1a-OH with a significant decrease in enantiomeric excess
(98.5% f 92.2%) of methyl ketone 1a. n-Butyl ketone 1b
was prepared in 90% yield by treating (S)-2-phenylbutyric
tertiary alcohol ratio was observed as a function of
quenching protocol. In all cases, ketones synthesized using
a cyanocuprate were isolated in higher yields (1.5- to 3-fold
increase) than those synthesized from the corresponding
alkyllithium and in most cases required no purification
after aqueous workup.
Benzoic acid was converted cleanly in 91À97% yields
into methyl (2a), n-butyl (2b), and isopropyl (2c) ketones
using the corresponding cyanocuprate. Ester methyl
benzoate, however, was converted into a tertiary alcohol
acid with 5 equiv of n-Bu2CuLi LiCN in Et2O with only a
3
1.5% loss of enantiomeric excess (98.5% f 97.0%)
(Scheme 1B). By comparison, n-butyl ketone 1b was pre-
pared in 34% yield and 82.0% ee by treating (S)-
2-phenylbutyric acid with 2.1 equiv of n-BuLi, correspond-
ing to a 3-fold drop in yield and a substantial loss of ee as
compared to the cyanocuprate protocol; the starting car-
boxylic acid was recovered in 30% yield upon acidic
workup.
using excess Me2CuLi LiCN. 4-Fluoro-, 2-fluoro-, and
3
4-chlorobenzoic acids were converted into the correspond-
ing methyl (3a, 4a, 5a) and n-butyl (3b, 4b, 5b) ketones
without breaking the carbonÀhalogen bond. Also, most
noteworthy, 4- and 3-bromobenzoic acids were converted
into bromophenyl methyl ketones 6a and 7a without
breaking the carbonÀbromine bond. All halogenated
benzoic acids shown in Figure 1 are known to undergo
ortho-lithiation in the presence of alkyllithium reagents,24À26
a side reaction that is effectively suppressed by utilizing a
cyanocuprate reagent.
Other copper(I) sources (CuI, CuBr DMS, and CuSPh)
3
were tested, and only CuI gave any methyl ketone product
(∼10%). The use of cuprates derived from Grignard
reagents was also investigated, but unacceptable amounts
(20À40%) of tertiary alcohol were observed. Room tem-
perature was found to be optimal; when the cyanocuprate
reaction was held at 0 °C or lower, little or no reaction
occurred. Though as few as 3.5 equiv of Me2CuLi LiCN
3
produced methyl ketone in approximately 75% yield, the
use of 5 equiv of cyanocuprate was found to be optimal for
maximizing ketone formation. Substrate scope was tested
comparing this new cyanocuprate reaction and the litera-
ture alkyllithium protocol (Figure 1, yields in parentheses
4-Methoxybenzoic acid and 2-picolinic acid, two sub-
strates that are also prone to ortho-lithiation by organo-
lithium reagents,27À29 were converted cleanly into the
corresponding ketones 8a, 8b, and 9a in high yields using
this cyanocuprate procedure. 1,3-Thiazole-4-carboxylic
acid was transformed into methyl ketone 10a in 74% yield
using Me2CuLi LiCN; however, when the same acid was
3
Scheme 1. Yields and ee’s are the average of quadruplicate
experiments.
treated with MeLi, complete decomposition of the thiazole
ring was observed.
3-Phenylpropionic acid was converted cleanly into the
corresponding methyl (11a), n-butyl (11b), and isopropyl
(11c) ketones in 86À99% yields. Attempts to treat 3-phe-
nylpropionic acid with Ph2CuLi LiCN were unsuccessful
3
at yielding phenyl ketone. This result can be attributed to
the instability of Ph2CuLi LiCN at room temperature,
3
where the cuprate readily decomposes into biphenyl.30,31
Also, treatment of 3-phenylpropionic acid with t-Bu2Cu-
Li LiCN yielded the tert-butyl ketone in only 17% yield
3
with the remaining mass balance being unreacted car-
boxylic acid. Other carboxylic acids, such as cyclohexyla-
cetic, cyclohexylcarboxylic, phenylacetic, and the sterically
hindered 2-phenylisobutyric acids, were converted into
methyl ketones 12a, 13a, 14a, and 15a and n-butyl ketones
12b, 13b, and 14b in 56À99% yields. Treatment of pheny-
lalanine with Me2CuLi LiCN yielded only recovered
3
starting material upon aqueous workup.32
(24) Gohier, F.; Castanet, A.; Mortier, J. Org. Lett. 2003, 5, 1919.
(25) Gohier, F.; Mortier, J. J. Org. Chem. 2003, 68, 2030.
(26) Gohier, F.; Castanet, A.; Mortier, J. J. Org. Chem. 2005, 70,
1501.
(27) Mortier, J.; Moyroud, J. J. Org. Chem. 1994, 59, 4042.
(28) Nguyen, T.; Chau, N. T. T.; Castanet, A.; Nguyen, K. P. P.;
Mortier, J. J. Org. Chem. 2007, 72, 3419.
(29) Mongin, F.; Trecourt, F.; Queguiner, G. Tetrahedron Lett. 1999,
40, 5483.
(30) Rahman, M. T.; Hoque, A. K. M. M.; Siddique, I.; Chowdhury,
D. A. N.; Nahar, S. K.; Saha, S. L. J. Organomet. Chem. 1980, 188, 293.
(31) Janssen, M. D.; Corsten, M. A.; Spek, A. L.; Grove, D. M.; van
Koten, G. Organometallics 1996, 15, 2810.
(32) Arnold, L. D.; Drover, J. C. G.; Vederas, J. C. J. Am. Chem. Soc.
1987, 109, 4649.
are for RLi reactions). Although there is no universal
literature protocol (equivalents of RLi range from 2 to 5
and temperature from À78 °C to reflux),12À17 the alkyl-
lithium reaction conditions we chose were intended to
maximize ketone formation and to minimize tertiary alco-
hol formation. Quenching protocols also were screened
including inverse addition of the reaction mixture into cold
dilute HCl;16 no effect on stereochemistry or on ketone/
Org. Lett., Vol. 13, No. 19, 2011
5359