R. Soni et al. / Tetrahedron Letters 56 (2015) 6397–6401
6399
O
O
the (R) enantiomer would be the major product. Using a standard
substrate concentration of 2M in FA/TEA (5:2 azeotrope), the
temperature and substrate to catalyst ratio were varied as shown
in Table 1.
MeO
O
O
O
Cl
MeO
MeO
14
18
15
16
17
Initial attempts focused on the use of high temperature and low
catalyst loading. However this required long reaction times and
gave incomplete conversion. Increasing the loading to 1 mol % gave
almost complete conversion after 6 h at 45 °C and 21 h at 28 °C,
with almost no loss of enantioselectivity. Hence the conditions in
entry 3 were selected for further applications. Under the optimised
conditions, the new ester catalyst 8, along with its acid analogue 12
were tested in the ATH of a series of ketones 14–18 (Fig. 3).
The effect of varying the substrate substitution pattern was
Figure 3. Substrates for ATH by complex 8.
Table 2
Ketone reduction using catalysts 8 and 12a
Entry
Ketone
Catalyst
Time (h)
Convb (%)
eeb (%)
1
Acetophenone
Acetophenone
8
12
8
12
8
6
20
99
99
100
100
98
95 (R)
95 (R)
88 (S)
89 (S)
79 (R)
79 (R)
95 (R)
96 (R)
92 (R)
95 (R)
42d (S)
42d (S)
2c
3
14
14
15
15
16
16
17
17
18
18
1.5
1.5
5.5
5.5
5
explored by the use of a-chloro acetophenone 14, as well as ortho,
4
5
6
7
8
9
10
11
12
meta and para (15, 16, 17) isomers of methoxyacetophenone.
Cyclohexylmethylketone 18 was also chosen to investigate how
the selectivity changed when reducing di-alkyl ketones (Table 2).
12
8
97
92
12
8
12
8
5
95
88
89
99
As expected, an
a-chloro substituent on the ketone increases the
23.5
23.5
22
rate of reaction due to the increased electrophilicity of the carbonyl
carbon. This increased rate came at the penalty of selectivity,
with reduced enantiomeric excess when compared to those for
acetophenone. Upon introducing a methoxy substituent onto the
aromatic ring at the meta position there was little effect other than
reducing the reaction rate; which was due to the increased
electron density on the aromatic ring making the carbonyl less
electrophilic.13g This effect was strongly increased at the para
position, where full conversions could not be achieved even after
24 h. Enantioselectivity was unaffected however and the
obtained ee remained high. Ortho substitution appeared to have
no additional effect on the rate, but significantly reduced the
enantiomeric excess, as had been seen before,2,13 possibly due to
a steric clash with the catalyst.
12
22
96
a
b
c
Reaction conditions: Ru-Catalyst (1 mol %), FA/TEA (5/2 azeotrope), 45 °C.
Conversion and ee determined by chiral GC.
Reduction performed at 28 °C.
d
ee determined by chiral GC after acetylation of the alcohol product.
Ru
N
TsN
Ph
Ru
N
TsN
Ph
Cl
Cl
Ph
Ph
OMe
H
H
19
20
Reduction of cyclohexyl methyl ketone is challenging for most
ATH catalysts, as the high enantioselectivities achieved in reduc-
tion of aryl alkyl ketones are generally dependent on the aromatic
Ph NHTs
N
Ru
N
TsN
N
N
N
Ph
N
H
Cl
CH-p interaction with the ruthenium arene (5, Fig. 1). Without this
Ru2+
Ph
OMe
H
N
2Cl-
interaction the two faces of the ketone are solely distinguished by
the steric bulk of its substituents, hence the ee is frequently lower,
and selectivity is reversed, as observed.2–4
Ph
22
21
N
The basic nitrogen atom of the TsDPEN ligand was also func-
tionalised through the addition of a number of other groups with
potential functionality including a cyclohexadiene (19), anthracene
(20), and even a ruthenium complex (21) (Fig. 4).
Complex 19 was prepared by the reaction between 22 and
[(C6H6)RuCl2]2. Complex 20 was synthesised through in situ forma-
tion of the triflate derived from 3-(1-anthracen-9-yl)propan-1-ol
23,18 followed by reaction with TsDPEN, and reaction of the result-
ing ligand 24 with [(C6H6)RuCl2]2 (Scheme 3). The BIPY–containing
Figure 4. N-Alkylated catalysts for ATH.
black solution from the original bright orange. Hence 12 was used
in its crude form.
Reduction of acetophenone to 1-phenylethanol (Scheme 1) is
the standard benchmark to evaluate the performance of ATH
catalysts, hence this reaction was used for the optimisation of
reduction conditions. Using the (R,R) catalyst it was expected that
ii)
i)
73%
45%
CO2Me
Cl
CO2Me
CO2Me
91%
Ru Cl
TsN
iii)
H
Ph
Ph
NHTs
N
Ph
v)
iv)
Ph
N
H
75%
68%
20
24
23
OH
Scheme 3. Reagents and conditions: (i) dimethyl malonate, NaOMe, MeOH, reflux, 30 h; (ii) LiCl, H2O, DMSO, 130 °C, 18 h; (iii) LiAlH4, THF, 0 °C–rt, 5 h; (iv) 2,6-lutidine, triflic
anhydride, DCM, 0 °C, 30 min, rt, 60 min, then (R,R)-TsDPEN, TEA, DCM, 0 °C, 30 min, rt, 18 h; (v) [C6H6)RuCl2]2, IPA, TEA, 80 °C, 1 h.