Investigation into the enantioselection mechanism of ruthenium-arene- diamine transfer hydrogenation catalysts using fluorinated substrates

Article abstract of DOI:10.1016/j.tet.2011.05.099

The effects of both steric and electronic properties of ketones on the selectivity in asymmetric transfer hydrogenation have been studied with aryl alkyl/fluoroalkyl ketones using four ruthenium based catalysts and two different media. The 1-arylethanones, 1-aryl-2-fluoroethanones and 2,2- difluoroacetophenones could be reduced with medium to high ee (86-99%), while the 1-aryl-2,2,2-trifluoroethanones were reduced with low selectivity in most systems. The change in enantioselectivity upon structural variation has been rationalised aided by regression analysis with substituent constants and the partial charge of the carbonyl carbon as predictors. The steric bulk of the alkyl/fluoroalkyl chain was found to be the major factor in determining selectivity in formic acid/triethylamine, while for reduction of a series of substituted 1-arylethanones and 1-aryl-2-fluoroethanones, the selectivity was found to depend on the electronic properties of the aromatic ring, supporting previous evidence that π-π interaction between the substrate and catalyst is important in determining the selectivity. For reductions in water using sodium formate as the hydrogen donor, altered and more complex selectivity mechanisms were observed. Experiments and regression focused on the variation of the alkyl/fluoroalkyl group of phenyl and 1-naphthyl ketones, showed that the selectivity correlated with the size of the substituent, but also the partial charge of the carbonyl carbon.

Full text of DOI:10.1016/j.tet.2011.05.099

Tetrahedron 67 (2011) 5642e5650
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Investigation into the enantioselection mechanism of rutheniumeareneediamine
transfer hydrogenation catalysts using fluorinated substrates
a
a
a,
Sigrid Volden Slungard , Tor-Arne Krakeli , Thor Hakon Krane Thvedt b, Erik Fuglseth a, Eirik Sundby b,
Bard Helge Hoff
ꢀ
ꢀ
a,
*
ꢀ
a Department of Chemistry, Norwegian University of Science and Technology, Høgskoleringen 5, NO-7491 Trondheim, Norway
b Sør-Trondelag University College, E.C. Dahls Gate 2, NO-7004 Trondheim, Norway
a r t i c l e i n f o
a b s t r a c t
Article history:
The effects of both steric and electronic properties of ketones on the selectivity in asymmetric transfer hy-
drogenation have been studied with aryl alkyl/fluoroalkyl ketones using four ruthenium based catalysts and
two different media. The 1-arylethanones,1-aryl-2-fluoroethanones and 2,2-difluoroacetophenonescould be
reduced with medium to high ee (86e99%), while the 1-aryl-2,2,2-trifluoroethanones were reduced with low
selectivity in most systems. The change in enantioselectivity upon structural variation has been rationalised
aided by regression analysis with substituent constants and the partial charge of the carbonyl carbon as
predictors. The steric bulk of the alkyl/fluoroalkyl chain was found to be the major factor in determining
selectivity in formic acid/triethylamine, while for reduction of a series of substituted 1-arylethanones and
1-aryl-2-fluoroethanones, the selectivity was found to depend on the electronic properties of the aromatic
Received 2 February 2011
Received in revised form 4 May 2011
Accepted 23 May 2011
Available online 30 May 2011
Keywords:
Asymmetric reduction
Fluoroketone
1,2-Fluorohydrin
Ruthenium catalysis
Transfer hydrogenation
ring, supporting previous evidence that pep interaction between the substrate and catalyst is important in
determining the selectivity. For reductions inwater using sodium formate as the hydrogen donor, altered and
more complexselectivitymechanismswereobserved. Experimentsand regressionfocused onthevariationof
the alkyl/fluoroalkyl group of phenyl and 1-naphthyl ketones, showed that the selectivity correlated with the
size of the substituent, but also the partial charge of the carbonyl carbon.
Ó 2011 Elsevier Ltd. All rights reserved.
1. Introduction
With respect to fluorinated ketones our laboratory has pre-
viously investigated the ATH of 1-aryl-2-fluoroethanones15 and
Asymmetric transfer hydrogenation (ATH) has been recognised as
a valuable method for obtaining enantioenriched alcohols and
amines.1,2 Considerable progress has been made in the areas of cat-
alyst/ligand development1e5 and in the understanding of the cata-
lytic processes.6e9 Despite the appearance of new catalysts, the
systems developed by Noyori et al.,10,11 are still attractive due to
experimental simplicity and the availability of ligands in both enan-
tiomeric forms. The enantioselection mechanisms of these catalysts
have been investigated previously.7,12e14 Aryl ketones are generally
reduced with higher selectivity and rates than alkyl ketones. This has
2,2,2-trifluoro-1-phenylethanone have also been reduced using the
ATH concept in a low 33e38% ee.16,17 Although useful protocols
exist for the asymmetric reduction of 1-aryl-2,2-difluoroethanones
and 1-aryl-2,2,2-trifluoroethanones,18e22 asymmetric reduction of
the such fluoroketones often gives the product alcohols in mediocre
ee.16e18,23,24 Thus, increased knowledge of the effects caused by
fluorine in asymmetric processes is needed.
On this background we have undertaken the ATH of 1-aryl-1-
alkanones and fluoroalkanones, with the aim of developing enan-
tioselective reductions and gaining an increased understanding of
the enantioselection mechanism of the rutheniumeareneedi-
amine catalysts.
been rationalised by a favourable edge to face
pep interaction
between the arene of the ligand and the substrate.12,13 The more
congested transition state is assumed favoured by these interactions,
and could explain why a lower ee results when the aromatic moiety
contains electron withdrawing substituents. In structurally related
systems, the enantioselectivity has been found to originate primarily
from unfavourable steric interactions between the substrate and the
ligand part of the catalyst,7 or to be the result of a complex interplay
between steric, electrostatic, dispersion and solvation effects.14
2. Result and discussion
2.1. Substrates, catalyst and substituent constants
The ketones used in this study were purchased or prepared as
described previously.25e27 The substrates were subjected to ATH in
two different media; formic acid/triethylamine (5/2 mol ratio) and
* Corresponding author. E-mail address: bhoff@chem.ntnu.no (B.H. Hoff).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.05.099

ꢀ
S.V. Slungard et al. / Tetrahedron 67 (2011) 5642e5650
5643
in water using sodium formate as hydrogen donor. Each reduction
was performed using four different ruthenium based catalysts IeIV,
see Scheme 1.
2.2.1. Effect of alkyl/fluoroalkyl side chain on enantioselectivity using
catalyst I and III. The introduction of aliphatic fluorines alters both
the electronic and steric properties of the substrates. Catalysts I and
III were therefore first employed in reduction of 1b, 2b, 3b and 4b
(Scheme 2) to investigate how a modification of the alkyl/fluo-
roalkyl group affected selectivity and rate.
O
OH
Cat. I-IV (1 mol %)
Ar
1-12
R
Ar
R
HCO2X
Solvent
13-24
O
OH
R
40 o
C
Cat. I or III
R
HCO2H/NEt3
X= H, Na.
1b-4b
13b-16b
Ar:
Scheme 2. ATH of 1b, 2b, 3b and 4b in formic acid/triethylamine using catalyst I and
III, R¼CH3 (1, 13), CH2F (2, 14), CHF2 (3, 15), CF3 (4, 16).
R = CH3 (1, 13), CH2F (2, 14),CHF2 (3, 15), CF3 (4, 16),
Et (5, 17), i-Pr (6, 18), t-Bu (7, 19)
R1 = OMe (a), H (b), F (c), Br (d), CF3 (e), CN (f), NO2
(g), Cl (h), CH3 (i).
R1
ATH of 1b and 2b proceeded with a high selectivity (97.0e97.5%
ee), while reduction of 3b gave 15b in 90.0e93.0% ee. For both
systems a lower ee was observed as the number of fluorines in the
side chain increased, and ATH of 2,2,2-trifluoroacetophenone (4b)
gave (S)-16b in a low 14.0e44.0% ee. In all cases hydride delivery
occurred preferably from the same relative side leading to enan-
tioenriched (R)-13b and (S)-14b, (S)-15b and (S)-16b.
R = CH3 (8, 20), CH2F (9, 21),CHF2 (10, 22),
CF3 (11, 23), Et (12, 24)
Catalysts
Regression analysis with
D(DG) data for catalyst I with various
predictors, gave the best fit with the steric bulk of the substituents
as defined by Taft (Es), and slightly lower using the Charton volume
(ESꢀV). Analysis of the performance of catalyst III, also including
the compounds 5b (R¼Et) and 6b (R¼i-Pr),11 gave a better re-
gression using ESꢀV. Both cases suggest that the lower selectivity
experienced upon exchanging hydrogens with fluorines are mainly
a size effect. The ee-values, conversion data, the observed and
Cl
Cl
Cl
Cl
Ru
N
Ru
N
Ru
Ru
N
Ts
Ts
Ts
N H N
Ts
N H
N H
N H
Ph Ph
Ph Ph
I
II
III
IV
Scheme 1. Asymmetric reduction of 1e12 to 13e24 using catalyst IeIV.
modelled D(DG), and residuals are compiled in Table 1 (Note that Es
and ESꢀV parameters have different signs.).
The catalysts were made in situ by mixing of [RuCl2(p-cym-
ene)2]2 or [RuCl2(mesitylene)2]2 in combinations with each of the
ligands (1R,2R)-N-(p-toluenesulfonyl)-1,2-diphenylethylenediamine
((R,R)-TsDPEN) and (1R,2R)-N-(p-toluenesulfonyl)-1,2-cyclohexan-
ediamine ((R,R)-TsCYDN).
In order to reveal how structural variations affected the enan-
tioselectivity, regression analysis was performed using the differ-
ence in Gibbs free energy for the reactions leading to the two
Table 1
ATH of the phenyl ketones 1b, 2b, 3b and 4b using catalyst I and III in formic acid/
triethylamine: ee (%), conversion (Conv., %), reaction time in hours (h),
estimated Gmod.) in kcal/mol and the residuals (Res.)
D(
DG), model
D(D
f
Entry Substrate R
Cat.
ee (%) Conv.% (h.) R/S
D(
DG)
D(
DG
)
Res.f
mod.
1a
2b
3
1b
2b
3b
4b
1b
2b
3b
4b
5b
CH3
CH2F I
I
97.0
98 (24)
R
S
S
S
R
S
S
S
S
S
d
2.60 2.73
2.60 2.59
1.83 1.56
0.59 0.74
2.72 2.94
2.60 2.21
2.06 1.89
0.18 0.38
2.50 2.68
1.48 1.37
ꢀ0.12
0.02
0.26
ꢀ0.15
ꢀ0.22
0.32
enantiomers, D(DG), calculated from the ee-values and substituent
97.0 >99 (2)
90.0 >99 (6)
44.5 >99 (6)
97.5
97.0 >99 (2)
93.0 78 (24)
constants.28 The substituent constant evaluated includes param-
CHF2
CF3
CH3 III
CH2F III
CHF2 III
CF3 III
I
I
eters for size of groups: Taft’s Es (Es), Charton volume (ESꢀV),
4
5c
6
89 (24)
inductive parameters: s-para, s s
-paraꢀ, -paraþ, resonance: reso-
nance effect (R), resonance delocalisation of negative charge (Rꢀ),
resonance delocalisation of positive charge (Rþ), group dipole,
hydrophobicity, group electronegativity and molar refractivity. As
electronic substituents constants are less developed in the case of
7
0.17
8
14.0 >99 (24)
ꢀ0.20
ꢀ0.17
0.11
9d
Et
(S,S)-III 97.0
96 (60)
10e 6b
11e 7b
i-Pr (S,S)-III 83.0
t-Bu (S,S)-III ND
41 (NRe)
<1 (NRe)
d
d
d
€
fluoroalkyl groups, the calculated Huckel and Mulliken charge for
a
Previously done by Wu et al. in 97% ee.29
Data from Ref. 15.
the carbonyl carbon and oxygen were also used as predictors.
Regression was performed at a confidence level of 95%. The quality
of various models was evaluated by comparing the correlation
coefficient (r), the statistical significance (F) and the standard
deviation (s), whereas the number of n denotes the number of
experiments included in the model. The residuals (Res.) describe
the deviation of the modelled difference in Gibbs free energy
b
c
d
e
f
Previously done by Fujii using (S,S)-III, 98% ee.10
Data from Ref. 10 reaction at 28 ꢁC.
Taken from Ref. 11, reaction time not reported (NR).
Regression: catalyst I:
D
(
D
Gmod.)¼4.85þ(1.72ꢂEs), n¼4, s¼0.226, r¼0.981,
F¼51.04, catalyst III:
D(D
Gmod.)¼6.36e(6.57ꢂESꢀV), (n¼6, s¼0.259, r¼0.971,
F¼66.31).
D(DGmod.) as compared to each experimental value. A good model
is characterised by having a r-value close to 1, a small s-value and
a large F-value.
The size of a phenyl group is not that different to a tri-
fluoromethyl group. Therefore, the lower selectivity observed in
reductions moving from 1b to 4b could in principal be attributed to
the minor size differences between the two substituents. Re-
ductions were therefore performed using the larger naphthyl ke-
tones 8e12, Scheme 3.
As compared with their phenyl analogues, reductions of the
1-naphthyl ketones 8e10 proceeded to give products with lower
ee-values. Catalyst I displayed a higher selectivity in reduction of
8e10 and 12 than catalyst III, giving the products 20e22 and 24 in
2.2. ATH in formic acid/triethylamine using catalyst I and III
Due to the structural similarity of catalysts I and III, their per-
formance are often comparable in terms of rates and selectivity.15
Therefore, a comparison is useful in rationalising their enantiose-
lection mechanisms. In the following sub-chapters their efficiency
as catalysts in ATH in formic acid/triethylamine are discussed.

ꢀ
S.V. Slungard et al. / Tetrahedron 67 (2011) 5642e5650
5644
determining selectivity was then investigated. These in-
HO
R
O
R
Cat. I or III
termolecular binding forces should be strengthened by increasing
the donating properties in the aromatic part of the substrate ke-
tone.31,32 Previously reported data on reduction of 1-arylethanones
and 1-aryl-2-fluoroethanones (2aeg)15 were complemented by
performing additional reactions with the p-substituted 1-aryleth-
anones 1aee and 1-aryl-2,2,2-trifluoroethanones 4aee, for un-
reported substrate/catalyst/solvent combinations.
HCO2H/NEt3
8-12
20-24
Scheme 3. ATH of the 1-naphthyl ketones 8e12 using catalyst I and III, R¼CH3 (8, 20),
CH2F (9, 21), CHF2 (10, 22), CF3 (11, 23), CH2CH3 (12, 24).
Reductions of 1aeg and 2aeg using catalyst I and III gave the
product alcohols in 82.5e97.5% ee, and the highest selectivity and
rates were observed using catalyst III. The 2,2,2-trifluoroketones
4aee were reduced in a low 0.5e44.5% ee. The selectivities and
conversions obtained are shown in Table 3. Regression analysis
performed for each substrate class-catalyst combination, revealed
the best correlation when combining both inductive and resonance
descriptors. For catalyst I/substrate 1beg, a statistically significant
77.0e94.0% ee. Reduction of the 2,2,2-trifluoroketone 11, preferably
occurred from the opposite side as compared to that of 8e10 and
12. The change in selectivity was most pronounced using catalyst III
(77.0% ee). The fluoro-containing substrates reacted with the
highest rate.
The selectivity,
of the R-substituent, and the best regression was found to be with
the Charton volume. The ee-values, conversions, G), regression
D(DG) in the reduction correlated with the size
D(D
model was obtained by combining the inductive s
-paraꢀ parameter
equations and residuals are shown in connection to Table 2.
with the resonance term, R, Fig. 2. With the same predictors,
a lower fit was obtained for the combination of catalyst III/sub-
strate 1aeg, (n¼8, r¼0.952, s¼0.139, F¼24.06).
Table 2
ATH of the 1-naphthyl ketones 8e12 using catalyst I and III in formic acid/trie-
thylamine: ee (%), conversion (Conv., %), reaction time in hours (h),
estimated Gmod.) in kcal/mol and the residuals (Res.)
D(DG), model
D(D
Table 3
c,d
Entry Substrate R Cat. ee (%) Conv. % (h.) R/S DD
1a
2
3
4
G
D
(
D
Gmod.
)
Res.c
0.01
ATH of 1aeg, 2aeg and 4aee to the alcohols 13aeg, 14aeg and 16aee using catalyst
I and III in formic acid/triethylamine: ee (%), conversion (conv.) and reaction time in
hours (h)
8
9
CH3
CH2F I
CHF2
CF3
Et
CH3 III 78.5
CH2F III 62.0 >99 (21)
CHF2 III 28.0 >99 (2)
CF3 III 71.0
Et III 75.5
I
94.0
37 (22)
R
S
S
R
R
R
S
S
R
R
2.16 2.15
1.90 1.58
1.27 1.23
ꢀ0.16 ꢀ0.09
1.61 1.92
1.32 1.43
0.90 0.78
0.36 0.40
ꢀ1.10 ꢀ1.08
1.23 1.17
91.0 >99 (2)
77.0 >99 (2)
12.5 >99 (2)
0.33
0.04
ꢀ0.07
ꢀ0.31
ꢀ0.11
0.12
ꢀ0.04
ꢀ0.02
0.05
10
11
12
8
I
I
I
Entry R1
Cat. CH3 (13)
CH2F (14)
CF3 (16)
5
86.0
4 (20)
18 (21)
ee, %a Conv. %, (h) ee, %b Conv. ee, % Conv. %, (h) R/S
6b
7
%, (2 h)
9
1
2
3
4
5
6
7
8
MeO (a)
H (b)
F (c)
Br (d)
CF3 (e)
CN (f)
NO2 (g)
MeO (a) III 96.5 23 (24)
H (b)
F (c)
I
I
I
I
I
I
I
d
<1 (24)
95.5 >99
97.0 >99
92.0 >99
90.5 >99
91.0 >99
84.5 >99
85.0 >99
96.0 >99
97.0 >99
93.5 >99
91.0 >99
90.5 >99
88.0 >99
84.5 >99
41.0 >99 (6)
44.5 >99 (6)
2.5 >99 (2)
0.5 >99 (2)
1.0 >99 (2)
S
S
S
R
8
9
10
10
11
12
97.0 69 (24)
92.5 67 (24)
93.0 64 (24)
93.0 >99 (24)
75 (2)
15 (21)
a
Previously done by Liu et al., 95% ee.30
S
c
c
b
c
Previously done by Fujii et al. with (S,S)-III at 28 ꢁC, 83% ee.10
Regression: catalyst I:
d
82.5d >99 (4)
d
d
d
S
S
R
R
R
d
c
d
D
(
D
Gmod.)¼5.13e(5.74ꢂESꢀV), n¼5, r¼0.967, s¼0.263,
27.5 92 (24)
14.0 >99 (6)
11.5 >99 (2)
12.5 >98 (2)
3.5 >99 (6)
F¼44.59, catalyst III:
D(D
Gmod.)¼4.78e(6.45ꢂESꢀV), n¼5, r¼0.996, s¼0.102,
9
III 97.5 89 (24)
III 94.0 96 (24)
III 95.5 100 (24)
F¼374.84.
10
11
12
13
14
Br (d)
Apparently, the lower selectivity for the reduction of the di- and
trifluoroketones 3b and 4b, is not related to the relative size of the
aromatic/fluoroalkyl group, but rather to an increase in the size of
the fluoroalkyl moiety. The cartoon drawing in Fig. 1 accounts for
some of the observations.
CF3 (e) III 95.5 100 (24)
CN (f)
III 90.0e 99 (14)
NO2 (g) III 86.0e 99 (24)
c
d
c
d
d
a
All products had the (R)-configuration.
Data from Ref. 15, all products had the S-configuration.
Data not obtained.
Data from Ref. 30.
Data from Ref. 11.
b
c
d
e
R
R
F
Ts
N
Ts
N
F
F
Ru
N
Ru
N
H
H
H
H
CH3
Ph
Ph
O
O
Ph
Ph
TS1
TS2
Fig. 1. Assumed orientation of the ketones in the transition states leading to product
alcohols with opposite stereochemistry.
The lower ee-values for ATH of the naphthyl ketones as com-
pared to the phenyl ketones might be related to less favourable
pep interactions due to steric effects. As the size of the fluoroalkyl
group increases there could be unfavourable interactions with the
ligand, causing a distortion of the normally favoured transition
state (TS1). These repulsive interactions might be more pronounced
in catalysis using the mesitylene-containing catalyst III.
2.2.2. Effect of aromatic substituents using catalyst I and III. The
Fig. 2. ATH of 1bef using catalyst I in formic acid/triethylamine. The selectivity model
plotted as
Gmod.)¼2.56e(1.07ꢂ
a function of the s
-paraꢀ and resonance (R). Regression equation:
importance of the previously postulated edge to face
pep in-
teraction between the arenes of the catalyst and the substrate in
D(
D
s
-paraꢀ)þ(1.30ꢂR), n¼5, r¼0.996, s¼0.054, F¼122.43.

ꢀ
S.V. Slungard et al. / Tetrahedron 67 (2011) 5642e5650
5645
Table 4
Analysis of the data for the combinations of catalyst I/substrate
2aeg and of catalyst III/substrate 2aeg (Fig. 3) gave a better fit
ATH of 1-aryl-2-ethanones in formic acid/triethylamine using catalyst II: ee (%),
conversion (conv. %) and reaction time in hours (h). The data for reduction of 2aeg
can be found in Ref. 15
when using the inductive term
as predictors.
s-para and the resonance term (R)
Entry
Substrate
R1/Ar
R
ee %
Conv. % (h)
Product
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
3d
4a
4b
4c
4d
4e
8
MeO
H
F
Br
CF3
H
MeO
H
F
Br
CF3
1-Naphthyl
1-Naphthyl
1-Naphthyl
1-Naphthyl
1-Naphthyl
CH3
CH3
CH3
CH3
CH3
CHF2
CF3
CF3
CF3
CF3
CF3
d
<1 (24)
11 (24)
53 (24)
<1 (24)
61 (24)
77 (24)
94 (24)
99 (24)
99 (24)
>99 (24)
>99 (6)
11 (23)
22 (23)
>99 (23)
>99 (22)
<1 (24)
d
83
86
d
(R)-13b
(R)-13c
d
86
61
40
39
17
14
18
79
76
53
21
d
(R)-13f
(S)-15b
(S)-16a
(S)-16b
(S)-16c
(S)-16d
(S)-16e
(R)-20
(S)-21
(S)-22
(R)-23
d
9
10
11
12
13
14
15
16
CH3
CH2F
CHF2
CF3
9
10
11
12
Et
has been correlated with substituent constants and the partial
charge of the carbonyl carbon.
2.4.1. Effect of alkyl/fluoroalkyl side chain on enantioselectivity using
catalyst IeIV. The ee of the products for the ATH of 1b, 2b, 3b and 4b
in water, depended on the catalyst and substrate structure, see Fig. 4.
A decrease in selectivity was again observed when increasing the
number of aliphatic fluorines. In all cases the hydride was delivered
preferably from the same side providing an excess of (R)-13b, (S)-
14b, (S)-15b and (S)-16b, respectively. Transfer hydrogenations
catalysed by III gave the highest selectivity for all substrates. The
Fig. 3. ATH of 2aeg in formic acid/triethylamine using catalyst III. The selectivity
model plotted as a function of the inductive -para and resonance (R). Regression
equation: -para)e(1.06ꢂR), n¼7, r¼0.992, s¼0.057, F¼131.42.
Gmod.)¼2.58e(1.57ꢂ
s
D
(
D
s
Although the effect is slightly different in the systems studied,
the results strongly suggest that the electron density of the aro-
matic ring, expressed as a combination of inductive and resonance
contributions, affects enantioselectivity in formic acid/triethyl-
amine. This is most likely due to more favourable pep interaction
between some substrates and the arene part of the catalyst.
Low ee-values were obtained in reduction of 4aec and especially
so in reduction of 4d and 4e. A background reaction was observed for
4d and 4e (Scheme 4), however at such a low rate that this alone
could not alone account for the low selectivity. More likely, formic
acid delivers a hydride to the ruthenium-complexed ketone, effi-
ciently competing with the intramolecular hydride transfer reaction.
OH
CF3
O
HCO2H/NEt3
CF3
24 h 40% conv.
CF3
16e
CF3
4e
Scheme 4. Reduction of 4e in formic acid/triethylamine.
Fig. 4. Enantiomeric excess (ee, %) of products after ATH of 1b, 2b, 3b and 4b using cat.
IeIV in water with sodium formate as hydrogen donor.
2.3. ATH in formic acid/triethylamine using catalyst II and IV
Table 5
ATH of 1aee, 2aeg, 3b, 4ae4e and 8e12 in formic acid/trie-
thylamine using catalyst II gave products with mediocre ee-values
and with a low turnover. Catalyst IV appeared even worse and can
be regarded as useless in this solvent system. Data for the reduction
of the 1-aryl-2-fluoroethanones 2aeg have been reported pre-
viously,15 while data for reduction of the other compounds using
catalyst II is shown in Table 4.
ATH of the phenyl ketones 1b, 2b, 3b and 4b using catalyst I and III in water: ee (%),
conversion (Conv.), reaction time in hours (h),
kcal/mol and the residuals (Res.)
D(DG), model estimated D(DGmod.) in
Entry Substrate R
1a
2
Cat. ee (%) Conv.% (h.) R/S
D(
DG)
D
(D
G mod.)c Res.c
1b
2b
3b
4b
5b
1b
2b
3b
4b
CH3
CH2F I
CHF2
CF3
Et
CH3 III 96.0 >99 (6)
CH2F III 95.5 >99 (2)
CHF2 III 93.0 >99 (6)
CF3 III 22.0 >99 (6)
I
94.0 >99 (2)
91.5 >99 (2)
89.0 >99 (6)
12.0 >99 (2)
86.0 >99 (2)
R
S
S
S
R
R
S
S
S
2.16 2.19
1.94 1.94
1.77 1.75
0.15 0.16
1.61 1.59
2.42 2.44
2.35 2.28
2.06 2.11
0.28 0.28
ꢀ0.03
0.00
0.02
3
I
I
I
4
ꢀ0.01
5b
6
0.02
ꢀ0.02
0.07
2.4. ATH in water using catalyst IeIV
7
8
9
ꢀ0.05
ATH can also be performed in water using sodium formate as
hydrogen donor. This system is attractive due to the higher rate
observed in reduction of less activated ketones,5,29 In contrast to
the formic acid/triethylamine system, reactions in water also gave
acceptable rate of reactions using the TsCYDN based catalysts II and
IV. The use of all catalyst is therefore discussed, and their selectivity
0.00
a
Data corresponds with that obtained by Wu et al.29
Reported by Wu Ref. 29.
Regression equations: catalyst. I:
b
c
D(
D
Gmod.)¼9.94e(7.44ꢂESꢀV)e(6.82ꢂMul.C),
n¼5, r¼0.999, s¼0.030, F¼1220.77, catalyst III:
e(9.84ꢂMul.C), n¼4, r¼0.999, s¼0.087, F¼200.75.
D
(D
Gmod.)¼12.6e(8.78ꢂESꢀV)

ꢀ
S.V. Slungard et al. / Tetrahedron 67 (2011) 5642e5650
5646
data obtained using catalysts I and III are compiled in Table 5, and
the results using catalyst II and IV are shown in Table 6.
Table 6
ATH of the phenyl ketones 1b, 2b, 3b and 4b using catalyst using catalyst II and IV in
water: ee (%), conversion (Conv.), reaction time in hours (h),
Gmod.) in kcal/mol and the residuals (Res.)
D(
DG), model estimated
D(D
b
Entry Substrate Cat. ee (%) Conv.% (h.) R/S
R
D(
DG)
D(
DGmod.
)
Res.b
1a
2
3
4
5
6
7
8
1b
2b
3b
4b
1b
2b
3b
4b
CH3 II
CH2F II
CHF2 II
CF3 II
CH3 IV 93.0
CH2F IV 90.5 >99 (2)
88.0
81.5 >99 (24)
78.0 >99 (6)
78 (24)
R
S
S
S
R
S
S
S
1.71 1.72
1.42 1.34
1.30 1.31
0.15 0.15
2.06 2.11
1.87 1.77
1.77 1.80
0.10 0.11
ꢀ0.01
0.02
ꢀ0.01
ꢀ0.00
ꢀ0.05
0.09
12.0
99 (24)
99 (24)
CHF2 IV 89.0
CF3 IV 8.0
99 (6)
92 (24)
ꢀ0.03
ꢀ0.01
a
Performed previously in 85% ee.33
Regression equations: Cat. II:
b
D(
DGmod.)¼10.5e(6.73ꢂESꢀV)e(11.2ꢂHuc.C),
Fig. 6. ATH of 1b, 2b, 3b and 4b using catalyst II in water with sodium formate as
n¼4, r¼0.999, s¼0.026, F¼1071.55. Cat. IV:
e(25.0ꢂHuc.C), n¼4, r¼0.997, s¼0.109. F¼103.32.
D
(
D
Gmod.)¼19.7e(11.1ꢂESꢀV)
hydrogen donor. The selectivity model
D(DG) plotted as a function of the Charton
€
volume (ESꢀV) and the Huckel charge of the carbonyl carbon (Huc.C).
Regression analysis for each catalyst gave an acceptable corre-
lation of G) with the size of the alkyl/fluoroalkyl group as de-
Table 7
D(D
ATH of 8e12 in water using sodium formate using catalyst IeIV: ee (%), conversion
fined by the Charton volume (ESꢀV). Further, when including the
(conv.), reaction time in hours (h),
and residuals (Res.)
D(
DG), model estimated
D
(D
Gmod.) in kcal/mol
€
Mulliken charge (catalyst I and III) or the Huckel charge (catalyst II
a
Entry Substrate
R
Cat. ee (%) Conv. % (h) R/S
D
(D
G)
D(
DGmod.
)
Res.a
and IV) of the carbonyl carbon as predictor, improved models of the
D(DG) were obtained, see Figs. 5 and 6.
1b
2
3
8
9
10
11
12
8
CH3
CH2F
CHF2
CF3
I
I
I
I
I
87.5
85 (168)
R
S
S
R
R
R
S
1.69 1.74
1.27 1.19
1.08 0.85
ꢀ0.06
0.08
0.23
ꢀ0.10
ꢀ0.15
0.02
77.0 >99 (4)
70.0 >99 (4)
40.5 >99 (9)
80.0
73.5
58.0 >99 (27)
50.0 97 (28)
41.5 >99 (21)
4
ꢀ0.53 ꢀ0.43
1.37 1.52
1.17 1.15
0.82 0.74
0.68 0.49
ꢀ0.55 ꢀ0.47
0.76 0.98
1.37 1.54
0.71 0.84
0.63 0.42
ꢀ1.27 ꢀ1.20
1.42 1.26
1.02 1.21
0.45 0.49
0.29 0.05
ꢀ1.71 ꢀ1.61
1.01 0.92
5b
6b
7
Et
28 (168)
23 (166)
CH3 II
CH2F II
CHF2 II
CF3 II
9
0.09
0.20
8
9
10
11
12
8
S
R
R
R
S
ꢀ0.08
ꢀ0.22
ꢀ0.17
ꢀ0.13
0.22
10b
11b
12
13
14
15b
16b
17
18
19
20b
Et
II
54.5
3 (167)
74 (167)
CH3 III 80.0
9
CH2F III 51.5 >99 (23)
CHF2 III 47.0 >99 (7)
CF3 III 77.0 >99 (7)
10
11
12
8
S
R
R
R
S
S
R
R
ꢀ0.07
Et
III 81.5
24 (167)
38 (161)
0.16
CH3 IV 67.5
ꢀ0.19
ꢀ0.04
0.24
9
CH2F IV 34.5 >99 (16)
CHF2 IV 23.0 >99 (16)
CF3 IV 88.0 86.0 (6)
10
11
12
ꢀ0.10
Et
IV 67.0
20 (161)
0.09
a
Regression equations: catalyst I:
D
D
(
D
Gmod.)¼4.65e(5.59ꢂESꢀV), n¼5, r¼0.983,
Gmod.)¼3.31e(4.15ꢂESꢀV), n¼5, r¼0.970,
Gmod.)¼5.19e(7.02ꢂESꢀV), n¼5, r¼0.987,
Gmod.)¼4.98e(7.25ꢂESꢀV), n¼5, r¼0.989,
s¼0.179, F¼91.23, catalyst II:
s¼0.186, F¼46.86, catalyst III:
s¼0.204, F¼111.07, catalyst IV:
D(
D(D
D(D
s¼0.193, F¼131.72.
Fig. 5. ATH of 1b, 2b, 3b and 4b using catalyst III in water with sodium formate as hy-
drogendonor. The selectivity model plottedas a function of thepredictorsChartonvolume
(ESꢀV) and the Mulliken charge of the carbonyl carbon (Mul.C). Regression equation:
b
The long reaction time in case of substrate 8 and especially 12 might lead to
change in catalyst structure over time, thus possibly changing the selectivity.
D(D
Gmod.)¼12.6e(8.78ꢂESꢀV)e(9.84ꢂMul.C), n¼4, r¼0.999, s¼0.087, F¼200.75.
Although the number of data points is limited, the selectivity
displayed by the catalysts IeIV seems to be affected by the size of the
fluoroalkyl chain, but also by an electronic component, which is
related to the charge density of the carbonyl carbon. Moreover,
according to the model, the increase in partial charge upon fluori-
nation affects the selectivity in the opposite direction to the size
effect. This might explain the low selectivity for the trifluoroketones.
Turning to the ATH of the 1-naphthyl ketones, 8e12 (Table 7),
the use of water in general gave a lower selectivity as compared to
the use of formic acid/triethylamine as reaction medium. The se-
lectivities obtained with the catalysts IeIV are visualised in Fig. 7.
Interestingly, reduction of the trifluoroketone 11 by catalyst IV
gave a product originating from reduction from the opposite face,
Fig. 7. Enantiomeric excess (ee, %) of products in asymmetric reduction of 8e12 using
catalyst IeIV in water with sodium formate as hydrogen donor. Negative ee-values are
assigned for products were hydride delivery occurred from the opposite face.

ꢀ
S.V. Slungard et al. / Tetrahedron 67 (2011) 5642e5650
5647
Table 9
giving as high as 86% ee. Thus, the combination of the bulky tri-
fluoromethyl group in the substrate, the mesityleneeTsCYDN cat-
alyst and water as reaction medium, favours reduction from the
opposite face of the ketone, as compared to a methyl substituent,
p-cymene-TsDPEN based catalyst and formic acid/triethylamine as
solvent.
Regression analysis revealed a fairly good correlation between
D(DG) and the Charton volume, again indicating size as the major
factor in enantioselection. However, better models were also for
catalyst I and II obtained by including the Mulliken charge as
predictor.
ATH of 1aee and 4aee in water using catalyst II and IV: ee (%), conversion (conv. %)
and reaction time in hours (h)
Entry
R
Cat.
CH3, 13
CF3, 16
ee (%)a
Conv. (h)
ee (%)
Conv. (h)
R/S
1
MeO (a)
H (b)
F (c)
Br (d)
CF3 (e)
MeO (a)
H (b)
F (c)
Br (d)
CF3 (e)
II
II
II
II
88.5
88.0
84.5
87.5
90.0
93.0
93.0
90.6
95.5
92.5
9 (24)
78 (24)
100 (2)
86 (24)
99 (2)
31 (24)
99 (24)
98 (24)
96 (24)
99 (24)
1.0
12.0
19.5
10.0
0.0
97 (24)
99 (24)
>99 (6)
90 (24)
>99 (2)
>99 (6)
92 (24)
>99 (24)
86 (24)
>99 (2)
R
S
R
R
ND
ND
S
R
R
2b
3
4
5
6
7
8
9
10
II
IV
IV
IV
IV
IV
0.5
14.0
20.5
9.5
2.4.2. Effect of aromatic substituents (water). The selectivity in re-
duction of the 1-aryl-2-fluoroethanones 2aeg using water as sol-
vent was governed by different factors than in formic acid/
triethylamine.15 Therefore, reduction of 1aee and 4aee were also
undertaken to complement previous reported data. The experi-
mental results for reduction of the 1-arylethanones, 1aee and 1-
aryl-2,2,2-trifluoroethanones 4aee using catalyst I and III are
compiled in Table 8.
3.5
R
a
(R)-Stereochemistry.
b
Previously reported in 85% ee.33
chirality method.15,34e37 The exciton chirality method depends on
two interaction chromophores and the alcohols were therefore
derivatised to their benzoate esters 25e27 (Fig. 8) by standard
benzoylation.
Table 8
ATH of 1aee and 4aee in water using catalysts I and III: ee (%), conversions (conv.)
and reaction time in hours (h)
Entry
R1
Cat.
CH3, 13
CF3, 16
R
O
ee (%)a
Conv. % (h)
ee (%)
Conv. % (h)
R/S
O
1b
2b
3
MeO (a)
H (b)
F (c)
Br (d)
CF3 (e)
Cl (h)
CH3 (i)
MeO (a)
H (b)
I
I
I
I
I
I
I
III
III
III
III
III
94.5
94.0
91.0
89.0
92.5
91.0
90.0
95.5
95.5
95.0
98.0
95.0
77 (24)
100 (2)
100 (2)
98 (24)
100 (2)
>99 (2)
98 (2)
96 (24)
>99 (6)
>99 (6)
>99 (24)
>99 (24)
5.0
99 (6)
S
S
R = CH3 (25), CH2F (26), CHF2 (27)
11.5
22.0
11.0
>99 (2)
99 (24)
90 (24)
R
R
R
d
d
S
25-27
4b
5b
6c
7c
8
3.5
>99 (2)
Fig. 8. Structure of the benzoates 25e27.
d
d
d
d
5.0
>99 (6)
>99 (6)
>99 (24)
>99 (24)
>99 (2)
An energy minimisation using Molecular Modelling Pro (MM2)
showed that the benzoates, 25e27, had similar preferred confor-
mations. A negative first cotton effect was predicted and later
confirmed by CD measurements.
9
22.5
21.5
15.0
2.0
S
10
11
12
F (c)
Br (d)
CF3 (e)
R
R
R
a
(R)-Stereochemistry.
Previously reported in ee of 95% (R¼OMe), 94% (R¼H), 93% (R¼Br), 94%
b
(R¼CF3).29
3. Conclusion
c
Data from Ref. 29.
Not performed.
d
The scope and limitations for the use of the Noyori type ruth-
eniumeareneediamine catalysts in ATH of fluorinated ketones
have been investigated in two different solvent systems. Generally,
good rates and high selectivity were obtained in reduction of
1-aryl-2-fluoroethanones and 2,2-difluoroacetophenone using ru-
thenium p-cymene or mesitylene in combination with the TsDPEN
ligand and formic acid/triethylamine as reaction medium. ATH with
the Ruemesitylene complex in most cases gave the highest selec-
tivity. This catalyst was also found to be highly selective for re-
ductions of 1-arylethanones in water (95.0e98.0% ee). Reduction of
the trifluoroketones 4aee gave the products in a low ee. However,
the rutheniumemesityleneeTsCYDN catalyst, unsuited in ATH of
1-acetonaphthone, in water reduced the trifluorinated analogue in
86% ee. Thus, suitable conditions for enantioselective ATH of tri-
fluoroketones cannot be found by employing the aryl methyl ke-
tones as model systems.
The ATH of 1aee with catalyst I gave products with ee-values
ranging from 89.0 to 94.5%. Even higher enantioselectivity was
obtained with the less commonly used mesitylene based catalyst III
(95.0e98.0% ee).
The 1-aryl-2,2,2-trifluoroethanones, 4aee, were all reduced in
low ee, with the stereochemistry varying depending on the R1-
group. No background reduction was observed in water and the
modest enantioselectivity is assumed to be due to comparable rates
for the two possible transition states.
The use of catalyst II and IV (Table 9) in the reduction of the 1-
arylethanones 1aee resulted in decent selectivity (84.5e93.0% ee),
and rate in contrast to that observed using formic acid/triethyl-
amine as solvent. However, the 1-aryl-2,2,2-trifluoroethanones,
4aee were reduced in low 0e20.5% ee. Attempts to correlate the
D(DG) with a number of substituent constants did not lead to any
In formic acid/triethylamine the enantioselection mechanism
was found to depend mainly on the size of the fluoroalkyl group.
However, within series of 1-arylethanones and 1-aryl-2-
fluoroethanones, regression analysis identified electronic proper-
ties of the aromatic group to be a determinating factor for the
trustworthy models. Possibly, in water, additional hydrophobic and
solvation effects are operating disguising electronic effects exerted
by the para-substituent.
2.5. Determination of absolute stereochemistry of
compounds 9 and 10
selectivity. This is most likely due to
catalyst and the substrate.
pep interaction between the
Also in water, the steric bulk of the alkyl/fluoroalkyl substituent
was found to be a major factor in determining selectivity. However,
an electronic effect related to the partial charge of the carbonyl
The absolute configuration of the alcohols 9 and 10 was de-
termined by circular dichroism spectroscopy (CD) using the exciton

ꢀ
S.V. Slungard et al. / Tetrahedron 67 (2011) 5642e5650
5648
carbon, was also found to be of significance. The regression analysis
indicated this effect to operate in the opposite direction to the size
effect, in part explaining the low selectivity obtained in reduction of
the trifluoroketones. The enantioselection mechanism for ATH in
water within a series of 1-arylethanones and 1-aryl-2-fluor4oeth-
anones, seemed more complex and could not be revealed by the
present study.
30 mL/min, isothermal programs, 13b (100 ꢁC): (R)-13b: 18.3 min,
(S)-13b: 21.2 min, 13c (125 ꢁC): (R)-13c: 6.3 min, (S)-13c: 7.1 min,
13d (145 ꢁC): (R)-13d: 10.7 min, (S)-13d: 12.0 min; 13e (115 ꢁC): (R)-
13e: 13.1 min, (S)-13e: 16.5 min. The ee of the fluoroalcohols 14aeg
was determined by HPLC using
a
Chiracel OD column
(0.46 cmꢂ25 cm), mobile phase: hexane/2-propanol, 98/2, flow
rate 1.0 mL/min42 2,2-Difluoro-1-phenylethanol (15b) was ana-
lysed using HPLC and a Chiracel OD column (0.46 cmꢂ25 cm),
mobile phase: hexane/2-propanol (95/5), flow rate 1.0 mL/min, (S)-
15b: 12.7 min, (R)-15a: 13.7 min.
4. Experimental
The trifluoroalcohols 16aee were analysed using HPLC and
a Chiralcel OD column with a flow rate of 1 mL/min 16a,b: mobile
phase: hexane/2-propanol (99/1): (S)-16a: 45.7 min, (S)-16a:
52.4 min, (S)-16b: 37.5 min, (R)-16b: 41.4 min 16cee: mobile phase:
hexane/2-propanol (95/5): (R)-16c: 7.2 min, (S)-16c: 8.8 min, (R)-
16d: 8.2 min, (S)-16d: 10.4 min, (R)-16e: 6.8 min, (S)-16e: 7.7 min.
The trifluoroalcohol 16a was also analysed using GC and a CP-
Chirasil-Dex CB column, 80e200 ꢁC, 10 ꢁC/min, column pressure:
10 psi, split flow: 30 mL/min. Retention times: (R)-16a: 8.2 min, (S)-
16a: 8.4 min.
The 1-naphthylalkanols 20e24 were analysed using HPLC and
a Chiralcel OD column eluting with hexane/2-propanol (87/13),
flow rate of 1 mL/min. Retention times: (S)-20: 8.10 min, (R)-20:
12.1 min, (R)-21: 8.9 min, (S)-21: 13.5 min, (R)-22: 8.9 min, (S)-22:
17.6 min, (R)-23: 7.3 min, (S)-23: 21.9 min, (S)-24: 6.9 min, (R)-24:
11.7 min.
4.1. Chemical and equipment
The solvents and reagents were used as received from the
suppliers. [RuCl2(p-cymene)]2, (R,R)-TsDPEN, (R,R)-TsCYDN, ruth-
enium(III) chloride hydrate, acetophenones (1aee), 1-aceton-
aphthone (8) 2,2,2-trifluoro-1-phenylethanone (4b) and 2,2,2-
trifluoro-1-(4-(trifluoromethyl)phenyl)ethanone (4e) were from
Aldrich, 1,3,5-trimethyl-1,4-cyclohexadiene was from Alfa Aesar, 4a
was from Fluchem, 4c and 4d were from Apollo. The 1-aryl-2-
fluoroethanones, 2aeg,25,26 the 1-naphthyl ketones 9e12,38 2,2-
difluoro-1-phenylethanone (3b)39 and [RuCl2(mesitylene)]2,40,41
were all synthesised as described previously. Asymmetric transfer
hydrogenations were performed in an incubator shaker from
€
Brunswic Scientific Co Inc. The Huckel and Mulliken partial charge
was estimated starting with MM2 minimised structure using
ChemBio3D, Version 12.0. The Mulliken charges were obtained via
the Games interface (RHF/3-21G). The used values are shown in
Table 10. Regression analysis was performed using Minitab 15 from
Minitab Inc. at a confidence level of 95%.
4.3. Asymmetric transfer hydrogenation in formic acid/
triethylamine (Investigation scale)
A suspension of the [RuCl2(arene)]2 (0.001 mmol) and ligand
(0.0027 mmol) in CH2Cl2 (0.5 mL) was stirred at 20 ꢁC for 30 min.
After removal of CH2Cl2 by a stream of N2, the ketone (0.1 mmol) in
a physical mixture of HCO2H/Et3N (5/2 mol ratio, 0.25 mL) was
added. The reaction mixture was stirred vigorously at 40 ꢁC for the
specified number of hours. Samples were withdrawn from the re-
action mixture and the solvent was removed under a stream of N2.
The samples were then dissolved in the HPLC eluent, filtered
through silica and analysed by HPLC for determination of conver-
sion and ee. Note: on scale-up the pressure built-up of CO2 must be
taken into account.
Table 10
€
Estimated values for the partial charge of the carbonyl oxygen and carbon: Huckel
charge of carbon (Huc.C), Huckel charge of oxygen (Huc.O), Mulliken charge of
carbon (Mul.C), Mulliken charge of oxygen (Mul.O)
€
Compound
Ar
R1
Huc.C
Huc.O
Mul.C
Mul.O
1b
2b
3b
4b
5b
6b
8
Ph
Ph
Ph
Ph
Ph
Ph
Naphthyl
Naphthyl
Naphthyl
Naphthyl
Naphthyl
CH3
CH2F
CHF2
CF3
0.471
0.440
0.413
0.378
0.464
0.462
0.452
0.424
0.390
0.353
0.448
ꢀ0.552
ꢀ0.552
ꢀ0.538
ꢀ0.526
ꢀ0.553
ꢀ0.562
ꢀ0.557
ꢀ0.547
ꢀ0.544
ꢀ0.534
ꢀ0.565
0.568
0.496
0.459
0.440
0.613
0.645
0.583
0.533
0.463
0.448
0.605
ꢀ0.587
ꢀ0.553
ꢀ0.541
ꢀ0.535
ꢀ0.587
ꢀ0.590
ꢀ0.588
ꢀ0.581
ꢀ0.556
ꢀ0.554
ꢀ0.594
Et
i-Pr
CH3
CH2F
CHF2
CF3
9
4.4. Asymmetric transfer hydrogenation in water
(Investigation scale)
10
11
12
Et
A suspension of [RuCl2(arene)]2 (0.001 mmol) and the ligand
(0.0027 mmol) in H2O (0.5 mL) were stirred at 40 ꢁC for 1 h. Sodium
formate (34 mg, 0.5 mmol) and the ketone (0.1 mmol) were then
added and the mixture was stirred vigorously at 40 ꢁC for the
specified number of hours. Samples were withdrawn from the re-
action mixture, extracted with Et2O and filtered through silica be-
fore analysis by HPLC for determination of conversion and ee.
4.2. Analyses
NMR spectra were recorded with Bruker Avance DPX 400 or
300 MHz 1H and 13C NMR chemical shifts are in parts per million
relative to TMS, while for 19F NMR the shift values are relative to
hexafluorobenzene. Coupling constants are in hertz. High resolu-
tion MS (EI/70 eV) was performed using a Finnigan MAT 95 XL. FTIR
spectra were recorded on a Thermo Nicolet Avatar 330 infrared
spectrophotometer. All melting points are uncorrected and mea-
4.5. Reference materials
The identity of the (R)-1-aryl-2-ethanols was confirmed by co-
injection of the (S)-1-aryl-2-ethanols, 13aee, prepared by enzyme
catalysed resolution using lipase B from Candida antarctica and
vinyl acetate as acyl donor. 1H NMR and optical rotation corre-
sponded with that reported previously. The enantioenriched 2-
fluoro-1-arylethanols 14aeg have been described previously.15,27
€
sured by a Buchi melting point instrument. HPLC was performed
using an Agilent 1100 series system with a DAD detector. GC was
performed using a Varian 3380. CD spectra were recorded on an
OLIS DSM 1000 spectrophotometer in a 1 cm cell, at concentration
0.01 mg/mL in MeCN.
The ee of 13a was determined using HPLC on a Chiracel
OD column (0.46 cmꢂ25 cm), mobile phase: hexane/2-propanol
(99/1), flow rate 1.0 mL/min, (R)-13a: 23.7 min, (S)-13a: 27.3 min.
Determination of the ee of 13bee was performed using GC and
a CP-Chirasil-Dex CB column, column pressure: 10 psi, split flow:
4.5.1. (S)-2,2-Difluoro-1-phenylethanol (15b)21,23. The reaction was
performed starting with [RuCl2(mesitylene)]2 (12 mg, 0.02 mmol),
TsDPEN (22 mg, 0.06 mmol) and 2,2,-difluoro-1-phenylethanone
(3b) (312 mg, 2.00 mmol) in formic acid/NEt3 (5/2 mol ratio,

ꢀ
S.V. Slungard et al. / Tetrahedron 67 (2011) 5642e5650
5649
5 mL) at 40 ꢁC. Full conversion was obtained after 2.5 h. The mixture
was diluted with CH2Cl2 (50 mL) and extracted with water
(4ꢂ25 mL) and brine (25 mL) and dried over Na2SO4 The crude
product was purified by silica-gel column chromatography (hex-
ane/EtOAc, 8/2, Rf¼0.27) and gave 290 mg (1.84 mmol, 92%) of
(5/2 mol ratio, 3.0 mL) and 2,2,2-trifluoro-1-(4-(trifluoromethyl)
phenyl)ethanone (4e) (287 mg, 1.19 mmol) were added and the
mixture was stirred vigorously at 40 ꢁC for 18 h. Work up as de-
scribed for 16a gave after silica-gel column chromatography
(CH2Cl2, Rf¼0.55) 168 mg (0.69 mmol, 58%) of a clear oil, ee¼13%,
a clear oil, ee¼90.0%, ½a D24
ꢃ
þ16.5 (c 1.00, CH2Cl2), lit.21 ee¼99%, ½a D24
ꢃ
½
a 2D0
ꢃ
þ4.6 (c 1.06, CH2Cl2). 1H NMR (400 MHz, CDCl3)
d: 7.68 (m,
þ19.4 (c 0.96, CH2Cl2). 1H NMR (300 MHz, CDCl3)
d: 7.45e7.37 (m,
2H), 7.63 (m, 2H), 5.12 (m, 1H), 3.17 (d, J¼4.5, 1H, OH).
5H), 5.77 (dt, J¼56.0, 4.8, 1H), 4.84 (m, 1H), 2.37 (d, J¼3.9, 1H, OH).
4.5.7. (R)-1-(Naphthalen-1-yl)ethanol (20)48. Due to a low rate in
formic acid/triethylamine the following protocol was used to pro-
vide a reference compound: a suspension of [RuCl2(cymene)]2,
(7.3 mg, 0.012 mmol) and TsDPEN (12.7 mg, 0.035 mmol) in water
(2.0 mL) was stirred for 1 h at 40 ꢁC. Formic acid (68 mg,
1.48 mmol), NEt3 (124 mg, 1.23 mmol) and 1-(1-naphthalen-1-yl)
ethanone (8) (210 mg, 1.23 mmol) were added and the mixture was
stirred for 18 h at 40 ꢁC. Work-up was performed as described for
16a and silica-gel column chromatography (hexane/EtOAc, 2/1,
Rf¼0.21) gave 150 mg (0.87 mmol, 71%) of a white solid, mp
4.5.2. (S)-2,2,2-Trifluoro-1-(4-methoxyphenyl)ethanol (16a)43. The
reaction was performed starting with [RuCl2(cymene)]2 (7.3 mg,
0.012 mmol), TsDPEN (11.8 mg, 0.032 mmol) and 2,2,2-trifluoro-1-
(4-methoxyphenyl)ethanone (4a) (241 mg, 1.18 mmol) in formic
acid/NEt3 (5/2 mol ratio, 3 mL) at 40 ꢁC. Full conversion was
obtained after 2.5 h. The mixture was diluted with CH2Cl2 (50 mL)
and extracted with water (4ꢂ25 mL) brine (25 mL) and dried over
Na2SO4 The crude product was purified by silica-gel column chro-
matography (hexane/EtOAc, 8/2, Rf¼0.36) and gave 112 mg
(0.54 mmol, 46%) of a clear oil, ee¼42.0%, ½a D20
þ13.5 (c 1.00,
ꢃ
62e63 ꢁC, lit.48 66e67 ꢁC, ee¼89.5%, ½a 2D0
ꢃ
þ76.6 (c 1.13, Et2O), lit.48
CH2Cl2), lit.44 (R)-4a, ee[41.0%, ½a 2D0
ꢃ
ꢀ8.9 (c 1.00, CH2Cl2). 1H NMR
ee>99.5%, ½a 2D5
ꢃ
þ82.1 (c 1.00, Et2O). 1H NMR (300 MHz, CDCl3)
d:
(400 MHz, CDCl3)
d
: 7.42 (d, J¼8.9, 2H), 6.96 (d, J¼8.9, 2H), 4.99 (m,
8.15 (d, J¼8.1,1H), 7.86 (m, 1H), 7.77 (d, J¼8.1,1H), 7.67 (d, J¼7.2, 1H),
7.54e7.46 (m, 3H), 5.71 (dq, J¼6.5, 3.6, 1H), 1.9 (d, J¼3.6, 1H, OH),
1.69 (d, J¼6.5, 3H).
1H, CH), 3.85 (s, 3H), 2.56 (d, J¼4.4, 1H, OH).
4.5.3. (S)-2,2,2-Trifluoro-1-phenylethanol (16b)44. The reaction was
performed as described for 16a starting with 2,2,2-trifluoro-1-
phenylethanone (4b) (205 mg, 1.18 mmol). The crude product was
purified by silica-gel column chromatography (CH2Cl2, Rf¼0.77)
4.5.8. (S)-2-Fluoro-1-(naphthalen-1-yl)ethanol (21). The reaction
was performed as described for 16a starting with 2-fluoro-1-
(naphthalen-1-yl)ethanone (9) (200 mg, 1.06 mmol). The reaction
time was 2 h. The crude product was purified by silica-gel column
chromatography (CH2Cl2, Rf¼0.27) and gave 170 mg (0.89 mmol,
and gave 91 mg (0.52 mmol, 44%) of a clear oil, ee¼49%, ½a D20
þ12.7
ꢃ
(c 0.40, CH2Cl2), lit.44 (R)-4b: ee¼56%, ½a 2D0
ꢀ12.5 (c 0.40, CH2Cl2).
ꢃ
1H NMR (400 MHz, CDCl3)
d: 7.47e7.40 (m, 5H), 5.00 (m, 1H), 2.63
84%) of a white solid, mp 66e67 ꢁC, ee¼89.0%, ½a D20
þ52.9 (c 1.07,
ꢃ
(d, J¼4.6, 1H, OH).
EtOH). 1H NMR (300 MHz, CDCl3)
d
: 8.04 (d, J¼8.4, 1H), 7.88 (m, 1H),
7.82 (d, J¼8.0, 1H), 7.73 (m, 1H), 7.57e7.48 (m, 3H), 5.84 (m, 1H),
4.73 (ddd, J¼46.7, 9.8, 2.9, 1H), 4.54 (ddd, J¼48.7, 9.8, 8.4, 1H), 2.65
4.5.4. (R)-2,2,2-Trifluoro-1-(4-fluorophenyl)ethanol (16c)44. [RuCl2-
(mesitylene)]2 (6.7 mg, 0.012 mmol) and TsDPEN (11.8 mg,
0.032 mmol) were suspended in H2O (6.0 mL) and stirred at 40 ꢁC for
1 h. To this mixture was added NaHCO3 (400 mg, 6 mmol) and 2,2,2-
trifluoro-1-(4-fluorophenyl)ethanone (4c) (226 mg, 1.18 mmol) and
the mixture was allowed to react at 40 ꢁC for 24 h. Work up as de-
scribed for 16a and purification using silica-gel column chroma-
tography (CH2Cl2, Rf¼0.41) gave 146 mg (0.75 mmol, 64%), of an oil,
(s, 1H, OH). 13C NMR (100 MHz, CDCl3)
d: 133.7, 133.6, 130.4, 129.1,
128.9, 126.5, 125.8, 125.5, 124.2 (d, J¼2.0), 122.4, 87.7 (d, J¼175.1),
70.1 (d, J¼19.1). 19F NMR (376 MHz, CDCl3)
: ꢀ219.5 (dt, J¼46.6,
d
14.4, 1F). HRMS (EI): 190.0790 (calcd 190.0788, Mþ). IR (KBr, cmꢀ1):
3388, 2954, 1103.
4.5.9. (S)-2,2-Difluoro-1-(naphthalen-1-yl)ethanol (22). The re-
action was performed as described for 16a starting with 2,2-
difluoro-1-(naphthalen-1-yl)ethanone (10) (210 mg, 1.02 mmol).
The reaction time was 2 h. Purification by silica-gel column chro-
matography (CH2Cl2, Rf¼0.47) gave 170 mg (0.82 mmol, 80%) of
ee¼27.0%, ½a 2D0
ꢃ
ꢀ5.7 (c 1.05, CH2Cl2), lit.44 ee¼57%, ½a D20
ꢀ20.0 (c
ꢃ
0.02, CH2Cl2). 1H NMR (400 MHz, CDCl3)
5.00 (m, 1H), 2.56 (s, 1H, OH).
d: 7.45 (m, 2H), 7.09 (m, 2H),
4.5.5. (R)- and (S)-1-(4-Bromophenyl)-2,2,2-trifluoroethanol
(16d)45. The compound was prepared as described for 16c using
a white solid, mp 48e50 ꢁC, ee¼76.5%, ½a D20
ꢃ
þ17.1 (c 1.00, EtOH). 1H
NMR (300 MHz, CDCl3)
d
: 8.08 (d, J¼8.4,1H), 7.92e7.87 (m, 2H), 7.72
1-(4-bromophenyl)-2,2,2-trifluoroethanone
(4d)
(300
mg,
(d, J¼7.0, 1H), 7.58e7.50 (m, 3H), 6.02 (ddd, J¼55.1, 55.1, 4.7, 1H),
1.19 mmol), and the mixture was allowed to react for 2.5 h. Work up
as described for 16a and purification using silica-gel column
chromatography (CH2Cl2, Rf¼0.30) gave a 174 mg (0.68 mmol, 57%),
5.66 (m, 1H), 2.53 (s, 1H, OH). 13C NMR (100 MHz, CDCl3)
d: 133.8,
131.7 (dd, J¼2.5, 2.5), 131.0, 129.6, 129.0, 126.6, 125.9, 125.3, 125.2,
122.9 (d, J¼1.5), 115.8 (t, J¼246.5), 70.8 (dd, J¼23.0, 23.0). 19F NMR
of a white solid, mp 54e56 ꢁC, lit.45 55e56 ꢁC, ee¼12.0%, ½a 2D0
ꢀ1.3
ꢃ
(376 MHz, CDCl3)
d
: ABMX-spin system ꢀ125.8 (ddd, JAB¼282.5,
(c 1.04, EtOH), lit.45 ee¼92.5%, ½a D20
ꢃ
ꢀ27.5 (c 1.06, EtOH). 1H NMR
JAM¼54.9, JAX¼7.1, 1F), ꢀ127.3 (ddd, JAB¼282.5, JAM¼56.4, JAX¼12.4,
1F). HRMS (EI): 208.0703 (calcd 208.0694, Mþ). IR (KBr, cmꢀ1):
3402, 1139.
(400 MHz, CDCl3) d: 7.55 (m, 2H), 7.36 (m, 2H), 5.00 (m, 1H), 2.78 (d,
J¼4.6, 1H, OH).
Due to the low ee and the low optical rotation, a sample of the
(S)-enantiomer was prepared by reduction using Geotrichum can-
didum acetone powder as described by Nakamura et al.46 Starting
with 4d (100 mg, 0.40 mmol) this gave 21 mg (0.08 mmol, 21%) of
4.5.10. (R)-2,2,2-Trifluoro-1-(naphthalen-1-yl)ethanol (23)43,49. A
suspension of [RuCl2(mesitylene)]2 (5.6 mg, 0.01 mmol) and (R,R)-
TsCYDN (6.7 mg, 0.025 mmol) in water (4.5 mL) was stirred for 1 h
at 40 ꢁC. To this mixture was added sodium formate (0.30 g,
4.48 mmol) and 2,2,2-trifluoro-1-(naphthalen-1-yl)ethanone (11)
(200 mg, 0.89 mmol). Reaction for 24 h at 40 ꢁC, work up as de-
scribed for 16a, and silica-gel column chromatography (hexane/
EtOAc, 5/1, Rf¼0.18) gave 113 mg (0.50 mmol, 56%) of a white solid,
a white solid, mp 54e56 ꢁC, ee¼98%, ½a D20
ꢃ
þ28.0 (c 1.00, EtOH), lit.21
½
a 2D4
ꢃ
þ30.25 (c 0.862, EtOH). 1H NMR was identical to that described
for (R)-16d.
4.5.6. (S)-2,2,2-Trifluoro-1-(4-(trifluoromethyl)phenyl)ethanol
(16e)47. [RuCl2(cymene)]2 (7.3 mg, 0.012 mmol) and TsCYDN
(8.7 mg, 0.032 mmol) in CH2Cl2 (6.0 mL) were stirred vigorously for
30 min. After removal of the solvent by using N2 gas, HCO2H/NEt3
mp 44e46 ꢁC, lit.49 51.6e53.2 ꢁC, ee¼85.0%, ½a 2D0
ꢀ20.8 (c 1.05,
ꢃ
EtOH), lit.,49
½
a 2D5
ꢃ
ꢀ25.8 (c 5.1 EtOH). 1H NMR (300 MHz, CDCl3)
d:

ꢀ
S.V. Slungard et al. / Tetrahedron 67 (2011) 5642e5650
5650
8.06 (d, J¼8.4, 1H), 7.94e7.89 (m, 2H), 7.83 (d, J¼7.1, 1H), 7.59e7.52
(m, 3H), 5.90 (m, 1H), 2.61 (d, J¼4.8, 1H, OH).
References and notes
1. Ikariya, T.; Blacker, A. J. Acc. Chem. Res. 2007, 40, 1300e1308.
2. Gladiali, S.; Alberico, E. Chem. Soc. Rev. 2006, 35, 226e236.
3. Hayes, A. M.; Morris, D. J.; Clarkson, G. J.; Wills, M. J. Am. Chem. Soc. 2005, 127,
7318e7319.
4. Wu, X.; Li, X.; Zanotti-Gerosa, A.; Pettman, A.; Liu, J.; Mills, A. J.; Xiao, J. Chem.dEur.
J. 2008, 14, 2209e2222.
5. Wu, X.; Xiao, J. Chem. Commun. 2007, 2449e2466.
6. Wu, X.;Liu,J.;Tommaso,D.D.;Iggo,J. A.;Catlow, C. R.A.;Bacsa,J.;Xiao,J.Chem.dEur.
J. 2008, 14, 7699e7715.
7. Petra, D. G. I.; Reek, J. N. H.; Handgraaf, J. W.; Meijer, E. J.; Dierkes, P.; Kamer, P.
C. J.; Brussee, J.; Schoemaker, H. E.; Van Leeuwen, P. W. N. M. Chem.dEur. J.
2000, 6, 2818e2829.
4.5.11. (R)-1-(Naphthalen-1-yl)propan-1-ol (24)50. A reference
compound was prepared as described for 20 starting with 1-
(naphthalen-1-yl)propan-1-one (12) (230 mg, 1.25 mmol). This
gave after silica-gel column chromatography (hexane/EtOAc, 2/1,
Rf¼0.28) (R)-24 as an oil, 60 mg (0.32 mmol, 26%), ee¼78.0%, ½a D20
ꢃ
þ61.6 (c 0.50, benzene), lit.51 ee¼77%, ½a D18
þ61.1 (c 0.44, benzene).
ꢃ
1H NMR (300 MHz, CDCl3)
d
: 8.11 (d, J¼9.3, 1H), 7.86 (d, J¼7.2, 1H),
7.77 (d, J¼8.1, 1H), 7.63 (d, J¼7.1, 1H), 7.53e7.46 (m, 3H), 5.42 (m,
1H), 2.04e1.91 (m, 3H), 1.04 (t, J¼7.4, 3H).
8. Chen, Y.; Liu, S.; Lei, M. J. Phys. Chem. C 2008, 112, 13524e13527.
9. Chen, Y.; Tang, Y.; Liu, S.; Lei, M.; Fang, W. Organometallics 2009, 28,
2078e2084.
4.5.12. (R)-1-(Naphthalen-1-yl)ethyl benzoate (25)52. To a mixture
of (R)-1-(naphthalen-1-yl)ethanol (20) (517 mg, 3.00 mmol) and
NEt3 (1.37 g, 13.53 mmol) in CH2Cl2 (20 mL) at 0 ꢁC was added
benzoyl chloride (0.67 g, 4.77 mmol). After stirring for 12 h at room
temperature, brine (25 mL) was added, and the mixture was
extracted with diethyl ether (3ꢂ25 mL). The combined organic
fraction was washed with aq HCl soln (5%, 3ꢂ15 mL), brine (25 mL)
and satd aq NaHCO3 (25 mL). Drying over Na2SO4, concentration in
vacuum, and silica-gel column chromatography (hexane/EtOAc, 4/
1, Rf¼0.47), gave 334 mg (1.21 mmol, 40%) of a clear oil, ee¼98%,
10. Fujii, A.; Hashiguchi, S.; Uematsu, N.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc.
1996, 118, 2521e2522.
11. Noyori, R.; Hashiguchi, S. Acc. Chem. Res. 1997, 30, 97e102.
12. Yamakawa, M.; Yamada, I.; Noyori, R. Angew. Chem., Int. Ed. 2001, 40,
2818e2821.
13. Noyori, R.; Yamakawa, M.; Hashiguchi, S. J. Org. Chem. 2001, 66, 7931e7944.
14. Brandt, P.; Roth, P.; Andersson, P. G. J. Org. Chem. 2004, 69, 4885e4890.
15. Fuglseth, E.; Sundby, E.; Hoff, B. H. J. Fluorine Chem. 2009, 130, 600e603.
16. Sterk, D.; Stephan, M.; Mohar, B. Org. Lett. 2006, 8, 5935e5938.
17. Gamez, P.; Fache, F.; Lemaire, M. Tetrahedron: Asymmetry 1995, 6, 705e718.
18. von Arx, M.; Mallat, T.; Baiker, A. Tetrahedron: Asymmetry 2001, 12, 3089e3094.
19. Ramachandran, P. V.; Teodorovic, A. V.; Brown, H. C. Tetrahedron 1993, 49,
1725e1738.
20. Yong, K. H.; Chong, J. M. Org. Lett. 2002, 4, 4139e4142.
21. Matsuda, T.; Harada, T.; Nakajima, N.; Itoh, T.; Nakamura, K. J. Org. Chem. 2000,
65, 157e163.
22. Ohkuma, T.; Koizumi, M.; Doucet, H.; Pham, T.; Kozawa, M.; Murata, K.; Ka-
tayama, E.; Yokozawa, T.; Ikariya, T.; Noyori, R. J. Am. Chem. Soc. 1998, 120,
13529e13530.
½
a 2D5
ꢃ
ꢀ101.8 (c 1.00, EtOH). CD (MeCN): D3¼ꢀ12.6 (223 nm). 1H
NMR (300 MHz, CDCl3)
d: 8.18 (d, J¼8.2, 1H), 8.10 (m, 2H), 7.87 (m,
1H), 7.80 (d, J¼8.3, 1H), 7.70 (d, J¼7.1, 1H), 7.60e7.42 (m, 6H), 6.90
(q, J¼6.6, 1H), 1.85 (d, J¼6.6, 3H).
23. Goushi, S.; Funabiki, K.; Ohta, M.; Hatano, K.; Matsui, M. Tetrahedron 2007, 63,
4061e4066.
24. Mallat, T.; Bodmer, M.; Baiker, A. Catal. Lett. 1997, 44, 95e99.
25. Fuglseth, E.; Thvedt, T. H. K.; Førde Møll, M.; Hoff, B. H. Tetrahedron 2008, 64,
7318e7323.
26. Thvedt, T. H. K.; Fuglseth, E.; Sundby, E.; Hoff, B. H. Tetrahedron 2009, 65,
9550e9556.
27. Thvedt, T. H. K.; Fuglseth, E.; Sundby, E.; Hoff, B. H. Tetrahedron 2010, 66,
6733e6743.
28. Hanch, C.; Leo, A.; Hoekman, D. Exploring QSAR-Hydrophobic, Electronic and
Steric Constants; ACS Professional Reference Book; American Chemical Society:
Washington, DC, 1995.
29. Wu, X.; Li, X.; Hems, W.; King, F.; Xiao, J. Org. Biomol. Chem. 2004, 2, 1818e1821.
30. Liu, W.; Cui, X.; Cun, L.; Zhu, J.; Deng, J. Tetrahedron: Asymmetry 2005, 16,
2525e2530.
31. Sinnokrot, M. O.; Sherrill, C. D. J. Am. Chem. Soc. 2004, 126, 7690e7697.
32. Meyer, E. A.; Castellano, R. K.; Diederich, F. Angew. Chem., Int. Ed. 2003, 42,
1210e1250.
33. Wu, X.; Vinci, D.; Ikariya, T.; Xiao, J. Chem. Commun. 2005, 4447e4449.
34. Bervova, N.; Nakanishi, K. Circular dichroism. Principles and Applications In.
Exciton Chirality Method: Principles and Applications; Berova, N., Nakanishi, K.,
Woody, R. W., Eds.; Wiley-VCH: New York, NY, 2000; pp 337e382.
35. Harada, N.; Nakanishi, K. Circular Dichroism Spectroscopy. Exciton Coupling in
Organic Stereochemistry; University Science Books: Mill Valley, 1983; 1e460.
36. Adam, W.; Lukacs, Z.; Viebach, K.; Humpf, H. U.; Saha-Moeller, C. R.; Schreier, P.
J. Org. Chem. 2000, 65, 186e190.
37. Berova, N.; Di Bari, L.; Pescitelli, G. Chem. Soc. Rev. 2007, 36, 914e931.
38. Fuglseth, E.; Ottherholt, E.; Høgmoen, H.; Sundby, E.; Charnock, C.; Hoff, B. H.
Tetrahedron 2009, 65, 9807e9813.
39. Verniest, G.; Van Hende, E.; Surmont, R.; De Kimpe, N. Org. Lett. 2006, 8,
4767e4770.
4.5.13. (S)-2-Fluoro-1-(naphthalen-1-yl)ethyl benzoate (26). The
synthesis was performed as described for 25, starting with (S)-21
(105 mg, 0.55 mmol). This gave 120 mg (0.41 mmol, 74%) of a white
solid, mp 74e75 ꢁC, Rf¼0.41, ee¼89.0%, ½a 2D0
ꢀ171.6 (c 1.00, EtOH),
ꢃ
CD (MeCN): D3¼ꢀ28.8 (227 nm). 1H NMR (400 MHz, CDCl3)
d: 8.25
(d, J¼8.8, 1H), 8.16 (m, 2H), 7.89 (d, J¼8.4, 1H), 7.85 (d, J¼8.0, 1H),
7.69 (d, J¼6.8, 1H), 7.64e7.58 (m, 2H), 7.54 (m, 1H), 7.51e7.46 (m,
3H), 7.14e7.07 (m, 1H), 4.90 (ddd, J¼48.1, 10.5, 7.8, 1H), 4.83 (ddd,
J¼46.6, 10.5, 3.0, 1H). 13C NMR (100 MHz, CDCl3)
d: 165.6, 133.8,
133.3, 131.0, 130.4, 129.8 (2C), 129.8, 129.4, 129.1, 128.5 (2C), 126.9,
126.0, 125.3, 124.7, 122.6, 84.8 (d, J¼180.1), 72.1 (d, J¼21.1). 19F NMR
(376 MHz, CDCl3)
d
: ꢀ220.2 (dt, J¼47.4, 15.9, 1F). HRMS (EI):
294.1059 (calcd 294.1051, Mþ). IR (KBr, cmꢀ1): 3062, 1717, 1287,
1104.
4.5.14. (S)-2,2-Difluoro-1-(naphthalen-1-yl)ethyl benzoate (27). The
synthesis was performed as described for 25, starting with (S)-22
(110 mg, 0.53 mmol). Work up and purification by silica-gel column
chromatography (CH2Cl2/hexane, 4/1, Rf¼0.56) gave 140 mg
(0.45 mmol, 85%) of a white solid, mp 86e88 ꢁC, ee¼76.5% (based
on 22), ½a 2D5
ꢃ
ꢀ178.6 (c 1.09, EtOH), CD (MeCN): D3¼ꢀ28.0 (225). 1H
NMR (400 MHz, CDCl3)
d
: 8.24 (d, J¼8.4, 1H), 8.14 (m, 2H), 7.90 (d,
J¼7.6, 2H), 7.76 (d, J¼7.2, 1H), 7.65e7.59 (m, 2H), 7.57e7.47 (m, 4H),
7.00 (dt, 1H, J¼10.9, 4.2, 1H), 6.26 (dt, J¼55.2, 4.2, 1H). 13C NMR
40. Palmer, M. J.; Kenny, J. A.; Walsgrove, T.; Kawamoto, A. M.; Wills, M. J. Chem.
Soc., Perkin Trans. 1 2002, 416e427.
(100 MHz, CDCl3) d: 164.9, 133.8, 133.6, 131.1, 130.0, 129.9 (2C),
41. Bennett, M. A.; Smith, A. K. Dalton Trans. 1974, 233e241.
42. Fuglseth, E.; Sundby, E.; Bruheim, P.; Hoff, B. H. Tetrahedron: Asymmetry 2008,
19, 1941e1946.
43. Xu, Q.; Zhou, H.; Geng, X.; Chen, P. Tetrahedron 2009, 65, 2232e2238.
44. Zhao, H.; Qin, B.; Liu, X.; Feng, X. Tetrahedron 2007, 63, 6822e6826.
45. O’Shea, P. D.; Chen, C. Y.; Gauvreau, D.; Gosselin, F.; Hughes, G.; Nadeau, C.;
Volante, R. P. J. Org. Chem. 2009, 74, 1605e1610.
46. Nakamura, K.; Matsuda, T. J. Org. Chem. 1998, 63, 8957e8964.
47. Hughes, G.; O’Shea, P.; Goll, J.; Gauvreau, D.; Steele, J. Tetrahedron 2009, 65,
3189e3196.
48. Theisen, P. D.; Heathcock, C. H. J. Org. Chem. 1988, 53, 2374e2378.
49. Pirkle, W. H.; Hoekstra, M. S. J. Org. Chem. 1974, 39, 3904e3905.
50. Hatano, M.; Miyamoto, T.; Ishihara, K. J. Org. Chem. 2006, 71, 6474e6484.
51. Nakamura, Y.; Takeuchi, S.; Okumura, K.; Ohgo, Y. Tetrahedron 2001, 57,
5565e5571.
129.2, 129.0, 128.9 (t, J¼2.5), 128.5 (2C), 127.0, 126.2, 126.0, 125.2,
122.9, 114.1 (t, J¼246.3), 71.0 (t, J¼26.2). 19F NMR (376 MHz, CDCl3)
d
: ꢀ126.2 (dd, J¼55.5, 11.3, 1F), ꢀ126.4 (dd, J¼55.0, 10.2, 1F). HRMS
(EI): 312.0958 (calcd 312.0956, Mþ) IR (KBr, cmꢀ1): 3058, 1719,
1284, 1110.
Acknowledgements
The group is thankful to the Anders Jahres foundation for fi-
nancial support. Ingemund M.F. Engøy and Roger Aarvik are
thanked for their contributions. Susana Villa Gonzalez is acknowl-
edged for the HRMS experiments.
52. Suzuki, Y.; Muramatsu, K.; Yamauchi, K.; Morie, Y.; Sato, M. Tetrahedron 2005,
62, 302e310.