
45, 6037. (d) Yadav, J. S.; Reddy, B. V. S.; Premalatha, K. Synlett 2004,
963. (e) Matsushita, Y.; Sugamoto, K.; Matsui, T. Tetrahedron Lett.
2004, 45, 4723. (f) Ollevier, T.; Lavie-Compin, G. Tetrahedron Lett.
2004, 45, 49. (g) Arnold, J. N.; Hayes, P. D.; Kohaus, R. L.; Mohan, R.
S. Tetrahedron Lett. 2003, 44, 9173.
(22) Bismuth triflate was purchased from Lancaster Chemical Co.
It can also be synthesized in the laboratory from triphenylbismuth and
triflic acid. See: Labrouille`re, M.; Le Roux, C.; Gaspard, H.; Laporterie,
A.; Dubac, J.; Desmurs, J. R. Tetrahedron Lett. 1999, 40, 285. Recently,
a synthesis of bismuth triflate from bismuth oxide and triflic acid in
aqueous ethanol has been reported. Re´pichet, S.; Zwick, A.; Vendier,
L.; Le Roux, C.; Dubac, J. Tetrahedron Lett. 2002, 43, 993. A more
convenient procedure for the synthesis of bismuth triflate uses
chlorobenzene as the solvent. See: Peyronneau, S. M.; Arrondo, C.;
Vendier, L.; Roques, N.; Le Roux, C. J. Mol. Catal. A 2004, 211, 89.
(23) A one-pot method for the conversion of aldehydes to homoallyl
ethers catalyzed by 10 mol % of Sc(OTf)3 has been reported (ref 13).
J. Org. Chem, Vol. 70, No. 6, 2005 2093