EtOAc, and EtOH), but reaction rates varied and the optimal
choice depends on the ketone examined. Regarding the
hydrogen pressure, 120 psi (8.3 bar) allowed short reaction
times (6-11 h) to be achieved. Greater hydrogen pressures
were not examined, and in general lower hydrogen pressures
(60 psi) resulted in longer reaction times.
to be converted to the corresponding enantioenriched primary
amine 3a (Scheme 1).
While the above ketones provided encouraging results
(Table 1), all of the studied substrates were 2-alkanones. To
investigate the scope of our new method further, we next
examined the asymmetric reductive amination of 2-methyl-
3-hexanone (5) (Scheme 3). This ketone is sterically con-
High diastereoselectivity (93-98% de) is achieved when
a moderately bulky alkyl group resides on an 2-alkanone,
e.g., cyclohexyl (1a), isopropyl (1b), or isobutyl (1c); see
the corresponding amine products 2 (Table 1). Interestingly,
even when a nondirecting straight chain alkyl group is
present, e.g., as in 2-octanone (1e), the diastereoselectivity,
while moderate (72% de, d.r. ) 86:14), is high for the
substrate under consideration. For all ketones examined, the
(R,R)-2 amine is the major diastereomer formed when (R)-
R-MBA is employed. This was confirmed by comparison
of the spectroscopic data obtained for the analytically pure
major diastereomer13 with that reported in the literature.14
The use of (R)- or (S)-R-MBA has previously been
explored for the synthesis of R-chiral aliphatic amines.5 The
previous approaches have relied on first forming the corre-
sponding N-R-methylbenzyl ketimine, isolating that chiral
imine, and finally reducing it by a variety of methods. Of
the few studies available for direct comparison, our system
is far superior to the above two-step strategy. For example,
when the (R)-R-methylbenzyl imine of ketone 1b is reduced
with Zn(BH4)2 in THF at 0 °C, a 48% overall yield of amine
2b (from 1b) is achieved in 66% de.14b,15 Another study using
NaBH4 at -78 °C allowed a 78% overall yield of 2b with
84% de.5f Our one-step method allows room-temperature
synthesis of 2b in 78% yield and 98% de.
Scheme 3. Examination of a Sterically Hindered Ketone
gested, and under our standard reaction conditions no reaction
occurred. When the reaction temperature was raised to 60
°C and the amount of Ti(OiPr)4 was increased to 2.0 equiv,
only 6 area % (GC) of ketone (5) remained after 24 h and
the amine product (6) was obtained in 87% de.
All previously discussed reactions were performed on a
2.5-5.0 mmol scale with Raney Ni loadings of 70-100 wt
%. We then examined these reactions on a larger scale (30-
40 mmol), with the expectation that lower Raney Ni catalyst
loadings would be possible. Thus, for 5.00 g of 2-octanone
(1e) the Raney Ni loading was reduced from 100 to 20 wt
%. Additionally we used 97.0% (instead of 99.999%) grade
Ti(OiPr)4.12 The de remained consistent, but the reaction time
lengthened from 6 to 17 h. Gratifyingly, an isolated yield of
90% was observed on scale-up. Note that no column
chromatography is required. Simple acid-base extraction
techniques followed by HCl salt formation are sufficient to
obtain the diastereomeric product mixture in qualitative
purity.
Another relevant comparison can be made with the only
other reported aliphatic asymmetric reductive amination
study. Using the chiral ammonia equivalent (S)- or (R)-tert-
butanesulfinamide, Ellman and co-workers have converted
ketone 1b to the chiral sulfinamide 4 in 66% yield and 94%
de (Scheme 2).6b Similarly, he converted ketone 1c to a chiral
Scheme 2. Asymmetric Reductive Amination; Comparative
Methods
Ketone 5 proved to be more challenging on scale-up (30
mmol) than our small scale (2.5 mmol) screening reactions
implied. To completely consume ketone 5 in the presence
of 1.05 equiv of (R)-R-MBA, an 80 wt % loading of Raney
(13) In all instances the major amine diastereomer, (R,R)-2, could be
isolated in analytically pure form using flash chromatography.
(14) For ketone 1a, see: (a) Alvaro, G.; Savoia, D.; Valentinetti, M. R.
Tetrahedron 1996, 52, 12571. For ketone 1b, see: ref 5f and (b) Cimarelli,
C.; Palmieri, G. Tetrahedron: Asymmetry 2000, 11, 2555. For ketone 1c,
see: ref 6b and (c) Andres, C.; Nieto, J.; Pedrosa, R.; Villamanan, N. J.
Org. Chem. 1996, 61, 4130. For ketone 1d, see: refs 4c and 14b. For ketone
1e, see: (d) Singaram, B.; Fuller, J. C.; Belisle, C. M.; Goralski, C. T.
Tetrahedron Lett. 1994, 35, 5389.
sulfinamide product (74% yield, 84% de). Our de is higher
in both instances (1b, 98% vs 94%; 1c, 93% vs 84%).
Additionally, the lower yield efficiency, much higher cost
of the chiral ammonia equivalent (S)- or (R)-tert-butane-
sulfinamide, and low temperature (-48 °C) requirements of
the Ellman method make our new method an attractive
alternative.6b,16 An additional reaction step allows 2b or 4
(15) Reference 14b does not provide the yield for the imine step. Based
on ref 5f it is assumed to be 85%.
(16) Ellman has extensively developed the use of (R)-(+)-tert-butane-
sulfinamide (2-methyl-2-propanesulfinamide) for the elegant synthesis of
R-secondary (via sulfinyl aldimines) and R-tertiary (via sulfinyl ketimines)
aliphatic chiral amines in high overall yield and ee in three synthetic steps
from aldehydes and ketones, respectively; see: ref 4a and (a) Liu, G.; Cogan,
D. A.; Ellman, J. A. J. Am. Chem. Soc. 1997, 119, 9913. (b) Cogan, D. A.;
Liu, G.; Ellman, J. Tetrahedron 1999, 55, 8883.
Org. Lett., Vol. 7, No. 22, 2005
4969
Nugent, Thomas C.
