256
K. Sz o˝ ri et al. / Journal of Catalysis 244 (2006) 255–259
Table 1
a
Hydrogenation of FCN over CD modified Pt/Al O
2
3
b
c
d
d
Solvent
Reaction time Conversion ee
de
eecis
eetrans
FCN
(min)
(%)
(%)
(%) (%)
(%)
e
Toluene
Toluene
80
96
–
39
–
–
10
52
92
14
62
74
75
17
9
5
14
25
THF
50
42
60
88
15
29
62
81
82
82
26
25
15
9
21
26
80
140
AcOH
15
37
53
82
67
18
39
73
44
79
76
74
79
52
46
31
35
22
26
40
35
30
Scheme 1. General scheme of hydrogenation of 2-fluorocyclohexanone.
120
5
f
g
SM 1/1
10
34
52
88
39
77
22
40
90
28
81
81
81
77
85
83
55
50
27
55
36
26
30
42
26
46
2
-methoxycyclohexanone was kinetically resolved by hydro-
20
genation over 10,11-dihydrocinchonidine-modified Pt/Al2O3;
the S enantiomer reacted much faster than the R, leading to
high optical purity of the hydrogenated product [20]. Encour-
aged by these results, we chose to study the hydrogenation of
racemic 2-fluorocyclohexanone (FCN) over cinchona alkaloid-
modified Pt/Al2O3 for testing the asymmetric hydrogenation of
ketones activated by one α fluorine atom. The possible isomers
of the hydrogenated product 2-fluorocyclohexanol (FCL) are
shown in Scheme 1.
120
h
h
30
120
a
Reaction conditions: see Section 2, 0.1 MPa H , 296 ± 2 K.
2
b
c
d
The (R)-FCN enantiomer was in excess.
The cis-FCL isomers were formed in excess.
The cis-(1R,2S)-FCL and the trans-(1R,2R)-FCL enantiomers were
formed in excess.
e
Reaction without modifier.
f
Reaction under 4 MPa H pressure.
Solvent mixture: AcOH/THF 1/1.
Reaction carried out at 273 K.
2
g
h
2
. Experimental
The ee of the unreacted FCN (eeFCN) and the de and ee of the
Commercial alumina-supported Pt catalyst, 5% Pt/Al2O3
Engelhard 4759), was used as received. Cinchonidine (CD,
98%, Fluka) was used without purification. Cinchonidine hy-
drochloride (CD×HCl), O-methylcinchonidine (OMeCD), and
N-methylcinchonidinium chloride (NMeCD) were prepared as
described previously [24,25]. Tetrahydrofurane (THF, ꢀ99.5%,
Fluka) was dehydrated using LiAlH4 and distilled before use;
toluene (ꢀ99.5%) and acetic acid (AcOH, ꢀ99.5%) were used
as received.
cis- and trans-hydrogenated products (eecis and eetrans) were
calculated by the following formulas: de(%) = |[cis-FCL] −
(
ꢀ
[
trans-FCL]| × 100/([cis-FCL] + [trans-FCL]); eeFCN(%) =
|
|
[R-FCN]−[S-FCN]|×100/([R-FCN]+[S-FCN]); eecis(%) =
[1R,2S-FCL]−[1S,2R-FCL]|×100/([1R,2S-FCL]+[1S,2R
-
1
FCL]); and eetrans(%) = |[1R,2R-FCL] − [1S,2S-FCL]| ×
00/([1R,2R-FCL] + [1S,2S-FCL]). The absolute configu-
ration of the FCN enantiomer accumulated in the reaction
mixture was determined by measuring the optical rotation
2
-Fluorocyclohexanone (FCN) was prepared by ring-open-
(
Polamat A polarimeter, c = 1, benzene) of the unreacted
ing of cyclohexene oxide using KHF2 [26] and Jones oxidation
of the resulting trans 2-fluorocyclohexanol (trans-FCL) [27] in
0% overall yield after distillation under reduced pressure, as
FCN isolated by column chromatography (silica gel, eluent:
hexane/dichloromethane 1/2) and comparison with published
optical rotation data [28].
4
shown in Scheme 2. The purity of FCN identified by GC-MS
1
(
Agilent Techn. 6890N GC-5973 MSD) and H NMR (Bruker
3
. Results and discussion
AVANCE DRX-500 spectrometer) analysis was >99% as de-
termined by gas chromatography (using an HP 5890 II equipped
with a flame ionization detector).
Results obtained in the hydrogenation of FCN over CD-
The hydrogenations were carried out in a conventional glass
hydrogenation apparatus or a stainless steel autoclave equipped
with a glass liner. First, 50 mg of catalyst, 3 ml of solvent, and
modified Pt/Al2O3 in different solvents are presented in Table 1.
In the absence of the chiral modifier, the hydrogenation of FCN
results in the selective formation of the cis-FCL (39% de in
toluene). However, in the presence of CD the de increased, with
>80% obtained in THF. In the presence of CD, one of the two
enantiomers of the starting material accumulated, and the ee
of the unreacted FCN increased with reaction time. Thus, ki-
netic resolution of the racemic FCN occurred, (S)-FCN was
hydrogenated at a higher rate, and (R)-FCN accumulated in
the reaction mixture, as determined by optical rotation mea-
surement of the purified compound. A clear sign of the kinetic
resolution was the much higher hydrogenation rates obtained
5
1
mg of modifier were loaded into the reactor. After flushing for
5 min and stirring for 5 min under H2 atmosphere, 0.1 ml of
FCN was added. The hydrogen uptake was followed; samples
were withdrawn at prespecified reaction times and analyzed.
The compounds FCN and cis- and trans-FCL found in the reac-
1
tion mixture were identified by GC-MS and H NMR analysis.
Conversions and selectivities were determined by GC analysis
using a Cyclosil-B (30 m×0.2 mm, J&W Scientific) chiral cap-
illary column.