Asymmetric reduction of cyclic enones to allylic alcohols

Article abstract of DOI:10.1055/s-2002-19747

Asymmetric transfer hydrogenation of cyclic enones results in highly enantioselective reduction to cyclic allylic alcohols.

Full text of DOI:10.1055/s-2002-19747

LETTER
263
Asymmetric Reduction of Cyclic Enones to Allylic Alcohols
a
a
b
a
A
J
symmetr
é
ic Reductio
r
n
of Cyc
ô
lic Enones
t
m
o Allylic
A
lcohols e Hannedouche, Jennifer A. Kenny, Tim Walsgrove, Martin Wills*
a
Department of Chemistry, University of Warwick, Coventry, CV4 7AL, UK
Fax +24(7652)4112; E-mail: m.wills@warwick.ac.uk
b
Glaxo SmithKline Pharmaceuticals, Old Powder Mills, Nr Leigh, Tonbridge, Kent, TN11 9AN, UK
Received 11 December 2001
clic enone due to geometric restraints (Figure 1). In con-
trast acyclic enones have a less rigid structure that may
allow a 1,4-reduction to take place and so to compete with
the 1,2-reduction pathway (Figure 2).
Abstract: Asymmetric transfer hydrogenation of cyclic enones re-
sults in highly enantioselective reduction to cyclic allylic alcohols.
Key words: asymmetric catalysis, enones, enantioselective hydro-
gen transfer, regioselectivity, ruthenium
X
H
M
Asymmetric transfer hydrogenation has recently emerged
as a powerful catalytic process for the asymmetric reduc-
tion of ketones to enantiomerically enriched secondary al-
Y
O
1
Figure 1 Preferred 1,2-reduction process for cyclic enones due to
cohols. Usually very low catalyst quantities are required
rigid structure (M = metal; X = O, NH = Y; auxiliary ligands are
(
typically < 1 mol%) with non-hazardous co-reagents and
2
omitted for clarity).
mild conditions (atmospheric pressure and temperature).
Of the many catalytic systems developed for this process,
the most commonly used are ruthenium(II) or rhodi-
X
H
M
um(III)-arene complexes in conjunction with chiral
-
Y
aminoalcohols or monotosylated diamine ligands and in
the presence of a hydrogen source such as propan-2-ol or
formic acid.
H
O
R
Figure 2 Proposed 1,4-reduction pathway for acyclic enones that
compete with 1,2-reduction process. (M = metal; X = O, NH = Y; au-
The majority of published work has focused on the reduc-
2
2a–c
tion of aromatic ketones, mainly alkyl aryl ketones and
xiliary ligands are omitted for clarity.)
2
d–g
substituted alkyl aryl ketones,
affording the corre-
sponding secondary alcohols with high enantioselectivies
in some cases. Only very limited examples of highly se-
lective reduction of alkyl alkyl ketones have been pub-
In view of this, we wished to examine the scope of the re-
duction of , -unsaturated ketones by asymmetric trans-
fer hydrogenation and so extend the range of useful chiral
compounds accessible by this process. We concentrated
on the reduction of cyclohex-2-en-1-one derivatives bear-
ing an aryl group or a substituted heteroatom at the -po-
sition, using ruthenium(II) or rhodium(III)-arene
3
lished. Surprisingly, despite the intense research work in
this field, few reports have appeared on the asymmetric
transfer hydrogenation of conjugated ketone substrates.4
With the exception of Noyori’s propargylic ketone reduc-
4
a
tion, these few studies focus mainly on the reduction of
4
b,c
cyclic , unsaturated ketones which give consistently
better results than acyclic substrates in term of selectivity
towards allylic alcohols. Noyori has also reported an ef-
fective kinetic resolution of racemic secondary allylic al-
complexes in conjunction with a monotosylated diamine
2a,2b
or aminoalcohol 2 (Figure 3).5
1
4
d
cohols by asymmetric transfer hydrogenation, affording
the enantiomerically enriched allylic alcohol and the un-
saturated ketone (oxidation product). This kinetic resolu-
tion affords better enantioselectivity with cyclic
Ph
Ph
NHTs
NH2
OH
NH2
(1R,2S)-2
4
d
(R, R)-1
substrates than acyclic substrates.
In view of these reports, cyclic enones appear to be more Figure 3
suitable substrates for the synthesis of allylic alcohols via
asymmetric transfer hydrogenation reaction than do acy-
The first investigated unsaturated ketone was the -phe-
nyl substituted cyclohex-2-en-one derivative, 3a, which
clic enones. This may reflect a preference for hydride
transfer to occur via a 1,2-reduction mechanism for a cy-
13
was selectively reduced at the carbonyl moiety by ruthe-
nium(II)-catalysed asymmetric transfer hydrogenation
with ligand 1, affording the corresponding allylic alcohol
4a, the product of the 1,2 reduction, in modest yield and
enantioselectivity (Table 1: entry 1) (conditions described
Synlett 2002, No. 2, 01 02 2002. Article Identifier:
1
©
437-2096,E;2002,0,02,0263,0266,ftx,en;D21001ST.pdf.
Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

2
64
J. Hannedouche et al.
LETTER
in Scheme 1 and Table 1). Similar selectivity and enantio-
O
selectivity were observed with ligand 2 (Table 1, entry 2).
0.25 mol % [Cp*RhCl2]2
.5 mol % (R,R)-1
HCO2H/Et3N (5:2)
rt, 9 days
2
NHCO Me
3c
4c
0
rac-5
O
H
OH
R
R
0.25 mol % [RuCl2(p-cymene)]2
OH
0.5 mol % (R,R)-1 or 1 mol % (1R,2S)-2
NHCO2Me
see table 1
cis/trans-6
3
3
3
a R = Ph
4a R = Ph
b R = OCH2Ph
4b R = OCH2Ph
c R = NHCO2Me
4c R = NHCO2Me
Scheme 2 Reduction of 3c using rhodium (III)-ligand 1 complex.
Table 2 Reduction of 3c Using Rhodium (III)-Ligand 1 Complex.
Scheme 1 Reduction of cyclohex-2-en-1-one derivatives using rut-
henium(II)-ligand 1 and 2 complexes (configuration. predicted for
2f
(
R,R)-1 from ref. ).
Entry
Product
4c
% Yieldb
% ee
Table 1 Reduction of Cylohex-2-en-one Derivatives 3a–c Using
Ruthenium(II)-Ligand 1 and 2 Complexes.
92c
1
2
3
7
5
20
26
13
0c
–d
–d
Entry
Ketones Ligand
% Yieldb % ee
Configura-
tion
g
6 cisa
6 transa
1
2
3
3a
3a
3b
3c
1a
47
27
78
54f
72c
70c
R
S
4
a
1
9
2d
Carried out by H NMR, ref.
Isolated yield.
b
c
1a
1a
99c
R
R
Determined by HPLC analysis using a Daicel chiralcel OD 4.6 250
–
1
mm column [ethanol:hexane (10:90), 0.5 mL min ].
4
>99e
d
No separation of enantiomers by HPLC analysis and by chiral shift
reagent [Eu(hfc)3].
a
The reaction was carried out overnight at 28 °C using 1 mol L–1 so-
lution of unsaturated ketone in formic acid: triethylamine mixture
(
5:2).
Isolated yield.
Whilst 3a, which has an aryl group at the -position, was
reduced by this catalytic system, the introduction of a tert-
butyl group (R = t-Bu) at the -position gave an incom-
patible substrate for ruthenium(II)-catalysed transfer hy-
drogenation and no reduction occurred using either ligand
b
c
Determined by HPLC analysis using a Daicel chiralcel OD 4.6 250
14
–
1
mm column [propan-2-ol: hexane (10:90), 0.5 mL min ].
d
The reaction was carried out for 3 h at room temperature using 0.1
mol L–1 solution of unsaturated ketone in propan-2-ol with KOH (2.5
mol%).
1
or 2/hydrogen donor systems.
e
HPLC analysis using Daicel chiralcel OD [ethanol:hexane (10:90),
.5 mL min ].
–
1
0
In contrast to the ruthenium(II)-catalysed transfer hydro-
genation, the asymmetric reduction of 3c using a rhodi-
um(III)-ligand 1 complex was not as regioselective and
the allylic alcohol 4c was obtained in low yield and excel-
lent enantioselectivity (Table 2, entry 1). The isolated by-
products were the ketone 5 (the 1,4-reduction product),
found to be racemic and the saturated alcohol carbamate
derivative 6 (the overreduction product of 5). In order to
gain an insight into the mechanism of formation of 6, ke-
tone 5 and allylic alcohol 4c were synthesised15 and re-
duced under the same conditions as 3c (Scheme 2). From
these experiments, it was deduced that 6 was formed from
the reduction of 5 via over-reduction as had been pre-
sumed, and not from the reduction of the carbon-carbon
double bond of 4c.
f
Some traces of 6 were also isolated.
g
Predicted using ref.2f,5
Replacement of the -aryl group of 3a by a benzyloxy
3b) or carbamate group (3c) does not influence the regi-
oselectivity of the ruthenium(II)-ligand 1 catalytic system
(
1
4
towards the carbonyl moiety.Thus 3b,c were selectively
reduced to allylic alcohols 4b,c by 1,2-reduction process
with very high enantioselectivities (99%) and modest to
good yields (Table 1, entries 3 and 4).
Attempts to reduce 3b,c using the ruthenium(II)-ligand 2
system afford the starting material back. We have already
observed similar failed reductions with the same catalytic
system on the reduction of -substituted ketones ( -chloro
2
g
Finally, we examined the asymmetric transfer hydro-
genation of some acyclic enones. Attempts to reduce chal-
cone with ruthenium(II)-ligand 1 complex gave the 1,4-
reduction product 7 in low yield (Figure 4). The rutheni-
um(II) and rhodium(III)-catalysed transfer hydrogenation
of 8 using ligand 1 afforded 9 (1,4-reduction product: 25%
isolated yield with ruthenium and 71% with rhodium)
with no 1,2-reduction product. The ruthenium(II)-based
catalyst reduces further the ketone 9 to the corresponding
and -methoxy acetophenone).
In the case of the reduction of -methoxy acetophenone,
we suggested an inhibition of the reduction process due to
the chelation of either the methoxy substituted derivative
or the reduction product to the catalyst. The chelation of
ketone 3b,c or 1,2-reduction product 4b,c to the rutheni-
um, followed by decomposition of the catalyst, may be a
possible explanation for the non-reduction of these sub-
strates.
2
g
Synlett 2002, No. 2, 263–266 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Asymmetric Reduction of Cyclic Enones to Allylic Alcohols
265
+
O
O
O
(CH ); m/z (EI) 171 [(M) , 47%], 143 (29), 140 (8), 127 (37), 114
2
(
100), 100 (15), 95 (10), 76 (22), 59 (24) and 41 (21); m/z (CI) 172
+
Ph
Ph
[(M + H) , 100%], 143 (9), 140 (12), 127 (10) and 114 (19); HRMS
m/z calculated for C H O N [M + H] 172.0973; Found 172.0974.
+
NHCOMe
NHCOMe
8
14
3
6
cis: [ ]D26 -20.1 (c 0.8, ethanol); IR (nujol, cm ) 3410, 3278,
–1
7
8
9
1
693, 1308 and 1051; (300 MHz; CDCl ; Me Si) 5.09 (1 H, br s,
H
3
4
Figure 4
NH), 3.98–3.92 (1 H, m, CHOH), 3.67 (4 H, br s, CHNHCOOCH3),
.00 (1 H, br s, OH), 1.80–1.69 (1 H, m, CH(OH)CHH), 1.69–1.46
5 H, m, CH(OH)CHH-CHHCHHCH CHNH)] and 1.49–1.30 [2 H,
2
[
2
saturated alcohol (33% yield) while no over-reduction m, CH(OH)CH
CHHCHHCH
];
(75.5 MHz; CDCl ; Me Si)
2
2
C 3 4
was observed with the rhodium(III)-based catalyst.16
157.28 (NHCO), 68.4 (CHOH), 52.9 (CH ), 52.4 (CHNH), 32.0,
3
2
8
7.7, 24.1 and 20.1(CH ); m/z (EI) 141 (46%), 114 (45), 98 (100),
2
8 (20), 69 (14), 56 (16) and 41 (14); m/z (CI) 174 [(M + H) , 29%],
In conclusion, the results reported herein highlight a better
+
suitability of cyclic enones than acyclic enones towards 159 (8), 142 (100), 98 (14) and 35 (34); HRMS m/z calculated for
+
selective reduction by asymmetric transfer hydrogenation
C H O N [M + H] 174.1130; Found 174.1130.
8 16 3
–
1
using ruthenium(II)/1/formic acid system to obtain chiral 6 trans: IR (CHCl , cm ) 3436; 3350; 1706; 1320 and 1035;
3
H
allylic alcohols in high enantioselectivities.
(300 MHz; CDCl ; Me Si) 4.78 (1 H, br s, NH), 3.69 (3 H, s, Me),
3 4
3
.40–3.24 [2 H, m, CH(OH)CHNH], 2.09–1.92 [2 H, m,
CH(OH)CHHCH CH CHHCHNH], 1.79–1.65 [2 H, m,
2
2
CH(OH)CH CHHCHH] and 1.41–1.09 [4 H, m, CH(OH)CHHCH-
2
Data for Reduction Products:
HCHHCHH];
(75.5 MHz; CDCl ; Me Si) 164.7 (NHCO), 75.4
C
3 4
2
3
4
a: (Table 1, entry 1) mp 59 °C; [ ]D +9.8 (c 1.6, chloroform);
(CHOH), 57.4 (CHNH), 52.7 (CH ), 34.5, 32.2, 25.04 and 24.45
3
(
Found C, 82.5; 8.1. C H O requires C, 82.7; H, 8.1); IR (nujol,
1
2
14
(CH ); m/z (EI) 141 (26%), 114 (67), 98 (93), 88 (37), 69 (66), 56
2
–
1
cm ) 3172 (OH), 2341, 1160 and 1052;
(300 MHz; CDCl3;
+
H
(100) and 41 (96); m/z (CI) 174 [(M + H) , 76%], 159 (33), 142
Me Si) 7.23–7.38 (5 H, m, ArH), 6.12 (1 H, br s, CH), 4.38 (1 H, br
4
(100), 98 (16) and 35 (14); HRMS m/z calculated for C H O N [M
8
16
3
s, CHOH), 2.29–2.53 (2 H, m, CH ), 1.84–2.00 (2 H, m, CH ) and
+
2
2
+ H] 174.1130; Found 174.1129.
1
1
.59–1.80 (3 H, m, CH and OH);
41.7 (CPh), 140.5, 128.7 (Ar), 127.8 (CH), 127.0, 125.8 (Ar), 63.7
(75,5 MHz; CDCl ; Me Si)
2
C
3
4
+
(
CHOH), 32.08, 27.9 and 19.9 (CH ); m/z (EI) 174 [(M) , 98%], Acknowledgement
2
1
1
45(100), 131 (93), 115 (89), 91 (91), 77 (70) and 51 (56); m/z (CI)
75 [(M + H) , 10%], 157 (100), 146 (31) and 91 (22). 72% ee de-
+
We thank the EPSRC for a research studentship (J.H.) and EPSRC/
GlaxoSmithKline pharmaceuticals for
(J.A.K.).
a CASE studentship
termined by HPLC analysis [OD column; EtOH: hexane (10:90);
–1
0
.5 mL min ]; r.t. 17.1 min (S) and 42.0 min (R).
b: [ ] 21 +74.4 (c 1, ethanol); IR (film, cm ) 3425, 3029, 2926,
–1
4
D
1
663, 1452 and 1180; (300 MHz; CDCl ; Me Si) 7.28–7.41 (5 H,
H
3
4
References
m, ArH), 4.85 (1 H, t, J = 4.1 Hz, CH CH), 4.74 (2 H, s, CH Ph),
4
1
2
2
(
1) (a) Zassinovich, G.; Mestroni, G.; Gladiali, S. Chem. Rev.
992, 92, 1051. (b) Noyori, R.; Hashiguchi, S. Acc. Chem.
.22 (1 H, m, CHOH), 2.37 (1 H, s, OH), 1.99–2.02 (2 H, m, CH2),
.63–1.97 (3 H, m, CH CHH) and 1.49–1.62 (1 H, m, CH CHH);
1
2
2
C
Res. 1997, 30, 97. (c) Palmer, M. J.; Wills, M. Tetrahedron:
Asymmetry 1999, 10, 2045.
(
75,5 MHz; CDCl ; Me Si) 155.0 (COCH ) 137.5, 128.9, 128.3,
3 4 2
1
1
28.0 (Ar), 97.9 (CH), 69.3 (CH Ph), 66.8 (CHOH), 31.4, 24.3 and
2
+
+
(2) (a) Hashiguchi, S.; Fujii, A.; Takehara, J.; Ikariya, T.;
Noyori, R. J. Am. Chem. Soc. 1995, 117, 7562. (b) Fujii,
A.; Hashiguchi, S.; Uematsu, N.; Ikariya, T.; Noyori, R. J.
Am. Chem. Soc. 1996, 118, 2521. (c) Alonso, D. A.;
Guijarro, D.; Pinho, P.; Temme, O.; Andersson, P. G. J. J.
Org. Chem. 1998, 63, 2749. (d) Gamez, P.; Fache, F.;
Mangeney, P.; Lemaire, M. Tetrahedron Lett. 1993, 34,
9.2 (CH ); m/z (CI) 222 [(M + NH ) , 52%], 204 [(M) , 39], 187
2
4
(
25), 108 (67) and 91 (72); HRMS m/z calculated for C H O N [M
13 20 2
+
+
NH ] 222.1494; Found 222.1492. 99% ee determined by HPLC
4
-
1
analysis [OD column; propan-2-ol: hexane (10:90); 0.5 mL min ];
r.t. 16.7 min (S) and 22.7min (R).
c: [ ] 11 +37.5 (c 0.2, ethanol); IR (film, cm ) 3403, 2942, 2867,
–1
4
1
6
D
709, 1522, 1448, 1248 and 1049;
(300 MHz; CDCl ; Me Si)
H
3 4
6
897. (e) Everaere, K.; Carpentier, J. F.; Mortreux, A.;
Bulliard, M. Tetrahedron: Asymmetry 1999, 10, 4663.
f) Kawamoto, A. M.; Wills, M. Tetrahedron: Asymmetry
000, 11, 3257. (g) Kenny, J. A.; Palmer, M. J.; Smith, A. R.
.65 (1 H, br s, NH), 5.78 (1 H, t, J = 4.1 Hz, CH), 4.22 (1 H, br s,
CHOH), 3.69 (3 H, s, Me), 3.57 (1 H, br s, OH), 2.20–1.96 (2 H, m,
CHCH ), 1.94–1.64 [3 H, m, CH(OH)CH CHH] and 1.64–1.49 [1
H, m, CH (OH)CH CHH];
(
2
2
2
(75.5 MHz; CDCl ; Me Si) 155.4
2
C
3
4
C.; Walsgrove, T.; Wills, M. Synlett 1999, 1615.
3) (a) Nishibayashi, Y.; Takei, I.; Uemura, S.; Hidai, M.
Organometallics 1999, 18, 2291. (b) Jiang, Y.; Jiang, Q.;
Zhu, G.; Zhang, X. Tetrahedron Lett. 1997, 38, 215.
4) (a) Matsumura, K.; Hashiguchi, S.; Ikariya, T.; Noyori, R. J.
Am. Chem. Soc. 1997, 119, 8738. (b) Katho, A.; Carmona,
D.; Viguri, F.; Remacha, C. D.; Kovacs, J.; Joo, F.; Oro, L.
A. J. Organomet. Chem. 2000, 593, 299. (c) Hennig, M.;
Püntener, K.; Scalone, M. Tetrahedron: Asymmetry 2000,
11, 1849. (d) Hashiguchi, S.; Fujii, A.; Haack, K.-J.;
Matsumura, K.; Ikariya, T.; Noyori, R. Angew. Chem. Int.
Ed. 1997, 36, 288. (e) Bianchini, C.; Glendenning, L.;
Zanobini, F.; Farnetti, E.; Graziani, M.; Nagy, E. J. Mol. Cat.
A: Chemical 1998, 132, 13.
(
2
1
CO), 134.5 (CNH), 113.4 (CH), 65.9 (CHOH), 52.6 (CH ), 31.9,
3
(
(
+
4.6 and 18.7 (CH ); m/z (EI) 171 [(M) , 83%], 153 (18), 139 (72),
2
12 (89), 96 (29), 84 (46), 68 (73), 59 (32), 54 (100) and 41 (61); m/
+
z (CI) 172 [(M + H) , 53%], 157 (62), 154 (100), 140 (85), 112 (24),
9
+
analysis [OD column; ethanol: hexane (10:90); 0.5 mL min ]; r.t.
1
6 (17), 84 (9) and 68 (9); HRMS m/z calculated for C H O N [M
8 14 3
+
H] 172.0973; Found 172.0973. >99% ee determined by HPLC
–
1
7.3 min (R) and 35.6 min (S).
–1
5
: IR (nujol, cm ) 1738, 1720, 1318, 1270 and 1051; (300 MHz;
H
CDCl ; Me Si) 5.68 (1 H, br s, NH), 4.30–4.22 (1 H, m, CHNH),
3
2
1
3
4
.67 (3 H, s, Me), 2.70–2.46 [2 H, m, (CO)CHHCH CH CHH],
2 2
.46–2.31 [1 H, m, (CO)CHH], 2.20–2.08 [1 H, m, (CO)CH CHH],
2
.95–1.56 [3 H, m, (CO)CH CHHCH ] and 1.50–1.39 (1 H, m,
2
2
CHHCH NH);
(75.5 MHz; CDCl ; Me Si) 208.6 (CO), 157.7
(5) Palmer, M.; Walsgrove, T.; Wills, M. J. Org. Chem. 1997,
62, 5226.
C
3
4
(NHCO), 59.8 (CHNH), 52.5 (CH ), 41.4, 36.2, 28.3 and 24.4
3
Synlett 2002, No. 2, 263–266 ISSN 0936-5214 © Thieme Stuttgart · New York

2
66
J. Hannedouche et al.
LETTER
(
(
(
6) Chidambaram, N.; Chandrasekaran, S. J. Org. Chem. 1987,
2, 5048.
7) Ponaras, A. A.; Younus Meah, M. Tetrahedron Lett. 1986,
7, 4953.
8) (a) Magnus, P.; Lacour, J.; Coldham, I.; Mugrage, B.; Bauta,
B. W. Tetrahedron 1995, 51, 11087. (b) Magnus, P.; Barth,
L. Tetrahedron 1995, 51, 11075.
(14) 1,2-cyclohexanedione was treated with benzyloxytri-
7
5
methysilane to afford diosphenol ether 3b ; the enone 3c
was synthesised in 3 steps starting from cyclohexanone
8
2
via 2-azido-cyclohexan-1-one. The enone derivated 3 with
R = t-Bu was obtained by addition of t-BuLi to 7-oxabi-
cyclo[4.1.0]heptan-2-one, followed by water elimination.
1
1
(15) Ketone 5 was prepared by radical addition of N-chloro-
1
0
(9) Baranov, S. V.; Mochalin, V. B. Chem. Heterocyl. Compd.
carbamate to 1-methoxycyclohexene; racemic 4c was
(
Engl.Transl.) 1981, 8, 779.
10) Driguez, H.; Vermes, J. P.; Lessard, J. Can. J. Chem. 1978,
6, 119.
obtained by reduction of 3c with sodium borohydride and
cerium chloride. The acyclic enone 8 was synthesised in 2
steps starting from the commercially available 3-chloro-
(
5
1
2
(
(
11) Cossy, J.; Bouzbouz, S. Tetrahedron Lett. 1996, 37, 5091.
12) (a) Aelterman, W.; De Kimpe, N.; Kulinkovich, O. Bull. Soc.
Chim. Belg. 1997, 106, 703. (b) Effenberger, F.;
butan-2-one via 3-azidobutan-2-one.
(16) Acyclic enones 7 and 8 were reduced under the same
conditions as displayed in Schemes 1 and 2 for cyclic enones
Beisswenger, T.; Rainer, A. Chem. Ber. 1985, 118, 4869.
13) The unsaturated ketone 3a was prepared by allylic oxidation
of 1-phenylcyclohex-2-ene using a 1:1 molar combination of
3a–c.
(
6
tert-butyl hydroperoxide and pyridinium dichromate.
Synlett 2002, No. 2, 263–266 ISSN 0936-5214 © Thieme Stuttgart · New York